Jump to content

Relearning me overclocking bits (9900K, Gigabyte AORUS Master Z390 board)

Hello guys!

I haven't messed with overclocking much in about a year now, I'm rewatching a lot of overclocking guides as a refresher, wondering if anyone has any suggestions on particular videos to watch or guides to read.

 

My goal is 5 or 5.1GHz, on a 9900K of course, setup is in my signature (Gigabyte Z390 AORUS Master).

 

Right now I'm testing 5GHz @ 1.32V, seems to be completely stable, it's passed every benchmark I've thrown at it so far, but I want to enable the C-state/downclock modes so it can idle at lower speeds, I tried doing that last night & it froze during a pretty standard benchmark, but that was at 1.31V.

 
Is there any settings in particular I need to make sure are ON to get the speedstep/C-state lowering speed while idle stuff to work properly that I might've missed?

Previously I got this chip to run 5GHz as low as 1.29V & pass (but I don't remember just how much actual stability testing I did on that) and I can't remember my power level/VRM settings or any of that stuff so I'm having to start from scratch.

 

So far temps are decent, I'm not even seeing 80C on any cores during non-AVX stress testing, think the highest I saw was 78C on the "hottest" of my cores.
This is with a Thermaltake Floe 280MM AIO with Push/Pull configuration fans at 70% fan speed (I've found the "flow" difference between 70 & 100% does not justify the increase in noise) on the top of my case.

 

Any suggestions would be greatly appreciated, as well as any tests I should run for stability checking.  Right now I'm using Intel Extreem Tuning Utility's stress testing, and 3DMark (like, all of them, Port Royal, Time Spy, FireStrike, Sky Diver, Night Raid, on all the different "difficulty" settings for Time Spy & FireStrike, as I've found sometimes OCs will pass Time Spy Extreme & FireStrike Ultra, but fail regular FireStrike or Time Spy because the CPU is more loaded on those).

 

I remember last time I tested extensively I also played about 3 hours of GTA V's single player with zero issues.  

 

I also plan on fiddling with RAM timings & speed (Buildzoid's video will get me there lul), but I'll be doing that last after I make sure 5GHz at least is stable with regular XMP settings being on.

This is all in preparation for getting a 3080 or 3090 sometime in the next month.

 

Right now I've tested 1.3V but not super extensively, and it seems to be stable with these voltage regulator settings, wondering if anyone has any suggestions on which ones I could possibly tweak:

 

200923112750.thumb.jpg.a727da6e4d092ec69073ea68783bfbb6.jpg

 

At those settings it passed 3 different stress tests including an AVX test, and all the different 3DMarks (FireStrike, TimeSpy, Night Raid, etc on all the different presets).

Save up to 10% on eVGA.com purchases, use my Associates Code: 47AQQ6KPU2IZNXY

My system on ModRigs: https://www.modsrigs.com/detail.aspx?BuildID=42686 

Primary Rig:

CPU: AMD 5950X at 4.65GHz 1.275V - Mobo: Asus Crosshair VIII Hero - PSU: eVGA P2 1200W

CPU Cooler: EK Quantum Velocity Block (480mm XE Radiator with push/pull EK Vardar D-RGB fans)

2 x 16GB G.Skill Trident Z Royal 3600MHz CL14 RAM (14-15-15-35-1T at 1.45V)

GPU: eVGA RTX 3090 Kingpin Hybrid - Core @ 2160MHz @ 1068mV, VRAM +1000MHz

Case: Thermaltake View 91 - SSDs/HDDs: Too many to list; Samsung 980 Pro 1TB & 2TB M.2s, Samsung 4TB SATA SSD

Monitor: Samsung Odyssey G9 49" Super-Ultrawide 240Hz Monitor

Link to comment
Share on other sites

Link to post
Share on other sites

dunno about all the fancy stuff you wanna do, if you can get the c states working with the overclock thatd be interesting. i prefer to use prime95 for stress testing but i like my stability, if you think thats overkill then thats fine. but i can never think "man maybe my cpu overclock is unstable" when i have problems, its almost completely out of the picture when im troubleshooting

topics i need help on:

Spoiler

 

 

my "oops i bought intel right before zen 3 releases" build

CPU: Ryzen 5 3600 (placeholder)

GPU: Gigabyte 980ti Xtreme (also placeholder), deshroud w/ generic 1200rpm 120mm fans x2, stock bios 130% power, no voltage offset: +70 core +400 mem 

Memory: 2x16gb GSkill Trident Z RGB 3600C16, 14-15-30-288@1.45v

Motherboard: Asus ROG Strix X570-E Gaming

Cooler: Noctua NH-D15S w/ white chromax bling
OS Drive: Samsung PM981 1tb (OEM 970 Evo)

Storage Drive: XPG SX8200 Pro 2tb

Backup Storage: Seagate Barracuda Compute 4TB

PSU: Seasonic Prime Ultra Titanium 750W w/ black/white Cablemod extensions
Case: Fractal Design Meshify C Dark (to be replaced with a good case shortly)

basically everything was bought used off of reddit or here, only new component was the case. absolutely nutty deals for some of these parts, ill have to tally it all up once it's "done" :D 

Link to comment
Share on other sites

Link to post
Share on other sites

Use the lowest LLC that you can. With the 9900K it will probably be a medium setting. I don't know what "extreme" means exactly since I haven't had a Gigabyte board in a while, but it sounds rather... extreme... 

 

You'll have to raise the voltage in BIOS to compensate, but you'll be able to get a lower stable vcore under load. Basically, VDROOP is good.

 

 

12 minutes ago, HoneyBadger84 said:

Intel Extreem Tuning Utility's

In my testing, I found that XTU was overwriting the settings that I set in BIOS, even when I didn't run the application. I was getting very strange behavior that went away after uninstalling it. I'd recommend that you uninstall it.

 

12 minutes ago, HoneyBadger84 said:

Is there any settings in particular I need to make sure are ON to get the speedstep/C-state lowering speed while idle stuff to work properly that I might've missed?

