Overclock.net › Forums › Graphics Cards › Graphics Cards - General › [BSN] State of 4K
New Posts  All Forums:Forum Nav:

[BSN] State of 4K - Page 48  

post #471 of 493
Even if we assumed there are Shills, the situation is rotten. Why do you become "fans" of AMD or NVIDIA or Intel in the first place? AMD and Intel are a duopoly on x86, they hold x86 intellectual property by the testicular spheroids, and NVIDIA and AMD are both good GPU designers that beg TSMC to release a new process node and they're pretty much mainly depended on it and they will always be "almost equal" until that situation changes.
PC
(9 items)
 
  
CPUMotherboardGraphicsRAM
4790K 4.6G 1.24v/1.74v, 5.1G 1.35v validation. GA-Z97X-Gaming 7 Tri-X R9 290 1100/1350 +0.012v G.Skill 2400 c10 
CoolingMonitorPowerCase
Noctua NH-D15 24EA53 IPS 76Hz OC EVGA 1000 G2 Phanteks Enthoo Pro 
Audio
SoundMagic E10 
  hide details  
PC
(9 items)
 
  
CPUMotherboardGraphicsRAM
4790K 4.6G 1.24v/1.74v, 5.1G 1.35v validation. GA-Z97X-Gaming 7 Tri-X R9 290 1100/1350 +0.012v G.Skill 2400 c10 
CoolingMonitorPowerCase
Noctua NH-D15 24EA53 IPS 76Hz OC EVGA 1000 G2 Phanteks Enthoo Pro 
Audio
SoundMagic E10 
  hide details  
post #472 of 493
Thread Starter 
Quote:
Originally Posted by looniam View Post

i guess you only scanned the articles headlines, eh?

try reading content - you'll go farther:
Frame Rating Dissected: Full Details on Capture-based Graphics Performance Testing
Vsync and its Effect on Frame Rating – Does it fix CrossFire?

the bottom line is, ryan isn't on AMD's payroll so why should HE be fixing AMD's problems?

as a journalist he only needs to report what he finds.


ZZZZZzzzzzzzzzz

Fair enough, he did run some basic vsync testing.(if you can find another article please do) Unfortunately though in the top test he's using settings that drop below 60fps without triple buffering. That immediately means the frame time variance is jumping between 16.6 and 33. Not ideal and not what ive been posting about above. However he does admit it cures the problem of runt frames which i also mentioned above. Infact all the problems he brings up for the AMD side would have been solved just by using an fps cap and or radeonpro, or what i said above.

What's even more interesting to note here is that if you read the article further down he goes to great lengths to explain the possible solutions for Nvidia user, but doesn't even mention one for Amd USERS.Aside from using standard vsync without triple buffering it would seem.

Solutions can be offered for Nvidia via adaptive vsync and smooth vsync. What are the solutions for AMD users? You guessed it adaptive vsync and smooth vsync via radeonpro. Are they mentioned, no. Any hint to them, no.



This should have been researched and presented. Again this is blatantly one sided. Its so obviously its painful to see.

What would have been great if pcper had gone to AMD and said look this is the situation. You can see the issues here. Currently they can be solved or greatly improved by using Radeonpro. As Nvidia currently offer this solution via their control panel we think you should do the same and here's why. They present the results to them. That could be made clear in the article with a before and after. Now that is good, helpful honest journalism.

I would have no objections to pcper saying it can be fixed for both sides but currently Nvidia offer the easiest solution as its part of their control panel. As it stand AMD users would have to use radeonpro to get the same effect.

I know what people will say. Pcper shouldn't have to report it as if its not included in CCC then its tough. Well that is an example or poor, one sided journalism in my eyes.This is about informing your readers of problems, making people aware of the problems as well as the companies. It should also offer possible solutions for both sides. To only do that for one side is bias.

Quote:
Originally Posted by NeedProof View Post

You are saying both sides are wrong, I'm saying the same thing. Yet somehow we're wrong.

