zlacker

Interestingly, the transistor density of GPUs has been following a roughly logarithmic curve since 2000, compared to the linear increase in x86 processors [1].

I totally agree that the incremental innovations observed in earlier GPU platforms felt much, much more ‘obvious’ though.

It’s as if the ‘wow factor’ of graphics hardware doesn’t scale at the same rate as density.

Or perhaps releases were more spread out than they are today (compared to the annual release cycle expected today) making the jumps more obvious.

[1] https://www.researchgate.net/figure/Comparison-of-NVIDIA-gra...

>>greggs+(OP)
I was designing Mac graphics accelerators in the early 90s, one thing I learned is that perceived performance follows a sort of S curve, there's an area where everything is terribly slow and any change is good but it's slow, then you hit the curve every performance increase is obvious, wonderful, makes new stuff possible - that's where you start selling big if you're leading the pack, then you hit an area where performance is "good enough", double the performance and the perceived increase is far less - now you're fighting for "cheapest", not so good for your company.

That was for 2D, bigger faster 3D enables new sorts of games so that market has been growing for far longer.

>>greggs+(OP)
I wonder if part of that is because GPU tasks are ridiculously parallelisable?

If you can split the screen into 64 equal chunks, there's nothing except silicon real estate stopping you splitting it into 128, or 256, or 2048. Think about how SLI worked, in the Voodoo II olden days.

>>Taniwh+a2
It feels like we are two or three generations away from that in gpus for games.

>>christ+69
I admire your optimism but the next big with ray/path tracing, voxels and whatnot, I'd say not even ten generations away.

Good thing nice graphics do not equal good games. My favourite multiplayer FPS games I prefer in glorious picmip 5 detail.

>>beebee+Dc
Yeah the amount of sameness open world games is getting boring. Seems big budget games are afraid to take risks. Maybe generative AI will help smaller teams build bigger worlds for us.

>>christ+pD
I've been enjoying the wave of excellent boomer shooters very much. Turbo Overkill, for example, has the best movement mechanics, and fantastic gun play. Art and enemies, while totally acceptable, could be better. Not sure how AI would help but imagination is not among the very short list of my talents