If you want your CPU to fully idle, just make sure that C States is on auto, speed step and speed shift are enabled, and use adaptive voltage rather than manual (just make sure it is behaving the way you want). This is the way that I run my 9900K.

BabyBlu (Primary): 

  • CPU: Intel Core i9 9900K @ up to 5.3GHz, 5.0GHz all-core, delidded
  • Motherboard: Asus Maximus XI Hero
  • RAM: G.Skill Trident Z RGB 4x8GB DDR4-3200 @ 4000MHz 16-18-18-34
  • GPU: MSI RTX 2080 Sea Hawk EK X, 2070MHz core, 8000MHz mem
  • Case: Phanteks Evolv X
  • Storage: XPG SX8200 Pro 2TB, 3x ADATASU800 1TB (RAID 0), Samsung 970 EVO Plus 500GB
  • PSU: Corsair HX1000i
  • Display: MSI MPG341CQR 34" 3440x1440 144Hz Freesync, Dell S2417DG 24" 2560x1440 165Hz Gsync
  • Cooling: Custom water loop (CPU & GPU), Radiators: 1x140mm(Back), 1x280mm(Top), 1x420mm(Front)
  • Keyboard: Corsair Strafe RGB (Cherry MX Brown)
  • Mouse: MasterMouse MM710
  • Headset: Corsair Void Pro RGB
  • OS: Windows 10 Pro

Roxanne (Wife Build):

  • CPU: Intel Core i7 4790K @ up to 5.0GHz, 4.8Ghz all-core, relidded w/ LM
  • Motherboard: Asus Z97A
  • RAM: G.Skill Sniper 4x8GB DDR3-2400 @ 10-12-12-24
  • GPU: EVGA GTX 1080 FTW2 w/ LM
  • Case: Corsair Vengeance C70, w/ Custom Side-Panel Window
  • Storage: Samsung 850 EVO 250GB, Samsung 860 EVO 1TB, Silicon Power A80 2TB NVME
  • PSU: Corsair AX760
  • Display: Samsung C27JG56 27" 2560x1440 144Hz Freesync
  • Cooling: Corsair H115i RGB
  • Keyboard: GMMK TKL(Kailh Box White)
  • Mouse: Glorious Model O-
  • Headset: SteelSeries Arctis 7
  • OS: Windows 10 Pro

BigBox (HTPC):

  • CPU: Ryzen 5800X3D
  • Motherboard: Gigabyte B550i Aorus Pro AX
  • RAM: Corsair Vengeance LPX 2x8GB DDR4-3600 @ 3600MHz 14-14-14-28
  • GPU: MSI RTX 3080 Ventus 3X Plus OC, de-shrouded, LM TIM, replaced mem therm pads
  • Case: Fractal Design Node 202
  • Storage: SP A80 1TB, WD Black SN770 2TB
  • PSU: Corsair SF600 Gold w/ NF-A9x14
  • Display: Samsung QN90A 65" (QLED, 4K, 120Hz, HDR, VRR)
  • Cooling: Thermalright AXP-100 Copper w/ NF-A12x15
  • Keyboard/Mouse: Rii i4
  • Controllers: 4X Xbox One & 2X N64 (with USB)
  • Sound: Denon AVR S760H with 5.1.2 Atmos setup.
  • OS: Windows 10 Pro

Harmonic (NAS/Game/Plex/Other Server):

  • CPU: Intel Core i7 6700
  • Motherboard: ASRock FATAL1TY H270M
  • RAM: 64GB DDR4-2133
  • GPU: Intel HD Graphics 530
  • Case: Fractal Design Define 7
  • HDD: 3X Seagate Exos X16 14TB in RAID 5
  • SSD: Inland Premium 512GB NVME, Sabrent 1TB NVME
  • Optical: BDXL WH14NS40 flashed to WH16NS60
  • PSU: Corsair CX450
  • Display: None
  • Cooling: Noctua NH-U14S
  • Keyboard/Mouse: None
  • OS: Windows 10 Pro

NAS:

  • Synology DS216J
  • 2x8TB WD Red NAS HDDs in RAID 1. 8TB usable space
Link to comment
Share on other sites

Link to post
Share on other sites

It really depends on the CPU, my i5 8600k gets an AVX2 stable overclock (5GHz) at 1.31v, but it's a really good chip.

Most 9900k's should get to 5.1GHz at under 1.4v.

For stress testing, if you really want to destroy that cpu use Prime95 small fft, no AVX (unless you really want to recreate a new star in the solar system).

If you want a quick and dirty way to do it, intel burn test seems pretty reliable.

Also, I'm not sure if that AIO can get your CPU to 5.1GHz, depends on the quality of the silicon

Main PC [The Rig of Theseus]:

CPU: i5-8600K @ 5.0 GHz | GPU: GTX 1660 | RAM: 16 GB DDR4 3000 MHz | Case: Lian Li PC-O11 Dynamic | PSU: Corsair RM 650i | SSD: Corsair MP510 480 GB |  HDD: 2x 6 TB WD Red| Motherboard: Gigabyte Z390 Aorus Pro | OS: Windows 11 Pro for Workstations

 

Secondary PC [Why did I bother]:

CPU: AMD Athlon 3000G | GPU: Vega 3 iGPU | RAM: 8 GB DDR4 3000 MHz | Case: Corsair 88R | PSU: Corsair VS 650 | SSD: WD Green M.2 SATA 120 GB | Motherboard: MSI A320M-A PRO MAX | OS: Windows 11 Pro for Workstations

 

Server [Solution in search of a problem]:

Model: HP DL360e Gen8 | CPU: 1x Xeon E5-2430L v1 | RAM: 12 GB DDR3 1066 MHz | SSD: Kingston A400 120 GB | OS: VMware ESXi 7

 

Server 2 electric boogaloo [A waste of electricity]:

Model: intel NUC NUC5CPYH | CPU: Celeron N3050 | RAM: 2GB DDR3L 1600 MHz | SSD: Kingston UV400 120 GB | OS: Debian Bullseye

 

Laptop:

Model: ThinkBook 14 Gen 2 AMD | CPU: Ryzen 7 4700U | RAM: 16 GB DDR4 3200 MHz | OS: Windows 11 Pro

 

Photography:

 

Cameras:

Full Frame digital: Sony α7

APS-C digital: Sony α100

Medium Format Film: Kodak Junior SIX-20

35mm Film:

 

Lenses:

Sony SAL-1870 18-70mm ƒ/3.5-5.6 

Sony SAL-75300 75-300mm ƒ/4.5-5.6

Meike MK-50mm ƒ/1.7

 

PSA: No, I didn't waste all that money on computers, (except the main one) my server cost $40, the intel NUC was my old PC (although then it had 8GB of ram, I gave the bigger stick of ram to a person who really needed it), my laptop is used and the second PC is really cheap.

