Return of AMD FX: My OC'd AMD FX 8150 with OC'd 6990 Review - First Results Up!

Discussion in 'AMD Processors' started by polyzp, Jan 4, 2012.

  1. polyzp

    polyzp n00bie

    Messages:
    48
    Joined:
    Jan 4, 2012
    [​IMG]

    Thanks for the sticker OCZ!


    So with the money I made from selling my Sapphire 6970 (which I used to have in crossfire with my 6990) and my old OCZ revodrive 120gb I decided to buy OCZ's consumer flagship PCI-X SSD the Revodrive 3 X2 240 Gb.


    Just how fast is this thing compared to other SATA 6 SSDs?





    RESULTS:


    ATTO Disk Benchmark:



    OCZ Vertex 3 Max IOPS

    [​IMG]

    OCZ Revodrive 3 X2

    [​IMG]

    Only up until 16Kb read/write , the Rovodrive 3 X2 actually trails its much cheaper brother the OCZ Vertex 3 Max IOPS, however as soon as it hits 32Kb read/write it leaves it in the dust!


    Kind of like this Video:


    http://www.youtube.com/watch?v=RXqSedWSu2k



    PassMark Disk Mark:


    [​IMG]


    That's more like it!
     
  2. Tsumi

    Tsumi [H]ardForum Junkie

    Messages:
    12,187
    Joined:
    Mar 18, 2010
    It has been delayed for almost a year by the time they released it. I doubt they could have really delayed any longer.
     
  3. polyzp

    polyzp n00bie

    Messages:
    48
    Joined:
    Jan 4, 2012
    winrar benches tmw! ^^
     
  4. polyzp

    polyzp n00bie

    Messages:
    48
    Joined:
    Jan 4, 2012
    ROUND 5: WinRar Benchmark


    So when bulldozer was officially released Winrar was one of the benchmarks where FX raced ahead of the 2600k. (Example1)(Example2)

    But due to a newly discovered bug where Windows disables HT for intel processors, CORE Parking must be enabled to get the full potential out of compressing and decompressing with Winrar.

    RESULTS:

    [​IMG]


    Bulldozer only barley beats a stock i7 870k with core parking turned off.
    Before the bug was discovered AMD FX 8150 appeared to have beat even a 3960x.


    source : http://www.xtremehardware.it/


    If you have any questions feel free to ask!

    Link to Blog:
    http://AMDFX.blogspot.com
     
  5. -Strelok-

    -Strelok- [H]ardForum Junkie

    Messages:
    10,740
    Joined:
    Dec 2, 2010
    It was a bug that inhibited the performance of the Intels, once again Bulldozer getting completely raped :eek:
     
  6. polyzp

    polyzp n00bie

    Messages:
    48
    Joined:
    Jan 4, 2012
    7 zip benchmarks up!
     
  7. polyzp

    polyzp n00bie

    Messages:
    48
    Joined:
    Jan 4, 2012
    heaven 2.5 benchmarks up!
     
  8. AntiSocialMunky

    AntiSocialMunky Gawd

    Messages:
    590
    Joined:
    Nov 4, 2006
    That's good I suppose 7z > RAR.
     
  9. jreinlie4

    jreinlie4 Limp Gawd

    Messages:
    321
    Joined:
    Dec 18, 2011
    Keep up the review bro, let's see some res evil 5 and lost planet 2 benchies! Just make sure you use the fixed benchmarks on both of them if you do decide to do them.
     
  10. polyzp

    polyzp n00bie

    Messages:
    48
    Joined:
    Jan 4, 2012
    cinebench results up!
     
  11. polyzp

    polyzp n00bie

    Messages:
    48
    Joined:
    Jan 4, 2012
    WEI / cpu and gpu temps up on blog. Benchmarks continue tomorrow!
     
  12. polyzp

    polyzp n00bie

    Messages:
    48
    Joined:
    Jan 4, 2012
    AID64 benchmarks later today! stay tuned! Also, doing patch benchmark comparison!
     
  13. teletran8

    teletran8 2[H]4U

    Messages:
    2,225
    Joined:
    Jan 12, 2011
    This.

