A few quotes from JonnyGuru on Overclocking the 8800GT ...

[ManWithNoName]

Found an interesting discussion going on over at the Hardforum regarding overclocking the 8800GT. JonnyGuru who is fairly well know for his expertise when it comes to power supplies made an appearance in this thread and issued a warning against overclocking the 8800GT's memory to 2000 Mhz. or above. Whether this is true or not, you can decide for yourself ..............

JonnyGuru Quote:

"2000MHz memory OC is easy with this card BUT the overclock won't hold long term.

The memory is actually 2GHz, but the interface can't handle the speed for long periods. This has been tested by Nvidia, their OEM's and some partners over several weeks. The card will run like a bat out of hell for a while and then suffers irrepairable damage.

Just look at XFX. These guys overclock anything regardless of yield (I guess they don't take the overhead of support into account when they sell stuff) and they're only at 1950MHz on their highest OC card (Alpha Dog).

I would only crank the memory up to 2GHz plus to get a killer benchmark and then crank it back down to 1900~1950MHz for day to day use."

http://www.hardforum.com/showp...031607668&postcount=13

Later on in the thread, the EVGA SCC 8800GT comes up, his opinion remained unchanged.

Edit:

2nd Post ......

JonnyGuru: http://www.hardforum.com/showp...031615854&postcount=47

"Look.. all I'm saying, or all I'm going to say, is that there's a reason why the memory can't run at 2GHz for a long period. Don't take my word for it. Use the fact that everyone is using the same boards, same RAM, same fansinks, and nobody is running the 2GHz RAM at 2GHz. You can do it and when you have to RMA your card, RMA your card. It's your time and money. I'm not one to tell you how to spend either."

3rd Post ......

JonnyGuru: http://www.hardforum.com/showp...031615960&postcount=51

"I never said anything about memory dying.

What's I've said is just my opinion based on what I've seen. That's all. I've probably said too much and since people hold what I say in high regard (God knows why) it's been blown out of propotion. Sorry. Move on."

 

[AzN]

Any overclock can damage your card. You just have to keep the temps down and don't always run at full capacity on a daily basis.

 

[ManWithNoName]

Originally posted by: Azn
Any overclock can damage your card. You just have to keep the temps down and don't always run at full capacity on a daily basis.

Yes of course any overclock can damage your card. This warning just seems a little different and a bit more dire in my opinion. I interpreted his post as you can be overclocking the GT doing all the "right" things, with the card "appearing" to work fine, and then card failure sneaks up and bites you on the ass when you least expect it.

Also as I briefly mentioned the EVGA 8800GT SSC. This card comes overclocked to 2000 Mhz. effective to begin with and according to JonnyGuru this card isn't going to last too long either. I found this suprising since the card is shipped with a 2000 Mhz. memory default clock. JonnyG still stood by his previous statement regard the memory clocks.

 

[Yzzim]

I was just looking at getting the EVGA SSC but the $300 sucked and now this just kills the deal.

Thanks for the heads up.

 

[Munky]

I'm still trying to understand why memory rated at 1000mhz will get damaged from running at its rated speed, without any warning or hint when initially testing for stability. Is it caused by the memory controller, or the specific brand of memory used, or what? Also, is this issue similar to all the 7900gt and 7900gto failures that occured when that card launched?

 

[deerhunter716]

As long as I can get my 1950 memory out of it for about 30 days until I step up to the GTS I will be happy, hehe

 

[ManWithNoName]

Originally posted by: munky
I'm still trying to understand why memory rated at 1000mhz will get damaged from running at its rated speed, without any warning or hint when initially testing for stability. Is it caused by the memory controller, or the specific brand of memory used, or what? Also, is this issue similar to all the 7900gt and 7900gto failures that occured when that card launched?

Same here. The questioning of JonnyGuru on the subject was never pushed that far and his answers were more or less accepted as gospel given his level of knoweledge and experience. So above is about as deep as the conversation went.

