| Time |
Nick |
Message |
| 00:01 |
kahrl |
it might not be as easy as that |
| 00:01 |
kahrl |
if you have a furnace open, you want fuel to go to the fuel list and cookables to go to the cook list |
| 00:01 |
PilzAdam |
so, we officially have the Lua menu now? |
| 00:02 |
hmmmm |
hey pilzadam, how's that VBO code coming? |
| 00:02 |
PilzAdam |
its idling arround |
| 00:04 |
kahrl |
there were some things I would have changed in the lua formspec before merging |
| 00:04 |
kahrl |
menu* |
| 00:04 |
kahrl |
but oh well |
| 00:04 |
hmmmm |
like what? |
| 00:05 |
PilzAdam |
also we already agreed on removing Finite Liquids from the GUI |
| 00:05 |
hmmmm |
all of these things can be fixed |
| 00:05 |
kahrl |
hmmmm: yeah they can |
| 00:05 |
hmmmm |
lol. besides, if it wasn't merged, nobody would ever talk about these things. |
| 00:05 |
kahrl |
I would eventually, it's just that I'm busy at the moment |
| 00:06 |
hmmmm |
understandable |
| 00:10 |
PilzAdam |
http://mg.viewskew.com/mgoblin_media/media_entries/59/FormspecMenu.png |
| 00:10 |
PilzAdam |
^ win build |
| 00:11 |
PilzAdam |
(mingw) |
| 00:11 |
PilzAdam |
we should have tested it on windows first... |
| 00:13 |
PilzAdam |
when I leave a world to the menu it either segfaults or it looks correctly |
| 00:14 |
PilzAdam |
but its always like the screenshot when starting Minetest |
| 00:15 |
PilzAdam |
seems like the segfault is not reproduceable |
| 00:18 |
PilzAdam |
hmmmm, I get the "two entries in one line" in the credits tab too |
| 00:18 |
PilzAdam |
oh, actually thats even 3 in one line :-) |
| 00:22 |
hmmmm |
hrmmmmm |
| 00:22 |
hmmmm |
can you readily reproduce it? |
| 00:23 |
PilzAdam |
yes |
| 00:23 |
hmmmm |
that's good at least |
| 00:23 |
PilzAdam |
the last 3 entries of "Active Contributors" are in the same line |
| 00:23 |
hmmmm |
we'll see if sapier has a windows testing setup |
| 00:24 |
PilzAdam |
also the Esc key doesnt work always to quit Minetest |
| 00:24 |
PilzAdam |
some focus thingy I guess |
| 00:24 |
hmmmm |
never knew you could do that |
| 00:25 |
PilzAdam |
oh dear, the "Configure" window is completly messed up |
| 00:26 |
PilzAdam |
its not sorted, modpacks are not "packed", the list doesnt work correctly (sometimes the selection follows the mouse all the time) |
| 00:26 |
hmmmm |
the configure window completely fails for me |
| 00:27 |
hmmmm |
obviously this is a work in progress and it's upstream |
| 00:27 |
hmmmm |
we'll just fix all of these things.... what did you honestly expect for an 8000 line commit? |
| 00:27 |
PilzAdam |
and the game icons are still in a lower resolution |
| 00:28 |
kahrl |
why is this https://github.com/minetest/minetest/issues/801 marked won't add? |
| 00:29 |
PilzAdam |
I wondered about that too |
| 00:30 |
kahrl |
I'll change it to low-priority enhancement |
| 00:30 |
PilzAdam |
I guess sapier has a long TODO list now ;-) |
| 00:32 |
kahrl |
heh, sapier snuck the change in the debug build from -O1 to -O0 into upstream |
| 00:33 |
PilzAdam |
again!? |
| 00:33 |
hmmmm |
... |
| 00:33 |
hmmmm |
god damn him |
| 00:33 |
PilzAdam |
its like the third time he does that |
| 00:33 |
hmmmm |
anyone want to change it back? |
| 00:33 |
hmmmm |
well it might not've been intentional |
| 00:33 |
kahrl |
yeah, O0 is terribly slow for me |
| 00:34 |
hmmmm |
i mean, here he's working on main menu things and he needs his debugger to output actual information |
| 00:34 |
hmmmm |
changed it, forgot to change it back |
| 00:34 |
kahrl |
although O0 does make it easier to see real warning messages |
| 00:34 |
hmmmm |
honestly, debugging with -O1 is a farce |
| 00:34 |
hmmmm |
it helps kinda |
| 00:34 |
hmmmm |
but you can mostly get the same results by adding printfs everywhere |
| 00:35 |
kahrl |
I personally would probably use -D CMAKE_BUILD_TYPE=RelWithDebInfo if minetest supported it |
| 00:36 |
kahrl |
unless I need to dig deeper with gdb, of course |
| 00:36 |
hmmmm |
i suppose we could add that too |
| 00:38 |
kahrl |
I would add but I have no idea about MSVC options |
| 00:39 |
hmmmm |
pretty sure it's the same, with the caveat that you can't use whole program optimization with debugging symbols or something along those lines |
| 00:44 |
hmmmm |
alright: http://dev.minetest.net/TODO#Formspec_main_menu |
| 00:45 |
|
dexter0 joined #minetest-dev |
| 00:48 |
hmmmm |
hey... at least the bugs aren't as hidden as those in the STL container migration |
| 00:52 |
kahrl |
I'm procrastinating so I fixed #803 and #782 in a less hacky way: https://github.com/kahrl/minetest/commit/4aba0d5f0f73c560f831486a7e86278ce7ad1fff |
| 00:53 |
|
Neological joined #minetest-dev |
| 00:54 |
kahrl |
02:50:05: INFO[main]: Remaining textures : 6 <-- after running two singleplayer games |
| 00:54 |
kahrl |
no more leftover _RTT textures |
| 00:57 |
|
BrandonReese_ joined #minetest-dev |
| 00:57 |
|
hmmmmm joined #minetest-dev |
| 00:57 |
|
DemonRaiser joined #minetest-dev |
| 00:58 |
VanessaE |
error: the new formspec menu doesn't respect HDX's menu overlays. |
| 00:59 |
VanessaE |
(presumably any other texture pack as well) |
| 00:59 |
kahrl |
we should release 0.4.8 right now to piss everyone off :D |
| 00:59 |
VanessaE |
hah |
| 01:00 |
|
smoke_fumus|2 joined #minetest-dev |
| 01:03 |
|
OWNSyouAll_DESKT joined #minetest-dev |
| 01:03 |
|
dexter00 joined #minetest-dev |
| 01:04 |
kahrl |
if someone is okay with https://github.com/kahrl/minetest/commit/4aba0d5f0f73c560f831486a7e86278ce7ad1fff I'll merge it |
| 01:05 |
|
drizz joined #minetest-dev |
| 01:06 |
VanessaE |
oh sure, don't even bother fixing my issue :D |
| 01:06 |
VanessaE |
(kidding!) |
| 01:07 |
PilzAdam |
kahrl, ehm, that doesnt build for me |
| 01:07 |
PilzAdam |
/home/adam/Minetest/minetest/src/tile.h:65:2: error: 'IMesh' in namespace 'irr::scene' does not name a type |
| 01:07 |
|
BackupCoder_ joined #minetest-dev |
| 01:10 |
kahrl |
hmm |
| 01:10 |
PilzAdam |
adding "namespace irr {namespace scene {class IMesh;}}" works |
| 01:10 |
kahrl |
it builds for me, strange, maybe because of irrlicht 1.8 |
| 01:10 |
kahrl |
adding that looks hacky |
| 01:10 |
PilzAdam |
I have Irrlicht 1.7.3 |
| 01:10 |
VanessaE |
irrlicht 1.8 is evil. |
| 01:10 |
PilzAdam |
^ |
| 01:11 |
kahrl |
mesh stuff is in irrlichttypes_extrabloated.h |
| 01:11 |
kahrl |
the 1.8 headers probably implicitly include IMesh.h |
| 01:12 |
PilzAdam |
we do the same for e.g. ISceneManager in main.h |
| 01:12 |
kahrl |
tile.h is included everywhere so include irrlichttypes_extrabloated.h in tile.h is not the greatest of ideas |
| 01:12 |
kahrl |
PilzAdam: ah |
| 01:14 |
kahrl |
I'll do that then |
| 01:14 |
VanessaE |
nice name for a header file. |
| 01:14 |
|
yno joined #minetest-dev |
| 01:18 |
kahrl |
force pushed: https://github.com/kahrl/minetest/commit/681ae0e273faed3a49a25079d986d9934d1114dd |
| 01:18 |
|
darkrose joined #minetest-dev |
| 01:18 |
|
darkrose joined #minetest-dev |
| 01:22 |
PilzAdam |
kahrl, seems to work fine now |
| 01:22 |
PilzAdam |
anything special to test here? |
| 01:22 |
kahrl |
not really |
| 01:22 |
PilzAdam |
valgrind anyone? ;-) |
| 01:23 |
kahrl |
not me :P |
| 01:23 |
PilzAdam |
its good then |
| 01:30 |
VanessaE |
kahrl: not to harass, but can you fix what sapier broke in those menu overlays? :) |
| 01:30 |
kahrl |
nope ;) |
| 01:30 |
VanessaE |
:P |
| 01:51 |
hmmmmm |
grrr |
| 01:52 |
hmmmmm |
i can't even continue working on what i was working on |
| 01:52 |
hmmmmm |
make a new world |
| 01:52 |
hmmmmm |
new world isn't highlighted |
| 01:52 |
hmmmmm |
and i have to actually really pick "v7" from the menu every time? |
| 01:52 |
VanessaE |
ugh, that single-click-starts-a-world is fucking irritating! |
| 02:00 |
hmmmmm |
the "dropdown" formspec element type is completely undocumented along with all the other new ones added |
| 02:18 |
hmmmmm |
sapier, if you read this: there's absolutely no way to get a list of available mapgens, it's just impossible with the architecture of minetest. having the list mirrored in the main menu lua scripts is simply going to have to do |
| 02:28 |
kahrl |
hmmmmm: wouldn't something like server->getEmergeManager()->getMapgenFactoryNames() be possible? |
| 02:28 |
kahrl |
in theory |
| 02:28 |
hmmmmm |
the "server->" part is the problem |
| 02:29 |
kahrl |
oh, in the main menu? |
| 02:29 |
hmmmmm |
yeah.. |
| 02:29 |
kahrl |
oh you said that |
| 02:29 |
hmmmmm |
it just isn't possible unless the server things are always compiled along with the client |
| 02:30 |
kahrl |
well, they are, but the Server object isn't created at that point in time |
| 02:31 |
hmmmmm |
they are?? |
| 02:31 |
kahrl |
to be able to play singleplayer |
| 02:31 |
hmmmmm |
i thought you were able to make a multiplayer only version |
| 02:32 |
kahrl |
I don't think so |
| 02:32 |
hmmmmm |
mm. |
| 02:33 |
hmmmmm |
and if there's a lua error, it goes right into the game apparently |
| 02:34 |
hmmmmm |
after you fix the NULL drop() on the clouds |
| 02:37 |
hmmmmm |
and apparently GUIEngine isn't destroyed if there's a lua error either, so if you give any input after it errors, it'll try to crash because you're using an old lua state |
| 02:42 |
hmmmmm |
how do you get irrlicht to stop calling GUIFormSpecMenu::OnEvent()? just destroy the GUIFormSpecMenu instance, correct? |
| 02:45 |
hmmmmm |
ugh, different problem actually |
| 02:48 |
hmmmmm |
frankly i have no idea what i'm doing here |
| 04:06 |
hmmmmm |
wait a minute, what was the whole point of making the main menu formspec based exactly? |
| 04:18 |
VanessaE |
to make it expandable in the future |
| 04:18 |
VanessaE |
themeable games and the like, I guess |
| 04:39 |
VanessaE |
another error I'm seeing is signing onto a server is no longer possible by just hitting enter after typing your password |
| 04:39 |
VanessaE |
(you must click "Connect") |
| 04:39 |
hmmmmm |
i'll fix that while i'm at it |
| 04:39 |
VanessaE |
ok |
| 04:40 |
hmmmmm |
er, not as easy as i thought |
| 04:50 |
hmmmmm |
i can't even connect to anything |
| 04:50 |
hmmmmm |
ERROR: couldn't ersolve address: non-recoverable failure in name resolution |
| 04:50 |
VanessaE |
try my server as a test, it's working okay |
| 04:51 |
hmmmmm |
screw this, i'm done for today |
| 04:51 |
VanessaE |
(aside from some odd bug that I'm not sure if it's UI or what) |
| 04:51 |
hmmmmm |
i assume this is due to proller's recent changes.. |
| 04:52 |
VanessaE |
*shrug* no idea |
| 04:52 |
hmmmmm |
oh |
| 04:53 |
hmmmmm |
derp |
| 04:53 |
VanessaE |
wrong tab? :) |
| 04:53 |
hmmmmm |
the address/port is being filled with the description from the serverlist |
| 04:53 |
VanessaE |
oops! |
| 04:53 |
hmmmmm |
if that resolves i think that'd be pretty amazing |
| 04:53 |
hmmmmm |
alright, simple fix |
| 04:53 |
hmmmmm |
the "enter to connect" thing is not easy to fix at all though |
| 04:53 |
hmmmmm |
i'd wait for sapier |
| 04:59 |
hmmmmm |
alright so what i fixed so far should make it at least usable |
| 05:00 |
|
neko259 joined #minetest-dev |
| 05:00 |
hmmmmm |
pressing enter to start the game isn't possible though |
| 05:01 |
VanessaE |
ok that's gonna be a problem. |
| 05:01 |
VanessaE |
(not for me, I can adjust) |
| 05:01 |
hmmmmm |
lol, that's a very small problem compared to what else there is to fix |
| 05:01 |
VanessaE |
well sure |
| 05:02 |
VanessaE |
I'm just saying to expect a lot of ... colorful, less-than-positive commentary :) |
| 05:02 |
hmmmmm |
configure is fucked completely, any lua errors on init is a crash if you press any key after the game starts |
| 05:02 |
hmmmmm |
and everything else i have on the todo |
| 05:02 |
VanessaE |
I wonder how much of this is because of that last rebase he had to do? |
| 05:03 |
hmmmmm |
none |
| 05:03 |
VanessaE |
oh ok |
| 06:59 |
|
Calinou joined #minetest-dev |
| 07:16 |
|
Anchakor_ joined #minetest-dev |
| 07:27 |
|
darkrose joined #minetest-dev |
| 07:27 |
|
darkrose joined #minetest-dev |
| 07:37 |
|
EdB joined #minetest-dev |
| 08:28 |
|
Calinou joined #minetest-dev |
| 08:49 |
|
darkrose joined #minetest-dev |
| 09:03 |
Taoki |
Anyone with upstream access wish to merge this? https://github.com/minetest/minetest/pull/799 Already been up for testing several days, no more issues and complaints apparently |
| 09:04 |
* Taoki |
then figures many of the core devs are in the US and probably sleeping right now |
| 09:21 |
|
Zeg9 joined #minetest-dev |
| 09:54 |
|
proller joined #minetest-dev |
| 10:01 |
|
neko259 joined #minetest-dev |
| 10:26 |
|
PilzAdam joined #minetest-dev |
| 10:43 |
|
Calinou joined #minetest-dev |
| 10:45 |
PilzAdam |
Taoki, AFAIK only hmmmm lives in the US |
| 10:46 |
Taoki |
oh |
| 10:47 |
PilzAdam |
Russia and Germany are the countries you are looking for :-) |
| 10:50 |
Exio |
nobody is from the south hemisphere though :( |
| 10:51 |
PilzAdam |
Taoki, it looks way better now |
| 10:52 |
Taoki |
I worked on fixing all issues as much as possible, so it should be as good as it can get |
| 10:52 |
PilzAdam |
though the coloring starts too early (at ~90°) |
| 10:52 |
PilzAdam |
it should be more like 45° |
| 10:52 |
Taoki |
That's intended. If it would start less early, it would also end sooner. The moon fog would also last less |
| 10:53 |
Taoki |
As well as the purple horrizon which I find most beautiful :) |
| 10:53 |
Taoki |
It already ends a bit too soon for the sun, so I found this the best balance |
| 10:54 |
PilzAdam |
it doesnt feel right if you look in one direction and the fog starts getting yellow without any visible cause |
| 10:54 |
Taoki |
PilzAdam: In real life, the horizon stays a bit purple-ish after the sun is over the horrizon |
| 10:55 |
PilzAdam |
in real life the horizon doesnt change the color based on your view direction |
| 10:55 |
PilzAdam |
I still think that its worth a try to make it not yaw dependend |
| 10:56 |
Taoki |
We must consider it like this: Intensity isn't tweaked individually for each fog, it's the same for both (the entire blend). It also doesn't tell between sunrise and sunset. So the center of the effect must be when the sun and moon are aligned perfectly. Over that the effect needs a certain amount of time |
| 10:56 |
Taoki |
(both back and forth) |
| 10:57 |
Taoki |
As for not yaw dependent, that would require a different system. I'm considering it when we (possibly I) might attempt to make sky planets defined in Lua. In which case this would also need to adapt to each position |
| 10:58 |
Taoki |
Right not I think that would be very difficult and not worth it, for a fixed sun and moon where we know where it rises and sets |
| 11:00 |
Taoki |
When sun & moon will be defined in Lua, each sky planet would have a "fog influence" you can set for each position. Not now though... for the hard-coded sky this is best and easiest. |
| 11:01 |
PilzAdam |
can you make a quick patch that makes the coloring start at ~45° so we can see how it looks? |
| 11:01 |
Taoki |
PilzAdam: Easier if I just change the values and pastebin the changes, since it's a few values in a simple location. Will try that now |
| 11:02 |
ironzorg |
/home/lenorm_f/games/minetest/src/mapgen.cpp:663:19: warning: multi-character character constant [-Wmultichar] |
| 11:02 |
ironzorg |
if (signature != 'MTSM') {who the fuck does that |
| 11:02 |
PilzAdam |
ironzorg, hmmmm :-) |
| 11:02 |
Taoki |
PilzAdam: But again, I heavily doubt it will work. Because the system is centered. Maximum intensity must be when sun and moon are aligned, distance equal both ways around that. If we reduce the effect before a planet rises, we also do so after it sets (and for setting it's already too short) |
| 11:04 |
Taoki |
Either way, going to test it now still |
| 11:04 |
ironzorg |
tell hmmmm to stop coding in javascript :| |
| 11:05 |
ironzorg |
minetest should be coded with -pedantic, so I can compile on other devices without having to fix those things |
| 11:06 |
Taoki |
PilzAdam: Tried making it twice shorter. It looks really bad and barely even visible like I expected |
| 11:08 |
|
proller joined #minetest-dev |
| 11:10 |
Taoki |
PilzAdam: Making it 0.75 shorter compared to my current branch instead if a bit more acceptable. Still feels a bit too short, but acceptable at least |
| 11:12 |
Taoki |
But as compensation, I'd increase intensity a bit |
| 11:15 |
Taoki |
PilzAdam: No... that doesn't help either. Even at 0.75 the effect is too late for the sun set, and ends too quickly for the moon rise |
| 11:17 |
PilzAdam |
why is the horizonzal and vertical part linked? |
| 11:17 |
Taoki |
In what sense? |
| 11:18 |
PilzAdam |
I asked you to decrease the horizontal range, but you say that it starts too late and ends too early; I dont see how these two thinks are linked |
| 11:18 |
Taoki |
Ahhh. You mean how the colors are mixed by the pitch? |
| 11:18 |
PilzAdam |
no |
| 11:20 |
Taoki |
Oh, sorry :P You meant the bounds between which the transition starts (yaw). Sorry, my mind being slow again >.< |
| 11:21 |
Taoki |
PilzAdam: I need to go eat now. To change those limits, tweak line 489 in sky.cpp. The one under the line: // bound view angle to determine where transition starts and ends |
| 11:21 |
Calinou |
talking about fog, it would be nice if we had "fancy fog" |
| 11:21 |
Calinou |
like minecraft does |
| 11:21 |
Taoki |
It's a simple one-line formula and you only need to change the bound value. But make sure the numver of division is always the same |
| 11:21 |
PilzAdam |
Calinou, what would that be? |
| 11:21 |
Taoki |
