Where are they bringing it to? CEX?
-
Competitor rules
Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.
OcUK Ampere RTX3000 series review thread
- Thread starter CAT-THE-FIFTH
- Start date
Where are they bringing it to? CEX?
Is it even possible to have two without being under water?
Thats winter heating sorted
It's when your glass side panel starts to look like those 100 year old window panes you have to worry!
I think the same guy posted a video, or at least I watched one where it was claimed Nvidia tried to bluff TSMC for a lower price. They did the arrogant, "we are Nvidia and we will go elsewhere if you don't give us a better price than all your other customers". TSMC basically said "knock yourself out, we can sell all the 7nm wafers we can make".
If it's true then fair play to TSMC not giving in to the bullyboy tactics. It left Nvidia on an inferior node with clear power issues.
Only if you live in siberia or keep your PC in the fridge.
Associate
- Joined
- 31 Mar 2016
- Posts
- 1,412
- Location
- Moonbase Alpha
So what why the cards are the way they are? What is there is what is on offer.
People will either buy them or not. I think every generation could have been different.
Associate
- Joined
- 27 Apr 2007
- Posts
- 974
It's called a negotiation which any company will do!
The bottom line is that companies are supposed to prioritise profits. Capitalism 101.
I dont think they are else you wouldn't get numbers like this at 1080p low settings in CPU benchmarks.
Especially when dealing with RTX, I think that is the story.
I'll be waiting on AMD based on the power consumption alone.
One thing that is obvious, is that Nvidia will for sure, be pushing RTX for all it's worth this generation to make use of the crazy amount of RT cores. If they don't then the gains don't look anywhere near as good.
One thing that is obvious, is that Nvidia will for sure, be pushing RTX for all it's worth this generation to make use of the crazy amount of RT cores. If they don't then the gains don't look anywhere near as good.
Associate
- Joined
- 31 Jan 2012
- Posts
- 2,323
- Location
- Droitwich, UK
When are AIB card reviews expected? Probably asked several times everywhere but I'm too lazy to search.
Shows what a lead TSMC have over everyone else, I think. I did read or hear though, that nVidia's partnering with Samsung could actually drive Samsung forward, which would eventually mean less of a monopoly on the top tier of fabrication for TSMC. Could be a good thing long term if that is true. Not something I know a great deal about though, tbh.
Makes me wonder if we've reached peak core scaling. Ampere doesn't have much IPC improvement it's all in the extra cores and those cores help for Compute - the 3080 is 100% faster than the 2080ti in Cuda accelerated work loads, but only 15% in games...ouch!
I think it just highlights a few things:
- Diminishing returns for conventional rasterization. The old hacks and fake lighting tricks just don't scale and hence the push to raytracing which is massively lore scalable.
- The 3080 is nearly twice as fast ss the 5700xt, and considering these diminish returns Bug navi is going to be more of a 3070 competitor.
- Ampere is massively CPU limited. And that extra few percent from intel CPUs in Games makes a difference.
- Ampere will probably show a bigger gain in future games
Peaks could reach about 450 watts per card going on how the 3080 performs and adding another 20%.
Actually, i would say it is the exact opposite.
Nvidia doubled transistor count from an already massive die and increased clocks while keeping power close (performance per watt improved).
The Samsung 8nm must be incredibly close to TSMC 7nm for these numbers, and all with a much cheaper and simpler process.
if you see the transistor density and clocks AMD achieved on 7nm then this lookd like a good move from Nvidia. The samsung 8nm must provide almost the same performance st much less complexity
Not really considering the A100 PCIe has double the transistors and only uses 250w.
Oh well.