This may or may not help. This is how my BIOS is set. I adjust up and down from this with ThrottleStop profiles. If theses settings don't work for you, don't jump to conclusions... it doesn't mean your CPU is no good... just tinker until you find some settings that do work. You will rarely ever find two CPUs that like identical settings at high clock speeds.
And, I'm sorry for being lazy... recorded this with the laptop in my lap, sitting my recliner with the phone in one hand and the other hand on the keyboard. I didn't remove the audio distractions, so mute the sound. Will make something more professional by my own standards later on.
-
-
-
-
Here's a quick no-edit video showing power draw in 3DMark 11 Graphics Tests 1 through 4 and Physics test. I don't have 5.2GHz dialed in good enough for 3DMark 11 yet, so it freezes at 41 FPS in the Physics Test. I can run the Physics Test by itself with a score over 14K in Windows 7, but if I run the full benchmark it always freezes once it hits 41 FPS and I haven't been able to figure out why.
My 7700K pulls about 125-130W for itself at 5.2GHz running Cinebech or wPrime. About 105W package power in 3DMark 11 or Fire Strike Physics tests.
Watch full-screen and you can read the power draw in watts on the UPS meter.
Last edited: Jan 15, 2017jaybee83, afloyd, FredSRichardson and 3 others like this. -
Yeah, I expect that with 52x. But my Q was about lower clock speed like 4.6 or around vs. same for 6700K. You said Kaby was hotter with this clock speed.ajc9988 likes this.
-
what do you get at 4.6 ghz?
-
Yes, it is WAY HOTTER... like +15 to 20°C compared to 6700K at 4.6GHz. I rarely ever used hit 80°C at 4.7GHz with 6700K benching without AC and it will hit ~99° benching at 4.7GHz without AC cooling.
-
-
-
-
There should be an explanation for this. Normally, lower voltage = lower wattage → Lower heat. There should be no major difference between these two chips. The only differences is iGPU and IHS as I understand.jaybee83 likes this.
-
Brother @Papusan - note stock load voltage in fine, but higher watts and ludicrous temps at only 4.5GHz after delid with CLU. And, no... iGPU is not getting hot. I enabled the iGPU sensor to show it is inconsequential.
Having HWiNFO64 or AIDA64 running during benchmarks causes a very abnormal and measurable performance hit on CPU with 7700K. I am guessing both software applications need an update of some kind.
At 4.7GHz CPU is around 109W.
Last edited: Jan 15, 2017jaybee83, afloyd, ajc9988 and 1 other person like this. -
-
Are you absolutely sure that the sensors show correctly? Same for TS as well(Only use TS monitor in bench)? This is crazy numbers. Both Power and temp.jaybee83 likes this.
-
http://valid.x86.fr/vwsrm6
5ghz -
Cinebench :
Screenshot app decided to take a stroll ..
@Johnksss@iBUYPOWER How much of a difference do you think the Vapor Chamber heatsink makes vs the standard solo CPU heat sink ? (When it comes to only CPU loads)
@Papusan the max power draw i saw during Cinebench @ 5ghz was 110W.Last edited: Jan 15, 2017 -
Thanks. Please test same clock speed and bench as previous post from @Mr. Fox and compare. Do the same as him. I don't know if hi used max fans. See if you get the same numbers.
-
Which is to say it works period? xD
This falls in line with what I heard elsewhere. The CPU is fine for benching and such but multitasking specifically appears to be really bad.
Hey @Donald@HIDevolution do you think you all could re-add Skylake as an option to the P870 models on HIDevolution? I checked earlier and even the P870DM3 only had Kaby Lake as an option, but until we get to the bottom of the multitasking issue with Kaby Lake, it might be best to let people choose the previous gen. Even if it's the same price. Yes, I know I discovered this idea at 3am my time, but... you can do it on monday or whenever
jaybee83 and FredSRichardson like this. -
Sure thing, do you need me to use the exact same OC settings? or just clocks?Papusan likes this.
