GTX 970@1.5Ghz and 2500k vs Q9550 in 70+ benchmarks (+bonus i7-860)

Discussion in 'Videocards - NVIDIA GeForce' started by psolord, Apr 20, 2015.

  1. dhwang

    dhwang Member

    Messages:
    38
    Likes Received:
    0
    GPU:
    nvidia GTX 560 ti
    Nice work. I'd like to see the following tests. Maybe next time? :)

    1. Same clock (4GHz) comparisons using different CPUs. (To see how much the architectures will differ)
    2. Different clock (3GHz~5GHz) comparision using the same CPU. (To see how much CPU OC will differ)
    2. Stock vs OC GTX 970. (To see how much GC OC will differ)
     
  2. mahanddeem

    mahanddeem Master Guru

    Messages:
    660
    Likes Received:
    27
    GPU:
    Asus GTX1080Ti StrX
    Nice work but these 2 "old CPUs" deserve that extensive analysis?
    And it's a common sense an i5 is better performance than q9xxx
     
  3. orion24

    orion24 Master Guru

    Messages:
    836
    Likes Received:
    0
    GPU:
    GTX 680 @ 1254/1703
    Clock-4-clock would be nice, but as long as he is using a DDR-2 board for the Q9550 it wouldn't be fair. I expect a major amount of the ~66% performance difference to be from DDR-2.

    BTW very good job. I can't even image how much work the whole thing needed to pull off. Reminds me of myself 10 years ago. It is great to see PC enthusiasts still out there.
     
  4. Rafael_Morales

    Rafael_Morales Member

    Messages:
    49
    Likes Received:
    0
    GPU:
    GTX 1070
    Congratulations !!
    What a fantastic job!
     

  5. rflair

    rflair Don Commisso Staff Member

    Messages:
    4,172
    Likes Received:
    490
    GPU:
    5700XT
    I think he did an awesome job. It shows the Q9xxx is still a usable processor in many games. In DX10+ the newer CPUs do push ahead, sometimes a fair bit but the Q9550 is still playable even then.

    I've got one running in my HTPC/Server and it does a great job, just checked and its uptime is 16 weeks 6 days............

    I love that old rig.
     
    Last edited: May 2, 2015
  6. Undying

    Undying Ancient Guru

    Messages:
    16,429
    Likes Received:
    5,366
    GPU:
    Aorus RX580 XTR 8GB
    True but some games (D3,BF4,WD) got really bottlenecked by C2Q.

    What is really interesting in those charts is how overclocked 2500k still pushes 970 to the max. I would like too see comparison against new shiny Haswell i5/i7's.
     
  7. psolord

    psolord Active Member

    Messages:
    55
    Likes Received:
    6
    GPU:
    1070/970/7950/5850
    Hello again.

    I received no email notification after my post #11, so I thought no one cared, :p.

    I came to post my 860 sum up. Only the sum up for now, since I still need quite some time finish my uploads. But I guess most people what would have any interest in these runs, would mostly care for the results themselves.

    Still I would like to address:

    @dhwang, this would be the most complete way to go. Still I would have to do another round of too many runs. The test itself is based on what the systems can give as a whole, IPC and Fab Process included.

    There is a reason I record CPU usage and GPU usage graphs at the end of each video. If a viewer sees too high cpu usage and the game is still cpu limited, a lower clock would only make things worse, and so on and so forth.

    I decided to include videos, because I wanted the viewer to have a more complete picture of what is happening, instead of just sterilized numbers. :)

    Still some normalized clock results, will be posted at a later time. Read below.

    @mahanddeem, sure my friend newer stuff will be better, but then we wouldn't need to do any testing, based on this fact alone.

    Truth be told, such large analysis was not needed, but
    - I am unemployed, bored and need to do something creative to get my mind off things
    - Most people cannot afford high end processors, so this huge benchmark suite will show them what to expect from similar powered system as mine
    - Many cards will come in the future, that will be similarly powered as the 970. This test will still provide some info of how a GTX 1060 (the imaginary 1060 is equal to the 970) would perform with a Lynnfield or an old Yorkfield
    - Also many games have different balances regarding their cpu/gpu needs, so hopefully some guy can find some info for a specific game he likes. For example he wants to play Watch Dogs at 60fps with max settings, but has a Yorkfield or similar. Will he get what he wants from a newer card,etc?
     
  8. psolord

    psolord Active Member

    Messages:
    55
    Likes Received:
    6
    GPU:
    1070/970/7950/5850
    Ok guys, since I have had some requests to post some i7-860 results already, here is the sum up of the 2500k, Q9550 and Core i7-860, on the most cpu limited benchmarks of the original benchmark suite.

    Here is the list sorted by the 2500k vs Q9550 delta, from highest to lowest


    [​IMG]
    [​IMG]

    and here are the same results, sorted by the 2500k vs core i7-860 delta, from highest to lowest.

    [​IMG]
    [​IMG]

    As you can see, the 2500k/Q9550 delta dropped from 63% to only 13% for the 2500k/i7-860 comparison.

