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

#1
Eagleye
Most will have a different experience by just changing one component between two exact same systems; some people have problems and some don’t with exact same hardware. I can tell you I had a terrible experience with nvidia 670, it was a stuttering mess I tell you. EVGA mass called shipments back because of this (Google it), but that dont mean the same experience for all (as above). Just have a look at the forums below and realize that both are having problems.

Nvidia 600 series
https://forums.geforce.com/default/board/34/geforce-600-series/

AMD 7 series
http://forums.amd.com/game/categorie...454&entercat=y
Posted on Reply
#2
qubit
Overclocked quantum bit
by: Eagleye
I can tell you I had a terrible experience with nvidia 670, it was a stuttering mess I tell you. EVGA mass called shipments back because of this
If EVGA recalled the cards then it sounds like the stutters are due to faulty cards ie bad hardware rather than inherently bad GPUs or drivers.
Posted on Reply
#3
eidairaman1
by: 3870x2
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.
Goes back to NFSU and U2, certain points in race event maps they would stutter at the same point all the time
Posted on Reply
#4
erocker
I can confirm this issue... Though I found playing Guild Wars 2 to be smooth, here are my results which almost mirror TechReports findings.



Granted I get higher framerates with a 7970 oc'd. You can use FRAFS bench viewer to make these graphs using frametimes from FRAPS.

Also, the graph doesn't lie however my experience with this card and any game have been a good one and I don't find "smoothness" to be any different from using a Nvidia card.


Get FRAFS bench viewer here: http://sourceforge.net/projects/frafsbenchview/
Posted on Reply
#5
lopane
AMD is not for justin bieber(Love AMD)

No problem with my 7970 IceQ His, it never heats and yet I also amd fx 8350 and at all as I overclock to 4.6 ghz without any problems and does not heat up that much even with the fan original, the only problem AMD chips currently have drivers ...
Posted on Reply
#7
Widjaja
Wondered where this read out came from.

I looked at the youtube latency thing and saw nothing that I can discern as micro stutter.
If so both cards had it and none of them seemed to be worse than the other.
Plus Skyrim does have a micro stutter on Windows 7 even when I use an HD4850 if I don't use windowed mode with SWB mod.

Either way AMD see it to be a serious enough issue and enough to work on it.
Maybe this will resolve the more significant DX9c issue which some people are plagued with still.

Maybe it's certain monitors which bring out the problem more than others?
Posted on Reply
#8
cadaveca
My name is Dave
Not monitors.

It's due to VGA memory management. It's even worse in crossfire, because the effect doubles, leaving what appears to be worse performance than a single card, even though the actual frame rate is higher.


And frankly, this problem is present in ANY high-end card from AMD, since 2900XT. I can't comment on anything other than high-end cards, because that's all that I have used.

What is even funnier, is that I can run 5760x1200(6912000 pixels) far easier than 2560x1600 (4096000 pixels). This was the key indicator that memory management, to me, was at fault, but since I've brought up this issue for years now, and none have seemed to care, I haven't reported this fact to anyone but a few users here.


How is it possible, that rendering 50% more pixels in Eyefinity performs better than 50% less pixels on a single monitor?


:wtf:


There's more to this than anyone wants to admit, unfortunately.

Truly, I dunno know WTF is the problem, really, but there HUGE issues here. I almost bought a GTX690, because I'm sick of AMD and their issues as well, but I feel I need I need GTX690 performance because of problems with performance on AMD cards.
Posted on Reply
#9
eidairaman1
by: Widjaja
Wondered where this read out came from.

I looked at the youtube latency thing and saw nothing that I can discern as micro stutter.
If so both cards had it and none of them seemed to be worse than the other.
Plus Skyrim does have a micro stutter on Windows 7 even when I use an HD4850 if I don't use windowed mode with SWB mod.

Either way AMD see it to be a serious enough issue and enough to work on it.
Maybe this will resolve the more significant DX9c issue which some people are plagued with still.

Maybe it's certain monitors which bring out the problem more than others?
Q is why the piss are we still on DX9?
Posted on Reply
#10
qubit
Overclocked quantum bit
by: eidairaman1
Q is why the piss are we still on DX9?
Hey wut, someone summoned me? :wtf:
Posted on Reply
#11
Widjaja
by: qubit
Hey wut, someone summoned me? :wtf:
I thought the Q stood for question.:wtf:
Posted on Reply
#12
Prima.Vera
Guys, just a recommendation. This stuttering issue is more common to Intel processors with Hyperthreading. Try to find a tool that enable parked cores in Windows, this was a big fix for everyone.
Posted on Reply
#13
eidairaman1
by: qubit
Hey wut, someone summoned me? :wtf:
nah sorry i was a lil sleepy at time, I meant The QUestion is why are we still on direct X 9C when 11 has been out for 2 years now?

