Pages: 1 2 3 [4] 5 6 ... 8
  Print  
Author Topic: OpenArena Community Mappack Volume 1 v3 Re-Release  (Read 348047 times)
fromhell
Administrator
GET A LIFE!
**********

Cakes 35
Posts: 14520



WWW
« Reply #75 on: January 26, 2014, 06:50:39 PM »

I'm really confused now where this 'final' map pack is going.  I thought it was just a texture problem holding up the release?
Logged

asking when OA3 will be done won't get OA3 done.
Progress of OA3 currently occurs behind closed doors alone

I do not provide technical support either.

new code development on github
Neon_Knight
In the year 3000
***

Cakes 49
Posts: 3775


Trickster God.


« Reply #76 on: January 26, 2014, 09:03:13 PM »

Yeah, at first, but other problems surfaced. As I'm not going to release a v5, this is the last time to report and fix everything before v4 ships on 30/1. Just in case there are other... "surprises".
Logged


"Detailed" is nice, but if it gets in the way of clarity, it ceases being a nice addition and becomes a problem. - TVT
Want to contribute? Read this.
Gig
In the year 3000
***

Cakes 45
Posts: 4394


WWW
« Reply #77 on: January 27, 2014, 01:59:42 AM »

It can be maked with GTKradiant or NETradiant.
In case you didn't notice, just opening a map perfectly optimized with Q3Radiant (q3map) and re-compiling it from GTK/NetRadiant (q3map2) may skrew up optimizations such as lighting. Unfortunately.  Sad And in some cases, it may even not compile successfully.   Embarrassed

I think that optimizing all maps for q3map2 may have been a good thing, but probably we should have done the switch something like August 2013... having done this in 2014 (I mean, at final release time) seems to have caused brightness problems in various maps...
« Last Edit: January 27, 2014, 03:54:38 AM by Gig » Logged

I never want to be aggressive, offensive or ironic with my posts. If you find something offending in my posts, read them again searching for a different mood there. If you still see something bad with them, please ask me infos. I can be wrong at times, but I never want to upset anyone.
Gig
In the year 3000
***

Cakes 45
Posts: 4394


WWW
« Reply #78 on: January 27, 2014, 08:14:14 AM »

First of all, I just realized an important thing about "corpse not gibbing" glitch. The corpse DOES explode also in V3 versions of those maps (DM4 and CTF3), but you have the time to see the beginning of the death animation, unlike what happens in DM7, where you immediately explode. I can guess there is some difference in the position of the triggers, or something similar. But in my case, the corpse did not explode at all was also related to the fact I forgot DO NOT LINK[/b]) h t t p s : / / openarena . wikia . com/wiki/Dmflags#Dmflags]DMFLAGS 1032 active (no self damage). However: with dmflags 1032, in V3 DM4 and CTF3 your corpse does not get gibbed at all when you fall in the void... while in DM7 it does. With dmflags 0, your corpse gets gibbed in all three maps, but in DM4 and CTF3 you did have the time to see the initial part of the death animation. V4 CTF3 has got immediate gibbing in both cases. I have no problem if DM4 is kept unfixed, the problem is less noticeable than I though before.

Giving a quick look to the V4-beta maps (quick run in gametype 0, no bots):

