Outline ·
[ Standard ] ·
Linear+
TCCD/TCC5 discussions, memory tuning, voltage, etc.
|
babyelf
|
 Aug 25 2005, 12:03 AM
|
|
QUOTE(kelvin_hata @ Aug 25 2005, 12:00 AM) i didt set all...onli set the timing...cos last time i buy all stuff, n i cant find gskill ram in lyn tat y i tempory use my old bh-5 to run 1st... last times buy tis baby at rm400 onli 256 corsair 3200 rev1.1, on b'cos run at single channel n didt do any benmark, jus use tat ram to burn in my proc  but ur superpi using sandy vy fast wat...abt 27sec finish 1mb jor... so 265 just 2-2-2-5 in bios? other timings on auto? that superpi1mb is suicide shot.. nothing near stable.. maybe will get it higher soon.. just got this  Attached thumbnail(s)
|
|
|
|
|
|
TSsoulfly
|
Aug 25 2005, 12:53 PM
|
revving towards 10,000 rpm
|
My current setting at 2.8V with very minimum error in Memtest86+, maybe need more burn-in  (Using DFI nF3 250Gb 4/15 BIOS w/ 4Vdimm mod) QUOTE Genie BIOS Settings:
FSB Bus Frequency - 300
AGP Bus Frequency - 67mhz
Clock Spread Spectrum - Disabled
LDT Downstream Width - Auto
LDT Upstream Width - Auto
LDT/FSB Frequency - Auto
CPU/FSB Frequency Ratio - 8x
K8 Cool 'n' Quiet Support - Disable
CPU VID Control - 1.4V
CPU VID Special Control - Above VID * 0.126%
Chip Set Voltage Control - 1.90v
AGP Voltage Control - 1.60v
DRAM Voltage Control - 2.80v
DRAM Configuration:
DRAM Frequency Set - 200 (DRAM/FSB:1/1)
Command Per Clock (CPC) - Enable
CAS Latency Control (Tcl) - 2.5
RAS# to CAS# delay (Trcd) - 04 Bus Clocks
Min RAS# active time (Tras) - 7 Bus Clocks
Row precharge time (Trp) - 03 Bus Clocks
Row Cycle time (Trc) - 07 Bus Clocks
Row refresh cyc time (Trfc) - 14 Bus Clocks
Row to Row delay (Trrd) - 03 Bus Clocks
Write recovery time (Twr) - 03 Bus Clocks
Write to Read delay (Twtr) - 02 Bus Clocks
Read to Write delay (Trwt) - 03 Bus Clocks
Refresh Period (Tref) - 4708 Cycles
Write CAS Latency (Twcl) - 1
DRAM Bank Interleave - Enabled
DQS Skew Control - Increase
DQS Skew Value - 255
DRAM Drive Strength - Level 2
Max Async Latency - 7.0ns
Read Preamble Time - 5.0ns
IdleCycle Limit - 016 Cycles
Dynamic Counter - Disable
R/W Queue Bypass - 16 x
Bypass Max - 07 x
32 Byte Granularity - Disable(8 Bursts) |
|
|
|
|
|
kelvin_hata
|
Aug 26 2005, 12:14 AM
|
|
|
cannot run at ???
Row to Row delay (Trrd) - 02 Bus Clocks
Write recovery time (Twr) - 02 Bus Clocks
Write to Read delay (Twtr) - 02 Bus Clocks
Read to Write delay (Trwt) - 02 Bus Clocks
|
|
|
|
|
|
TSsoulfly
|
Aug 26 2005, 12:23 AM
|
revving towards 10,000 rpm
|
haven't tried that though i tot lower timing is more stable  |
|
|
|
|
|
kelvin_hata
|
Aug 26 2005, 12:28 AM
|
|
|
QUOTE(soulfly @ Aug 26 2005, 12:23 AM) haven't tried that though i tot lower timing is more stable dunno loh...cos i can run 3dmark wit tis setting...if i put 3,3,2,3 or 3,2,2,3 got problem... but put it 2,2,2,2 then no problem jor  i oso pelik  mb u can try lah hmm...here my old bh-5 run at 2-2-2-7@267@3.7v@1:1@1t This post has been edited by kelvin_hata: Aug 26 2005, 12:37 AM Attached thumbnail(s)
|
|
|
|
|
|
pcmoddingmy
|
Aug 26 2005, 06:56 PM
|
|
|
Eh? Whats BH-5 doing in a TCCD thread? Anyways, I find that it is much easier to overclock 2x256MH TCCDs to 300+MHz with 1T than 2x512MB TCCDs. Here's my setup doing 300MHz 2.5-4-3-7-1T @ 2.7V on Corsair 3200XL 2x256MB TCCD 513  My current highest stable record at 320MHz 2.5-4-4-8-1T 2.7V on 3200XL 2x256MB TCCD modules  My Setup: A64 3000+ WinChester - DFI LP UT nF3 Ultra-D (BIOS 705) - Corsair XMS 3200XL 2x256MB 2.7v - LTD 3x 1.7v - CPU 1.6v - vAGP 1.6v Edit: All tests done in DUAL CHANNEL. This post has been edited by pcmoddingmy: Aug 26 2005, 06:57 PM |
|
|
|
|
|
TSsoulfly
|
Aug 26 2005, 07:53 PM
|
revving towards 10,000 rpm
|
nice one there dude
but 256MB modules sure can clock high easily due to the single bank factor
|
|
|
|
|
|
pcmoddingmy
|
Aug 26 2005, 08:19 PM
|
|
|
QUOTE(soulfly @ Aug 26 2005, 07:53 PM) nice one there dude but 256MB modules sure can clock high easily due to the single bank factor Yup.. its either that or my board hates dual sided memory module. |
|
|
|
|
|
PCcrazy
|
Aug 29 2005, 11:10 PM
|
|
|
I found this thread informative so I have decided to pinned it up for easy reference.