by: Prima.Vera
Guys, just a recommendation. This stuttering issue is more common to Intel processors with Hyperthreading. Try to find a tool that enable parked cores in Windows, this was a big fix for everyone.
that or just turn off hyper threading
Posted on Reply
#14
Frick
Fishfaced Nincompoop
by: Widjaja
I thought the Q stood for question.:wtf:
Qubit has a big ego.
Posted on Reply
#15
TRWOV
by: eidairaman1
nah sorry i was a lil sleepy at time, I meant The QUestion is why are we still on direct X 9C when 11 has been out for 2 years now?
This is why:

Posted on Reply
#16
qubit
Overclocked quantum bit
by: Widjaja
I thought the Q stood for question.:wtf:
by: eidairaman1
nah sorry i was a lil sleepy at time, I meant The QUestion is why are we still on direct X 9C when 11 has been out for 2 years now?
I wuz just messin'! :) This qubit kinda wakes up when people stick Q in their sentences, lol.

by: Frick
Qubit has a big ego.
This ^^ :laugh:
Posted on Reply
#17
SK-1
by: Prima.Vera
Guys, just a recommendation. This stuttering issue is more common to Intel processors with Hyperthreading. Try to find a tool that enable parked cores in Windows, this was a big fix for everyone.
Looking into this....
Posted on Reply
#18
cadaveca
My name is Dave
by: Prima.Vera
Guys, just a recommendation. This stuttering issue is more common to Intel processors with Hyperthreading. Try to find a tool that enable parked cores in Windows, this was a big fix for everyone.
Except me. I even bought a 3570k, a chip with no HT by default, just so I could check into that, and it didn't fix a thing. I tried the same with P55, too bought i5 760 and then i7 870.

What's more interesting is that system memory speed seems to smooth things out a wee bit for me. Like 2400 MHz on IVB and then some.
Posted on Reply
#19
erocker
by: Prima.Vera
Guys, just a recommendation. This stuttering issue is more common to Intel processors with Hyperthreading. Try to find a tool that enable parked cores in Windows, this was a big fix for everyone.
While HT can cause stuttering in some games/applications, it isn't the case here. I've tried testing with HT both on and off and the same results. Granted, I haven't once noticed stuttering and am going by what the frametimes/FRAFS viewer say .
Posted on Reply
#20
qubit
Overclocked quantum bit
by: erocker
Granted, I haven't once noticed stuttering and am going by what the frametimes/FRAFS viewer say .
It looks to me that if the delay isn't long enough to cause the graphics card to skip a frame then you won't see it?
Posted on Reply
#21
erocker
by: qubit
It looks to me that if the delay isn't long enough to cause the graphics card to skip a frame then you won't see it?
Correct. Generally, one will notice it if it is above 80ms and if a few happen at around the same time. Guild Wars 2 is one of those games that has a large issue with latency, however it still seems rather smooth.
Posted on Reply
#22
qubit
Overclocked quantum bit
by: erocker
Correct. Generally, one will notice it if it is above 80ms and if a few happen at around the same time. Guild Wars 2 is one of those games that has a large issue with latency, however it still seems rather smooth.
So the problem can be mitigated by throwing lots of horsepower at it, making the overall processing time shorter, thus reducing the "really big" delays, which would tend to show up as little hitches in the movement.

That's not so bad then and helps to explain why AMD didn't bother fixing it for a year. Still naughty of them though, tsk.
Posted on Reply
#23
Calin Banc
Not really, it depends. In Skyrim, at least with v-sync ON, more power (meaning less/low GPU usage for 60FPS) usually means more stutter. In some games that works, more power, less lag or stuttering.
Posted on Reply
#24
erocker
by: qubit
So the problem can be mitigated by throwing lots of horsepower at it, making the overall processing time shorter, thus reducing the "really big" delays, which would tend to show up as little hitches in the movement.

That's not so bad then and helps to explain why AMD didn't bother fixing it for a year. Still naughty of them though, tsk.
by: Calin Banc
Not really, it depends. In Skyrim, at least with v-sync ON, more power (meaning less/low GPU usage for 60FPS) usually means more stutter. In some games that works, more power, less lag or stuttering.
Calin gives a good answer. I'm going to run a couple "non-intensive" games and see what comes up. AMD states (and I believe them) that this issue has to do with how memory (controller I assume) is being utilized with GCN.
Posted on Reply
#25
radrok
by: Calin Banc
Not really, it depends. In Skyrim, at least with v-sync ON, more power (meaning less/low GPU usage for 60FPS) usually means more stutter. In some games that works, more power, less lag or stuttering.
Yeah I've noticed that, running CFX on Skyrim increases my stuttering notably because the GPU usage on both chips is around 40%.

When running a single GPU it goes up to 80%-90% and it "seems" smooth.

Still can't figure out why quadCFX doesn't work on Skyrim but this doesn't concern this topic.
Posted on Reply
Add your own comment