 

[ArchAngel777]

Originally posted by: ManWithNoName

Originally posted by: munky
I'm still trying to understand why memory rated at 1000mhz will get damaged from running at its rated speed, without any warning or hint when initially testing for stability. Is it caused by the memory controller, or the specific brand of memory used, or what? Also, is this issue similar to all the 7900gt and 7900gto failures that occured when that card launched?

Same here. The questioning of JonnyGuru on the subject was never pushed that far and his answers were more or less accepted as gospel given his level of knoweledge and experience. So above is about as deep as the conversation went.

Then why don't you question him before spreading the word? Sure, he knows a lot about power supplies, but I am not sure that qualifies him as being an expert in graphics cards and/or overclocking.

I think we can be sure that the interface can handle it, otherwise you are going to see RMA's through the roof. Also, keep in mind that damage done to parts isn't an on/off type of thing, it is gradual. Thus, if 2000 memory causes these things to fail at 2 weeks, you would expect to see 1950 fail after 4 weeks, maybe 6 and so on. It may not be linear, but you can be sure damage does in fact occour. So who knows... Maybe, he is right, maybe these cards are going to die on you after 2 weeks. But I doubt it.

 

[ManWithNoName]

Originally posted by: ArchAngel777
Then why don't you question him before spreading the word? Sure, he knows a lot about power supplies, but I am not sure that qualifies him as being an expert in graphics cards and/or overclocking.

I think we can be sure that the interface can handle it, otherwise you are going to see RMA's through the roof. Also, keep in mind that damage done to parts isn't an on/off type of thing, it is gradual. Thus, if 2000 memory causes these things to fail at 2 weeks, you would expect to see 1950 fail after 4 weeks, maybe 6 and so on. It may not be linear, but you can be sure damage does in fact occour. So who knows... Maybe, he is right, maybe these cards are going to die on you after 2 weeks. But I doubt it.

So sorry, thought I might be doing a few of you a favor or at least giving you a heads-up on a "potential" issue and I thought I would leave you to make up your own mind on the validity of his comments since "most" of everyone here on the forum appears to be intelligent. As to why I didn't question him myself, see post above where I already stated the reason. I provided a link to the thread, please feel free to question him at your leisure.......

 

[nitromullet]

This certainly reflects my own experiences with my GT. I can push the core/shader clock above 700/1750 with an increase in fan speed, but the memory doesn't like to run at even 2000 no matter when speed the fan is at. I prefer near silence over an imperceptible performance gain, so I run mine at 670/1735/1900 with stock fan speeds.

As far as JonnyGuru goes, I would definitely consider him to knowledgeable and trustworthy, plus I think he works for BFGTech now so he's probably got a pretty good scoop on video cards as well.

 

[ManWithNoName]

Originally posted by: nitromullet
This certainly reflects my own experiences with my GT. I can push the core/shader clock above 700/1750 with an increase in fan speed, but the memory doesn't like to run at even 2000 no matter when speed the fan is at. I prefer near silence over an imperceptible performance gain, so I run mine at 670/1735/1900 with stock fan speeds.

As far as JonnyGuru goes, I would definitely consider him to knowledgeable and trustworthy, plus I think he works for BFGTech now so he's probably got a pretty good scoop on video cards as well.

Really? I wasn't aware of this. I'm not suprised however.

 

[AzN]

Originally posted by: ManWithNoName

Originally posted by: ArchAngel777
Then why don't you question him before spreading the word? Sure, he knows a lot about power supplies, but I am not sure that qualifies him as being an expert in graphics cards and/or overclocking.

I think we can be sure that the interface can handle it, otherwise you are going to see RMA's through the roof. Also, keep in mind that damage done to parts isn't an on/off type of thing, it is gradual. Thus, if 2000 memory causes these things to fail at 2 weeks, you would expect to see 1950 fail after 4 weeks, maybe 6 and so on. It may not be linear, but you can be sure damage does in fact occour. So who knows... Maybe, he is right, maybe these cards are going to die on you after 2 weeks. But I doubt it.

