EVGA

Helpful ReplyHot!Concensus on safe daily mesh voltage

Page: 123 > Showing page 1 of 3
Author
glocked89
Superclocked Member
  • Total Posts : 187
  • Reward points : 0
  • Joined: 2019/01/16 23:48:00
  • Status: offline
  • Ribbons : 0
2019/12/30 12:25:01 (permalink)
Thoughts on 3.2 at 1.2v?
 
 
#1
jasoncodispoti
SSC Member
  • Total Posts : 628
  • Reward points : 0
  • Joined: 2013/03/18 14:40:28
  • Location: Nashville, TN
  • Status: offline
  • Ribbons : 14
Re: Concensus on safe daily mesh voltage 2019/12/30 13:01:04 (permalink)

Intel Core i7-9800x | EVGA X299 Dark| EVGA RTX 2080ti FTW3 Ultra HydroCopper | 32GB Corsair Dominator Platinum | EKWB Liquid Cooled Loop | Phanteks Enthoo Primo SE | EVGA SuperNOVA 1000 P2


#2
GGTV-Jon
FTW Member
  • Total Posts : 1813
  • Reward points : 0
  • Joined: 2017/11/25 14:11:43
  • Location: WA, USA
  • Status: offline
  • Ribbons : 19
Re: Concensus on safe daily mesh voltage 2019/12/30 18:38:09 (permalink) ☄ Helpfulby Cool GTX 2024/01/17 09:47:12
Which CPU and what are you using for cooling?
 
My 7940x @ 4.8ghz (X299 Dark )has been running 32 mesh with voltage set to override / 1.35 volts - same as vcore - custom loop
 
Biggest part is testing till you find how little you can get away with and be stable - this is just where I ended up


#3
glocked89
Superclocked Member
  • Total Posts : 187
  • Reward points : 0
  • Joined: 2019/01/16 23:48:00
  • Status: offline
  • Ribbons : 0
Re: Concensus on safe daily mesh voltage 2019/12/31 08:46:27 (permalink)
Thanks jasoncodispoti! That thread was actually very informative. Although Intel maximum and minimum voltages are probably conservative, its nice to read something directly from them.
 
Hey Jon, I'm on a 10980xe using a 360 clc. This many cores get quite hot so I'm being very conservative with 4.5ghz all core at 1.1v. I'm looking into getting a custom loop as well for this beast.
 
My mesh is at 32 with 1.2v with VCCIO at 1.1.
 
Another thing I'd like to ask is safe daily VCCIO voltage? The overclocking guide at xDevs said to start at 1.250, but I keep reading posts online talking about how thats too much. Intel's spec sheet that was also posted by jasoncodispoti's thread also listed VCCIO max voltage at ~1.050v
#4
GGTV-Jon
FTW Member
  • Total Posts : 1813
  • Reward points : 0
  • Joined: 2017/11/25 14:11:43
  • Location: WA, USA
  • Status: offline
  • Ribbons : 19
Re: Concensus on safe daily mesh voltage 2019/12/31 09:17:35 (permalink)
The newer chips may have a different voltage range. You will have to find the balance between how fast you can run it with the lowest voltage and temps. If it takes more voltage and produces too much heat for your aio to keep up with then lower the OC.

You may want to do a bios reset and run a few stress tests with whatever you are running to watch min / max temps and voltages

Personally I like HWiNFO64 for watching my numbers, others may suggest something else.

Get the baseline to see what voltages it runs on it's own and where the temps land. If it is too hot still then you need to go back and look at your cooler mounting


#5
glocked89
Superclocked Member
  • Total Posts : 187
  • Reward points : 0
  • Joined: 2019/01/16 23:48:00
  • Status: offline
  • Ribbons : 0
Re: Concensus on safe daily mesh voltage 2019/12/31 09:44:16 (permalink)
Oh yes, I haven't listed any temps. I'm trying to stay at around 80c all core which I am able to do with my current settings so I know I do have some headroom here.
 
4.8ghz, 1.35v. What are your temps on a like a cinebench r20 (lets say ambient temps 22c)?
 
