Not Global foundries, not TSMC .. but Samsung will be fabbing next yearsÂ*Arctic Islands family of GPUs. Since it's Samsung it'll be a 14nm FinFET fab, opposed to Nvidia who will be using 16nm for t... Samsung to fab 14nm GPUs for AMD
This will be interesting! On the one hand, TSMC has a proven track record in the PC space, while Samsung does not. On the other hand TSMC's track record is, quite frankly, spotty at best. Plus it will be for the first time in quite a while that AMD and nVidia ship their GPUs on different nodes, if this is true. And although the two processes are pretty much equivalent in size, despite the difference in nomenclature, it'll still be interesting to see what comes out of it. Bring on Q2 2016 then! I really need an upgrade.
I'm just hoping Pascal and others don't jump too far ahead of Maxwell as i want my GTX980 G1 to last a bit longer.
Nooooo! I was going to buy a G1 as well, but resisted the temptation successfully. So, I want the next-gen noooooooooooooooooow!!! :banana:
Can be even more interesting than we now anticipate. Or it doesn't make any difference. Hard to say. Even if AMD could have a small edge on manufacturing process, it is still questionable if they can beat NV in the end. NV has a track record of having better architecture except for bitcoin mining...
I read that same article and even at a slightly larger node the tsmc used way less power and generated less heat, it was just a much more efficient product. I would like to see another big player making gpu's but amd is already known for hot inefficient gpu's so i hope this doesnt end up the same.
Our nvidia drivers are so bad that we fear a new generation being released and making us all second class citizens.
Aside from the initial Windows 10 drivers being a little flaky, I haven't had any problems lately with their drivers.
Yeah but you have a 980, a lot of people with 780s have been crying because they feel like nvidia brushed them aside along as collateral damage to make maxwell look good compared to amd with the use of gameworks .
I actually saw articles claiming the exact opposite results, so I wouldn't really put much faith in these results either way. Among those who have looked into the issue, the most "authoritative" sources I can find are Tom's Hardware, which are the ones giving the (slight) nod to the Samsung Chip: tomshardware.com/news/iphone-6s-a9-samsung-vs-tsmc,30306.html And Anandtech who came to the conclusion that "There is no reasonable way to accurately measure this": anandtech.com/show/9708/analyzing-apple-statement-for-tsmc-and-samsung-a9
Those are funny 0.5~2W chips at best. It tells no story how each technology works once you use it to 150~250W monster.
Tom's hardware is known to be often biased and opinionated, I wouldn't put much faith in their articles. Anantech on the other hand is great, I have yet to encounter any article there that seems to favor any one side based on brand. (guru3d is the best though love this site)
This. The process isn't even the same with bigger chips. It's impossible to compare mobile chipsets and make assumptions on significantly larger GPU's.
Don't believe what you hear. Do your own research. The 700 series is going mega strong. http://www.hardwarecanucks.com/foru...ws/70125-gtx-780-ti-vs-r9-290x-rematch-8.html Qft.