-
Try all the same
Ps. Got my best number with 46x in Cinebench R11.5 - 11.25
Could you as well compare Kaby with same clock speed. You should get the same as well.
bloodhawk likes this. -
What do you get for temps at 4.8GHz? HWiNFO64 and ThrottleStop show within 1°C of the same, which is close enough (could be data polling rate with minor variance). I lowered watts and temps were maybe 2-3°C less, but not enough to matter. Limiting watts also limits performance.
Edit: @Papusan - Here is the 4.6GHz you keep asking for. Stock voltage. Max fans above and below. Temps are basically the same at 4.6 and 4.8 after limiting watts (along with performance). So, as you can see, MUCH MUCH HOTTER than 6700K. With more watts the temps just go higher, but so does the performance.
Last edited: Jan 15, 2017afloyd, FredSRichardson and Papusan like this. -
And the Wprime 32M still have the weird behavior on Kaby lake?
Earlier bench 4.8GHz. The power draw is higher as expexted. Around +13W for Skylake with 48x
Edit: You get 86w vs almost 100W power draw from Skylake.
This is the numbers I don't understand. Why in a so short bench?
Last edited: Jan 15, 2017Prema likes this. -
-
Sorry my English
, but what you meant by this?
-
He limited power draw. I.E. his CPU was prevented from pulling 100W by him, manually.Papusan likes this.
-
There are still problems with Wprime 32? You must run several times before the scores will be correct? See @Mr. Fox 32M score.
[email protected]
-
Sometimes I think you forget who you are talking too.
Ps. Your os was optimized for your 1125. We are not.
-
Maybe I did not speak in layman correct English, sir
-
The Wprime 32 was click and run. With antivirus and Hwinfo up and go. What more could I do? Fire up virus scan?
It was this test I talked about.
-
How about you run it right now?
Edit nevermind. I see what is going on.
ajc9988 likes this. -
-
This is @ 4.6Ghz , Ring @ 4.2Ghz -
EDIT : WHOOPS! Forgot to reset BCLK. BRB -
No worries. I think it was I that was misunderstanding the situation.
Edit:
@bloodhawk Bclk at 102.07 my friend. -
-
LOL!
I'm Starting to wonder if the first gen machine is better at cpu cycles than the dm2/3....ajc9988 likes this. -
You shave of 10W vs. 6700K. Maybe you can shave of more? You like playing with 7700K in DM1?
Johnksss likes this. -
@Papusan - Here is the proper 4.6Ghz / 4.2Ghz @ 100BCLK
Dont jinx it yo!
Yeah im not sure whats up with everyone's results being all over the place. Loving the clocks so far. And i think the system can actually be used @ 4.8 to 5.0 Ghz as 24/7 clocks.
Let me know if you need more tests.afloyd, jaybee83, Johnksss and 1 other person like this. -
Nice. 67W ~ aka almost 13W lower than 6700K
I want this as well. Can you go lower?
FredSRichardson and Johnksss like this. -
-
-
Thank you Mr. Fox. I'll give this a shot. My silicon lottery CPU was rated for 5ghz. Now to keep it cool.
Sent from my iPhone using Tapatalk -
So I'm curious, Are any of you using the x264 benchmark to test for temps / stability or just quick apps like wPrime and Cinebench? The temps I get out of x264 are much hotter than the other two (obviously). I don't plan on encoding a lot of video on this but that's what I was using as my "medium" stress of worst case scenario.
-
lctalley0109 Notebook Evangelist
Did Sager send you the new bios, are you waiting for the bios or are you using a different version? -
I got what I needed from Mr. Fox. I'm on the Prema BIOS and EC.
Sent from my iPhone using Tapatalk -
To clarify, I got the BIOS from my reseller, I got sample BIOS settings from the video that Mr. Fox posted. Sorry for the confusion here.
Sent from my iPhone using Tapatalk -
Meaker@Sager Company Representative
Well my liquid metal has not shorted anything yet. Hopefully I can get back to both PSUs and give it a proper test.
-
-
FredSRichardson Notebook Groundsloth
@bloodhawk - wow! What machine are you running the 7700K in?
-
According to his screenshot CPU-Z says its the P870DM
Clevo Overclocker's Lounge
Discussion in 'Sager/Clevo Reviews & Owners' Lounges' started by Spartan@HIDevolution, Mar 4, 2016.
