| Time |
Nick |
Message |
| 00:41 |
|
Hijiri joined #minetest-dev |
| 00:43 |
|
ANAND joined #minetest-dev |
| 00:57 |
|
Foz joined #minetest-dev |
| 01:15 |
|
Foz joined #minetest-dev |
| 02:15 |
|
paramat joined #minetest-dev |
| 02:19 |
paramat |
"WARNING[Main]: Client side scripting is disabled by client." meh, this is annoying, no need to warn about that |
| 02:29 |
|
Miner_48er joined #minetest-dev |
| 02:31 |
paramat |
WARNING[Main]: OMG dude! One of your settings is set to the default value! |
| 02:53 |
|
leolrrj joined #minetest-dev |
| 02:56 |
leolrrj |
what are the challenges to make minetest playable in browsers through webassembly? is there any branch trying to achieve that? |
| 03:26 |
paramat |
i don't know of anyone working on that, but a fork of MT might be |
| 04:17 |
|
paramat joined #minetest-dev |
| 05:56 |
|
kaeza joined #minetest-dev |
| 06:39 |
sofar |
any reason why I shouldn't merge #8140 myself? asking for more reviews is wasting everyone's time |
| 06:39 |
ShadowBot |
https://github.com/minetest/minetest/issues/8140 -- Force send a mapblock to a player. by sofar |
| 07:37 |
|
proller joined #minetest-dev |
| 07:53 |
|
celeron55_ joined #minetest-dev |
| 08:02 |
|
proller joined #minetest-dev |
| 08:14 |
|
Beton joined #minetest-dev |
| 08:41 |
ANAND |
!tell paramat Please take a look at https://github.com/minetest/minetest/pull/7557#issuecomment-485060382 when you have the time. Thanks. |
| 08:41 |
ShadowBot |
ANAND: O.K. |
| 08:42 |
ANAND |
:) |
| 09:11 |
ANAND |
rubenwardy: What are your thoughts on #8409, with respect to https://github.com/minetest/minetest/pull/8409#issuecomment-482250041? |
| 09:11 |
ShadowBot |
https://github.com/minetest/minetest/issues/8409 -- Rewrite camera modes; implement per-player camera modes by ClobberXD |
| 09:44 |
|
proller joined #minetest-dev |
| 09:51 |
|
YuGiOhJCJ joined #minetest-dev |
| 10:29 |
|
Wuzzy joined #minetest-dev |
| 10:31 |
|
Fixer joined #minetest-dev |
| 10:48 |
|
proller joined #minetest-dev |
| 11:00 |
|
calcul0n_ joined #minetest-dev |
| 11:17 |
|
proller joined #minetest-dev |
| 11:56 |
|
pyrollo joined #minetest-dev |
| 12:04 |
p_gimeno |
I completely disagree with paramat's comment preceding the one you've linked, ANAND. When you're talking zoom factor, you're not talking FOV angle factor, but projected image size factor, and that's calculated with the formula I posted. |
| 12:04 |
|
Darcidride joined #minetest-dev |
| 12:27 |
p_gimeno |
Multiplying the angle by a factor is essentially useless for wide FOV angles, because the result is visually unpredictable. If a user has a wide FOV configured in their client, setting a multiplier of, say, 0.5 will have a radically different result than if they have a smaller FOV. It does not produce uniform results. |
| 12:27 |
p_gimeno |
The formula I gave would yield a zoom factor (which is what FOV is all about) of 2.0x. |
| 12:28 |
|
robdrake joined #minetest-dev |
| 12:28 |
|
robdrake left #minetest-dev |
| 12:50 |
p_gimeno |
I think after re-reading him I understand part of his concern. A solution that will hopefully make everyone happy is to let set_fov be used to set the FOV angle to a value, and only that, without the extra boolean, and have an extra function e.g. set_zoom that sets the zoom factor relative to the current setting. |
| 12:55 |
p_gimeno |
I also believe that set_fov is useless, and set_zoom is the only realistic use of this feature. Despite what he said earlier, scopes work by zoom factor. You have a 3x scope, for example. From a quick search: https://www.opticsplanet.com/howto/how-to-choose-a-riflescope-the-numbers.html |
| 13:00 |
rubenwardy |
Does anyone happen to know the name of the algorithm that Minetest switched to recently to do raycasting? |
| 13:00 |
rubenwardy |
in #4346 |
| 13:00 |
ShadowBot |
https://github.com/minetest/minetest/issues/4346 -- Improved getPointedThing() by juhdanad |
| 13:07 |
p_gimeno |
Quick experiment: set your FOV to 160 (maximum), take a look around then press the zoom button, and then tell me that the zoom factor is reasonable and intuitive. |
| 13:14 |
|
Fixer_ joined #minetest-dev |
| 13:28 |
|
proller joined #minetest-dev |
| 13:39 |
ANAND |
!tell p_gimeno While splitting the multiplier part into a separate set_zoom method sounds good, it probably complicates things a bit. For starters, we'd be having two setters (and maybe even two getters) for the exact same feature - paramat's going to turn red with rage :) |
