Please report all spam threads, posts and suspicious members. We receive spam notifications and will take immediate action!
Page 3 of 4 FirstFirst 1234 LastLast
Results 21 to 30 of 37

Thread: tREAD and CAS Latency?




  1. #21
    Join Date
    Jun 2009
    Posts
    57

    Default Re: tREAD and CAS Latency?

    Quote Originally Posted by Chike View Post
    Is it possible to use 450x9@2.40B DDR-1080 5-5-5-18 instead?
    This should give more bandwith with no or minimal increase of latency.
    very early on in this process,
    I struggled heavily with attempting a 9.5x400@2.66D config. And also a few failed attempts at 9.5x400@2.40B. I choked it up to the amount of DRAM I was working with (4x2gb) as an insurmountable FSB>333 obstacle, running anything other than 1:1.

    But, I've learned quite a bit since then,
    and may revisit a few strap options before calling it quits.

    At the very least, I'll try the 2.00B strap. I've read that if you can get stable with it, it usually is less stressing to the MCH and may allow one-notch lower tRD and/or tighter timings.
    case: Gigabyte 'Triton 180'
    power: Corsair 'HX-620'
    motherboard: Gigabyte 'GA-EP45-UD3P' (rev.1.1, F9)
    cpu: Intel 'e8500' (E0)
    cooling: Xigmatek 'HDT-S1283'
    memory: OCZ 'Reaper' 4x2gb (OCZ2RPR10664GK)
    storage: Western Digital 'Black' (500gb)
    video: EVGA 'GeForce' GTX-260
    display: Samsung 'SyncMaster' 2333sw

  2. #22
    Join Date
    Jun 2009
    Posts
    57

    Default Re: tREAD and CAS Latency?

    Quote Originally Posted by corlay View Post
    But, I've learned quite a bit since then, and may revisit a few strap options before calling it quits.
    I played with the B-strap last night,
    and may actually be on to something:

    9.0x450@2.40B, tRD=8, 5-5-5-15

    I'm closing-in on stability,
    already having established the best voltage range for vMCH (1.18-1.20) and vDRAM (2.26-2.28); I've achieved 1HR. stability in p95 blended testing. That usually means that I'm getting close. Tonight I'll play around with Vtt, MCH Ref and Driving Strength Profiles to see if that helps any. If not, I'll loosen my Standard Timings to 5-5/6-5/6-16/18, and I bet you I'll get there.

    What's nice about moving to a divider greater than 1:1, is that the vMCH required is much lower. Especially on the B-Strap. I needed 1.26-1.28 on the 2.00D strap.Although vDRAm required for this increased bandwidth is significant. I only needed 2.16-2.18 in the 2.00D, vs. 2.28v on the 2.40B. But I'm still just under max. spec for my kit, so that's ok with me.
    case: Gigabyte 'Triton 180'
    power: Corsair 'HX-620'
    motherboard: Gigabyte 'GA-EP45-UD3P' (rev.1.1, F9)
    cpu: Intel 'e8500' (E0)
    cooling: Xigmatek 'HDT-S1283'
    memory: OCZ 'Reaper' 4x2gb (OCZ2RPR10664GK)
    storage: Western Digital 'Black' (500gb)
    video: EVGA 'GeForce' GTX-260
    display: Samsung 'SyncMaster' 2333sw

  3. #23
    Chike is offline Senior Member
    Join Date
    Jun 2009
    Location
    Israel
    Posts
    6,531

    Default Re: tREAD and CAS Latency?

    Quote Originally Posted by corlay View Post
    I played with the B-strap last night,
    and may actually be on to something:

    9.0x450@2.40B, tRD=8, 5-5-5-15
    As I read in the forum that D strap is doing better at the low 400s FSB while B strap is doing better at the high 400s.

    Quote Originally Posted by corlay
    I'm closing-in on stability,
    already having established the best voltage range for vMCH (1.18-1.20) and vDRAM (2.26-2.28); I've achieved 1HR. stability in p95 blended testing. That usually means that I'm getting close. Tonight I'll play around with Vtt, MCH Ref and Driving Strength Profiles to see if that helps any. If not, I'll loosen my Standard Timings to 5-5/6-5/6-16/18, and I bet you I'll get there.
    I personally came to the conclusion it does not worth all the hassle. If you decide you want 100MHz more CPU by raising FSB and any teaking you made become useless.
    I leave my memory timings on auto with EPP selected and only tRD set manually (BIOS thinks 10, I think 7 is better :-) )
    But a picture worth a thousand words, so here a quick test I've made, 400FSB vs 411FSB with timings on auto. It looks like timings go south, but do I suffer any lose?
    Decide for youtself:

  4. #24

    Default Re: tREAD and CAS Latency?

    It would be great if you can get her stable with the 2.4D. Even if you have to settle with 5-6-6-18, the memory performance should be superior to 2D 4-5-5-15 by around 500M/s read and 2.5ns latency. Good luck.
    QX9650 batch L739A761/ GA-EP45-UD3P/ Kingston KHX9200 4x1G

  5. #25
    Join Date
    Jun 2009
    Posts
    57

    Default Re: tREAD and CAS Latency?

    Quote Originally Posted by gampamu View Post
    It would be great if you can get her stable with the 2.4D. Even if you have to settle with 5-6-6-18, the memory performance should be superior to 2D 4-5-5-15 by around 500M/s read and 2.5ns latency. Good luck.
    Hey, thanks for the encouragement.