- I have not understood: DM4 is the version from Akom, or has been modified? This post said it has not been modified, but in the .pk3 I can find some .ase "mapobjects" with their name starting with "akom4-".  Undecided .bsp "last modified" date reports 05/12/2013.
- DM5: I also noticed the thin "white dots line" Pelya noticed in this post. More, I just realized that some jumps from the outer platforms to the central platform are possible with 125 Hz physics, but not (or at least, I have not been able to perform them with a few tries) with 90 Hz or "accurate" physics. See first screenshot attached. I'm sorry, I should have tested this before. You decide what to do.
- DM7: The V4Beta version has got some items dropped to the bottom of the void, and has got some parts too dark (especially the lower side of the arena). TEST version has still got too dark areas, and some new lights have something weird (are some red and yellow lights mixed?) -see third and fourth screenshots-. I don't know about the new item placement, maybe Moixie should tell... maybe this one may even be better, but however isn't it a bit too late to change map flow?  Huh
- DM9: Now that it is without fog, that "end of the world" looks bad. Maybe Akom's original one was a bit better, with the sanctuary suspended in the air? Other ideas are welcome, too...
- DM10: I'm not sure what Pelya means exactly with "it's a little dimmer": I think it's still the same map than V3, no? .bsp file modify date reports 15/01/2014, just like in V3 and V2 (15/01/2014 is the date I released udestruction beta version 18, which added "ambient 2"). Do you mean there is less difference between light and dark? May be the side effect of ambient light (I purposely used a low "ambient" number trying to limit this side effect, but...). Or maybe you did mean something else?
- CTF2 I fear the bug I noticed here with bots standing still in that place, suspended from the ground, has not been fixed yet. I tried the map in Overload Harvester mode again (the only V4 map I tried with bots, for the moment), and after some time I found again Gargoyle where you see it in the second screenshot attached (standing still, suspended from the ground).  Sad

OTHER (packaging, minor):
- Zip file -> sources/mapsources.txt says "The maps are located inside of the z_oacmp-volume1.pk3 file.". I can guess it should mention "z_oacmp-volume1-v4.pk3".
- Maybe the .pk3 file "sources" folder may contain a similar txt file, mentioning textures sources? However, that's already at the beginning of the main readme file, so it's not stricly necessary. Decide yourself.
« Last Edit: January 27, 2014, 08:53:48 AM by Gig » Logged

I never want to be aggressive, offensive or ironic with my posts. If you find something offending in my posts, read them again searching for a different mood there. If you still see something bad with them, please ask me infos. I can be wrong at times, but I never want to upset anyone.
Neon_Knight
In the year 3000
***

Cakes 49
Posts: 3775


Trickster God.


« Reply #79 on: January 27, 2014, 09:26:39 AM »

Yep, oacmpdm4 wasn't modified at all, but I was modifying it. In the end I've decided to rollback those changes and keep the v3 version untouched, so those models are leftovers of those changes which didn't made the cut.

That part on dm5, I guess, shouldn't be reachable at all.

dm7 has lighting as a pending improvement. I can rollback oacmpdm7 to the previous placement, but there was a huge overload of items.

dm10 hasn't changed from the v1 release at all.

About the ctf2 bug, are you sure you don't see the flag platforms? Because the three models are aligned with them, and those platforms are already aligned with the obelisk models.
Logged


"Detailed" is nice, but if it gets in the way of clarity, it ceases being a nice addition and becomes a problem. - TVT
Want to contribute? Read this.
Gig
In the year 3000
***

Cakes 45
Posts: 4394


WWW
« Reply #80 on: January 27, 2014, 09:52:19 AM »

That part on dm5, I guess, shouldn't be reachable at all.
Note sure. Maybe we should ask what was Jan's idea?

Quote
About the ctf2 bug, are you sure you don't see the flag platforms? Because the three models are aligned with them, and those platforms are already aligned with the obelisk models.
Where did I mention missing "flag platforms"? :-/ I said there are bots appearing camping in the air, but I did not mention "flag platforms". The place of the screenshot is NOT next to flags/skull receptacles, but is on the extreme part of the map (near to that lightning gun spawn point). It's not in the middle of the room, but on the platform you get after a few stairs. Please give a better look to the first two screenshots at http://openarena.ws/board/index.php?topic=4905.msg49900#msg49900 to better distinguish about which place I'm talking about. Then, I have no idea which bug you did identify the other day (the one you were surprised you didn't notice before)... :-/ I'm a bit confused...

Maybe I misunderstood something from your post? Do you have some screenshot about the part you are talking about?
« Last Edit: January 27, 2014, 09:58:13 AM by Gig » Logged

I never want to be aggressive, offensive or ironic with my posts. If you find something offending in my posts, read them again searching for a different mood there. If you still see something bad with them, please ask me infos. I can be wrong at times, but I never want to upset anyone.
Akom74
Member


Cakes 9
Posts: 906


Q3A/OA Mapper


WWW
« Reply #81 on: January 27, 2014, 10:30:46 AM »

It can be maked with GTKradiant or NETradiant.
In case you didn't notice, just opening a map perfectly optimized with Q3Radiant (q3map) and re-compiling it from GTK/NetRadiant (q3map2) may skrew up optimizations such as lighting. Unfortunately.  Sad And in some cases, it may even not compile successfully.   Embarrassed

I think that optimizing all maps for q3map2 may have been a good thing, but probably we should have done the switch something like August 2013... having done this in 2014 (I mean, at final release time) seems to have caused brightness problems in various maps...