    I personally found a Res Evil 5 ;) Ignore pic if you own a SB, because a Phenom II can spank it in this game also :D

    [​IMG]
     
  14. noko

    noko 2[H]4U

    Messages:
    2,508
    Joined:
    Apr 14, 2010
    This is a great thread! Also shows when OC this arch scales well and continues to scale well the faster it goes. All the OC tricks for BD I don't think is well know yet which I believe is leading some with less OCing range then possible with their rigs. For example I am limited to 4.8ghz unless I up my fsb, once I do that (of course with other stuff too) I can go to 5.3ghz. Not sure if particular to just the chip I got but for mine that is the case. People stick great water coolers on the cpu and then cause another problem. Those VM's surrounding the cpu also gets cooled by that CPU fan which was just taken away, so those VM's can start getting rather toasty and causing one's OC to be much less then it should be. There are bios settings which left in auto will not take you above 4.8 ghz as well. Other settings which have a huge impact on memory performance and thus impact on memory intensive tasks. I am thinking of starting another thread dealing with all the OCing tricks or techniques that I've found to work with BD which should work for all the BD line and hopefully help others get the most out of their setup. Plus others can share their successes and failures so we can all win.
     
  15. jreinlie4

    jreinlie4 Limp Gawd

    Messages:
    321
    Joined:
    Dec 18, 2011
    Do it.

    I was able to get my 8150 4.62ghz x 210mhz stable with 1.51 Vcore.
    Temps max in high 40's in Overdrive Stability test.

    But I can't get it to post @ 4.7. I have to clear Cmos every time.
     
  16. polyzp

    polyzp n00bie

    Messages:
    48
    Joined:
    Jan 4, 2012
    My chip wont go past 1.47V without freezing. Im at 4.81 Ghz with 1.42 V core atm.
     
  17. noko

    noko 2[H]4U

    Messages:
    2,508
    Joined:
    Apr 14, 2010
    I looked for what motherboard you are using and couldn't find it (what are you using?). Anyways if you havn't, up your fsb and decrease your cpu multiplier, make sure ram speed is OK. I hit a multiplier wall right around 4.8ghz with a 200mhz fsb. This multiplier wall (don't quit remember if it was 23x stays constant) fsb OCing is then needed plus more stability I gained if I was much lower then let say 23x multiplier.

    If your motherboard has CPU Power Phase control settings like ASUS, use the extreme setting (ASUS has an optimize setting which is ment for low power and not high OCing and good luck if you can OC over 4ghz with it)

    Another setting key, with ASUS is CPU Voltage Fequency if left in auto it will limit OC to like 4.9ghz, manually adjusting the VRM fixed frequency to greater then 450 allows a higher OC.

    Turn off all Spread Spectrum settings except maybe PCIx.

    There are also current limit settings, voltage droop settings which I can go into but if you don't have an ASUS board it might not make any sense.

    Edit: Unfortunately, everything I told you will increase heat load on your computer and VRMs, so cooling not only the cpu is the key.
     
    Last edited: Jan 18, 2012
  18. noko

    noko 2[H]4U

    Messages:
    2,508
    Joined:
    Apr 14, 2010
    double post, deleted.
     
  19. polyzp

    polyzp n00bie

    Messages:
    48
    Joined:
    Jan 4, 2012
    yes I do use asus crosshair V. Thanks for the tips, but i have literally tried almost everything except for maybe manually setting voltage frequency. I found extreme settings for cpu voltage pushes the voltage too high and my comp wont handle it. Past 1.47 v my temps skyrocket as well. I honestly think my chip cant properly oc past 1.47v without over heating or crashing.

    AIDA64 benchmarks up!! Patch vs. no Patch comparison!!
     
  20. noko

    noko 2[H]4U

    Messages:
    2,508
    Joined:
    Apr 14, 2010
    Cool! I can compare bios settings then, I have an ASUS Sabertooth, give you what I have and any findings which may help. Yes when you start getting towards 5ghz heat does become an issue. Monitor your VRM temps using ASUS monitor, when they get above 85c things can go astray. I think I use the Very high setting for CPU voltage droop and not the Extreme since it jumps the voltage up too high when OCing with the Very high setting the CPU voltage goes up very little when at 100% load like .02v. But the phase one is one of the most important setting since it determins how many VRMs are used, four settings (ASUS Optimize, Standard, Extreme, Manual) I use the Extreme with T-Probe sensing, havn't tried the current sensing parameter with that one. The CPU voltage frequency if left in Auto limits me to around 4.9ghz, I run it manually around 470-480.
     
  21. polyzp

    polyzp n00bie

    Messages:
    48
    Joined:
    Jan 4, 2012
    updated h264 scores up! WITH PATCH!
     