    The clock difference of the 2500k@4.8Ghz and the i7-860@4.0Ghz is 4.8/4=20%, so for 20% more clock we only have 13% more performance. Of course this is due to the fact that this whole benchmark suite, is primarily designed to be GPU limited, with real life, very high settings, hence we have many GPU limits in all runs.

    At this point, I believe it would be interesting to redo every 2500k benchmark that has a delta of 20% or more compared to the 860, which would correspond to the operating frequency difference of the two cpus, but this time clocked at 4Ghz, as the i7-860 is.

    To stretch things even more, I could also disable the HyperThreading of the i7-860 for another run of these above 20% deltas, so we can see what a Lynnfield can do without the help of hyperthreading, but this is a story for another time I guess!

    Of course I will come back to post the complete version of this test, with graphs, bbcode with video links etc. :)
     
  9. Blacklac

    Blacklac Member Guru

    Messages:
    144
    Likes Received:
    0
    GPU:
    EVGA GTX 960 SSC 2GB
    Nice! Amazing work. I'm actually surprised the 860 is that much better than the Q9550. Not surprised the 860 is that close to the Sandy Bridge, though.
     
  10. Noisiv

    Noisiv Ancient Guru

    Messages:
    8,026
    Likes Received:
    1,321
    GPU:
    2070 Super
    yes, run both 860 and 2500k @4GHz, with HT
    and without( for the science!)
     

  11. CrazY_Milojko

    CrazY_Milojko Ancient Guru

    Messages:
    2,320
    Likes Received:
    763
    GPU:
    Asus STRIX 1070 OC
    OP, thanks for benchmarks, a lot of work for sure, much appreciated. I did something similar maybe 10 years ago, but no way I could do that now, time needed for something like that I don't have atm, my wife would probably kill me :D

    2500K is Core i5, it doesn't have HT
     
  12. Netherwind

    Netherwind Ancient Guru

    Messages:
    7,754
    Likes Received:
    1,600
    GPU:
    MSI 3080 Gaming X
    Well done!

    Next up 3570K vs 4790K vs 5280K pretty please? ;)
     
  13. Undying

    Undying Ancient Guru

    Messages:
    16,429
    Likes Received:
    5,366
    GPU:
    Aorus RX580 XTR 8GB
    Well difference between 2500k and 3570k is like 200mhz clock to clock. You'll probably need 4.5-4.6ghz to match those Sandy charts here.
     
  14. Dorlor

    Dorlor Banned

    Messages:
    1,708
    Likes Received:
    0
    GPU:
    2x evga 780 ti sc acx
    In a purely cpu bottlenecked situation, the 5820k will be roughly 60% faster than the 3570k (assuming the game has perfect multicore scaling, such as frostbite or cryengine) :)
     
  15. Netherwind

    Netherwind Ancient Guru

    Messages:
    7,754
    Likes Received:
    1,600
    GPU:
    MSI 3080 Gaming X
    Witch wont happen since my CPU wont take more punishment :)

    That's interesting, thanks.
     

  16. 0blivious

    0blivious Ancient Guru

    Messages:
    3,134
    Likes Received:
    630
    GPU:
    5700XT / GTX1070
    Good stuff. :)


    I have a ton of benchmark info (gathered over a few years) that doesn't necessarily match properly. Under my roof I have an i7 4790K, i7 2600K, i7 860, and i3 2120 rigs with a gtx970, gtx760, HD6970 & HD5770. My kids have i5 4670K and q6600 rigs at their Mom's place with a gtx970 & HD6950. I'm testing a budget gamer I built yesterday with an i3 4160 & gtx560.

    I may try to organize and post these results in a thread someday if they are of any interest to other users.
     
  17. cc5037

    cc5037 New Member

    Messages:
    6
    Likes Received:
    0
    GPU:
    290 / 980
    This is really good stuff man, thanks!
     
  18. Mumriken

    Mumriken New Member

    Messages:
    1
    Likes Received:
    0
    GPU:
    780 Ti Matrix Platinum
    Great testing, I really appreciate it since i still use an oced x3363 (same as q9550)as an HTPC to play project cars on my tv with g25 wheel.
    It is struggling when there is a lot of cars and gpu usage drops to around 30%.
     
  19. ravaged

    ravaged Member Guru

    Messages:
    196
    Likes Received:
    2
    GPU:
    ASUS 2080 Super
    Nice work, man. These are very useful. Not everyone is in the same financial situation, so being able to draw data from this sort of testing is very valuable. I know at certain points I wish I had something this extensive to compare, to be better able discern where I can skimp a little and where I should be putting more of my budget.
     
  20. Blacklac

    Blacklac Member Guru

    Messages:
    144
    Likes Received:
    0
    GPU:
    EVGA GTX 960 SSC 2GB
    I'm just curious, does your Q9550 go passed 4Ghz? That's a great OC, don't get me wrong, but its usually fairly easy to hit that mark, too. Depending on the Mobo, or stepping...
     

Share This Page