Thursday, January 3rd 2013

AMD to Fix GCN Latency Issues with Driver Updates

Last month, an investigative report by The TechReport found out that despite being faster, AMD's Radeon HD 7950 graphics card isn't "smoother" than NVIDIA GeForce GTX 660 Ti, in that shows signs of higher frame-delivery latency, a theory proven by high-speed camera recordings. Over the holiday, AMD's David Baumann responded in discussions around the web talking about the issue, in which he put AMD's stand.

Apparently, AMD Catalyst drivers still have refinement left in working perfectly with GPUs based on the Graphics CoreNext (GCN) architecture. Baumann explained that GCN, and AMD's older Very Long Instruction Word (VLIW) architectures feature fundamentally different memory management, and drivers that make the most of it are still a work in progress. Baumann stated "Over the early part of the year you'll see a few driver updates help this across a variety of games."

He continued "Additionally, when we switched from the old VLIW architecture to the GCN core there was a significant updates to all parts of the driver was needed - although not really spoken about the entire memory management on GCN is different to prior GPU's and the initial software management for that was primarily driven by schedule and in the meantime we've been rewriting it again and we have discovered that the new version has also improved frame latency in a number of cases so we are accelerating the QA and implementation of that."
Source: The TechReport
Add your own comment

95 Comments on AMD to Fix GCN Latency Issues with Driver Updates

#26
amdftw
I have no latency issues with my 7970.
But i will happy if any performance boost come again.
GTX680 ass kicked for a while, with new driver NV again fall deeper and deeper...:laugh:
Posted on Reply
#27
TRWOV
Jstn7477Yeah, I can't even overclock it anymore now after repasting it because it will suddenly shoot up to 95c if I add a little voltage without clock increase and the PC blanks out. Idle temp is around 40c and BOINC still runs it at 55-60c but fire up Far Cry 3 and the card practically reaches meltdown after a minute or two.
Have you tried to change the screws? I've had to change some on my own because the standard screws were longer than needed and the die had poor contact with the heatsink. Even 1/10 of a mm can make a difference in cooling performance.
Posted on Reply
#28
qubit
Overclocked quantum bit
So, the GCN cards have been out for over a year and AMD is only now just starting to get a grip on the drivers? :eek: That's rubbish. The drivers should be pretty much optimised on the day of release, with only a few bugs here and there to fix. They're certainly not getting my money, that's for sure.

I remember when I compared my then current HD 4870 512MB with an old GTX 8800 I got used three years ago, how much smoother and more responsive the nvidia card generally felt even though performance was lower. I see that nothing's changed and I'm sticking to nvidia.
Posted on Reply
#29
Ravenas
They need to fix the overdrive features... Occasionally, its gets buggy and you have to uninstall the drivers. On top of that use driver sweeper. Reinstall the driver suite. It works again. AMD needs to fix this.
Posted on Reply
#30
Tonim89
Am I the only Radeon user who never noticed any kind of sttutering?
Posted on Reply
#31
Ferrum Master
RavenasThey need to fix the overdrive features... Occasionally, its gets buggy and you have to uninstall the drivers. On top of that use driver sweeper. Reinstall the driver suite. It works again. AMD needs to fix this.
They haven't fixed it since I owned Radeon 8500/9200 oh pardon... 2600XT?, oh 3850?, oh 4850? nope... was it ever been working like it should? :D
Posted on Reply
#32
cadaveca
My name is Dave
SK-1Well kudos to The TechReport for getting AMD off their butts.
I doubt it was TechReport alone. the 12.11 betas with big performance boosts say hello, nice to meet you, do you like my memory management?


That's why, I think, this beta is so long in approval...they have to test literally EVERYTHING, since such a change can drastically affect how well things run.


If it's not...well...maybe it WAS them. Funny though, since Microstutter has been complained about for years now, and AMD has finally admitted to it?

That is literally what microstutter is, large variations in render times, which many said they could not detect, so AMD ignored it.

What Tech Report did very well was present this in a way that was very understandable, and with benchmarks. That deserves a pat on the back, for sure.
Posted on Reply
#33
Widjaja
qubitSo, the GCN cards have been out for over a year and AMD is only now just starting to get a grip on the drivers? :eek: That's rubbish. The drivers should be pretty much optimised on the day of release, with only a few bugs here and there to fix. They're certainly not getting my money, that's for sure.

