The video game benchmarks really show that it's still not powerful enough to play less than 5 years old video games in good conditions. Old games are fine, as always.
Just to clarify for anyone who at first misunderstood the comment as I did: the integrated graphics are not so powerful, not the CPU itself. So you'll still want a dGPU in your laptop if you plan to game.
It's a really misleading comment (if I'm being charitable) considering that the review points out, quite clearly, how the laptop being reviewed has a dedicated GPU, which is pretty powerful, though arguably maybe 30% slower than the full powered desktop RTX 2060 (and a third the power consumption.)
The test that is linked is about the integrated GPU though.
Which is pretty much a given with integrated graphics. The Vega series iGPUs are worlds faster than Intel's, even going back to the 3000 series laptops.
I assume you mean that AMD's integrated graphics still do not compete with discrete GPUs.
Which isn't surprising - integrated graphics are never going to beat discrete GPUs with their own power delivery, cooling and dedicated high speed memory.
This is not intrinsically true. Cooling and power delivery certain have it beat, but it seems feasible to beat based on memory bandwidth. I bet a GPU uses roughly the same order of power/cooling that say half a threadripper does?
No, how do you beat the wide bus memory bandwidth of a discrete GPU with the standard dram bandwidth? Even if your processor was not competing with the GPU for bandwidth you'd still have a small fraction of the memory bandwidth available.
You'd have to come up with a new architecture to make this work, but perhaps you could have CPU/GPU on the same chip with dedicated GDDR memory lanes.
PS4/XBOne have an APU with powerful (ish) graphics - how does their memory work?
They work with graphics memory rather than regular memory if my memory serves. Hehe.
Aw, I was thinking bw between the CPU and GPU directly, my bad.
Aren't console GPU chips on the same die as the CPU ?
Might be more accurate to say that consoles put the CPU on the same die as the GPU. The GPU portion of the console SoC die is bigger than the CPU portion and the memory controller is very much a GPU memory controller not a CPU memory controller.
Yes sorry to not be explicit enough. It's still way too slow compared to a dedicated GPU.
HW-wise the Vega is better, but the SW, OpenCL compared to Cuda at least on Linux, sucks. It rarely works, and you get the crashes at runtime, not compile-time.