My previous post was last post addressing him. I suggest you do the same. smile.gif

I know. Unfortunately it doesn't matter we say or how we try to explain it. He won't change his mind on the matter as 'he's right' and that's it.
post #473 of 493
Why are people so obsessed with this one reviewer is just beyond me? Come on folks, you all are giving it way too much attention than its worth. What difference does it make what this reviewer said or how it was said. Not like it will influence anyone's decision making process. rolleyes.gif
Edited by provost - 9/22/13 at 5:21am
Simplicity
(11 items)
 
Apotheosis
(10 items)
 
 
CPUMotherboardGraphicsRAM
4770k Asus Z87 Pro TBD Corsair Vengeance (2x8GB) DDR3 1600 RAM 
OSMonitorKeyboardPower
Windows 7 Pro Dell U2713HM Alienware TactX gaming Seasonic 850W Gold  
CaseMouse
Cooler Master HAF XB Alienware TactX premium mouse 
  hide details  
Simplicity
(11 items)
 
Apotheosis
(10 items)
 
 
CPUMotherboardGraphicsRAM
4770k Asus Z87 Pro TBD Corsair Vengeance (2x8GB) DDR3 1600 RAM 
OSMonitorKeyboardPower
Windows 7 Pro Dell U2713HM Alienware TactX gaming Seasonic 850W Gold  
CaseMouse
Cooler Master HAF XB Alienware TactX premium mouse 
  hide details  
post #474 of 493
Quote:
Originally Posted by looniam View Post

sure radeon pro works by limiting the cap if you don't mind:
setting it for every game
having your avg FPS lower
having to go down to the worse FPS if you go from inside to outside (when your FPS will drop)
increased input lag.


this was all covered in febuary of this year when the first frame pacing articles were publish
1) No. You do not need to set it for every game, you can have a global profile
2) Avg fps? Are you kidding me?
3) Wat?
4) It doesn't increase input lag. It's Dynamic Framerate Control AND NOT VSYNC. It's a variable frameCAP not a VSYNC
Quote:
Originally Posted by looniam View Post

good journalists DID cover that, including ryan - sorry but its late and i don't feel like digging back 8 months for a link. i did a quick search for radeon pro up in the search bar - there are number of threads that have a mixed bag of results but most of them were satisfied.

and IIRC the consensus was at the time of reviews from different sites: why should people have to rely on a third party software to get consistent performance from the hardware they spent good money on?

that is the hardware manufacture's responsibility. personally i wouldn't want to do a "voodoo dance" before starting to play a game. i want to just load it and start shooting/looting/whatever. but that is just my opinion
Are you implying that we shouldn't use Windows because IE is so terrible we have to use 3rd party browsers? Fedora comes with Firefox as standard.
Ayase Eli
(20 items)
 
Testbench
(10 items)
 
MC Server
(13 items)
 
CPUMotherboardGraphicsRAM
AMD Ryzen 7 1700 ASRock X370 SLI Killer Zotac GTX1070 "FE" with ID Cooling Frostflow 240G Crucial BallistiX Sport LP 16GB 8x2GB 
Hard DriveHard DriveHard DriveCooling
Zotac SSD Premium 480GB OCZ Vector 180 480GB Hitachi 7K3000 3TB 7200rpm ID Cooling Frostflow 240L 
CoolingCoolingCoolingCooling
ID Cooling Frostflow 240G Phanteks F140MP x2 CoolerMaster Silencio FP120 x2 Phanteks F140SP 
CoolingMonitorKeyboardPower
Delta AFB1212SH x2 Dell U2713HM CM Storm MECH FSP Aurum PT 1200 
CaseMouseMouse PadAudio
Phanteks Enthoo Pro Logitech G102 Tesoro Aegis X3 Virtus.Pro AKG K7XX 
CPUMotherboardGraphicsRAM
Intel i7 6700k Gigabyte Z170X-UD5 Depends Depends 
Hard DriveCoolingMonitorKeyboard
Plextor M5S Depends Dell U2312HM CM Storm Quickfire Rapid-i 
PowerMouse
FSP Aurum PT 1200 CM Storm Mizar 
CPUMotherboardGraphicsRAM
Intel i7 2600 4GHz Gigabyte Z77X D3H Intel HD Graphics 2000 obviously Kingston HyperX Genesis 4X4GB DDR3 1600 
Hard DriveCoolingCoolingCooling
Intel 530 240GB Deepcool Captain 240 Arctic Cooling Bionix 140 Delta AFB1212L 
OSPowerCase
Windows 10 Pro CoolerMaster G750M NZXT Switch 810 
  hide details  
Ayase Eli
(20 items)
 
Testbench
(10 items)
 
MC Server
(13 items)
 
