Wednesday, July 23rd 2014
NVIDIA Timing GeForce GTX 880 Launch with Gamescom?
NVIDIA is reportedly preparing new graphics cards based on the "Maxwell" architecture, which judging by the purported 3DMark FireStrike screenshots, could be high-end parts, maybe even the GeForce GTX 880. The cards could launch along the sidelines of Gamescon, a gaming expo held each year in Cologne, Germany, around mid-August. Given how most information to date points to a later-Q3, maybe even mid-Q4 launch of these cards, it wouldn't surprise us if NVIDIA merely teases these cards. NVIDIA's next big consumer GPU is widely expected to be the GM204, a performance-segment chip based on the "Maxwell" architecture.
Source:
VideoCardz
20 Comments on NVIDIA Timing GeForce GTX 880 Launch with Gamescom?
and yes im also waiting for this to turn into a amd fanboy section :p
I must do it for my country :D
If this thread follows just about any other graphics story since 1998, You'll probably be seeing comments along the lines of " too expensive" (even though no pricing is mentioned), "I don't like Nvidia because [insert random proprietary tech]", followed by some rambling about drivers, SLI scaling, the absolute necessity of DirectCompute and GPGPU functionality, and never-to-be-attained-theoretical FLOPS performance, relative price-per-perf, and perf -per-watt of cards that aren't the GTX 880. Cue backlash. Cue unrelated comparisons and anecdotal sob stories of dashed dreams.
Something like that....but it won't just be AMD fanboys involved.
Let me just start it off by saying it simply looks like both AMD/nvidia will have fairly-optimized 64 ROP parts (more-or-less 4x the ps4 gpu spec, the same way Hawaii/GK110 products are ~4x xboxone). Both look to be coming relatively soon and on 28nm. That both companies seemed to make that feasible is pretty astounding, even if HPL saves a bunch of power...should be some big freakin' chips. I'm sure they will be very nice, bump each current series down a peg (while being themselves refreshed in one form or another) and truly usher in 4k gaming while elevating the baseline...
But yeah...How much $? Prolly a lot. Not just kind of a lot, prolly a lot a lot. Also, if in a similar power envelope, at what point does more units at a lower clock really make a lot more sense than less overclocked to a higher one, even if more efficient?
On a purely fundamental level, I wonder how each achieves this feat architecturally speaking. I could see AMD doing 512-bit and 7gbps ram, but what about nvidia? Tons of cache (bigger chip) and slower ram? Should be interesting. /#jointhespeculation.
L2 cache - if GM 204 follows GM107's example, will be sizeable and should allow for a reasonable speedup. The GM 107 suffers in heavily in comparison to GK 106's bandwidth, bus width, memory speed, core count, ROP count, and texture address units, but the cache structure certainly mitigates what should be a major performance disparity.
The NV UK's facebook post was never attached to the image above it. The T-27, at the date it was posted, matches to the date Shield-tablet will be available in UK.
This picture of 3DMark looks completely legit and there's no way this isn't a GTX 880. Clearly.
:|
being that low power is maxwell's bread and butter I'd expect the mobile solutions to be fully vetted before the desktop high ends come out.
AMD may have the tech crown for their 295 but that probably doesn't worry NV. I could see NV sitting on their Maxwell cards until AMD 'next gen' rumours started.