![]() |
|
|
#1 |
|
Oct 2005
22 Posts |
Hello.
I did a stress test with the prime 95 v.24.14 to check my ram and had stange results. Here is my material: MB: Asus P4P800-Deluxe (with last bios revision) CPU: Intel P4 2.6C (800MHz FSB) RAM: 2x Kingmax 512Mb DDR400 MPWC22D-38KT3R 2x Kingston 512mB DDR400 KVR400X64C25 GC: ATI 9800 Power: LC-Power 550W HD: IMB/Hitachi deskstar 120Mb I launched the torture test and had directely the message "FATAL ERROR: Rounding was 0.5, expected less than 0.4" It was strange because my computer is not overclocked and all specs of the RAM are corrects (2.5 3 3 8 timing, 2.65V, turbo disabled and Hyper path disabled) I tried only with the two Kingston RAM (slot 2 and 4) and results are OK after 3hours of test. I tried also with the 2 Kingmax only and obtain the same results (slot 2 and 4). I change the 2 Kingmax to the slot 1 and 3 and same results. So my conclusion is that the 4 RAM are ok. But why the problem occurs only when the 4 RAM are present? I checked the voltage with EVEREST and every thing seems to be stable with all configuation of RAM (CPU voltage@1.52V +3.3@3.39V +12@11.92V). Difference between Kingston and Kingmax RAM is that normal voltage is 2.5V for the KM and 2.6V for the KS. So I plugged the 2 KS to the 2 slots closer to the CPU and the 2 KM to the further slots. This configuration is not remcomanded by Asus but now it works!!! Asus datasheets recommand to use same memory for slots 1-3 and 2-4 but you don't need to use the same at all (in my case KM of KS in slots 1-3 and KS or KM in slot 2-4). I changed bios settings (2.5 3 3 7, turbo mode) and it still working (tested around 9hours). It's very stange. Has someone any idea about these results? |
|
|
|
|
|
#2 |
|
Oct 2005
1002 Posts |
I'm sorry I have some problems with my keyboard. I said it's STRANGE!!!
|
|
|
|
|
|
#3 |
|
3×13×97 Posts |
It is your motherboard that are crap!!!
I have the same in my server I use 2 Corsair 256Mb PC3200 and 2 corsair PC3500 256Mb sticks When I run the server with just 2 sticks, it run as it should (400Mhz at the RAM), when I use all 4 it will only run at 320MHz. I have talked to Asus about this and they can't help with this motherboard as it is discountinued... Alot of people on Asus forum have strange problems with this motherboard
Last fiddled with by Bikerman on 2005-10-19 at 14:36 |
|
|
|
#4 |
|
Oct 2005
22 Posts |
I asked Asus about my problem and they told me to reset the Bios (unplug the PSU, remove the battery and move the jumper from 1-2 to 2-3) and reconfigure the bios. Now it works with standard placement of my 4 RAMs
So try to do this. Maybe it will work for you too. |
|
|
|
|
|
#5 |
|
Jul 2004
Nowhere
80910 Posts |
that seams a pointless request removeing cmos battery is just like shorting pin 2 to pin 3....
|
|
|
|
|
|
#6 |
|
Oct 2004
232 Posts |
Glad it seems sorted. In event of problems it is often a good idea to get latest motherboard bios by download and flash to it in case prob was fixed.
Also would have been interesting if you could characterise the working / not working scenarios in terms of measured memory throughput in Memtest 86 PLUS. Just out of interest of course. |
|
|
|
|
|
#7 |
|
Oct 2005
22 Posts |
My MB has the lastest bios (V19). The problem I had is written in my first post.
To resume the situation, I have 4x 512MB dual channel RAM (2x Kingston and 2x Kingmax value ram). Kingmax in slots 1 and 3, Kingston in slots 2 and 4. When I launched the torture test (nothing was o/c and timming were manually well set at 2.5-4-4-8) it made an error after few seconds. I moved the RAM is this configuration: Kinston in slot 1 and 3, kingmax in slot 2 and 4. result: same error. I moved the RAM is this configuration: Kinston in slot 1 and 2, kingmax in slot 3 and 4 (it is not recomanded by asus) but after that it was working. I reset the bios (like described in last post) moved Kinston in slot 1 and 3, kingmax in slot 2 and 4. result: now it works. When I say it works, it means that torture test made no error during more than 10 hours of test. So my conclusion is that the reset made really something good but I don't know why. |
|
|
|
![]() |
Similar Threads
|
||||
| Thread | Thread Starter | Forum | Replies | Last Post |
| Results | ET_ | Operazione Doppi Mersennes | 627 | 2021-01-31 18:06 |
| Where have all the TF results gone?... | lycorn | PrimeNet | 22 | 2017-10-02 02:40 |
| PGS Results | danaj | Prime Gap Searches | 0 | 2017-08-14 18:35 |
| CPU Results last 24 hrs | Unregistered | Information & Answers | 3 | 2010-07-26 00:49 |
| 0x results... | Mike | PrimeNet | 11 | 2004-05-23 12:55 |