CPUMotherboardGraphicsRAM
AMD Ryzen 7 1700 ASRock X370 SLI Killer Zotac GTX1070 "FE" with ID Cooling Frostflow 240G Crucial BallistiX Sport LP 16GB 8x2GB 
Hard DriveHard DriveHard DriveCooling
Zotac SSD Premium 480GB OCZ Vector 180 480GB Hitachi 7K3000 3TB 7200rpm ID Cooling Frostflow 240L 
CoolingCoolingCoolingCooling
ID Cooling Frostflow 240G Phanteks F140MP x2 CoolerMaster Silencio FP120 x2 Phanteks F140SP 
CoolingMonitorKeyboardPower
Delta AFB1212SH x2 Dell U2713HM CM Storm MECH FSP Aurum PT 1200 
CaseMouseMouse PadAudio
Phanteks Enthoo Pro Logitech G102 Tesoro Aegis X3 Virtus.Pro AKG K7XX 
CPUMotherboardGraphicsRAM
Intel i7 6700k Gigabyte Z170X-UD5 Depends Depends 
Hard DriveCoolingMonitorKeyboard
Plextor M5S Depends Dell U2312HM CM Storm Quickfire Rapid-i 
PowerMouse
FSP Aurum PT 1200 CM Storm Mizar 
CPUMotherboardGraphicsRAM
Intel i7 2600 4GHz Gigabyte Z77X D3H Intel HD Graphics 2000 obviously Kingston HyperX Genesis 4X4GB DDR3 1600 
Hard DriveCoolingCoolingCooling
Intel 530 240GB Deepcool Captain 240 Arctic Cooling Bionix 140 Delta AFB1212L 
OSPowerCase
Windows 10 Pro CoolerMaster G750M NZXT Switch 810 
  hide details  
post #475 of 493
Quote:
Originally Posted by LtMatt View Post

I hope Ryan is reading this. I think someone suggested it before so credit to whoever that was.
Warning: Spoiler! (Click to show)
It would be nice if pcper tried to show or make an effort to discover a solution to problems they report in the meantime. Someone else said its important to report the problems, but its also helpful to try and provide a fix or a temp solution if possible. This is something i definitely agree with. Now obviously its up to AMD to fix it, no one is disputing that. But there are several solutions to the frame pacing issue for example. Something ive never once see pcper mention in ANY of their articles. Its all just crappy crossfire experience this, stuttering that, runt frames etc.

An easy way to dramatically improve things regarding frame pacing, runt frames etc is Vsync. At least that's how it felt to me using crossfire months back before frame pacing existed. Vsync removed microstutter for me, completely in some cases. I also think it might have had an effect on runt frames. Of course i have no data for this, i can only go on how it felt. On some titles it did not remove microstuttering, sleeping dogs was one of them. The majority it did though.

I don't think ive ever seen pcper offer this as a solution. I remember once reading a pcper article that said they were thinking of testing vsync in the future to see if it solved any problems. As far as i know, that never happened. If it did im sure someone will correct me on that shortly.

Another solution is to limit the fps, with or without using vsync. Though i prefer vsync where possible as i hate tearing. Lets say you use vsync on a game. You would then set a fps limit of 59. This will remove ALL stutter, and id bet all runt frames. Providing you can maintain 60fps your frame time variance will be zero. Nada. This may not be perfect in all scenarios. But this is what we did. Us being xfire users. Before we had frame pacing this is what we would do if the game we played had stutter, hitching etc. Suddenly there was no stutter. I mean none, nothing. Why was this never presented? Tested? Good journalism would be trying to help readers with a temporary solution until AMD fixed the problem. In my opinion if pcper had done this, they would get much less grief.

Let me show you an example.




Source
http://www.tomshardware.com/reviews/radeon-hd-7990-devil13-7970-x2,3329-11.html


Now pcper. Why not show that there is another side, there is another solution. Sure its not ideal, but its a placeholder until the real frame pacing driver came out. Pcper were saying crossfire is/was unsuitable for such a long time. Yet there were so many of us using it and thinking but all we need is radeonpro and a few tweaks here and there. With that we have better frame times than a titan and no microstutter at all. This was never explored and it should have been. The articles would have been looked on so much more favourably. Pcper could have tried to work with AMD and say hey look what we can do with radeonpro or with this fix or that tweak. This is how bad it was, but with using this (insert possible fix/tweak here) it suddenly becomes so much better.

Again im not trying to say its a perfect solution, but there are many capable solutions. Unfortunately none of this was ever presented and it would have greatly helped many users if it was. Its never too late to change ways.