  22. polyzp

    polyzp n00bie

    Messages:
    48
    Joined:
    Jan 4, 2012
    thanks alot for the tips man! im going to utilize this and see what happens! A higher OC and windows 7 patch is a nice boost in performance ^^.
     
  23. jreinlie4

    jreinlie4 Limp Gawd

    Messages:
    321
    Joined:
    Dec 18, 2011
    I hit P10338 on 3DMark11
    FX-8150 @ 4.73 x 215.0
    6970 CFX 990/1500
    http://3dmark.com/3dm11/2599334

    Try the Catalyst 12.1a Preview drivers, they gave me about a 500+ point boost on GPU score.
     
  24. polyzp

    polyzp n00bie

    Messages:
    48
    Joined:
    Jan 4, 2012
    memory benchmarks up! ^^
     
  25. Blacklash

    Blacklash [H]ard|Gawd

    Messages:
    1,899
    Joined:
    Jun 16, 2004
    Thanks for sharing :)
     
    Last edited: Jan 22, 2012
  26. K1tty

    K1tty [H]ard|Gawd

    Messages:
    1,962
    Joined:
    Mar 11, 2006
    Awesome, thanks for this! :)
     
  27. polyzp

    polyzp n00bie

    Messages:
    48
    Joined:
    Jan 4, 2012
    PCMARK 7 scores up! Pre patch vs with Patch!
     
  28. polyzp

    polyzp n00bie

    Messages:
    48
    Joined:
    Jan 4, 2012
    Winrar / Cinebench 11.5 revisited. Patch scores up!
     
  29. polyzp

    polyzp n00bie

    Messages:
    48
    Joined:
    Jan 4, 2012
    7-zip patch scores up! tomorrow Dirt 3!
     
  30. Omen7288

    Omen7288 [H]Lite

    Messages:
    85
    Joined:
    Nov 20, 2006
    Awesome stuff. Thanks for posting these. I keep checking back for more and some of the results are surprising.
     
  31. polyzp

    polyzp n00bie

    Messages:
    48
    Joined:
    Jan 4, 2012
    DIRT 3 benchmarks up!
     
  32. malih

    malih Limp Gawd

    Messages:
    196
    Joined:
    Jan 2, 2009
    Wonderful, never thought an FX could be made quite competitive,
    Now if they would just release a Piledriver CPU (or APU) that is more power efficient, I'll be jumping onboard.
     
  33. polyzp

    polyzp n00bie

    Messages:
    48
    Joined:
    Jan 4, 2012
    TrueCrypt benchmarks up!
     
  34. polyzp

    polyzp n00bie

    Messages:
    48
    Joined:
    Jan 4, 2012
    Symark SiSoftware 2012 Benchmarks up! tomorrow Alien Vs. Predator Revisited!
     
  35. polyzp

    polyzp n00bie

    Messages:
    48
    Joined:
    Jan 4, 2012
    AVP revisited with patch up!
     
  36. polyzp

    polyzp n00bie

    Messages:
    48
    Joined:
    Jan 4, 2012
    POV Ray 3.7 RC3 benchmarks up!
     
  37. polyzp

    polyzp n00bie

    Messages:
    48
    Joined:
    Jan 4, 2012
    more benchmarks starting tomorrow! Revisiting DIRT 3 by request with FX @ 4 Ghz
     
  38. polyzp

    polyzp n00bie

    Messages:
    48
    Joined:
    Jan 4, 2012
    DIRT 3 revisited ..AGAIN! FX @ 4.0 Ghz. Tomorrow Fritz Chess benchmark, then after Cinebench 10, then SpecviewPerf 11
     
  39. polyzp

    polyzp n00bie

    Messages:
    48
    Joined:
    Jan 4, 2012
    Fritz Chess Benchmark 4.3 results up! Tomorrow Cinebench 10
     
  40. pgaster

    pgaster [H]ard|Gawd

    Messages:
    1,134
    Joined:
    May 17, 2008
    Before you get to Specviewperf 11 I wanted to point something out.
    Specviewperf really only runs well if you have a professional video card, in other words Quadro or Firepro.
    Take a look at this older review of a Quadro FX 4800 vs a Geforce GTX 280.
    http://www.tomshardware.com/reviews/quadro-fx-4800,2258-10.html
    The Quadro was 10X faster in the Solidworks portion of the test. This is mainly due to the difference in drivers between the Quadro and Geforce.
    Anyway, my point is that since I think you are running a gaming card, then your scores will be low and will not tell you much of anything of value. Specview is so dependent on the gpu that I do not see it as a good cpu test at all.