Top Richland 28nm APU is A10 6800K, DDR3 2133 support
The successor to AMD's Trinity line of APUs, codenamed Richland, just got a name. It wont surprise many that the top Richland processor will get the A10 6800K moniker. It will succeed AMDs current flagship APU, the A10 5800K, and continue the similar naming convention.
The A10 6800K is a quad-core APU with DirectX 11 support and it will fit well known 100W TDP envelope. It continues to support FM2 motherboards and the rest of AMDs Trinity infrastructure. The K suffix indicates that the processor will come with an unlocked multiplier and it will offer easy overclocking for the willing. Naturally Richland A-series APUs do support AMD Turbo core and its graphics is based on Radeon HD 8000 series.
The expected launch date is June, but the production should be ready by late March 2013. Production candidate samples were scheduled for early December, while late January should be the time for production ready samples. The Richland generation currently consists of four quad-cores and two dual-core processors, but at this time we still dont know the final clock speeds. We can just pass on that there are 100W and 65W TDP parts, in both dual and quad-core flavours.
Top Kabini 28nm APU is Kabini X4 5110
Kabini is AMDs next generation entry level APU set to replace Brazos and Brazos 2.0 in the essential desktop market. Kabini 28nm will come in two flavours, X Series and E Series. The top notch processor is called X4 5110 and it rocks a TDP of 25W. There will be some 18W parts as well. They all use Jaguar cores, have DirectX 11.1 graphics and use the FT3 BGA infrastructure. The Kabini APU branded as X4 5110 supports DDR3 1866 as well as Turbo Core.
You will need a board with Yangtze FCH chipset to get it going. Our sources tell us that the graphics that will get this chip going is branded as HD 8310G. Currently we know that the launch is scheduled for June but some additional information that we gathered so far might indicate that the X4 5110 might come out even before the end of Q1 2013.
Sources close to AMD indicate that production ready silicon should be ready in March 2013, or very late Q1 2013, but full volume production with hundreds of thousand, or millions of chips is more likely to happen in June.
Unfortunately we dont have exact clock speeds just yet, and this part of the spec will probably be revealed as the launch date draws closer.
Kabini is AMDs next generation entry level APU set to replace Brazos and Brazos 2.0 in the essential desktop market. Kabini 28nm will come in two flavours, X Series and E Series. The top notch processor is called X4 5110 and it rocks a TDP of 25W. There will be some 18W parts as well. They all use Jaguar cores, have DirectX 11.1 graphics and use the FT3 BGA infrastructure. The Kabini APU branded as X4 5110 supports DDR3 1866 as well as Turbo Core.http://www.fudzilla.com/home/item/29934-top-kabini-28nm-apu-is-kabini-x4-5110
@ mrmt
Where do you see Kabini being only 18W/25W product?
Even with Bobcat and Trinity you have certain overlaps and Bobcat really cannot compete with Trinity in the same TDP bracket (maybe it draws a bit less power but perf. wise it's still slower).
Same will go for Kabini/Richland. Kabini's cores might be better per clock but its power and clock target are lower and its GPU would be a noticeably slower.
They CAN make GPU-less Trinity. From various reviews we see that at most Vishera can be 15% faster than L3-less Piledriver bu this is corner case. In reality the difference is closer to ~4-5% on average
(...)
Imagine if AMD had access to intel's money and fabs: PD or even better SR on 22nm intel-level fab tech with hand optimized layouts and 8 different dies(2/4M with and without L3 for server/desktop;
15% can be substantial for workloads where they aimed Bulldozer in the 1st place. There you need that L3 cache,unlike desktop or mobile where you don't(and Trinity therefore has none).
If AMD had access to intel's funds the race would be much closer and delays would be much rarer occurrences. With money they have they compete(or try at least) very well.
- can't ship a GPU less Trinity as that would cannibalize Vishera which would be more than twice the die size and even worse yields in all l.
Found a Video from HC24 on Jaguar presentation . Interesting Q&A (I embedded the time,just click). Head of the Jaguar project basically states they achieved >10% mircoarchitectural frequency gain versus Bobcat if it was done on 28nm . That was very surprising to me since what we heard before from sites covering HC was just : >15% singlethread IPC gain(~2x fp throughput), 2x more cores, >10% clock gain (assumed versus 40nm Bobcat). Now this new info from the guy in the video is revealing in a way that we have no clue at what clock would a 28nm Bobcat run. It could be the same 1.7Ghz as top product was running at or it could have been significantly more. Also in the very presentation the fellow mentioned that by doing various optimizations to the added die area(more xtors) they've increased the die size of Jaguar by about 10% versus a Bobcat if it was done on 28nm (also a very interesting information).
