TWINX1024-3200LLPT on A7V600

[JasonPW]

Hi All, well I'm stuck. Although the abouve board only has a single channel memory controller I wanted to replace my not so LL memory modules with something a little more nippy. system setup: Barton 2800+ OC'd to 3200+ speed (100% stable) A7V600 rev1.0 (1005 BIOS) Crucial Radeon 9800 Pro (Cat 3.8) Audigy 2 OEM PCI 450watt enermax PSU My original memory was 2x 512MB Crucial 3200 DDR (Cas3) set to 2.5,2,2,6 (1T not enabled) and was 100% stable. Anyway the abouve "Corsair" modules I selected to give me the same capacity will not run even individually (tested all configs, 1 or 2 modules different slots) at the settings set by their SPD. only when I set the CAS to 2.5 will they pass the below tests. memtest86 - Passed. (voltage had to be increased to 2.75V or failed) Microsoft Windows mem test - Passed. (voltage had to be increased to 2.75v or failed) 3Dmark 2001SE (Fails - crashes to desktop at SPD 2,3,2,6) Basically I can only get the new ram to work at 2.5,3,2,6 Switching back to my original (cheaper) ram and going back to their best timings restores stability and passes with memory volts on "Auto". Basically as far as I'm concerned ram should work within the timings specified out of the box without over volting. What I don't want to do is back them off so far that my cheaper "branded" and overclocked ram matches or excedes the new rams performance, without stability issiues. What are the recomended timings for this MB and the abouve ram? Any other suggestions?

[RAM GUY]

The MB is not a dual channel MB and running 2 modules may need to be set a little more relaxed. I would try with our modules installed in slots 1 and 3 to reset the bios or just load setup defaults and then set the timings manually to Cass 2.5-3-2-6 with the Rass to Cass set to "3" and please set the Dim Voltage to 2.7 Volts. Then please run [url]www.memtest86.com[/url] and if you get errors please test the modules one at a time with [url]www.memtest86.com[/url] and let's make sure it's just a bad module.

[JasonPW]

Hi, I only use slots 1 and 3 anyway. It was the only reliable way to run 2 modules on my previous A7V333 so I re-used it on this MB. Both the CPU and memory are using 1:1 ratio and are in sync. The "OS" is XP SP1 Pro. However neither of the TwinX modules will run at their rated specification "individually" in any slot, regardless of voltage. They will pass (single test run only) both memtest86 and windows memory tester but 3d apps CTD or lockup and I get registry and NTFS corruption problems (recoverable). I may have to run multiple passes to show the errors in memory test programs which I will try later. There is also a BIOS update for the MB which adds compatibility for Kingston and Kingmax 3200 DDR so I may try that as well. At the moment you ram will only run at your suggested (and allready tested and in use at the moment) 2.5-3-2-6. In this mode they are 100% reliable but slower than my CAS 3 memory overclocked which are just as reliable. And there is now way I would even consider using what is badged as "low latency" ram at cas 3. My "Crucial" brand original memory which was only rated at cas 3 will run 100% relaible at 2.5-2-2-6. This memory cost me less and is currently performing better than your "low latency" ram. This with the voltage at "auto" which is 2.75V anyway on this MB (2.7 is not an option on the ASUS BIOS). If this were a simple 2 module on a single channel controller problem then I would expect the TWINX modules to run at their rated SPD individually which they are not.

[RAM GUY]

JasonPW I would try testing the modules in another MB if possible as that would help to isolate if it's the memory or MB. But honestly I would run AIDA32 with the modules and see what the timings are actually running. But it may be that the LL modules are just too aggressive for this MB. You might try setting the timings to Cass 2.5-3-3-7 and see if the performance in your benchmark improves if so that would suggest that the MB may not run well with more aggressive memory. But I also would suggest the latest bios as well as many times they will refine performance in new bios releases And I have no problem replacing your modules but first I would try testing on another MB. Please send us a email with a copy of the [URL=http://www.corsairmemory.com/main/rma_request.xls]form[/URL] or all of your info name address and phone# and the Module part# and copy the link to this post and email it to [email]Warranty@corsairmemory.com[/email]. Or you can also use the [URL=http://www.corsairmemory.com/main/rma_request.html]On Line RMA Request Form[/URL] as well. If after 1 day or 24 hours excluding weekends you do not get the rma please email the same to [email]warranty@corsairmemory.com[/email] and we will help to resolve it.

[JasonPW]

Hi again, Well I have tried all I can with these modules on this PC. Just for your reference I was using the 1005 BIOS on this board. I have now tried the 1006 (latest release version). This is the Kingston/Kingmax fix release. However this version definatly reads the SPD wrong or deliberatly miss reads it. It reads these modules as 2.5-3-3-8 but checking with AIDA32 it reports 2.5-3-3-6. Anyway it dosen't fix any of the issiues i have been having. Also because I don't have any time for long term testing and knowing 1005 isn't causing any issiues on this system I have now rolled back to 1005. I will try on another motherboard when I have more time. It will not however be a Nforce2 chipset due to the amount of problems I have had setting one up in the past. (Different Story - bugged drivers). Are you aware of any issiues on the VIA KT600 chipset? I was wondering if it is a dram chip/chipset compatibility problem. Just to confirm that these modules are running fine now at 2.5-3-2-6 on the A7V600 rev 1.0. However this is dropping my memory performance by around 5%. I will see if I can get these modules to run at their SPD on another system before contacting the supplier or selling them on. Thanks.

[RAM GUY]

Honestly that's about right from the reports I have had within the forum. And for the most part probably about as good as it will get because of the chipset. But I think you have to consider a few things about Nforce2. First it’s been the hottest selling chipset for AMD platform since it's release about 12 months ago. So it would stand to reason that there was a lot having trouble. And for the most part what I am seeing is configuration issues and not compatibility problems. It's still a great chipset for AMD and runs well. I have 3 systems with Nforce2 and all run extremely well. FYI this is the main cause of problems or actually because most are not aware of chipset limitations. If the FSB of your CPU is 400 For Intel and 100 MHz for AMD, the memory speed supports only DDR 266. If the FSB of your CPU is 533 For Intel and 133 MHz for AMD, the memory speed supports DDR 266/333 If the FSB of your CPU is 800 MHz for Intel and 200 MHz for AMD, the memory speed supports DDR 333/400 But I would give this MB some time to mature and a new bios may help performance and over-clock-ability

[JasonPW]

Hi, Well just posting a followup after my testing. I have been loaned another pair of DDR from another manufacturer to test. These are specced at 2-3-2-6 1T. At these settings they would pass memtest86 and microsoft memory test with zero errors. However I would get CTD's with 3Dmark 2001SE and 2003. The only way to pass these tests was to drop to CAS 2.5. In this mode due they were like the Corsair XMS LL modules and passed and were 100% stable. Despite altering BIOS and voltage settings I was unable to get to CAS2 with this chipset. The only thing I can say is that due to these modules being 1T command capable I get higher performance than with the previous "corsair" modules so I am trying to swap them for the pair loaned. Basically I'm concluding that this is either a KT600 chipset or A7V600 rev1 issiue and not the fault of the DDR. Although I may get my hands on a MSI KT600 chipset board over the holidays to test your modules on. Thanks for your help anyway. Merry Christmas and a Happy New Year.