[Contest Entry] pl_zig

[Rexy]

I can't load it. Engine hunk overflow error crashes TF2 when trying to. But apparently everybody else on for gameday can.... >____>

Time for a video card upgrade. What card are you using now?

 

[Cerious]

One of the reasons no one here is big on fpsb. I'm not big on maps like this either but i do love your cave's with those crystal's.

Are you kidding me? FPSB falls over on their knees for Arhurt and aerospace, lol.

 

[grazr]

I'm not sure what you are responding to Cerious.

One of the reasons no one here is big on fpsb. was a response to the rating system implemented into FPSB, and the childish behaviour that surrounds it, revenge ratings etc. If you read that quote containing 2 sperate sentances as one thing that is, and looks like you did (when they were two seperate points).

 

[A Boojum Snark]

Time for a video card upgrade. What card are you using now?

While I am using an older card, are you sure that's it? I can never find any info on just what the error means, niether today nor in the past when I've ran into it. Either way, I've never had it before in TF2, so something is still different with your map, and if I had it someone else will.

 

[Rexy]

A member of my gaming community has had the same issue, because the card he is using is ancient. This map nearly maxes facelights (LDR worldlights 8186/8192 720368/720896 (99.9%) VERY FULL!)and I suspect that cards without enough memory can't handle loading it. I'm typically not too concerned with this error, as it seems that it's only effecting players with older video cards. I feel bad that some can't play it, but it's not worth sacrificing quality to cope with older cards. Performance-wise, I still feel like the map has plenty of fps for players with older video cards, and I do speak from experience on that matter. Try playing TF2 with a 5200 geforce for a whole year, then get a good upgrade.

If anyone has stuff to say about this map, please let me know! I crave feedback!

I'll get some beta 6 screenshots up in a while.

 

[Lepock]

There are some colored hallways where its hard to tell which team people are on. I understand the ambiance of the lightning and it looking nice. Do you think you could point the colored lighting more toward the walls and put regular lighting in the center of the hallways without messing up the looks? Not that big of a deal though.

 

[Rexy]

There are some colored hallways where its hard to tell which team people are on. I understand the ambiance of the lightning and it looking nice. Do you think you could point the colored lighting more toward the walls and put regular lighting in the center of the hallways without messing up the looks? Not that big of a deal though.

I can't point the colored lighting away because the colored lighting is produced from self-illuminating textures (meaning they are omnidirectional). What I CAN do is decrease the brightness of them, and add in more white lighting in places that need it. Specifically, the inside areas at D and just after C and right before B. That seemed to be the major complaint.

As for time for each cap, how did those feel? Not enough? Too much? I may add 30 seconds to each cap and 1 minute to the last cap.

 

[Apom]

I can't point the colored lighting away because the colored lighting is produced from self-illuminating textures (meaning they are omnidirectional).

A question about that, if I may: what method did you use for your self-illuminated textures? A self-illum shader in the .vmt? This one does not seem to properly light the world. Did you go for the fancy .rad settings?

 

[Rexy]

A question about that, if I may: what method did you use for your self-illuminated textures? A self-illum shader in the .vmt? This one does not seem to properly light the world. Did you go for the fancy .rad settings?

Lol, I guess you can call them 'fancy' rad settings, but they aren't that complicated. I'll spit out a line on one specifically. It goes like this:

In lights.rad:

MQN/GlowMetal01B 65 216 255 150

and so on, for textures that are self-illuminating, with the first 3 values representing the RGB color and the last determining brightness of the colored RGB values. The rest of the work deals with making a NICE texture, setting the parameters correctly in the .vmt and having a working alpha channel.

Rgr?

If the textures are popular enough, I suppose I could pack the entire set for this map and release them, there's like 80+ custom textures. I'm willing to part with so that my next project has entirely new materials. I also fear that they could go into the wrong hands...more poorly made 'orange maps' with my textures? I haven't decided.

 

[A Boojum Snark]

This map nearly maxes facelights (LDR worldlights 8186/8192 720368/720896 (99.9%) VERY FULL!)and I suspect that cards without enough memory can't handle loading it.

I see. In that case I toss out the question: have you raised the lightmap scale on faces that don't need detailed shadows?