I like tinkering with computers and have a personal hatred towards phones and everything they represent (I daily drive an iPhone 7, or a 6, depends on which one works that day)

Link to comment
Share on other sites

Link to post
Share on other sites

2 minutes ago, VeganJoy said:

dunno about all the fancy stuff you wanna do, if you can get the c states working with the overclock thatd be interesting. i prefer to use prime95 for stress testing but i like my stability, if you think thats overkill then thats fine. but i can never think "man maybe my cpu overclock is unstable" when i have problems, its almost completely out of the picture when im troubleshooting

I personally prefer intel burn test to prime 95.

prime 95 is a bit too harsh if you use small FFT AVX2 and requires a LOT of time if you want to be sure you have a completely stable oc

Main PC [The Rig of Theseus]:

CPU: i5-8600K @ 5.0 GHz | GPU: GTX 1660 | RAM: 16 GB DDR4 3000 MHz | Case: Lian Li PC-O11 Dynamic | PSU: Corsair RM 650i | SSD: Corsair MP510 480 GB |  HDD: 2x 6 TB WD Red| Motherboard: Gigabyte Z390 Aorus Pro | OS: Windows 11 Pro for Workstations

 

Secondary PC [Why did I bother]:

CPU: AMD Athlon 3000G | GPU: Vega 3 iGPU | RAM: 8 GB DDR4 3000 MHz | Case: Corsair 88R | PSU: Corsair VS 650 | SSD: WD Green M.2 SATA 120 GB | Motherboard: MSI A320M-A PRO MAX | OS: Windows 11 Pro for Workstations

 

Server [Solution in search of a problem]:

Model: HP DL360e Gen8 | CPU: 1x Xeon E5-2430L v1 | RAM: 12 GB DDR3 1066 MHz | SSD: Kingston A400 120 GB | OS: VMware ESXi 7

 

Server 2 electric boogaloo [A waste of electricity]:

Model: intel NUC NUC5CPYH | CPU: Celeron N3050 | RAM: 2GB DDR3L 1600 MHz | SSD: Kingston UV400 120 GB | OS: Debian Bullseye

 

Laptop:

Model: ThinkBook 14 Gen 2 AMD | CPU: Ryzen 7 4700U | RAM: 16 GB DDR4 3200 MHz | OS: Windows 11 Pro

 

Photography:

 

Cameras:

Full Frame digital: Sony α7

APS-C digital: Sony α100

Medium Format Film: Kodak Junior SIX-20

35mm Film:

 

Lenses:

Sony SAL-1870 18-70mm ƒ/3.5-5.6 

Sony SAL-75300 75-300mm ƒ/4.5-5.6

Meike MK-50mm ƒ/1.7

 

PSA: No, I didn't waste all that money on computers, (except the main one) my server cost $40, the intel NUC was my old PC (although then it had 8GB of ram, I gave the bigger stick of ram to a person who really needed it), my laptop is used and the second PC is really cheap.

I like tinkering with computers and have a personal hatred towards phones and everything they represent (I daily drive an iPhone 7, or a 6, depends on which one works that day)

Link to comment
Share on other sites

Link to post
Share on other sites

1 minute ago, mbntr said:

I personally prefer intel burn test to prime 95.

prime 95 is a bit too harsh if you use small FFT AVX2 and requires a LOT of time if you want to be sure you have a completely stable oc

i mean thats kinda the point, sure the average user probably wont have any issues with IBT or some other stress test but i wouldnt personally be comfortable with any alternative

topics i need help on:

Spoiler

 

 

my "oops i bought intel right before zen 3 releases" build

CPU: Ryzen 5 3600 (placeholder)

GPU: Gigabyte 980ti Xtreme (also placeholder), deshroud w/ generic 1200rpm 120mm fans x2, stock bios 130% power, no voltage offset: +70 core +400 mem 

Memory: 2x16gb GSkill Trident Z RGB 3600C16, 14-15-30-288@1.45v

Motherboard: Asus ROG Strix X570-E Gaming

Cooler: Noctua NH-D15S w/ white chromax bling
OS Drive: Samsung PM981 1tb (OEM 970 Evo)

Storage Drive: XPG SX8200 Pro 2tb

Backup Storage: Seagate Barracuda Compute 4TB

PSU: Seasonic Prime Ultra Titanium 750W w/ black/white Cablemod extensions
Case: Fractal Design Meshify C Dark (to be replaced with a good case shortly)

basically everything was bought used off of reddit or here, only new component was the case. absolutely nutty deals for some of these parts, ill have to tally it all up once it's "done" :D 

Link to comment
Share on other sites

Link to post
Share on other sites

I'm only using XTU for it's stress testing.  Everything is being done in BIOS of course.

 

Every time I turn the C-states on it seems to bug out the stability quite a bit, and I don't think my board has an "adaptive" voltage setting for the CPU.

 

This is what the Voltage area looks like: 

 

200923112842.thumb.jpg.b0c5b8e856da1a40696644c98fa9d9f4.jpg

 

... Oh duh, Dynamic = Adaptive doesn't it?

 

Any suggestions on where to start given that 1.3V seems to be "stable"?  I guess I'll start fiddling with that next time I'm doing testing.

 

I do plan on doing some Prime 95/similar testing for at least an hour once I really get locked in settings... I remember when 24hrs of Prime95 was "the thing" back in the day, don't think I want to have my CPU rad spewing the exhaust of ~72-80C core temps for a day straight, it's not THAT cold here at night yet LUL

Save up to 10% on eVGA.com purchases, use my Associates Code: 47AQQ6KPU2IZNXY

My system on ModRigs: https://www.modsrigs.com/detail.aspx?BuildID=42686 

Primary Rig:

CPU: AMD 5950X at 4.65GHz 1.275V - Mobo: Asus Crosshair VIII Hero - PSU: eVGA P2 1200W

CPU Cooler: EK Quantum Velocity Block (480mm XE Radiator with push/pull EK Vardar D-RGB fans)

2 x 16GB G.Skill Trident Z Royal 3600MHz CL14 RAM (14-15-15-35-1T at 1.45V)

GPU: eVGA RTX 3090 Kingpin Hybrid - Core @ 2160MHz @ 1068mV, VRAM +1000MHz

Case: Thermaltake View 91 - SSDs/HDDs: Too many to list; Samsung 980 Pro 1TB & 2TB M.2s, Samsung 4TB SATA SSD

Monitor: Samsung Odyssey G9 49" Super-Ultrawide 240Hz Monitor

Link to comment
Share on other sites

Link to post
Share on other sites

I don't know what Gigabyte changed in the background settings for the BIOS going from F8 to F11C, but it is crazy how much more difficult it is to get an OC stable on F11C vs F8... in F8 I can set literally like 5 settings different & 5GHz "just works"... in F11C I can do the exact same settings & I get blue screens, I spent over an hour fine tuning things, looks like I've gotten it decently stable again at 5GHz 1.3V on F11C, haven't done full stability testing yet but it's passed what I've done so far.

 

Room is kinda warm at the moment so I'm saving full stability testing for tonight when it's cool outside & I can leave the AC on in the computer's room so it isn't getting in to the mid 80s load temps.

 

Oh, also, I tried "Normal" voltage with an offset of .1V (to try and get 1.3V vs 1.2V that is "normal" at stock) to try & see what it would do for 5GHz... rebooted, and the CPU was set at 1.5V in BIOS on F11C... so yeah, don't do that. lol luckily I reset to default & rebooted quickly so it didn't get too warm or have that voltage running through it for very long.

 

I'll add my "currently testing" settings BIOS screenshots to this post or a later one this evening when I'm doing testing again, for now I'm back at "stock".

Save up to 10% on eVGA.com purchases, use my Associates Code: 47AQQ6KPU2IZNXY

My system on ModRigs: https://www.modsrigs.com/detail.aspx?BuildID=42686 

Primary Rig:

CPU: AMD 5950X at 4.65GHz 1.275V - Mobo: Asus Crosshair VIII Hero - PSU: eVGA P2 1200W

CPU Cooler: EK Quantum Velocity Block (480mm XE Radiator with push/pull EK Vardar D-RGB fans)

2 x 16GB G.Skill Trident Z Royal 3600MHz CL14 RAM (14-15-15-35-1T at 1.45V)

GPU: eVGA RTX 3090 Kingpin Hybrid - Core @ 2160MHz @ 1068mV, VRAM +1000MHz

Case: Thermaltake View 91 - SSDs/HDDs: Too many to list; Samsung 980 Pro 1TB & 2TB M.2s, Samsung 4TB SATA SSD

Monitor: Samsung Odyssey G9 49" Super-Ultrawide 240Hz Monitor

Link to comment
Share on other sites

Link to post
Share on other sites

I run my 9900K at 5GHz all cores without any other tweaks. Just used the Intel Tuning utility to increase the multiplier. No voltage tweaks, no fuss. Then disabled Multicore Enhancement in the BIOS and set Windows power to Balanced and speedstep works properly. Downclocks to 800MHz when idle.

Link to comment
Share on other sites

Link to post
Share on other sites

On 9/28/2020 at 3:36 PM, Internet Person said:

I run my 9900K at 5GHz all cores without any other tweaks. Just used the Intel Tuning utility to increase the multiplier. No voltage tweaks, no fuss. Then disabled Multicore Enhancement in the BIOS and set Windows power to Balanced and speedstep works properly. Downclocks to 800MHz when idle.

So you left voltage on Auto?  Out of curiousity what does it read out at in Windows (CPUz or HWInfo) when under load?  What board are you using?

Save up to 10% on eVGA.com purchases, use my Associates Code: 47AQQ6KPU2IZNXY

My system on ModRigs: https://www.modsrigs.com/detail.aspx?BuildID=42686 

Primary Rig:

CPU: AMD 5950X at 4.65GHz 1.275V - Mobo: Asus Crosshair VIII Hero - PSU: eVGA P2 1200W

CPU Cooler: EK Quantum Velocity Block (480mm XE Radiator with push/pull EK Vardar D-RGB fans)

2 x 16GB G.Skill Trident Z Royal 3600MHz CL14 RAM (14-15-15-35-1T at 1.45V)

GPU: eVGA RTX 3090 Kingpin Hybrid - Core @ 2160MHz @ 1068mV, VRAM +1000MHz

Case: Thermaltake View 91 - SSDs/HDDs: Too many to list; Samsung 980 Pro 1TB & 2TB M.2s, Samsung 4TB SATA SSD

Monitor: Samsung Odyssey G9 49" Super-Ultrawide 240Hz Monitor

Link to comment
Share on other sites

Link to post
Share on other sites

Yes, auto, I haven't touched the voltage settings in the BIOS. I did all my overclocking by moving the sliders to 5.00GHz in the Intel Extreme Tuning Utility:) My board is Gigabyte Z390 Aorus Pro WiFi.

 

So just to add to this. Speedstep wasn't working at first, was stuck at 5GHz. I did a little googling and someone said that Multicore Enhancement should be disabled in the BIOS and Windows Power Plan set to Balanced and that did the trick

 

BTW, the system crashes at 5.1 GHz, all cores. I haven't played with it any further. At 5.00GHz it's perfectly stable. I had it running like this for almost a year now. I've run any stress test I could find and it's 100% fine.

 

Idle:.

