Contribute
Register

Do I have a bad 3770k chip?

Status
Not open for further replies.
...I would rather have one that operates normally, then one that requires obscure voltages but stays reasonably cool.
Normally means default settings... Overclocking = "bragging rights", no more. Wondering what apps you use that you could possibly benefit from a few 100 Mhz. Not thinking you use any, most people don't. Property setup a 3770k will get you Geek-bench scores of approx. 15000, if you clock the crap out of it with proper cooling you may start seeing mid 16000's. At mid 16000's there's no way you're going to realize a better desktop experience, not going to happen. Enjoy your build, this is not an argument, just a perspective from someone that's been screwing around with OCing for 15+ years.
 
Normally means default settings... Overclocking = "bragging rights", no more. Wondering what apps you use that you could possibly benefit from a few 100 Mhz. Not thinking you use any, most people don't. Property setup a 3770k will get you Geek-bench scores of approx. 15000, if you clock the crap out of it with proper cooling you may start seeing mid 16000's. At mid 16000's there's no way you're going to realize a better desktop experience, not going to happen. Enjoy your build, this is not an argument, just a perspective from someone that's been screwing around with OCing for 15+ years.
Thanks for your reply. My geekbench score is a 13500 without OC'ing, overclocked it goes into mid 15000s and I notice handbrake video encoding is noticeably faster (32 bit geekbench). By no means would I normally notice any difference from normal usage, but converting video is something I notice and it benefits from it substatiably. I am not afraid of over clocking in any way, I am just afraid I got a bad chip and will be having problems with it in the future.

Ive bridge almost always offers free performance with a light overclock. However for some reason I must of gotten a weird chip that requires lots of voltage to be stable. What concerns we is that it is a lot more then I have ever read anybody needing to achieve 4.5ghz (my goal). Since you have been over clocking for 15 years, can you tell me how much vcore its taken you to achieve 4.5 ghz on your ivy bridge chips?
 
No, too many variables. OCing has always been luck of the draw. If you're not happy with the cpu return it, but the chip is doing what Intel intended it to do. If you return it because it doesn't reach unrealistic goals you have no assurance the next one will do any better. I for one don't believe in hosing a manufacturer when a product works as advertised, but to each his own.
 
No, too many variables. OCing has always been luck of the draw. If you're not happy with the cpu return it, but the chip is doing what Intel intended it to do. If you return it because it doesn't reach unrealistic goals you have no assurance the next one will do any better. I for one don't believe in hosing a manufacturer when a product works as advertised, but to each his own.
Thanks for your reply. I just don't understand why you consider "4.5ghz an unrealistic goal" most people can go up to 4.3ghz on stock voltages. So by no means would 4.5ghz seem unrealistic to me. An unrealistic goal would be 5ghz.

I'm not happy with these voltages so I think I will be exchanging it, I'm sure any cpu will have better voltages then this one. Once again I appreciate your quick replies.
 
I really appreciate your insight!:)

Yeah, I guess, I'm sure voltages play a part too. But I dont understand how a better die-heatspreader interface can have a such a drastic effect on temperatres. Maybe it has bad temp sensors? I just don't feel comfortable having to push 1.32V through an ivy bridge for 4.5ghz, while almost everybody can do it with 1.25V. It just seems way off, almost as if its a sandy bridge marked as an ivy bridge. I tried underclocking it at stock settings and it was stable only up till 1.1V, I've read people going as low as 0.9V at the stock settings and stable.

Because the die-heatspreader interface is often terrible on IB (do a search, tons of pages on it, IB uses paste for the interface as opposed to solder used in SB). But how terrible is variable. CPUs die when they short out. When the current is too much for the 'correct' pathways to accommodate, the current will find an alternate path (arcing). Resistance is the limiting factor, and temperature is a good surrogate for resistance. Liquid nitrogen allows such high overclocks because at lower temperatures, the circuitry can handle more current.

Let me put it another way, which would you be happier with, 4.5Ghz on 1.25v and temps in the 90's, or your set up? Obviously yours is better. So, do some searches to see what kinds of temps (ignore voltage) people have seen at certain levels of over clocking using a CM212. I doubt you will find any better than your result.

There is always a lottery when getting a chip. Prior to IB, the lottery only related to the die itself, how many volts it needed for a certain OC. With IB they added another variable, how well the die-heatspreader interface ended up. It seems you got a poor result in the first lottery, but a great result in the second. If you exchange your chip you will most likely get a better result in the first, and a worse result in the second. If your main goal is lower voltage, by all means exchange away. If your main goal is actual results, I'd think long and hard about it.
 
Because the die-heatspreader interface is often terrible on IB (do a search, tons of pages on it, IB uses paste for the interface as opposed to solder used in SB). But how terrible is variable. CPUs die when they short out. When the current is too much for the 'correct' pathways to accommodate, the current will find an alternate path (arcing). Resistance is the limiting factor, and temperature is a good surrogate for resistance. Liquid nitrogen allows such high overclocks because at lower temperatures, the circuitry can handle more current.

Let me put it another way, which would you be happier with, 4.5Ghz on 1.25v and temps in the 90's, or your set up? Obviously yours is better. So, do some searches to see what kinds of temps (ignore voltage) people have seen at certain levels of over clocking using a CM212. I doubt you will find any better than your result.

There is always a lottery when getting a chip. Prior to IB, the lottery only related to the die itself, how many volts it needed for a certain OC. With IB they added another variable, how well the die-heatspreader interface ended up. It seems you got a poor result in the first lottery, but a great result in the second. If you exchange your chip you will most likely get a better result in the first, and a worse result in the second. If your main goal is lower voltage, by all means exchange away. If your main goal is actual results, I'd think long and hard about it.
Thanks so much. this is exactly the kind of answer I was looking for.

After reading numerous reviews and through tons of forums even with a 3770k with the worst die-heatspreader interface, I should not be anywhere close to 90C at 4.5ghz using the CM212. I find the worst case scenario is I get a chip that runs at basically the same temp at a much lower voltage.

It's unfortunate intel didnt go with solder. Even if I get an average chip after exchanging it, I think it would be still be significantly better in terms of at least voltages. And because their is less voltage, the temps would be lower which would make up for the poorer die-heatspreader interface. I have thought about it, and I think I'm going to exchange it, because it now is requiring 1.332V with a low LLC (like you recommended).
 
Just an update. I exchanged the chip and now I got to my 4.5ghz with a voltage of 1.188V, I got a really good chip. Power consumption from the wall down 20 watts. Temperatures with my cm212 is under 70C. Thank you guys for all your help. I'm glad I exchanged it:clap:
 
Just an update. I exchanged the chip and now I got to my 4.5ghz with a voltage of 1.188V, I got a really good chip. Power consumption from the wall down 20 watts. Temperatures with my cm212 is under 70C. Thank you guys for all your help. I'm glad I exchanged it:clap:

Very nice, congrats =)
 
Status
Not open for further replies.
Back
Top