Burpo
Diamond Member
- Sep 10, 2013
- 4,223
- 473
- 126
7980XE @ 4800Mhz
Asus ROG Rampage VI Apex
32Gb G.SKILL - 3200Mhz 12-12-12-28 1t
Wow! Wicked System!
Last edited:
7980XE @ 4800Mhz
Asus ROG Rampage VI Apex
32Gb G.SKILL - 3200Mhz 12-12-12-28 1t
This is like when your best buddy takes on the school bully and opens a can of whoopass on his *ss!! Man!!!!
7980XE @ 4800Mhz
Asus ROG Rampage VI Apex
32Gb G.SKILL - 3200Mhz 12-12-12-28 1t
Ladies and gentlemen, we have a new champion, a single CPU score that is not only fastest in its class, but also faster than any multiprocessor score posted here to date! Congratulations, @Totalnet , care to share any extra information about your build? I'd be interested in how how keep it all cool.
7980XE @ 4800Mhz
Asus ROG Rampage VI Apex
32Gb G.SKILL - 3200Mhz 12-12-12-28 1t
Don't think you'll find any 11.5 reporting more than 16 cores per CPU and AFAIK in the case of multi socket not more than 32. If you can post a link showing otherwise, please do.I was joking about the earlier photoshop/pentium 4 joke, but why does Cinebench report 16C/32T?
I've seen it scale as high as 48 cores on an intel chip, so it's not cinebench, did you seriously doctor up a benchmark just to try and get to the top of the list? I know a guy with 2 less cores than you, but the same clockspeed as you, and he gets around half that score. It's doctored or you glitched it.
I was frankly shocked to read that; calling someone out like that without knowing all the facts. Not good.Don't think you'll find any 11.5 reporting more than 16 cores per CPU and AFAIK in the case of multi socket not more than 32. If you can post a link showing otherwise, please do.
That's a terrible thing to say. Take my 3.15GHz 14 core Haswell at 23.14pts and do some math to convert to 18 cores at 4.8GHz
23.14 * (18 / 14) * (4.8 / 3.15) = 45.3pts
Scaling probably not perfect from 14 to 18 but that CPU is several generations older and limited to 2133MT/s DRAM. Seems either that 'guy' is just not doing it right (TR seems to give people a lot of trouble with CBR11.5) or just about everybody else is doctoring their scores.
I was joking about the earlier photoshop/pentium 4 joke, but why does Cinebench report 16C/32T? I've seen it scale as high as 48 cores on an intel chip, so it's not cinebench, did you seriously doctor up a benchmark just to try and get to the top of the list? I know a guy with 2 less cores than you, but the same clockspeed as you, and he gets around half that score. It's doctored or you glitched it.
Because I work a full time job. Your 2 windows did not match up with each other, hence why I called you out. An above poster provided a reasonable explanation, so I'm a bit less skeptical. However, being an Intel user all my life, I still find that score a bit hard to believe and that's NOT because I have an AMD chip in my system currently. As an example, HWBot provides even more reasons to be skeptical of your result, mainly that anyone wishing to get over 5 GHz must resort to LN2, so your claim of hitting nearly 5.5 GHz on water (and therefore beating all the results listed above on HWbot) is also suspicious to me. Call me what you want, the facts here are clear. If you wish to argue, feel free to provide more evidence (or not). Just understand that you are claiming to be one of the top 10 overclockers on HWbot...and possibly the world for your CPU...on water. That is something that requires extraordinary proof, regardless of your opinion of me.To be called a scammer by some badly informed sore loser is difficult to digest, it takes a lot of effort (and skill) to get this kind of results.
Even worse is that you are not even able to say that you were wrong.
Just my 2 cent..
Because I work a full time job. Your 2 windows did not match up with each other, hence why I called you out. An above poster provided a reasonable explanation, so I'm a bit less skeptical. However, being an Intel user all my life, I still find that score a bit hard to believe and that's NOT because I have an AMD chip in my system currently. As an example, HWBot provides even more reasons to be skeptical of your result, mainly that anyone wishing to get over 5 GHz must resort to LN2, so your claim of hitting nearly 5.5 GHz on water (and therefore beating all the results listed above on HWbot) is also suspicious to me. Call me what you want, the facts here are clear. If you wish to argue, feel free to provide more evidence (or not). Just understand that you are claiming to be one of the top 10 overclockers on HWbot...and possibly the world for your CPU...on water. That is something that requires extraordinary proof, regardless of your opinion of me.
7980XE @ 4800Mhz
Asus ROG Rampage VI Apex
32Gb G.SKILL - 3200Mhz 12-12-12-28 1t
Thank you!! the bios setting was 1.350vCore voltage returns to near-idle when not under load.
4.8GHz at 1056mV...?
FTR on air at 950mV Hardware.fr get 3.8GHz and at those settings they have low die temp, yet at this temp they wouldnt get higher than 4.65GHz@1056mV even if the frequency/voltage curve of the CPU was an ideal one, wich would imply frequency increasing as the square of the voltage..
Besides their 7980X didnt clock higher than 4.2 and here we re talking of 4.8 at surprisingly low voltage, as for 5.5 for this CPU i dont think that it s possible without LN2, the reason is that in a CPU and in Cinebench most of the heat arise from the FPU wich is small part of the die, hence the temperature gradient within the die is too high.
It s different in a GPU where the exe units take most of the die area and allow for homogeneous heating within the whole area.
Last, but not least, 1056mV/4.8GHz would drain barely 300W in CB, far from the stated 600-700W wich would imply 1.5V at this frequency...
http://www.hardware.fr/articles/969-4/overclocking.html
CPU-Z in your link is not measuring core voltage, it is measuring core VID, two different things. Since core voltage is read through a relatively slow ADC it represents an integration of the voltage (average) unlike the VID which is an instantaneous reading that needs the core active to be able to take it. On top of that the CPU supports per core p-states which means each core can be at a different voltage at the same moment in time but we have only one single measurement. IOW don't put too much faith into core voltage.4.8GHz at 1056mV...?
FTR on air at 950mV Hardware.fr get 3.8GHz
http://www.hardware.fr/articles/969-4/overclocking.html
I'm not surprised at the disbelief coming from the AMD camp. For months they went on and on about how Threadripper was going to trump this 18-Core chip because SKL-X runs hot, is a power hog, VRM won't hold etc, etc. The truth is these chips are an engineering feat (the main reason I want one). Common overclocks on water are around 4.4 - 4.6GHz. Others have even pushed it to 4.8Ghz on water https://www.tweaktown.com/news/58970/core-i9-7980xe-overclocked-4-8ghz-18c-36t/index.html and 4.9Ghz here:http://www.legitreviews.com/intel-core-i9-7980xe-18-core-processor-review_197903/10 Your chilled water overclock is well in the ballpark.
I just improved my score, thanks guys.. for pushing me...lol
I'm not surprised at the disbelief coming from the AMD camp.
I thought eek2121 was a lifetime Intel user?