I remember when I compared my then current HD 4870 512MB with an old GTX 8800 I got used three years ago, how much smoother and more responsive the nvidia card generally felt even though performance was lower. I see that nothing's changed and I'm sticking to nvidia.
I agree, regardless of new architecture the drivers should be at an optimal performance without compromises such as noticeable image corruption which has lead to pointless RMAs for some people.

Since I am noticing large improvements since the 12.8s I'm sticking to AMD.
Was pretty close to going to nVidia though near the end of 2012 despite my last nVidia card was a POS.
Posted on Reply
#34
m1dg3t
With all the driver problems? ATi has/had at least they never put out a self destruct driver. :o

Good job to the people who presented the issue and good job to ATi for stepping up. GCN looks like it's only gonna get much better in the future :rockout:
Posted on Reply
#35
Nordic
qubitSo, the GCN cards have been out for over a year and AMD is only now just starting to get a grip on the drivers? :eek: That's rubbish. The drivers should be pretty much optimised on the day of release, with only a few bugs here and there to fix.
Since when is anything pretty much optimized on release. I will only use games for example. How many patches on average do AAA titles have in their first month of release. Skyrim had 3.
Posted on Reply
#36
Eagleye
From TR`s own benches, the 7970 GE is faster and smoother (99% sites confirm this including TR). The 7870 and 7850, 7770 or 7750 is not affected. The only card TR stating is effected is the stock speed 7950 @ 925mhz (overclocked ones have not shown to have problems). The 660TI was the highest clocked card tested against. The games tested were few and BF3 and other major games have not been shown to have any problems but 2 games which are BL and Skyrim (Skyrim doesn’t count since it was run without VSync which breaks the game), so we can assume AMD is faster and smoother apart from the few cherry picked benches TR have shown until otherwise proven or shown. What people forget is; nvidia were not smoother in all the few games tested :banghead:

On TR`s site, Amd 6 series were smoother than the nvidia 5 series according to the benches they have, yet we had people saying the nvidia cards were smoother? So which is it? I think it’s all in the head, or people imagining stuff.:laugh:
Posted on Reply
#37
eidairaman1
The Exiled Airman
EagleyeFrom TR`s own benches, the 7970 GE is faster and smoother (99% sites confirm this including TR). The 7870 and 7850, 7770 or 7750 is not affected. The only card TR stating is effected is the stock speed 7950 @ 925mhz (overclocked ones have not shown to have problems). The 660TI was the highest clocked card tested against. The games tested were few and BF3 and other major games have not been shown to have any problems but 2 games which are BL and Skyrim (Skyrim doesn’t count since it was run without VSync which breaks the game), so we can assume AMD is faster and smoother apart from the few cherry picked benches TR have shown until otherwise proven or shown. What people forget is; nvidia were not smoother in all the few games tested :banghead:

On TR`s site, Amd 6 series were smoother than the nvidia 5 series according to the benches they have, yet we had people saying the nvidia cards were smoother? So which is it? I think it’s all in the head, or people imagining stuff.:laugh:
thing is not everyone experienced microstutter, but from sounds the driver probably just needs to be built from the ground up and a separate set be made for GCN and VLIW
Posted on Reply
#38
Prima.Vera
I only noticed big stutter on crossfire configs with my 5870/5850. Even if I was running 60+ fps I can feel sometime that is getting sluggish for a fraction of a sec, like the movement is choking, Really annoying especially on car games.
Posted on Reply
#39
qubit
Overclocked quantum bit
EagleyeSo which is it? I think it’s all in the head, or people imagining stuff.:laugh:
How can it all be in people's heads when AMD's David Baumann confessed to this problem himself, saying it was due to driver issues with GCN?
Posted on Reply
#40
EarlZ
At least they are admitting that they have terrible drivers, though this issue has been around for the longest of time. This is actually good news if they can fix this, gives nvidia more reason to lower their card prices :laugh:
Posted on Reply
#41
Jadawin
Now, I don't care about AMD vs. Nvidia or AMD vs. Intel or any of that crap... but if I need a high-speed camera to detect latencies I can't notice on my own and which therefore didn't bother me at all, wouldn't a fix for this "problem" be just as undetectable for me without a high-speed camera? An undetectable fix for a undetectable problems (for most people)?
Posted on Reply
#42
EarlZ
Maybe what they ment is that its easily detectable by a high speed camera that AMD's rendering interval is pretty bad. at 60fps with a dual 7950 I can easily feel/notice micro-stuttering even on a single 7950.
Posted on Reply
#43
Fluffmeister
qubitHow can it all be in people's heads when AMD's David Baumann confessed to this problem himself, saying it was due to driver issues with GCN?
Denial and brand loyalty is a powerful thing.