What are you talking about ?? The only difference is in the light that has to be incremented in the .shader (in case of skybox) or by entity light (in closed maps).

The geometry optimized rest as it is.

By the way, my Sanctuary map was surely better in the original version, but i think that for optimize it, it was necessary to do some changes....... Embarrassed.....

I think that Neon_Knight does not matter about brightness problems, he likes dark maps  Tongue Tongue

Wink
Logged

...sorry for my English, i'm Italian... Tongue



Gig
In the year 3000
***

Cakes 45
Posts: 4394


WWW
« Reply #82 on: January 27, 2014, 10:53:04 AM »

What are you talking about ?? The only difference is in the light that has to be incremented in the .shader (in case of skybox) or by entity light (in closed maps).

The geometry optimized rest as it is.
Maybe not exactly. I think Neon_Knight said he had do to some "under the hood" hacks to be able to correctly compile some of your maps, starting from "making detail" many brushes (a thing I fear may negatively affect in-game performances). When he did a try with a customized version of Under Destruction, apart the quite different lighting, a few strange things happened: a part of the water started "blinking"... and some textures were scaled up (those textures scaled up also appeared to me when I tried to move to q3map2 last summer), and framerate was lower. His version was not so bad in general, but it would have required to spend a lot of time for fixing it.
Logged

I never want to be aggressive, offensive or ironic with my posts. If you find something offending in my posts, read them again searching for a different mood there. If you still see something bad with them, please ask me infos. I can be wrong at times, but I never want to upset anyone.
Akom74
Member


Cakes 9
Posts: 906


Q3A/OA Mapper


WWW
« Reply #83 on: January 27, 2014, 11:15:03 AM »

When he did a try with a customized version of Under Destruction, apart the quite different lighting, a few strange things happened: a part of the water started "blinking"... and some textures were scaled up (those textures scaled up also appeared to me when I tried to move to q3map2 last summer), and framerate was lower. His version was not so bad in general, but it would have required to spend a lot of time for fixing it.

This happen in GTKradiant or in NETradiant ?

Wink
Logged

...sorry for my English, i'm Italian... Tongue



Gig
In the year 3000
***

Cakes 45
Posts: 4394


WWW
« Reply #84 on: January 27, 2014, 11:30:30 AM »

Hi! I've prepared FIXED (meaning "corrected") mappools, that should be working (a quick try I did was successful).

Usage:
place this pk3 in your baseoa folder

/exec oacmp1_enable
 to set g_mappools variable
/exec oacmp1_disable
 to reset it to its default (load a map first).
Oacmp1_enable.cfg and oacmp1_disable.cfg files include brief usage infos. They do not control g_autonextmap variable, which should be set to 1 to use DO NOT LINK[/b]) h t t p s : / / openarena . wikia . com/wiki/G_autonextmap]auto change map feature.

I renamed the .cfg mappool files NK previously did, to make thir names shorter: now the g_mappools variable does not become too long and do not cause the game to crash anymore. I haven't modified the maplists themselves.

I hope I did not make mistakes, however you all are invited to check and test!  Wink

NK, could you please just remove the old .cfg files from the OACMP package, and place those from the pk3 attached here instead? I think that should be enough. Thank you!  Smiley

------------
@Akom: Ask NK.  Smiley
When I tried, last Summer, I had problems compiling from GTK Radiant, NetRadiant and Q3map2build, IIRC. They did compile the map, but the map showed some differences I did not know how to manage.
« Last Edit: January 27, 2014, 12:04:27 PM by Gig » Logged

I never want to be aggressive, offensive or ironic with my posts. If you find something offending in my posts, read them again searching for a different mood there. If you still see something bad with them, please ask me infos. I can be wrong at times, but I never want to upset anyone.
Neon_Knight
In the year 3000
***

Cakes 49
Posts: 3775


Trickster God.