| 13:39 |
ShadowBot |
ANAND: O.K. |
| 13:41 |
ANAND |
!tell p_gimeno I also can't come up with usecases for FOV multipliers apart from fixed zoom factor like weapon optics, binoculars, or telescopes. Therefore, I don't see why your formula shouldn't be used to implement FOV multipliers... |
| 13:41 |
ShadowBot |
ANAND: O.K. |
| 13:46 |
ANAND |
I wish HexChat supports linking to irc.minetest.net logs out of the box... :) |
| 13:47 |
ANAND |
!tell paramat Also, some discussion on FOV multipliers: http://irc.minetest.net/minetest-dev/2019-04-23#i_5533315 |
| 13:47 |
ShadowBot |
ANAND: O.K. |
| 14:18 |
|
p_gimeno joined #minetest-dev |
| 14:24 |
p_gimeno |
I can only see a use for a zoom factor modifier, I can't see a use for setting a fixed FOV. |
| 14:26 |
p_gimeno |
The way I thought it would work, these would not be setters for the same property. One would set the FOV (which is useless IMO) and the other would set the zoom factor, which is a modifier for the FOV. |
| 14:41 |
|
AntumDeluge joined #minetest-dev |
| 14:49 |
|
proller joined #minetest-dev |
| 15:02 |
p_gimeno |
made my calculations: from a 160° FOV to an 80° FOV the zoom is 6.75X. From 80° FOV to 40° FOV the zoom is 2.3x. The first zoom factor is hugely disproportionate. And in general, the shorter the angle, the closer halving the angle will be to doubling the zoom. |
| 15:07 |
p_gimeno |
So, a case can be made that too wide FOVs are rare, and that it's not a huge concern misbehaving at these extreme angles. I'm not too comfortable with that argument, but it's not my call. |
| 15:22 |
|
proller joined #minetest-dev |
| 15:58 |
|
Sokomine joined #minetest-dev |
| 16:20 |
|
AntumD joined #minetest-dev |
| 16:22 |
|
Antum joined #minetest-dev |
| 16:24 |
|
AntumDeluge joined #minetest-dev |
| 17:08 |
|
pyrollo joined #minetest-dev |
| 17:32 |
|
AntumDeluge joined #minetest-dev |
| 17:48 |
|
p_gimeno joined #minetest-dev |
| 18:23 |
|
kaeza joined #minetest-dev |
| 18:52 |
|
pyrollo joined #minetest-dev |
| 18:59 |
|
proller joined #minetest-dev |
| 19:00 |
|
kaeza joined #minetest-dev |
| 19:41 |
|
ensonic joined #minetest-dev |
| 20:07 |
|
Fixer joined #minetest-dev |
| 20:16 |
|
calcul0n_ joined #minetest-dev |
| 20:32 |
|
proller joined #minetest-dev |
| 20:45 |
|
paramat joined #minetest-dev |
| 22:12 |
|
Ruslan1 joined #minetest-dev |
| 22:51 |
p_gimeno |
paramat: it's all about the reason why people change FOV. If you have a big screen, and you're relatively close to it, you want to widen your FOV in order to not have to turn your head in order to get the whole picture. |
| 22:51 |
p_gimeno |
However, if you're forced a fixed-angle FOV, it makes you have too much zoom, and you have to step back and look at the whole monitor in order to see the whole screen. |
| 22:53 |
p_gimeno |
That's why I think that a zoom factor-based FOV control is useful. Ideally, if you *really* want to make it work like a real scope, you *additionally* would limit the view using a black circle. |
| 22:53 |
p_gimeno |
That would accomplish the FOV angle that you are referring to. |
| 23:00 |
p_gimeno |
Then there's also those who alter the FOV just in order to be able to have a viewer angle to see potential threats. I think it's acceptable in that case to assume that such view is their "eyes", and the magnification factor is applied on it. They would also get the benefit of a wider angle that they want, unless the black circle is added. With the method of restricting the angle, they are getting a bigger magnification factor, which can be an |
| 23:00 |
p_gimeno |
advantage or an inconvenience depending on the situation. |
| 23:00 |
p_gimeno |
viewer -> wider |
| 23:13 |
p_gimeno |
there's also the case that in wide screens, since the default FOV is applied vertically and the horizontal one is the one that corresponds to the aspect ratio (with the tangent formula), that can cause too much of a "fisheye" effect and be considered uncomfortable or cause dizziness. |
| 23:15 |
p_gimeno |
Those people will reduce the FOV. In that case, applying a factor to the angle will asymptotically tend to the same factor applied on the tangent, so I don't think it's a concern. |
| 23:18 |
p_gimeno |
Forcing a fixed FOV on them is also probably OK. They just get a wider view than the one specified in the FOV, and the zoom factor they get is different. |
| 23:19 |
p_gimeno |
Anyway, I'll drop the discussion here. |
| 23:58 |
|
paramat joined #minetest-dev |
