EVGA Nforce 650i Ultra issues

[Highlanderx]

Ok, I'm at my wits end. It all started back in the first week of October when I received my shipment from Newegg:

Intel Core 2 Duo E6750
EVGA Nforce 650i Ultra T1
G Skill RAM DDR2 800
Creative X-Fi Extreme Gamer sound card
2 WD Sata2 320GB HD's

I had at the time a 7900GTO vid card from EVGA as well. Soon after getting everything loaded onto the newly built machine I noticed that I would get lockups while playing 3d games and ONLY during gaming. Running Orthos for hours on end was fine, Memtest fine. A couple weeks later I purchased an EVGA 8800GT SC. Put that in and was having the same results.

I've read the EVGA forums relentlessly for the past 6 weeks and even broke down and called their support line when I had tried everything I could think of. Voltages, replacing parts, formatting, etc

I've replaced the PSU (had Silverstone 600W, bought Seasonic M12 700w), after it was suggested by looking at the voltages in windows that perhaps my 12v rail was too low. New power supply displayed same voltages, explanation, sucky reporting software.

I formatted the Boot RAID 0 array and reloaded Win XP SP2, no updates, just loaded the Nforce drivers off of EVGA's website (8.44) and the latest 169.02 drivers, that lasted for a while, then the lockups/freeze ups happened yet again. So I reghosted my drives back to my original configuration.

I've RMA'd the board which I received last Friday and put it in. Hours later, a freeze up happened..sigh

I finally pulled off the Northbridge passive Heatsync and replaced it with an active cooling solution and arctic Silver 5....4 hours later on the next day and another freeze up.

I've moved the RAM to different slots, same issue.

I haven't replaced the RAM since I had no problems in Memtest, nor have I replaced the CPU since Orthos for hours on end proved fine.

Temps look fine and I have an Antec P180 case with multiple fans. The ambient temp in the room is 70F.

Called EVGA again today and was put in touch with their 2nd tier support. He suggested at first that it might be the RAM or CPU, but after I told him the extent to which I have tested things, he sort of backed off. In the end, had no ideas or solutions and I was basically told that I could do the "step up" to a different board. Although they were somewhat helpful and kept saying, we'll do what we can to help you. I felt as if I was shoved out the door. I even asked to speak to someone who could authorize something else. I've done the RMA thing and it's still not working. I felt that I'd done all the testing and troubleshooting that needed to be done and didn't want to RMA yet another board. I'm a bit disappointed with EVGA's support on this matter.

I can't continue to replace parts, funds are running out. Anyone have any help or advice? perhaps a shotgun I can borrow?

 

[jaggerwild]

Couple questions, I see your posting in EVGA site. What bios are you using and what is the ratting on the 650 board as that CPU has a 1333FSB ratting.
Also Welcome to the forum too!

 

[Highlanderx]

Hi Jagger, nice to see a familiar face

I'm using the latest P03 BIOS on both the original board and the RMA'd one. EVGA support confirms that the board supports the 1333FSB CPU's. But I did have this issue with the P02 BIOS too, which is what it was shipped with on the original board.

 

[jaggerwild]

I wouldn't think it was the mem as I have 800 G-skill and have no issues but do have a different board (gigabyte 680).
So all these problems while your having them you also have the 8800 GT right? Lock ups durring third party on mine is usualy cause I'm clocking it to high, if your not clocking it then you might need to bump up the CPU voltages a little. Also I have the L4 mem they need 2.2 G-skill says.
The artic silver you put on the north bridge was it put on sparringly? Cause this could cause an issue as it is conductive and if it ossed out from under the chip and touched the pins this could be causing the issues too. (I'm kinda thinking out loud here to try and help ya).
I'm pretty sure that this maybe a T1 issue here, not certain has I'd have to have my hands on it to see. But I mean you have tried everything else here, and they said you could step up if you wanted to?
Do this first try under volting the chip, then test don't go too low but bump it down a step if this doesn't change then try one more If still no change then go one up from the default setting one or the other should but not certainly fix it. I'd also try the mem, if the cpu didn't do it then put the CPU to default voltages then try the mem don't go over 2.2volts here!!!! I'd first try going lower fer chits and giggles too.
Your multi is @8 times right? What is the front side bus @? With this ram don't go over 400 just to see for now. If you have two boards then try this if the one that isn't in the case is a known good unit then try hooking it up outside the case, as a last resort to see if the one in the case has a short.
If this fails you then yeah ask then to step it up to an A1, as they said they would so why the heck not then.......
Couple last minute thoughts, try running it with only one stick of ram, if this is no good with just the one reset all bios settings to defaults hit save (f10) then re-enter the bios on boot up and reconfigure the RAID settings. Oh and one other thing(told you I was thinking out loud here can you tell?) You have a raid, see this link make sure the sata cables are hooked up in the correct locations also do you have the locking cables as ive seen them look ok but be lose and this could cause an issue similar to yours.
See the first photo for the true cable settings, you did load the raid drivers before loading windows or the OS right? By hitting the F6 button while booting into loading of the OS?

http://www.evga.com/forums/tm....09&mpage=1&key=&#81409
Sorry its way to much info all at once, take yer time with it as this aint rocket science. Witch means its usually the little things that we over look, and really basic too with me! Power plug on the video card? Sorry have to ask, Check all the plugs while the system is off and unpluged too.
Post back and let me know, we'll figure this out.

 

[Mr Fox]

Originally posted by: Highlanderx
Ok, I'm at my wits end. It all started back in the first week of October when I received my shipment from Newegg:

Intel Core 2 Duo E6750
EVGA Nforce 650i Ultra T1
G Skill RAM DDR2 800
Creative X-Fi Extreme Gamer sound card
2 WD Sata2 320GB HD's

I had at the time a 7900GTO vid card from EVGA as well. Soon after getting everything loaded onto the newly built machine I noticed that I would get lockups while playing 3d games and ONLY during gaming. Running Orthos for hours on end was fine, Memtest fine. A couple weeks later I purchased an EVGA 8800GT SC. Put that in and was having the same results.

I've read the EVGA forums relentlessly for the past 6 weeks and even broke down and called their support line when I had tried everything I could think of. Voltages, replacing parts, formatting, etc

I've replaced the PSU (had Silverstone 600W, bought Seasonic M12 700w), after it was suggested by looking at the voltages in windows that perhaps my 12v rail was too low. New power supply displayed same voltages, explanation, sucky reporting software.

I formatted the Boot RAID 0 array and reloaded Win XP SP2, no updates, just loaded the Nforce drivers off of EVGA's website (8.44) and the latest 169.02 drivers, that lasted for a while, then the lockups/freeze ups happened yet again. So I reghosted my drives back to my original configuration.

I've RMA'd the board which I received last Friday and put it in. Hours later, a freeze up happened..sigh

I finally pulled off the Northbridge passive Heatsync and replaced it with an active cooling solution and arctic Silver 5....4 hours later on the next day and another freeze up.

I've moved the RAM to different slots, same issue.

I haven't replaced the RAM since I had no problems in Memtest, nor have I replaced the CPU since Orthos for hours on end proved fine.

Temps look fine and I have an Antec P180 case with multiple fans. The ambient temp in the room is 70F.

Called EVGA again today and was put in touch with their 2nd tier support. He suggested at first that it might be the RAM or CPU, but after I told him the extent to which I have tested things, he sort of backed off. In the end, had no ideas or solutions and I was basically told that I could do the "step up" to a different board. Although they were somewhat helpful and kept saying, we'll do what we can to help you. I felt as if I was shoved out the door. I even asked to speak to someone who could authorize something else. I've done the RMA thing and it's still not working. I felt that I'd done all the testing and troubleshooting that needed to be done and didn't want to RMA yet another board. I'm a bit disappointed with EVGA's support on this matter.

I can't continue to replace parts, funds are running out. Anyone have any help or advice? perhaps a shotgun I can borrow?

The issues that you are running into are characteristic to the Crappy NForce Platform drivers what of the drivers do you install ?

The SW IDE drivers are notorious for giving issues.

I would recommend acquiring a few versions, and testing what works, and what does not.

It is this item right here that led me away from NVIDIA Chipsets.

Let me know if you need more info.....or further support.

 

[Highlanderx]

Jagger - Yeah running G Skill, but in looking at their supported memory list, my particular model number isn't on EVGA's list, but in going to their forums, someone asked if it was supported and they say yes.

Lockups are happening on the 8800GT and a 7900GTO - nuff said there

I've tried every variation of voltage that I could, I can't do more than 2.1 on my RAM, in fact going 2.1 yields very unstable results. 1.83 or 1.9 is the only stable voltage I was able to attain. I've tried underclocking, but not undervolting.

Yes, I used the AS5 sparingly But this problem existed before and after I did that mod.

I have moved the RAM around, but did not try using 1 stick at a time.

