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

Weird varying results unigine superposition

Joined
Aug 9, 2019
Messages
1,812 (0.87/day)
Processor 7800X3D 2x16GB CO
Motherboard Asrock B650m HDV
Cooling Peerless Assassin SE
Memory 2x16GB DR A-die@6000c30 tuned
Video Card(s) Asus 4070 dual OC 2610@915mv
Storage WD blue 1TB nvme
Display(s) Lenovo G24-10 144Hz
Case Corsair D4000 Airflow
Power Supply EVGA GQ 650W
Software Windows 10 home 64
Benchmark Scores Superposition 8k 5267 Aida64 58.5ns
I run my 4070@2610@915mv and +1000 on vram. When running 1080p extreme I get anywhere from 8200 to 9100 points. It seems the gpu throttles sometimes. I also tried disabling vram oc, same result. Sometimes score is just 10% below. But it is never in the middle. Like a powersaving being active or something. I run windows at max performance. I have tried 8k optimized now and here score is consistent. 0 vram oc gives 4900, +500 5000, +1000 5100, +1500 5200. Any ideas?
 
My guess is your 6-core CPU being a bottleneck for that 4070.
 
My guess is your 6-core CPU being a bottleneck for that 4070.
I have 7800X3D, forgot to update spec. Even with 6 core should not vary like this.
 
1080P is all about system memory and cpu.

4KO is all about gpu, less cpu.
 
Yeah, but I dont understand why score harirs wildly. Gpuload is 99%, cpu is very low at 1080p. Clesn windows. Top notch aida, geekbench, cyberpunk etc scores. Weird.
 
Could be clockstretching. Try giving it more voltage or lower frequency and see if it smoothes out.
 
Could be clockstretching. Try giving it more voltage or lower frequency and see if it smoothes out.
For gpu it is strange as 8k tesults are 100% consistent, but since it doesnt appear to affect gamingperformance it may no matter that much, just weird :)
 
For gpu it is strange as 8k tesults are 100% consistent, but since it doesnt appear to affect gamingperformance it may no matter that much, just weird :)
To be fair, clockstretching isn't necessarily inconsistent, just worse performance. Different boosting behaviour on the two different test could explain running into issues on one but not the other. Only real way to test would be to increase voltage and see if results improve. However if everything works in the scenarios you care about then all is good.
 
Back
Top