The Corsair User Forums  

Go Back   The Corsair User Forums > Hardware Related (Non Corsair Questions) > Motherboards, Processors and Overclocking

Notices

Reply
 
Thread Tools Rating: Thread Rating: 5 votes, 5.00 average. Display Modes
  #1  
Old 12-08-2007, 02:38 PM
winKSU winKSU is offline
Registered User
winKSU's PC Specs
 
Join Date: Dec 2007
Posts: 8
POST ID # = 321185
winKSU Reputation: 10
Default Memory, MB, or Me?

Gigabyte GA-P35-DS4 (Rev 2) F9 BIOS
Q6600 @ 3600 MHz (450 x 8)
2 x TWIN2X2048-8500C5D
Cooler Master Real Power Pro 650w

I'm having a problem with running Prime95--it keeps reporting a rounding error. Every other stability test has been OK (WPrime, SuperPi, Hot CPU Tester). I've tried memory multipliers 2.0 (900), 2.4 (1080), and 2.5 (1125) with no difference--higher multipliers than 2.5 won't post. I've manually set the memory timings to 5/5/5/15 (all others auto). The overvoltage for DDR2 is set to +0.40 and the MCH is +0.20.

I downloaded and ran Memtest86+. In 15 minutes of run time, it reported 90 errors at address 000dfeedf18 (3582.8)--dfeedf18 good and dfeedf19 bad (Err bits 00000001). Thinking that I had a bad memory module, I removed 2 modules (leaving it in Dual Channel) and ran Memtest86+ again. This time it reported the error at 0007feedf18 (7feedf19 bad, Err bits 00000001). Testing each one of the modules individually produced errors at 0003feedf18 (3feedf19 bad, Err bits 00000001). I then tested the other pair of modules that I had originally removed--each one reported the same problem.

The odds of 4 memory modules being bad must be low. Also, why wouldn't I get 4 different address errors when all the modules were installed? I've played with overvoltage on the DDR2 going as high as +0.65--how much is safe with these modules? I've do have the dominator airflow. I also tried going higher on the MCH voltage with no different results (lower causes instability).

Any advice? Bad motherboard? Have I missed a setting?
Reply With Quote
  #2  
Old 12-08-2007, 03:25 PM
bigbuck08 bigbuck08 is offline
Crysis Gamer
bigbuck08's PC Specs
 
Join Date: Aug 2007
Location: Wisconsin
Posts: 509
POST ID # = 321188
bigbuck08 Reputation: 10
Send a message via MSN to bigbuck08
Default

It sounds like a bad MB. Like you said, it is highly unlikely that all four modules are broke. I would look into RMA'ing the MB if its still under warranty. When you tested the modules was it at BIOS defaults first? When Corsair tested this RAM the Maximum DDR2 recommended VDIMM is 2.2 Volts. However, on the spec sheet for this RAM it was tested at 5-5-5-12-2T at 1066MHz @ 2.1v.

Last edited by bigbuck08; 12-08-2007 at 04:21 PM.
Reply With Quote
  #3  
Old 12-08-2007, 08:30 PM
winKSU winKSU is offline
Registered User
winKSU's PC Specs
 
Join Date: Dec 2007
Posts: 8
POST ID # = 321223
winKSU Reputation: 10
Default

By default, the PS4 is setting 1.8v and I've tried every combination from 1.8 to 2.45 volts. This is a completely new system, so the MB is under warranty--I wonder if it would be less hassle to RMA it with NewEgg?

I wish I knew for sure. I'd hate to take everything apart and rebuild only to have the same problem.
Reply With Quote
  #4  
Old 12-08-2007, 09:42 PM
bigbuck08 bigbuck08 is offline
Crysis Gamer
bigbuck08's PC Specs
 
Join Date: Aug 2007
Location: Wisconsin
Posts: 509
POST ID # = 321229
bigbuck08 Reputation: 10
Send a message via MSN to bigbuck08
Default

If would RMA it to Newegg. It will be faster and proabably cheaper too.
Reply With Quote
  #5  
Old 12-09-2007, 02:24 PM
winKSU winKSU is offline
Registered User
winKSU's PC Specs
 
Join Date: Dec 2007
Posts: 8
POST ID # = 321271
winKSU Reputation: 10
Default

I was able to get CPU stability with 9 x 400 for the same 3.6 GHz (8 x 450). This allowed me to set the memory multiplier to 2.5 to get 1000 MHz on the memory (just short of its rated 1066). The next multiplier is 3.0 which is 1200 and I can't get it to post.

