15.4 vs 1018.1 "Vbs" Benchmarks, Quad and Dual Core

Discussion in 'Videocards - AMD Radeon Drivers Section' started by PrMinisterGR, May 9, 2015.

Thread Status:
Not open for further replies.
  1. PrMinisterGR

    PrMinisterGR Ancient Guru

    Messages:
    7,004
    Likes Received:
    137
    GPU:
    Sapphire 7970 Quadrobake
    Because the subject was getting a bit out of hand in the DX11 overhead thread, I decided to start this thread to concatenate my benchmark results in this one, so that we at least have some kind of point of reference.

    My software is Windows 8.1.1 x64 with all the updates as of today (08/05/2015), my hardware (thanks Blackfyre :D) is as you see it on the left. The game settings are everything to the max except non-post-processing antialiasing, unless I provide a link about the specific title's settings. The resolution is a humble 1080p60 (GPU donations accepted :D ).

    I do not include min/max framerates, as they are completely useless and they clout the charts. I have gone the other way around and tried to find some kind of metric that would show the relative variance and smoothness of the experience, instead of only the framerates (which don't mean much from a point on). The 99th percentile number shows in under how many ms the 99% of the frames is presented, and after Vbs' suggestions I saw the light and I'm removing the Standard Deviation that was too linear to represent anything very meaningful, in favor of Variance. It can be better explained here, by people who know much more than me. Variance and the 99th percentile are by far the most important metrics here, as the smaller they are the smoother the frame delivery is.

    Let's start with the games then.

    Assassin's Creed Unity
    [​IMG]

    Game Settings:
    Environmental: Quality High
    Texture Quality: Ultra High
    Shadow Quality: Low
    Ambient Occlusion: HBAO+
    Anti-Aliasing Quality: FXAA
    Bloom: On

    Test Route.
    You can see in the frame chart, this driver is much better at delivering frames as both the curves and the numbers are indicating much smoother deliveries. The dual core peformance seems worse though.

    Dying Light
    [​IMG]

    Game Settings, Benchmark Route.

    Same story here, with many more gains in the quad core test. This will become a pattern with this driver, as you'll see in the next tests too.

    Total War Rome 2
    [​IMG]

    Rome 2 is a badly written game (still lags horribly by just selecting a unit in the campaing map). These numbers are taken using the game's benchmark. The 1018.1 plays and feels worse with a dual core and the Variance number confirms it. It is the opposite with a quad though. Whatever these drivers have, it is not focused on dual+HT scenarios.

    StarCraft II: Heart of the Swarm
    [​IMG]

    Althoug StarCraft II is a DirectX 9.0 title, we curiously see a difference in peformance, towards the bottom. Again dual core performance is harmed with the new drivers, and the game more or less performs in a similar manner to Rome 2.

    Unigine Heaven
    [​IMG]

    Here things are neck to neck and within all kinds of statistical error. For all intents and purposes, the performance is the same, with the 15.4 driver being worse in a Dual Core scenario for the first time, and being more "spiky" in general.

    Unreal 4 Elemental Demo
    [​IMG]

    Another case where everything is within the margin of error.

    Metal Gear Solid V: Ground Zeroes
    [​IMG]

    Here again we take the paces of the intro sequence of the game (and it is Hojima-long), again the performance is the same between the driver sets, except the 1018.1 being worse in the dual core scenario.

    Final Fantasy XIV: Heavensward Benchmark
    [​IMG]

    This is one of the examples where the 1018.1 driver simply gave worse results, especially in the quad core scenario. I run the test three times, the results were consistent with what you see. Much lower average framerate, a tiny bit better frame delivery.

    Monster Hunter Online Benchmark
    [​IMG]

    Another one where it is more of the same. Practically identical performance, except that in the dual core scenario the 1018.1 is a bit worse again.

    Project C.A.R.S.
    For the first test everything was at Ultra except Track Detail which was Low, and Detailed Grass which was off.

    [​IMG]

    Practically identical performance from both drivers, with a tiny bit more of frame stability for 1018.1 in the quad core scenario.

    For the second test Track Detail and Detailed Grass were at Ultra. There is something wrong with this game, nothing being displayed could ever possibly justify the performance hit.

    [​IMG]

    Still, this is one of the cases where the 1018.1 driver carries its own weight, and for the first time in a dual core scenario too. The 15.4 is in practice unplayable.

    Civilization V
    With its support for DX11 multithreading, Civilization V is the personification of "last, but not least". The game provides its own benchmarks and metrics, with the LateGameView simulating a game of 300 rounds played already, and the Units test filling the screen with, well... units.

