Sauerbraten 2006-04-26 release!

by Aardappel_ on 04/26/2006 08:01, 106 messages, last message: 09/02/2006 04:19, 120817 views, last view: 05/03/2024 08:04

http://sourceforge.net/project/showfiles.php?group_id=102911

# tweaks & fixes to the capture gameplay mode
# added high resolution skyboxes, and some of the g_pack textures to the standard set
# fixed newent sending invalid attributes in coopedit mode
# models with a specularity of 0 now have a specialized shader ("nospecpvmodel") for faster drawing (mainly for vegetation)
# grenadelauncher damage increased to max 75
# multiplayer players & flag models will not be culled no matter how far away they are
# added waterfall effect for water material sides
# added pistol ammo (ent type "cartridges")
# added linearly-interpolated normals for lighting (controlled by "lerpangle", "lerpsubdiv", and "lerpsubdivsize" vars)
# added HW occlusion culling support (requires NV GeForce3-class hardware or better)
# height map mode. will hopefully allow for easier terrain editing.
# reduced overdraw dramatically
# reduced size of vertex data by using short instead of float
# removed ZYX order inconistency from math classes
# fixed lights placed outside the world not casting shadows
# fixed grenades being stopped by clip material and being stuck in walls
# made networking protocol use less bandwidth
# calclight now uses multi-sampling and adaptive sampling
# faster mapmodel ray intersects with spheretrees instead of bsp
# shorter fuse on grenades (2 seconds)
# fixed hudgun light computation related crash

   Board Index   

#19: Re: OSX and CVS

by >driAn<. on 04/26/2006 20:30, refers to #16

The webinterface and the anonymous cvs were not synced the last weeks.

reply to this message

#20: Re: OSX and CVS

by absinth on 04/26/2006 20:58, refers to #19

btw, the mac release is done:

http://prdownloads.sourceforge.net/sauerbraten/sauerbraten_2006_04_26_occlusion_edition_mac.dmg?download


mac-specific changes include:
• added option to the launcher to turn off shaders
• the launcher now remembers it's settings
• compiled sauerbraten with debug symbols, so the crash logs are actually useful

reply to this message

#21: Re: fb capture & krpg1 kill sauer with SDL parachute

by eihrul on 04/26/2006 21:31, refers to #13

Do "/floatvtx 1" from the console and see if that stops the crashing.

reply to this message

#22: Visual artifacts

by guest57 on 04/26/2006 23:02

Latest version seems to have introduced a few minor visual artifacts, e.g. on \"capture maps.../face_capture\" when looking from one tower to the other then pieces of it flicker visible/invisible.
OSX 10.4.6, 2x2Ghz G5, ATI x800 (1.0.1f46 driver)

Meanwhile, is anyone running a server (with the latest protocol) that I can connect to :-)

reply to this message

#23: Re: Visual artifacts

by absinth on 04/27/2006 01:02, refers to #22

yes i would also be interested if the artifacts are a cross-platform issue, or mac specific...

there were always some artifacts and even more in the RPG maps, i wouldn't say that they got worse in this version...

reply to this message

#24: Re: Visual artifacts

by Aardappel_ on 04/27/2006 01:04, refers to #23

can you make a screenshot?

reply to this message

#25: Server up n runnin

by Passa_familypc on 04/27/2006 03:24

I just upgraded my server to the new version, so who ever asked that, there is a server now.

Aardappel: You were right, I tested the latest verion on this comp, which does not have BETA drivers, and there was no problem. Only thing is this comp has a 6600GT so the result is not conclusive until I test normal drivers on my pc.

reply to this message

#26: Re: Visual artifacts

by kurtis84 on 04/27/2006 03:32, refers to #24

I do still see those oddities with my GF6800 Aard, and I've tracked it down to the driver option "conformant texture clamp" which was enabled by default....disabling this option removed those greatly stretched cubes problems completely...for me.

Not sure if this is very helpfull, since I'm the only one to see these things.