Now i know these Eyefinity issues that pcper mention are different. There may not be any specific or temp fix out there that radeonpro or vsync or anything else can solve. This may be an issue that only AMD can fix. I don't know enough about it. What i do know is i have friends using Eyefinity that have never mentioned these problems. I'm now in contact with them and have directed them to the pcper article and will see if they have these problems. If they don't ill be sure to ask them why and what, if anything, they do differently.

No matter how it may seem, i respect Ryan and pcper for helping AMD and testing their drivers to improve crossfire. It has benefitted me greatly, but things can improve on both sides regarding this and i would like to see that happen more in the future
.
It was me. And you are quite correct. And the only debate offered by most is just that, not the writers responsibility. But the writers always claim they work for the consumer not the companies, so how does just telling them how bad such a purchase is helping their consumers that already have said bad product. In this regard it isn't telling them anything new and even less helpful other than to quantify what they are experiencing. Trust me what Ryan does is cool and all the work involved in getting the data is helpful is some ways. But as I read his articles it seems even he knows little of the problem in and of itself. He points out the issues but has no scientific reason for what result eventuates. Again it is like going up to a person on the side of the road changing a tire and telling them they have a flat. Unfortunately, and more so in this case, very obvious.

Now whether Nvidia had anything to do with the articles timing or not I think any reasonable person would agree the timing does in fact open the possibility of that being an actual possibility here. Doesn't mean it is, but does mean it has credible merit. It could be as simple that Ryan was working on the story then Nvidia rounds up all these writers and then Ryan has the need to rush his article to press to beat all others knowing now that more would soon follow. That is the most likely scenario.
post #476 of 493
The only real problem I see with this whole expose on crossfire issues is that a bigger stink wasn't made about it years ago.

Crossfire has been plagued with microstutter issues since AFR became the preferred multi-GPU rendering mode for most games in 2006-2007 (supertiling and SFR don't have comparable issues, but they were depreciated in the pursuit of more, often useless, frame rate scaling), and people with crossfire/SLI settups have been complaining about microstutter for just as long. SLI's microstutter problems largely evaporated by the GTX 400 series, but ATI/AMD's persisted for years more with only occasional mention.
Quote:
Originally Posted by LtMatt View Post

Now pcper. Why not show that there is another side, there is another solution. Sure its not ideal, but its a placeholder until the real frame pacing driver came out. Pcper were saying crossfire is/was unsuitable for such a long time. Yet there were so many of us using it and thinking but all we need is radeonpro and a few tweaks here and there.

Dynamic vsync is a joke and RadeonPro is not the magic fix many claim it to be. Yes, it can sometimes result in perceptible improvements, but real frame pacing is much better, and you don't need to forfeit performance to take advantage of it.

Crossfire was so unsuitable that I, an individual who simultaneously owned four 7950s, two 6950s, and four 5750s, all installed in pairs, for quite awhile (several months to several years) before the frame pacing driver fix, normally ran his systems without Crossfire enabled. And yes, I know how to use RadeonPro, and have used RadeonPro, occasionally to good effect, but often it doesn't help.
Quote:
Originally Posted by LtMatt View Post

But if consistent frame times are so important then this is the way to get the most consistent frame times possible.

Frames displayed at consistent times do not imply frames rendered at consistent times. Frame pacing will discard frames rendered too closely together, vsync will just spread them apart. The latter results in perfect frame times, but still shows a stutter in the actual motion being displayed.
Quote:
Originally Posted by LtMatt View Post

With most dual gpu setup holding 60fps in games is not a problem at all.

Almost none of my recent games consistently maintain 60+ fps at the IQ I prefer to use, even on a single 1920x1080 display, and this is with two (stable and not-throttling) 7950s at 1053/1600 and a 4.3GHz 3930k or 4.7GHz 2700k.

The higher frame rate goes the less consistency matters because of diminishing returns. Low frame rate scenarios are precisely where microstutter is most problematic, so setting a 60 fps cap means that for the short times where I'm at 40 fps, and need consistency the most, is simultaneously the time I have the least of it.
Quote:
Originally Posted by LtMatt View Post

I would rather have a consistent frame rate ie 60fps locked than fps going dramatically up or down. That keeps my frame times consistent.

Again, consistent frame times is less relevant than lack of wild/erratic changes to present calls/render times.
Quote:
Originally Posted by LtMatt View Post

You set the fps to your minimum fps that way it won't drop below it.

