Kyle was one of the first to really push the idea of real-world gameplay over canned/synthetic benchmarks, and that was one of the main reasons I found myself on [H] circa 2001. I remember when [H] was also blowing the whistle on GPU driver cheating with synthetic benchmarks later in the 2000s.I guess I share in Kyle's distaste for "Canned benchmarks".
My previous job was working for a freelance software QA company. We tested everything from video games to mobile phone apps/games. nVidia was one of our biggest clients. We primarily did GPU driver testing for them. So through that I've gotten familiar with all the 3DMarks since 2001, and various other random-@ss synthetic benchmarks.I ran Time Spy just for ****s and giggles when I picked up my 6900xt, but that's the first time I've run a 3DMark benchmark since 3DMark 2001.
Yeah, that was their only real useful purpose, to go "oooh" and "aaaaah" at all the pretty tech shinies. Not representative of real gameplay, but interesting to see an actual implementation of new tech in APIs and hardware. I see 3DMark is keeping up the tradition busting out more DX12 Ultimate benchmarks.You know, I used to LOVE 3DMark back 20 years ago.
But more as a technology demo than an actual benchmark.
It was so cool to see what was theoretically possible if you challenged the hardware of the day to its absolute limits.
It was weird some when games like the first Max Payne included 3DMark scores with the minimum and recommended system requirements on the retail packaging/game cases.