I am guessing the 285 was a salvaged chip after they messed up the fine grained power gating intended for their 300 series update. They have had to fix their power control circuitry twice already: trinity (fixed in richland) and kabini (fixed in beema).
So the 380 could be a fixed Tonga.
It is clear DX12 and Vulkan are largely based on Mantle, but that isn't really the important point. People somehow fail to realize that the console APIs and these new PC APIs are all similar because they are all low level APIs targeting similar hardware. And the reason Nvidia has no problem...
Seems clear that AMD are prepping architecture and uncore improvements for carrizo. The frame buffer compression, 16-bit float support, and improved tesselation all allow for better performance in low bandwidth situations. Combined with DDR4 we could see a decent increase in graphics...
Seems like Carrizo is a good candidate for this technology. Finally solves the memory bandwidth problems of existing APUs and in mobile they could even potentially forgo any additional system RAM.
I could also see them updating their mobile GPU line up on GF's 28nm process. Beema and Mullins...
I am surprised no one mentioned "Adelaar". It is described as a "discrete" GPU, but it can embed 2GB of memory with MCM packaging. This must be the solution AMD is aiming for to overcome the memory bottleneck on their APU designs.
The additional flexibility and ability to detect and recover from corruption should be of interest to anyone putting in the effort to create a RAID array.
You are just plain wrong. Top of the line hardware RAID controllers stick two PPC cores and 1GB of RAM on a card with a battery. There is...
ZFS addresses a lot of issues that have nothing to do with RAID, overall RAIDZ is a relatively small part of ZFS. That said, the performance of RAIDZ isn't really better than a similar configuration of software RAID (similar amount of RAM writeback cache, perhaps SSD writeback cache).
Also it...
Well Amdahl's Law actually applies to single threaded performance per clock too. All IPC scaling comes from caching and parallelization of a single instruction stream. For IPC gains the situation is actually much worse because the parallelization has to be completely automatic, so it is quite...
I agree. It makes sense to use a low resolution to compare the general performance of CPUs (for the type of processing games need). But you need to put that in context. If I can get 95% of the performance for 50% of the price when running at the resolution I need, that is infinitely more...
One of the most interesting possibilities is for AMD to allow customers to mix and match architectures in a single system. Allow people to run a VM for memcached on top of some low power ARM cores, and a VM for your database server on top of some high performance x86 cores, and have them all...
This post is so laughable I had to register and account so I could respond:
1) This chart is only for double precision, a known weakness of GPUs. Also, generally not very important in consumer applications.
2) The Cypress card is the equivalent of a 5870 (released Sept. 2009). A fair...
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.