 

HWIndfo01.jpg.0eb34cf9b91415f0671f58573a9af508.jpg

 

Under load (Cinebench R20):

 

screenshot.1601567835.jpg.139900eee421b627a3c78ee86a5d269b.jpg

 

And these are the Intel Utility settings:

 

screenshot.1601569863.jpg.2f41dcf93b6b231b564d23318f89bc75.jpg

 

Link to comment
Share on other sites

Link to post
Share on other sites

Your temps are actually pretty similar to mine under 5GHz load, what cooler are you using out of curiosity?

 

I'm gonna fiddle with Adaptive voltage again, I'm just scurd after that "booting up at 1.53V" incident I had when I left it on auto and set to 5GHz last time.

Save up to 10% on eVGA.com purchases, use my Associates Code: 47AQQ6KPU2IZNXY

My system on ModRigs: https://www.modsrigs.com/detail.aspx?BuildID=42686 

Primary Rig:

CPU: AMD 5950X at 4.65GHz 1.275V - Mobo: Asus Crosshair VIII Hero - PSU: eVGA P2 1200W

CPU Cooler: EK Quantum Velocity Block (480mm XE Radiator with push/pull EK Vardar D-RGB fans)

2 x 16GB G.Skill Trident Z Royal 3600MHz CL14 RAM (14-15-15-35-1T at 1.45V)

GPU: eVGA RTX 3090 Kingpin Hybrid - Core @ 2160MHz @ 1068mV, VRAM +1000MHz

Case: Thermaltake View 91 - SSDs/HDDs: Too many to list; Samsung 980 Pro 1TB & 2TB M.2s, Samsung 4TB SATA SSD

Monitor: Samsung Odyssey G9 49" Super-Ultrawide 240Hz Monitor

Link to comment
Share on other sites

Link to post
Share on other sites

On 10/1/2020 at 9:10 AM, Internet Person said:

Yes, auto, I haven't touched the voltage settings in the BIOS. I did all my overclocking by moving the sliders to 5.00GHz in the Intel Extreme Tuning Utility:) My board is Gigabyte Z390 Aorus Pro WiFi.

 

So just to add to this. Speedstep wasn't working at first, was stuck at 5GHz. I did a little googling and someone said that Multicore Enhancement should be disabled in the BIOS and Windows Power Plan set to Balanced and that did the trick

 

BTW, the system crashes at 5.1 GHz, all cores. I haven't played with it any further. At 5.00GHz it's perfectly stable. I had it running like this for almost a year now. I've run any stress test I could find and it's 100% fine.

 

For some reason my XTU looks a bit different than yours, but I got basically the same settings as you set... the core voltage is going a bit higher than I'd prefer with it set that like that, but temps are in check (varying from 60s to low 70s under load) so far and it seems stable.

I'll throw some benchmarks at it & report back after the Stress Tests pass on XTU.

 

Settings screen I have looks like dis:

 

5GHzTestingXTU.thumb.jpg.32b5ca2b874e4ffa673c9e435c642bc1.jpg

Save up to 10% on eVGA.com purchases, use my Associates Code: 47AQQ6KPU2IZNXY

My system on ModRigs: https://www.modsrigs.com/detail.aspx?BuildID=42686 

Primary Rig:

CPU: AMD 5950X at 4.65GHz 1.275V - Mobo: Asus Crosshair VIII Hero - PSU: eVGA P2 1200W

CPU Cooler: EK Quantum Velocity Block (480mm XE Radiator with push/pull EK Vardar D-RGB fans)

2 x 16GB G.Skill Trident Z Royal 3600MHz CL14 RAM (14-15-15-35-1T at 1.45V)

GPU: eVGA RTX 3090 Kingpin Hybrid - Core @ 2160MHz @ 1068mV, VRAM +1000MHz

Case: Thermaltake View 91 - SSDs/HDDs: Too many to list; Samsung 980 Pro 1TB & 2TB M.2s, Samsung 4TB SATA SSD

Monitor: Samsung Odyssey G9 49" Super-Ultrawide 240Hz Monitor

Link to comment
Share on other sites

Link to post
Share on other sites

😄 I went in to BIOS to try & adjust the VRM/voltage controls to really low settings in attempt to lower the voltage the CPU was getting, bricked Windows until I reset to defaults & went back to my normal stock/stable settings & disabled XTU from loading it's service on startup.

I'm retesting again with a -.04V offset so it's getting up to 1.34V load instead of 1.38V... so far so good, that's about where 5GHz without speedstep enabled tested stable, the load voltage was peaking around 1.34V reading out in CPUz / XTU.

Load temps are about the same, unfortunately, slightly lower at best, but if it's stable, I'm cool with that.

 

Edit: It passed XTU stress testing... and crashed as soon as I tried to run Port Royal.

I even tried going back to no voltage offset, BSOD... dunno what happened but RIP that working out, oh well... I may try reinstalling it and starting over again tomorrow, right now I'm tired of messing with it for the evening.

I suppose it was working before I started messing with it too much, but that voltage was too high for my liking.

Save up to 10% on eVGA.com purchases, use my Associates Code: 47AQQ6KPU2IZNXY

My system on ModRigs: https://www.modsrigs.com/detail.aspx?BuildID=42686 

Primary Rig:

CPU: AMD 5950X at 4.65GHz 1.275V - Mobo: Asus Crosshair VIII Hero - PSU: eVGA P2 1200W

CPU Cooler: EK Quantum Velocity Block (480mm XE Radiator with push/pull EK Vardar D-RGB fans)

2 x 16GB G.Skill Trident Z Royal 3600MHz CL14 RAM (14-15-15-35-1T at 1.45V)

GPU: eVGA RTX 3090 Kingpin Hybrid - Core @ 2160MHz @ 1068mV, VRAM +1000MHz

Case: Thermaltake View 91 - SSDs/HDDs: Too many to list; Samsung 980 Pro 1TB & 2TB M.2s, Samsung 4TB SATA SSD

Monitor: Samsung Odyssey G9 49" Super-Ultrawide 240Hz Monitor

Link to comment
Share on other sites

Link to post
Share on other sites

11 hours ago, HoneyBadger84 said:

Your temps are actually pretty similar to mine under 5GHz load, what cooler are you using out of curiosity?

 

I'm gonna fiddle with Adaptive voltage again, I'm just scurd after that "booting up at 1.53V" incident I had when I left it on auto and set to 5GHz last time.

It's Corsair H100I RGB PRO XT. It's a 240mm cooler.

Link to comment
Share on other sites

Link to post
Share on other sites

10 hours ago, HoneyBadger84 said:

😄 I went in to BIOS to try & adjust the VRM/voltage controls to really low settings in attempt to lower the voltage the CPU was getting, bricked Windows until I reset to defaults & went back to my normal stock/stable settings & disabled XTU from loading it's service on startup.

I'm retesting again with a -.04V offset so it's getting up to 1.34V load instead of 1.38V... so far so good, that's about where 5GHz without speedstep enabled tested stable, the load voltage was peaking around 1.34V reading out in CPUz / XTU.

Load temps are about the same, unfortunately, slightly lower at best, but if it's stable, I'm cool with that.

 

Edit: It passed XTU stress testing... and crashed as soon as I tried to run Port Royal.

I even tried going back to no voltage offset, BSOD... dunno what happened but RIP that working out, oh well... I may try reinstalling it and starting over again tomorrow, right now I'm tired of messing with it for the evening.

I suppose it was working before I started messing with it too much, but that voltage was too high for my liking.

I find voltage tweaking intimidating and scary :) Good luck!

 

Link to comment
Share on other sites

Link to post
Share on other sites

Finally got around to doing some additional testing on BIOS settings & grabbing screenshots.

 

This tested stable in all the 3DMark tests (Time Spy (Ex & regular), Port Royal, Firestrike (Ultra, ex & regular) & Night Raid), as well as AIDA64 loading CPU FPU & RAM all at the same time.  Temps got pretty warm during that since it's an AVX workload, but still never saw 90C on any core so I'm pretty happy with that.  Voltage readout in CPUz/HWInfo maxes out at 1.32V under load (usually reads 1.298-1.304) so fiddling with the VRM settings definitely helped lower the load voltage which should help with load temps, I'll have to test non-AVX workload temps at some point since for me they're typically around 8-10C lower.

 

BIOS screens:

 

MainTweakerScreen.thumb.jpg.bedd384b1db476411a996a99208f471d.jpg

Advanced CPU Screen part 1:

AdvancedCPUScreen.thumb.jpg.a9486dde2e4aee46d709bcfa01d3defb.jpg

Advanced CPU Screen part 2:

AdvancedCPUScreen2.thumb.jpg.7d7efcb67545648f2eaa6919ffc3dc2b.jpg

Advanced Voltage Screen:

AdvancedVoltageScreen.thumb.jpg.2713f02541a4998268a7fd89f2175aaa.jpg

 

Also pretty happy with only having to have the CPU Load Line at Performance/Turbo instead of Turbo/Extreme like I did before.  I plan on testing at 1.300V next to see if it still passes with the same VReg settings.

Only thing that's changed on the RAM from default XMP settings is tRFC is set manually to 450 (cuz at default it sets it to 731 which is crazy loose), and the last smol timings at the very bottom are 60-60-120-120-40-40.

 

Broke some personal best records due to the GPU change in some of the 3DMarks, it's interesting to see the 2070 Super beat the 1080 Ti in some tests but not others.

Save up to 10% on eVGA.com purchases, use my Associates Code: 47AQQ6KPU2IZNXY

My system on ModRigs: https://www.modsrigs.com/detail.aspx?BuildID=42686 

Primary Rig:

CPU: AMD 5950X at 4.65GHz 1.275V - Mobo: Asus Crosshair VIII Hero - PSU: eVGA P2 1200W

CPU Cooler: EK Quantum Velocity Block (480mm XE Radiator with push/pull EK Vardar D-RGB fans)

2 x 16GB G.Skill Trident Z Royal 3600MHz CL14 RAM (14-15-15-35-1T at 1.45V)

GPU: eVGA RTX 3090 Kingpin Hybrid - Core @ 2160MHz @ 1068mV, VRAM +1000MHz

Case: Thermaltake View 91 - SSDs/HDDs: Too many to list; Samsung 980 Pro 1TB & 2TB M.2s, Samsung 4TB SATA SSD

Monitor: Samsung Odyssey G9 49" Super-Ultrawide 240Hz Monitor

Link to comment
Share on other sites

Link to post
Share on other sites

Anyone have any experience with whether or not hyperthreading in these newer processors has any effect on temperatures?  Like if I disable it I may be able to push more speed or lower temps?  I know a fair amount of games actually perform better with HT disabled, in particular CS:GO which I play a fair bit.

Since my 3080 FTW3 Ultra will be here Tuesday, I figure I'll spend the next 3 days fiddling with overclocking as much as I have time for & get things as tight & clocked up as I can before that thing arrives.

Save up to 10% on eVGA.com purchases, use my Associates Code: 47AQQ6KPU2IZNXY

My system on ModRigs: https://www.modsrigs.com/detail.aspx?BuildID=42686 

Primary Rig:

CPU: AMD 5950X at 4.65GHz 1.275V - Mobo: Asus Crosshair VIII Hero - PSU: eVGA P2 1200W

CPU Cooler: EK Quantum Velocity Block (480mm XE Radiator with push/pull EK Vardar D-RGB fans)

2 x 16GB G.Skill Trident Z Royal 3600MHz CL14 RAM (14-15-15-35-1T at 1.45V)

GPU: eVGA RTX 3090 Kingpin Hybrid - Core @ 2160MHz @ 1068mV, VRAM +1000MHz

Case: Thermaltake View 91 - SSDs/HDDs: Too many to list; Samsung 980 Pro 1TB & 2TB M.2s, Samsung 4TB SATA SSD

Monitor: Samsung Odyssey G9 49" Super-Ultrawide 240Hz Monitor

Link to comment
Share on other sites

Link to post
Share on other sites

29 minutes ago, HoneyBadger84 said:

Anyone have any experience with whether or not hyperthreading in these newer processors has any effect on temperatures?  Like if I disable it I may be able to push more speed or lower temps?  I know a fair amount of games actually perform better with HT disabled, in particular CS:GO which I play a fair bit.

