• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

ASUS Z490 Performance Oddities - Need your thoughts...

SonicMojo

New Member
Joined
Mar 15, 2021
Messages
6 (0.04/day)
Location
Canada
All

During the first half of 2021 - it was time to upgrade a pair of my key workstations with new parts. PC1 is my recording digital audio workstation and PC2 is my personal "work" machine.

I outfitted both with an ASUS z490 board AND the identical CPU AND the identical NMVe system drive - but the performance of one machine vs the other is quite noticeable. The key takeaways are below:

PC1 = ASUS ProArt Z490 Creator with i5 10500K
PC2 = ASUS Prime Z490-P with i5 10500K

Each PC has the same system drive (Samsung 970 EVO Plus NVMe)

PC1 is just slightly overclocked (9%) (CPU Clock 4500 MHz (original: 4100 MHz, overclock: 9%)
PC1 has Hyperthreading = OFF
PC1 has 32 GB of RAM
PC1 has system drive housed within an ASUS Hyper 16 4x NVMe card
PC1 uses on board (UHD 630 Graphics) to a circa 2009 Samsung 40 inch TV (Via HDMI)
PC1 System Drive Temp = Samsung SSD 970 EVO Plus 500GB 34 °C / 31 °C

PC2 is more than slightly overclocked (CPU Clock 4798.8 MHz (original: 4100 MHz, overclock: 17%)
PC2 has Hyperthreading = ON
PC2 has 16 GB of RAM
PC2 has system drive housed within M2_1 slot (on motherboard)
PC1 uses on board (UHD 630 Graphics) to a 2016 era Dell U2415 (Via HDMI)
PC2 System Drive Temp = Samsung SSD 970 EVO Plus 500GB 40 °C / 46 °C

Given the above - you would think they would be close in performance but PC2 feels noticeably slower than PC1 for even the most standard stuff from startup to actual program execution.

PC1 has a "snap" to it that makes everything seem almost instant/immediate whereas PC2 sometimes feels like the circa 2015 ASUS z97 board that I just replaced.

Samsung Magician performance tests on each system drive are almost identical.

I even placed PC2's system drive into an NVMe PCIx4 controller (StartTech) to see if that would make any difference and the drive still felt the same.

Can anyone tell me where I need to look to determine why PC2 feels so pokey?

Power Supply?
Hyperthreading?
32GB vs 16GB of RAM?
AntiVirus? PC2 has ESET while PC1 uses Windows Defender.
Or is the ASUS z490-P just a forever slower stepchild of the ProArt-Creator?

Curious to hear what you think.

Cheers

Sonic.
 
Joined
Feb 6, 2021
Messages
614 (3.57/day)
Location
Germany
Processor AMD Ryzen 9 5950X (and a LOT of other CPUs)
Motherboard MSI B550 Tomahawk
Cooling Corsair H150i RGB Pro XT
Memory G.Skill Trident Z Neo 3600 Mhz CL16 (2x16GB)
Video Card(s) a lot... 6900XTs, 3090s,80s,70s and 60s, almost all Turing and Pascal Cards + a few older ones.
Storage Samsung 970 Evo Plus 500GB, 2x Samsung 870 QVO 4TB
Display(s) Samsung Odyssey G7 32"
Case Fractal Design Meshify S2 White TG / Noctua NF-A14 iPPC-3000RPM Fans
Audio Device(s) Bose Companion Series 2 III, Sennheiser GSP600 running through a SoundblasterX G6
Power Supply bequiet! Dark Power Pro 12 1200w Titanium
Mouse Glorious PC Gaming Race Model O, G903 Hero, G502 Hero SE, Deathadder V2, EC2-A
Keyboard ASUS ROG Strix Scope TKL DLX, Corsair K95, Logitech G815
VR HMD Oculus Rift S
"AntiVirus? PC2 has ESET while PC1 uses Windows Defender."

Probably this.
 
Joined
Jan 13, 2021
Messages
46 (0.23/day)
Processor AMD 5800X
Motherboard MSI MPG B550 GAMING EDGE WIFI
Cooling Water Cooler SilverStone PF360-ARGB
Memory HyperX Fury, 32GB (4x8GB), 3600MHz, DDR4, CL17, HX436C17FB3K4/32
Video Card(s) RX 6800 XT Sapphire Nitro+ OC
Display(s) Acer Predator X34 34"
Power Supply Sharkoon 850W, 80 Plus Gold, SilentStorm CoolZero
Software Windows 10 x64 Build 19042.804 (20H2)
Probably the anti-virus, but I would also benchmark the CPUs.
 

SonicMojo

New Member
Joined
Mar 15, 2021
Messages
6 (0.04/day)
Location
Canada
Thanks for the updates!

Here are some basic benches done with UserBenchMark...and yes - I know I can get much deeper on this

PC1 - my studio machine

2021-06-16_12-52-13.png



And my work machine (this machine)

2021-06-16_12-47-13.png




All the benchmarks for the "work" CPU indicate biggest, better and faster - but it does not feel that way when I am in front of it (compared to the other machine - which flies).

Sonic.
 
Joined
May 8, 2021
Messages
558 (6.89/day)
System Name R&D Machine #1
Processor Intel Core i5 10400F
Motherboard Gigabyte B460M Aorus Pro
Cooling Scythe Choten
Memory 2x8GB G.Skill Aegis 2666 MHz
Video Card(s) Powercolor Red Dragon V2 RX 580 8GB 107 watt
Storage 512GB WD Blue + 256GB WD Green
Display(s) BenQ BL2420PT
Case Cooler Master Silencio S400
Audio Device(s) Topping D10
Power Supply Chieftec A90 550W (GDP-550C)
Mouse Steel Series Rival 100
Keyboard Steel Series 6GV2
Software Windows 10 Enterprise
Could be PL1 and PL2, could be different C state support, could be VRM temperature.
 

SonicMojo

New Member
Joined
Mar 15, 2021
Messages
6 (0.04/day)
Location
Canada
Could be PL1 and PL2, could be different C state support, could be VRM temperature.

C states should be the same (I will check) on both machines.

Can you expand on where you are going with PL1 and PL2 and VRM Temps?

Cheers

Sonic.
 
Joined
May 8, 2021
Messages
558 (6.89/day)
System Name R&D Machine #1
Processor Intel Core i5 10400F
Motherboard Gigabyte B460M Aorus Pro
Cooling Scythe Choten
Memory 2x8GB G.Skill Aegis 2666 MHz
Video Card(s) Powercolor Red Dragon V2 RX 580 8GB 107 watt
Storage 512GB WD Blue + 256GB WD Green
Display(s) BenQ BL2420PT
Case Cooler Master Silencio S400
Audio Device(s) Topping D10
Power Supply Chieftec A90 550W (GDP-550C)
Mouse Steel Series Rival 100
Keyboard Steel Series 6GV2
Software Windows 10 Enterprise
C states should be the same (I will check) on both machines.
Despite up to 10 C states available, not all are actually functional.

Can you expand on where you are going with PL1 and PL2 and VRM Temps?
If different PL values are set, then one CPU will reduce clock speed more to fit within power limit. High VRM temps don't let chip to perform as well as it should and i5 10600K does draw over 125 watts. I may be wrong but it can draw around 180 watts and not all boards can deliver that without overheating, Asus Prime boards aren't exactly most robust, however Z490-P seems to perform well:

I would check VRM temperatures to see if it doesn't throttle. You should also watch this video to understand PL1, PL2 and Tau:
 

unclewebb

ThrottleStop & RealTemp Author
Joined
Jun 1, 2008
Messages
3,720 (0.77/day)
PC1 has Hyperthreading = OFF
Hard to compare two computers when they are not setup equally.

Are both computers using the Windows High performance power plan with both CPUs locked to their maximum speed?
 
Last edited:
Joined
Nov 7, 2009
Messages
3,704 (0.87/day)
Location
Denmark
System Name The work PC /4790 fold&WCG /2700x fold&WCG/2700x full time WCG
Processor 3900X stock/ 4790K 4.4/2x2700x stock
Motherboard Gigabyte AORUS Master X570/Maximus Hero/2xMSI X470 M7 AC
Cooling Custom WC/Noctua NH-D15/ 2xNH-U12S SE
Memory G Skill Trident Z F4-4000C18D-16GTZ 3800 CL 16 /8 GB Corsair Vengance/2x16 GB XSkill Flare CL14
Video Card(s) 2070 Super X MSI/2070 Z/2060 X MSI + 2060 Ventus MSI/GTX 970 MSI
Storage 1 TB SSD / 480 GB SSD/2x250 GB SSD
Display(s) Dell UltraSharp U2518D/2408WFP
Case Corsair 800D / none/Lian test bench/NZXT 500 crap
Power Supply AX 850/AX 860/AX 860i/AX 760
Software Dual boot/Win 7/Linux for AMD
Benchmark Scores Cherry pie at one point ;)

SonicMojo

New Member
Joined
Mar 15, 2021
Messages
6 (0.04/day)
Location
Canada
Why does PC1 show 6 cores - 6 threads while the second computer shows the correct 6 cores - 12 threads? Did you deliberately or accidentally turn off hyper threading in the BIOS which disables half of your CPU?

Open up msconfig, click on the Boot tab, Advanced Options... and make sure the Number of processors box is not checked. Reboot so Windows can find all of your CPU.

Run CPU-Z to confirm that it shows 6 cores 12 threads on both computers.
Hyperthreading is deliberately turned off on PC1. Hyperthreading results in higher load and potentials for crackling and dropouts in many DAW applications so we (the audio community) generally turn that off.

Again - I realize there are differences between the two builds and I did list Hyperthreading as being one of those differences in the first post. I am totally open to making changes in one area or many to get PC2 to be as responsive as PC1 is right now.

I do find it bizarre that the machine with no hyperthreading on and a lower overclock (9% vs 17%) is running faster and smoother than a machine that (judging by the current settings) - should be faster.

Sonic.

Are both computers using the Windows High performance power plan with both CPUs locked to their maximum speed?
PC1 is running Windows 10 Pro (20H2) using the Ultimate Performance plan where as PC2 is running Windows 10 ENT (20H2) and using the same Ultimate plan.

CPU: As mentioned in the first post - PC1 CPU is just slightly overclocked (9%) (CPU Clock 4500 MHz (original: 4100 MHz, overclock: 9%) and PC2 is more than slightly overclocked (CPU Clock 4798.8 MHz (original: 4100 MHz, overclock: 17%)

By the book here - PC2 should at least feel as fast (or faster) than PC1

Sonic.
 
Top