Currently it's 1.375. Try 1.25, 1.5, etc instead and see which looks better |
| 11:21 |
Taoki |
BRB in 15 minutes or so |
| 11:22 |
Taoki |
But tweak that line and see if another value looks better. I found the sky corner most balanced (1.375) |
| 11:24 |
Calinou |
optifine offers 1) fast or fancy fog 2) fog start distance (between 0.2 and 0.8) |
| 11:24 |
Calinou |
default is 0.2 in minecraft |
| 11:25 |
Calinou |
screenshots with fog start 0.8, comparison between fast and fancy fog: https://dl.dropboxusercontent.com/u/82342922/temp/2013-07-03_13.22.57.png https://dl.dropboxusercontent.com/u/82342922/temp/2013-07-03_13.23.00.png |
| 11:26 |
PilzAdam |
I dont really see the point of this |
| 11:27 |
Calinou |
and again 8) |
| 11:27 |
Calinou |
with fog start 0.2 it is way less obvious |
| 11:28 |
Calinou |
minetest uses a "sharper" fog start |
| 11:29 |
proller |
is possible later make fog depend on humidity ? |
| 11:36 |
Taoki |
Back. Will try more options now |
| 11:36 |
PilzAdam |
Taoki, that doesnt really reach what I want |
| 11:36 |
Taoki |
PilzAdam: Ok. Can't think of any other way |
| 11:37 |
Taoki |
For the current system that's the only tweak I can imagine |
| 11:37 |
PilzAdam |
lemme draw a picture :-) |
| 11:37 |
Taoki |
Sometimes only pictures can explain what's difficult to explain in words :P |
| 11:38 |
Taoki |
But I don't know what else could be done |
| 11:38 |
PilzAdam |
http://mg.viewskew.com/u/pilzadam/m/my-awesome-gimp-drawing-skills/ |
| 11:39 |
Taoki |
ok. What about that? |
| 11:39 |
PilzAdam |
I want the fog is colored only if looking at the yellow/blue part |
| 11:39 |
PilzAdam |
so if I e.g. look directly to the right then the fog should have the default color |
| 11:40 |
Taoki |
Oh, that value I mentioned should tweak it then |
| 11:40 |
Taoki |
Just need to extend the sphere bounds |
| 11:40 |
PilzAdam |
well, it doesnt work in the extend I want it to be |
| 11:41 |
Taoki |
Should But highs might need to be mor eextended than ends |
| 11:41 |
PilzAdam |
it currently looks more like this: http://mg.viewskew.com/u/pilzadam/m/my-awesome-gimp-drawing-skills2/ |
| 11:41 |
Taoki |
Also, on that drawing, where should the fading start and end from? As in, during which area should it still fade? |
| 11:42 |
Taoki |
Fading between the empty spaces in the first one? |
| 11:42 |
PilzAdam |
the end points yellow/blue lines are the fade start/end points |
| 11:42 |
PilzAdam |
+of the |
| 11:42 |
PilzAdam |
so there is no coloring at all when looking at the left or right |
| 11:42 |
Taoki |
It should work then. I'll try to tweak it |
| 11:43 |
Taoki |
Ah, wait |
| 11:43 |
Taoki |
No coloring when looking to the side you mean. Yes, the formula isn't optimized for that... |
| 11:43 |
Taoki |
And I'm not sure how it would look. The transition time is short enough as is\ |
| 11:44 |
Taoki |
Wouldn't want the fog to change color too suddenly |
| 11:44 |
Taoki |
But, I'll try to obtain a formula that does this there, to at least test it out |
| 11:44 |
PilzAdam |
the problem is that when looking at the right and turning the camera a bit (lets say 10°) the colors change without any visible reason |
| 11:45 |
Taoki |
Ah. That's meant to represent a gradient, not necessarily the color "changing". |
| 11:45 |
Taoki |
Since like discussed at the start of the topic, we can't have volumetric fog. So we rely on this changing to simulate you're seeing colors already to the left or right |
| 11:46 |
Taoki |
I also re-checked in Minecraft BTW, and it does the same thing. Just more discretely for some reason |
| 11:47 |
Taoki |
PilzAdam: But the idea is that the horrizon color is "already there". The color only changes to represent a gradient we can't have volumetirically. So it makes sense that when looking halfway, it's the middle color. At least what it's meant to simulate |
| 11:48 |
PilzAdam |
but it looks weird |
| 11:48 |
Taoki |
Bounding the halfway area might look even weirder. But I'll try to and see what I come up with |
| 11:56 |
|
darkrose joined #minetest-dev |
| 11:56 |
|
darkrose joined #minetest-dev |
| 11:56 |
celeron55 |
Taoki: i request you to stop calling it "volumetric" |
| 11:57 |
celeron55 |
there's nothing volumetric in varying fog color |
| 11:57 |
Taoki |
Yeah, directional is more correct |
| 11:57 |
celeron55 |
this is volumetric fog: http://udn.epicgames.com/Three/rsrc/Three/FogVolumes/AutomaticConstDensityFogVolume.jpg |
| 11:58 |
Calinou |
^ |
| 12:00 |
Taoki |
true. Not something I'd mine Minetest having BTW :) |
| 12:00 |
Taoki |
Think about water fog |
| 12:03 |
|
iqualfragile joined #minetest-dev |
| 12:08 |
Taoki |
PilzAdam: Oh, something else to note. The sun / moon coming into view can also depend on the FOV setting. So the formula won't be the same for everyone |
| 12:09 |
Taoki |
We need an average logical bounding sphere |
| 12:11 |
Taoki |
AAAH! |
| 12:11 |
Taoki |
Never mind, even. What I'm trying won't work >.< |
| 12:11 |
Taoki |
Color blend is calculated entirely separate from color intensity. it wouldn't be possible to add a break at the middle. |
| 12:13 |
Taoki |
The code has two main calculations: Intensity of the sun / moon color over the normal colors (determined by daytime and influenced by pitch) and the blend between the sun and moon color in this offset color (calculated by yaw). |
| 12:13 |
Taoki |
Only way to add a break to the sides is to apply the offset to both. But that wouldn't keep them from mixing |
| 12:14 |
PilzAdam |
thats not good then |
| 12:14 |
Taoki |
Well, actually... I can do a hard snap between the colors past 0.5 yaw, since if the blend isn't applied there no one sees the lack of a transition |
| 12:14 |
Taoki |
I'll try to go with that |
| 12:15 |
|
RealBadAngel joined #minetest-dev |
| 12:15 |
RealBadAngel |
hi |
| 12:16 |
Taoki |
hi |
| 12:16 |
RealBadAngel |
i updated bumpmapping to current tree |
| 12:16 |
RealBadAngel |
https://github.com/minetest/minetest/pull/805 |
| 12:16 |
Taoki |
Ok, my head hurts... I'd need to remake part of the whole formula for the fog to fix this, and a lot of other new maths |
| 12:17 |
Taoki |
PilzAdam: Does the halfway blend really look that bad? With the default fov it shouldn't even be visible there's a transition. We could leave it like this for now |
| 12:17 |
PilzAdam |
yes, it looks bad |
| 12:19 |
Taoki |
I'll probably give up on doing anything any more then. I worked daily for 3 days on a code, for nothing. Because no matter what I do it's still bad |
| 12:19 |
RealBadAngel |
i had a little delay thx to kahrl, he removed atlas and i was using some atlas methods before |
| 12:20 |
RealBadAngel |
Taoki, whats the problem? |
| 12:20 |
Taoki |
RealBadAngel: I worked 3 whole days all of my free time to add properly colored fog. No matter what I do, it won't look good and it's not ok |
| 12:21 |
Taoki |
I need to re-make complex maths every few hours, and it takes me 1-2 hours to even figure out a few lines. |
| 12:21 |
RealBadAngel |
i saw effects of your work and i find it quite nice |
| 12:21 |
Taoki |
I can't take a break either, cuz if I do the code will either be forgotten or new features will cause a GIT conflict |
| 12:21 |
Taoki |
Whenever I contribute, this is the situation I end up in |
| 12:21 |
Taoki |
(not only in Minetest if that's any consolation) |
| 12:22 |
Calinou |
http://forum.minetest.net/viewtopic.php?pid=97878#p97878 |
| 12:22 |
Calinou |
I remember trying to interpolate the yaw changes, and epicly failed |
| 12:22 |
Taoki |
I fixed a dozen suggestions from many people for days. Now I need to re-make even more maths to fix another. |
| 12:25 |
Taoki |
Problem is that I think changing it would also look worse. In ints initial state, the transition was done around the middle. It looked ugly because the colors flashes too quickly while turning around small areas |
| 12:27 |
Taoki |
RealBadAngel, darkrose, celeron55, or other core devs: Anyone want to try my code and tell me if you agree with current functionality and appearance? And if you find the current coloring I used acceptable. Would like more votes on this |
| 12:28 |
|
proller joined #minetest-dev |
| 12:28 |
Taoki |
https://github.com/minetest/minetest/pull/799 |
| 12:29 |
Taoki |
I'm curious if you agree with the fog transition, or that I should do as PilzAdam suggested and apply no influence halfway. Which was what it worked like during my first test but I rememver and can say it looked worse |
| 12:30 |
RealBadAngel |
Taoki, i will test the current pull later today and let you know what i think about current status of it |
| 12:30 |
Taoki |
Thanks, I appreciate it. If other devs agree with the current looks it might be good |
| 12:31 |
Taoki |
Not that I want to disregard PilzAdam's opinion illy intended. But for one thing I can't remake the whole formula gain, and on the second hand the way he's suggesting was already tested by me and it looked worse. |
| 12:31 |
Taoki |
Worse because the colors would flash and change so quickly between two nouns as you turned around, it looked even more illogical |
| 12:31 |
celeron55 |
...are you seriously requesting me to test it? well okay |
| 12:32 |
Taoki |
celeron55: I understand it must be so bad it has to be a joke I'm asking you to bother :P |
| 12:32 |
celeron55 |
i'm going to test it and i am going to have an opinion on it |
| 12:33 |
celeron55 |
just beware |
| 12:33 |
Taoki |
ok |
| 12:34 |
RealBadAngel |
celeron55, you could take a look also at bumpmapping, i have pulled it |
| 12:34 |
Taoki |
If you'll disagree on too many lajor things, I'll give up on continuing this code and leave it to other devs to do it as it's considered it must be done. I can't make any more major changes at this point |
| 12:34 |
Taoki |
**major |
| 12:35 |
celeron55 |
RealBadAngel: does it work on a GMA 950? |
| 12:35 |
Taoki |
Not a ragequit BTW, but it means I can't do it how the team considers it must be done. That and after several days of working on Minetest, I can't allocate even more mental resources to re-making part of the code |
| 12:35 |
RealBadAngel |
celeron55, whats GMA 950? |
| 12:36 |
celeron55 |
an old integrated intel GPU |
| 12:36 |
RealBadAngel |
no way |
| 12:36 |
celeron55 |
no way then |
| 12:36 |
RealBadAngel |
its a modern GPU's thingy |
| 12:37 |
RealBadAngel |
it can be turned on/off depending on players hardware and choice |
| 12:38 |
RealBadAngel |
moreover its not for 16x texture packs |
| 12:38 |
RealBadAngel |
but 128x + |
| 12:38 |
celeron55 |
by the way, has somebody kept track of build times? |
| 12:38 |
PilzAdam |
Taoki, if you dont have fun when coding then you are wrong in open source development |
| 12:38 |
celeron55 |
i feel like building this taoki's pull request takes longer than sometime before |
| 12:39 |
PilzAdam |
celeron55, do you compile Lua mainmenu the first time? |
| 12:39 |
celeron55 |
yes but i take that into consideration |
| 12:39 |
Taoki |
PilzAdam: I have fun with some of the codes (and ith initial implementations). It's only not fun when you spend hours trying to figure small things out and it takes days to even do anything basic |
| 12:39 |
|
proller joined #minetest-dev |
| 12:40 |
Taoki |
celeron55: The code is very small and shouldn't increase compile time by more than a milisecond :) |
| 12:40 |
RealBadAngel |
celeron55, will you be against the option to use it too? |
| 12:40 |
Taoki |
Likely something else is increasing it |
| 12:40 |
celeron55 |
RealBadAngel: i take no position on that, unless it hinders something when being off |
| 12:41 |
RealBadAngel |
when off, nothing changes |
| 12:41 |
celeron55 |
i don't really understand why something like that would be used though, considering there are no dynamic lights |
| 12:41 |
RealBadAngel |
have you seen the screenshots? |
| 12:41 |
celeron55 |
because all could be just baked into textures anyway |
| 12:42 |
celeron55 |
of course |
| 12:42 |
RealBadAngel |
when lights will become reality, code will require almost no changes to work with them |
| 12:42 |
Taoki |
Liiiightsssss.... |
| 12:43 |
RealBadAngel |
im using now fake light source |
| 12:43 |
Taoki |
RealBadAngel: Can envmap and / or dynamic reflections, as well as refraction (for water) be done until we have HW lighting? |
| 12:43 |
RealBadAngel |
no |
| 12:44 |
Taoki |
ok |
| 12:44 |
RealBadAngel |
for this i need light sources |
| 12:44 |
Taoki |
Even envmap reflections? Those shouldn't require any lighting, just blending two textures and giving camera view coordinates to one |
| 12:44 |
RealBadAngel |
i tried to fake it, but results were not acceptable |
| 12:45 |
RealBadAngel |
sometimes it was lookin ok, sometimes it was a shit |
| 12:45 |
Taoki |
RealBadAngel: What's the progress on the hardware light shader? Any even partly working test up? |
| 12:46 |
RealBadAngel |
i need first to finish the projects i was workin on |
| 12:46 |
Taoki |
ok |
| 12:46 |
RealBadAngel |
bumpmapping is ready, major technic changes are partially ready |
| 12:46 |
RealBadAngel |
i want to raise technic onto another level |
| 12:47 |
celeron55 |
Taoki: i think it's good |
| 12:47 |
celeron55 |
as for what it does |
| 12:47 |
celeron55 |
i don't have time to look at the code though, somebody else must review it |
| 12:47 |
Taoki |
ok. Glad you agree about it |
| 12:48 |
Taoki |
The code is rather simple. Only thing that could be considered dirty are the formulas I had to use, but I guess those are needed in any major code |
| 12:53 |
celeron55 |
RealBadAngel: have you tried making faked light sources based on analyzing the environment? |
| 12:53 |
celeron55 |
eg. placing a couple to nearest torches and a cave opening if such is found with some fast algorithm |
| 12:55 |
RealBadAngel |
not yet |
| 12:55 |
RealBadAngel |
i just assumed we will go for irrlicht lights |
| 12:55 |
RealBadAngel |
is that an idea? |
| 12:55 |
celeron55 |
it's not going to happen anytime soon, don't rely on it |
| 12:56 |
RealBadAngel |
that could be possible |
| 12:56 |
RealBadAngel |
take sun pos (visible or not) and something like neighbours |
| 12:56 |
Taoki |
celeron55: RBA plans to make HW lights a reality, so it might happen soon |
| 12:57 |
Taoki |
But Irrlicht lights I doubt will work, RealBadAngel |
| 12:57 |
celeron55 |
irrlicht light = hw light |
| 12:57 |
Taoki |
We already suggested and discussed them, and there's no way to mask non-shader lighting |
| 12:57 |
Taoki |
celeron55: Yes. But if not done with our own shader, the masking might not work |
| 12:57 |
Taoki |
I shader it should be possible to cut lights off using param1 |
| 12:57 |
RealBadAngel |
i could make simplex lights faking that fits the current schema |
| 12:58 |
Taoki |
If Irrlicht lights can be made to do that, I'd prefer them too, so sure thing |
| 12:58 |
RealBadAngel |
if i will suceed with hw too that could give players option to choose from |
| 12:59 |
RealBadAngel |
with hw lights much more can be easily done |
| 12:59 |
Taoki |
RealBadAngel: Problem is that for any type of light, we must keep at least sunlight from shining indoor and in caves |
| 13:00 |
Taoki |
Preferably torches too from shining through walls |
| 13:00 |
RealBadAngel |
for example water reflections and surfaces are built-in irrlicht stuff |
| 13:00 |
RealBadAngel |
simple one liner |
| 13:01 |
Taoki |
RealBadAngel, celeron55: If you want an image of how I dream Minetest will look like someday (not that it matters what I think xD), THIS: http://www.youtube.com/watch?v=RM5PRtDrbW8 |
| 13:02 |
RealBadAngel |
i have all the shaders needed for such effects |
| 13:03 |
RealBadAngel |
and found out im able to make just bumpmapping without lights |
| 13:03 |
RealBadAngel |
but c55 gave me an idea actually |
| 13:03 |
Taoki |
OMG, that sounds like too good news to be true :P |
| 13:03 |
RealBadAngel |
and im gonna test it out |
| 13:04 |
Taoki |
If we add god's rays, my fog system might turn out useless too :P (not really, it's its own effect still) |
| 13:06 |
Taoki |
Oh... and there's another shader we can add for now BTW. Already tried it but my knowledge in shaders was still too bad |
| 13:07 |
Taoki |
We should be able to add a wavery deformation shader. To use for water and drawtype_plantlike |
| 13:07 |
Taoki |
So plants deform as if wind is beating and water is well... wavery |
| 13:09 |
Taoki |
BTW... shaders bring me back to something else, and I wonder: How possible would it be to use them to smoothen out map blocks. Yes, I'm talking about my idea to use something similar to marching cubes to (optionally) have smooth detailed terrain. Although I know that's not what Minetest is intended for |
| 13:09 |
RealBadAngel |
im thinkin about rebuilding shaders system |
| 13:09 |
RealBadAngel |
so one could be able to add effect |
| 13:09 |
Taoki |
http://forum.minetest.net/viewtopic.php?pid=28827 |
| 13:09 |
Taoki |
sounds good |
| 13:10 |
RealBadAngel |
which could be turned on/off |
| 13:10 |
RealBadAngel |
and one final shader |
| 13:11 |
Taoki |
http://www.volumesoffun.com/phpBB3/download/file.php?id=90&sid=6452d9930e0df1d13b5645c1c932640f This is the type of thing I'd be hoping to achieve with a "rounding" shader |
| 13:11 |
RealBadAngel |
so there wont be a need to copy the same code in all the shaders |
| 13:11 |
RealBadAngel |
but thats rather big project and for later |
| 13:12 |
Taoki |
RealBadAngel: Well, smoothing of node surfaces could be done with a shader. Just needs a subdivide call I think... but it needs to be done properly |
| 13:12 |
Taoki |