Since my 3080 FTW3 Ultra will be here Tuesday, I figure I'll spend the next 3 days fiddling with overclocking as much as I have time for & get things as tight & clocked up as I can before that thing arrives.

 

I recommend leaving some headroom so your rig is stable rather than pushing to the extreme limit. 95-97% to the limit vs 99% is going to make minor difference that's only noticeable in benchmarks. Stay in the efficient range of your CPU temp wise rather than going max to the limit. 

 

When I get my 3080, I'm going to undervolt it to lower power usage. 

Link to comment
Share on other sites

Link to post
Share on other sites

1 hour ago, bomerr said:

 

I recommend leaving some headroom so your rig is stable rather than pushing to the extreme limit. 95-97% to the limit vs 99% is going to make minor difference that's only noticeable in benchmarks. Stay in the efficient range of your CPU temp wise rather than going max to the limit. 

 

When I get my 3080, I'm going to undervolt it to lower power usage. 

I was mostly refering to trying to tighten up my RAM again, I don't think I'll be pushing past 5GHz/1.31 or 1.3V since that's very good voltage for 5GHz compared to most people I've found that don't have high-bin 9900K chips.

 

I'd also probably only be running those overclocks for benchmarks, and possibly games that maybe need the extra performance to really push the 3080, since I game at 3440 by 1440, CPU is less of an issue in the majority of games.

 

As far as power draw goes, definitely not at all concerned about that given what my PSU is, and the board can definitely handle a lot as it's one of the best reviewed VRM setups, both in cooling & power delivery/performance.

 

Hoping I can find some decent information on the typical range of voltage people achieve 5.1 or 5.2GHz with, just to see if I can actually get those speeds "stable" tested, since I've never gotten 5.1 to pass a stability test yet, but I also never went past 1.35V because I'm a chicken... apparently anything under 1.4V is "safe" for 24/7 usage, from what I've seen on reddit, these forums & others.

 

Wondering if that extra 100-200MHz would be worth it in CPU-dependent games.

Save up to 10% on eVGA.com purchases, use my Associates Code: 47AQQ6KPU2IZNXY

My system on ModRigs: https://www.modsrigs.com/detail.aspx?BuildID=42686 

Primary Rig:

CPU: AMD 5950X at 4.65GHz 1.275V - Mobo: Asus Crosshair VIII Hero - PSU: eVGA P2 1200W

CPU Cooler: EK Quantum Velocity Block (480mm XE Radiator with push/pull EK Vardar D-RGB fans)

2 x 16GB G.Skill Trident Z Royal 3600MHz CL14 RAM (14-15-15-35-1T at 1.45V)

GPU: eVGA RTX 3090 Kingpin Hybrid - Core @ 2160MHz @ 1068mV, VRAM +1000MHz

Case: Thermaltake View 91 - SSDs/HDDs: Too many to list; Samsung 980 Pro 1TB & 2TB M.2s, Samsung 4TB SATA SSD

Monitor: Samsung Odyssey G9 49" Super-Ultrawide 240Hz Monitor

Link to comment
Share on other sites

Link to post
Share on other sites

15 minutes ago, HoneyBadger84 said:

I was mostly refering to trying to tighten up my RAM again, I don't think I'll be pushing past 5GHz/1.31 or 1.3V since that's very good voltage for 5GHz compared to most people I've found that don't have high-bin 9900K chips.

 

I'd also probably only be running those overclocks for benchmarks, and possibly games that maybe need the extra performance to really push the 3080, since I game at 3440 by 1440, CPU is less of an issue in the majority of games.

 

As far as power draw goes, definitely not at all concerned about that given what my PSU is, and the board can definitely handle a lot as it's one of the best reviewed VRM setups, both in cooling & power delivery/performance.

 

Hoping I can find some decent information on the typical range of voltage people achieve 5.1 or 5.2GHz with, just to see if I can actually get those speeds "stable" tested, since I've never gotten 5.1 to pass a stability test yet, but I also never went past 1.35V because I'm a chicken... apparently anything under 1.4V is "safe" for 24/7 usage, from what I've seen on reddit, these forums & others.

 

Wondering if that extra 100-200MHz would be worth it in CPU-dependent games.

the issue with power is more so that it means your room will get hot rather than psu or vrm load. 

Link to comment
Share on other sites

Link to post
Share on other sites

11 minutes ago, bomerr said:

the issue with power is more so that it means your room will get hot rather than psu or vrm load. 

Indeed, this room has a dedicated one-room AC unit though, and it's also full on winter at night here already (temps are already getting below freezing at night), so I'm not too-too worried about that, though of course that dynamic is going to change when I get the 3080 in the system because, unlikely my current card (and the 1080 Ti Hybrid I had in for 2 years before it), the 3080 FTW3 Ultra is air-cooled, whereas both the 2070 Super I'm using now & the 1080 Ti were Hybrid cooled, so less heat created overall.

 

I'm looking forward to comparing my temps with other people's once again on an air-cooled card, I typically have lower temps than just about anyone on air... also will be interesting to see how the AIB cooler on the 3080 FTW3 Ultra effects my M.2 temps, since right now they sit in the high 20s/low 30sC unless they're under heavy load, thanks to all the GPU's "heat" going out the back of the card or up and out of the radiator.

I sort of posed that question to people that already have the 3080 from eVGA, but most of them went from air cooled card to air cooled card, so their temps "didn't change", which is a good thing from the perspective of the cooler doing it's job well I suppose.

