Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.
Is the benchmark for the redux the same as the standard? I only have redux......
Is the benchmark for the redux the same as the standard? I only have redux......
You must think us all stupid? he said toward the end, thats 0.5 seconds worth of the end. The last 25 frames, the fade to black.
Yep look at the actual graph provided by the benchmark. There is more than 1 moment where the game tanks.
Even with 12 threads Borderlands 2's PhysX brings my PC to its knees. I can imagine the single-figure FPS seen with Core i5s!Its shared between the CPU and GPU, or at least it was, older PhysX games are broken on newer nVidia GPU's, like Pascal, Like it does with AMD it is running on the CPU.
Its the same on Borderlands 2, i also play MLL, trust me, with PhysX on the CPU gets overloaded, especially a 4 core.
We can see that on the Graph, no point in highlighting it in Excel, i told you MLL picks up and scores frame hiccups, thats one, literally one frame, those long coaster dips are not frame hiccups, they are CPU bottlenecks.
Except that I can't see it in the graph. The graph is automatically plotted using this csv file. You probably want to accuse me manipulating the csv file next?
Its here...
Yeah, isn't it within the first five seconds?
Yes you can stop posting screencaps now...., its just a glitch, nothing to do with the performance of anything....
But those Roller Coaster dips, those are CPU bottlenecks.
Metro Last Light is perfect for testing the performance of the GPU and how well the CPU is able to feed the GPU, (Draw Calls) its also a very well threaded game.
Those areas where the CPU is put under a full load, those are the 'canyon shaped dips' on the graph, the Ryzen chip compared with the 6600K, at least, has a lot more compute threads and with that much higher performance to feed the GPU, that is what you are seeing in my screencap with 6600K @ 100 FPS and the Ryzen 1600 @ 160 FPS.
This does matter for longevity in gaming because it means the Ryzen 1600 is capable of driving much more powerful GPU's than the 6600K is in MLL and similarly optimised games, which is true for the future as they become better threaded.
You may have seen another screencap where the 8 compute threads on a 5Ghz 7700K in BF1 are completely saturated? do i need to post that again?
You completely missed my point. I never disagreed that there are some games (and more games in the future) which can support more than 4 threads. If you like playing 720p low settings I'm fine with that. I'm only claiming that there exists no single GPU capable of doing 4k max settings at over 60 fps, not in the near future of the first generation Ryzens. I just showed that the 1080 (non-Ti) can do only 50 fps lower bound in the LL Redux benchmark, which translates to 12.5 fps for 4k. If each generation of GPU improves by 37% then it would take five generations after 1080 (or four generations after 1080 Ti) to reach 60 fps for the lower bound in the LL Redux benchmark. And yet 60 fps is still not CPU-bound. You'll need over 100 fps to see it to be CPU-bound, and that's gonna be 5.6 generations after 1080 Ti.
Yes.... YES YES YES this is exactly the point, when you test a CPU's true performance in games you make the CPU Not the GPU work for it.
What do you think will happen when GPU's get faster, and faster and....? the CPU will have to work harder and harder to keep up, the way you test for that with weaker GPU, or one from today is to offload the work onto the CPU by turning the resolution down, it mimics the faster GPU making the CPU work hard to keep up.
The result is as that screenshot proves the Ryzen Chip with 50% more cores, 75% more threads is 60% faster than the i5 there.
So the better CPU for longevity is....??????? yes, its the Ryzen chip.
Not at all, i have already addressed this, using low resolution and settings is a long time tried and tested way to highlight the CPU's performance in games, how well the CPU is able to feed the GPU with Draw Calls when the GPU is fast enough to Call Draws at a fast enough rate to stress the CPU.
This is my original response the first time you said this to me. different people using that account? make sure you keep track of what each of you have already been over.
You completely missed my point. I never disagreed that there are some games (and more games in the future) which can support more than 4 threads. If you like playing 720p low settings I'm fine with that. I'm only claiming that there exists no single GPU capable of doing 4k max settings at over 60 fps, not in the near future of the first generation Ryzens. I just showed that the 1080 (non-Ti) can do only 50 fps lower bound in the LL Redux benchmark, which translates to 12.5 fps for 4k. If each generation of GPU improves by 37% then it would take five generations after 1080 (or four generations after 1080 Ti) to reach 60 fps for the lower bound in the LL Redux benchmark. And yet 60 fps is still not CPU-bound. You'll need over 100 fps to see it to be CPU-bound.