 

[Rexy]

I see. In that case I toss out the question: have you raised the lightmap scale on faces that don't need detailed shadows?

Previously, the lightmaps set were lower for quality purposes, but the plain fact is that the custom content of the map out-values the size of a map with lower lightmap scale. A major complaint was the size of the map. Uncompressed, beta 6 is 117 MB, beta 3? 123 MB that has nothing to do with lightmap scales. I simply can't afford to lower the lightmap scale anymore. In beta 6, all surfaces rendering lighting start at the default 16. I don't want that, but I don't have much of a choice. People don't like sitting in line for 10 minutes downloading maps. Personally, I wouldn't mind waiting in line to download a map if I know the quality is going to be really good, but not everyone feels the same way.

So in answer to your question, yes, the entire map has raised lightmap scales (16 to be specific, in all areas). About 100 mb of the map is custom content. :\

 

[grazr]

I think what Boojam was getting at is that lowering the lightmap scales to reduce system load is never a bad thing and on faces with no detail shadows (fully lit, fully shaded) faces, you can afford to lower the lightmap scale to 64. Whilst the rest of your map can have a better lightmap scale. The final point certainly looks like you could effectively implement lightmap scale optimisation significantly. There will be absolutely no difference in appearence.

 

[A Boojum Snark]

Just so we have the terminology right.... raising = larger number = lower quality. Not quite sure if you read my question the right direction. However, grazr had my point right. If there are places you can afford a bigger lightmap, it will reduce the filesize and maybe make your map accessible to everyone.

 

[Rexy]

Nah I understand, I'll be sure to work these things into the final release.

 

[Rexy]

Alright, I got everything taken care of and DONE. I am very satisfied to be done.

Contest Entry Link:
http://forums.tf2maps.net/showthread.php?t=4833

Download Link:
http://forums.tf2maps.net/downloads.php?do=file&id=924

From the readme:

Payload Zig

*************************************

Change log for beta 7/Final version:

*Complete revision of the environment lighting (looks better imo).
*Complete revision of lights.rad to remove team identity confusion in colored areas (blu players now appear blue in the red areas and vice-versa!).
*More skybox detail
*Complete revision of lightmap scale on ALL surfaces (optimization).
*Added more time to each control point (balance)
*Reworked mission briefing screen
*Added a new 'view' area outside of the alternate path to D.
*Added fog and skybox fog
*corrected the shadow_control control.
*Added another staircase to the last point for easier access to red spawn.

Change log for beta 5 closed, beta 5a and beta 6:

*Inclusion of 3-D skybox!
*Moved red spawn back into the crystal cave (balance).
*Reconfigured respawn wave times for each time and control point (balance).
*Added alternate route to D.
*Sectioning of 'outside' areas for fps improvements (previously was one big box! O.
*Added section specific 'crates' for cover.
*More dectional arrows so players are even less likely to be lost.
*Changed all doors to 'glass' doors, brush based func_doors that fit the style of the map more appropriately than the tf2 models.

Change log for beta 4:

*Slightly reduced red respawn time for cap A (balance)
*COMPLETE REVISION of the cave section (BRAND NEW), now twice as wide and tall, with better features and lighting, as well as better for team balance. See new screenshot
*Increased the spawn area for purple spawn (exploit fix)
*Changed spawn areas, purple spawn is now used by both teams, the tunnel from red spawn to cap B is obsolete and has been removed
*Fixed red and blu spawn orientation issues
*Fixed displacement issues in the cap C section
*Added pop-up signs! Players should not get lost now!
*Added large glass doors that open on captures. Also keeps players from going the wrong direction
*Reduced # of reflective textures (subtlety, I am working on it...)
*Fixed "No entry" hand for blu spawn 2/red spawn 1. Should be gone by the time blu owns the spawn
*Fixed blu spawn 2 ownership issue (no more suicide when changing class)
*Fixed some texture inconsistencies
*Added player clipping to ALL metal door frames (smoother for players)
*Overall "white" lighting added to colored areas to reduce team identity confusion.
*REDUCED FILESIZE (on request) and I'm working to make it smaller!