Capping performance at the minimum they see is completely impractical for most people.

I would need to throw thousands more dollars at my systems to ensure that non of my games ever dipped below 50 or 60 fps, and that's with one mid-resolution display.

I would need powerful magic to make such a feat possible on an Eyefinity setup or even 2560x1200 display.
Primary
(15 items)
 
Secondary
(13 items)
 
In progress
(10 items)
 
CPUMotherboardGraphicsRAM
5820K @ 4.2/3.5GHz core/uncore, 1.175/1.15v Gigabyte X99 SOC Champion (F22n) Gigabyte AORUS GTX 1080 Ti (F3P) @ 2025/1485, 1... 4x4GiB Crucial @ 2667, 12-12-12-28-T1, 1.34v 
Hard DriveHard DriveHard DriveCooling
Plextor M6e 128GB (fw 1.06) M.2 (PCI-E 2.0 2x) 2x Crucial M4 256GB 4x WD Scorpio Black 500GB Noctua NH-D15 
OSMonitorKeyboardPower
Windows 7 Professional x64 SP1 BenQ BL3200PT Filco Majestouch Tenkeyless (MX Brown) Corsair RM1000x 
CaseMouseAudio
Fractal Design Define R4 Logitech G402 Realtek ALC1150 + M-Audio AV40 
CPUMotherboardGraphicsRAM
X5670 @ 4.4/3.2GHz core/uncore, 1.36 vcore, 1.2... Gigabyte X58A-UD5 r2.0 w/FF3mod10 BIOS Sapphire Fury Nitro OC+ @ 1053/500, 1.225vGPU/1... 2x Samsung MV-3V4G3D/US @ 2000, 10-11-11-30-T1,... 
RAMHard DriveHard DriveHard Drive
1x Crucial BLT4G3D1608ET3LX0 @ 2000, 10-11-11-3... OCZ (Toshiba) Trion 150 120GB Hyundai Sapphire 120GB 3x Hitachi Deskstar 7k1000.C 1TB 
CoolingOSPowerCase
Noctua NH-D14 Windows 7 Pro x64 SP1 Antec TP-750 Fractal Design R5 
Audio
ASUS Xonar DS 
CPUMotherboardGraphicsRAM
i7-6800K @ 4.3/3.5GHz core/uncore, 1.36/1.2v ASRock X99 OC Formula (P3.10) GTX 780 (temporary) 4x4GiB Crucial DDR4-2400 @ 11-13-12-28-T2, 1.33v 
Hard DriveHard DriveCoolingOS
Intel 600p 256GB NVMe 2x HGST Travelstar 7k1000 1TB Corsair H55 (temporary) Windows Server 2016 Datacenter 
PowerCase
Seasonic SS-860XP2 Corsair Carbide Air 540 
  hide details  
Primary
(15 items)
 
Secondary
(13 items)
 
In progress
(10 items)
 
CPUMotherboardGraphicsRAM
5820K @ 4.2/3.5GHz core/uncore, 1.175/1.15v Gigabyte X99 SOC Champion (F22n) Gigabyte AORUS GTX 1080 Ti (F3P) @ 2025/1485, 1... 4x4GiB Crucial @ 2667, 12-12-12-28-T1, 1.34v 
Hard DriveHard DriveHard DriveCooling
Plextor M6e 128GB (fw 1.06) M.2 (PCI-E 2.0 2x) 2x Crucial M4 256GB 4x WD Scorpio Black 500GB Noctua NH-D15 
OSMonitorKeyboardPower
Windows 7 Professional x64 SP1 BenQ BL3200PT Filco Majestouch Tenkeyless (MX Brown) Corsair RM1000x 
CaseMouseAudio
Fractal Design Define R4 Logitech G402 Realtek ALC1150 + M-Audio AV40 
CPUMotherboardGraphicsRAM
X5670 @ 4.4/3.2GHz core/uncore, 1.36 vcore, 1.2... Gigabyte X58A-UD5 r2.0 w/FF3mod10 BIOS Sapphire Fury Nitro OC+ @ 1053/500, 1.225vGPU/1... 2x Samsung MV-3V4G3D/US @ 2000, 10-11-11-30-T1,... 
RAMHard DriveHard DriveHard Drive
1x Crucial BLT4G3D1608ET3LX0 @ 2000, 10-11-11-3... OCZ (Toshiba) Trion 150 120GB Hyundai Sapphire 120GB 3x Hitachi Deskstar 7k1000.C 1TB 
CoolingOSPowerCase
Noctua NH-D14 Windows 7 Pro x64 SP1 Antec TP-750 Fractal Design R5 
Audio
ASUS Xonar DS 
CPUMotherboardGraphicsRAM
i7-6800K @ 4.3/3.5GHz core/uncore, 1.36/1.2v ASRock X99 OC Formula (P3.10) GTX 780 (temporary) 4x4GiB Crucial DDR4-2400 @ 11-13-12-28-T2, 1.33v 
Hard DriveHard DriveCoolingOS
Intel 600p 256GB NVMe 2x HGST Travelstar 7k1000 1TB Corsair H55 (temporary) Windows Server 2016 Datacenter 
PowerCase
Seasonic SS-860XP2 Corsair Carbide Air 540 
  hide details  
post #477 of 493
Thread Starter 
Quote:
Originally Posted by Blameless View Post

