Jump to content
The Dark Mod Forums

OpenGL perf. on AMD/ATi gpu's: WIP fix.


Bikerdude

Recommended Posts

  • Replies 252
  • Created
  • Last Reply

Top Posters In This Topic

He did, with the previous driver 40, latest driver 47.

Update: "40" was the FRAPS FPS value, not the Doom3 console value. I checked now with FRAPS and nothing is changed.

Maybe we must decide to avoid FRAPS and use the Doom 3 FPS counter.

Edited by lowenz

Task is not so much to see what no one has yet seen but to think what nobody has yet thought about that which everybody see. - E.S.

Link to comment
Share on other sites

Yeah, the small boost from changing the executable name means that the drivers still aren't profiling TDM as well as Chronicles of Riddick

but at least it's not a massive difference (though folks who have Crossfire may not have any choice and will probably need to rename the executable).

 

I guess we are still awaiting confirmation about which release was supposed to have the fix \ improvement.

Please visit TDM's IndieDB site and help promote the mod:

 

http://www.indiedb.com/mods/the-dark-mod

 

(Yeah, shameless promotion... but traffic is traffic folks...)

Link to comment
Share on other sites

the equivelent nVidia card (GTX660) would be flat out @60fps at your highest setting. I have emailed my contact to see what version the driver team will include The Darkmod fix in.

 

I don't understand why you keep saying Nvidia runs at 60 fps on max out settings, when in reality it does not. I have GF 670GTX and I posted my results in this thread. It's nowhere near to be close to 60 fps.

Link to comment
Share on other sites

I don't understand why you keep saying Nvidia runs at 60 fps on max out settings, when in reality it does not. I have GF 670GTX and I posted my results in this thread. It's nowhere near to be close to 60 fps.

 

We don't have many nvidia results, it's true. Just two results from Biker at 60fps (using different nvidia cards) and one from you (not 60 fps), which shows that their cards vary too. Biker is the only one who can compare the cards under controlled conditions, but his result is enough to know that there's something that could be done other than speeding up the cpu side.

Link to comment
Share on other sites

We don't have many nvidia results, it's true. Just two results from Biker at 60fps (using different nvidia cards) and one from you (not 60 fps), which shows that their cards vary too. Biker is the only one who can compare the cards under controlled conditions, but his result is enough to know that there's something that could be done other than speeding up the cpu side.

 

That's not how comparison is done o.O You can't just have 1 guy saying Nvidia runs at 60 fps, and go bug AMD to make a questionable "fix".

 

By looking at the test results, you can see the weaker the CPU, the lower the fps. Doom 3 engine was optimized for Intel and Nvidia. So perhaps it's the combination of powerful Intel CPU and Nvidia GPU that allows Bikerdude to have 60 fps. Although I'd love to see a screenshot made with that 60 fps counter.

Edited by motorsep
Link to comment
Share on other sites

That's not how comparison is done o.O You can't just have 1 guy saying Nvidia runs at 60 fps, and go bug AMD to make a questionable "fix". By looking at the test results, you can see the weaker the CPU, the lower the fps. Doom 3 engine was optimized for Intel and Nvidia. So perhaps it's the combination of powerful Intel CPU and Nvidia GPU that allows Bikerdude to have 60 fps. Although I'd love to see a screenshot made with that 60 fps counter.

 

That is how tests are done properly. You restrict the comparison to one variable, in this case by using the identical system and situation and varying only the card between tests. And it's not just "1 guy" bugging AMD, it's a guy who provided specific, detailed and relevant information, so they could rep it themselves which makes the number of reporters irrelevant. Biker knows what he's talking about and there's nothing he needs to prove to us with screenshots.

 

Yes, CPU will have an effect too. TDM is neither CPU nor GPU-bound, or rather, it's both when it's not getting full FPS because it's single-processor and the application has to finish its job before drawing starts. Speeding up either CPU or the drawing process will push FPS back up, it's not a case of one process being the bottleneck.

Link to comment
Share on other sites

eeh, I disagree. If that was the case, there would be no QA departments. One person with one specific hardware setup could test it all. However, software is tested on wide variety of hardware. So you don't just take one isolated case and proclaim: "That's it, AMD has issues with drivers because TDM run at 60fps on all Nvidia GPUs just because it does for me on my PC!"

 

sUk1jwH.jpg

Edited by motorsep
Link to comment
Share on other sites

eeh, I disagree. If that was the case, there would be no QA departments. One person with one specific hardware setup could test it all. However, software is tested on wide variety of hardware. So you don't just take one isolated case and proclaim: "That's it, AMD has issues with drivers because TDM run at 60fps on all Nvidia GPUs just because it does for me on my PC!"

 

 

I'm with Steve here: If you want to test the GPU, you keep everything else (monitor, CPU, motherboard, memory, HDD) the same and swap the GPU and driver. If it then becomes much slower - you have found a culprit.

 

Testing on a different PC would not prove anything in this case.

