AMD Unveils Carrizo APU With Excavator Core Architecture

MojoKid writes: AMD just unveiled new details about their upcoming Carrizo APU architecture. The company is claiming the processor, which is still built on Global Foundries' 28nm 28SHP node like its predecessor, will nonetheless deliver big advances in both performance and efficiency. When it was first announced, AMD detailed support for next generation Radeon Graphics (DX12, Mantle, and Dual Graphics support), H.265 decoding, full HSA 1.0 support, and ARM Trustzone compatibility. But perhaps one of the biggest advantages of Carrizo is the fact that the APU and Southbridge are now incorporated into the same die; not just two separates dies built into and MCM package.

This not only improves performance, but also allows the Southbridge to take advantage of the 28SHP process rather than older, more power-hungry 45nm or 65nm process nodes. In addition, the Excavator cores used in Carrizo have switched from a High Performance Library (HPL) to a High Density Library (HDL) design. This allows for a reduction in the die area taken up by the processing cores (23 percent, according to AMD). This allows Carrizo to pack in 29 percent more transistors (3.1 billion versus 2.3 billion in Kaveri) in a die size that is only marginally larger (250mm2 for Carrizo versus 245mm2 for Kaveri). When all is said and done, AMD is claiming a 5 percent IPC boost for Carrizo and a 40 percent overall reduction in power usage.

Comments

Re:Operating at 20W gives zero improvement.

[hairyfeet]

Just type "Cinebench rigs benchmarks" into the search engine of your choice friend, plenty of articles on the subject. You can also go to the channel of the first video I linked to, Tek Syndicate, as he has video up talking about it IIRC. You can trust the guys at Tek Syndicate as they buy their own gear and don't favor one manufacturer over the other.

You might also want to look up "Intel cripple compiler" as to this day their ICC still puts out code that will cripple any program compiled on it by using CPUID to detect AMD chips and turning off all SSE optimizations if AMD is detected. Considering that both chip makers have had SSE 1-3 for over a decade? Its a pretty big smoking gun. You can also look up "EU investigates Intel" for several articles with the charges laid out, it is shaping up to be as nasty as the OEM bribery scandal which netted AMD 1.2 billion in payoff from Intel.

I can't speak for the server market as I no longer do corporate but on the desktop? Its pretty dramatic, if you compare chips based on final unit cost then pretty much anything under $1K is gonna favor AMD by a fairly large amount as the money you save by going AMD can be used for more RAM, a bigger GPU, and often an SSD on top. I mean when you can pick up the FX8320 for $140 and if you shop around the FX8300 for less than $120? There really is no comparison, nothing Intel has at that price point even comes close.

BTW I put my money where my mouth is, as not only do I now sell AMD exclusively at the shop (nothing I hate more than bribery and dirty tricks) but my entire family is on AMD, with me and the oldest boy running Phenom II X6, the wife is happy with my hand me down Phenom II X4, the youngest rocking his games on the FX8300 and dad pulling up the rear with his Phenom I X4 he's had for nearly a decade because "Keep your mitts off my PC, it runs great so don't you dare touch it!" LOL. We couldn't be happier and despite the Phenom IIs being 6+ years old they play all the latest games without issue, hell the oldest is tearing into Shadows Of Mordor right now and its smooth as butter. This is why I have no trouble being an AMD only family, the bang for the buck is just nuts!