tripenfall said:
This article offers a formula but it's gamespot so take it with a grain of salt.
https://www.gamespot.com/articles/console-specs-compared-xbox-one-x-ps4-pro-switch-a/1100-6443665/
Their formula is The basic formula for computing teraFLOPS for a GPU is: (# of parallel GPU processing cores multiplied by peak clock speed in MHz multiplied by two) divided by 1,000,000 Let's see how we can use that formula to calculate the teraFLOPS in the Xbox One. The system's integrated graphics has 768 parallel processing cores. The GPU's peak clock speed is 853MHz. When we multiply 768 by 853 and then again by two, and then divide that number by 1,000,000, we get 1.31 teraFLOPS. Anyone care to weigh in on this formula I honestly have no idea... |
Yeah, this is usual formula that's been used for quite some time - and it is correct for more modern GPU architectures - I'd say, anything from ATI's unified shaders onward...
Pemalite said:
Number of cores * Number of SIMD units * ((Number of mul-add units*2) + Number of mul units) * Clockrate. It just reinforces the fact that flops is a useless metric. |
Yeah, I know about that one too...or variation of it...yet I have no idea how anyone come up with GC and XBOX numbers.
GC has 4:1:4:4 @162MHz, while XBOX has 4:2:8:4 core config @233MHz (pixel:vertex:TMU:ROP)...how one comes to actual FLOPS is beyond me without knowing particular architectures.
For example, I still can't figure out how to convert PS3's RSX GPUs specs into FLOPS (24:8:24:8 part), since, to me at least, something seems to be off with quoted numbers, as if they are conflicting each other. For example, current GFLOPS at wiki are 192 for pixel shaders (I remember this being changed numerous time), and this is quoted from K1 whitepaper, which states 192GFLOPS for whole PS3's GPU.
- 24 parallel pixel-shader ALU pipelines clocked at 550 MHz
- 5 ALU operations per pipeline, per cycle (2 vector4, 2 scalar/dual/co-issue and fog ALU, 1 texture ALU)
- 27 floating-point operations per pipeline, per cycle
- Floating Point Operations per a second : 192 GFLOPs
- 8 parallel vertex pipelines
- 2 ALU operations per pipeline, per cycle (1 vector4 and 1 scalar, dual issue)
- 10 FLOPS per pipeline, per cycle