|
|
|
|
|
|
ahpaul82
|
 Aug 30 2005, 03:23 PM
|
|
|
Dont know whether is it a suitable place to post or not .. Anything idea what cause failed ? 2x512MB Corsair TwinX DDR400 v1.4, running at 270MHz, 2.5-4-3-7,1T, 3.2v memtest run >7hours same setting with 3.0v no error. but in window i need to adjust to 3.2v   here is my tweaker setting :  |
|
|
|
|
|
kianwee
|
Aug 30 2005, 04:35 PM
|
|
|
Try:
Max Async Latency to 8
Read preable to 6
Idle Cycle Limit to 256
|
|
|
|
|
|
TSsoulfly
|
Aug 30 2005, 05:01 PM
|
revving towards 10,000 rpm
|
ahpaul... i think your voltage is too high for only 270mhz ler...
try back down to 2.7v or 2.8v
yours are tccds right?
|
|
|
|
|
|
ahpaul82
|
 Aug 30 2005, 05:04 PM
|
|
|
Which one more realible ? or should i said more trusted ? Prime95 or SP2004 ?  The Prime95 failed but SP2004 still running no error yet.. This post has been edited by ahpaul82: Aug 30 2005, 05:16 PM |
|
|
|
|
|
ahpaul82
|
Aug 30 2005, 05:05 PM
|
|
|
QUOTE(soulfly @ Aug 30 2005, 05:01 PM) ahpaul... i think your voltage is too high for only 270mhz ler... try back down to 2.7v or 2.8v yours are tccds right? voltage too high? but i think mine's TCCD is not as good as you guy one leh..  ok i will try lower the voltage and kianwee setting . |
|
|
|
|
|
kianwee
|
Aug 30 2005, 07:24 PM
|
|
|
QUOTE(ahpaul82 @ Aug 30 2005, 05:04 PM) Which one more realible ? or should i said more trusted ? Prime95 or SP2004 ? The Prime95 failed but SP2004 still running no error yet.. Which test did you run? |
|
|
|
|
|
TSsoulfly
|
Aug 30 2005, 07:35 PM
|
revving towards 10,000 rpm
|
find your maximum memory oc and tune the memory timings to get memtest86 pass at test #5, tune until you can get it run at 2.8V or less. if you're trying to run at 300mhz, put the vdimm at 2.9v
from there, after you can pass memtest86 #5, try to enter windows. if you cannot stable in windows, you might need to increase your chipset, agp or cpu voltage. or maybe u need to bump your agp speed to 67mhz. no need to alter your vdimm anymore.
|
|
|
|
|
|
pcmoddingmy
|
Aug 30 2005, 07:49 PM
|
|
|
QUOTE(ahpaul82 @ Aug 30 2005, 05:04 PM) Which one more realible ? or should i said more trusted ? Prime95 or SP2004 ? The Prime95 failed but SP2004 still running no error yet.. Actually both are the same, using the same Lucas-Lehmer iterations. To stress the RAM, use In-place Large FFTs or Blend Tests. However, Blend test will use up all of your RAM and causing extreme lags. So, I prefer Large FFTs. If you have Prime95 for some reason, try Windows MemTest (http://hcidesign.com/memtest/). Let it pass atleast 3000% coverage or 2 hours error free for maximum stability. Edit: Screw BIOS Memtest!  This post has been edited by pcmoddingmy: Aug 30 2005, 07:50 PM |
|
|
|
|
|
TSsoulfly
|
Aug 30 2005, 08:01 PM
|
revving towards 10,000 rpm
|
bios memtest purely tests the ram, that's why even OCZ still use this utility for testing.
when you go into windows..... any memtest based tool is affected by the processor and overall system stability also, that's why you can pass bios/dos memtest but cannot pass windows memory tester.
|
|
|
|
|
|
ahpaul82
|
 Aug 30 2005, 08:09 PM
|
|
|
Im using Blend Test to test my RAM stability. Thanks for the kianwee, soulfly and pcmoddingmy's sugguestion. I will give it a try. Thanks a lot. I think the max speed for my Corsair RAM can run is 275MHz only . last time did a test with my Winchester and can run >8hours no problem but with kinda loose timing. i think is 2.5-4-4-8,1T, 3.2v. I forgot how im able to do that timing last time..  >>>  Now don't think about RAM first, test my processor how high can go but with low voltage .. |
|
|
|
|
|
gsan
|
Aug 31 2005, 03:40 AM
|
|
|
QUOTE(soulfly @ Aug 30 2005, 07:35 PM) after you can pass memtest86 #5, try to enter windows. if you cannot stable in windows, you might need to increase your chipset, agp or cpu voltage. or maybe u need to bump your agp speed to 67mhz. no need to alter your vdimm anymore. what do you mean by "cannot stable in windows" ? my GH run 235 @ 2-2-2-5 stable in memtest65#5 for more than 13hours and 240 @ 2-2-2-5 for more than 9hours but with 2 errors only. but come to windows, few minute later, the pc automatic reboot, increase the agp voltage or speed will make it stable? |
|
|
|
|
Quote
0.0236sec
0.43
6 queries
GZIP Disabled