Not anything by far urgent of course, but still thinking :) |
| 13:14 |
Taoki |
Heh... even better idea: Don't smoothen out the whole terrain surface, just the topside. To still make them blocky but a pillow-like type of block design or something. But again, distant thoughts |
| 13:15 |
Taoki |
Need to learn shader functions and do this myself :) |
| 13:17 |
|
proller joined #minetest-dev |
| 13:17 |
RealBadAngel |
btw, http://www.youtube.com/watch?v=nNsN_konRbs |
| 13:18 |
RealBadAngel |
i suggest some1 start thinkin bout Oculus Rift |
| 13:18 |
Taoki |
http://fc03.deviantart.net/fs71/f/2012/099/7/e/minecraft__the_grass_by_lordpankake-d4vmvzv.png Something of this sort is what we could do with deformation vertex shaders :) |
| 13:19 |
Taoki |
RealBadAngel: http://th06.deviantart.net/fs70/200H/f/2011/347/1/6/minecraft_grass_block_full_3d_textured_model_by_xhazard78x-d4j192g.png This would look cool to do with your parallax mapping system |
| 13:19 |
RealBadAngel |
and it wont work on intel cards too ;) |
| 13:19 |
Taoki |
AHH, eyephones. I want some badly too :P |
| 13:19 |
Taoki |
side-by-side separation should be enough to support that tho |
| 13:20 |
RealBadAngel |
minecraft and terasology have already Oculus Rift support |
| 13:20 |
Taoki |
RealBadAngel: If ou have an intel card, you don't plan to render anything 3D except maybe glxgears |
| 13:20 |
Taoki |
nice |
| 13:20 |
RealBadAngel |
my gfx is 4 years old and capable of all this stuff |
| 13:21 |
Taoki |
http://fc00.deviantart.net/fs71/i/2013/146/8/0/minecraft___grass_block_by_lemmino-d66ooxy.png A node in minetest decades from now ;) |
| 13:21 |
RealBadAngel |
even amd thinks my card is legacy and not supported anymore |
| 13:22 |
RealBadAngel |
nowadays parts are so cheap that it shouldnt be any point |
| 13:23 |
RealBadAngel |
Taoki, lol |
| 13:23 |
Taoki |
Depends what cheap means to everyone. I'll need to put money aside for an year to upgrade my PC |
| 13:23 |
Taoki |
Well in my case I only get top-edge hardware |
| 13:24 |
Taoki |
RealBadAngel: That last link was partly a serious statement :P Think of a simple deformation shader + very detailed parallax map. Then a dense particle system |
| 13:24 |
RealBadAngel |
i just got better laptop than my box from my cell phone provider as a bonus for not quiting |
| 13:24 |
Taoki |
Really? |
| 13:24 |
Taoki |
Your cell phone company gave you a laptop as a gift? |
| 13:24 |
RealBadAngel |
i could get a new phone or tablet or a laptop |
| 13:24 |
RealBadAngel |
ive chosen laptop |
| 13:25 |
Taoki |
You live in one happy country, I'm glad for you :P |
| 13:25 |
RealBadAngel |
its not really a gift |
| 13:25 |
|
proller joined #minetest-dev |
| 13:25 |
Taoki |
still |
| 13:25 |
RealBadAngel |
they use it to make you sign for another year |
| 13:25 |
RealBadAngel |
but still |
| 13:26 |
Taoki |
Even so |
| 13:27 |
RealBadAngel |
i will have Acer Aspire V5 |
| 13:27 |
RealBadAngel |
quite nice toy |
| 13:27 |
Taoki |
RealBadAngel: Oh, and regarding a better (and automated) shader system. sfan5 has this pull request for some time, and I don't know why it's not in yet: https://github.com/minetest/minetest/pull/595 |
| 13:27 |
Taoki |
RealBadAngel: My laptoip is an Acer Aspire too :) |
| 13:28 |
Taoki |
celeron55, PilzAdam: Any reason why this pull is waiting? It would make shader development a lot easier by not requiring code changes |
| 13:28 |
RealBadAngel |
about sfan's pull, its definitely wrong |
| 13:28 |
RealBadAngel |
shaders are not a texture you can put somwhere or not |
| 13:28 |
PilzAdam |
Taoki, yes, it doesnt work |
| 13:28 |
Taoki |
Ahh, ok |
| 13:29 |
RealBadAngel |
code needs variables to be passed |
| 13:29 |
celeron55 |
Taoki: it's mostly useless because of what you know yourself too |
| 13:29 |
celeron55 |
just read your comment on it |
| 13:29 |
Taoki |
RealBadAngel: Why couldn't they work like a texture you can put somewhere? I know shaders require parameters, but base shaders wcould be auto-detected |
| 13:29 |
celeron55 |
practically *all* remotely useful shaders need special handling from C++ side |
| 13:29 |
RealBadAngel |
and thats a hardcoded connection between c++ and shader code |
| 13:30 |
celeron55 |
also there is just one shader per one material |
| 13:30 |
Taoki |
celeron55: I admit I'm not experienced with shaders. I know they take properties from code via uniform variables, and I assume those would need mentioning between the code and each shader |
| 13:30 |
celeron55 |
so it becomes even more useless |
| 13:30 |
* Taoki |
ponders shaders from Lua, just to be more evil :P |
| 13:31 |
RealBadAngel |
to make shaders plug n play we would need to make some standards |
| 13:31 |
PilzAdam |
I dont really understand why so many people care about graphics, its not the main point of Minetest at all |
| 13:31 |
RealBadAngel |
nah, pointless |
| 13:31 |
RealBadAngel |
shaders wont do anything on their own |
| 13:32 |
RealBadAngel |
thats game deciding what shaders should do |
| 13:32 |
PilzAdam |
inb4 shader API ;-) |
| 13:32 |
Taoki |
IIRC game needs to point out a mesh to the shader, and the shaders applies the effects |
| 13:32 |
celeron55 |
PilzAdam: because graphics is a thing one can do without actually having any nontrivial goal |
| 13:34 |
RealBadAngel |
shaders on their own are not able to decide whats water, where it is etc |
| 13:34 |
RealBadAngel |
with such plug in one you can make shader that turns everything grayscale for example |
| 13:34 |
celeron55 |
PilzAdam: altough they will be wasted if _nobody_ has a nontrivial goal 8) |
| 13:34 |
Taoki |
That's true, the game needs to tell them what surface types to act one |
| 13:35 |
Taoki |
Well, for 3D shaders at least. 2D processing shaders can decide on their own |
| 13:35 |
Taoki |
Which BTW we don't have yet and need to :P |
| 13:35 |
Taoki |
Those are fot things like bloom, motion blur, etc. which don't need 3d info, and can just post-process each frame before rendering it |
| 13:35 |
Taoki |
**fog |
| 13:36 |
RealBadAngel |
my long term goal is to allow different shaders work on the surface and gather all the effects in one final shader |
| 13:36 |
Taoki |
RealBadAngel: I know I'm getting overly excited. But what about depth of field? Does that require lighting too to calculate properly? |
| 13:37 |
RealBadAngel |
so game shall pass surface type, light data to shaders code |
| 13:37 |
RealBadAngel |
and let all the shaders work on it |
| 13:37 |
|
proller_ joined #minetest-dev |
| 13:37 |
RealBadAngel |
so no switching between shaders and copy pasta between them |
| 13:38 |
RealBadAngel |
Taoki, im not sure about DOF |
| 13:39 |
Taoki |
Why not? |
| 13:39 |
Taoki |
Actually, never mind. DOF is also a 2D shader. Though it uses 3D data |
| 13:39 |
Taoki |
Since 3D shaders affect the mapped texture and geometry itself. FOF iss still post-processing done to the rendered frame |
| 13:39 |
RealBadAngel |
i was workin on god rays, reflections, refraction, bumpmapping, occlusion and this cartoon effect shader |
| 13:40 |
Taoki |
nice |
| 13:40 |
Taoki |
God rays are the ones that would take the cake after HW lighting ^^ |
| 13:41 |
RealBadAngel |
but now i really have to go, i have a date :) |
| 13:41 |
RealBadAngel |
c ya |
| 13:41 |
Taoki |
lok, later |
| 13:41 |
Taoki |
*ok |
| 13:42 |
PilzAdam |
Taoki, btw, Terasology has volumentric fog |
| 13:42 |
Taoki |
Haven't checked it out in some time, nice |
| 13:47 |
Taoki |
We could use that one for water |
| 13:56 |
celeron55 |
i really wonder why people don't just go for terasology or something instead of MT |
| 13:56 |
celeron55 |
as the only thing they seem to care about is graphics |
| 13:56 |
celeron55 |
it just doesn't make any sense |
| 13:57 |
Taoki |
It's not the only thing. But graphics matter a lot too |
| 13:57 |
Taoki |
The real art is having great graphics with a great system and functionality. But both are important... though graphics are indeed secondary |
| 13:58 |
Taoki |
If you focus ONLY on graphics, you end up with something like Unreal Tournament 3: Fantastic visuals, c**p gameplay and barely anything fun to do |
| 13:59 |
Taoki |
If you focus only on gameplay, you have the primary purpose properly done. But it feels empty and wrong without great visuals after some time (namely when you know they can be improved) |
| 14:00 |
Taoki |
celeron55: As for why I like Minetest over terasology, mainly two reasons. One is that TS is still in java, and java sucks for games IMO. Second and most important one, it has no modding API like we have Lua |
| 14:00 |
Taoki |
And of course I've been with Minetest for a while, I wouldn't want to leave and get started with another engine |
| 14:00 |
celeron55 |
ut3 has "fantastic visuals"? lol |
| 14:00 |
Taoki |
MT is also simple and flexible, yet able to do a lot if you want it to |
| 14:01 |
celeron55 |
laughably generic and boring they are |
| 14:01 |
Taoki |
celeron55: The graphics in UT3 are pretty good. But only that |
| 14:01 |
Taoki |
ok I did find them impressive at the time at least :P |
| 14:03 |
celeron55 |
i'd hope that all people who have interest in gameplay would develop gameplay |
| 14:04 |
Taoki |
Well, "gameplay" is mostly developed now I'd think. In the sense that you can define and use items like you wish. So I'm most interested in other mechanics that go with it now |
| 14:04 |
Taoki |
Only additions I'd see possible to what could be called the purpose of MT is more abilities to how items work. But the gameplay is mostly done in Lua |
| 14:05 |
Taoki |
So adding to the gameplay would mostly mean adding new items. Which is easy to do and many mods already have (but never got upstream) |
| 14:05 |
Taoki |
The "paintings" mod for example has been ready for I think almost 2 years. Flowers mod took very long to get in too |
| 14:07 |
Taoki |
There's also already a workbench mod, and we still have no workbench and the 3 x 3 crafting grid in inventory (because, le sigh, that would make us too much like Minecraft) |
| 14:09 |
celeron55 |
in my viewpoint, gameplay *isn't* developed |
| 14:10 |
celeron55 |
there's not much depth in even the best game attempts currently |
| 14:10 |
Taoki |
I don't even know how it can be further developed and the wanted direction. Only thing I do know is that we need a lot more items, decorations, things that spawn naturally, etc... that without doubt |
| 14:11 |
Taoki |
If you mean that then yeah |
| 14:11 |
celeron55 |
i also know the community has developed to have people who don't even expect depth by now |
| 14:12 |
Taoki |
Funy thing is I heard opposite views too. That instead MT should be kept relatively simple. Though I don't think that |
| 14:12 |
Taoki |
At the same time, I think ignoring mods that could be integrated officially is a big cause |
| 14:12 |
celeron55 |
depth and simplicity aren't opposites |
| 14:12 |
Taoki |
Example: Why isn't mesecons in yet? Or the paintings mod... |
| 14:12 |
celeron55 |
that's not depth |
| 14:12 |
Taoki |
Also PilzAdam's simple mob framework, though I can see where that needs more work |
| 14:12 |
Taoki |
ok |
| 14:13 |
Taoki |
I do think that's adding to the environment and things you can do still |
| 14:13 |
celeron55 |
i acknowledge this wishing could be way off of any development reality by now though |
| 14:14 |
celeron55 |
well, mobs for sure are a key thing in blocking depth |
| 14:14 |
celeron55 |
gameplay depth, to be clear |
| 14:15 |
Taoki |
One major difference I noticed between Minecraft and Minetest is that in MC, you are more intrigued to play because you have more things to do. In MT it's harder to really feel like you can have the same building a house, farming, etc. It just doesn't feel... ready in a sense. |
| 14:15 |
Taoki |
It doesn't feel the same to put work into a house on an online server in Minetest same as on MC. Something is missing |
| 14:15 |
celeron55 |
that feel could be quite close to the feel of missing depth |
| 14:15 |
Taoki |
yeah |
| 14:16 |
Taoki |
It's not just the lack of items though. MT still lacks a lot of immersion. I think the "fancy visuals" discussion contributes a bit to this too |
| 14:16 |
celeron55 |
...and it isn't fixed by modifying sunsets or adding shaders |
| 14:16 |
Taoki |
Personally at least, I find a combination between lack of items and activities, and the overall feel isn't quite as... pulling and captivating yet |
| 14:17 |
Taoki |
celeron55: That's a small part that can contribute too |
| 14:17 |
Taoki |
My sunset fog for instance fixes one thing I hated; The dull and dead visuals at sunrise / sunset. It adds more life and I can tell the difference |
| 14:17 |
Taoki |
A small step, but the difference can be felt |
| 14:18 |
Taoki |
More sounds (including environment ones) would help too. Though a proper ambience mod doesn't exist yet |
| 14:18 |
Taoki |
(I know about the "immersive sounds" mod, but I have reasons to say proper) |
| 14:18 |
Anchakor |
MT isn't a game now, it's purely sandbox atm |
| 14:19 |
Taoki |
Anchakor: Sort of. I'd say it does feel slightly like a "3D voxel editor with realistic physics and logics" |
| 14:21 |
Taoki |
Sadly, the word "a voxel editor" qualifies it most at this stage, IMHO. But don't worry much about that... I can also tell it's deemed to become one of the best open-source virtual space engines in time |
| 14:21 |
Taoki |
Or "world engines" if that's better |
| 14:22 |
Taoki |
It's very easy to add things, and a lot of people want to add things. It will get a lot further I can tell |
| 14:22 |
celeron55 |
gameplay depth is about something like progressing interaction |
| 14:23 |
celeron55 |
you don't get that from people adding things they want from the same standpoint |
| 14:23 |
Anchakor |
I can say that sandbox isn't attractive to me |
| 14:23 |
Taoki |
That too. For me it's having enough items and activities to do. A lot of complex things to go about |
| 14:24 |
Taoki |
Many biomes, many tools, many devices and systems (technic mod helps a lot with that) |
| 14:24 |
Taoki |
And of course many mobs... both villagers, animals, or enemy mobs |
| 14:24 |
celeron55 |
that's called variety, not depth |
| 14:25 |
Anchakor |
when I first played MC I played it because there were monsters, I could explore and discover new things and places and I could build crazy fast minecart rides :) |
| 14:25 |
Taoki |
Having a lot of things to do creates depth though. It's not the only factor still |
| 14:26 |
|
punsit joined #minetest-dev |
| 14:26 |
Anchakor |
I think simple addition which adds depth is mobs which give you reason to build things to properly defend yourself and your things |
| 14:26 |
Taoki |
Anchakor: I got into MC because it was booming over the internet. Unfortunately, I started out as a hater... because I didn't understand why everyone was so exciting AND paying money over a game in jave with 8 bit graphics |
| 14:26 |
Taoki |
But when I understood I was retarded I fixed it |
| 14:26 |
Taoki |
**java |
| 14:27 |
Taoki |
By that time I didn't understand the wonders of what you could do with voxels yet. |
| 14:27 |
Taoki |
Think this was about 3-4 years ago |
| 14:27 |
Anchakor |
I think even sandbox games need gameplay mechanics which gives them reason to build and do things |
| 14:27 |
celeron55 |
anyway, gameplay depth is the thing where game development actually is, and what is hard about game development |
| 14:28 |
celeron55 |
everything else is just engines, graphics and whatever items |
| 14:29 |
Taoki |
Idea is that MC (and I think that's the aim for MT too) aren't just sandbox games. I could say they also have a bit of RP, but the correct term is it's more of a worlds simuillator where you can do things |
| 14:30 |
Taoki |
Hence why frequently, I'm tempted to not call MC / Mt games at all. More like virtual spaces |
| 14:31 |
Taoki |
Games are usually those where you have an objective to accomplish (cross the finish line first, frag as many people as possible, etc). This is more of a worlds simulation system where you can invent a lot of things |
| 14:32 |
Taoki |
People who've been into Second Life as well can easily tell what I mean by virtual space where you can do things |
| 14:33 |
celeron55 |
it's not as different as you think, and it's separate from mechanics |
| 14:33 |
celeron55 |
when there are gameplay mechanics in place in a sandbox world, the players can easily imagine "finnish lines" by themselves |
| 14:34 |
celeron55 |
when there aren't a lot of mechanics (as in current MT), players can't really do that |
| 14:34 |
Taoki |
Yeah. But if they make games and contests within such an engine, the engine itself isn't a game engine... just that construction in it |
| 14:34 |
Taoki |
yeah |
| 14:35 |
Taoki |
And sadly I have to go now. BBL in 1-2 hours or so. Hopefully can finish the fog thing with RBA then tale a look at his bumpmaps |
| 14:35 |
celeron55 |
monsters are an easy and well working example of a mechanic which allows setting many kinds of objectives |
| 14:35 |
Taoki |
yeah |
| 14:35 |
celeron55 |
just one among many, but anyway |
| 14:37 |
celeron55 |
and as for monsters, depth you start getting by eg. requiring some drops from monsters in a crafting recipe, and more as you can't kill that monster without something, and more with eg. having more difficult monsters deeper, or having some tricks to defeat them, or pretty much anything that creates "progress" like that |
| 14:38 |
Anchakor |
Taoki: good open world RPGs are also not games according to your definition then |
| 14:41 |
celeron55 |
also, actually i can't even think of anything as good as monsters - it's just such an inherently obvious thing for poking up interaction from players |
| 14:41 |
celeron55 |
but for example mining could be used to achieve gameplay depth |
| 14:42 |
celeron55 |
being able to succesfully design and implement stuff like that is the expertise of a game developer |
| 14:48 |
sokomine |
building complex buildings, enjoying when it's done and perhaps doing contests (we don't really have those on mt servers) is worth something. beeing able to change the world - not having to go with a pre-generated world. that's where mt and mc are both great at - and that is valuable. in a way, right now the game is mostly creative and less of a game as such. but is that bad? i don't think so |
| 14:50 |
celeron55 |
i think it *is* bad, but not in a sense that it shouldn't be configurable to work like currently |
| 14:50 |
celeron55 |
it leaves so many people out now because it doesn't have anything to offer for them while it could |
| 14:51 |
sokomine |
admittedly, more and more young players join the game, and only a small fraction really wants to excell in creative building. gathering materials and trading them adds a bit of fun. some of those players may need more entertainment, and there mobs come in. i heard that on a mc server, players tend to hunt mobs more these days than participate in building contests |
| 14:51 |
celeron55 |
trading is a lot more fun if you trade stuff to people who need it for fighting monsters 8) |
| 14:51 |
celeron55 |
or for mining, or any other objective |
| 14:53 |
sokomine |
hm. i've seen players play hide and seek on a minecraft classic server. and similar things. but does that really add depth? to me, it only adds more entertainment (which is not bad either) |
| 14:56 |
sokomine |
mobf is a very intresting factor in that regard because it could eventually allow both - "decorative" farm animals, traders and inhabitants for your shops and houses, and monsters for those who like hunting them. even complex rpg stories could be done in theory. unfortionately, mobf is just at the beginning. simple mobs hasn't even started in that direction (it just aims at offering mobs that roam around and are thre to be killed |
| 14:58 |
celeron55 |
well, one problem is also that the current "ActiveObject" system of MT isn't particularly good |
| 14:58 |
celeron55 |
or, actually i don't even know |
| 14:58 |
|
Jordach joined #minetest-dev |
| 14:58 |
sokomine |
and in that context, i'm thinking of the mobs as another creative element - as mobs which you can place to extend the decoration/make the world feel filled with life - without personally caring about the shooter-aspect |
| 14:58 |
celeron55 |
if following certain design principles, it actually should work |
| 14:59 |
celeron55 |
but it really isn't a piece of cake |
| 14:59 |
celeron55 |
the largest problem is that it's buggy :P |
| 15:00 |
sokomine |
it helped a lot when entity-entity-collusion was added. that mobs wandered through each other was quite scary. and the 2d shapes for players where also too strange |
| 15:00 |
sokomine |
yes :-( hope it will improve :-) |
| 15:02 |
sokomine |
improvements in graphics are also very welcome. beautiful sunsets are likely to make builders happy |
| 15:03 |
sokomine |
in a way you can say that the mt gameplay is extended by *modding* - not only can you dig/place blocks in the world, you can also change quite easily how those blocks behave. it may sound strange, but that's one of the strength of the game |
| 15:06 |
sokomine |
in another sense than modding in other games works like. here, it can be considered part of the gameplay |
| 15:07 |
sokomine |
it extends what the game is about with diffrent means |
| 15:09 |
celeron55 |
yes it does; it's probably very odd to outsiders though |
| 15:11 |
celeron55 |
but really, it should be just a happy side benefit of having something otherwise good |
| 15:12 |
sokomine |
for most other game types it would be like this. but here, it really is part of it. just see how many people want to write mods - even if it doesn't do much and just adds a few blocks. it doesn't mean that they consider the world incomplete - it's just a way of changing it. like putting a building on an otherwise natural landscape |
| 15:13 |
|
twoelk joined #minetest-dev |
| 15:16 |
twoelk |
hi, this topic finally got me out of lurking |
| 15:16 |
sfan5 |
..hello |
| 15:17 |
twoelk |
I have some mc maps and I always play peacefull |
| 15:18 |
twoelk |
I think building in itself is a great play subject |
| 15:18 |
twoelk |
maybe some sort of questsystem might attract gameplay designers |
| 15:21 |
celeron55 |
hmm, i don't really think so |
| 15:21 |
celeron55 |
i mean, of course some people would try it out and make something |
| 15:22 |
celeron55 |
but the open world isn't really optimal at all for premade maps and quests |
| 15:22 |
|
hmmmm joined #minetest-dev |
| 15:25 |
twoelk |
maybe not when starting the game as such |
| 15:26 |
twoelk |
but I have a nephew who as an example enjoys enrichening mc maps with quests |
| 15:26 |
twoelk |
and then sharing these |
| 15:29 |
twoelk |
if you don't have some static goal like a drogaon on a pillar, all you have is searching for some rare item or difficult build |
| 15:30 |
twoelk |
*dragon |
| 15:30 |
twoelk |
or maybe advancing through a level system |
| 15:33 |
celeron55 |
i don't really like level systems, at least if they are plainly visible |
| 15:34 |
celeron55 |
and if they are like the ultimate goal |
| 15:34 |
twoelk |
I was rather thinking of an achievement levelsystem for the player |
| 15:34 |
celeron55 |
how does that differ from what you tought i understood? |
| 15:35 |
celeron55 |
because i see no difference :P |
| 15:35 |
twoelk |
I thought you understood different realms and such |
| 15:35 |
celeron55 |
no no |
| 15:37 |
celeron55 |
as an acceptable progression system i could see for example the vertical expansion of the player-controlled environment (by making players overcome all kinds of challenges getting further down and up) |
| 15:37 |
twoelk |
well the good thing of an open system is you can change the goal |
| 15:38 |
celeron55 |
or, actually, my original vision still stands as that |
| 15:38 |
proller_ |
difficulty = distance_from_0,0,0 |
| 15:38 |
twoelk |
what I like about mt is you can use it as fps or creative or technic simulation or whatever |
| 15:38 |
proller_ |
by x,z too |
| 15:39 |
proller_ |
stronger monsters, etc |
| 15:40 |
twoelk |
actually I'm personally totally uninterested in fictional monsters, sorry |
| 15:40 |
celeron55 |
proller_: i feel the only thing preventing it from happening is people lacking some kind of an initial motivation boost |
| 15:41 |
twoelk |
I'd rather have wolfs, lions mamoths ... |
| 15:42 |
celeron55 |
twoelk: from a gameplay perspective, all the same; it doesn't really mattter in this context what they're called and look like |
| 15:42 |
celeron55 |
-t |
| 15:42 |
twoelk |
uh? |
| 15:43 |
twoelk |
sorry, new to irc, what does all black mean? |
| 15:43 |
celeron55 |
no idea what you're talking about 8) |
| 15:44 |
celeron55 |
did your client do something funny because i mentioned your name? |
| 15:44 |
twoelk |
might be |
| 15:45 |
twoelk |
cant read your name |
| 15:45 |
twoelk |
its all black |
| 15:45 |
celeron55 |
anyway, i said "from a gameplay perspective, all the same; it doesn't really matter in this context what they're called and look like" |
| 15:46 |
twoelk |
the behavior is different |
| 15:46 |
twoelk |
but that doesn't matter , thats personal taste |
| 15:46 |
celeron55 |
well, it'll no doubt be moddable to anyone's taste once the expertise of making good mobs exists in the first place |
| 15:47 |
celeron55 |
and systems |
| 15:48 |
celeron55 |
really the question is: is there even any other viable alternative than starting to develop mobs |
| 15:48 |
twoelk |
don't think that will add depth though just more excitement/fun |
| 15:49 |
twoelk |
some sort of static message system similar to signs but on book scale might be of value |
| 15:49 |
celeron55 |
i feel like just continuing to provide a platform for people to build houses kind of wastes much of the available developer work |
| 15:49 |
celeron55 |
build houses in* |
| 15:50 |
twoelk |
like go to some place read the riddle and solve the mystery |
| 15:50 |
celeron55 |
what'd the solving involve? |
| 15:51 |
twoelk |
donu, build a bridge accros lava withy no fly to get to some other hint? |
| 15:52 |
twoelk |
oops brebuild map |
| 15:52 |
celeron55 |
sounds boring if there aren't any monsters or anything to bother you |
| 15:52 |
proller_ |
celeron55, in rpg motivation is to find more powerful items, without levels system we can make more frequent diamonds-mese-.. in hard monsters areas |
| 15:53 |
twoelk |
not excluding that ;-) |
| 15:54 |
proller_ |
or kill rare strong monster.. |
| 15:55 |
twoelk |
so you want some random gameplay gain by generating a random map with a random danger to suit every random player? |
| 15:56 |
twoelk |
to get some random game depth? |
| 15:56 |
proller_ |
yes! |
| 15:57 |
twoelk |
we need some random genius then |
| 15:57 |
proller_ |
but low-level in spawn area (100-200 radius from spawn) |
| 15:59 |
proller_ |
and volatile world around player with seasons, weather, hard-to survive hot-cold areas, volcano biomes. ........ |
| 16:00 |
proller_ |
it will be the best game |
| 16:00 |
proller_ |
but after 1-3 years of develop |
| 16:01 |
twoelk |
I like building stuff and as such enjoy the graphics (+for seasons) but when in a fighting game have no time to gaze at the surroundings |
| 16:01 |
twoelk |
I would minetest to be best platform for both |
| 16:01 |
twoelk |
I would like ..... |
| 16:02 |
|
dexter0 joined #minetest-dev |
| 16:02 |
proller_ |
disable damage and build unlimited |
| 16:02 |
twoelk |
maybe switchable ingame |
| 16:02 |
twoelk |
I do that in singleplayer |
| 16:02 |
celeron55 |
twoelk: in the tradition of minecraft, we will always support some kind of a creative/peaceful mode |
| 16:02 |
proller_ |
twoelk, its smallest problem |
| 16:02 |
celeron55 |
we _know_ there are people who like such |
| 16:03 |
twoelk |
I said both didn't I? |
| 16:03 |
twoelk |
I got nephews to please with the games I suggest |
| 16:04 |
|
Calinou joined #minetest-dev |
| 16:04 |
twoelk |
one is for guns and tnt, one is for roleplay and the girl for building |
| 16:09 |
celeron55 |
do people think it's a good idea to start a channel for having people who support such a thing join and discuss there until something more exact is agreed on? |
| 16:09 |
celeron55 |
really there are two issues: can something be agreed on, and are there people who actually are willing to work towards whatever is agreed |
| 16:11 |
celeron55 |
it could be done here too but i feel some people don't like the publicity of this channel for letting their thoughts flow freely |
| 16:11 |
celeron55 |
(and the officialness) |
| 16:12 |
|
proller_ joined #minetest-dev |
| 16:12 |
Anchakor |
people who are willing to work on such a game mode and are willing to collaborate can get people and start such a channel themselves |
| 16:13 |
celeron55 |
it's not a matter of starting a subgame, as it requires open collaboration from the engine side |
| 16:13 |
Anchakor |
well I mean that it wouldn't be bad if 2 or more groups with different gameplay visions emerged |
| 16:16 |
celeron55 |
...well that depends |
| 16:16 |
celeron55 |
in any case, i'm open to the idea of having deeper gameplay in the spirit of what was talked about and am on the side of whoever who dares to attempt such |
| 16:20 |
celeron55 |
Anchakor: as far as i understand, you too would prefer such development to start? |
| 16:20 |
twoelk |
programming and gamedesigning skills often don't live on the same planet from what I have seen and played |
| 16:22 |
celeron55 |
yeah 8) |
| 16:22 |
celeron55 |
you can learn both by hard work though |
| 16:23 |
celeron55 |
(as long as you're able to see your weaknesses) |
| 16:24 |
Anchakor |
yes |
| 16:25 |
celeron55 |
MT doesn't have a good game design background at all |
| 16:25 |
Anchakor |
I studied quite game design in my free time quite a lot, and I code C++ for money, but unfortunately I am quite burnt out as a coder |
| 16:25 |
Anchakor |
on the other hand I enjoy learning and doing game design |
| 16:26 |
twoelk |
a game designer might sketch possible directions and the coder will have to stake out what is possible |
| 16:26 |
twoelk |
and that should be published somewhere for all to read |
| 16:27 |
twoelk |
it might stop peoble from requesting things that are difficult to achieve |
| 16:27 |
celeron55 |
how much are you involved with gamedev stuff? |
| 16:27 |
twoelk |
who me? |
| 16:28 |
celeron55 |
yes |
| 16:28 |
celeron55 |
i can't say your name because apparently you couldn't see what i write then 8) |
| 16:28 |
twoelk |
first time on this channel |
| 16:28 |
celeron55 |
i'm a hobbyist game developer these days and read and follow and do gamedev stuff a lot, but i don't really focus my efforts in terms of that to MT |
| 16:29 |
twoelk |
I used to design obscure boardgames long time ago |
| 16:29 |
twoelk |
none ever got anywhere |
| 16:30 |
Anchakor |
I did boardgame design too :) |
| 16:30 |
celeron55 |
MT is so disconnected from real game development that it's hard to get anything relevant to that through here |
| 16:30 |
celeron55 |
i mostly react to that by being passive about it and sometimes rant here like now |
| 16:30 |
Calinou |
real game development = DRM development, graphics, "how to make stuff more realistic" |
| 16:30 |
Calinou |
that's all :P |
| 16:31 |
celeron55 |
Calinou: that's called shit |
| 16:31 |
twoelk |
with computers I am more on the building and cad side |
| 16:31 |
Calinou |
celeron55: if by real you mean AAA |
| 16:31 |
celeron55 |
AAA is shit development |
| 16:31 |
celeron55 |
it's a different thing! |
| 16:31 |
twoelk |
that is building with virtuall lego for example |
| 16:31 |
Anchakor |
are people opposed to a resource based combat system in MT? |
| 16:31 |
Calinou |
well, +1 |
| 16:31 |
Jordach |
how about minetest works more like Age of Empires |
| 16:32 |
Jordach |
a really good RTS never hurts |
| 16:32 |
Anchakor |
-_- |
| 16:32 |
celeron55 |
lol, MT isn't going to be an RTS by any stretch |
| 16:32 |
PilzAdam |
minetest should be more like <insert random game> |
| 16:32 |
Calinou |
...minecraft! |
| 16:32 |
celeron55 |
Anchakor: what'd that mean in practice? |
| 16:32 |
Anchakor |
minetest should be more like tetris! |
| 16:33 |
twoelk |
oh I was hoping for some 3d "die Siedler" |
| 16:33 |
twoelk |
among other directions |
| 16:33 |
Anchakor |
celeron55: system based on different kinds of attacks, buff and debuffs, that kind of thing |
| 16:34 |
celeron55 |
Anchakor: you know, you've suggested that before too, yes? |
| 16:35 |
Anchakor |
it's possible, I don't know how much of my design of a procedurally generated action RPG game have I spilled here |
| 16:35 |
celeron55 |
in any case, i think anything is better than what MC and MT have currently |
| 16:36 |
twoelk |
wouldn't be here if that was so |
| 16:36 |
celeron55 |
have you even used the combat system, once? 8) |
| 16:37 |
twoelk |
oops not interested |
| 16:37 |
celeron55 |
Anchakor: seriously, my only concern here overall is that nobody will actually do anything |
| 16:37 |
celeron55 |
...and that some people won't understand this "game thing" at all, but i think they'll learn to understand |
| 16:39 |
twoelk |
maybe you should explain somewhere "big" with lots of examples |
| 16:40 |
celeron55 |
seriously, if there is anything that keeps someone from experimenting with anything like that, shout it here like it was the most important thing in the world because it is |
| 16:40 |
Anchakor |
celeron55: yeah, well even small steps are good, just get some monsters, half-decent combat system and everyone understands survival :) |
| 16:41 |
celeron55 |
also yes, i feel bad for not just doing it my self |
| 16:41 |
celeron55 |
myself* |
| 16:41 |
twoelk |
is there some ai engine one could plug in? |
| 16:42 |
PilzAdam |
twoelk, of course there isnt, otherwise we would have proper mob mods |
| 16:42 |
twoelk |
ah Pilzadams name I can read he's all blue |
| 16:43 |
celeron55 |
he just said your name with some stuff, can you read it? |
| 16:44 |
celeron55 |
(you really should fix your client if not) |
| 16:44 |
twoelk |
I can read the stuff but his name is difficult |
| 16:45 |
twoelk |
sure got lots to learn like tweaking this programm I'm using for the first time |
| 16:45 |
PilzAdam |
what client do you use? |
| 16:45 |
twoelk |
KVIrc portable |
| 16:48 |
twoelk |
ok undone name highlighting if thats the issue |
| 16:49 |
|
sapier joined #minetest-dev |
| 16:52 |
sapier |
I'm not sure what pilzadam built but last time I tested on win32 it did show correct menu ... so I need more information |
| 16:52 |
PilzAdam |
it seems to be only broken in wine |
| 16:52 |
sapier |
ok that's an explanation |
| 16:53 |
PilzAdam |
and before someone asks, we support wine |
| 16:53 |
sapier |
ok then drop formspec I don't even have a clue where to look for it |
| 16:55 |
sapier |
supporting wine for a open source app buildable on linux ... whoever decided this must have been drinking |
| 16:55 |
|
serengeor joined #minetest-dev |
| 16:57 |
twoelk |
btw outdated info: the 0.4.7 install version for windows seems to behave differently to the runinplace zip, being much slower on my system |
| 16:59 |
sapier |
kahrl is there a reason why you call a function parameter m_device? |
| 16:59 |
hmmmm |
sapierrrr |
| 17:00 |
hmmmm |
make some sort of lua error on init, it'll start the last world immediately. then press escape, instacrash |
| 17:00 |
sapier |
there shouldn't be a lua error in mainmenu ;-) |
| 17:00 |
hmmmm |
GUIFormSpecMenu::OnEvent is still getting events after it errored out |
| 17:01 |
hmmmm |
i looked and you never clean up any m_menu things in the dtor |
| 17:01 |
sapier |
yes but the only way of catching a lua error is closing application either so it's not quite different ;-) |
| 17:01 |
sapier |
its the mainmenu if there's no working mainmenu nothing can be done ;-) |
| 17:01 |
hmmmm |
see, i would've fixed it or come up with some other behavior that doesn't involve instantly crashing |
| 17:02 |
hmmmm |
but i don't really understand how everything fits together |
| 17:02 |
sapier |
ok I'll fix it to crash in a more controled way :-) |
| 17:02 |
hmmmm |
like, what's a formspecsource, when do you delete it if ever, and how about m_menu |
| 17:02 |
hmmmm |
if you do delete m_menu, guienv crashes when it tries to draw that in draw_load_screen |
| 17:02 |
hmmmm |
or something along those lines |
| 17:04 |
hmmmm |
well those are things i saw when i was trying to fix some other things that are real problems, like you used to be able to press enter in the multiplayer tab and it'd press connect |
| 17:04 |
sapier |
not sure what you mean? |
| 17:04 |
hmmmm |
so how do you detect enter button presses in the button handlers? |
| 17:04 |
hmmmm |
(or if you really can't, how do you force a button element to have focus) |
| 17:06 |
sapier |
I guess enter button could be done similar to escape key but I need to have a look |
| 17:07 |
hmmmm |
and who comes up with the event.typ codes? |
| 17:07 |
sapier |
I'll fixup the todo list points first some are minor |
| 17:07 |
hmmmm |
looked through all files for "DCL" or whatever it was, nothing |
| 17:08 |
hmmmm |
also, pilzadam is able to consistently get the listbox append bug |
| 17:08 |
sapier |
it's mapped in lua |
| 17:09 |
hmmmm |
ah okay, i figured it was in guiEngine.cpp because you had all those similar 3-letter color codes |
| 17:09 |
sapier |
stop no DCL is set inm guiFormSpecMenu.cpp line 2032 |
| 17:09 |
hmmmm |
okay |
| 17:09 |
sapier |
evaluation of listbox events |
| 17:09 |
hmmmm |
by the way, you *can* use if-elseif-else |
| 17:09 |
sapier |
problem is a listbox is a somehow special button as it can be clicked different ways |
| 17:10 |
sapier |
in lua ? really? |
| 17:10 |
PilzAdam |
of course you can |
| 17:10 |
hmmmm |
well in lua, but i was referring mostly to in C++ :) |
| 17:10 |
PilzAdam |
elseif is just one word, not two |
| 17:11 |
sapier |
what's that todo with the delete button? |
| 17:11 |
sapier |
ahh server tab |
| 17:11 |
hmmmm |
if (string == "some constant") { return SVideo::Color(255, blah blah); } else if (string == "some other constant") { ... |
| 17:12 |
sapier |
I'll change it |
| 17:12 |
hmmmm |
eh. |
| 17:12 |
hmmmm |
not really necessary now, i'm just saying |
| 17:13 |
hmmmm |
lol. if it were me coding that, what i'd probably do would be to read the color code as an integer and use a switch statement on it |
| 17:14 |
hmmmm |
speaking of which, i was reading the logs and it seems like someone here doesn't like multi-character constants |
| 17:14 |
hmmmm |
i should add -Wno-multichar to shut up the schematic file signature checking (and hopefully complaints as well) |
| 17:14 |
sapier |
I didn't code it as int because I cant remember 7 beeing red and 3 beeing blue ;-) |
| 17:16 |
hmmmm |
no, i mean like color = readU32(color.c_str()); switch (color) { case 'RED\0': return blah; case 'YLW\0': ... |
| 17:16 |
|
BrandonReese joined #minetest-dev |
| 17:17 |
sapier |
that wouldn't work at all |
| 17:17 |
sapier |
switch case only supports integers but not strings |
| 17:17 |
hmmmm |
they are integers |
| 17:17 |
sapier |
'RED\0' isn't an integer |
| 17:17 |
hmmmm |
why not |
| 17:18 |
sapier |
hmm 4 bytes .. ok ok ... it could be an integer yes |
| 17:18 |
sapier |
but only on 32bit architecture |
| 17:18 |
hmmmm |
i was thinking about this the other day |
| 17:18 |
hmmmm |
basically, minetest won't run at all on anything that uses a 16 bit int |
| 17:18 |
sapier |
in any case it's a really really dirty hack :-) |
| 17:19 |
hmmmm |
if only because it'd lack the performance necessary |
| 17:19 |
hmmmm |
no doubt we'd have a lot of integer overflows as well |
| 17:19 |
hmmmm |
so a 32 bit integer is safe to assume |
| 17:19 |
sapier |
yes it's not that performant true but we're in menu .... user is way more slow |
| 17:20 |
hmmmm |
what's the point of a menu if you can't play the game? |
| 17:20 |
sapier |
I guess the listbox concatenation is a wine issue too? |
| 17:20 |
sapier |
while playing game no menu is shown ;-) |
| 17:20 |
sapier |
and it's no difference if a menu takes 1 or 2 ms to show up |
| 17:21 |
hmmmm |
but in regards to a color switch statement, i'd do that purely for the syntactical benefits |
| 17:21 |
hmmmm |
you get to use a switch statement instead of a huge if-elseif chain |
| 17:21 |
hmmmm |
real shame that lua has no switch statement at all |
| 17:21 |
sapier |
the vertical text is/was there in standard menu too ... as anyone requested to show the menu exactly as before its in formspec too |
| 17:22 |
|
RealBadAngel joined #minetest-dev |
| 17:22 |
hmmmm |
er.. i wanted to remove the vertical text a while ago |
| 17:22 |
hmmmm |
never got around to it |
| 17:22 |
sapier |
no I can't use a switch statement if you have text identifiers? |
| 17:22 |
RealBadAngel |
hi |
| 17:22 |
sapier |
ok if there's consens about removing it I'll remove it but I wont add it back once removed it |
| 17:26 |
RealBadAngel |
we do have now formspec based menu, right? |
| 17:27 |
hmmmm |
yup, just avoid clicking 'configure' for now ;) |
| 17:27 |
RealBadAngel |
hehe why? |
| 17:27 |
celeron55 |
hmmmm: in languages like lua, you're really supposed to make a table with mapping from keys to callback functions |
| 17:28 |
PilzAdam |
hmmmm, do you have some example usage of register_biome()? |
| 17:28 |
celeron55 |
or if you only need to translate values, then keys to values |
| 17:28 |
hmmmm |
it either lua errors, segfaults, or has misplaced formspec buttons |
| 17:28 |
hmmmm |
segfaults on windows |
| 17:28 |
hmmmm |
pilzadam, just the ones i gave you |
| 17:28 |
hmmmm |
i still discourage people from using biomes and v7 |
| 17:28 |
PilzAdam |
I only have this old one http://pastebin.com/xwmCD3ae and it doesnt work anymore |
| 17:29 |
hmmmm |
ahh.. i pasted what i currently have to somebody in #minetest |
| 17:30 |
hmmmm |
hm, celeron, makes sense, they really take the asssociative array idea to the maximum |
| 17:32 |
RealBadAngel |
i would like to make shaders configuration menu, by now with one option to choose, but in future for many settings. shall i wait for more settings aviable or start with just one? what do you guys think? |
| 17:32 |
hmmmm |
http://pastebin.com/qQ2uw3Nm |
| 17:32 |
PilzAdam |
thx |
| 17:32 |
hmmmm |
wait for more settings... and it'd be nice if the menu got stable before people started trying to add features |
| 17:33 |
RealBadAngel |
ok |
| 17:33 |
RealBadAngel |
so by now i will add just config file setting for bumpmapping |
| 17:34 |
sapier |
can someone explain to me what's up with "Button positioning ... Configure menu" |
| 17:35 |
hmmmm |
http://mg.viewskew.com/mgoblin_media/media_entries/59/FormspecMenu.png |
| 17:36 |
hmmmm |
er not button, what are those things called |
| 17:36 |
hmmmm |
tabs |
| 17:36 |
hmmmm |
tab buttons |
| 17:36 |
sapier |
ok that's the wine problem ? |
| 17:37 |
hmmmm |
i don't know what pilzadam used... i guess wine |
| 17:37 |
PilzAdam |
yep |
| 17:37 |
sapier |
I don't have any idea what wine is messing up to get that result |
| 17:44 |
PilzAdam |
hmmmm, how do I tell the mapgenerator to put dirt_with_grass on top of dirt? |
| 17:46 |
hmmmm |
see, that's something i was going to change today |
| 17:47 |
hmmmm |
right now you can't. i have the grow dirt into dirt with grass hard coded |
| 17:48 |
hmmmm |
i am redoing the entire top node placement mostly because it's inflexible and also it's partially broken when i added in positive 3d noise mountains |
| 17:52 |
sapier |
ok I guess I fixed all non wine related issues currently known about formspec menu |
| 17:52 |
hmmmm |
how did you fix the enter button press? |
| 17:53 |
sapier |
ok that one is missing ... i was talking about the todo list |
| 17:53 |
hmmmm |
oh okay |
| 17:53 |
hmmmm |
the listbox line concatenation too? |
| 17:54 |
sapier |
I can't reproduce this |
| 17:54 |
hmmmm |
me neither |
| 17:54 |
sapier |
I can't even guess what's happening there |
| 17:54 |
sapier |
I assume this happens on wine only too? |
| 17:55 |
hmmmm |
apparently not... i saw it once |
| 17:56 |
|
Calinou joined #minetest-dev |
| 17:56 |
sapier |
if someone finds a way to reproduce it I may be able to fix it |
| 17:59 |
sapier |
enter key should start game in singleplayer and server tab right? |
| 17:59 |
hmmmm |
yes |
| 17:59 |
hmmmm |
you fixed the address issue for the server tab, right? |
| 17:59 |
Calinou |
oh, that |
| 17:59 |
sapier |
address issue? |
| 17:59 |
Calinou |
when you disconnect, the preselected server is not used |
| 17:59 |
Calinou |
instead, the server you previously connected to is used |
| 17:59 |
Calinou |
to fix, you have to select another server then select the first server again |
| 18:00 |
hmmmm |
click on something from the favorites list, the server name gets put in the address box, not the address |
| 18:00 |
hmmmm |
so then it tries to connect to Foobar's Awesome Server:30000 |
| 18:00 |
sapier |
oh ok I'll fix it |
| 18:00 |
hmmmm |
I tried fixing it but i got frustrated and gave up after it started saving the name to the settings for no apparent reason |
| 18:02 |
sapier |
enter key in server tab is a little bit tricky |
| 18:02 |
hmmmm |
yeah, i thought it was a simple fix yesterday too |
| 18:03 |
sapier |
in singleplayer it's not an issue |
| 18:03 |
sapier |
but in server tab the textfields catch the enter button |
| 18:03 |
sapier |
I guess I know how to fix it |
| 18:03 |
hmmmm |
i suppose you can see if all of the textfields are part of the entry but no other elements are |
| 18:04 |
sapier |
I already pass editbox enter event too |
| 18:05 |
sapier |
ok fixed |
| 18:08 |
hmmmm |
heh. what's tricky is to see if your choices for perlin noise values are good or not, if they fall into your range most of the time... i've been assuming this whole while that they form a normal distribution, but this might not be the case :( |
| 18:09 |
sapier |
ok I think I fixed the server address issue too |
| 18:10 |
sapier |
anything else? |
| 18:10 |
|
moistersatcg joined #minetest-dev |
| 18:10 |
hmmmm |
not to my knowledge |
| 18:10 |
hmmmm |
do you think we can make a bold font? |
| 18:10 |
hmmmm |
for the credits tab i wanted to use a bold face instead of yellow text, but the latter was simpler |
| 18:11 |
Calinou |
doesn't freetype support multiple fonts already? |
| 18:11 |
* Calinou |
would rather like outlined text :P |
| 18:11 |
Calinou |
outlined text can be read on any color |
| 18:15 |
|
BackupCoder joined #minetest-dev |
| 18:18 |
Taoki |
Anchakor: RPG's are RPG's. They have fixed environments and objectives / purposes such as quests and leveling |
| 18:18 |
sapier |
irrlicht listbox doesn't support setting item font |
| 18:23 |
hmmmm |
oh, yeah, i forgot about that |
| 18:23 |
hmmmm |
irrlicht guis are very limited in what they can do |
| 18:23 |
sapier |
sadly irrlicht gui elements are quite limited |
| 18:23 |
hmmmm |
and if you bring it up to the developers they'll say "go draw it yourself, you have the tools' |
| 18:23 |
sapier |
there are some minor glitches I can't fix due to it ... I'm just waiting for someone to recognize ;-) |
| 18:24 |
Taoki |
hmmmm: http://forum.minetest.net/viewtopic.php?pid=97920 |
| 18:24 |
hmmmm |
oh.... taoki, you know i can't do that |
| 18:25 |
Taoki |
Do what? |
| 18:25 |
hmmmm |
add facedir rotation as well |
| 18:25 |
Taoki |
Why? I already havbe the formula to fix it (lua version at last) |
| 18:25 |
Taoki |
**have |
| 18:25 |
hmmmm |
yeah, this is a lua thing |
| 18:25 |
hmmmm |
load up the area in a vmanip |
| 18:25 |
hmmmm |
scan through things that have a facedir |
| 18:26 |
hmmmm |
edit those |
| 18:26 |
hmmmm |
oh wait i see the problem, can't modify the facedir |
| 18:26 |
Taoki |
Why can't they be fixed on schematic creation? |
| 18:26 |
Taoki |
yes |
| 18:26 |
hmmmm |
because that's not something for the schematic api to do |
| 18:26 |
celeron55 |
8D |
| 18:26 |
hmmmm |
seriously, give them an inch, they ask for a mile |
| 18:26 |
hmmmm |
zz |
| 18:26 |
Taoki |
heh |
| 18:26 |
hmmmm |
well |
| 18:26 |
* celeron55 |
orders a lightyear |
| 18:27 |
hmmmm |
set_node calls |
| 18:27 |
hmmmm |
ah right, that's what i said about this |
| 18:27 |
Taoki |
"seriously, give them an inch, they ask for a mile" - What reporting a bug in an implementation means |
| 18:27 |
hmmmm |
this is a feature request |
| 18:27 |
celeron55 |
it's not a bug when it's not even intended to do it; but it's a bug in the interface that it's hard to do by oneself |
| 18:28 |
Taoki |
Any scheatic with a torch or stair will be imported incorrectly. But ok, I just reported it is all |
| 18:28 |
hmmmm |
it's not imported |
| 18:28 |
hmmmm |
it wasn't made for importing things |
| 18:28 |
Calinou |
Taoki: except if the torch's direction is already right :P |
| 18:28 |
hmmmm |
again with this |
| 18:28 |
Taoki |
Creating, sorry |
| 18:28 |
Calinou |
1/6 chance :P |
| 18:28 |
hmmmm |
i didn't add the rotation because you requested it either, by the way |
| 18:29 |
hmmmm |
i did it because that was part of my original plan |
| 18:29 |
kahrl |
<sapier> kahrl is there a reason why you call a function parameter m_device? |
| 18:29 |
kahrl |
yes; it is called sed s/ |
| 18:29 |
kahrl |
can you tell I was tired when writing that patch? |
| 18:30 |
hmmmm |
right so my official stance on facedir is the same as my official stance on node metadata: the nodes that do have these are not present in bulk, and encountering them are rare exceptions that need to be handled individually |
| 18:30 |
hmmmm |
schematic api and voxelmanip api are there to handle bulk node data |
| 18:30 |
Taoki |
hmmmm: That's very ok and not relevant. I'm not pointing this bug out like I ordered a feature from you (I did ask it for my mod but also because I considered it would be needed overall) |
| 18:30 |
sapier |
no problem kahrl :) |
| 18:31 |
hmmmm |
so I really don't see what's *wrong* with scanning through the things you need to give special attention to with a voxelmanip, and then calling set_node() on it |
| 18:32 |
Taoki |
Only thing I'd see wrong with that is that one has to use a different system to "repair" the results of a bug with the original creation of the schematic |
| 18:32 |
hmmmm |
... |
| 18:32 |
celeron55 |
stop calling it a bug or someone will explode |
| 18:32 |
hmmmm |
where did i ever say that this fixes up facedirs |
| 18:33 |
hmmmm |
and it'll mow your lawn as well, right? |
| 18:33 |
hmmmm |
it doesn't? |
| 18:33 |
hmmmm |
bug! |
| 18:33 |
hmmmm |
gotta fix this BUG! |
| 18:33 |
hmmmm |
need to apply a BUGFIX |
| 18:33 |
Taoki |
You totally misunderstand what I'm saying |
| 18:34 |
Taoki |
I called it a bug because the incorrect facedirs cause broken schematic creation. I consider that a bug as much as skipping blocks and spawning cincompletely. But of course it's not for me to say "fix it now" and it's not what I meant... just pointed it out |
| 18:35 |
hmmmm |
when i move into an apartment and find that it doesn't already have a washing machine, i start calling it broken |
| 18:35 |
Taoki |
Any schematic with a torch, stair, or door is created incorrectly (eg: is broken). If that's not a bug then I don't know what is |
| 18:35 |
Taoki |
Again, not to say "fix it now", but as a principle |
| 18:36 |
Taoki |
Might as well just import dirt nodes instead of the node type in the schematic |
| 18:37 |
hmmmm |
it's a big WONTFIX here |
| 18:37 |
celeron55 |
you're both fighting about a different thing |
| 18:38 |
celeron55 |
that's useless |
| 18:38 |
Taoki |
If that's your point on it, I understand. But by a normal standard, that is still a bug. |
| 18:38 |
Taoki |
yeah |
| 18:38 |
Taoki |
Went into explaining why I think it's a bug is all |
| 18:38 |
Taoki |
Gonna post on the forum for anyone else who wants to try fixing it (since it likely can be fixed) |
| 18:39 |
celeron55 |
it would be much more useful if hmmmm noted how he thinks the thing should be implemented in builtin like it was decided |
| 18:39 |
celeron55 |
and if Taoki then made that, or something |
| 18:39 |
celeron55 |
Taoki: it won't be fixed in C++, it will be implemented in builtin as a wrapper function, as far as i understand |
| 18:39 |
hmmmm |
yes, i originally proposed a builtin wrapper function for this |
| 18:39 |
hmmmm |
"place schematic things the elegant but slow way" |
| 18:40 |
Taoki |
Only thing that upset me is that hmmmm considers it's not a bug although the created structures are created incorrectly / brokenly. Just a thing of theory, but I don't get how that could be not a bug :P |
| 18:40 |
hmmmm |
not a single doorknob left unpolished |
| 18:40 |
Taoki |
Builtin sounds good too if that works better, sure |
| 18:40 |
Taoki |
I already have the Lua-side formula |
| 18:40 |
celeron55 |
Taoki: it's not a bug in hmmmm's code, you're just using the wrong interface (the interface you want to use doesn't exist yet and won't be made by hmmmm) |
| 18:41 |
celeron55 |
or, well, this is how i understand it |
| 18:41 |
hmmmm |
basically yeah |
| 18:41 |
celeron55 |
i will of course come and punch each of you in the face if you don't somehow solve this |
| 18:41 |
celeron55 |
go on |
| 18:41 |
hmmmm |
i called upon others to write wrappers for voxelmanip/schematic things |
| 18:41 |
Taoki |
celeron55: I thought minetest.create_schematic is the command you use directly in a mod. If that's intended to go through a buildin first, and not used as the raw function, that is different then |
| 18:41 |
hmmmm |
voxelarea.lua was made by me for testing and i just happened to be kind enough to include that helper class |
| 18:41 |
Taoki |
The Lua API specifies that as the command to use to create schematics though |
| 18:42 |
Taoki |
If it's just something that wasn't added in place yet and is a known thing, that's ok too. Again I'm not saying "Aaah the feature I wanted is broken and you suck". |
| 18:43 |
Taoki |
Just one thing I don't understand though; Why can't it be fixed in C++ in the schematic_import function? By adding the code that flips facedirs based on angle |
| 18:44 |
hmmmm |
why can't it be fixed in Lua? |
| 18:44 |
hmmmm |
the reason why it doesn't do it in C++ is because of all the extra checks required for the sake of handling a special edge case that's very rare |
| 18:45 |
PilzAdam |
hmmmm, I have a problem to remove the air from my schematic file; I loop through the whole area with this https://gist.github.com/PilzAdam/5921533 and set every air nodes probability to -1, but the air nodes are still placed |
| 18:45 |
celeron55 |
hmmmm: i'm curious though: how much more code does it even need? |
| 18:45 |
celeron55 |
if done in C++, that is |
| 18:45 |
hmmmm |
a special edge case that's only really present when used for things aside from its actual purpose |
| 18:45 |
Taoki |
hmmmm: It can. I just thought it would be most correct for the import function to do it all correctly from the core. If it's fixed in Lua I don't mid (though that would require using the create_)schematic function through a different function, but again that's good too) |
| 18:46 |
Taoki |
Also, it's not an "extremely rare case". There are probably lots of schematics that could use things with facedirs. |
| 18:46 |
hmmmm |
PilzAdam, it depends on how you place it actually |
| 18:46 |
hmmmm |
are you using place_schematic or placing it through the mapgen? |
| 18:46 |
Taoki |
hmmmm: place_schematif cuntion |
| 18:46 |
hmmmm |
i have place_schematic explicitly place air nodes |
| 18:46 |
Taoki |
**function |
| 18:46 |
PilzAdam |
place_schematic for testing, but its meant to be in mapgen later |
| 18:47 |
PilzAdam |
oh, thats strange |
| 18:47 |
Taoki |
Ah you meant PilzAdam, thought you asked me |
| 18:47 |
hmmmm |
right, when it's being placed in the mapgen, i have air nodes not placed |
| 18:47 |
hmmmm |
so it doesn't carve out odd pieces of map |
| 18:47 |
hmmmm |
but i figured it'd be okay to force place them with place_schematic because it was completely intentional |
| 18:48 |
hmmmm |
if you'd like me to allow you to toggle this behavior with place_schematic, i guess i can do that |
| 18:48 |
PilzAdam |
well, if people want air to be placed then they dont set the probability to -1 |
| 18:48 |
hmmmm |
celeron, as for that, it's more or less the slippery slope i'm afraid of going down |
| 18:49 |
celeron55 |
hmmmm: why? |
| 18:49 |
celeron55 |
what do you expect to happen |
| 18:49 |
hmmmm |
because then i handle a special param2 case |
| 18:49 |
hmmmm |
now i have to handle everybody elses' special param2 cases |
| 18:49 |
celeron55 |
ehm |
| 18:49 |
celeron55 |
it's a param2 case that is also hardcoded in the client |
| 18:49 |
hmmmm |
and now there's no excuse for me to not handle node metadata while i'm at it |
| 18:50 |
celeron55 |
then i request you to proceed to making the lua wrapper |
| 18:50 |
celeron55 |
if it doesn't exist in a matter of days, there will be only trouble |
| 18:51 |
celeron55 |
or at least make an empty function in the right place so that someone can make it in the way you've thought it should be |
| 18:51 |
sapier |
https://github.com/minetest/minetest/pull/806 fixes for formspec menu rebased to current master |
| 18:51 |
hmmmm |
conceivably, though, it'd take a if (ndef->get(c).facedir) vm->m_data[i].setFaceDir(rotateFaceDir(blah, rot)); |
| 18:51 |
celeron55 |
currently nobody really knows exactly what you think you want |
| 18:51 |
celeron55 |
and it's a problem |
| 18:52 |
hmmmm |
i'll think about what to do |
| 18:52 |
RealBadAngel |
https://github.com/minetest/minetest/pull/805 i have added config setting to enable bumpmapping and bash tool to create basic normal maps with gimp |
| 18:52 |
hmmmm |
i don't really want to cave on this |
| 18:52 |
hmmmm |
PilzAdam, anyway sorry for not focusing before |
| 18:52 |
hmmmm |
i want to really solve the problem now |
| 18:53 |
hmmmm |
so you go through all that, check for air, and set it to 0 probability if that happens |
| 18:53 |
hmmmm |
sounds reasonable |
| 18:53 |
PilzAdam |
actually to -1 |
| 18:53 |
RealBadAngel |
that tool can make normal maps quickly but its highly advised to fine tune them later, different surfaces needs different settings |
| 18:53 |
PilzAdam |
the docs say its never placed then |
| 18:53 |
hmmmm |
well zero probability, i mean always placed if the prob parameter is 0 |
| 18:54 |
PilzAdam |
yea |
| 18:54 |
hmmmm |
alright, i'll just say "never placed" or "always placed" from now on |
| 18:54 |
hmmmm |
hrm, that looks right |
| 18:54 |
hmmmm |
those hardcoded coordinates are absolute coordinates, right? |
| 18:54 |
hmmmm |
not relative |
| 18:55 |
PilzAdam |
yes |
| 18:56 |
hmmmm |
and this table is being passed along to a create_schematic call, right? |
| 18:56 |
hmmmm |
just making sur |
| 18:57 |
PilzAdam |
yes |
| 18:57 |
PilzAdam |
heres the whole code: https://gist.github.com/PilzAdam/5921687 |
| 18:59 |
hmmmm |
that should be working |
| 19:01 |
hmmmm |
now it could be that there are duplicate entries in the probability table |
| 19:01 |
PilzAdam |
nope |
| 19:01 |
PilzAdam |
oh well, Im pretty sure there arent |
| 19:02 |
hmmmm |
and somehow the probability entries that set the node position to 186 come after the one that sets it to never place |
| 19:02 |
hmmmm |
but that doesn't even matter because it'd be internally set to content ignore.... |
| 19:02 |
hmmmm |
so after the 'never place' probability is applied you can overwrite the probability with anything and it'll never stick |
| 19:02 |
hmmmm |
hmmm |
| 19:03 |
Taoki |
hmmmm: Still taking suggestions on mapgen v7 (which for safety doesn't mean "do it now" or " do it cuz I want to")? |
| 19:03 |
hmmmm |
taoki, i'm all ears |
| 19:03 |
PilzAdam |
hmmmm, just tested it in mapgen with register_decoration() and it works fine |
| 19:03 |
Taoki |
Wanted to ask if you could make the water node also a parameter of the biome definitions. So special biomes can have their different water types |
| 19:04 |
hmmmm |
sure, why not i guess |
| 19:04 |
Taoki |
Nice :) Could be useful for realms |
| 19:05 |
hmmmm |
pilzadam, of course it works fine when placing it as a decoration because air is never placed |
| 19:06 |
PilzAdam |
DecoSchematic::resolveNodeNames: node 'ignore' not defined |
| 19:06 |
hmmmm |
well that'll explain why that happens |
| 19:07 |
hmmmm |
okay WTF. |
| 19:07 |
hmmmm |
we know what the problem is at least |
| 19:07 |
Taoki |
PilzAdam: I get that warning too with Structures mod when using schematics. Thought it's just me |
| 19:08 |
Taoki |
But if not, I can confirm |
| 19:08 |
hmmmm |
that's literally not right though. "ignore" is always present, always always |
| 19:08 |
hmmmm |
oh |
| 19:08 |
hmmmm |
i see what the problem is |
| 19:08 |
hmmmm |
:) |
| 19:08 |
PilzAdam |
hmmmm, it writes "ignore" to the schematic file when prob=-1, right? |
| 19:08 |
hmmmm |
yes, i'll fix that |
| 19:09 |
hmmmm |
it's the way i handle errors in resolveNodeName() |
| 19:09 |
hmmmm |
now only one question remains... why didn't i encounter this error when i was testing it? |
| 19:10 |
PilzAdam |
if 2 biomes have the same heat and humidity then only the first one is generated |
| 19:10 |
hmmmm |
yup |
| 19:11 |
hmmmm |
you were expecting something else? |
| 19:11 |
PilzAdam |
no, just making sure that Im not blind |
| 19:11 |
PilzAdam |
what are the ranges for heat and humidity? |
| 19:12 |
PilzAdam |
0 to 100? |
| 19:12 |
hmmmm |
i don't really know |
| 19:12 |
hmmmm |
fixing the ranges to be 0 to 100 was on my todo list |
| 19:13 |
PilzAdam |
and the average is 50? |
| 19:14 |
|
BackupCoder joined #minetest-dev |
| 19:14 |
hmmmm |
i *assume* that perlin noise values have a normal distribution, and that the max value is M = p^0 + p^1 + p^2 + ... + p^n where p = persistence, and n = number of octaves |
| 19:14 |
hmmmm |
and the minimum is -M |
| 19:15 |
hmmmm |
(1 - pow(np->persist, np->octaves + 1)) / (1 - np->persist) would be the maximum |
| 19:16 |
RealBadAngel |
anybody with suggestions/ideas on #805? |
| 19:16 |
PilzAdam |
hmmmm, I see a problem with trees when placed as schematic: some trunks are blocked by the leaves of another tree |
| 19:16 |
hmmmm |
and of course, 0 would be the average value |
| 19:17 |
hmmmm |
pilzadam, that's a bit tougher to solve |
| 19:17 |
hmmmm |
now you're saying you want placement priorities |
| 19:18 |
PilzAdam |
:-) |
| 19:19 |
PilzAdam |
or just say that schematics override other schematics |
| 19:19 |
PilzAdam |
oh, that wouldnt work |
| 19:22 |
RealBadAngel |
sapier i got reports that latest changes broke completely formspec backgrounds |
| 19:23 |
RealBadAngel |
ERROR[main]: Invalid pos for element background specified |
| 19:23 |
hmmmm |
oh.... actually sorry, the max value would be M = p^0 + p^1 + ... + p^(n-1) |
| 19:24 |
hmmmm |
so the max value of perlin noise is (1 - pow(np->persist, np->octaves) / (1 - np->persist) |
| 19:24 |
sapier |
rba I'll have a look |
| 19:25 |
|
SpeedProg joined #minetest-dev |
| 19:25 |
RealBadAngel |
sapier, ask VanessaE about details, i operating now on the tree that is a day behind, so im not familiar with the errors |
| 19:26 |
hmmmm |
i'm not sure if 50 is the average value, but probably not.. i made it so that it resembles an average value but again i need to do actual work on it |
| 19:27 |
sapier |
VanessaE can you give additional information about the background error? I changed almost everything there but it should work as before ;-) |
| 19:27 |
hmmmm |
to translate your noise into having the average value you'd want in your distribution, you'd need noise * scale + offset, where scale is the std dev of your distribution, and offset is the average you'd like to have |
| 19:28 |
hmmmm |
just don't bother coding things for biomes yet, all of it still blows a lot |
| 19:29 |
sokomine |
hmmmm: i'd still love to have a parameter node_replacement_list for place_schematic where you could tell it to e.g. use "ignore" for all nodes of "air" in the .mts file, and "air" for e.g. all nodes marked with wool:yellow (air inside a house is very important - outside it can be other scenery), or to use default:cobble instead of default:wooden_planks and so on. that would allow more variety. and the .mts file seems to be constr |
| 19:30 |
|
BackupCoder joined #minetest-dev |
| 19:30 |
|
khonkhortisan joined #minetest-dev |
| 19:30 |
hmmmm |
i'll consider it |
| 19:31 |
|
proller joined #minetest-dev |
| 19:33 |
VanessaE |
sapier: ok, two things. One, menu backgrounds are refusing to display anything but what's part of my game modes or the defaults. Images supplied by HDX are no longer being displayed in the menu. |
| 19:34 |
VanessaE |
two, and I don't know if this is related, formspec background images, such as those supported by Unified Inventory, are not being displayed either, citing the above "invalid pos" error (many such) every time I open the formspec e.g. the inventory display. |
| 19:34 |
sapier |
can you give an example for background formspec? the message rba posted indicates the formspec not beeing correct (at least what I expected to be correct) |
| 19:34 |
VanessaE |
easiest way to see both in effect by installing any resolution of HDX and visit my server. |
| 19:35 |
VanessaE |
s/by installing/is to install/ |
| 19:35 |
VanessaE |
code-wise, I cannot (but RBA can :P ) |
| 19:36 |
sapier |
RealBadAngel can you give me an example for the background formspec tag now considered wrong? ;-) |
| 19:36 |
VanessaE |
the formspec backgrounds might just be a mod bug, but I'm pretty sure the main menu backgrounds are your fault :P |
| 19:36 |
VanessaE |
:) |
| 19:37 |
RealBadAngel |
idk which are causing errors, but i can give you sample usage |
| 19:37 |
sapier |
no formspec was almost completely rewritten for mainmenu too sometimes guessing what code was intended to do |
| 19:37 |
VanessaE |
so it's possible that breaking one broke both. |
| 19:37 |
sapier |
yes |
| 19:37 |
sapier |
possible but from what you tell I'd expect different errors |
| 19:38 |
RealBadAngel |
http://pastebin.com/0DqLgUzL |
| 19:38 |
PilzAdam |
hmmmm, for now v7 does exactly what I want, just generate ocean, sand grasslands and forest |
| 19:38 |
RealBadAngel |
all the backgrounds used by UI |
| 19:38 |
PilzAdam |
+, |
| 19:38 |
sapier |
why is there an additional "," after y pos? |
| 19:38 |
VanessaE |
RealBadAngel: maybe the final comma? |
| 19:38 |
VanessaE |
what sapier said. |
| 19:39 |
sapier |
that's the problem I check if position is exactly 2 elements |
| 19:39 |
RealBadAngel |
hallo, VanessaE, its YOUR code |
| 19:39 |
sapier |
with additional , its 3 x,y,<empty> |
| 19:40 |
PilzAdam |
hmmmm, just one thing, where does minetest search for the schematic file? |
| 19:40 |
PilzAdam |
I currently use the full path, wich isnt a good solution |
| 19:41 |
hmmmm |
it's relative to the executable |
| 19:41 |
sapier |
ok this one is clear now the other one backgrounds not shown how are hdx images supplied vanessa? |
| 19:41 |
hmmmm |
a solution would be to call get_modpath() during the initialization, store it, and then use that to get the schematic from your mod |
| 19:42 |
RealBadAngel |
VanessaE, can you confirm removal of comma helps? |
| 19:43 |
VanessaE |
I'll try it, hold. |
| 19:43 |
PilzAdam |
hmmmm, oh, and I think would just encourage that <modname>_<biomename> format for biomes |
| 19:43 |
PilzAdam |
s/_/: |
| 19:45 |
VanessaE |
RealBadAngel: confirmed, removing the commas fixed the issue. |
| 19:45 |
VanessaE |
there were about 8 or so instances among that section of code, suggest you search for others also. |
| 19:45 |
|
celeron55 joined #minetest-dev |
| 19:45 |
RealBadAngel |
ok i will update technic and technic_game soon |
| 19:46 |
VanessaE |
sapier: correct on the HDX part - those menu splash/background images have been there for months now, and stopped displaying with my latest pull, so I can only assume it was your menu formspec code. |
| 19:46 |
PilzAdam |
hmmmm, Im sure you know that already, but there are problems with the height interpolation between bioms (i.e. 2 node difference for me) |
| 19:47 |
sapier |
yes most likely ... we're talking about singleplayer tab only right? |
| 19:47 |
VanessaE |
sapier: also, the "Client" tab keeps losing the last Address/Port I connected to. |
| 19:47 |
VanessaE |
no sir, all tabs display the "only shows the default images" isuse. |
| 19:47 |
VanessaE |
issue* |
| 19:48 |
VanessaE |
also, the game icons need to stay at the bottom of the window regardless of window size. |
| 19:48 |
sapier |
ok currently background is only set for singleplayer tab didn't know there should be something shown on other tabs too |
| 19:49 |
sokomine |
hmmmm: thank you for considering it! such a node replacement would be best done at that location imho. i need it for all my houses. spawning them always with the same materials would be boring, and getting ignore-nodes would otherwise be pretty difficult |
| 19:49 |
sapier |
last one can't be fixed without major formspec rewrite vanessa |
| 19:49 |
VanessaE |
when I maximize, the icons are moved up several cm, as if the whole thing is matted onto an 800x600 canvas. |
| 19:50 |
VanessaE |
sorry, this menu formspec thing, honestly, is slightly worse than what came before |
| 19:50 |
sapier |
it's a single formspec either everything is scaled or nothing ... first version did scale menu too that wasn't accepted |
| 19:51 |
sapier |
but back to things that can be fixed |
| 19:51 |
sapier |
the hdx image how are they expected to behave? |
| 19:51 |
VanessaE |
sapier: identically to the default images. |
| 19:52 |
|
BackupCoder joined #minetest-dev |
| 19:52 |
VanessaE |
they just replace them when the texture pack is configured to be used via minetest.conf |
| 19:52 |
VanessaE |
(I don't use textures/all except for a couple of font files) |
| 19:53 |
VanessaE |
sapier: download the 16px size of HDX (it's about 20MB), configure minetest.conf to point to it (texture_path=/foo/bar/HDX-16px) and you'll see what I mean |
| 19:53 |