Edited by Tels

"The reasonable man adapts himself to the world; the unreasonable one persists in trying to adapt the world to himself. Therefore, all progress depends on the unreasonable man." -- George Bernard Shaw (1856 - 1950)

 

"Remember: If the game lets you do it, it's not cheating." -- Xarax

Link to comment
Share on other sites

I'm with Steve here: If you want to test the GPU, you keep everything else (monitor, CPU, motherboard, memory, HDD) the same and swap the GPU and driver. If it then becomes much slower - you have found a culprit.

 

Testing on a different PC would not prove anything in this case.

 

You aren't being serious stating this, are you?

 

If that's the case (all rigs should be the same), why is it all test done on AMD have completely different PCs ? Where in the real-world did you see gamers with all the same rigs?

 

If you want to prove that Nvidia outperforms AMD, you'd need one rig, and several Nvidia and AMD GPUs. All tested on the same rig. And if there is consistent difference in performance between comparable GPUs, then you can claim that AMD has screwy drivers.

 

How about for once post a screenshot, where Nvidia 660GTX shows 60 fps on max settings and 40 fps on AMD R9?

Link to comment
Share on other sites

You are talking to eachother, but you're not communicating here. What Tels and Steve are talking about is proving that a specific component is causing an issue in a certain context. The way to go about this is indeed to leave all variables the same and change the components to see if that resolves the issue. However, what motorsep is getting at is that testing has to be done on multiple rigs if you want to make a claim about the behaviour of a processor in general, meaning independent from the context.

The issue with the AMD test as it has been done, is the following: the performance drop might be caused by the interaction of the CPU with the rest of the components. ie. for a different GPU/HDD/Whatever setup, this problem might be resolved, and that would point towards a problem with the context rather than the specific component. For instance, the GPU might have a bad time interacting with a specific type of processor but not with another, and in this case the problem would be with the GPU rather than the CPU. So the test is not conclusive.

You can call me Phi, Numbers, Digits, Ratio, 16, 1618, or whatever really, as long as it's not Phil.

Link to comment
Share on other sites

Now you're making up stuff. No one made any claims regarding the performance of Nvidia vs AMD in general, or about any cards other than the two specified in the OP, in the situation specified in the OP. Biker found a specific problem with two specific cards in a specific situation with a valid test for that situation. A problem that AMD agreed was worth their attention.

 

Regarding testing, you are conflating two different situations. It takes a single test find a problem. It takes multiple tests to prove it's fixed. I find it hard to believe you don't understand that really, so I'm out of this asinine conversation.

 

Edit: ninja'd! My reply was to motorsep's post.

Edited by SteveL
Link to comment
Share on other sites

Calm down guys, I have a twin PC with a 750ti (but Win8.1) so I can test for NVidia HW too ;)

Same specs.

Edited by lowenz

Task is not so much to see what no one has yet seen but to think what nobody has yet thought about that which everybody see. - E.S.

Link to comment
Share on other sites

You aren't being serious stating this, are you?

 

If that's the case (all rigs should be the same), why is it all test done on AMD have completely different PCs ? Where in the real-world did you see gamers with all the same rigs?

 

If you want to prove that Nvidia outperforms AMD, you'd need one rig, and several Nvidia and AMD GPUs. All tested on the same rig. And if there is consistent difference in performance between comparable GPUs, then you can claim that AMD has screwy drivers.

 

How about for once post a screenshot, where Nvidia 660GTX shows 60 fps on max settings and 40 fps on AMD R9?

 

Why is this even being debated when AMD have 'confirmed' the issue themselves?

 

As a professional tester, there are cases that require testing across multiple devices, and others that are better off isolated to a control system where you swap out the suspected hardware / software to see what is going on. In this case, the latter is perfectly fine.

Link to comment
Share on other sites

Suggestion: can anyone code and add to TDM a script to write down (on a txt file), every X ms, the "com_showFPS" return values so we can make some serious ( :laugh:) Nobel-oriented maths??

Edited by lowenz

Task is not so much to see what no one has yet seen but to think what nobody has yet thought about that which everybody see. - E.S.

Link to comment
Share on other sites

Calm down guys, I have a twin PC with a 750ti (but Win8.1) so I can test for NVidia HW too ;)

Same specs.

Update!

 

FPS range: 63-68 (in screenshot). With or without adding TDM to NVidia Doom 3 profile, there's no difference. The TDM folder was simply *cloned*, so every setting is the same.

Verdict: the good Radeon 7850 is outperformed by a lesser GeForce 750ti.

Edited by lowenz

Task is not so much to see what no one has yet seen but to think what nobody has yet thought about that which everybody see. - E.S.

Link to comment
Share on other sites

  • 3 weeks later...

Just got an update from my contact -

The Darkmod Issue is still being worked on. While the issue has been reproduced and identified, the solution is pretty complex and is taking longer than initially estimated. I’ve not been given a firm date for the fix but it should be in Q1 2015.
Link to comment
Share on other sites

I have sh*ty Samsung NP535U3C laptop, but still, it should be capable to run TDM, I belive.

