Hey guys if you have uploaded your CPU-Z screenshot go ahead and put the code in your signature. Welcome to the dark side.
I've been messing around with my benchmarking program and the algorithms. That way I can continue to bring the best results so people know what to expect from the X58 platform and the GPU I benchmark. Currently it's the Fury X since that the only card I have on hand at the moment. I've been trying to get rid of bad data or as I call it "rotten" data during benchmarking. I won't go to deep detail. I've noticed that the minimum FPS [FPS min] is pointless since it can change depending on a lot of different factors.
For instance when the game is loading, saving. sending data\retrieving data or whatever it's doing can cause inaccurate results. I've tested this with the same games and different games. I got different results every time during loading or checkpoint sections. Very misleading results that I have experienced for years. Pointless data. This is the reason I created the "FPS Min Caliber ™" category last year. It will give a better representation of the rig setup + GPU based on "good" actual data. There's a lot of other things I've done to get accurate results as well.
I was playing Ryse: Son of Rome tonight at 4K. Instead of using my high 4.8Ghz overclock that usually run for benchmarks, I decided to run my 24/7 daily 4Ghz + DDR3-1400Mhz [9-9-9-24-1T]. I normally play games at 4Ghz, but I never benchmark at 4Ghz. I ran Ryse: Son of Rome at it's max settings @ 4K. The gameplay was great and the game is beautiful. That CryEngine 3 is gorgeous. I'm using the Crimson 16.1 Drivers.Ryse: Son of Rome [100% Maxed] – 4K
AMD R9 Fury X @ Stock Settings [Crimson 16.1 Drivers]
RAM: DDR3-1400Mhz [9-9-9-24-1T]
Gameplay Duration: 25 minutesFPS Avg: 33.4fps
FPS Max: 47.4fpsFPS Min Caliber ™: 24.3fpsFrame time Avg: 29.9msFury X Info:
GPU Temp Avg: 41.5c
GPU Temp Max: 42c
GPU Temp Min: 34cCPU info:
CPU Temp Avg: 38.5c
CPU Temp Max: 46c
CPU Temp Min: 36c
CPU Usage Avg: 17.77%
CPU Usage Max: 30.60%
The level I played was Chapter 4. Gorgeous graphics and plenty of physics during the levels. My benchmark program overhead isabout a 3% - 5% so it's very minor. 4Ghz + lower clocked RAM did pretty good with a demanding game. No micro-stutter, no screen tearing and button input was great. The Frame Time was smooth and consistent. I'll probably run the benchmark again some other time with Fury X overclocked to see how many FPS I can pick up. Anyways the stock Fury X does well @ 4K. The GPU temps were great as usual and the CPU temps were as expected.
Notice that I did not add the FPS min in the results above. If you read the paragraph above you'll know why I left that stat out. It's not consistent data. There's other inconsistent data that has to be weeded out. The first several dips you see in the chart are loading\saving\checkpoint areas. The actually FPS Min should be approx. 19fps-20fps which is literally 0.1%....meaning you'll nearly NEVER see the frame rate this low. The FPS Min Caliber ™
: 24.3fps can be thought of as the 0.9% or 1%. It's not random data either, it's actual data that can be proven. I'm focusing more on data that matters and excluding info that is not consistent or needed. I normally would clean up the chart, but I posted the actual RAW data. I'm still doing a lot of testing and re-programming.
It's difficult, but my program isn't perfect just yet. I'm still working with it. So far so good. For instance if 115fps pops up [and it has], I know that info is incorrect. No GPU is putting out 100+fps @ 4K in this game 100% maxed out. With that being said Ryse: Son of Rome needs to be in every benchmark in my opinion. It's a demanding game and pretty fun.