Overclocking The RX 580 - Weird Results?

[alanp1995]

Hi, I'm trying my hand at overclocking my XFX RX 580 GTS XXX 8GB GPU model and having strange results.

I managed to find a very acceptable stable overclock with the Core Clock at 1425MHz and Memory Clock at 2250MHz.

 

I'm using 3D Mark to generate performance reports and comparing them with others online. Only others with the same specifications as myself.

 

My best benchmark was 4812 points, the highest results with my specs is 5397 points.

5397 points was achieved on a lower Core and Memory clock than my setup, how is that possible?

 

[emosun]

Theres a lot of ways to cheat scores in those cheapo benchmarks so im not sure why a miniscule % increase in some other randomly submitted score on this website would be a problem.

 

If they had scored 20,000 instead of 5400 would you still care or would it obviously not matter then? 

 

It could also be simply luck of the draw and for whatever reason they simply got a higher score in this useless benchmark for no reason. 

 

Was the machine purposely built to only run 3d mark and nothing else?

[alanp1995]

1 hour ago, emosun said:

Theres a lot of ways to cheat scores in those cheapo benchmarks so im not sure why a miniscule % increase in some other randomly submitted score on this website would be a problem.

 

If they had scored 20,000 instead of 5400 would you still care or would it obviously not matter then? 

 

It could also be simply luck of the draw and for whatever reason they simply got a higher score in this useless benchmark for no reason. 

 

Was the machine purposely built to only run 3d mark and nothing else?

I am just commenting on the fact that I'm trying to see if there's more I can squeeze out of my benchmark or it's just simply luck of the draw and their cards, yet identical, perform better on average - regardless of the testing method. Thanks for your reply.