Bookmark and Share


Advanced Micro Devices on Tuesday unveiled additional details about its forthcoming code-named A-series “Kaveri” accelerated processing unit and also confirmed plans to launch them on January 14, 2014. As it appears, Kaveri will fully support all AMD’s latest technologies, including unified CPU-GPU memory, TrueAudio, DirectX 11.2, Mantle as well as HSA [heterogeneous system architecture].

AMD A-series “Kaveri” will feature two dual-core x86 Steamroller modules (two integer units per module), one FPU with two 128-bit FMACs, one MMX unit, Radeon R7 graphics processing unit with 512 stream processors (eight GCN compute units) as well as HUMA [heterogeneous unified memory architecture] shared memory controller. The total compute performance of AMD A10-7850K (3.7GHz x86 core clock-rate, 720MHz graphics engine frequency) “Kaveri” APU will be around 856GFLOPS, which is quite a lot.

The new series of APUs will support all the technologies that AMD introduced along with its latest family of graphics processors – which feature improved second-gen GCN architecture – including TrueAudio as well as DirectX 11.2 and Mantle application programming interfaces.

One of the key innovations of Kaveri is its HUMA shared memory controller that fetches data to both x86 and GCN processing units as well as acts like a dispatcher/hub. The unit supports heterogeneous queuing that defines how processors interact equally and enables uniform visibility into the entire memory space by both GPU and CPU.

Unlike in cases of previous-generation Fusion products, AMD will first release desktop versions of the accelerated processing units. The first bunch of such FM2+ chips will emerge on January 14, 2014, around two months from now. Later on Kaveri-like chips will be available inside notebooks, servers and embedded systems. Revenue shipments of the chips will begin this quarter.

Tags: AMD, Fusion, Kaveri, Radeon, GCN, HSA, TueAudio, Mantle, Steamroller


Comments currently: 24
Discussion started: 11/13/13 08:42:55 PM
Latest comment: 11/16/13 09:51:56 PM
Expand all threads | Collapse all threads


We know the theory, now we need to see the Windows 8.1 and Linux (ie with HSA support) benchmarks to see how this kicks. No doubt AMD will wait until CES Jan 7 - 10 in the new year to release these figures, with the timing strategic post holiday season. But I feel they should have released some tid bits to start a spark to the motherboard sales. For example, how about a Handbrake x264 second pass bench?
3 2 [Posted by: linuxlowdown  | Date: 11/13/13 08:42:55 PM]
- collapse thread

I'm more interested first pass than second pass. Second pass is only used in ABR (average bitrate) encoding where you trade video quality for predictable file size. Anything else uses first pass.
0 0 [Posted by: trumpet-205  | Date: 11/14/13 01:22:11 AM]
Second pass is where the heavy number crunching and video quality refinement happens. The second pass transcode process really shows up the difference in floating point calculating power between APUs. This is where HSA will shine. I would put down money that Kaveri will out do Intel's finest in this department.
1 1 [Posted by: linuxlowdown  | Date: 11/14/13 01:56:08 AM]
ABR can both be done in first pass and second pass. When doing second pass the first pass becomes analyzing video. Second pass takes the result from first pass and determine which part to compress more which part to compress less. There is no refinement per say.

If you do ABR with first pass only it compress video on the go.

If you care about video quality you would do CRF or QP encoding instead. Which encodes each video frame based on quality factor. Since there is no need to maintain an average bitrate there is no need for second pass.
0 0 [Posted by: trumpet-205  | Date: 11/14/13 11:55:06 AM]
I used to do a lot of transcoding in my late-twenties circa 2000. But you've shown I'm not completely up to date with this anymore. I only do a transcode once a year now for the Sydney NYE fireworks TV broadcast (that runs about 20 minutes), and put it up as a torrent on Pirate Bay. My challenge every year is to see how fast I can get it posted after the fireworks end (that's if I'm not there in person, of course). I'm hoping Kaveri + Handbrake (GPU accelerated version) will break my previous records.
2 1 [Posted by: linuxlowdown  | Date: 11/15/13 07:54:31 AM]

5 1 [Posted by: Tukee44  | Date: 11/13/13 09:38:32 PM]
- collapse thread

Do you really think AMD is going to show its Ace card to Intel and Nvidia just yet? I will wait for CES announcements in January before I believe this is their most powerful APU at 95W TDP for 2014. I suspect there is a big cat and mouse game going on with this.
1 2 [Posted by: linuxlowdown  | Date: 11/14/13 02:10:36 AM]

AMD must survive to balance the monopoly from Intel & nVidia....
3 1 [Posted by: tks  | Date: 11/13/13 10:19:12 PM]

what....856GFlops amd said kaveri will hit 1TFlops means 15% performance loss,,,,,,,,, something went wrong or anything else?
1 1 [Posted by: mudi1  | Date: 11/14/13 12:10:24 AM]

Kaveri won't disappoint anyone but the AMD haters. Intel is sure to copy or lose even more market share.
2 2 [Posted by: beenthere  | Date: 11/14/13 08:12:12 AM]


Add your Comment

Related news

Latest News

Wednesday, November 5, 2014

10:48 pm | LG’s Unique Ultra-Wide Curved 34” Display Finally Hits the Market. LG 34UC97 Available in the U.S. and the U.K.

Wednesday, October 8, 2014

12:52 pm | Lisa Su Appointed as New CEO of Advanced Micro Devices. Rory Read Steps Down, Lisa Su Becomes New CEO of AMD

Thursday, August 28, 2014

4:22 am | AMD Has No Plans to Reconsider Recommended Prices of Radeon R9 Graphics Cards. AMD Will Not Lower Recommended Prices of Radeon R9 Graphics Solutions

Wednesday, August 27, 2014

1:09 pm | Samsung Begins to Produce 2.13GHz 64GB DDR4 Memory Modules. Samsung Uses TSV DRAMs for 64GB DDR4 RDIMMs

Tuesday, August 26, 2014

10:41 am | AMD Quietly Reveals Third Iteration of GCN Architecture with Tonga GPU. AMD Unleashes Radeon R9 285 Graphics Cards, Tonga GPU, GCN 1.2 Architecture