sapier |
ok I guess thats easy to fix |
| 19:53 |
VanessaE |
http://digitalaudioconcepts.com/vanessa/hobbies/minetest/screenshots/Screenshot%20-%2003092013%20-%2001:46:02%20PM.png |
| 19:53 |
VanessaE |
her's what it used to look like |
| 19:53 |
VanessaE |
(except I've since changed that text and added a resolution identifier at the bottom) |
| 19:54 |
VanessaE |
here* |
| 19:55 |
hmmmm |
pilzadam, what do you mean? |
| 19:55 |
hmmmm |
about the format for biomes (you never told me about anything like this before) |
| 19:55 |
hmmmm |
and you're using a different number of topnodes i take it |
| 19:56 |
PilzAdam |
http://mg.viewskew.com/mgoblin_media/media_entries/62/MapgenBug.png |
| 19:57 |
hmmmm |
for now, set the number of top nodes to the same value |
| 19:57 |
PilzAdam |
is that because of a top_depth difference of one? |
| 19:57 |
hmmmm |
yes |
| 19:57 |
hmmmm |
i'm redoing that entire thing soon |
| 19:58 |
PilzAdam |
k |
| 19:58 |
|
iqualfragile joined #minetest-dev |
| 19:58 |
hmmmm |
the number of top nodes will fill up to the height at that point |
| 19:58 |
hmmmm |
so instead of the ground height being 0, and 2 top nodes, making the new ground height 2, it'll just subtract 2 from the ground height and it'll be 0 |
| 19:59 |
hmmmm |
also i'm thinking of varying depth of top nodes by perlin noise like v6, but it'll be varied depth instead of height (like minecraft does iirc) |
| 20:07 |
sapier |
VanessaE game specific backgrounds override texture specific ones right? |
| 20:07 |
VanessaE |
if there's no texture pack, yeah |
| 20:08 |
VanessaE |
if there's a texture pack, then the pack's images override all |
| 20:08 |
sapier |
even game? |
| 20:08 |
VanessaE |
per menu, of course |
| 20:08 |
VanessaE |
*nod* even the game |
| 20:08 |
VanessaE |
HDX has images for the default menu, minetest_game and my game. |
| 20:08 |
sapier |
sounds wrong to me but if it's already done this way I'm gonna rebuild it |
| 20:08 |
VanessaE |
(others will follow sooner or later) |
| 20:09 |
sapier |
wait you post additional images within the texture pack? |
| 20:09 |
VanessaE |
yep. |
| 20:09 |
sapier |
great .... why do we always need to do same thing 100 different ways |
| 20:09 |
VanessaE |
several of us kinda hemmed and hawed over it until we settled on this method. |
| 20:10 |
VanessaE |
default imagery first, then game-specific imagery if it's provided by the game, then texture-pack-specific imagery overrides those. |
| 20:10 |
sapier |
dosn't change the fact that this adds lots of useless complexity |
| 20:11 |
sapier |
how are filenames specified within a texture pack? |
| 20:14 |
VanessaE |
menu_header.png, menu_overlay.png, minetest_menu_header.png, minetest_menu_overlay.png, vanessae_menu_overlay.png |
| 20:16 |
sapier |
doesn't help I can't hardcode names ;-) first is gameid? |
| 20:16 |
VanessaE |
so if not the default menu, then it's [gameid]_menu_xxxx.png |
| 20:16 |
VanessaE |
where xxx = header, overlay, etc. |
| 20:20 |
celeron55 |
i don't get sapier's fear of trying a couple of filenames |
| 20:21 |
celeron55 |
if you can't cope with that kind of complexity easily, you're a pretty useless coder |
| 20:21 |
VanessaE |
celeron55: hush, let the man fix what he broke instead of disparaging him. |
| 20:21 |
celeron55 |
seriously; i know it takes a bit more effort but it just is worth the bit |
| 20:21 |
sapier |
it's not a fear but a wast try this try that try something else then try next |
| 20:22 |
sapier |
if a texure pack doesn't supply a menu image fallback to default? |
| 20:24 |
celeron55 |
it's an order of priority |
| 20:24 |
sapier |
no it's an order of trial and error |
| 20:24 |
celeron55 |
1) texture pack, 2) game, 3) default |
| 20:24 |
celeron55 |
... |
| 20:24 |
sapier |
don't blame me for filling up the logs ;-P you wanted me to do it |
| 20:25 |
celeron55 |
do you realize all of that is *obviously* needed? |
| 20:25 |
celeron55 |
we can't drop out defaults, we can't drop out games, we can't drop out texture packs |
| 20:25 |
celeron55 |
because none of that would make any sense |
| 20:26 |
celeron55 |
and because of that, there's no reason to complain |
| 20:26 |
sapier |
no it's not it's pushing at least 2 different aproaches into a single menu |
| 20:26 |
sapier |
resulting in a menu that doesn't really fit any of the two aproaches |
| 20:26 |
sapier |
but I'm tired of argueing if you want it that way I'll put it in |
| 20:27 |
celeron55 |
i know it's inconsistent in that way, but that's not the programmer's problem |
| 20:27 |
celeron55 |
it's waiting for a proper redesign |
| 20:27 |
celeron55 |
designing a menu while coding it isn't really a good idea |
| 20:27 |
sapier |
the formspec menu gives anyone chance to adapt the menu to whatever she/he needs for his game |
| 20:28 |
sapier |
putting all in one results in a default menu is exactly opposit direction ;-) |
| 20:28 |
celeron55 |
if you want to redesign it, you could do it; just split it somehow so that it makes sense and allows that adaptation sanely |
| 20:29 |
sapier |
it already is separated any game can set it's own menu ... of course it'd need to adapt the main menu somehow but the more we blow up the main menu the more difficult for someone to strip of unneded parts |
| 20:30 |
sapier |
and there already are lots of things in not really needed for it to do purpose of a main menu |
| 20:31 |
celeron55 |
two options: make it work like before, or make it work better |
| 20:33 |
sapier |
I choose work as before as "work better" is a 100% subjective target that can't be achieved in general |
| 20:34 |
sapier |
but I deny any blames about complexity that's not my fault |
| 20:35 |
VanessaE |
sapier: it's your fault anyway, just because :) |
| 20:36 |
sapier |
I wonder if I somehow could hide a if vanessae then crash in code .... hmmm |
| 20:36 |
|
khonkhortisan_ joined #minetest-dev |
| 20:36 |
VanessaE |
:P |
| 20:37 |
VanessaE |
I manage to find ways to crash the game/server without any help anyways :D |
| 20:38 |
RealBadAngel |
so folks, can i count on any comments on my bumpmapping code? |
| 20:39 |
khonkhortisan_ |
I like it |
| 20:44 |
sapier |
what about the logo? |
| 20:45 |
VanessaE |
what about it? |
| 20:45 |
sapier |
texture pack specific too? |
| 20:45 |
VanessaE |
everything needs to be |
| 20:45 |
VanessaE |
imho |
| 20:45 |
VanessaE |
if it is supplied as an image somewhere under minetest/ or minetest_game/, it should be possible for a texture pack to replace it. |
| 20:45 |
VanessaE |
(or another game mode of course) |
| 20:46 |
sapier |
lol you don't really know how much texture operations this will cause ;-) |
| 20:46 |
VanessaE |
heh |
| 20:46 |
VanessaE |
I suppose not :) |
| 20:46 |
sapier |
hmm ... no actually I can't add a fallback for in formspec images |
| 20:47 |
VanessaE |
well then leave those be for now |
| 20:47 |
VanessaE |
what's more important is the imagery around/behind the menu |
| 20:48 |
sapier |
I'll make everything texturepackspecific except the logo |
| 20:49 |
VanessaE |
the logo? |
| 20:49 |
sapier |
in credits |
| 20:49 |
VanessaE |
oh in the credits tab? |
| 20:49 |
VanessaE |
meh, that's fine the way it is |
| 20:49 |
sapier |
and the icons in game selector will stay as game specifies too |
| 20:50 |
VanessaE |
note btw that with my "bigger, sharper" font, lots of stuff is now overwriting other stuff just a bit when it didn't before. |
| 20:50 |
sapier |
should be same for ingame formspecs too I guess |
| 20:50 |
VanessaE |
in-game formspecs work fine with texture packs. |
| 20:51 |
sapier |
I was talking about the font ;-) |
| 20:51 |
sapier |
ingame uses texture cache which isn't available before client startup |
| 20:51 |
sapier |
so all images used in mainmenu are explicitly loaded |
| 20:51 |
VanessaE |
http://digitalaudioconcepts.com/vanessa/hobbies/minetest/screenshots/Screenshot%20-%2002272013%20-%2007:48:22%20PM.png |
| 20:52 |
VanessaE |
this is provided by HDX, for example |
| 20:52 |
VanessaE |
and btw, that part still works after those comma changes. |
| 20:52 |
VanessaE |
right, so you probably need not to anything special about in-game formspec textures, those work fine |
| 20:53 |
VanessaE |
(as long as RealBadAngel doesn't fuck up ;) ) |
| 20:54 |
sapier |
ok plz have a look at that pull request I recently mentioned I just added the image fixes there |
| 20:55 |
sapier |
I couln't test the game/texture pack specific things as I don't have any supported game installed atm |
| 20:56 |
VanessaE |
ok, I'll pull it and see how it behaves. |
| 20:56 |
sapier |
at least it shows texture pack backgrounds &co for other tabs |
| 20:58 |
VanessaE |
sapier: *shakes head* that did nothing. |
| 20:59 |
sapier |
do you specify a full path in texture_path or a relative one? |
| 20:59 |
VanessaE |
in the singleplayer tab, I see the image from the selected game - but it's the one the game provides. In all other tabs, just the clouds and nothing else. |
| 20:59 |
VanessaE |
it's a full path. |
| 20:59 |
sapier |
ooohh |
| 20:59 |
sapier |
try relative to minetest folder I'll fix that |
| 20:59 |
VanessaE |
afaik that's how it has to be, since I'm using a system-wide install. |
| 21:00 |
VanessaE |
no good, relative to ~./minetest (means ../Minetest-textures-and-mods/......) does nothing to the menu; textures still work in-game though |
| 21:01 |
VanessaE |
did you make sure to set the image scaling properly? if it's wrong, an image simply isn't drawn I guess |
| 21:01 |
sapier |
no it's just looking at wrong location atm |
| 21:01 |
VanessaE |
ok |
| 21:05 |
sapier |
ok now using absolute path |
| 21:07 |
VanessaE |
oh sure, now github starts to flake out on my. |
| 21:07 |
VanessaE |
me* |
| 21:10 |
sapier |
:-) me too |
| 21:15 |
VanessaE |
ok, trying it now. |
| 21:15 |
VanessaE |
yay! |
| 21:15 |
VanessaE |
MUCH better |
| 21:15 |
sapier |
do texture pack specific game backgrounds work too? |
| 21:15 |
VanessaE |
yes |
| 21:15 |
sapier |
good |
| 21:15 |
VanessaE |
it's back to normal |
| 21:15 |
VanessaE |
now, about the game selection bar. |
| 21:16 |
VanessaE |
surely there's some way to detect where the bottom of the window is? |
| 21:16 |
VanessaE |
(and move it to there |
| 21:16 |
VanessaE |
) |
| 21:16 |
sapier |
imagebuttons are specified by pos |
| 21:16 |
sapier |
pos is relative to topleft corner of menu ... not even screen |
| 21:17 |
VanessaE |
this is going to cause a problem then |
| 21:17 |
sapier |
without adding a completely different positioning scheme it's not possible ... if adding it it needs to be added for all elements |
| 21:17 |
VanessaE |
because it basically means nothing can be reliably placed into the footer area |
| 21:18 |
VanessaE |
at 800x600 for example, it just overwrites the bottom edge of the title for my game, but maximized to 1280x1024, it overwrites the top half of that same title |
| 21:18 |
sapier |
true ... only way to do so requires redefinition of formspec syntax |
| 21:19 |
sapier |
of course we could do a quick n dirty hack but I hate those ones as they tend to grow and grow and grow ... resulting in such monsters as the engine <-> game <-> modengine triple purpose mainmanu currently implemented |
| 21:20 |
sapier |
imho if a game wants special menu features it should implement its own menu |
| 21:20 |
VanessaE |
I could move my title images to the top of course, but that doesn't fix the real issue |
| 21:20 |
VanessaE |
yeah but this happens on any game screen |
| 21:20 |
VanessaE |
the problem is where the game selection bar is placed, not the rest of the menu |
| 21:21 |
sapier |
imho there's no issue just a missing feature that can be set to todo ... but imho this is only worth adding if formspec is completely redesigned |
| 21:21 |
sokomine |
hmmmm: could you make the minetest schematic file format (.mts) accessible through lua? that is, add a minetest.read_schematic( filename ) function (that could be done evnen now with just the definition of the file format), allow manipulation of said data (e.g. i'd really like to rewrite the node-name-list part; reading how large the schematic is is also virtually important; rotation and handling of param2 could also be done wit |
| 21:21 |
sapier |
the game selection bar is part of the menu even if it looks not exactly like that |
| 21:21 |
VanessaE |
no, this has to be changed, because you're overwriting part of a game-supplied image in a way the game cannot predict |
| 21:22 |
sapier |
it's not placed at the end of screen that's just an illusion |
| 21:22 |
VanessaE |
the bar needs to be at the bottom left corner of the window regardless of the size of the window |
| 21:23 |
sokomine |
the point would then be to accept a differently formated lua table - one very close to the mts file format - to be passed on to place_schematic. i think that would help a lot! the table ought to contain all the variables with their values the same way as the .mts file (just as part of a lua table). expanding the table to the full {nodename, param1, param2} format as could be done right now seems very inconvenient and impractical |
| 21:24 |
sapier |
I'll put that at the end of my list as I don't consider this a bug, it's 100% exactly predictable where the bar is either |
| 21:24 |
VanessaE |
nonono you're overcomplicating it |
| 21:24 |
sapier |
no I'm not there is NO way to tell a formspec to place something relative to anything else then topleft corner of formspec menu |
| 21:24 |
VanessaE |
am I right in assuming negative values aren't allowed in a formspec item's coordinates? |
| 21:24 |
sapier |
they are |
| 21:25 |
VanessaE |
ok |
| 21:25 |
sapier |
but for most elements placed outside of menu box they won't be shown |
| 21:25 |
VanessaE |
then how about adding some other character instead of a - sign, say an underscore or something, that signifies "relative to lower left of the window"? |
| 21:25 |
sapier |
I've added special handling for some of them to avoid them beeing hidden |
| 21:26 |
|
smoke_fumus|2 joined #minetest-dev |
| 21:26 |
VanessaE |
so if I were to give a position of _123,_456 this should mean to draw the image starting at -123,-456 from the bottom left of the window. |
| 21:26 |
sapier |
ok A for topleft B for topright C for bottomleft D for bottomright a for topleftmenu b for toprightmenu c for bottomleftmenu and d for bottomrightmenu |
| 21:27 |
sapier |
and of course mixed coordinates are allowed too |
| 21:27 |
VanessaE |
well if you want to go that far, sure, but I'm just trying to replicate the previous layout. |
| 21:27 |
sapier |
no I won't add a new coordinate system for now at all ... once all important bugs are fixed I'm gonna considering that |
| 21:28 |
VanessaE |
this is a bug - it breaks the formatting of a person's game screen layout. |
| 21:28 |
sapier |
coordinate system in formspec is already crap as some are padded some aligned some use default offsets some not |
| 21:28 |
sapier |
no it's a change in behaviour |
| 21:29 |
RealBadAngel |
sapier, an idea, could you make main menu music dependent on game selected? |
| 21:29 |
RealBadAngel |
so the game should provide music file |
| 21:29 |
VanessaE |
sapier: it needs fixed. I don't care if it means taking the game selection icons out of the formspec and hard-coding them like before. |
| 21:29 |
VanessaE |
it's wrong and it breaks things. |
| 21:29 |
sapier |
no music isn't handled by mainmenu at all atm |
| 21:29 |
VanessaE |
RealBadAngel: don't change the subject. |
| 21:30 |
RealBadAngel |
it was the offtopic idea, sorry |
| 21:30 |
sapier |
vanessa I already tried that it's not possible to have two different menus shown at same time |
| 21:30 |
VanessaE |
then make it possible :) |
| 21:30 |
sapier |
plain no |
| 21:31 |
VanessaE |
well then I'd have to move to revert the formspec menu until it can be made possible. |
| 21:31 |
sapier |
make them revert it back to c++ menu or live with it until a sane solution has been found |
| 21:33 |
sapier |
I'm tired of having to readd all bugs until concept of a addition is completely lost so you can't threaten me by it if you thing the new features modmanager/gamemanager online modstore automated zip modinstall aren't worth giving up some bad habit just revert it |
| 21:33 |
VanessaE |
who's threatening? |
| 21:34 |
VanessaE |
I'm just saying not to replace something that works with something else that's less capable is all |
| 21:34 |
VanessaE |
(for once, celeron55 and I agree) |
| 21:34 |
sapier |
It's even more capable but different |
| 21:34 |
VanessaE |
yes, different enough to break existing stuff in ways that the existing stuff's authors *can't fix* |
| 21:37 |
VanessaE |
if the authors of the broken stuff could fix their stuff to work with the new model, that would have been perfectly fine to change the model. |
| 21:37 |
RealBadAngel |
VanessaE, damn u ;) i have copied that coma after you into 20+ formspecs ;) |
| 21:37 |
VanessaE |
but the new method breaks the ability to use a footer for anything meaningful (it's supposed to be for displaying some info about the game) |
| 21:38 |
VanessaE |
RealBadAngel: sucks to be you :P :) |
| 21:38 |
sapier |
what about replacing the bottombar by a sidebar? |
| 21:38 |
sapier |
no wont help |
| 21:38 |
VanessaE |
you mean the game selector? |
| 21:39 |
sapier |
bad idea won't help either |
| 21:39 |
RealBadAngel |
anyway technic is updated, technic_game update will follow soon |
| 21:39 |
VanessaE |
if you were to make the game selector part of the menu section with just a tiny offset, that would work |
