Read more.CTO Mark Papermaster urges foundries to intro EUV "as fast as they can".
Read more.CTO Mark Papermaster urges foundries to intro EUV "as fast as they can".
I don't know why but I am really excited for the 7nm node, it will be a big jump for both companies and because it's a completely different scale it will be interesting to see how both companies fight on a level playing field.
Well its more the case Samsung 14NM was a process more orientated towards mobile and is basically 20NM with finfets.
7NM should be realistically closer to Intel 10NM,and is a high performance process which had IBM involvement.
OFC,we will need to wait and see,but it MIGHT mean they get closer together in clockspeeds.
Can anyone reasonably guess the kind of increase in perf/watt we'll likely see?
Yes, but AMD was late to the 14nm party so Intel had the home advantage whereas if they both arrive onto the node together we will see genuine competition where time and experience with the node will put you ahead.
https://www.globalfoundries.com/tech...nce/7nm-finfet
This technology provides world-class performance, power, area and cost advantages from 7nm scaling. Based on 3D FinFET transistor architecture and optical lithography with EUV compatibility at key levels, 7LP technology delivers more than twice the logic and SRAM density, >40% performance boost and >60% total power reduction compared to 14nm foundry FinFET offerings.
How comes Skylake ate more power at 14nm than 22nm?
Skylake didn't exist on 22nm.
Perhaps I'm reading too much into this, and it's just a comment about historical relationships, but AMD uses solely GloFo for their CPUs and GPUs at the moment? I wonder if they're planning to move some capacity back to TSMC in the future? I'm still somewhat surprised they stayed with GloFo for Vega.AMD has been working with Globalfoundries for CPUs and TSMC for GPUs and Papermaster has been happy with this choice.
Yeah that's why I was expecting Vega back at TSMC - Polaris no doubt helped with capacity at GloFo but now Ryzen is in volume production, I doubt they're struggling there, not least with TR, EPYC and Ryzen mobile.
AFAIK the only things in production at TSMC are the console parts? It just seems a bit back-to-front - sure, there was probably less porting to do in keeping GCN and Jaguar at TSMC, but they stayed even after moving on from 28nm.
I could be wrong about the performance delta (or the significance of it), but most of what I've seen points to GF14 being better suited to lower clocks, and running into a clock wall before TSMC16? However Nvidia's GP107 does well on 14nm (but apparently Samsung), albeit with lower specified clocks than the 16nm parts. But its existence in itself is curious - what was the motivation for porting Pascal to another foundry, maybe it's substantially cheaper and/or TSMC are nearing capacity on 16nm? That could also explain AMD's choice to stick with GloFo where they likely have preferential access to lots of capacity.
And just to confuse things further, this die photo shows 'Taiwain' etched on the die! https://www.techpowerup.com/reviews/...images/gpu.jpg
if neither company can get a 7nm chip to 5ghz non boost, then the idea is only good for waste measures.
Hmm, have we actually had it from an authoritative source that GP107 actually is 14nm? All I can find from Nvidia themselves is 'FinFET' (notable not 16nm FinFET shown on earlier Pascal slides) - has 14nm been explicitly stated in press kits or something? Because that 'Taiwain' etching is very suspect!
Also, I wonder if this is the reason for everything@GloFo: http://images.anandtech.com/doci/10631/WSA5.png
I wonder what those 'certain wafers' are? And I wonder how AMD arrived at the conclusion that having to pay GloFo to use another foundry was a good deal? They must have supreme confidence in performance and execution!
There are currently 1 users browsing this thread. (0 members and 1 guests)