-
Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!
Ordered an 8700k has anyone had heat issues with these chips?
- Thread starter J☼E
- Start date
- Thread Starter
- #3
This review had me worried
A hot mess and an overpriced paper launch: Intel's EPYC failure of an attempt of Ryzen to the occasion against AMD. Did you ever think the 7700K was slightly too hot for comfort in your build but barely managed anyhow? Take all the ills and missteps of the thermals of the 7700K and magnify them to unimaginable levels in this flagship mainstream processor. Though you would expect a mainstream-targeted processor outside of the HEDT X299 line to be entry-level builder friendly, the 8700K is actually more demanding than the 7800X and practically requires liquid cooling. In testing, the 8700K, with a $150+ premium AIO like the NZXT Kraken X62 or the Alphacool Eisbear 420, has been found to be up to 20 degrees hotter than its predecessor. Google "8700K temperature" and you will find numerous accounts of reviewers and users having temperatures in the 80 and 90 degree Celsius range while constantly hitting thermal throttling. Do not even remotely consider pairing a Cooler Master Hyper 212 EVO with this hot shot or throwing this fiery devil into a small form factor case like the RVZ02 or the DAN Cases A4-SFX. Intel is making the same mistakes from the darker times of its past and is returning to its Pentium 4 ways with this underwhelming, steaming pile of product. Practically speaking, Ryzen also achieves much greater than 60 fps in most games (not even the 8700K can achieve 144 fps+ for high refresh rate monitors unless you drop graphical settings to levels where Ryzen is practically on the same footing) and still has the upper hand in multi-threaded performance and power consumption. My strong suggestion: either (1) purchase a Ryzen 7 1700 processor and overclock it to 1800X levels, (2) wait until the 12nm Ryzen refresh comes out in February/March 2018, or (3) wait a bit longer for 7nm Zen 2 (SemiAccurate, a costly but highly reputable professional tech industry news outlet, affirms Zen 2 is still on schedule) to come out by the end of 2018. Less
A hot mess and an overpriced paper launch: Intel's EPYC failure of an attempt of Ryzen to the occasion against AMD. Did you ever think the 7700K was slightly too hot for comfort in your build but barely managed anyhow? Take all the ills and missteps of the thermals of the 7700K and magnify them to unimaginable levels in this flagship mainstream processor. Though you would expect a mainstream-targeted processor outside of the HEDT X299 line to be entry-level builder friendly, the 8700K is actually more demanding than the 7800X and practically requires liquid cooling. In testing, the 8700K, with a $150+ premium AIO like the NZXT Kraken X62 or the Alphacool Eisbear 420, has been found to be up to 20 degrees hotter than its predecessor. Google "8700K temperature" and you will find numerous accounts of reviewers and users having temperatures in the 80 and 90 degree Celsius range while constantly hitting thermal throttling. Do not even remotely consider pairing a Cooler Master Hyper 212 EVO with this hot shot or throwing this fiery devil into a small form factor case like the RVZ02 or the DAN Cases A4-SFX. Intel is making the same mistakes from the darker times of its past and is returning to its Pentium 4 ways with this underwhelming, steaming pile of product. Practically speaking, Ryzen also achieves much greater than 60 fps in most games (not even the 8700K can achieve 144 fps+ for high refresh rate monitors unless you drop graphical settings to levels where Ryzen is practically on the same footing) and still has the upper hand in multi-threaded performance and power consumption. My strong suggestion: either (1) purchase a Ryzen 7 1700 processor and overclock it to 1800X levels, (2) wait until the 12nm Ryzen refresh comes out in February/March 2018, or (3) wait a bit longer for 7nm Zen 2 (SemiAccurate, a costly but highly reputable professional tech industry news outlet, affirms Zen 2 is still on schedule) to come out by the end of 2018. Less
- Joined
- Jul 26, 2004
nice clocks enjoy your new system .
What is the max req voltage for CL ?
I see ppl quote 1.35v for haswell but am living proof that 1.4+ for years hasn't been a issuse .
What is the max req voltage for CL ?
I see ppl quote 1.35v for haswell but am living proof that 1.4+ for years hasn't been a issuse .
- Joined
- Jul 26, 2004
J☼E;8041592 said:
Are your voltages set on auto ? (not a good idea )
What are you using to stress test ? P95 ? AIDA64 ? occt . or just booting windows ( this doesnt meen it is any way stable)
- Joined
- Jul 26, 2004
J☼E;8041604 said:
what is that in C ? quote temps in C every one uses that for temps
* 48 degc wow that is nice @ 5ghz !
- Thread Starter
- #15
https://valid.x86.fr/4ql28g
yeah 50c load, I could get the voltage down a little more I think, going to do some more tomorrow
yeah 50c load, I could get the voltage down a little more I think, going to do some more tomorrow
Actually, thats pretty average, really. According to silicon lottery, 81% of 8700k they binned reach 5ghz at 1.4V. 
https://siliconlottery.com/collections/all/products/8700k50g?variant=224965885964
Hey Joe..... 50c at 1.4V without a delid???? What kind of cooling do you have?? That doesnt sound possible...which p95 version did you use? Which stress test...small fft is what you should be using.
https://siliconlottery.com/collections/all/products/8700k50g?variant=224965885964
Hey Joe..... 50c at 1.4V without a delid???? What kind of cooling do you have?? That doesnt sound possible...which p95 version did you use? Which stress test...small fft is what you should be using.
Last edited:
Laat i recall, they use realbench for an hour with offsets. Realbench uses avx instructions iirc... otherwise whats the point of adding a negative "avx" offset when it isnt tested? Its testing tends to be more thorough than p95 since its doing a more than just beating on it with different length fft. Real world loads. Regardless what it is, avx testing is part of their suite.
Edit: they use p95 no avx, linpack, and other according to a forum post there. It used to be RB only.
Ive moved away from aida to rb for my personal stability testing.
Id have no idea how he managed 50c at those clocks and voltages with an avx p95 or non for that matter.....something is off there.
Edit: they use p95 no avx, linpack, and other according to a forum post there. It used to be RB only.
Ive moved away from aida to rb for my personal stability testing.
Id have no idea how he managed 50c at those clocks and voltages with an avx p95 or non for that matter.....something is off there.
Last edited:
- Thread Starter
- #20
Speedfan was reporting my temps incorrectly, its actually around 90 load
- - - Updated - - -
and thats prime small fft test
also if i turn avx off in local.txt it only gets to 83C
CpuSupportsAVX=0
CpuSupportsAVX2=0
CpuSupportsFMA3=0
- - - Updated - - -
and thats prime small fft test
also if i turn avx off in local.txt it only gets to 83C
CpuSupportsAVX=0
CpuSupportsAVX2=0
CpuSupportsFMA3=0
Last edited:
