[Hank]

Plagman, on Feb 13 2011, 12:20 AM, said:

I'm all for solving this issue, but I don't know how. I need some time with a setup where the issue reproduces to be able to pinpoint exactly what's going wrong and work around it.

Then should that not be a new thread? Polymer + White Squares. Honestly. I'm a n00b here, sifting through 70 pages and collecting valid data is hard already, anyone after me will face an impossible task.
Also, since Polymer is a new engine using OpenGL, does anyone plan to contact AMD? It's a simple question, before 10.4 drivers; the games worked without white squares, after 10.5 drivers; the game shows white squares, the renderer is using OpengGL 2. Please advice what further info you require to solve our dilemma, code or whatever. Again just a suggestion. The worst thing is that AMD does not respond, but if they do - hey Bob's Your Uncle!

[Mikko]

Tea Monster, on Feb 13 2011, 06:43 PM, said:

If you present a valid counter argument that's fine. But if you issue trollish statements that seem intended more to incite than to persuade, then damn straight.

You made no valid arguments in the first place. You started trolling by slipping in an ATI-bashing comment. The idiot choir then joined you.

I had no intention of starting an NVIDIA vs. ATI war as I have used both GPUs and have nothing to complain about either one of them. Tea Monster and other trolls instead took the opportunity to bash ATI by making an utterly silly and simplistic suggestion.

Mblackwell said:

It also escapes him that Polymer is a new engine, not a 15 year old one.

Really? Several people suggested that to get around the "white boxes" issue is to get a better GPU. No references were made to other games; it was implied that people ought to purchase a new GPU in order to get the most out of a 15-year-old game. I didn't argue that ATI is better than NVIDIA or that ATI handles old games better. You're fighting strawmen, which usually results from poor reading comprehension.

[Hendricks266]

Hank, on Feb 13 2011, 12:27 PM, said:

Then should that not be a new thread? Polymer + White Squares. Honestly. I'm a n00b here, sifting through 70 pages and collecting valid data is hard already, anyone after me will face an impossible task.

There's a sticky thread for it in the problems forum.

Mikko_Sandt, on Feb 13 2011, 12:38 PM, said:

Several people suggested that to get around the "white boxes" issue is to get a better GPU. No references were made to other games; it was implied that people ought to purchase a new GPU in order to get the most out of a 15-year-old game.

I doubt that many of those suggestions were completely serious. Either way, there's no reason to be arguing back and forth about this. I've taken your points, Mikko, and I certainly hope others have. Plagman's tested ATI cards before and implemented workaround. Hopefully this could be fixed a similar way.

[Hank]

Hendricks266, on Feb 13 2011, 02:04 PM, said:

There's a sticky thread for it in the problems forum.

No shit. But it is not discussesd there.

b.t.w. I was refering to this post, to make a separate one.

Plagman, on Feb 11 2011, 07:48 PM, said:

Performance isn't what's being discussed here and way off-topic since it's pretty much on par with both vendors. The issue is trying to render characters and getting a white box instead. That only happens with the ATI driver.

As you can see the topic is already missed. Anyways, carry on.

[Plagman]

Mikko_Sandt, on Feb 13 2011, 05:57 AM, said:

So, one-sided ATI bashing is generally accepted but when someone points out the idiocy (and obvious bias) of some of the arguments you call that trolling?

The "arguments" were mostly humorous until you showed up. This was also off-topic and not helping to resolve the issue at hand, but much less annoying. Of course now that everyone walked right into your bullshit, the thread is ruined for good.

I find it awesome that everyone is so glad to jump in and make it sound like Polymer ATI users are constantly rounded up and sent to extermination camps, yet despite my various attempts no-one has been willing to provide me with a remote shell to an affected system so that I could actually fix the problems. But sure, maybe enough whining will make the problems vanish on their own.

Now, PLEASE only reply to this thread if you have constructive stuff to say about EDuke32, for this is what the thread is about. Feel free to open another dedicated, constructive thread about fixing the particular issues you're having if that's what you want to do. Stop doing this, though.

[Plagman]

Hank, on Feb 13 2011, 10:27 AM, said:

Also, since Polymer is a new engine using OpenGL, does anyone plan to contact AMD? It's a simple question, before 10.4 drivers; the games worked without white squares, after 10.5 drivers; the game shows white squares, the renderer is using OpengGL 2. Please advice what further info you require to solve our dilemma, code or whatever. Again just a suggestion. The worst thing is that AMD does not respond, but if they do - hey Bob's Your Uncle!

