By using this site, you agree to our Privacy Policy and our Terms of Use. Close
haxxiy on 24 January 2017
Edit Thread
Snoopy said:
Barkley said:

Try using proportions, it's better then numerics as the difference between a 4.2TF and a 6TF GPU isn't as big as the difference between a 0.1TF and a 1.9TF GPU for example.

XBO - 1.31TF :  PS4 - 1.84TF      Difference: 40.4%

Pro - 4.2TF     :  Scorpio - 6TF     Difference: 42.8%

Percentage isn't very telling in this situation. For example, the difference between 1mm and 2mm is twice as much, but to us there is hardly a difference because its a mm. However, the difference between 1 mile and 2 mile is huge to us, despite being the same percentage difference between 1mm and 2mm. 2flops difference between xbox scorpio and ps4 pro vs .5 tflop difference between ps4 and xbox one.

It doesn't work like that, otherwise graphics would get exponentially better since every new die shrink would mean graphics become twice as good, over and over. In reality it's sort of the opposite, for you need exponentially more computational power to get the appearance of graphics improving on a linear rate.

Unless you want to believe the 200 GFLOPS difference between the Dreamcast and the PS3 yielded the same perceived graphical benefits as the PS4 Pro GPU against the very same GPU theoretically clocked a mere 4% slower.