It's a combination of nodes losing efficiency gains like we got in the past, coupled with Nvidia accepting higher leakage dies while maximizing performance/clock speeds to prioritize profits.
As we saw with Ampere, an end user could usually shave 30% off the power draw while only losing ~3-5%...
Everyone here trying to say Ampere's vram is a huge source of its power problems - that doesn't explain how much power is saved by undervolting the GPU and also the fact that the 3070 is only 5-6% more efficient than the 3080.
But also, has anyone ran benchmarks with the 3080 at 16gb/s vram? If...
This guy has literally made nearly every plausible prediction possible since Ampere has come out. There are higher vram cards, no there isn't. There is a GA103, no there isn't. There are TI models, no there aren't. There are super models, no there aren't.
When a person makes this many...
Nvidia definitely pushed Ampere to the limit, but given that Navi 21 is about 15% more dense on TSMC 7nm AND AMD's huge perf/w increase within the same TSMC 7nm node, I think it's a forgone conclusion that Samsung's 8nm is noticeably inferior to TSMC. A straight port to TSMC would likely yield a...
Ampere is quite literally the first architecture from Nvidia since before Fermi that I could see getting a straight forward port to 5nm with only very minor tweaks in the architecture. Scaling is currently much worse with bigger dies / more cores than with past architectures and efficiency is in...
I'm incredibly impressed with how much AMD has fought back and gained vs. Nvidia, but overall I'm disappointed with both camps this round.
Ray tracing on AMD is exactly the same performance as Turing - a feature everyone rightfully called a tech demo then. I imagine AMD will work with...
But it is "that simple." With games currently available, Ampere is experiencing clear scaling issues. Turing, Pascal, Maxwell, and even Kepler all had noticeably better scaling with more cores and bigger chips.
If it's a matter of getting developers to code specifically to Ampere's strengths...
80% power without a performance drop is great.
It's a shame we've gone from overclocking 15+% with Fermi & Maxwell to undervolting just to shave off heat.
Solid card at $400.
It seems to me that Ampere as an architecture has significant scaling issues.
The RTX 3070 has 21% more cores at a slightly higher clock speed, yet is only 14% faster in a best case (4k) scenario.
The RTX 3080 has 80% more cores at a slightly higher clock speed, yet can...
The perf/$ is tied at 4k for rasterization and much better for Nvidia with RT on. Also Nvidia gets the DLSS advantage until AMD can respond with an equally impressive solution with the same developer support.
Wow I hate to so it, but I'm a little disappointed. I was expecting the 6800xt to be faster than the 3080 in rasterization but its actually a little slower. And in RT it gets straight up smoked.
The 3070 gets spanked in rasterization vs. the 6800, but again AMD chokes in RT. DLSS is the 3070's...
NAVI21 GPU details
7nm
519.8 sqmm
26.8 Billion Transistors
And there we have it - more proof that TSMC 7nm is vastly superior to Samsung 8nm. Navi 21with better perf/w and better xtr/mm2 (Navi 51.6m xtrs / mm2 vs. Ampere 45m xtrs / mm2). AMD is going to have the trifecta win this round...
This is all spot on. CP2077 will drive hardware sales for the next two years and will probably attain better FPS with RT on Ampere but Nvidia pooped all over efficiency this round and gimped their vram amounts.
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.