Even linking to original article induced rage.
Posted on Reply
#44
Jadawin
EarlZI can easily feel/notice micro-stuttering even on a single 7950.
Well, THEN it is a real problem for you, no doubt about that. I just wondered because if it would be a huge big problem for everybody, it would have been noticed earlier and would be headlines everywhere. So it's a real problem for people with real good perception.
Posted on Reply
#45
Calin Banc
EagleyeSkyrim (Skyrim doesn’t count since it was run without VSync which breaks the game)
Skyrim was stuttering on my 6970 when the load on the card dropped into some scenes due to v-sync and I think the same issue I'm having with 7950, but a little bit less annoying.

My last two nVIDIA cards were a couple of GTX470 (not in SLI). I believe that in most of the games that I've put them through, there was some amount of stutter if PhysX or other Cuda functions (JC 2 for instance) were ON. With 5850 + GT240 for PhysX it was smooth as butter, so I suppose it was a problem regarding resource allocation or something like that. BFBC 2 run with less stutter on my 5850 than on GTX470 although the last one gave higher FPS, especially with AA.
Other "weird" experiences were thanks to Ambient Occlusion, in some games with that option activated, it started to lag or stutter like it was a 20-25FPS or worse even though the actual FPS hit 33-35FPS or more, in others, there was no problem. A powerful card/GPU, in the exact same game, gave a better experience where others lag although the difference shouldn't be that high.

Right now with a 7950@1170/1600MHz I feel some lag or stutter in Sleeping Dogs with Extreme AA, The Witcher 2 with US, Skyrim at times, BF 3 under certain settings/maps (6970 was unplayable on Ultra despite it's 40-45FPS), GTA 4, MP 3 with MSAA, Far Cry 3 under 60FPS, Hitman Absolution with high AA levels etc. I think the problem hangs on AMD/nVIDIA heads but also on the game developers hands and how well they optimize the code. It's not that big of a tragedy as long as the stutter can be avoided by turning of a graphic or other kind of option.

BTW, nVIDIA also had a "stutter problem" with gtx670 and adaptive v-sync I think.

Thankfully AMD is working on a solution for it.
Posted on Reply
#46
Tonim89
JadawinWell, THEN it is a real problem for you, no doubt about that. I just wondered because if it would be a huge big problem for everybody, it would have been noticed earlier and would be headlines everywhere. So it's a real problem for people with real good perception.
I keep saying this all time. If it's a big deal, why would it come to discussion only 7 or 8 months after the card was released?

Of course the 7950 had bigger latencies, it was stuck at 850/925 MHz, while the 660 Ti can boost to almost 1100 MHz. The card with higher fps will have the lower latencies...

How can the fps count be higher if the frames comes delayed to the screen? OC Both cards and measure the timings again.
Posted on Reply
#47
Calin Banc
Well, you can have 30FPS with 33,3ms needed for render every frame or 45FPS with render time from 5ms all the way to 50ms and so on. MHz/MHz performance is not equal between AMD and nVIDIA, so it's not that the problem. It MAY be, if the core/architecture needs X MHz in order to give constant frame times rate and the card was clocked lower than that to begin with.
Posted on Reply
#48
3870x2
Prima.VeraI only noticed big stutter on crossfire configs with my 5870/5850. Even if I was running 60+ fps I can feel sometime that is getting sluggish for a fraction of a sec, like the movement is choking, Really annoying especially on car games.
I know exactly what you mean, where it kinda slows down (like rubber banding or something) then speeds right back up. This is not limited to ATI, I first noticed it on Need for Speed games on my x1300 then 7900GS. This is probably an example of coding being the culprit.
Posted on Reply
#49
Widjaja
3870x2I know exactly what you mean, where it kinda slows down (like rubber banding or something) then speeds right back up. This is not limited to ATI, I first noticed it on Need for Speed games on my x1300 then 7900GS. This is probably an example of coding being the culprit.
I had this with the 12.8 drivers and my current card in NFSMW 2012 and to a degree in Saints Row The Third.
Disappeared in 12.11 BETA 11.
Posted on Reply
#50
xenocide
I'm very pleased AMD is finally addressing their Graphics Drivers. I went with a Kepler GPU because after 2 AMD cards and constant driver issues I wanted something that just worked and Nvidia offered that. :/
Posted on Reply
Add your own comment
Apr 19th, 2024 06:17 EDT change timezone

New Forum Posts

Popular Reviews

Controversial News Posts