So sorry, thought I might be doing a few of you a favor or at least giving you a heads-up on a "potential" issue and I thought I would leave you to make up your own mind on the validity of his comments since "most" of everyone here on the forum appears to be intelligent. As to why I didn't question him myself, see post above where I already stated the reason. I provided a link to the thread, please feel free to question him at your leisure.......

Did any 8800gt die yet? Does Johnny have Proof that the cards will die?

NO! It hasn't been out long enough to warrant these issues. There's no issue except Johnny made a statement regarding what he thought about overclocking the memory too high. Did you even bother reading the whole thread? Later Johnny says people are blowing things way out of proportion. 1 guy says the card will die and everyone follows like the plague.

 

[CrystalBay]

My understanding it was the 256bus that was frying at the higher memory clocks..

 

[ArchAngel777]

Quote from JonnyGURU

I never said anything about memory dying.

What's I've said is just my opinion based on what I've seen. That's all. I've probably said too much and since people hold what I say in high regard (God knows why) it's been blown out of propotion. Sorry. Move on.

There you go, it is what he has seen, known as a 'fair warning'. It doesn't mean it is fact, or that it will happen. I agree, and I think that is great that he popped in there to give his opinion on it. I just am not passive enough, or sheepish enough to believe it without questioning it. No one person is infallable and if you simply look at someone's creditionals and 'presume' that based on those, everything that comes out of their mouth is 100% accurate, then.. Well, I won't even go there.

 

[kylebisme]

Check your B.S. detector. First the guy claimed he was stating a fact discovered by Nvidia and their partners own testing; but two pages after making that claim to authority he states that what he has said is nothing but his own opinion, and suggests that people shouldn't hold what he says in high regard. Obvioius FUD.

 

[ArchAngel777]

Originally posted by: TheSnowman
Check your B.S detector. First the guy claimed he was stating a fact discovered by Nvidia and their partner's own testing; but two pages after making that claim to authority he states that what he has said is nothing opinion, and suggests that people shouldn't hold what he says in high regard. Obvioius FUD.

I WAS going to mention that, but I didn't want to create a flame war...

 

[kylebisme]

Heh, yeah, your post was far more diplomatic. I tend to cut to the point.

 

[ManWithNoName]

Originally posted by: ArchAngel777
Quote from JonnyGURU

I never said anything about memory dying.

What's I've said is just my opinion based on what I've seen. That's all. I've probably said too much and since people hold what I say in high regard (God knows why) it's been blown out of propotion. Sorry. Move on.

There you go, it is what he has seen, known as a 'fair warning'. It doesn't mean it is fact, or that it will happen. I agree, and I think that is great that he popped in there to give his opinion on it. I just am not passive enough, or sheepish enough to believe it without questioning it. No one person is infallable and if you simply look at someone's creditionals and 'presume' that based on those, everything that comes out of their mouth is 100% accurate, then.. Well, I won't even go there.

As I said before I posted it here for the "intelligent" posters at Anand to make up their own minds. For some strange reason which I don't understand this offends you. I never said JonnyGuru was a God, I never said don't buy an overclocked EVGA card and I infact own one. I never said don't overclock the memory or your card or it will explode.

 

[DrMrLordX]

If Jonny works for BFGTech then he probably has some behind-the-scenes knowledge which may or may not be applicable to cards "in the wild" or from cards being sold by other companies. We also need to keep in mind that Jonny tends to favor caution with respect to PSUs (he follows the 50% rule instead of the prevailing 70-80% rule, for example), so it is logical to assume that he will be similarly-cautious with video card memory interfaces.

It's not really fair to say that he recanted or to assume that he's talking out of his arse based on all the things we don't know and what he hasn't gone out of his way to prove. He probably didn't (and doesn't) see any point in trying to prove that this problem could exist with such a new product that hasn't seen months of overclocked operation in end-user's machines. All it was was a warning which is something many of us would rather not heed without solid proof that there is indeed a fragile memory interface on the 8800GT. I mean, heck, I wouldn't want to hear it about my 8800GT, especially if it seemed to run "just fine" at memory speeds of 2000mhz or beyond.