    Played around with settings a bit more last night, and got 3hrs. stability with vMCH=1.16. I set it to run p95 blended again that same config but with looser 5-6-6-16 DRAM timings. Hopefully it's still going when I return home tonight....
    case: Gigabyte 'Triton 180'
    power: Corsair 'HX-620'
    motherboard: Gigabyte 'GA-EP45-UD3P' (rev.1.1, F9)
    cpu: Intel 'e8500' (E0)
    cooling: Xigmatek 'HDT-S1283'
    memory: OCZ 'Reaper' 4x2gb (OCZ2RPR10664GK)
    storage: Western Digital 'Black' (500gb)
    video: EVGA 'GeForce' GTX-260
    display: Samsung 'SyncMaster' 2333sw

  6. #26
    Join Date
    Jun 2009
    Posts
    57

    Default Re: tREAD and CAS Latency?

    ok,
    really closing-in on stability now.

    vMCH=1.16 w/ 5-5-5-15 timings yielded 3hrs. stability.
    vMCH-1.16 w/ 5-6-6-16 timings yielded 6hrs. stability.
    and finally,
    vMCH=1.16 w/ 5-6-6-18 timings yielded 7min.of stability.

    This is maddening work.

    But last night, I reverted my timings back to 5-5-5-15, and started trying different Driving Strength Profile (DSP) settings (all previous configs used 'Auto'...).

    DSP=1066 yielded 1hr. stability
    DSP=OC-1200 is currently into the 10hr. mark and still going strong!

    This just might be the one.

    Side note: Ever notice that, seemingly with a torture test that is destined to fail, the test-sequence in one core that is lagging behind the other will skip the last test in the FFT group? I suppose this is p95's attempt to "Catch up" the lagging core, or maybe just an indication of an instability in the current config. Every time I notice this behavior, the test does eventually fail. Anyone else notice this?
    case: Gigabyte 'Triton 180'
    power: Corsair 'HX-620'
    motherboard: Gigabyte 'GA-EP45-UD3P' (rev.1.1, F9)
    cpu: Intel 'e8500' (E0)
    cooling: Xigmatek 'HDT-S1283'
    memory: OCZ 'Reaper' 4x2gb (OCZ2RPR10664GK)
    storage: Western Digital 'Black' (500gb)
    video: EVGA 'GeForce' GTX-260
    display: Samsung 'SyncMaster' 2333sw

  7. #27
    Join Date
    Mar 2009
    Location
    Romania :(
    Posts
    95

    Default Re: tREAD and CAS Latency?

    I think the lagging core happens because windows and other application are still using your CPU, thus making it incapable of keeping track with the other core. But I've noticed the test skipping even while none of the cores was falling behind.
    GA-EP45-UD3R (F12)
    Zalman CNPS-7000C-CU
    E8400@3.6GHz
    2x2GB A-DATA Vitesta 800+ CL4 Extreme Edition @ 1066MHz@2.0V (5-5-5-15)
    Asus EN9800GT Ultimate @ 810/2000/2200
    Seagate 250GB + WD 500GB + WD 640GB
    Asus DRW 1608-P
    BenQ G2010WP
    Thermaltake TR2 500W
    Windows 7 Pro x64

  8. #28
    Chike is offline Senior Member
    Join Date
    Jun 2009
    Location
    Israel
    Posts
    6,531

    Default Re: tREAD and CAS Latency?

    Quote Originally Posted by corlay View Post
    Side note: Ever notice that, seemingly with a torture test that is destined to fail, the test-sequence in one core that is lagging behind the other will skip the last test in the FFT group? I suppose this is p95's attempt to "Catch up" the lagging core, or maybe just an indication of an instability in the current config. Every time I notice this behavior, the test does eventually fail. Anyone else notice this?
    I have noticed the lag. not the fact it would eventually fail. Sometimes the lagging test will catch up.
    Prime95 run it's test at idle or screen saver priority, that mean that almost any process in the system will be given priority over them. I couldn't determine if the tests are affined to a specific CPU or roaming between them but it seems like the second when I run only one thread.
    Another hing I observed is that the test is very likely to fail, if it should fail, when starting p95 as soon as possible when windows starts, while the system is still loading stuff and disk activity is high.

  9. #29
    Join Date
    Jun 2009
    Posts
    57

    Default Re: tREAD and CAS Latency?

    just to clarify: I've had tourtue tests run 24hrs. stable, where one (of 2...) cores, at times, has lagged behind the other by as many as 4 tests; and then at other times was perfectly in synch.

    It's the behavior of a skipped test, that seems to indicate probable instability, that I'm suggesting - not the just that there is a lagging core.
    Last edited by corlay; 07-23-2009 at 10:00 PM.
    case: Gigabyte 'Triton 180'
    power: Corsair 'HX-620'
    motherboard: Gigabyte 'GA-EP45-UD3P' (rev.1.1, F9)
    cpu: Intel 'e8500' (E0)
    cooling: Xigmatek 'HDT-S1283'
    memory: OCZ 'Reaper' 4x2gb (OCZ2RPR10664GK)
    storage: Western Digital 'Black' (500gb)
    video: EVGA 'GeForce' GTX-260
    display: Samsung 'SyncMaster' 2333sw

  10. #30
    Chucko is offline Member
    Join Date
    Jan 2009
    Location
    6 blocks from AMD HQ
    Posts
    112

    Default Re: tREAD and CAS Latency?

    I'm trying to optimize my OC now - just got a Q9550 and plugged it into the -UD3P. 8GB RAM is stable at 5-4-4-13 @ 890MHz; these Reapers won't do CL4. All at 2.00B multiplier.

    Static tRead of 8 didn't work. Neither did 9. I had to restore it to Auto. I haven't played with the Driving Strength Profiles yet with this CPU. I'm hopeful I can get Static tRead down to 8 at some point. So I'll be following this thread.

Page 3 of 4 FirstFirst 1234 LastLast

Thread Information

Users Browsing this Thread

There are currently 1 users browsing this thread. (0 members and 1 guests)

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •