Behold Kaveri, in which CPUs and GPUs are equal 'compute unit' partners (click to enlarge)
Posted on 03/03/2014 11:48:29 AM PST by Ernest_at_the_Beach
AMD has released its long-awaited Kaveri processor, the first accelerated processing unit (APU) to incorporate both on-die CPU and GPU cores in a heterogeneous system architecture (HSA) with a shared memory architecture.
Behold Kaveri, in which CPUs and GPUs are equal 'compute unit' partners (click to enlarge)
"We're going to need to create a new term," AMD product CTO Joe Macri said at a Kaveri briefing, "because HSA has really changed what can be done on the graphics part of the die."
That new term, Macri explained, is compute cores. "The compute cores can either be CPU cores or graphics cores," he said. "We based them on the historical definition of what a CPU is," meaning that in an HSA system, each compute core – CPU or GPU – is programmable, capable of running at least one process in its own context in its own virtual memory space, and do so independently from all the other cores on the 2.41-billion transistor, 245mm2 die.
(Excerpt) Read more at theregister.co.uk ...
Can someone translate this??
To use 22 nm they need FinFET.
which may not be ready for Carrizo,,,,according to discussion on Semi-Accurate Forum.
Is this the recent TV ad for a tablet on which you can do four things at once? I think it was a Samsung ad.
Hmmm....just about ready to build a new machine to replace my aging but still useful Intel dual core 3.4ghz machine.
This will be out of my budget but other chips will drop.
Might have to wait a bit...
The chip can have up to 12 cores, a mix of CPU (used for regular tasks) of GPU (graphics processing units; used for rendering in games or video, and also for certain intensive calculating tasks).
In short, it can do a lot more stuff in parallel.
“of GPU” -> “or GPU”
Interesting that they actually say that 22nm will be slower.
Scaling below 28nm is REALLY hard.
Without finfet power goes up, performance drops.
Comments about it being a compromise indicates they are having problems getting power low enough for laptops.
Should work for desktop though.
ISA is Instruction Set Architecute,.
[ Can someone translate this?? ]
You need to bounce the graviton particle beam from the secondary nadion emiiter through the primary EPS conduiot into the Secondary plasma intake manifold via the Impulse generator bypass....
oh, why didn’t they just say so! Now I get it.
They now have seriously improved how they can work together.
With proper programming.
It is a very BIG DEAL for what can be done on a laptop.
I would make an awesome Starfleet engineer, it is all about BSing your way out of technical problems...
You forgot the gizmo. No technical explanation should be w/o a gizmo.
You need to bounce the graviton particle beam from the secondary nadion emiiter through the primary EPS conduiot into the Secondary plasma intake manifold via the Impulse generator bypass....
Thank you, Sheldon, that was very illuminating.
Many complain that does not have enough serious traditional CPU horsepower,...no much complaining about the serious graphics horsepower
If you can run flighty Simulators and ...you can understand this stuff.
Absolutely! For me graphics are pretty much secondary. But if Kaveri becomes the hot thing, maybe I can get a better I5.
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.