The only real problem I see with this whole expose on crossfire issues is that a bigger stink wasn't made about it years ago.

Crossfire has been plagued with microstutter issues since AFR became the preferred multi-GPU rendering mode for most games in 2006-2007 (supertiling and SFR don't have comparable issues, but they were depreciated in the pursuit of more, often useless, frame rate scaling), and people with crossfire/SLI settups have been complaining about microstutter for just as long. SLI's microstutter problems largely evaporated by the GTX 400 series, but ATI/AMD's persisted for years more with only occasional mention.
Dynamic vsync is a joke and RadeonPro is not the magic fix many claim it to be. Yes, it can sometimes result in perceptible improvements, but real frame pacing is much better, and you don't need to forfeit performance to take advantage of it.

Crossfire was so unsuitable that I, an individual who simultaneously owned four 7950s, two 6950s, and four 5750s, all installed in pairs, for quite awhile (several months to several years) before the frame pacing driver fix, normally ran his systems without Crossfire enabled. And yes, I know how to use RadeonPro, and have used RadeonPro, occasionally to good effect, but often it doesn't help.
Frames displayed at consistent times do not imply frames rendered at consistent times. Frame pacing will discard frames rendered too closely together, vsync will just spread them apart. The latter results in perfect frame times, but still shows a stutter in the actual motion being displayed.
Almost none of my recent games consistently maintain 60+ fps at the IQ I prefer to use, even on a single 1920x1080 display, and this is with two (stable and not-throttling) 7950s at 1053/1600 and a 4.3GHz 3930k or 4.7GHz 2700k.

The higher frame rate goes the less consistency matters because of diminishing returns. Low frame rate scenarios are precisely where microstutter is most problematic, so setting a 60 fps cap means that for the short times where I'm at 40 fps, and need consistency the most, is simultaneously the time I have the least of it.
Again, consistent frame times is less relevant than lack of wild/erratic changes to present calls/render times.
Capping performance at the minimum they see is completely impractical for most people.

I would need to throw thousands more dollars at my systems to ensure that non of my games ever dipped below 50 or 60 fps, and that's with one mid-resolution display.

I would need powerful magic to make such a feat possible on an Eyefinity setup or even 2560x1200 display.

Frame pacing is great and i love it, but you still get a smoother experience with vsync and an fps cap in my opinion. The frame time variance which i monitor via OSD would also back this up. That said in some games i prefer to leave vsync and an fps cap off and just use frame pacing, battlefield 3 is one of those games and i do the same on some other fps shooter games. On the whole though i use frame pacing and an fps cap/vsync. Its just smoother.

Why is Dynamic vsync a joke? Its a nice idea in theory of disabling vsync when the fps drops below 60. Its very handy i think ,though i don't use it much personally. I know exactly what performance i get in games as i spend a lot of time working out what clocks i need to deliver the fps i need to deliver the performance and smoothness i seek, depending on the title.

Most of my games have no trouble maintaining 60fps at 1440p with 2x7950's @1125/1575. I play a variety of games and i run them all at the ultra/highest preset with varying amounts of AA depending on performance. These range from Tomb Raider, to Hitman, to Crysis 3 to farcry 3 to battlefield 3 and many more. Most games i will use FXAA or MLAA (AMD's own version of FXAA, much better) or small levels of AA. Maintaining 60 fps is no problem.

I ran some benchmarks on another forum a while back. Although i used more AA than i would normally, my averages all remain over 60 fps at the highest possible in game settings, at 1440p. I was also running slightly lower clocks back then on the memory, but i doubt that would make too much difference. I appreciate the metro scores are under 60 fps, but in game they are over. Also Metro games are not as well optimised i find as gaming evolved titles. For example i see gpu usage drops in both metro benchmarks during the benching run which would indicate some kind of cpu bottleneck/physx crap.
Warning: Spoiler! (Click to show)
Specs

i7 2700k@ 4.8ghz
8Gb DDR3 2133Mhz
2x7950 @1125/1500
Windows 8
13.10 Drivers
Frame pacing enabled

Benchmarking Procedure & Settings

FOV (where applicable) 82
Ultra/Extreme Preset used from in game options.

Battlefield 3 - http://www.youtube.com/watch?v=j2b5b3vK-TY
Bioshock - Used in game benchmark
Crysis 3 - http://www.youtube.com/watch?v=mxrjqhSAnAo
Dirt Showdown - Used in game benchmark
Farcry 3 - http://www.youtube.com/watch?v=mW2gDhEMfyk
Hitman Absolution - Used in game benchmark
Metro 2033 - Used in game benchmark
Metro Last Light - Used in game benchmark
Tomb Raider - Used in game benchmark
Sleeping Dogs - Used in game benchmark
3DMark 11 - Firestrike
Unigine Valley - Extreme HD


7950 Crossfire - Overclocked
Warning: Spoiler! (Click to show)








12660 with AMD Radeon HD 7950(2x) and Intel Core i7-2700K Processor
Graphics Score 16376
Physics Score 12213
Combined Score 4784

http://www.3dmark.com/3dm/1181662?






Anyway my point was that its not a perfect solution but if stutter is a problem there are ways around it where you can get an excellent performance if you just use vsync/fps limit or other little tweaks. As i said before you need to be able to deliver those fps but generally thats not a problem as long as you run reasonable settings and don't try whacking on silly amounts of AA. As you have discovered though if you set a fps cap and cannot maintain it then you won't get the desired smoothness or performance.

You don't need a fortune either. My 2x7950's cost £350 or $560 dollars and cost about the same as a 770 give or take.
Warning: Spoiler! (Click to show)





Edited by LtMatt - 9/22/13 at 7:32am
post #478 of 493
Thread Starter 
Did a bit more digging and what do you know. AMD's invite for the Hawaii launch event says...

Ultra High Gaming Resolutions (4k??)

Multiple Display (Eyefinity??)





I wonder if Ryan will be wrong and his contact in the AMD driver team story is now looking a bit questionable. The driver won't be ready for months, possibly never.

Looks like the news is spreading fast and now the man who wrote the BSN article has been made aware of the 'claims' against him here and elsewhere. (http://neogaf.com/forum/showthread.php?t=682225)





Grab some popcorn folks, there might be more to this one than meets the eye.
post #479 of 493
-nevermind-
PC
(9 items)
 
  
CPUMotherboardGraphicsRAM
4790K 4.6G 1.24v/1.74v, 5.1G 1.35v validation. GA-Z97X-Gaming 7 Tri-X R9 290 1100/1350 +0.012v G.Skill 2400 c10 
CoolingMonitorPowerCase
Noctua NH-D15 24EA53 IPS 76Hz OC EVGA 1000 G2 Phanteks Enthoo Pro 
Audio
SoundMagic E10 
  hide details  
PC
(9 items)
 
  
CPUMotherboardGraphicsRAM
4790K 4.6G 1.24v/1.74v, 5.1G 1.35v validation. GA-Z97X-Gaming 7 Tri-X R9 290 1100/1350 +0.012v G.Skill 2400 c10 
CoolingMonitorPowerCase
Noctua NH-D15 24EA53 IPS 76Hz OC EVGA 1000 G2 Phanteks Enthoo Pro 
Audio
SoundMagic E10 
  hide details  
post #480 of 493
Thread Starter 
Quote:
Originally Posted by fateswarm View Post

LtMatt, that popcorn also applies to you. Have you got a real name? Because at this point I believe if you accuse someone with a real name of the ominous act being a Shill, you have to have the spheres to do it in real.

What are you talking about? Matt is my real name.

Am i breaking any law by questioning things? Is free speech not allowed here?
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Graphics Cards - General
This thread is locked  
Overclock.net › Forums › Graphics Cards › Graphics Cards - General › [BSN] State of 4K