No one wrote that it is only Windows problem, so here it goes, linux:

  • Fps - 6-7
  • System - AMD A4-4355M APU with Radeon™ HD Graphics - Trinity [Radeon HD 7400G], DDR3 4GB (onboard), 120GB SSD,
  • Driver - Gallium 0.4 on AMD ARUBA, 3.0 Mesa 10.5.0-devel, GLX 1.4, GLU 1.3
  • Resolution - 1280x720
  • Settings - seta com_videoRam "512", seta com_showFPS "1", seta com_compressDemos "0", seta image_downSizeLimit "64", seta image_downSize "1", seta r_swapInterval "1", seta r_useIndexBuffers "0", seta r_customHeight "720", seta r_customWidth "1280", seta r_mode "5", seta tdm_wideScreenMode "6" - rest unimportant or default.

I know, my issue is probably related to not great APU, not perfect drivers (which are reported to be better than proprietary ones).

But i really want to try out TDM...

 

And my xorg if anyone interested:

Section "Device"																				  
	Identifier  "Radeon 7400G - AMD A4-4355M"
	Driver "radeon"
	#Option "MigrationHeuristic"  "greedy"
	Option  "SWcursor"			  "off" #software cursor might be necessary on some rare occa$
	#Option  "EnablePageFlip"		"on"  #supported on all R/RV/RS4xx and older hardware, and $
	Option  "Accel"				 "on"  #Enables or disables all hardware acceleration. The default is on except for HAWAII.
	Option  "AccelMethod"		   "EXA" #valid options are XAA, EXA and Glamor. Default value$
	Option  "RenderAccel"		   "on"  #enabled by default on all radeon hardware
	Option  "ColorTiling"		   "on"  #enabled by default on RV300 and later radeon cards
	Option  "ColorTiling2D"		 "on"
	Option  "EXAVSync"			  "off"  #default is off, otherwise on. Only works if EXA activ$
	Option  "EXAPixmaps"			"off"  #when on icreases 2D performance, but may also cause $
	Option  "AccelDFS"			  "on"  #default is off, read the radeon manpage for more inf$
	Option  "SwapBuffersWait"	   "off"
	Option "DynamicClocks" "true"
	Option "DRI"			"true"
	#Option "AGPMode" "8"
EndSection

 

Ah, and i've found solution for some issue: blank screen

"can't load image". As it is related to S3TC, solution is [link]:

 

force_s3tc_enable=true steam

 

to test (replace steam with ./thedarkmod.x86).

to make it permanent, use dricon and select 'Enable S3TC texture compression even if software support is not available'.

Edited by San
Link to comment
Share on other sites

  • System - AMD A4-4355M APU with Radeon HD 7400G

There is your potential issue straight away I'm afraid, the 7400G (160/192 cores) is about the same as a nVidia 9600GS(48 cores). You will need to lower the screen res to 1024*768 and drop all settings in TDM to thier lowest, then you should manage around 20fps.

 

Once AMD comes out with a fix for our little mod in 2015, all AMD/ATi user should see a nice FPS improvement.

Link to comment
Share on other sites

Once AMD comes out with a fix for our little mod in 2015, all AMD/ATi user should see a nice FPS improvement.

 

I am keeping my fingers crossed, because AMD sucks in all idTech 4 / 5 games :/

 

By the way, how did you get in touch with AMD about TDM issue? I might need to get with them about our engine performing horribly under Win 8 / 8.1 on some AMD GPUs.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


  • Recent Status Updates

    • Ansome

      Finally got my PC back from the shop after my SSD got corrupted a week ago and damaged my motherboard. Scary stuff, but thank goodness it happened right after two months of FM development instead of wiping all my work before I could release it. New SSD, repaired Motherboard and BIOS, and we're ready to start working on my second FM with some added version control in the cloud just to be safe!
      · 0 replies
    • Petike the Taffer  »  DeTeEff

      I've updated the articles for your FMs and your author category at the wiki. Your newer nickname (DeTeEff) now comes first, and the one in parentheses is your older nickname (Fieldmedic). Just to avoid confusing people who played your FMs years ago and remember your older nickname. I've added a wiki article for your latest FM, Who Watches the Watcher?, as part of my current updating efforts. Unless I overlooked something, you have five different FMs so far.
      · 0 replies
    • Petike the Taffer

      I've finally managed to log in to The Dark Mod Wiki. I'm back in the saddle and before the holidays start in full, I'll be adding a few new FM articles and doing other updates. Written in Stone is already done.
      · 4 replies
    • nbohr1more

      TDM 15th Anniversary Contest is now active! Please declare your participation: https://forums.thedarkmod.com/index.php?/topic/22413-the-dark-mod-15th-anniversary-contest-entry-thread/
       
      · 0 replies
    • JackFarmer

      @TheUnbeholden
      You cannot receive PMs. Could you please be so kind and check your mailbox if it is full (or maybe you switched off the function)?
      · 1 reply
×
×
  • Create New...