So new information (to me at least): core is 10% bigger than what 28nm Bobcat would have been if it was on 28nm ; clock gain by adding stage here and there(decoder and fp scheduler portions of pipe) cost them a bit IPC(didn't say how much,probably a few %s) while providing a >10% clock gain to the design. So a question is : how high does it clock in the end and how much power did they save(for Temash). Bobcat at 28nm would certainly clock much higher than 40nm Bobcat and/or have higher Turbo boost state at least. This new Jaguar core may very well significantly outclock 40nm Bobcat(20+%?) while performing much better at the same clock(SSE stuff would see major uplift,basically K10 level if not better).
In the end we may have pretty potent chip in Jaguar for even a desktop user,with decent GPU and decent CPU. Since it was supposed to be launching in the same year alongside SR core performance delta would have been in the favor of later(but by how much). As it is now, IF Richland is 28nm it could save it in direct comparison with this little fella,but even if Richland is a Piledriver+,from IPC standpoint(by as much as 5%), Jaguar would probably still own it in IPC department . Fun times ahead!
Found a Video from HC24 on Jaguar presentation . Interesting Q&A (I embedded the time,just click). Head of the Jaguar project basically states they achieved >10% mircoarchitectural frequency gain versus Bobcat if it was done on 28nm . That was very surprising to me since what we heard before from sites covering HC was just : >15% singlethread IPC gain(~2x fp throughput), 2x more cores, >10% clock gain (assumed versus 40nm Bobcat). Now this new info from the guy in the video is revealing in a way that we have no clue at what clock would a 28nm Bobcat run. It could be the same 1.7Ghz as top product was running at or it could have been significantly more. Also in the very presentation the fellow mentioned that by doing various optimizations to the added die area(more xtors) they've increased the die size of Jaguar by about 10% versus a Bobcat if it was done on 28nm (also a very interesting information).
So new information (to me at least): core is 10% bigger than what 28nm Bobcat would have been if it was on 28nm ; clock gain by adding stage here and there(decoder and fp scheduler portions of pipe) cost them a bit IPC(didn't say how much,probably a few %s) while providing a >10% clock gain to the design. So a question is : how high does it clock in the end and how much power did they save(for Temash). Bobcat at 28nm would certainly clock much higher than 40nm Bobcat and/or have higher Turbo boost state at least. This new Jaguar core may very well significantly outclock 40nm Bobcat(20+%?) while performing much better at the same clock(SSE stuff would see major uplift,basically K10 level if not better).
In the end we may have pretty potent chip in Jaguar for even a desktop user,with decent GPU and decent CPU. Since it was supposed to be launching in the same year alongside SR core performance delta would have been in the favor of later(but by how much). As it is now, IF Richland is 28nm it could save it in direct comparison with this little fella,but even if Richland is a Piledriver+,from IPC standpoint(by as much as 5%), Jaguar would probably still own it in IPC department . Fun times ahead!
That slide tells us nothing about 28nm Bobcat . What they say is basically >10% gain. Of course it's more than 10% gain, but 15% is >10% and 35% is also more than 10% . The new information is that >10% refers to 28nm Bobcat ,not the one we have today(and the lead of Jaguar project states so in the video,in no uncertain terms).......
That slide tells us nothing about 28nm Bobcat . What they say is basically >10% gain. Of course it's more than 10% gain, but 15% is >10% and 35% is also more than 10% .
If you are trying to sell a product that can be 15% faster than your product, you'll put on the slide ~15% or >14%, not >10%. Same for 35% or 200%.
If they are putting 10% on the slides then you can expect 10% at most.
The guy in the Video is Jeff Rupley, Chief Architect for Jaguar not a salesman.
Hes not trying to sell the CPU, hes trying to hid information for a product not yet released(information is under NDA).
So, 10% more than a Bobcat at 28nm doesn't mean 10% more than current Bobcat at 40nm
When I say sell, I mean sell in a broader sense. John Fruehe never sold you a processor, but a lot of people were wishing for a Bulldozer after reading his posts.
As for the 28nm bobcat, the results shouldn't be very good. If 28nm bobcat were that good a cash strapped company like AMD would have gone that route without thinking too much. I.stead they went for the hard route, improving the uarch.