 

That's why I have side-flow over the M.2/PCH/GPU area though, I'm going to test 1 or 2 fans in that area after the 3080 arrives & see which one performs better in terms of GPU & M.2/PCH temps:

 

20201009-232512smol.jpg

Save up to 10% on eVGA.com purchases, use my Associates Code: 47AQQ6KPU2IZNXY

My system on ModRigs: https://www.modsrigs.com/detail.aspx?BuildID=42686 

Primary Rig:

CPU: AMD 5950X at 4.65GHz 1.275V - Mobo: Asus Crosshair VIII Hero - PSU: eVGA P2 1200W

CPU Cooler: EK Quantum Velocity Block (480mm XE Radiator with push/pull EK Vardar D-RGB fans)

2 x 16GB G.Skill Trident Z Royal 3600MHz CL14 RAM (14-15-15-35-1T at 1.45V)

GPU: eVGA RTX 3090 Kingpin Hybrid - Core @ 2160MHz @ 1068mV, VRAM +1000MHz

Case: Thermaltake View 91 - SSDs/HDDs: Too many to list; Samsung 980 Pro 1TB & 2TB M.2s, Samsung 4TB SATA SSD

Monitor: Samsung Odyssey G9 49" Super-Ultrawide 240Hz Monitor

Link to comment
Share on other sites

Link to post
Share on other sites

Went a few chips lower on the voltage & it's passed everything I've thrown at it so far with the same Voltage Regulation settings, which is nice.  Temps were a bit "toasty" because of my ambient being a little bit warmer than normal.

 

AIDA64 CPU + Memory passed for an hour with the following results:

 

StabilityTest1-CPUMemoryOnly-MaxTempsetc.thumb.jpg.566383025fd4846a395fc816de8c38cf.jpg

 

CPU + FPU passed 10 minutes (I will run it for longer eventually when my ambient is cooler, you'll see why when you see the temps lol):

 

StabilityTest1-CPUFPU-MaxTempsetc.thumb.jpg.c355a99a551feb0cd508745eca2a9171.jpg

 

Also passed a whole suite of 3DMark testing (Time Spy, Port Royal, Firestrike, etc) but I'm not bothering posting those scores cuz they're the same as before...

 

BIOS Screens for posterity so I have something to refer back to if I lose my settings during the push for 5.1 or 5.2GHz:

 

5GHz-1p3V-PrimaryTweaksPage201010232110.thumb.jpg.9cbea9a8d4aabd1df544cdd6b1269ed4.jpg

 

Advanced CPU part 1:

 

5GHz-1p3V-AdvancedCPUsettings-1-201010232121.thumb.jpg.3b5aa589c2b2080389597d2149409011.jpg

 

Advanced CPU part 2:

 

5GHz-1p3V-AdvancedCPUsettings-2-201010232133.thumb.jpg.0b5da0d7fddcc502ffdfcbea6297482f.jpg

 

Memory Settings:

 

5GHz1p3V-MemoryTimings-1-201010232150.thumb.jpg.ebd5d9225b4e14dc1358b78fbfb29e55.jpg

5GHz1p3V-MemoryTimings-2-201010232215.thumb.jpg.2a09f6eb6b4c812d212dc5f42fa21f0f.jpg

 

Advanced Voltage Settings:

 

5GHz1p3V-AdvancedVoltageSettings201010232231.thumb.jpg.922a7f0ea00a9de93de22cb87ec9ed34.jpg

 

And now the real fun begins, pushing for never-before-stable numbers...

Save up to 10% on eVGA.com purchases, use my Associates Code: 47AQQ6KPU2IZNXY

My system on ModRigs: https://www.modsrigs.com/detail.aspx?BuildID=42686 

Primary Rig:

CPU: AMD 5950X at 4.65GHz 1.275V - Mobo: Asus Crosshair VIII Hero - PSU: eVGA P2 1200W

CPU Cooler: EK Quantum Velocity Block (480mm XE Radiator with push/pull EK Vardar D-RGB fans)

2 x 16GB G.Skill Trident Z Royal 3600MHz CL14 RAM (14-15-15-35-1T at 1.45V)

GPU: eVGA RTX 3090 Kingpin Hybrid - Core @ 2160MHz @ 1068mV, VRAM +1000MHz

Case: Thermaltake View 91 - SSDs/HDDs: Too many to list; Samsung 980 Pro 1TB & 2TB M.2s, Samsung 4TB SATA SSD

Monitor: Samsung Odyssey G9 49" Super-Ultrawide 240Hz Monitor

Link to comment
Share on other sites

Link to post
Share on other sites

Iiiiiiiit's heeeeeeere!

 

Note the CPU cooler is all janky tubing wise because it's going to be replaced sometime in the next month with a 360mm unit.

 

IMG-20201013-174731-218smol.jpg

 

20201013-190313smol.jpg

 

20201013-212400smol.jpg

Save up to 10% on eVGA.com purchases, use my Associates Code: 47AQQ6KPU2IZNXY

My system on ModRigs: https://www.modsrigs.com/detail.aspx?BuildID=42686 

Primary Rig:

CPU: AMD 5950X at 4.65GHz 1.275V - Mobo: Asus Crosshair VIII Hero - PSU: eVGA P2 1200W

CPU Cooler: EK Quantum Velocity Block (480mm XE Radiator with push/pull EK Vardar D-RGB fans)

2 x 16GB G.Skill Trident Z Royal 3600MHz CL14 RAM (14-15-15-35-1T at 1.45V)

GPU: eVGA RTX 3090 Kingpin Hybrid - Core @ 2160MHz @ 1068mV, VRAM +1000MHz

Case: Thermaltake View 91 - SSDs/HDDs: Too many to list; Samsung 980 Pro 1TB & 2TB M.2s, Samsung 4TB SATA SSD

Monitor: Samsung Odyssey G9 49" Super-Ultrawide 240Hz Monitor

Link to comment
Share on other sites

Link to post
Share on other sites

  • 5 months later...

If I may ask, what BIOS are you on? Because I have been trying for over a week to get a stable low voltage low heat oc to 5ghz and 99% of screen shots and or videos look nothing like what I see on my end, also is their a benefit or hindrance to rolling back to an older BIOS?

 

PS new here and sorry for reviving an old thread but I have read this like 100x's an it's the most comprehensive of any I have read.

Link to comment
Share on other sites

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×