    [​IMG]

    The results are the same. Apart from a tiny increase in the Units Full Render score, things are more or less the same.

    Unfortunately I didn't have more games/hardware available for testing, but the results are quite fascinating. In almost all the tests we have a drop in the dual core performance of the new drivers, which is simultaneously combined with smoother frame delivery in quad core scenarios. The StarCraft II results are surprising, as DirectX 9.0 performance was supposed to be the same and they might be an indication that this is a change in a lower part of the driver (possibly the scheduler itself).
    This doesn't solve the problem of low performance with lower end CPU's, it actually makes it worse. It is good news for people with substantial hardware though, as things seem to move in a direction where the GPUs seem to have more of leeway given by the higher-end CPUs.
    If this trend continues in subsequent releases, the stutter gap with NVIDIA will probably close, but the lower cpu/thread cap will probably remain and worsen.
    It took me some time to post this, as it has been a long time I had to run benchmarks and check things so thoroughly.
    I await your comments, extra benches and hardware for me to test. :infinity:
     
    Last edited: May 9, 2015
  2. Blackfyre

    Blackfyre Maha Guru

    Messages:
    1,017
    Likes Received:
    33
    GPU:
    MSI 1070 Gaming X
    Thanks again for writing this up. Just proof read it again though, you have some grammatical mistakes (I don't care about the small ones, I mean saying software instead of hardware for example) that you should fix. You must be really tired after doing all of this... Great job again.
     
  3. BuildeR2

    BuildeR2 Ancient Guru

    Messages:
    2,837
    Likes Received:
    129
    GPU:
    EVGA 1080 Ti FTW3
    Thank you for all of this work and effort. I really only spend large amounts of time in Civ 5 and PCARS these days, so those were the ones I focused on although I did check all the results out.

    All I know is that I hope this PCARS stuff is sorted soon, since I'm between a rock and a hard place right now. I don't want to buy a 290X just to try and play PCARS at 1080p/60 since the R9 3xx are right around the corner, but I also can't play the game the way I have been dreaming about for months.

    I tried running my dream race earlier and it was just barely maintaining 30FPS with a split of high/medium settings. 55 opponents, 20 laps at LeMans in the McLaren P1 with the heaviest fog and rain, starting the race at midnight with 2x time progression.
     
  4. Blackfyre

    Blackfyre Maha Guru

    Messages:
    1,017
    Likes Received:
    33
    GPU:
    MSI 1070 Gaming X
    My advice, don't upgrade right now. It's going to feel like a long wait but it'll be worth it (Either way, if AMD releases good cards you'll get a better upgrade that way, if not and AMD continue digging, you can switch to the green-side).

    EDIT:

    It has been confirmed that AMD's low performance in this game is due to driver issues (so you might get better frame-rates within a month if AMD optimize their drivers for the game). I am seriously starting to hope Samsung buys AMD, and throw a butt-load of cash into the driver team.

    http://www.tweaktown.com/news/45047/project-cars-having-serious-performance-issues-amd-radeon-users/index.html
     
    Last edited: May 9, 2015

  5. theoneofgod

    theoneofgod Ancient Guru

    Messages:
    4,079
    Likes Received:
    51
    GPU:
    RX 580 8GB
    Shouldn't Project CARS be written to work on AMD GPU's with current drivers?

    edit: That website is damn awful! :)
     
  6. xacid0

    xacid0 Master Guru

    Messages:
    441
    Likes Received:
    1
    GPU:
    Zotac GTX980Ti AMP! Omega
    Those graphs gave me cancer.

    [​IMG]
     
  7. PrMinisterGR

    PrMinisterGR Ancient Guru

    Messages:
    7,004
    Likes Received:
    137
    GPU:
    Sapphire 7970 Quadrobake
    Any specific problem? They are supposed to by cote/driver per game. There is also a15 image per post limitation. All the frametime lines are scaled.
     
  8. xacid0

    xacid0 Master Guru

    Messages:
    441
    Likes Received:
    1
    GPU:
    Zotac GTX980Ti AMP! Omega
    Nothing, just that the graphs did not start at 0 so i got deceived on quick read. :wanker:
     
  9. klaupe

    klaupe Member Guru

    Messages:
    157
    Likes Received:
    0
    GPU:
    Zotac GTX980Ti AmpExtreme
    Thanks. Good work.

    Anyway, could you bench assetto corsa too?