I am running the 84.21 drivers.

reply to this message

#27: Re: Visual artifacts

by Aardappel_ on 04/27/2006 05:03, refers to #26

the best description I found of what this feature actually does is:

The feature in question is the behavior of GL_CLAMP texture clamp mode, which by spec clamps the coordinate to 0.0 - 1.0 and with bilinear or other filtering modes takes the samples outside the texture from border color. NVidia hardware before Geforce 3 didn't support this, but behaved like wrap mode GL_CLAMP_TO_EDGE, which samples the texture edge pixels instead of border color. Many games that were coded on older NVidia cards assumed this wrong behavior (and displayed varying degrees of artifacts on Ati, Matrox etc. cards, like the infamous "cage" skybox in Tribes 2) so NVidia retained this buggy behavior as default and only recently added the option.

Thing is, the only clamped texture in sauer is the skybox, and we use GL_CLAMP_TO_EDGE for that (GL_CLAMP gives black border lines, as does the normal GL_REPEAT). So I am not sure what we can do here to fix it.

So maybe another bug I need to report? is this their absolutely latest drivers or something?

reply to this message

#28: Re: Visual artifacts

by absinth on 04/27/2006 15:35, refers to #27

besides the usual problems of texture edges not quite matching up and having some white pixels between them (clamping?) there is the problem that there is some kind of 'flickering' happening between some parts of the geometry being rendered or not (hard to screenshot, since you frequently get a "good" frame)

i made some screenshots trying to illustrate the problem...

http://stud4.tuwien.ac.at/~e0126505/

reply to this message

#29: Re: Visual artifacts

by eihrul on 04/27/2006 17:17, refers to #28

Let me guess, you're on an ATI card, right?

reply to this message

#30: Re: Visual artifacts

by eihrul on 04/27/2006 17:41, refers to #28

And for the record, your card/drivers, whichever, do not properly implement OpenGL occlusion queries.

You can just turn off occlusion with "/oqfrags 0".

reply to this message

#31: Re: Visual artifacts

by rpointon on 04/27/2006 17:42, refers to #28

Yeap, screenshot_5_1.png & 5_2 are what I see too - it sort of flickers rapidly between the two right?

Seems something new in the latest release as I'd not observed it before.

Graphics card is ATI x800 - sorry, I was "guest57" who start this thread - damn board was slow to send me the login token...

reply to this message

#32: Re: OSX and CVS

by rpointon on 04/27/2006 17:45, refers to #19

When will the anonymous CVS be synced?

reply to this message

#33: ..

by Mad Merv on 04/27/2006 17:58

the new maps arent compatible with the old; is there any way we could retain backwards compatibility or is there a converter?

reply to this message

#34: Re: Visual artifacts

by absinth on 04/27/2006 19:39, refers to #29

yes, i'm also on an ATI x800 (XT), but i don't think the Mac drivers for this or any other card have much in common with the linux drivers

reply to this message

#35: ..

by absinth on 04/27/2006 19:40

>it sort of flickers rapidly between the two right?
yes

reply to this message

#36: ..

by absinth on 04/27/2006 19:46

using "/oqfrags 0" indeed remedies the problem, but shouldn't this be turned off automatically if the card doesn't support it?

the other problems like the tree rendering in screenshot_7.png remain...

reply to this message

#37: Re: ..

by eihrul on 04/27/2006 20:56, refers to #36

Your driver is telling Sauer it supports the feature. We can't help it if it supports it incorrectly, and it's annoying to litter the code with a bunch of per-card/per-vendor/per-driver checks for features. It's a lost cause to keep track of all the quirks of every particular card/vendor/driver combo.

It would be kinda lame if we have to entirely disable occlusion for all ATI cards.

reply to this message

#38: ..

by Mad Merv on 04/27/2006 21:26

i also use an ATI, Radeon. I noticed some flicker on light maps on fb_capture

reply to this message

   Board Index