And yes HWinfo64 is my go to as well
#6
cdc-951
SSC Member
  • Total Posts : 520
  • Reward points : 0
  • Joined: 2012/04/27 02:26:30
  • Status: offline
  • Ribbons : 1
Re: Concensus on safe daily mesh voltage 2020/01/04 05:13:04 (permalink)
Mesh 31 1.075v on auto in motherboard.
#7
cdc-951
SSC Member
  • Total Posts : 520
  • Reward points : 0
  • Joined: 2012/04/27 02:26:30
  • Status: offline
  • Ribbons : 1
Re: Concensus on safe daily mesh voltage 2020/01/07 08:04:57 (permalink)
Any other mesh overclockers? I can post benchmarks if someone shows off a game like Rainbow 6 siege or even cpu benchmarks. Running 1080 TI and 1.45v 4000Mhz ram CL 17-17-17-36
#8
jasoncodispoti
SSC Member
  • Total Posts : 628
  • Reward points : 0
  • Joined: 2013/03/18 14:40:28
  • Location: Nashville, TN
  • Status: offline
  • Ribbons : 14
Re: Concensus on safe daily mesh voltage 2020/01/07 20:17:32 (permalink)
cdc-951
Any other mesh overclockers? I can post benchmarks if someone shows off a game like Rainbow 6 siege or even cpu benchmarks. Running 1080 TI and 1.45v 4000Mhz ram CL 17-17-17-36




I have a 9800x, have not pushed mesh too far, but from the testing that I have done thus far I have not really seen any significant performance increase in synthetic benchmarks when going from a Mesh Ratio of 24 to a Mesh Ratio of 29. However given that I am pushing 29 at 1.000v I feel that I have plenty of head room to push it higher. 

Intel Core i7-9800x | EVGA X299 Dark| EVGA RTX 2080ti FTW3 Ultra HydroCopper | 32GB Corsair Dominator Platinum | EKWB Liquid Cooled Loop | Phanteks Enthoo Primo SE | EVGA SuperNOVA 1000 P2


#9
HiLuckyB
New Member
  • Total Posts : 56
  • Reward points : 0
  • Joined: 2010/10/16 18:43:06
  • Location: MI
  • Status: offline
  • Ribbons : 0
Re: Concensus on safe daily mesh voltage 2020/01/08 06:36:37 (permalink)
What work's best to test mesh stability? My 10980XE is at 4.6GHz 1.142v, 32 mesh 1.2v Haven't tried dropping it yet, And VCCIO at 1.1v. This is just a quick OC after getting my X299 Dark bios updated to use the 10980XE.

CPU - Silicon Lottery Binned i9-10980XE 4.6GHz 1.137v
MB - EVGA X299 Dark
GPU - EVGA RTX 2080 TI FTW3
PSU - EVGA SuperNOVA 1600 T2
#10
jasoncodispoti
SSC Member
  • Total Posts : 628
  • Reward points : 0
  • Joined: 2013/03/18 14:40:28
  • Location: Nashville, TN
  • Status: offline
  • Ribbons : 14
Re: Concensus on safe daily mesh voltage 2020/01/08 08:50:53 (permalink)
HiLuckyB
What work's best to test mesh stability? My 10980XE is at 4.6GHz 1.142v, 32 mesh 1.2v Haven't tried dropping it yet, And VCCIO at 1.1v. This is just a quick OC after getting my X299 Dark bios updated to use the 10980XE.



I would follow the normal process you typically use for testing general CPU stability testing. I personally prefer OCCT; using a large data set and 1 hour test each for AVX2, AVX512, and non-avx instructions (three hours of testing total).

Intel Core i7-9800x | EVGA X299 Dark| EVGA RTX 2080ti FTW3 Ultra HydroCopper | 32GB Corsair Dominator Platinum | EKWB Liquid Cooled Loop | Phanteks Enthoo Primo SE | EVGA SuperNOVA 1000 P2


#11
dmann304
iCX Member
  • Total Posts : 410
  • Reward points : 0
  • Joined: 2008/07/16 22:44:20
  • Status: offline
  • Ribbons : 0
Re: Concensus on safe daily mesh voltage 2021/01/14 10:41:17 (permalink)
Running the 10980xe, and it runs near 93c in R20, with VCCIO at 1.2v, which i heard is ok, and these cascade lake x chips run 
very hot, but in idle only 36c.  Is it normal though to see a 201f temp under a cinabench run?
#12
jasoncodispoti
SSC Member
  • Total Posts : 628
  • Reward points : 0
  • Joined: 2013/03/18 14:40:28
  • Location: Nashville, TN
  • Status: offline
  • Ribbons : 14
Re: Concensus on safe daily mesh voltage 2021/01/14 11:38:57 (permalink)
dmann304
Running the 10980xe, and it runs near 93c in R20, with VCCIO at 1.2v, which i heard is ok, and these cascade lake x chips run 
very hot, but in idle only 36c.  Is it normal though to see a 201f temp under a cinabench run?




The higher the core count the hotter it will be and the better your cooling solution will need to be. So is it normal? I dont know that I would say that its 'normal' however it is certainly not unexpected. Keep in mind that Max TJunction for that CPU is 86C and personally I would not want anything from the X299 lineup running hotter than 80C. IMO your temps are too high.. 

Intel Core i7-9800x | EVGA X299 Dark| EVGA RTX 2080ti FTW3 Ultra HydroCopper | 32GB Corsair Dominator Platinum | EKWB Liquid Cooled Loop | Phanteks Enthoo Primo SE | EVGA SuperNOVA 1000 P2


#13
Sultan.of.swing
Superclocked Member
  • Total Posts : 174
  • Reward points : 0
  • Joined: 2012/12/14 20:58:21
  • Status: offline
  • Ribbons : 2
Re: Concensus on safe daily mesh voltage 2021/01/14 19:56:19 (permalink)
Running 32 mesh on my 10940X with Mesh voltage set to 1.2v adaptive.
You have to pay attention to the "Package" temp in HWinfo64 to be able to see the effects of the mesh voltage, Once you start pumping in the mesh voltage the package temp will start getting hotter.



#14
Sultan.of.swing
Superclocked Member
  • Total Posts : 174
  • Reward points : 0
  • Joined: 2012/12/14 20:58:21
  • Status: offline
  • Ribbons : 2
Re: Concensus on safe daily mesh voltage 2021/01/14 19:57:47 (permalink)
dmann304
Running the 10980xe, and it runs near 93c in R20, with VCCIO at 1.2v, which i heard is ok, and these cascade lake x chips run 
very hot, but in idle only 36c.  Is it normal though to see a 201f temp under a cinabench run?


36c is a pretty high idle temp, on my custom loop with a 23-24c ambient my CPU idles at 26-27c.  Load temps in R23 at 4.8ghz and 1.25 for Vcore has the cores all in the low 70's.
#15
dmann304
iCX Member
  • Total Posts : 410
  • Reward points : 0
  • Joined: 2008/07/16 22:44:20
  • Status: offline
  • Ribbons : 0
Re: Concensus on safe daily mesh voltage 2021/01/14 21:09:23 (permalink)
I'm using the 360 i150 corsair AIO, and idle atm now is 33c, and setting a manual tjmax, i don't go above 88 under 
the classroom render in blender. games like: GTA, and WatchDogs, or even Red Dead 2, i get 50c.
 
This AIO enough for this thing? and for some strange reason, in AIda64, under stress test, it says in red thermal Throttling heat detected, 
when i haven't even ran a test, but when i do, it does around 77c.
#16
Sultan.of.swing
Superclocked Member
  • Total Posts : 174
  • Reward points : 0
  • Joined: 2012/12/14 20:58:21
  • Status: offline
  • Ribbons : 2
Re: Concensus on safe daily mesh voltage 2021/01/14 21:26:07 (permalink)
dmann304
I'm using the 360 i150 corsair AIO, and idle atm now is 33c, and setting a manual tjmax, i don't go above 88 under 
the classroom render in blender. games like: GTA, and WatchDogs, or even Red Dead 2, i get 50c.
 
This AIO enough for this thing? and for some strange reason, in AIda64, under stress test, it says in red thermal Throttling heat detected, 
when i haven't even ran a test, but when i do, it does around 77c.


You need to pay attention to the "Package" temp, not just the temps of each core.
#17
dmann304
iCX Member
  • Total Posts : 410
  • Reward points : 0
  • Joined: 2008/07/16 22:44:20
  • Status: offline
  • Ribbons : 0
Re: Concensus on safe daily mesh voltage 2021/01/14 23:12:01 (permalink)
Yea, package temp under a R20 run was 93c, and idle at 34c, on a Corsair 150I 360 AIO full fan speed.
#18
Sultan.of.swing
Superclocked Member
  • Total Posts : 174
  • Reward points : 0
  • Joined: 2012/12/14 20:58:21
  • Status: offline
  • Ribbons : 2
Re: Concensus on safe daily mesh voltage 2021/01/14 23:36:55 (permalink)
I would try and get voltages lower, If you can get VCCIO lower try that as well. I run mine on Auto with no issues.
From my experience with these High Core Count HEDT chips a full custom loop is really the only way to go.
Even then they are hard to cool, I do know the Heatkiller and the Optimus Blocks are really the main 2 that seem to do well on these chips.
#19
dmann304
iCX Member
  • Total Posts : 410
  • Reward points : 0
  • Joined: 2008/07/16 22:44:20
  • Status: offline
  • Ribbons : 0
Re: Concensus on safe daily mesh voltage 2021/01/15 07:31:28 (permalink)
I heard that an XMP profile, even with Corsair 3600 RGB Dominator Kits, can 
ramp up temps, is that true? , and i switched VCCIO back to Auto, from the 1.2v i had
before.
 
I was Watching Jay do a undervolt vid for 10980XE, even at 47Ghz, and even with a low vcore  of 
1.23v, his highest temps were 110c, and thats with much lower vcore, than i have, and a 1.8v VDIN< wow, and 
he has an excotic custom cooling. These are beasts. 
#20
Sultan.of.swing
Superclocked Member
  • Total Posts : 174
  • Reward points : 0
  • Joined: 2012/12/14 20:58:21
  • Status: offline
  • Ribbons : 2
Re: Concensus on safe daily mesh voltage 2021/01/15 07:40:52 (permalink)
dmann304
I heard that an XMP profile, even with Corsair 3600 RGB Dominator Kits, can 
ramp up temps, is that true? , and i switched VCCIO back to Auto, from the 1.2v i had
before.
 
I was Watching Jay do a undervolt vid for 10980XE, even at 47Ghz, and even with a low vcore  of 
1.23v, his highest temps were 110c, and thats with much lower vcore, than i have, and a 1.8v VDIN< wow, and 
he has an excotic custom cooling. These are beasts. 

If he had to run 1.23 vcore for 4.7 either he had a bad chip or he's just showing his mediocre overclocking skills.
All core 4.7 can usually be done with less than 1.2v on these unless you have a really poor chip.
#21
cdc-951
SSC Member
  • Total Posts : 520
  • Reward points : 0
  • Joined: 2012/04/27 02:26:30
  • Status: offline
  • Ribbons : 1
Re: Concensus on safe daily mesh voltage 2021/01/15 07:51:39 (permalink)
Not true ^ I have seen a 1.23V CPU on a 9800X for 4.7 Ghz to be stable and I am no novice. This build was for someone else 
#22
dmann304
iCX Member
  • Total Posts : 410
  • Reward points : 0
  • Joined: 2008/07/16 22:44:20
  • Status: offline
  • Ribbons : 0
Re: Concensus on safe daily mesh voltage 2021/01/15 08:25:40 (permalink)
Ok, i use HWinfo64 from CPUID, and from the Jay2cents vid, he showed, there more voltages:  A cpu vin=1.25v,
vdin 1.8v, and another voltage CPU Vcore at 1.315v. so there must be more then 1 cpu core voltage, and how
to adjust the 1.315 setting. When i run CPU at 1.25v at 4800mhz
 
And as for Jay, he's a big timer in this stuff, since he alignes himself with notables like Vince AKA Kingpin, and even
Debaur. World OCers.
#23
glocked89
Superclocked Member
  • Total Posts : 187
  • Reward points : 0
  • Joined: 2019/01/16 23:48:00
  • Status: offline
  • Ribbons : 0
Re: Concensus on safe daily mesh voltage 2021/01/15 23:44:58 (permalink)
dmann304
Ok, i use HWinfo64 from CPUID, and from the Jay2cents vid, he showed, there more voltages:  A cpu vin=1.25v,
vdin 1.8v, and another voltage CPU Vcore at 1.315v. so there must be more then 1 cpu core voltage, and how
to adjust the 1.315 setting. When i run CPU at 1.25v at 4800mhz
 
And as for Jay, he's a big timer in this stuff, since he alignes himself with notables like Vince AKA Kingpin, and even
Debaur. World OCers.


damn i made this thread over a year ago. but please, jay is a joke among enthusiasts. i would do some more research on your own. i suggest Luumi and buildzoid as good starts. theres also some good info in this thread: 
 
https://www.overclock.net...z-on-18-cores.1737048/
#24
dmann304
iCX Member
  • Total Posts : 410
  • Reward points : 0
  • Joined: 2008/07/16 22:44:20
  • Status: offline
  • Ribbons : 0
Re: Concensus on safe daily mesh voltage 2021/01/16 08:30:58 (permalink)
Plus he gets all kinds of free stuff too.  I do watch Buildzoid, he has alot of stuff on these boards.
 
 
#25
dmann304
iCX Member
  • Total Posts : 410
  • Reward points : 0
  • Joined: 2008/07/16 22:44:20
  • Status: offline
  • Ribbons : 0
Re: Concensus on safe daily mesh voltage 2021/01/16 08:46:26 (permalink)
Looks like the only way to have a 5ghz daily driver, you need more than a corsair 150I 360.
 Seems they give the voltage for the vcore, but what about the vdinn ? My previous CPU, i had to 
run it at 2.1v but 1.25v on cpu itself
 
#26
Sultan.of.swing
Superclocked Member
  • Total Posts : 174
  • Reward points : 0
  • Joined: 2012/12/14 20:58:21
  • Status: offline
  • Ribbons : 2
Re: Concensus on safe daily mesh voltage 2021/01/16 13:09:32 (permalink)
cdc-951
Not true ^ I have seen a 1.23V CPU on a 9800X for 4.7 Ghz to be stable and I am no novice. This build was for someone else 




Yes that is on a 9800x, the higher core count chips seem to be binned better and generally require less Vcore.
My 10940x @ 4.7 all core requires less than 1.2v for stability. 
#27
glocked89
Superclocked Member
  • Total Posts : 187
  • Reward points : 0
  • Joined: 2019/01/16 23:48:00
  • Status: offline
  • Ribbons : 0
Re: Concensus on safe daily mesh voltage 2021/01/16 13:21:27 (permalink)
dmann304
Looks like the only way to have a 5ghz daily driver, you need more than a corsair 150I 360.
 Seems they give the voltage for the vcore, but what about the vdinn ? My previous CPU, i had to 
run it at 2.1v but 1.25v on cpu itself
 



I have the h150i pro as well.
https://imgur.com/a/S53KHIJ
#28
dmann304
iCX Member
  • Total Posts : 410
  • Reward points : 0
  • Joined: 2008/07/16 22:44:20
  • Status: offline
  • Ribbons : 0
Re: Concensus on safe daily mesh voltage 2021/01/16 14:27:03 (permalink)
Yea, but is it good enough for these high core count CPU's? 
#29
arestavo
CLASSIFIED ULTRA Member
  • Total Posts : 6916
  • Reward points : 0
  • Joined: 2008/02/06 06:58:57
  • Location: Through the Scary Door
  • Status: offline
  • Ribbons : 76
Re: Concensus on safe daily mesh voltage 2021/01/16 14:32:00 (permalink)
Sultan.of.swing
cdc-951
Not true ^ I have seen a 1.23V CPU on a 9800X for 4.7 Ghz to be stable and I am no novice. This build was for someone else 




Yes that is on a 9800x, the higher core count chips seem to be binned better and generally require less Vcore.
My 10940x @ 4.7 all core requires less than 1.2v for stability. 


I thought as much for mine, as I was running at 1.18V at 4.7. But it would occasionally crash out of AVX games like Kingdom Come Deliverance with a -3 offset until I bumped it up to 1.2V.
post edited by arestavo - 2021/01/16 14:35:04
#30
Page: 123 > Showing page 1 of 3
Jump to:
  • Back to Mobile