« Reply #85 on: January 27, 2014, 12:00:34 PM »

I'll replace them if they're correctly tested.

About GTKR, NetR and q3map2build, I still insist that there was something wrongly done with those installations of GTKR, NetR and q3map2build. I've been using those tools since I've started mapping for OA (I've even switched flawlessly from GTKR to NetR, which I currently use) and never had a problem with them.

The current guides on the wiki should be rewritten, though. Hopefully the Mapping manual will contain better info for GTKR/NetR configuration. And prior to that I have to finish the gamepacks.
Logged


"Detailed" is nice, but if it gets in the way of clarity, it ceases being a nice addition and becomes a problem. - TVT
Want to contribute? Read this.
Neon_Knight
In the year 3000
***

Cakes 49
Posts: 3775


Trickster God.


« Reply #86 on: January 27, 2014, 10:51:52 PM »

ctf2: Bots spawned inside of botclip. Looking and floating bug in the areas marked by Gig solved.
Logged


"Detailed" is nice, but if it gets in the way of clarity, it ceases being a nice addition and becomes a problem. - TVT
Want to contribute? Read this.
Neon_Knight
In the year 3000
***

Cakes 49
Posts: 3775


Trickster God.


« Reply #87 on: January 27, 2014, 11:01:39 PM »

White glitches in dm5: Won't be fixed. They're pretty complex to be fixed.

(I wonder why do they use fractionary grids to build maps -0.5, 0.25, 0.125-, when it isn't something recommended and for many good reasons...)
Logged


"Detailed" is nice, but if it gets in the way of clarity, it ceases being a nice addition and becomes a problem. - TVT
Want to contribute? Read this.
Gig
In the year 3000
***

Cakes 45
Posts: 4394


WWW
« Reply #88 on: January 28, 2014, 04:18:12 AM »

More tests with mappools performed (also thanks to "callvote nextmap"). They work very well!!!!  Smiley Smiley Smiley I think you can add them safely enough.  Wink

The only one which does not work at all is the "tourney" one. But it seems this is not due to the mappool, but due to some bug in gamecode (if Sago may give a look...  Wink), or simply due to how the "tournament" mode works. Does a "tournament" match ever end? However "/callvote nextmap" with g_gametype 1 causes the same map to be restarted also with default OA 0.8.8 mappoools: the problem is somewhere in gamecode... I suppose we should still include the tourney mappool, so if nextmap for tournament mode will ever be fixed, its mappool will start working. Other gametypes seem to work perfectly, instead.  Smiley

I also compared those mappool files with the map/gametype table in DO NOT LINK[/b]) h t t p s : / / openarena . wikia . com/wiki/OACMP/Volume_1]OACMP1 wiki page, and thus I added oacmpdm5 to Elimination pool file (it was listed in the table -and in the .arena file-, but not in the mappool file). So you can use the updated package you find attached to this post.  Wink
Changes against yesterday version are just that small addition to Elimination pool, and the creation of a ".txt" file about "OACMP vol 1 auto change map" (not strictly necessary, due to the fact most of those infos are already in the .cfg files... but people may find simpler to open a .txt file than a .cfg file with a text editor): decide yourself if you want to include it.  Smiley

I also repeated the "items start solid" check for BETA4 maps:
- OACMPDM5 reports two item_health startsolid.
- OACMPDM7 (V4-beta version) reports many health items startsolid. (NK-Test version does not).

Note for who want to try OACMPDM7 NK TEST (from this post), remember rename "zz_dm7-test.pk3" to something like "_zz_dm7-test.pk3", otherwise the standard version of the map will be used instead.

« Last Edit: January 28, 2014, 12:12:28 PM by Gig » Logged

I never want to be aggressive, offensive or ironic with my posts. If you find something offending in my posts, read them again searching for a different mood there. If you still see something bad with them, please ask me infos. I can be wrong at times, but I never want to upset anyone.
Neon_Knight
In the year 3000
***

Cakes 49
Posts: 3775


Trickster God.