Yes, that's the thing to do and I've done it several times in the past in this context (I have direct contacts there). However, you can't just report it like that. You have to make sure you understand exactly what's going wrong first and provide a simplistic test case that shows the issue happening just by toggling a few bits of functionality. Just dumping an application (which implies having to set it up, understanding how to reproduce the problem and investigating it inside of a flurry of other stuff that's happening at the same time) on a team that's strapped for manpower and resources will usually net you a fair amount of nothingness. This level of support can usually only be afforded by customers with paid contracts or issues receiving wide end-user exposure. To be able to submit a helpful bug report, I need to understand the issue enough to have it worked around on my end first.

[Tea Monster]

I know that Maya has been struggling with problems, especially with ATI's 'professional' cards. Because of my tech support background, I was called in to help a fellow artist who'd bought a firepro card and was having problems. After trying every different driver we could find on the net, I rang the firepro helpdesk. Now bear in mind that this card is sold on the basis of being a professional product with professional support - the guy on the other end of the phone ummed and ahhed for a minute and asked ME what the version number was of the latest driver. He said he had no clue what to do next. He took my name and number. I called back a week later to check up on the case and they had no record of my call. He called his studio eventually and it turns out that they had a few firepro cards in to test. They had the same problems and they got no satisfaction either. They wound up sending the cards back in the end (as did the artist I was working with).

Now, Im NOT saying that to wind Mikko up or to bash ATI. I'm saying that as my personal experience. The reason I'm trotting all this out is to explain why it was that I suggested Plagman check in with other OSS projects. I don't think he's going to get anywhere with their tech division. If he can prove me wrong and he has an inside contact, then great.

[Awesomebob]

Plagman, on Feb 13 2011, 01:59 PM, said:

I find it awesome that everyone is so glad to jump in and make it sound like Polymer ATI users are constantly rounded up and sent to extermination camps, yet despite my various attempts no-one has been willing to provide me with a remote shell to an affected system so that I could actually fix the problems. But sure, maybe enough whining will make the problems vanish on their own.

I would totally be willing to be a lab rat for you. I have no choice right now but to use ATI till finances become a little less strained. Just PM me details.

[The Commander]

Plagman, on Feb 14 2011, 09:59 AM, said:

I find it awesome that everyone is so glad to jump in and make it sound like Polymer ATI users are constantly rounded up and sent to extermination camps, yet despite my various attempts no-one has been willing to provide me with a remote shell to an affected system so that I could actually fix the problems. But sure, maybe enough whining will make the problems vanish on their own.

I know I was suppose to help you with this, but due to the moving I have done recently I have been forced to mobile internet and the my upload speed would not be the best. Unless upload speed won't matter to much?

[Scott_AW]

I finally got a new system.

Acer
Win7 Premium
Nvidia 1gb ? don't know too much about it, but I can run Oblivion full detail with HDR, yay!
AMD 3gz Dual Core
4gb DDR3 Ram
1tb HD

So with it the first thing I tried was Black Shadow in Polymost 32bit and was surprised with how fast it ran in 800x600, averaging 200fps even with voxels converted into models. Nice.

But of course this would be really about eduke if I didn't give DukePlus and HDR in polymer a spin, which was also super nice. Glad I can finally experience it. I think the only slow downs I noticed was with excessive jibs and explosions all at once, but funny enough that just made me remember the old days.

[Micky C]

I don't think polymer has HDR?

[Tea Monster]

I think he meant to say 'HRP' but mistyped it (got a bit confuseled).

[Scott_AW]

Exactly, Oblivion=HDR, Eduke=HRP.

[Spirrwell]

How about in the eduke32 GUI when it first runs it asks you if you have an ATI card? Based on your answer it could do the change to the filtration or whatever it is that was discussed as a fix.

[Roma Loom]

Crap. I'm experiencing black mirrors on GF9600GT with latest drivers on eduke32 bild #1783. WinXP. Setting r_pr_lighting 0 makes them work again. Also at certain random angles mirrors could work for a bit of second then they go into black.