RAID doesn't seem to be an issue, that works flawlessly. I did have one problem with it, but that was because a cable was loose after removing and installing a vid card, easily fixed.

I will try the RAM one stick at a time thing or replacing it altogether and see what that yields.

Running Orthos for hours on end using the blend test or the cpu test yields no issues nor does Memtest. Considering that my specific memory model isn't on the support list from EVGA but is on the GSkill forum, I'm wondering if that may indeed be the issue.

Mr. Fox - Yeah I thought the same thing too, remember VIA drivers? /shudder.

I've used two versions, 8.43, the latest version from Nvidia's web site and 8.44 which seem to be EVGA's own version, neither seemed to have any effect. In these versions of drivers, I didn't see an option NOT to install the SW IDE, I never even saw it come up..not sure if that particular drivers is actually even being installed. It has check mark boxes for the SMS Bus controller, NVRAID, Ethernet and Ntune. Then during the install asks if I want to install NAM, to which I know better and then it's done.

 

[o1die]

Since you're running xp, I would try a different version of directx. I had similar issues with a different board, and changing to another version fixed my lockups. I switched back to the first version of 9.0. Some of my old games run fine on 8.1, but you may not be able to find it.

 

[jaggerwild]

Hello Mr Fox and Mr O1die!
Yeah I have the LT board and this ram ran fine so I'd not think this was the problem either.Mr Fox had a very good suggestion also could possibly be video drivers too for that matter so might try messing with them as well.
Also as Mr O1die suggested make sure Direct X is updated and OpenGL to I just get"JAVA" service pack.
Have you tried the drivers that come on the EVGA disk or do they ask you to down load them from there site?
here's a link to other drivers too.

http://www.nvidia.com/page/home.html

 

[Highlanderx]

Ok, update. Bought some Crucial Ballistix ram off of the approved list and it took about 10 minutes of Crysis and 5 minutes of Call of Duty 4 before it locked up...sigh

I've tried multiple sets of drivers to no avail, I don't believe those are the problem or everyone and their brother would be screaming of this issue. Further messing around with voltages and I got another freeze, this time, it just corroded the screen with a mix of colors to the point where I could barely make out what was on the screen (It has done this on occasion before, but not that often) While that happened, I heard the infamous sound of an error popup happen in the background. I tried to ALT-TAB and was able to see what it was. It was the one that says that the display driver has become stuck in an infinite loop waiting for hardware to become available or something to that effect.

With that in mind, I decided to reload Rivatuner 2.06 and try configuring it for the fan profiles that I see referenced alot on these and other boards. It was fine for an hour and a half in one of my MMO games, which isn't as demanding as Crysis and COD4. I will try Crysis and COD4 again when I get home today.

I'm now leaning more towards it being a graphics card issue that just happens to happen on both the 7900 GTO and 8800GT. Since Nvidia's drivers set both cards fans speeds to 29% and don't ramp up at all, perhaps that is the overall issue.

I'll report back of any findings.

 

[Highlanderx]

Another update:

It appears that it was the overheating of the GPU/VRAM all along. Rivatuner has seemed to fixed those issues with configuring the fan profiles that I found on a guide. I found through testing that I could get the card to repeatedly lockup once it hit 71c..pretty low imo. My only conclusion as to why this was happening on two completely different cards is that perhaps my new system (C2D from AMD) was pushing the cards harder and therefore overheating them easier.

For some reason though, perhaps through my own messing around, Rivatuner profiles won't always start, have to give a boost, by going in and setting the fan to fixed and selecting one of the profiles...odd. I didn't like that, so I went out and bought a Zahlman VF900 and just installed that. Idle temps have dropped 11c down to 38c from 49c, nearly a 52F drop. Once I have a good burn in, I'll test the load temps (going to just let it run idle for a few hours).

I'm still miffed that I had to go out and buy a 3rd party item for another $50 to get the card to work stable in a game environment...wake up Nvidia and tune yer damn drivers.

 

[jaggerwild]

Yeah,
Everybody is doing it just to be safe, glade you got it figured out at least even though I know how frustrating these issues can be .
I was just reviewing everything posted cause I thought you still had issues :shocked:
OH well, now if you have any other issus or concerns you have your own thread to post into,
Take care man, see you in one forum or the other

 

[thesid]

oh!! great to hear you fixed the issue,
im planning on buying the exact same configuration
650i utlra 8800gt and e6750, + patriot extreme 4gb 4-4-4-12 PC2-6400
half way through the post i almost gave up on this config hehe.
It was like reading a suspense novel.. god! my heart started beating faster and faster with every line i read!