I ran Prime95 continuously for 4 hours so I think it's stable. MEMTEST86+ is still reporting the same error.

I wonder if I'll suffer any consequences by calling it "good". I'm a little disappointed that I couldn't get to 1066.
Reply With Quote
  #6  
Old 12-09-2007, 10:45 PM
RAM GUY's Avatar
RAM GUY RAM GUY is offline
Corsair Product Guru
 
Join Date: Apr 2001
Location: Fremont, CA
Posts: 118,289
POST ID # = 321321
RAM GUY Reputation: 10
Default

Thats about normal with this chipset when over clocking, have you checked some of the MB reviews and see how they might have ran their memory settings at these CPU speeds?
Reply With Quote
  #7  
Old 12-10-2007, 10:34 AM
winKSU winKSU is offline
Registered User
winKSU's PC Specs
 
Join Date: Dec 2007
Posts: 8
POST ID # = 321411
winKSU Reputation: 10
Default

I ran across a post that indicated the FSB needed overvoltage too. I added +0.2v to match the MCH. I reduced the CPU multiplier to 8 with 450 clock. I set the memory multiplier to 2.5. I was able to post at 1125 I ran Prime95 stable overnight.

I still get the same memtest error, but Windows Memory Diagnostic Tool runs clean.

It must be something with memtest--I'm current running the SANDRA burn-in loop and it hasn't found any errors.

I think I'll play around with multipliers, voltages, clocks and see what I can get.

What would be the safest overvoltage on the 8500C5D?
Reply With Quote
  #8  
Old 12-10-2007, 12:30 PM
mxxxv mxxxv is offline
Registered User
 
Join Date: Dec 2007
Posts: 1
POST ID # = 321422
mxxxv Reputation: 10
Default same problem

Hello from Czech,
I have same problem,same address - 0003feedf18 on P35 DS4 rev.1.1 bios F9 While default, oc, uderclock.... be all one.



GB P35 DS4 rev.1.1
C2D E6750
Patriot PDC22G6400LLK 2x1gb
Reply With Quote
  #9  
Old 12-10-2007, 02:32 PM
RAM GUY's Avatar
RAM GUY RAM GUY is offline
Corsair Product Guru
 
Join Date: Apr 2001
Location: Fremont, CA
Posts: 118,289
POST ID # = 321453
RAM GUY Reputation: 10
Default

What test are you getting errors on and are you using www.memtest.org?
Reply With Quote
  #10  
Old 12-10-2007, 04:21 PM
Wired's Avatar
Wired Wired is offline
Administrator
Wired's PC Specs
 
Join Date: Aug 2000
Posts: 32,000
POST ID # = 321474
Wired Reputation: 16
Default

Quote:
Originally Posted by mxxxv View Post
Hello from Czech,
I have same problem,same address - 0003feedf18 on P35 DS4 rev.1.1 bios F9 While default, oc, uderclock.... be all one.



GB P35 DS4 rev.1.1
C2D E6750
******* PDC22G6400LLK 2x1gb
This is Corsair memory's official forum, wrong place to post for that memory, sorry.
__________________
No, I don't work for Corsair
...you need to flow like a river...
Forum Rules
Admin of this forum / webdesignforums.net / Petri's IT Forum / The Corsair Memory Steam Group
Founder/Creator/Admin of ZE SECRET PROJECT (Coming Soon)
Reply With Quote
  #11  
Old 12-10-2007, 10:08 PM
winKSU winKSU is offline
Registered User
winKSU's PC Specs
 
Join Date: Dec 2007
Posts: 8
POST ID # = 321521
winKSU Reputation: 10
Default

Quote:
Originally Posted by RAM GUY View Post
What test are you getting errors on and are you using www.memtest.org?
I'm using version 1.70 from memtest.org. Test 5 was the only one to complete without error. Tests 9 & 10 I let run for 10 minutes with no errors. Tests 1, 2, 3, 4, 6, 7, and 8 all produced errors.

With the 8500C5D, the errors are at address:

1 stick - 0003feedf18
2 sticks - 0007feedf18
4 sticks - 000dfeedf18 (3582.8 M)

It was only that singular address that failed and it was actually the address that ended in 9 (dfeedf18 good, dfeedf19 bad).

I currently have the 4 sticks back in and the interesting thing (that I didn't notice before) is that the testing range was 116k - 3583 M. The error is occuring at the upper end of the tested range.

The Vista Memory Diagnostic Tool finds no errors. I guess that makes sense because with the 32-bit version of Vista, the memory limitation is around 3120 M.

Despite trying different voltages, multipliers, and clocks I couldn't get anymore than 1125. My final settings:

Core Voltage 1.264v
CPU Clock 450
CPU Multiplier 8 (3600 MHz)
Memory Multiplier 2.5 (1125 MHz)
DDR2 Overvoltage +0.45v (2.25v)
MCH Overvoltage +0.10v
FSB Overvoltage + 0.10v
Reply With Quote
  #12  
Old 12-10-2007, 10:38 PM
bigbuck08 bigbuck08 is offline
Crysis Gamer
bigbuck08's PC Specs
 
Join Date: Aug 2007
Location: Wisconsin
Posts: 509
POST ID # = 321533
bigbuck08 Reputation: 10
Send a message via MSN to bigbuck08
Default

Are these with all modules installed? If so test them individually.
Reply With Quote
  #13  
Old 12-11-2007, 12:48 PM
RAM GUY's Avatar
RAM GUY RAM GUY is offline
Corsair Product Guru
 
Join Date: Apr 2001
Location: Fremont, CA
Posts: 118,289
POST ID # = 321603
RAM GUY Reputation: 10
Default

That is still a great O.C. but I would add a bit more voltage to the CPU if it were me. But I cannot tell you to do that!
Reply With Quote
  #14  
Old 12-22-2007, 04:35 PM
ArchCorsair ArchCorsair is offline
Registered User
 
Join Date: Dec 2007
Posts: 1
POST ID # = 323196
ArchCorsair Reputation: 10
Default

Hey, I don't come from this forum. Today i googled "000dfeedf18" from the address my RAM was failing consecutively (all 4 sticks of them) and is stumbled upon this thread. I have the same exact motherboard as you, and a Q6600 @ 3.6Ghz. Did you ever resolve the issue?

this is my exact post from another thread:
Quote:
Yesterday I ran memtest86+ to test my RAM's new overclocked settings (1000mhz @ 4-4-4-12) they are Crucial Ballistix. After about 10 minutes I found the PERFECT settings, perfectly stable, and no errors on memtest86+ durring a 30minute test.

About 5-6 hours later I get BSOD and I was pretty sure it was the RAM so i boot back into memtest86+

Within the first 30 seconds it reports about 40 errors at

Code:
Address:
000dfeedf18 - 3582.8
I then reverted the clocks back to stock;
Then I tried to test each stick of ram (total of 4) 1 at a time and i STILL get the same error at the same address for ALL sticks! I don't think its likely for all sticks to go bad at the same time, reporting the same error, at the same address. Is it the motherboard? or is memtest86+ giving false errors?

Last edited by ArchCorsair; 12-22-2007 at 04:53 PM.
Reply With Quote
  #15  
Old 12-22-2007, 06:13 PM
winKSU winKSU is offline
Registered User
winKSU's PC Specs
 
Join Date: Dec 2007
Posts: 8
POST ID # = 323206
winKSU Reputation: 10
Default

Quote:
Originally Posted by ArchCorsair View Post
Did you ever resolve the issue?:
No, but on other threads (and I think there was another poster here whose messages got deleted) I've seen that this isn't a unique situation. It doesn't seem to be causing a problem, but when I get the time I'm going to e-mail Gigabyte.

I've been spending my time trying to fix two other annoyances. The first is with the new machine (Q6600). It's stable at core voltage 1.264 only after its warm. It won't post if its been sitting overnight. I had to bump it to 1.281 to get it to post. EasyPro 5 should be able to bump it back down (auto load last save on next reboot) but it doesn't seem to work. I could leave it at the higher voltage, but it runs a tad bit hotter.

The other problem is on my old PC (AMD XP 3200+) which I'm handing down to my kids. They like the Yahoo AT&T browser. Everytime I load it, I get an infinite loop of script errors on all browsers (even if I shut the setting off to report script errors). Very frustrating.
Reply With Quote
Reply

Thread Tools
Display Modes Rate This Thread
Rate This Thread:

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is On

Forum Jump


All times are GMT -5. The time now is 08:23 PM.


Powered by vBulletin® Version 3.8.7
Copyright ©2000 - 2014, vBulletin Solutions, Inc.