It will feature more than 50 cores, support 8GB+ of DDR5 memory and will be manufactured using the 22nm process. The architecture will come in PCIe form factor and will support Intel’s standard programing model and software tools.
Thats the crazy thing. No better information about the way we walk at the moment. We all need a lot of resources for a more Nebula use. Invidia release some great stuff for Cuda but is this in real a good solution for us? Short latency, Pure Cuda Version? Or better to use a big server system like other posts preferred here. The Nebula server edition is working great. Or better to wait for a solution like Xeon Phi? Maybe easy handling and the Nebula Core III engine is working fine with that? On the other side Giancarlo (or Enrique) wrote some things about a Nebula developing with less Ram and CPU use. Big things are coming but no information at this time. To be patient... Damned I dont.
I know it is an old post but I managed to find some pricing for this beast.
Intel is doing (to its credit) what it always does: putting a price tag on the cards. The passively cooled Xeon Phi 5110P is shipping for revenue at Intel now, and will be generally available on January 28 to the rest of us for $2,649. The actively cooled Xeon Phi 3120A card, which is hotter and yet has less memory and bandwidth, will be available sometime in the first half of 2013 with a price that is expected to be around $2,000.