[Spirrwell]

Roma Loom, on Feb 16 2011, 07:28 AM, said:

Crap. I'm experiencing black mirrors on GF9600GT with latest drivers on eduke32 bild #1783. WinXP. Setting r_pr_lighting 0 makes them work again. Also at certain random angles mirrors could work for a bit of second then they go into black.

Heh, same here, although it happens no matter what OS I have.

[Ryan Cordell]

I can confirm that, having a GeForce GTX 275. It doesn't bother me too much, though.

[LeoD]

Crash report (missing texture, self-compiled svn1800):

"Error: file 'highres/screen/fonts/redfont13/2929_8b.png' does not exist"

(gdb) run
Starting program: c:\Programme\_No_Installer\Duke Nukem/eduke32.debug.exe -game_dir DukePlus -cfg DukePlus/dp_custom.cfg
[New Thread 2804.0xf90]

Program received signal SIGSEGV, Segmentation fault.
0x004daa79 in defsparser (script=<value optimized out>) at src/defs.c:555
555 if (col->f < alphacut) { ftd[i*ysiz+j] = 255; continue; }

[Gambini]

Will, anytime, the incompatibility issues with 8bits be fixed?

You know, sprites stuck into walls (decals) tend to clip behind them. And sprites that are placed very near to other sprites, clip eachother. Also there´s this old problem with parallax fake ror doors that work in any renderer but polymer.

[Skulldog]

Can you add a way to select which Folder to use as Autoload Folder? Like the one used for Custom game directory. So that a folder for Original hrp can
be loaded or the one with polymer hrp can be loaded.

[TerminX]

No. Autoload is deprecated and should probably be removed altogether.

[Hank]

TX, on Mar 8 2011, 12:06 PM, said:

No. Autoload is deprecated and should probably be removed altogether.

Is there one or two wiki pages, or a blog, that summarizes all those Eduke32 changes? The only reason I will know this (No more Autoload) is, because I happened to be home. In two days this post will be lost with the other 2000 posts. Just a thought.

[Master Fibbles]

TX, on Mar 8 2011, 12:06 PM, said:

No. Autoload is deprecated and should probably be removed altogether.

When did that happen? I need to start paying more attention...

[Danukem]

Mr.Flibble, on Mar 8 2011, 09:18 AM, said:

When did that happen? I need to start paying more attention...

It's not something that "happened", TX is saying (more or less) that autoload was a bad idea and he wishes he hadn't added it in the first place.

[Skulldog]

TX, on Mar 8 2011, 12:06 PM, said:

No. Autoload is deprecated and should probably be removed altogether.

Remove it , that would work for me.

[Hendricks266]

TX, on Mar 8 2011, 11:06 AM, said:

No. Autoload is deprecated and should probably be removed altogether.

What is better?

[TerminX]

Putting files in their own separate mod dirs and loading them using the included drop-down selector in the startup window.

[Danukem]

But if you remove it now, a whole generation of Duke players who strained their tiny brains to understand the concept of the autoload folder will think the HRP is broken and will flood the forums asking how to make it work again.

Nah, just kidding. There aren't enough Duke players left to flood the forums.

[TerminX]

DeeperThought, on Mar 8 2011, 11:37 PM, said:

Nah, just kidding. There aren't enough Duke players left to flood the forums.

You'd be surprised. Server logs for dukeworld.duke4.net show 5000+ hits on eduke32_current.zip so far this month and 26500+ hits on stuff within the synthesis directory. I have no idea what these numbers would look like when adjusted for non-unique visitors but it's still going to be a pretty decent number.

Edit: WGR2 is doing pretty well too. WGR2_SIEGEBREAKER.zip is getting a download at least every couple of minutes, to the tune of more than 25000 log entries referencing it so far. Good job!

Edit 2: looks like fissile.duke4.net has moved 80 gigs of data so far this month.

[Skulldog]

TX, on Mar 9 2011, 12:17 AM, said:

Putting files in their own separate mod dirs and loading them using the included drop-down selector in the startup window.

Then you would have to put Hrp packs in each mod dierectory or the main dierectory with no way to turn if off without removing the hrp files.
That would make the eduke directory 50 gigs or so. It would be better to remove checkbox and use a combobox with option of None, or any folder in
the main directory. Add a command for users and mod devs -hrp_dir None or -hrp_dir [dir name] Autoload is good thing to have.