laserbeam Posted March 1, 2005 Share Posted March 1, 2005 I have just gone through the process of ramping the 1GB matched pair of modules I installed in my new system build (testing with memtest86 v1.51). Starting with timings of 2-2-2-10 (tRAS 10 seems optimal with my mobo) I got to 220Mhz FSB (CPU multiplier reduced to 8) before encountering any errors. Since then I've increased timings through 2.5-3-3-10 as I raised FSB. All frequencies generate errors on test number 8. I've tried raising DIMM voltage to 2.8V but that doesn't help either. So, I thought I'd test the modules at the certified timings (DDR550 2.5-4-4-8), turning the cpu multiplier down in the process and setting vdimm to 2.7V. Again, memtest blew out on test number 8. Seeing as these modules won't pass memtest86 at their certified timings I'm thinking they should be replaced unless there are bios settings I've overlooked that are causing me grief. Some help and/or warranty guidance will be greatly appreciated. Components in play are: Athlon64 3200+ Winchester (cooled with XP-120) DFI Lanparty UT NF4 Sli-D 1GC Corsair Twin-X 4400C25PT OCZ Powerstream 600W PSU Link to comment Share on other sites More sharing options...
Corsair Employees RAM GUY Posted March 1, 2005 Corsair Employees Share Posted March 1, 2005 From X86-secret.com - I'm getting errors in test #5 and/or #8 and have read a lot about it. Yes there are just about enough discussions on the topic to fill a book, but it all boils down to the answer given above. The only thing that can be said is that many times, when memory latencies are incorrectly set in the BIOS you will experience errors in test #5 and #8. (Though #8 does not exist anymore as of version 1.40 and might be reinstated as a different test in a later version.) This does however NOT mean that errors in these tests are always the cause of incorrect settings; your memory might just as well be defective. - I'm getting errors in test #x, what does that mean? Interpreting memtest results is as scientific an endeavo as testing whether a person is a witch by the methods used in Monty Python's Holy Grail. In short, don't even start, it's not going to get you anywhere. Just interpret any error as you should any other and use the methods described in the previous question to determine the cause. Bottom line please just run the default tests and then if you are having a problem with the system lets try and find the cause. Link to comment Share on other sites More sharing options...
laserbeam Posted March 1, 2005 Author Share Posted March 1, 2005 I've been running the default tests. Test 8 always seems to be the sticking point. Link to comment Share on other sites More sharing options...
Corsair Employees RAM GUY Posted March 1, 2005 Corsair Employees Share Posted March 1, 2005 Test 8 should not run by default. What version of memtest are you using and where did you get it? Link to comment Share on other sites More sharing options...
laserbeam Posted March 2, 2005 Author Share Posted March 2, 2005 I downloaded it from guru3d in the new download section. It's version 1.51. Link to comment Share on other sites More sharing options...
Corsair Employees RAM GUY Posted March 2, 2005 Corsair Employees Share Posted March 2, 2005 Please try http://www.memtest.org. But with AMD errors in test 5 or 8 usually suggest some other problem. Link to comment Share on other sites More sharing options...
laserbeam Posted March 2, 2005 Author Share Posted March 2, 2005 Have downloaded from the site you've listed and will try again. What other problems are you referring to? I've left the HTT multiplier at auto. Should I be manually setting this to 3? Then again, I'd have expected more fundametal problems if Hypertransport settings were the problem. Link to comment Share on other sites More sharing options...
Corsair Employees RAM GUY Posted March 2, 2005 Corsair Employees Share Posted March 2, 2005 It's hard to say where you would get errors, as no 2 systems will behave the same. But errors in test 8 at default settings at DDR400 with these modules which are tested in a ASUS P4C800-E with 2.4C at 275 MHz before they leave our factory would suggest some other problem, other than the modules. Link to comment Share on other sites More sharing options...
laserbeam Posted March 2, 2005 Author Share Posted March 2, 2005 Okay, I've made some progress. By turning down the CPU multiplier to 5.5 (from 7.0) I'm able to run memtest86 at 275Mhz with certified timings. of course, I find this surprising as it means I need to raise voltage for my CPU and that it was experiencing problems while clocked well below 2Ghz. As you're the RAM guru can you explain this interaction? I'm gonna continue trials and will keep you informed. Link to comment Share on other sites More sharing options...
Corsair Employees RAM GUY Posted March 2, 2005 Corsair Employees Share Posted March 2, 2005 This MB is still verry new and may take some time for the Bios to mature. Also what do you have the H/T set to? You might set it to 2 or 3 and see if that helps. Link to comment Share on other sites More sharing options...
laserbeam Posted March 3, 2005 Author Share Posted March 3, 2005 I'm finding that a CPU multiplier anywhere above 5.5 results in test 8 failing. And adjusting vcore doesn't seem to help either. I'm consulting other forums to see if this is a mobo issue. Link to comment Share on other sites More sharing options...
Corsair Employees RAM GUY Posted March 3, 2005 Corsair Employees Share Posted March 3, 2005 What do you have the hyper Transport set too? And what CPU core do you have? Link to comment Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.