    If not, could you tell me what tools you used to bench?
     
  10. Pill Monster

    Pill Monster Banned

    Messages:
    25,216
    Likes Received:
    7
    GPU:
    7950 Vapor-X 1100/1500
    Could someone please explain the purpose of all these threads.

    In all bar maybe 1 chart, none of which start at 0, there's less than 2fps between the GPU's, and no more than a couple of ms in the fcat results.

    Were I live we those results "on par", "equivelent", "identical", "neck and neck" etc...... So was it your intention to illustrate the similar performance? If so, u have suceeded.



    Your comment here confused me though,

    Not sure what u mean here, what gap?

    A gap is 30-50ms. 1ms-2ms is not a gap......


    Were you the one posting this stuff on other sites like Anantech and TPU, a few months ago?
     

  11. niczerus

    niczerus Master Guru

    Messages:
    290
    Likes Received:
    3
    GPU:
    MSI GamingX 580 4GB
    use 2 and 4 cores with or without ht ? , please add benchs for 15.4 and 1018.1 with 4 cores without ht .
     
    Last edited: May 9, 2015
  12. klaupe

    klaupe Member Guru

    Messages:
    157
    Likes Received:
    0
    GPU:
    Zotac GTX980Ti AmpExtreme
    Did you even watch the dying light/assassins creed benchmarks?
    There´s a huge difference. At others, there´s no difference. If he didn´t show, we wouldn´t know.

    What´s the problem? You are not interested, so the thread should be closed?
     
  13. Deathchild

    Deathchild Ancient Guru

    Messages:
    3,970
    Likes Received:
    2
    GPU:
    -
    Believe it or not the difference is actually noticeable in game Pill. :D
     
    Last edited: May 9, 2015
  14. WarDocsRevenge

    WarDocsRevenge Master Guru

    Messages:
    294
    Likes Received:
    0
    GPU:
    Fury-x Crossfire
    good thread good amount of info thank you very much for this post
     
  15. lorikano

    lorikano Master Guru

    Messages:
    323
    Likes Received:
    0
    GPU:
    2x Sapphire 7950@1150
    I recall there was one with GTA V as well ?
     

  16. Deathchild

    Deathchild Ancient Guru

    Messages:
    3,970
    Likes Received:
    2
    GPU:
    -
    Yes there was, I can post mine here. The improvements are a little better than on PrMinsters because I have crossfire so.. it shows a little bit more. :D Shows a better example.
     
  17. Pill Monster

    Pill Monster Banned

    Messages:
    25,216
    Likes Received:
    7
    GPU:
    7950 Vapor-X 1100/1500
    We must be looking at 2 different benchmarks because the Dying Lying graph I saw had a grand total of ~2.5fps between AMD and Nvidia. Frame latencey is same for both.

    So can u point out the huge difference as maybe I'm blind?





    [​IMG]
     
    Last edited: May 9, 2015
  18. Garwinski

    Garwinski Member Guru

    Messages:
    178
    Likes Received:
    4
    GPU:
    XFX Fury X
    @Pill Monster:

    I do not know much about frametimes, but I think less variation is better, right? So in the last graph for Dying Light frametimes, I see a lot less spikes/extreme spikes with the 1018.1 drivers, which should translate to smoother gameplay.

    Deathchild, do you have Dying Light as well? I am curious because I have crossfire as well, but I am hesitant to try the 1018.1 because of missing crossfire profiles in comparison to 15.4. In my experience, Crossfire always magnifies problems that occur in games, so for us CF users, these improvements would also mean more! :D
     
  19. OnnA

    OnnA Ancient Guru

    Messages:
    9,729
    Likes Received:
    2,046
    GPU:
    Vega 64 XTX LiQuiD
    Great Thread :D
    THX for sharing Bro... I have some stats on my Blog also ;-)
    But this driver is even more efficient as it has at least 6 cores. +8-12FPS/Average in a lot of Games.
     
  20. klaupe

    klaupe Member Guru

    Messages:
    157
    Likes Received:
    0
    GPU:
    Zotac GTX980Ti AmpExtreme
    The difference is that the standard deviation in frametimes is halved, when you use a quad core processor, which means, that you have a much more consistent frame rate, and less "hickups".

    Just look at the last graph, the "blue line" differs much more, than the other one, which mean that there is a pause between 2 frames, and that the next 2 frames come closely after another , which causes the experience to be much less fluid, although you have almost the same frames per second.
     
    Last edited: May 9, 2015
Thread Status:
Not open for further replies.

Share This Page