• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

P35c DS3R destroying ram

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.
D

DirtyIsh

Registered
‍
Joined
Aug 25, 2007
Initially was up to 400Mhz FSB on stock cooling. Ballistix 8500 was not stable at stock settings, RMA'd successfully. The new kit proved to be fine. With water, 425Mhz fsb (3.4Ghz), ram 1:1 using 4-4-4-12 under volted to 2.1v. System was 12hrs+ Orthos stable, and multiple passes of Memtest stable. Then suddenly started getting BSODs from the 'irql_not_less_or_equal' error which I understand to be related to either add-on card conflicts, CPU overheat/instability or memory. No add on cards in my system to cause IRQ conflicts, and the system was stable as noted above. Cpu is getting 1.54 volts before Vdroop at 41 degrees load. I was in no mood to reduce the overclock, so I ran Memtest with after getting the BSODs maintaining overclocked settings and found countless errors, reverted to stock bios settings, same result.

Could it be just another doomed ram kit, or is this mobo going to kill again. I have a set of Patriot 6400 that I can use temporarily. I tested this kit before and it passes, so if I test again and it fails, do I go through the hassle of getting a manufacturer RMA for the board, or blame all my problems on bad ram? I would hate dismantling after the annoyances I went through to install watercooling with 1/2" ID tubing in a P182 case, using 2 screw drivers.
 
rma your ram and if it does it again then rma the motherboard

One thing i have noticed is that the Ballistix is starting to decline in quality so it is quite possibly the ram

what are your case idle temps and do you have a fan blowing on the ram?
 
DirtyIsh said:
Initially was up to 400Mhz FSB on stock cooling. Ballistix 8500 was not stable at stock settings, RMA'd successfully. The new kit proved to be fine. With water, 425Mhz fsb (3.4Ghz), ram 1:1 using 4-4-4-12 under volted to 2.1v. System was 12hrs+ Orthos stable, and multiple passes of Memtest stable. Then suddenly started getting BSODs from the 'irql_not_less_or_equal' error which I understand to be related to either add-on card conflicts, CPU overheat/instability or memory. No add on cards in my system to cause IRQ conflicts, and the system was stable as noted above. Cpu is getting 1.54 volts before Vdroop at 41 degrees load. I was in no mood to reduce the overclock, so I ran Memtest with after getting the BSODs maintaining overclocked settings and found countless errors, reverted to stock bios settings, same result.

Could it be just another doomed ram kit, or is this mobo going to kill again. I have a set of Patriot 6400 that I can use temporarily. I tested this kit before and it passes, so if I test again and it fails, do I go through the hassle of getting a manufacturer RMA for the board, or blame all my problems on bad ram? I would hate dismantling after the annoyances I went through to install watercooling with 1/2" ID tubing in a P182 case, using 2 screw drivers.
Click to expand...

What are you reading memory voltage with, and from where on the board? From my experience the boards default for DDR2 is 2.0v.

And 1.54v for 3.4ghz on E64xx is ALOT.
 
Case idle temps are 42 degrees. From what I gathered (forums posts I've read) this boards start DDR2 voltage at 1.8v, so I have it set to +.4 = 2.2v. Also at a point in time when I had the Gigabyte monitoring program, it showed the mem voltage as 2.2v when set to +.4v in the bios. So software and bios voltage readings for the memory. Ranger you say 2.0v is the default, but at +.2v bios which if you are correct would be 2.2v, the Ballistix does won't boot correctly at 800 or 1066 using stock timings.
Just tested the Patriot 6400 sticks individually, and in different slots. They fail Memtest, specifically at test 4. Also got smiley faces during the boot screen. "Memory runs at single.... :( :) :)" Wierd.
So this would make it three 2 gig kits that went bad. I'll rma the Ballistix again, but this is odd.

And yea Ranger, I'm not pleased with that 1.54v.
 
DirtyIsh said:
Case idle temps are 42 degrees. From what I gathered (forums posts I've read) this boards start DDR2 voltage at 1.8v, so I have it set to +.4 = 2.2v. Also at a point in time when I had the Gigabyte monitoring program, it showed the mem voltage as 2.2v when set to +.4v in the bios. So software and bios voltage readings for the memory. Ranger you say 2.0v is the default, but at +.2v bios which if you are correct would be 2.2v, the Ballistix does won't boot correctly at 800 or 1066 using stock timings.
Just tested the Patriot 6400 sticks individually, and in different slots. They fail Memtest, specifically at test 4. Also got smiley faces during the boot screen. "Memory runs at single.... :( :) :)" Wierd.
So this would make it three 2 gig kits that went bad. I'll rma the Ballistix again, but this is odd.

And yea Ranger, I'm not pleased with that 1.54v.
Click to expand...

I measured with my DMM and auto voltage is 2.0v.
 
I don't think the reference voltage for the P35C is the same as that of the regular P35 which supports only DDR2. I thought the reference voltage for the P35C was 2.0V (due to DDR3) in which case u are putting 2.4V thru the RAM an thus killing them.
 
well rma your ram and motherboard if you have gone through 3 sets of ram on that motherboard

this is making me think about not getting this board
 
EmAn said:
well rma your ram and motherboard if you have gone through 3 sets of ram on that motherboard

this is making me think about not getting this board
Click to expand...

read my post it is very clear what is wrong here and it is not the board. Gigabyte should get rid of these delta voltages (+0.1, +0.2V, etc) and just use absolute voltages.
 
I am using the F4g bios, but I am pretty darn sure auto vdimm is 2.0v.

I have it set to +3 in the bios and my DMM reads 2.29.

Speedfan reads 2.27 and

Everest reads:

 
Last edited:
Ok so with 2.0v being the auto setting, tell me what you think about this.
When I first put the system together, it took 2.3v for me to get through the initial windows installation. With windows installed, I did initial overclocking to gauge the potential of the combination of components to decide how much to spend on cooling this time. So with the readings from the Gigabyte monitoring tool telling me that the ram was at 2.1v (2.3v according to wisdom) and the lack of stability at speeds and timings where there should be stability, I set it to +.3/.4v bios = 2.1/2.2v according to Gigabyte software = 2.3/2.4v according to you fellas.

What I'm saying is its taking 2.3/2.4v to be stable within spec.
Thank you for the responses so far.
 
Last edited:
I was on Gigabyte's site looking for RMA information (too late to get one from retailer) and 2-3 weeks turnout time is too long for me. On the spec page, http://www.gigabyte.com.tw/Products...board&ProductID=2631&ProductName=GA-P35C-DS3R, it says 2 x 1.5V DDR3 DIMM sockets and 4 x 1.8V DDR2 DIMM sockets. Then I was going through some notes I had from a conversation with a Gigabyte tech when the first set of Ballistix was being killed and he said "Ok, try setting the voltage down to 2.1 and Memtest again. Yea, change it to +.3"

Reading this forum, I learned how to build a NF7-S with Xp2100. Then a watercooled C2D and 8800GTS. But with the information I've given you all, you can see I am not seeing a definite answer to the problem. With school resuming, don't have the time or desire to fight this battle. So I'll order another board, rma the DS3R after the swap and determine the status of the ram kits with whatever new mobo I decide on. Gratzi.

 
Last edited:
I see nothing that makes me believe it is anything other than 1.8v. Do you seriously expect me to believe that Gigabyte shipped a motherboard that allows you to increase VDIMM to 2.7v? What, do they think we'll jam DDR in there?

I have memory that needed 2.3v to do over 1000+ at 4-4-4-8. That same memory needs +4 on the DS3R to run at 4-4-4-12. Now if you're suggesting that means I'm running 2.4v into these sticks, I'd call you crazy. They're warm, but no warmer than they were on a P5B Deluxe @ 2.3v

If they were 2.0v slots Gigabyte would have described them as such.

IMHO. I'll find more proof.
 
Did you run memtest with everything at stock settings, including OC and FSB? Also, 425 for FSB, how hot is your northbridge? Mine gets screaming hot so I dumped my u120 Extremes to go with SI-128s to get some airflow on the NB.
 
You must log in or register to reply here.

Similar threads

N
CPU unstable no matter what I do, and it's getting worse
2
Replies
25
Views
1K
mackerel
mackerel
L
Looking for DDR3 2x16GB RAM
2
Replies
25
Views
510
LordLiverpool
L
S
CMT32GX5M2X6200C36 *4 Sticks Z790-E GAMING WIFI i9-13900K XMP Unstable
Replies
4
Views
673
Woomack
W
R
Bad CPU, Bad Mobo, or 3 Bad Ram Kits?
Replies
14
Views
469
Rewire
R
N
Poor memory clocks caused by TG contact frame?
Replies
9
Views
757
nstgc
N
Back