 

[ManWithNoName]

Originally posted by: DrMrLordX
If Jonny works for BFGTech then he probably has some behind-the-scenes knowledge which may or may not be applicable to cards "in the wild" or from cards being sold by other companies. We also need to keep in mind that Jonny tends to favor caution with respect to PSUs (he follows the 50% rule instead of the prevailing 70-80% rule, for example), so it is logical to assume that he will be similarly-cautious with video card memory interfaces.

It's not really fair to say that he recanted or to assume that he's talking out of his arse based on all the things we don't know and what he hasn't gone out of his way to prove. He probably didn't (and doesn't) see any point in trying to prove that this problem could exist with such a new product that hasn't seen months of overclocked operation in end-user's machines. All it was was a warning which is something many of us would rather not heed without solid proof that there is indeed a fragile memory interface on the 8800GT. I mean, heck, I wouldn't want to hear it about my 8800GT, especially if it seemed to run "just fine" at memory speeds of 2000mhz or beyond.

After reading JonnyG's complete posts, I agree, it wasn't actually recant, so I removed the reference. Great common sense post by the way, taking the information given and evaluating it for what it's worth. Where were you when I needed you?

 

[ArchAngel777]

Originally posted by: DrMrLordX
If Jonny works for BFGTech then he probably has some behind-the-scenes knowledge which may or may not be applicable to cards "in the wild" or from cards being sold by other companies. We also need to keep in mind that Jonny tends to favor caution with respect to PSUs (he follows the 50% rule instead of the prevailing 70-80% rule, for example), so it is logical to assume that he will be similarly-cautious with video card memory interfaces.

It's not really fair to say that he recanted or to assume that he's talking out of his arse based on all the things we don't know and what he hasn't gone out of his way to prove. He probably didn't (and doesn't) see any point in trying to prove that this problem could exist with such a new product that hasn't seen months of overclocked operation in end-user's machines. All it was was a warning which is something many of us would rather not heed without solid proof that there is indeed a fragile memory interface on the 8800GT. I mean, heck, I wouldn't want to hear it about my 8800GT, especially if it seemed to run "just fine" at memory speeds of 2000mhz or beyond.

Great post and something I definately agree with.

 

[1ManArmY]

Fear, Uncertainty and Doubt over a video card?

 

[kylebisme]

Originally posted by: ManWithNoName
Serves me right for trusting a partial quote from the other poster.

Nonsense, ArchAngel quoted the whole post. Why are you so insistant pushing this FUD?

 

[ManWithNoName]

Originally posted by: 1ManArmY
Fear, Uncertainty and Doubt over a video card?

Ooh-rah, not here sir .....

The only fear I have is that I won't be able to kick my kid off the computer later so I can get some time with the 8800GT and Bioshock. (Yes, we already had some quality time together eariler today)

The only doubt I have is whether or not I'll "Step-up" once the new EVGA GTS or better comes out.

The only uncertainty I have is whether should I go for a second party cooling solution or keep with the stock cooling and the fan cranked to 70%

 

[ManWithNoName]

Originally posted by: TheSnowman

Originally posted by: ManWithNoName
Serves me right for trusting a partial quote from the other poster.

Nonsense, ArchAngel quoted the whole post. Why are you so insistant pushing this FUD?

Ill be damned, you're right and my post has been edited to reflect that.

Yeah, I never gave his post a complete read because he ticked me off when he came after me for quoting a verified/linked to post by someone who's intelligent and generally knows what he's talking about. That does not fit the definition of FUD in my book...

AND NO ONE HAS BEEN PUSHING FUD! ALLOW ME TO QUOTE MYSELF.....

I've said this at least twice in this thread ....

"As I said before, I posted it here for the "intelligent" posters at Anand to make up their own minds and for some strange reason which I don't understand this offends you. I never said JonnyGuru was a God, I never said don't buy an overclocked EVGA card and I infact own one. I never said don't overclock the memory or your card or it will explode."