Well, I killed another 680i SLI board.

I've been through multiple 680i boards from various brands and seen multiple ones go BANG out of nowhere and take themselves out, RAM out, or just go berserk.

I wouldn't blame this on D9 RAM either. I heard that and bought up 4gb of prmos based memory. Damn striker didn't want to clear CPU INIT (well that's normal), finally did and then flew into a fit of CMOS ERR and all sorts of erratic behavior. It spent the greater part of a week down before I finally got all 4 sticks working at the rated speeds with NO overclock. And then POW bye-bye RAM a month later.

Sadly I made the mistake of purchasing over $160 in water blocks for the striker, so now I've either got to ditch the blocks (shame really) or I'm stuck with the striker (not so bad since for all it faults it just seems to kill itself more often then it does kill my RAM).

Waiting on 780i boards and hoping for the best... and that my blocks fit them. Till then I'm on another striker, new RAM, and locked into CMOS ERR loop again.:p
 
Man , you've got more patience than me.

If this Striker finally dies (so far so good with new bios) , I'm smashing it to little tiny bits and mailing it back to asus.They can keep thier rma.

I wish you the best of luck on getting back up and running , I think we all need some.
 
BrainEater said:
Man , you've got more patience than me.

If this Striker finally dies (so far so good with new bios) , I'm smashing it to little tiny bits and mailing it back to asus.They can keep thier rma.

I wish you the best of luck on getting back up and running , I think we all need some.
Click to expand...

More stubborn then patient. Plunking for dual 8800's, and 400 on water cooling the board + cards is enough motivation to keep trying at it. ASUS rma blows but most memory vendors are really good about it.
 
sd11 said:
More stubborn then patient. Plunking for dual 8800's, and 400 on water cooling the board + cards is enough motivation to keep trying at it. ASUS rma blows but most memory vendors are really good about it.
Click to expand...

I hear you. I'm in the same boat. I would have given up on the 680i SLI's myself long ago, but I refuse to give up my SLI performance.
 
Dan_D said:
I hear you. I'm in the same boat. I would have given up on the 680i SLI's myself long ago, but I refuse to give up my SLI performance.
Click to expand...

Yeah the performance is a huge reason, but price is another.

Complete EK asus1 mobo kit cost me about 200 when all was said and done, 8800gts 640mb was 400, ek block for it 110. So if I ditched asus SLI I'm tossing around 710 bucks of perfectly good hardware... unless I decided to drop for a commando and throw it all onto that.

The memory failures are obnoxious, but corsair and ocz have been really good about sending me new sticks even though it's no fault of theirs. I'm sure they have to be pissed at the situation.

The real bitch of it all is the mobo failures. Having to take apart the loop each time and jump through the hoops again. Combined with ASUS rma that's several weeks turn around, and at times it's not possible. But the current 300 for a new striker vs 710 on lost parts makes it a no brainer.

I've got old computers I built still going strong, all the way back to a p2 computer with a rock solid OC till this day.

But each 680i feels like a ticking time bomb till it blows up in my face.
 
sd11 said:
More stubborn then patient. Plunking for dual 8800's, and 400 on water cooling the board + cards is enough motivation to keep trying at it. ASUS rma blows but most memory vendors are really good about it.
Click to expand...

why not cut your loses and ditch the nvidia? :)
 
jarthel said:
why not cut your loses and ditch the nvidia? :)
Click to expand...

SLI performance and cash invested in second GPU + water blocks. I could give a crap about "eats ram" comes out of corsairs pocket not mine, and while I feel bad for them in this situation they've been more then understanding and more help in making the damn thing run period that I'll buy from them and OCZ from now on.

I can RMA the strikers most of the time, it takes time and I'm stuck on a second PC, from where I'm posting now, but oh well.

