Wednesday, January 25th 2012
NVIDIA Taunts AMD's GCN Architecture Performance
As AMD's Radeon HD 7900 series is finding ground in the market, and NVIDIA's competitive product line still without a concrete launch schedule, the mind games have begun. In an interview to NordicHardware, a senior NVIDIA official said that NVIDIA expected more from AMD's new GPU family. "Honestly, we expected more from our competitor's new architecture," the official said, indicating two interrelated things:
Source:
NordicHardware
- AMD's Southern Islands GPU family's performance levels are well within NVIDIA's expectations
- NVIDIA's new architecture will be a lot more powerful than Southern Islands, because it was prepared keeping in mind a faster architecture from AMD than what Southern Islands ended up being
87 Comments on NVIDIA Taunts AMD's GCN Architecture Performance
It's very cute to release a chip that's say, 20% better than the competition's fastest chip, but when it costs 50% more, is it really THAT great an accomplishment?
This is nothing new, the Geforce 2 Ultra, back in... 2000 I think, cost $800 at launch. Bravo.
But nVidia's frothing-at-the-mouth fanbois never, ever cease to epeen with the "yeah but nV has THIS CHIP which is 20% faster than the fastest Radeon" preening like champions of righteousness while they try to run BF3 with their SLI GT 240*...
Nvidea has several extra months, they'd better perform better.
Nvidea has not produced any numbers. No numbers mean they haven't provided any proof.
Beyond those two points, Nvidea has done little more than cock slap the consumer. Proving that your chicken is the largest, without actually proving it, is worthless. Wake me when they release performance figures.
From what I've seen so far prices should be ~400eur in my region 420$ for US maybe!
Hardware speaks for its self and since they dont have shit to show their just blowing smoke.
Atleast AMD has something on the table nV :nutkick:
As far as their next release AMD has alot of overclocking headroom and at least another 3 months to work on it ;)
Sounds like bantar should've titled it "Wait for Me, please"... :shadedshu
Wake up, Nvidia won this round and that's it.
So nVidia doesnt even need to release hardware and they win :roll:
As much as I'd say to wait before saying anything I think Nvidia does not have to put much effort to surpass the HD 7970.
And no, don't say that the 7970 overclocks better because that's mostly thanks to the 28nm node and the next Nvidia chip will probably overclock as good as the 7970.
- HD7970 == 4.3 billion transistors.
- GTX580 == 3 billion transistors. +50% == 4.5 billion
- GTX 560 ti == 1.95 billion transistor. x2 == 3.9 billion (how does GTX560 ti SLI perform against HD7970 again??)
- Kepler more efficient than Fermi.
End of story.
Then, it'll be a while still before nVidia outs their supposed "high-end" Kepler, or a dual-gpu of their own.
But how much more has been in development we do not know. GCN is a new arch that was suposed to be released when Cayman, but was delayed until 28nm, so it's been in development for 1 more year than Kepler, probably.
EDIT: Also having x months more to develop means absolutely nothing in this market. What can be done is tighly related to the manufacturing process and the limits are the same today as they will in 9 months. Look at Cayman, released 1 year after Cypress, and it was faster than Cypress, but also increased die size and power consumption accordingly. They had 1 more year, but...