« Reply #89 on: January 28, 2014, 07:55:33 AM »

Oh, yeah, forgot to mention.
Pelya's reported bugs on ctf1 fixed as well.

Might as well upload the pack with all the fixed maps.
Logged


"Detailed" is nice, but if it gets in the way of clarity, it ceases being a nice addition and becomes a problem. - TVT
Want to contribute? Read this.
Neon_Knight
In the year 3000
***

Cakes 49
Posts: 3775


Trickster God.


« Reply #90 on: January 28, 2014, 08:07:01 AM »

Well, here's a small fixpack. Includes all the modified/fixed maps (ctf1, ctf2, dm5, dm7, dm9) It also includes the modified, compilable dm4.

http://www.onykage.com/files/armageddonman/!oacmp/z_oacmpv4-fixpack.pk3

Known issues:
- oacmpdm4 needs more lighting work.
- oacmpdm5 "sparklies", which won't be fixed.
Logged


"Detailed" is nice, but if it gets in the way of clarity, it ceases being a nice addition and becomes a problem. - TVT
Want to contribute? Read this.
Gig
In the year 3000
***

Cakes 45
Posts: 4394


WWW
« Reply #91 on: January 28, 2014, 09:24:22 AM »

Given a quick look to the edited maps...

- CTF1: Better now, although there are still a few small glitches (see the first three screenshots). In case you compile the map again, what about adding those famous few machinegun ammo boxes in the lateral tunnels?
- CTF2: the bots bug seems fixed now.  Wink
- DM4: Yes, more lighting work is required. However, how's possible that the sky light does not illuminate at all the vertical walls? Maybe some q3map2 options control how much light "bounces" and you disabled it completely, or something similar? It's very strange... However, I may also just live with Akom's original version of this map.
- DM5: the view-through polygon bug seems fixed. No, now the problem has just moved a few units below (a different brush), but is still there.  Embarrassed
- DM7: not sure what actually changed from your previous "test" version. It seems quite similar to your previous "test" version... maybe the lower side (visible when you fall down) is even darker now? Is it normal that no items spawn around the hatch? Those red/yellow lights (fourth screenshot) still have something "strange", IMHO...
- DM9: Maybe now the skybox effect is a bit better (with the dark instead of concrete)... however the upper part of the map is really too dark (I cannot attach more screenshots to this post, however just look at the upper part of the buildings...).
« Last Edit: January 28, 2014, 09:32:19 AM by Gig » Logged

I never want to be aggressive, offensive or ironic with my posts. If you find something offending in my posts, read them again searching for a different mood there. If you still see something bad with them, please ask me infos. I can be wrong at times, but I never want to upset anyone.
pelya
Member


Cakes 6
Posts: 399


WWW
« Reply #92 on: January 28, 2014, 09:32:51 AM »

dm4: as you've already said, more lighting work needed (ambient light perhaps).
dm5: another missing texture on the same place as before, and lighting bugs around bevels. Gig, are we using the same version?

 
Logged
pelya
Member


Cakes 6
Posts: 399


WWW
« Reply #93 on: January 28, 2014, 09:42:48 AM »

dm7: inconsistent lighting at the upper platform, it's pitch dark on one side, and illuminated on other side.
dm9: lower area is much better now, and I don't care much about dark upper part of building. Would be nice to fix, but still a minor bug.
ctf1: okay now, I've noticed some Z-fighting if you shoot rocket in the middle tunnel, but that's minor.
ctf2: still too dark.
Logged
Gig
In the year 3000
***

Cakes 45
Posts: 4394


WWW
« Reply #94 on: January 28, 2014, 10:29:32 AM »

dm5: another missing texture on the same place as before, and lighting bugs around bevels. Gig, are we using the same version?
We are using the same version... but in the first quick run with the map, I didn't notice the bug had moved to the brush below the old one (about this, I corrected my post before you posted yours). About (lighiting?) bugs (black dots and lines), probably in the first quick run I confused some of those with some marks on walls caused by weapons. I confirm there are.
More, now I also noticed another thing: for some reason, those bevels do not show "marks on walls".
Logged

I never want to be aggressive, offensive or ironic with my posts. If you find something offending in my posts, read them again searching for a different mood there. If you still see something bad with them, please ask me infos. I can be wrong at times, but I never want to upset anyone.
Neon_Knight
In the year 3000
***

Cakes 49
Posts: 3775


Trickster God.