| 21:39 |
VanessaE |
because then the game's/texture pack's menu footer could be moved to the bottom of the window and would never be overwritten |
| 21:40 |
sapier |
ok that'd be a minor change |
| 21:40 |
VanessaE |
make it the same width as a menu, with a really thin space between it and the menu, or part of the menu with a horizontal rule |
| 21:41 |
VanessaE |
you may want to consider shifting whole apparatus upwards by 20 or so pixels to keep it centered. |
| 21:41 |
VanessaE |
but that's purely cosmetic |
| 21:42 |
VanessaE |
and make the selection bar just a bit thinner - don't waste too much space above/below |
| 21:42 |
VanessaE |
and... and... and.. ;) |
| 21:45 |
RealBadAngel |
technic_game is also updated |
| 21:50 |
sapier |
have a look at new variant vanessae |
| 21:50 |
VanessaE |
oh sure, blame me eh :) |
| 21:51 |
VanessaE |
that's PERFECT |
| 21:51 |
VanessaE |
well, almost ;) |
| 21:51 |
sapier |
ok hope now pilzadam isn't upset the other one was the version he could accept |
| 21:51 |
VanessaE |
maybe a white horizontal line |
| 21:51 |
PilzAdam |
hm? |
| 21:52 |
VanessaE |
instead of the gap between the two parts |
| 21:52 |
sapier |
gamebar |
| 21:52 |
VanessaE |
http://digitalaudioconcepts.com/vanessa/hobbies/minetest/screenshots/Screenshot%20-%2007032013%20-%2005:52:40%20PM.png |
| 21:52 |
VanessaE |
PilzAdam: ^^^^^ |
| 21:53 |
RealBadAngel |
i would like to ask again, really no one core dev but c55 have an opinion on bumpmapping? :P |
| 21:53 |
VanessaE |
sapier: yeah, put a thin white line between the menu and the game selector, instead of the thin gap. that'll look better |
| 21:53 |
PilzAdam |
RealBadAngel, I wonder why you start throwing lighting related features in if we dont have hardware lighting yet |
| 21:54 |
RealBadAngel |
first of all because it works |
| 21:54 |
RealBadAngel |
and works perfectly |
| 21:54 |
RealBadAngel |
we do have already 3 texture packs that support it |
| 21:55 |
RealBadAngel |
code is tested for over a month |
| 21:55 |
RealBadAngel |
need more reasons? |
| 21:55 |
sapier |
this way? |
| 21:55 |
RealBadAngel |
lemme throw the next one, i dont want to rebase it a few time a week to use it |
| 21:58 |
RealBadAngel |
also i rebased it now after removal of atlas, so i think its a good point in the timeline |
| 22:00 |
RealBadAngel |
about hw lighting (or fake ones which is more propably) using them in shaders code will just mean to pass the coords instead of using fixed one |
| 22:01 |
RealBadAngel |
and im not starting throwing anythin. im workin on shaders for over a month already |
| 22:02 |
RealBadAngel |
repo was aviable for testin for over 4 weeks |
| 22:03 |
RealBadAngel |
and the code is damn well tested |
| 22:03 |
VanessaE |
sapier: yes, exactly so |
| 22:03 |
VanessaE |
that's good |
| 22:03 |
VanessaE |
now, one last tweak? |
| 22:04 |
VanessaE |
add a few pixels of space above/below the icons - you cropped the bar a little too thin |
| 22:04 |
RealBadAngel |
PilzAdam, have you even tried it? |
| 22:05 |
PilzAdam |
yea, I tried it some time ago |
| 22:05 |
hmmmm |
i have an opinion on bumpmapping, i think it looks really neat, and we'll be a step ahead once we *do* get hardware lighting |
| 22:05 |
sapier |
better? |
| 22:06 |
RealBadAngel |
gimme a bulb and it will be alive. |
| 22:06 |
RealBadAngel |
by now it just works |
| 22:06 |
VanessaE |
sapier: yes, that's good. I'd say go with exactly that |
| 22:06 |
RealBadAngel |
i dont really need a hw lights |
| 22:06 |
VanessaE |
squash those commits and merge the bastard :) |
| 22:07 |
RealBadAngel |
i need coords of them just |
| 22:07 |
sapier |
ok now lets hope next one doesn't require exactly oposite |
| 22:07 |
RealBadAngel |
we dont have even sun coords yet |
| 22:08 |
RealBadAngel |
with this one thing i could release a few more shaders including water surfaces |
| 22:08 |
VanessaE |
sapier: Abe (husband) gives his "thumbs up" also ;) |
| 22:09 |
hmmmm |
you people... |
| 22:09 |
hmmmm |
i'd be embarassed as hell if any of my family knew about minetest |
| 22:09 |
VanessaE |
lol |
| 22:10 |
VanessaE |
hmmmm: actually, the last couple of times someone else has seen minetest on my box, they were like "you play minecraft!" |
| 22:10 |
VanessaE |
(and they meant it in a good way) |
| 22:10 |
hmmmm |
if my mom saw it she'd be like "don't you do anything else on that damn computer except play video games?" |
| 22:10 |
hmmmm |
"i hate that video game" |
| 22:10 |
VanessaE |
that was my parents years ago |
| 22:10 |
RealBadAngel |
http://www.youtube.com/watch?v=UnEoyEgsgig |
| 22:10 |
hmmmm |
it's just best to not let them know anything at all |
| 22:10 |
Anchakor |
you should react to that with switching to porn |
| 22:11 |
RealBadAngel |
this is water surface made by shaders |
| 22:11 |
VanessaE |
excpet I was expected to somehow magically make money with my computer. |
| 22:11 |
sapier |
once modmanager is enabled by default minetest will become way more userfriendly |
| 22:11 |
hmmmm |
yeah |
| 22:11 |
sapier |
btw did anyone test it? |
| 22:11 |
hmmmm |
it's like if you're not making money by doing things on your computer, it's useless |
| 22:11 |
VanessaE |
hmmmm: yep. basically. |
| 22:12 |
RealBadAngel |
its not useless, its a hobby |
| 22:12 |
PilzAdam |
RealBadAngel, IIRC I have reported some problems with water when running your code |
| 22:12 |
VanessaE |
hmmmm: it comes from the attitude that if you're not making money with all of your available spare time, your activities are useless in general, and I think that's crap |
| 22:12 |
RealBadAngel |
PilzAdam, thats why code is not pulled yet |
| 22:13 |
hmmmm |
the whole 'time is money' idea you always hear lawyers say, yeah... richard stallman has a quote about this |
| 22:13 |
RealBadAngel |
i learned from my mistakes and im pulling now only things im 100% sure that are workin |
| 22:13 |
VanessaE |
hmmmm: and I'm sick-and-G*d-damned-tired of it, too |
| 22:15 |
RealBadAngel |
compile my pull, use https://github.com/RealBadAngel/sphax256 or https://github.com/RealBadAngel/HavenTP and see for yourself |
| 22:15 |
RealBadAngel |
or ask VanessaE for HDX link |
| 22:16 |
VanessaE |
https://github.com/VanessaE/hdx-256 |
| 22:16 |
VanessaE |
sapier: you missed a commit in your squash? |
| 22:16 |
RealBadAngel |
and feel the difference |
| 22:16 |
VanessaE |
oh I see what you did, nm |
| 22:17 |
sapier |
should be ok isn't it? |
| 22:17 |
VanessaE |
sapier: yeah, fine by me, but it's PilzAdam and/or hmmmm and/or RealBadAngel that have to make it offiical :) |
| 22:17 |
VanessaE |
(random pings ;) ) |
| 22:17 |
proller |
sapier, server - "public" checkbox lost |
| 22:17 |
hmmmm |
from the screenshot it seems really nice |
| 22:17 |
hmmmm |
screenshots rather |
| 22:18 |
RealBadAngel |
hmmm, i ask for gameplay feel |
| 22:18 |
RealBadAngel |
screenshots are worth shit |
| 22:18 |
sapier |
proller what does this checkbox do? |
| 22:18 |
hmmmm |
ev.. |
| 22:18 |
hmmmm |
eh....* i trust it's good |
| 22:18 |
proller |
sapier, enable server announce |
| 22:18 |
hmmmm |
if it's not good i can just turn it off |
| 22:19 |
RealBadAngel |
move your asses guys and check this out |
| 22:19 |
hmmmm |
busy |
| 22:19 |
RealBadAngel |
that really changes the gameplay feelin |
| 22:19 |
VanessaE |
proller: you know, I have to wonder if there's a point to that |
| 22:19 |
RealBadAngel |
world has a depth now |
| 22:19 |
sapier |
isn't that a setting? can you post a screenshot? I don't have curl compiled in normaly |
| 22:19 |
VanessaE |
proller: because you still need to be able to set the server URL, name, description, etc. for the announce to work properly |
| 22:20 |
proller |
no, its works without |
| 22:20 |
VanessaE |
imho there's no reason to announce on the public list from the GUI, because 99% of server owners won't use that |
| 22:21 |
RealBadAngel |
VanessaE, theres a reason, newbies will see that something is goin on here |
| 22:21 |
proller |
no reason in server tab, because 99% servers dont use that |
| 22:21 |
VanessaE |
well I guess. |
| 22:22 |
VanessaE |
proller: so you'd rather announce from the Client tab then? |
| 22:22 |
RealBadAngel |
let the newbies use it simple way |
| 22:22 |
proller |
RealBadAngel, +1 |
| 22:22 |
VanessaE |
ok, fair |
| 22:23 |
proller |
you can create server and share to friends in one click |
| 22:23 |
RealBadAngel |
genau |
| 22:24 |
proller |
sapier, it use enable_public |
| 22:24 |
RealBadAngel |
forget about this 99% |
| 22:24 |
VanessaE |
proller: plus multiple clicks through your firewall/router ;) |
| 22:24 |
RealBadAngel |
it is by now 99% hardcore users |
| 22:24 |
RealBadAngel |
not the newbies from outside |
| 22:24 |
proller |
not everyone have firewall |
| 22:25 |
proller |
and one step to make public server via ipv6 without routers |
| 22:25 |
|
iqualfragile_ joined #minetest-dev |
| 22:25 |
VanessaE |
proller: I was kidding :) |
| 22:26 |
RealBadAngel |
what minetest lacks the most by now is not hw lights, new menu or whatever |
| 22:26 |
RealBadAngel |
its a lack of click n play for total greenhorns |
| 22:27 |
RealBadAngel |
or the kids |
| 22:27 |
sapier |
proller only until ppl realize they need to setup their firewall properly for ipv6 |
| 22:27 |
RealBadAngel |
my friends family that lives in Berlin plays minetest on their home network |
| 22:28 |
proller |
after we can make game via sctp 8) |
| 22:28 |
RealBadAngel |
dad, mom and a kid |
| 22:28 |
RealBadAngel |
but i have to set everthin up for them |
| 22:29 |
sapier |
what about adding a lan discovery feature? |
| 22:29 |
RealBadAngel |
theyre simply fuckin unable to solve all the problem that we, developers just dont see |
| 22:30 |
sapier |
that's why I added modstore and modinstall gui ;-P |
| 22:30 |
RealBadAngel |
deps, builds, configs etc |
| 22:30 |
sapier |
first is "install a mod" ;) |
| 22:30 |
RealBadAngel |
then whole tragedy starts |
| 22:31 |
proller |
sapier, cant make screenshot. it was simple chechbox, named "Public" with value from "server_announce", and saves value too |
| 22:31 |
RealBadAngel |
"i did everythin right but it crashes!" |
| 22:31 |
sapier |
and where to get the server address from? |
| 22:31 |
proller |
no need |
| 22:31 |
proller |
its automatic |
| 22:32 |
sapier |
are you sure it's hardcoded? |
| 22:32 |
proller |
masterserver gets ip from announce request |
| 22:32 |
sapier |
announce request? do we already have lan discovery? |
| 22:33 |
proller |
http://servers.minetest.net/ |
| 22:33 |
sapier |
that'd be a hardcoded address |
| 22:33 |
proller |
no address |
| 22:33 |
proller |
empty in config by default |
| 22:33 |
sapier |
no it really is hardcoded |
| 22:34 |
sapier |
defaultsettings l 135 |
| 22:34 |
sapier |
but not now Its past midnight ;-P |
| 22:34 |
proller |
no! |
| 22:34 |
proller |
that'd be a hardcoded address |
| 22:34 |
sapier |
gn8 |
| 22:34 |
proller |
ups |
| 22:35 |
RealBadAngel |
hehe |
| 22:35 |
proller |
138 settings->setDefault("server_url", ""); |
| 22:35 |
proller |
139 settings->setDefault("server_address", ""); |
| 22:35 |
proller |
140 settings->setDefault("server_name", ""); |
| 22:35 |
proller |
141 settings->setDefault("server_description", ""); |
| 22:35 |
proller |
optional params |
| 22:35 |
|
sapier left #minetest-dev |
| 22:35 |
proller |
you can make gui for it |
| 22:35 |
RealBadAngel |
PilzAdam, so? have you tested current state? |
| 22:35 |
proller |
no, cant |
| 22:36 |
hmmmm |
hey proller, just wondering, was there any specific reason why you chose to use mapgen v7 for your math mapgen? |
| 22:36 |
proller |
hmmmm, to use biomes and trees in future |
| 22:36 |
hmmmm |
i see |
| 22:37 |
RealBadAngel |
when you guys need trees related info just let me know |
| 22:37 |
proller |
to create main structure from stone, and fill later with dirt, trees, etc |
| 22:37 |
proller |
to simply make ores |
| 22:38 |
proller |
and caves if player wants |
| 22:39 |
RealBadAngel |
VanessaE, i just scrolled up a bit, i missed that. Say "hallo" to Abe |
| 22:39 |
VanessaE |
he says hi :) |
| 22:41 |
proller |
if you set spawn point to top of mandelbox - it can be used as usual world but with different underground |
| 22:42 |
RealBadAngel |
hmmmm, could you turn green light for me to start publishin all those eye candy shaders? |
| 22:42 |
hmmmm |
you've had the green light for quite a while now... |
| 22:43 |
RealBadAngel |
but im following the rules |
| 22:43 |
proller |
after PilzAdam go to sleep? 8-) |
| 22:43 |
RealBadAngel |
2 of devs have to agree |
| 22:43 |
* proller |
agree but too small dev |
| 22:44 |
RealBadAngel |
proller, thats a best time to post somethin that is not minecraftish |
| 22:44 |
RealBadAngel |
;) |
| 22:44 |
RealBadAngel |
but still mc has Sonic's shaders |
| 22:45 |
RealBadAngel |
i do know their code |
| 22:45 |
RealBadAngel |
also terasology ones and a few more |
| 22:46 |
hmmmm |
zzz, just push it to upstream already |
| 22:46 |
RealBadAngel |
i have readed fucking lotsa shaders code last time |
| 22:46 |
hmmmm |
even if it's horribly broken, there's still an option to disable it |
| 22:46 |
RealBadAngel |
its disabled by default, even with shaders on |
| 22:47 |
RealBadAngel |
thats why i asked about separate shaders settings window |
| 22:47 |
RealBadAngel |
all those eye candies will consume lotsa computing power |
| 22:48 |
RealBadAngel |
but hey, thats why we are all workin for, to buy better boxes ;) |
| 22:48 |
proller |
now every mobile phone have gpu |
| 22:49 |
proller |
but on odroid still black screen 8( |
| 22:49 |
VanessaE |
Shaders as a button in the "Settings" tab, like the "Change keys" button, would make sense |
| 22:52 |
proller |
ingame public serverlist broken |
| 22:53 |
proller |
sapier wakeup! 8) |
| 22:57 |
PilzAdam |
RealBadAngel, your bumpmapping code is at RealBadAngel/master ? |
| 22:59 |
PilzAdam |
its broken |
| 23:00 |
PilzAdam |
everything is red with default textures and shaders enabled |
| 23:04 |
RealBadAngel |
when have you downloaded the code? |
| 23:04 |
RealBadAngel |
i already fixed that issue |
| 23:04 |
PilzAdam |
4 minutes ago |
| 23:04 |
RealBadAngel |
hold on a sec |
| 23:05 |
RealBadAngel |
ah fixed but not pushed |
| 23:05 |
PilzAdam |
so, this is "tested for 4 weeks" :-p |
| 23:09 |
hmmmm |
:-q |
| 23:11 |
sokomine |
idea for rollback: announce it on the server so that people may prepare for lag and won't log out in panic |
| 23:11 |
sokomine |
it's ok if the actual rollback takes time. just a server message might be helpful |
| 23:14 |
proller |
btw is anybody tries rollback with liquid_finite? |
| 23:16 |
proller |
rollback flag - its 3 minute to code |
| 23:17 |
proller |
i think to add full mod list with first announce (not include in every minute ping) |
| 23:18 |
sokomine |
the issue with rollback is mostly that rollback_check is incredably slow and unusable. the information it provides comes at a very high price (=long lag for everyone) and is of limited value as it shows only the latest action in the vicinity. i've written a simple mod that uses external grep for the check |
| 23:19 |
sokomine |
rollback as such is ok. just that message to all as a server announce before an actual rollback is done would be helpful |
| 23:23 |
RealBadAngel |
PilzAdam, this is todays request from c55 |
| 23:23 |
RealBadAngel |
to let it disable when user has not powerfull enough gfx card |
| 23:24 |
RealBadAngel |
its fixed already, please hold on for commit |
| 23:24 |
RealBadAngel |
i dont wanna screw anythin when commiting |
| 23:25 |
RealBadAngel |
hurry is good for youngsters, i like to think twice |
| 23:30 |
proller |
pu-pu-pu-push it 8) |
| 23:32 |
VanessaE |
real good? :) |
| 23:35 |
RealBadAngel |
checkin the commit now |
| 23:35 |
RealBadAngel |
dont push me ;) |
| 23:41 |
|
Weedy joined #minetest-dev |
| 23:41 |
|
Weedy joined #minetest-dev |
| 23:50 |
PilzAdam |
RealBadAngel, did you fell asleep? |
| 23:50 |
VanessaE |
fall* |
| 23:50 |
PilzAdam |
*fail |
| 23:52 |
RealBadAngel |
actually no |
| 23:52 |
RealBadAngel |
im ready to press the button |
| 23:52 |
RealBadAngel |
i just triple checked the commit |
| 23:52 |
VanessaE |
"Remember the little red button? Push the little red button!" |
| 23:52 |
PilzAdam |
you know, there is this thing called force push |
| 23:52 |
VanessaE |
"..and you better put on your seatbelt." |
| 23:53 |
RealBadAngel |
hehe |
| 23:53 |
PilzAdam |
you can push whatever crap you want and override it easily |
| 23:53 |
RealBadAngel |
i think i will compile it to be sure |
| 23:53 |
RealBadAngel |
gimme 2 minutes :) |
| 23:55 |
RealBadAngel |
yeah, thats the best test. compiling it now |
| 23:55 |
RealBadAngel |
i had to rebase two files that sapier changed |
| 23:56 |
PilzAdam |
inb4 Minetest 1.0 released |
| 23:57 |
RealBadAngel |
dont laugh, last time i thought that "dont push the green button" is a joke for a newbie :P |
| 23:57 |
sokomine |
strange behaviour of rollback: the server reports success - and then shuts down. unfortionately, i have no access to its logfiles. what is so strange is that the actions in my immediate vicinity are undone - and then the server shuts down - and the griefing is back again |
| 23:58 |
sokomine |
green button? :-) |
| 23:58 |
PilzAdam |
RealBadAngel, you push to your fork, nobody cares about that |
| 23:58 |
RealBadAngel |
"merge" |
| 23:58 |
RealBadAngel |
i thought that it was the joke because there were no other one to push ;) |