Change log from beta 3:

*Pushed back Cap B for balancing
*Solved problem of players getting stuck when riding ontop of the cart in the cavern.
*More optimization
*More detail
*More custom materials
*Map info

Change log from beta 2:

*added player clipping
*added detail, custom decals, props, stock decals
*added more areaportals for better optimization
*pushed back Cap B for better team balance.
*added 2 side tunnels to the cavern for better balance.

*************************************

Map Facts:

Compile Time: 52 minutes
Map Size: 121 MB - .bz2 Size: 44 MB
Custom Textures: 87 Unique Textures (including normal mapping), all developed by myself.
Build Time: Approx. 1 Month, 3 weeks (December 2008 - January 2008)

*************************************

Special thanks to tf2maps.net for the contest and encouragement to work my ass off!
Special thanks to Rad Lionheart for the Zig logo concept.
Thanks to all the admins and regulars of DuckSoup-Gaming.com who waited patiently in line to download each and every beta version of Zig, and then gave positive, useful feedback!

Play pl_zig on our TF2 Server:

Ducksoup-gaming.com: 66.150.164.129:27015

Thank you!

Big Rexy/Alex Kreeger
forcingace@gmail.com

 

[Rexy]

So I got this post on Zig's page on FPSB, and I'm wondering what you think about reworking the textures for smaller size...but I'm worried (as usual) about quality. My problem is in using the .vtf plugin for photoshop and that it was never fully understood what formats to use and why. I did some tests as well and I'm shaking my head that something as important as .vtf format could make such a difference...

I think I've worked out why the filesize is so large: Almost all of your custom textures are uncompressed and you are using formats with an alpha channel even for textures that don't need one.

Example: materials/MQN/Concrete01B.vtf

Original (BGRA8888): 1366kb (264kb with bzip2 at max)
Quick conversion to DXT1: 171kb (124 with bzip2 at max)

According to this page, what you should do is the following:
1) Use DXT1 for typical textures that have no transparency
2) Use DXT3 or DXT5 for typical textures with transparency, depending on what the gradients are like
3) Since your normal maps all (or most?) have no alpha channel, you can use BGR888 for them. Using DXT1/3/5 for normal maps can harm their quality, but it's still worth a try if you don't mind any quality loss that may occur

Based on the example I gave, you can reduce the bzip2 size by up to 30mb (assuming you DXT1 the normal maps. If you just BGR888 them, then it'll be about 19mb saved)

Please, please release a v2 that takes this into account. I'd love to try this on my server, but I don't know how patient my regulars will be regarding the file size :S

So what I'd like to know is how much will changing formats affect the quality of the textures. Previously, I felt filesize was a nasty nasty curse of custom content, but I'm upset with myself for not knowing the difference in formats.

Should I rework this map after the contest? And what about the naming thing? I hate putting v2 next to a map, that's just tacky...I'm wishing I wasn't so ready to finalize the map for the contest, I was thinking the maps had to be 'finished'.

So now I need some serious advice. What should I do about this? Forget it? Rework? And if so, what about the naming thing? I really need some advice here. I want to make this map as good as I can possibly make it, but I need advice.

 

[Icarus]

Normally you should compress everything that is not part of the HUD. The difference of quality is unnoticeable; the difference of filesize is phenomenal

Don't worry about naming, vilepicke had maps well into their 3rd and 4th revisions, as well as I. If the filesize difference is great enough, you could consider it worth a revision. not to mention you can work on that messy chokepoint right after CP1

 

[Rexy]

Normally you should compress everything that is not part of the HUD. The difference of quality is unnoticeable; the difference of filesize is phenomenal

Don't worry about naming, vilepicke had maps well into their 3rd and 4th revisions, as well as I. If the filesize difference is great enough, you could consider it worth a revision. not to mention you can work on that messy chokepoint right after CP1

I think you're right, I just don't like the v2 tag...I vehemently hate the idea of that...so what should I do? Suck it up or what?

 

[Rexy]

You guys are going to laugh at me, I just saved 70 mb....

*facepalms*

This makes the top ten list of things I SHOULD have known before I started...

Ok, so please please please, help me with the renaming thing.

 

[Icarus]

pl_zig2

This neatly presents that it is a revision of Zig. Anything else would confuse players.