Same here, HDMI 2.1 is my criteria for my next build so I can plug it into the living room and enjoy VRR and 4K at high frame rates on my future TV. My trusty 2500k+GTX680 will have to last me a bit longer, it had a great run.
I see you also have an RTX 2070 card.I am kind of regretting my RTX 2070 purchase right now. I could have waited for the Super lineup. Even if they will be expensive in my country at least I could have grabbed 2070 on the cheap. Coming from 980ti, playing games on 1440P ultra-ish settings @60 fps is great and all but the only game I used my card's RTX capabilities was Metro.
I believe they are talking about not having the restrictive voltage and power limits so AIB's can go a bit more nuts with their cards and push voltages higher without requiring custom solutions like they use on the Kingpin cards.
I think that the RTX 2080Ti core reference on the new RTX 2080 Super in the wccftech specs table is misleading.
oh bummer. Then I might look at the 2070 Super or 2070 Ti and hope for a good price there. 11GB on the RTX 2080 would be a small game changer for me and increase my willingness to pay RTX pricesI think that the RTX 2080Ti core reference on the new RTX 2080 Super in the wccftech specs table is misleading.
The most rumors I see have the TU104-450 chip for the RTX 2080 Super. That is the full TU104 chip with 3072 CUDA cores. For reference the RTX 2080 has 2944 CUDA cores. And also a speed increase for the VRAM. From 8GB 14Gbps to 8GB 16Gbps.
The RTX 2070 Super 2560 CUDA cores (from 2304). Same memory (capacity and speed).
The RTX 2060 Super 2176 CUDA cores (from 1920). 8GB VRAM from 6GB. Same speed (14Gbps).
I don't see how a card like that would be anything less than 500 dollars. No way their top of the line cards will be at reasonable prices.
I know. I'm just tired of these posts really. Anyone buying a higher end GPU should be ready for an even faster GPU coming out in less than a year - this is how it was pretty much always since the beginning of GPUs.
There's some conflicting info: some sources say that 2080 Super will use full TU104 chip (same as 2080) and have 8GBs VRAM, some point to it coming with a cut down TU102 chip (same as 2080Ti) and 11GBs.
There were higher and lower binned chips in 20 series, with higher ones getting an A postfix (TU104A, TU102A, etc). Presumably they mean that these chips will be of "A" variety from the start - but this is hardly something surprising or unexpected.
Until the stock dries up maybe. I'm expecting these to supplant current 20 lineup. The real question is on which price points these "super" cards will launch? It would make little sense to launch them on the same prices which current 20 series retails on.
Well it is a refresh on the existing product line. We can't expect huge changes.
Oh I know I was just commenting on the fact that the CUDA core bump on the 2080 Super would do pretty much nothing to perf (the 2060/70 Super fare better in that respect at 11-13%)
HDMI 2.1 and Witcher 3 performance above 60 fps at 4k is my requirement (2080 ti hits this, which is encouraging for future cards). Only downer is that 7nm cards may not see DP 1.5. Doubt the refreshes will change my plan and these new cards are what should have happened for last years launch.
But in the same rumor they say the 2080 super is a cut down 2080 ti.
I honestly think Nvidia is worthy of their position, the prices may be high but I haven’t been disappointed in any of my latest Nvidia cards. Haven’t jumped in to RTX yet though so maybe that would change my mind.
Totally agree on this, and I like it that way, the PC gaming industry needs to push things forward as much as possible, nothing good comes from holding back releases.
ATi would have been so much better without AMD. AMD gave RTG next to no money post acquisition while Nvidia have plowed cash into R&D. AMD basically gave ATi the busywork of sticking GPUs onto CPU dies which ended up being a low value wide goose chase in the end.
The flippant answer is "They could make obsolete gear work together well enough for $100 a chip." You can basically look at the "Enterprise, Embedded and Semi-Custom" section of AMD's earning report and see how much of a dud that one is at making money. Gross margin for the segment is something like 5%. Which makes sense because Nvidia have said they passed on it because Sony didn't want to pay enough (how much of that is sour grapes is left as an exercise for the reader), Intel passed because they don't do custom design work, and nobody in the ARM space could get the same horsepower out of an ARM chip that you were able to get out of a low end x86 core back in 2013.
Won't happen until Ampere next year. This is a refresh, the chips - and their features - will be essentially the same.
They sell them for a dollar. Also they have both high performance CPUs and GPUs - something which none other player on the market has right now.