Had I known what I know know, I'd have moved from 975x to x38 and run dual ATi parts. But at the time it was 8800 or... nothing, and then the 2900hd series was junk. I also though nforce6 would mirror nforce2/4/5 for quality, it didn't.
 
sd11 said:
Yeah the performance is a huge reason, but price is another.

Complete EK asus1 mobo kit cost me about 200 when all was said and done, 8800gts 640mb was 400, ek block for it 110. So if I ditched asus SLI I'm tossing around 710 bucks of perfectly good hardware... unless I decided to drop for a commando and throw it all onto that.

The memory failures are obnoxious, but corsair and ocz have been really good about sending me new sticks even though it's no fault of theirs. I'm sure they have to be pissed at the situation.

The real bitch of it all is the mobo failures. Having to take apart the loop each time and jump through the hoops again. Combined with ASUS rma that's several weeks turn around, and at times it's not possible. But the current 300 for a new striker vs 710 on lost parts makes it a no brainer.

I've got old computers I built still going strong, all the way back to a p2 computer with a rock solid OC till this day.

But each 680i feels like a ticking time bomb till it blows up in my face.
Click to expand...

Well I'm not into it for as much money as you, but I understand your perspective. Having to drain and dissassemble the loop just sucks.
 
You know, from following this thread, and looking at other members sigs, I noticed that just about everyone having these problems has a QXxxxx. To add to that everyone I know that has just a C2D has not experienced these issues. Something to look into maybe?
 
Proxy said:
You know, from following this thread, and looking at other members sigs, I noticed that just about everyone having these problems has a QXxxxx. To add to that everyone I know that has just a C2D has not experienced these issues. Something to look into maybe?
Click to expand...

I've made mention of that exact theory several times already in this thread. I've got one 680i SLI board running in another system running with an E6600 overclocked. I have no problems out of that machine at all. All the boards that have been used with my Q6600 have met an untimely death.
 
Dan_D said:
I've made mention of that exact theory several times already in this thread. I've got one 680i SLI board running in another system running with an E6600 overclocked. I have no problems out of that machine at all. All the boards that have been used with my Q6600 have met an untimely death.
Click to expand...

AH! We maybe on to something, to go back a bit, I made the statement that the buddy of mine with an OCed E6600 has had ZERO problems. In the same respect the only person I know of comes to my clans ventrilo, he has a QX6700 but the one thing is, he has an Asus Striker Extreme. Now we know it does not OC well, but he has had no problems with it.

I am thinking this is something the [H] team should look into.:cool:
 
I agree, after thinking about it and talking to pc299 on aim, I decided to stick it out for now with the 680i since I like to have the feeling that I have a great graphics setup now, hopefully the 780i will fix most of the issue, but we will see. I also agree with the fact that it seems that most people who use regular dual core chips have had less issues, I might try a 6850 and see how much difference I feel between the 2 chips.
 
the same theory about the boards with quads experiencing the problem and the boards with duals not has been bounced around the EVGA forums as well. That being the case, I'd almost lean towards maybe something along the lines of the power circuitry on the boards. Maybe it's not substantial enough to handle two more cores? I know the northbridge gets mighty toasty...which is bothersome but a quiet fan can cure that quickly (I put a 60mm maglev on mine. Runs at about 80% and isn't too loud compared to the rest of the system).

I wish some of the companies would actually address the issue instead of keep pushing it to the wayside with repetitive RMA's in the hopes that 780i will be better. I also wish AMD would make a chipset like their 790 for Intel...but we all know that won't happen.
 
nst6563 said:
the same theory about the boards with quads experiencing the problem and the boards with duals not has been bounced around the EVGA forums as well. That being the case, I'd almost lean towards maybe something along the lines of the power circuitry on the boards. Maybe it's not substantial enough to handle two more cores? I know the northbridge gets mighty toasty...which is bothersome but a quiet fan can cure that quickly (I put a 60mm maglev on mine. Runs at about 80% and isn't too loud compared to the rest of the system).

I wish some of the companies would actually address the issue instead of keep pushing it to the wayside with repetitive RMA's in the hopes that 780i will be better. I also wish AMD would make a chipset like their 790 for Intel...but we all know that won't happen.
Click to expand...

This would make some since in the respect that the 680i LT boards don't seem to have these issues, but aside from the LT's having a few less features what else sets them apart? ... The LT has two active fans to keep it cool.
 
Count me as one of the lucky people I guess.

Then again, I have an insane amount of 120s cooling my system.
 
If you look at my sig I have an e6850 not a quad, with at 1333 fsb dual you can still run into issues.

For the record my "fail safe" is to drop an e6400 1066 FSB chip into it along with some 667 mhz 1.8v ram and boot from that. It most often brings the strikers back into alive status.

The A1 is a "quad fixed" board, but not always. From personal experience I can safely say the abit and dfi options (though dfi wisely picked the LT version to go off of) seem to have zero issues with dual cores.

Now for quads, a lot of the boards that don't do well with them seem to fail on vregs, but my 8 phase striker has those on liquid and still will fry.

Another note on the striker. I've fried less of them then the evga parts then strikers. But the striker is so finicky and such a test OC board you can't really say it's the better option. Toss in the fact that evga will send you a board regardless, and fast... ASUS ermm... not a fun experience and if you dare use ASUS update to get a better bios prepare to send them 25 bucks for a new bios.

The real kicker with the nforce chipset is that it and the vregs run HOT. IMHO it's liquid or bust with this chipset. But even then it seems that all boards are prone to over volting things out of the blow resulting in an rma on some part, and there seems to be no logic to this. I set all values manually and still run into it on stock. evga is the worst here because their stock thermal application is horrible.

For the record the longest I have seen an nforce 680i last problem free was liquid cooled, nb, sb, both mosfets on the striker, with custom made back plates on the striker to make sure those suckers held (thanks dave). It's the one striker I've always had come back to life.

Frankly though all these boards are hit and miss, it's luck of the draw. You're best bet is to drop everything on liquid and pray.
 
Proxy said:
AH! We maybe on to something, to go back a bit, I made the statement that the buddy of mine with an OCed E6600 has had ZERO problems. In the same respect the only person I know of comes to my clans ventrilo, he has a QX6700 but the one thing is, he has an Asus Striker Extreme. Now we know it does not OC well, but he has had no problems with it.

I am thinking this is something the [H] team should look into.:cool:
Click to expand...

Well I don't know about us looking into it officially. These posts I make are based on my own experiences which haven't been catalogued scientifically. Nor have I documented all the issues very well. We'd likely have to start from scratch, and with the EVGA 680i SLI's becoming obsolete and replaced by newer 780i SLI designs such an article today would have far less impact than it would have 9 months ago. Hell even six months ago it would have been more worth while.

As for the Striker, mine still works, but the stability isn't very good. It has never died on me. When run with the right ram and dual core CPU it's perfectly stable though.

sd11 said:
If you look at my sig I have an e6850 not a quad, with at 1333 fsb dual you can still run into issues.

For the record my "fail safe" is to drop an e6400 1066 FSB chip into it along with some 667 mhz 1.8v ram and boot from that. It most often brings the strikers back into alive status.
Click to expand...

I've not personally experienced the issue with 1333MHz FSB CPUs as I haven't used one with any 680i SLI chipset based boards.

sd11 said:
The A1 is a "quad fixed" board, but not always. From personal experience I can safely say the abit and dfi options (though dfi wisely picked the LT version to go off of) seem to have zero issues with dual cores.
Click to expand...

The A1 version does indeed fix the problem of not being able to overclock the quad cores past 1300MHz FSB. However it doesn't seem to help much more than that. The longevity of the 680i SLI boards with quad cores seems to be just as hit or miss as ever. Though people running them at stock speeds seem to have less problems. Indeed overclocking the quad core processors might very well stress the boards too hard.

sd11 said:
Now for quads, a lot of the boards that don't do well with them seem to fail on vregs, but my 8 phase striker has those on liquid and still will fry.
Click to expand...

I've noticed that all my boards that died all ran considerably hotter than the boards that are still alive. The ones that ran too hot, (about 20c hotter than normal) all had their thermal paste re-done however it seemed to help very little. I've got no experience running the Striker on liquid.

sd11 said:
Another note on the striker. I've fried less of them then the evga parts then strikers. But the striker is so finicky and such a test OC board you can't really say it's the better option. Toss in the fact that evga will send you a board regardless, and fast... ASUS ermm... not a fun experience and if you dare use ASUS update to get a better bios prepare to send them 25 bucks for a new bios.
Click to expand...

I think you might have hit the nail on the head here. My ASUS board is very picky about the RAM that I use with it and the overclocking is pathetic compared to other boards. Even with dual core processors. I've also had the same problem with killing the BIOS chips on these things through ASUS update. Plus, the CPUINIT problem is about the most annoying problem I've ever faced consistently with motherboards of any model.

sd11 said:
The real kicker with the nforce chipset is that it and the vregs run HOT. IMHO it's liquid or bust with this chipset. But even then it seems that all boards are prone to over volting things out of the blow resulting in an rma on some part, and there seems to be no logic to this. I set all values manually and still run into it on stock. evga is the worst here because their stock thermal application is horrible.
Click to expand...

That's why I was glad to see the Black Pearl came with a VREG water block. So far so good with mine. So far the Black Pearl is the best one I've owned. Runs cooler, and so far it's been rock solid with the ram I've got in there now. I'm using OCZ Flex XLC PC2-9200 modules running at 1143MHz. Unfortunately that only leaves me with 2GB of ram which is unfortunate since I use Photoshop alot. I'll probably just deal with that until the new boards come out, and I'll buy RAM for either what I've got now, or a 780i SLI board in the near future. What would really kick ass is to use the other set of modules that match what I have now. I've got 4 of them and if I had a board that would run 4x1GB I'd really enjoy that.

sd11 said:
Frankly though all these boards are hit and miss, it's luck of the draw. You're best bet is to drop everything on liquid and pray.
Click to expand...

Really, this is about as true a statement as I've ever seen about these things.
 
hmz said:
Wow, thats not funny at all :eek:
Click to expand...

I think that frying memory can generally be avoided by ensuring your RAM voltage is not only manually set, but set to the absolute minimum voltages needed to keep the system stable. For me that's usually 2.1v.

Oh, as an update to my original post, I think that my RAID controller is ok. I just slapped it into another 680i SLI based system I have at home and it seemed to work. I haven't loaded the drivers in Windows, but I have a feeling that my problems with it are actually related to the upgrade to BIOS P31. The other system is still running BIOS P26 or P28 and my problems with my SAS card started when I upgraded the BIOS to version P30. Though I only did the update of the BIOS to try and stabalize the system as nothing else seemed to help. This was before I concluded my memory was toast.
 
Ah I'm not sure the whole quadcore thing applies , although it may be related.

I run an e6600 .stock.everything.

Still havin issues , I know of others not running quadcore who are too.

:(
 
Well I decided to test my modules to find out which one was bad. I installed them both in my other PC running an EVGA 680i SLI board and after 10 hours of Orthos testing I couldn't crash the machine. I even used the machine for a few hours and had no problems either.

So it would appear my Patriot 2GB modules are both still alive and kicking. An interesting note is that machine runs an older BIOS. If I put them in my Black Pearl running BIOS P31 they won't work no matter what I do. After experiecing that I threw my RAID controller into the same machine and it worked flawlessly. So it seems like BIOS P30 and P31 caused some kind of issue with my memory modules and RAID controller.
 
found this post over at EVGA forums... maybe give that a shot? dunno...he claims he's able to recover from C1 or -- errors that way. Not sure if that's the error your getting or what, but I figured if you had some time to kill ;) :p
 
nst6563 said:
found this post over at EVGA forums... maybe give that a shot? dunno...he claims he's able to recover from C1 or -- errors that way. Not sure if that's the error your getting or what, but I figured if you had some time to kill ;) :p
Click to expand...

I'll check it out but I'm not getting any errors on the board at all. It POSTs but it isn't stable and it has some annoying USB issue.
 
the voodoo in Mr.Natural's post may not do the trick then :(

luckily I haven't had any of those errors so far.

btw, do you happen to know what the "BR version" of the EVGA 680i board is? I had an AR version, did the RMA to get the A1 version and it showed up used. dust on it, even had a motherboard standoff still screwed into it as well as some residue on the back of the dimm sockets (looks like the same stuff that comes from bad caps). So I complain to evga and they setup another rma. And this is kinda what ticks me off. They got right on it and got another rma setup without me having to fill anything out, yet I originally paid for the 2-day option and they send the 2nd board out by ground...and took over a week to ship it. Not to mention that instead of A1 version it shows as a BR version.

I might be putting this thing on Ebay and gettin an x38 if money permits after the holidays.
 
nst6563 said:
the voodoo in Mr.Natural's post may not do the trick then :(

luckily I haven't had any of those errors so far.

btw, do you happen to know what the "BR version" of the EVGA 680i board is? I had an AR version, did the RMA to get the A1 version and it showed up used. dust on it, even had a motherboard standoff still screwed into it as well as some residue on the back of the dimm sockets (looks like the same stuff that comes from bad caps). So I complain to evga and they setup another rma. And this is kinda what ticks me off. They got right on it and got another rma setup without me having to fill anything out, yet I originally paid for the 2-day option and they send the 2nd board out by ground...and took over a week to ship it. Not to mention that instead of A1 version it shows as a BR version.

I might be putting this thing on Ebay and gettin an x38 if money permits after the holidays.
Click to expand...

I've never heard of a BR version but it's worth checking out.

I found this through a quick search of EVGA's website: http://www.evga.com/articles/377.asp The BR version of the 680i SLI is listed but the differences aren't.
 
i <3 my evga 680i board 0.0 I hope i don't run into any of these problems, I kiss it before i go to sleep everynight after reading all of these horror stories
 
Digital Viper-X- said:
i <3 my evga 680i board 0.0 I hope i don't run into any of these problems, I kiss it before i go to sleep everynight after reading all of these horror stories
Click to expand...

Just because some people have issues with them doesn't mean you will. I've got three EVGA boards one of which is jacked up, and two of them that are fine. One is new but the other one is nearly a year old and running strong.
 
Dan_D said:
I've never heard of a BR version but it's worth checking out.

I found this through a quick search of EVGA's website: http://www.evga.com/articles/377.asp The BR version of the 680i SLI is listed but the differences aren't.
Click to expand...

yep, I found that link, I also downloaded the manual and specs...both of which are the same documents for the other versions of the board. Wonder if this version is a newer one than the A1 or not? Or it's the red-headed step-version that they only bring out of the closet every so often...

here is another link I found on it
http://www.evga.com/products/moreinfo.asp?pn=122-CK-NF68-BR
 
nst6563 said:
yep, I found that link, I also downloaded the manual and specs...both of which are the same documents for the other versions of the board. Wonder if this version is a newer one than the A1 or not? Or it's the red-headed step-version that they only bring out of the closet every so often...

here is another link I found on it
http://www.evga.com/products/moreinfo.asp?pn=122-CK-NF68-BR
Click to expand...

It may be the lead free version. That seems reasonable considering the context that part number always comes up in.
 
Dan_D said:
It may be the lead free version. That seems reasonable considering the context that part number always comes up in.
Click to expand...

well damn...I'll have to find some other way to poison myself then....time to go dig out the ol' P2 motherboards I have :p
 
Dan_D said:
Just because some people have issues with them doesn't mean you will. I've got three EVGA boards one of which is jacked up, and two of them that are fine. One is new but the other one is nearly a year old and running strong.
Click to expand...

bootcode_usb_driver BSOD is a premature error of your 680i going out
Both my striker/680i evga did that before they went to sleep
 
mccartney said:
bootcode_usb_driver BSOD is a premature error of your 680i going out
Both my striker/680i evga did that before they went to sleep
Click to expand...

I know. My other two boards show no sign of that, only one of those is experiencing that issue.
 
I just thought I would chip in my story......
no overclocking or quad core for me........ it is E6750 at stock 1333 though
After 2 months, I could no longer run dual channel RAM or single chip above 533.

Seems to be a memory controller issue I guess....
Of course I tried many o'ram without luck. My board is at ASUS now in Indiana. Past few weeks they said they were out of stock to do advanced replacement.... wonder why, perhaps others having to RMA their boards every few months....

I need to see a thread dedicated to success stories to cheer me up.
 
Prodeje79 said:
I just thought I would chip in my story......
no overclocking or quad core for me........ it is E6750 at stock 1333 though
After 2 months, I could no longer run dual channel RAM or single chip above 533.

Seems to be a memory controller issue I guess....
Of course I tried many o'ram without luck. My board is at ASUS now in Indiana. Past few weeks they said they were out of stock to do advanced replacement.... wonder why, perhaps others having to RMA their boards every few months....

I need to see a thread dedicated to success stories to cheer me up.
Click to expand...

This is pure speculation on my part, but it is based on observation of many 680i SLI chipset based boards. It has been my experience (as I've stated before) that failing north bridges (680i SLI) or north bridges that are likely to fail will run hotter than they should regardless of thermal solution used. I have noticed a 15c to 20c higher operating temperature in the north bridges that were failing both prior and during the time they were showing problems.

If your north bridge is routinely running around 70c I suspect you will have problems shortly if you aren't already.

Anyone having these issues need to check those temps. There may be a voltage problem or it may simply be poor QC on the 680i silicon itself. I'm not sure. As I said it's just a theory, but I've owned too many of these things to not suspect something is up with them.
 
Dan_D said:
Well, I just killed yet another EVGA 680i SLI board.

I was having increasing stability issues, returned the board to stock, that didn't help, so I flashed it with the latest BIOS and now, it's BSOD hell. Swapped the ram, now the thing won't post anymore.

So I'm installing the EVGA Black Pearl as I type this.
Click to expand...

another? come on man, why dont you leave the poor things alone, and kill some other boards for a while?

jk

i had 2 ar's and 3 a1's and all ran great , but i swapped out to the ip35 when they would not oc the quad too tough. thinking of getting another, or maybe a dfi lt now that i have 2x 8800gt. how are the lt boards with the quad?
 
Dan_D said:
Just because some people have issues with them doesn't mean you will. I've got three EVGA boards one of which is jacked up, and two of them that are fine. One is new but the other one is nearly a year old and running strong.
Click to expand...
I don't care! I won't neglect my motherboard, hugs and kisses every night as long as she works fine :p


on the other hand, yea, no that you mention it, the one board that I worked with that did end up bad, was on a Q6600!, and more so, after a while ,it would boot with a celeron, and refused to boot with the q6600(it did at first) switched it for a new board and it works fine for now(fingers crossed as it's a friends machine)
 
Digital Viper-X- said:
I don't care! I won't neglect my motherboard, hugs and kisses every night as long as she works fine :p


on the other hand, yea, no that you mention it, the one board that I worked with that did end up bad, was on a Q6600!, and more so, after a while ,it would boot with a celeron, and refused to boot with the q6600(it did at first) switched it for a new board and it works fine for now(fingers crossed as it's a friends machine)
Click to expand...

Good luck with that.
 
You must log in or register to reply here.
Back
Top