« Reply #95 on: January 28, 2014, 10:42:56 AM »

All the mentioned bugs makes me even more convinced that good brushwork should always be the rule while mapping.

OA3 should definitely enforce good brushwork and patchwork.
Logged


"Detailed" is nice, but if it gets in the way of clarity, it ceases being a nice addition and becomes a problem. - TVT
Want to contribute? Read this.
Gig
In the year 3000
***

Cakes 45
Posts: 4394


WWW
« Reply #96 on: January 28, 2014, 10:58:40 AM »

Apart from better or worse patchwork, I don't understand the reason why your compilings gave such different results than those Jan (and Pelya) got. I mean, I can understand some difference with maps that were previously compiled with Q3Radiant (mine and Akom's)... but Jan's were always compiled with Q3MAP2, isn't it?

And I still don't understand why do your compilings have such completely dark areas (like in DM4)... how's possible the sky does not illuminate vertical walls at all?  Undecided
« Last Edit: January 28, 2014, 11:52:10 AM by Gig » Logged

I never want to be aggressive, offensive or ironic with my posts. If you find something offending in my posts, read them again searching for a different mood there. If you still see something bad with them, please ask me infos. I can be wrong at times, but I never want to upset anyone.
Neon_Knight
In the year 3000
***

Cakes 49
Posts: 3775


Trickster God.


« Reply #97 on: January 28, 2014, 11:05:19 AM »

I base the final compile settings (sans -dark and -dirty) on those which appear on the Q3Map2 page.
http://en.wikibooks.org/wiki/Q3Map2

Code:
"q3map2.exe" -meta -v "mapname.map"
"q3map2.exe" -vis -v "mapname.map"
"q3map2.exe" -light -fast -patchshadows -samples 3 -bounce 8 -gamma 2 -compensate 4 -dirty -v "mapname.map"

What I also use is some other switches, so this is really the final compile I use for the maps I compile:

Code:
"q3map2.exe" -flares -meta -patchmeta -skyfix -v -verboseentities "mapname.map"
"q3map2.exe" -vis -v "mapname.map"
"q3map2.exe" -light -fast -patchshadows -samples 3 -bounce 8 -gamma 2 -compensate 4 -v "mapname.map"

(-flares in maps with tons of light sources like oacmpdm7 is omitted)

For BSPC I use this commandline in all of the maps:

Code:
"bspc.exe" -bsp2aas "mapname.bsp" -forcesidesvisible -optimize -grapplereach
Logged


"Detailed" is nice, but if it gets in the way of clarity, it ceases being a nice addition and becomes a problem. - TVT
Want to contribute? Read this.
Neon_Knight
In the year 3000
***

Cakes 49
Posts: 3775


Trickster God.


« Reply #98 on: January 28, 2014, 11:35:22 AM »

Well, it's time to confess this.

I'm a dumb idiot. -.-

I forgot to add Akom's shaders to my shaderlist.txt. -.-
Logged


"Detailed" is nice, but if it gets in the way of clarity, it ceases being a nice addition and becomes a problem. - TVT
Want to contribute? Read this.
Akom74
Member


Cakes 9
Posts: 906


Q3A/OA Mapper


WWW
« Reply #99 on: January 28, 2014, 11:37:19 AM »

dm4: as you've already said, more lighting work needed (ambient light perhaps).

I think a higher value in the script will solve the problem (it can work also for the other maps of mine)

Quote
dm5: another missing texture on the same place as before, and lighting bugs around bevels. Gig, are we using the same version?

It's not a texture missing, it's a "brutal" graphic bug, i don't know why.
See it in spectator mode:



Wink
« Last Edit: January 28, 2014, 11:40:45 AM by Akom74 » Logged

...sorry for my English, i'm Italian... Tongue



Pages: 1 2 3 [4] 5 6 ... 8
  Print  
 
Jump to: