EVGA

How relevant are the voltage measuring points on the SR-2?

Author
Post
Essentials Only Full Version
Bepzinky
iCX Member
  • Total Posts : 281
  • Reward points : 0
  • Joined: 2016/04/30 13:15:31
  • Status: offline
  • Ribbons : 2
2020/07/06 00:55:57 (permalink)
I was wondering...since the HWInfo64/Aida64 reported values are way off from what both my DMMs say when measuring at the test points, even when running SmallFFT's the voltage is still very close to what it's set to in the BIOS.
For example - it would take 1.36 and 1.34Vcore setting in BIOS for HWInfo to report both Vcore values as equal, roughly 1.352V. However, measuring at the test points on the MB show the Vcore is bang on what you set it up to in BIOS.
VTT seems to overshoot a lot, 1.275 setting giving 1.30+ at the TP on the DMM while HWInfo reports 1.269V. HWInfo only shows equal VTT for both CPU's if you give .025mV more to CPU1.
The DIMM voltage has the worst overshoot, 1.61 in the BIOS gives 1.64 and 1.66V(!) on the TP.
 
I can't get my head around on why folks state one CPU needs higher Vcore than the other (and which is confirmed by HWInfo vcore discrepancy) while the test points show Vcore being nearly equal. I know there are differences in silicon quality but still, it shouldn't be the rule that CPU0 would always need like .050 or .075 more juice.
post edited by Bepzinky - 2020/07/06 01:08:59
#1

3 Replies Related Threads
    Robbanswe
    iCX Member
    • Total Posts : 338
    • Reward points : 0
    • Joined: 2011/09/18 00:19:28
    • Status: offline
    • Ribbons : 4
    Re: How relevant are the voltage measuring points on the SR-2? 2020/07/07 00:10:48 (permalink)
    I have always set the same voltage for both cpu's.
    post edited by Robbanswe - 2020/07/07 00:17:11

    Attached Image(s)


    2x EVGA SR-2 with X5690, 48/96GB RAM.
    1x intel Platinum 48core system with 9GPUS, 192GB ram all cpus\gpus watercooled 
    2x AMDThreadripper builds with 48GB ram,nvme raid.
    Intel 12900k,5xNVME raid, 3080,ddr5 
     
     
     
     
     
     
     
    #2
    Bepzinky
    iCX Member
    • Total Posts : 281
    • Reward points : 0
    • Joined: 2016/04/30 13:15:31
    • Status: offline
    • Ribbons : 2
    Re: How relevant are the voltage measuring points on the SR-2? 2020/07/07 04:37:04 (permalink)
    What do you set in BIOS as Vcore for both? To get equal Vcore (1.352V) I have to set CPU0 to 1.375V and CPU1 to ~1.35V.
    The auto IOH voltage is also set darn high, 1.35V. Is it a 5520 thing as compared to X58?
    post edited by Bepzinky - 2020/07/07 15:06:38
    #3
    Robbanswe
    iCX Member
    • Total Posts : 338
    • Reward points : 0
    • Joined: 2011/09/18 00:19:28
    • Status: offline
    • Ribbons : 4
    Re: How relevant are the voltage measuring points on the SR-2? 2020/07/08 04:49:56 (permalink)
    equal Vcore ?
    Im not measuring volt at mobo, i test/OC until it works.
    Better to burn some 5690 than the mobo so dont overvoltage the IHO more than 1,3-1,35.
     
     My DAW SR-2: 
    Big watercoolingsystem
    Vcore 1,47500v
    VTT 1.325v
    Dimm 1,51v
    CPU PLL 1,65v
    IOH 1,30v
    HT OFF, so 12 real cores at 4800mhz, perfect for DAW and Gaming, not suitable at all for 3D rendering ( 3d rendering 4200-4400mhz HT core on )
     
    12+12HT cores + original CPU PLL voltage generating a lot of heat.
    HT OFF and 1,65v  PLL voltage lower the cpu heat and make SR-2 more OC friendly.
    HT cores is great in benchmarking,rendering and calculating 25% up, but in DAW its worse, i get better preformance with only real cores than those fake HT  
     
     
     
     
    post edited by Robbanswe - 2020/07/08 05:01:49

    2x EVGA SR-2 with X5690, 48/96GB RAM.
    1x intel Platinum 48core system with 9GPUS, 192GB ram all cpus\gpus watercooled 
    2x AMDThreadripper builds with 48GB ram,nvme raid.
    Intel 12900k,5xNVME raid, 3080,ddr5 
     
     
     
     
     
     
     
    #4
    Jump to:
  • Back to Mobile