My discussion of the Carrizo architecture went up a couple of days ago. The post did not include specific SKUs because we did not have those at the time. Now we do, and there will be products: one A8-branded, one A10-branded, and one FX-branded.
All three will be quad-core parts that can range between 12W and 35W designs, although the A8 processor does not have a 35W mode listed in the AMD Dual Graphics table. The FX-8800P is an APU that has all eight GPU cores while the A-series APUs have six. The A10-8700P and the A8-8600P are separated by a couple hundred megahertz base and boost CPU clocks, and 80 MHz GPU clock.
Also, we have been given a table of AMD Radeon R5 and R7 M-series GPUs that can be paired with Carrizo in an AMD Dual Graphics setup. These GPUs are the R7 M365, R7 M360, R7 M350, R7 M340, R5 M335, and R5 M330. They cannot be paired with every Carrizo APU, and some pairings only work in certain power envelopes. Thankfully, this table should only be relevant to OEMs, because end-users are receiving pre-configured systems.
Pricing and availability will depend on OEMs, of course.
Cheap 35W APUs might make
Cheap 35W APUs might make sense as Intel left this category. Dual graphics would be competitive only when AMD CPU has better thermals then Intel so not going to happen anytime soon.
I do not want this APU to
I do not want this APU to only be offered in thermally constrained “Ultrabook” style laptops, put it in a regular laptop and allow it the full 35 watts, the possible discrete mobile GPU pairing needs to be more equal in GPU resources to the FX-8800P’s GPU, or a little higher like offering an FX-8800P paired with a discrete R9 M360 or above. Hopefully this dual graphics is only restricted via ANDs driver based dual graphics, and not Microsoft’s multi-adaptor based DX12 graphics API, as DX12’s multi-adaptor appears to be able to pair GPUs of various asymmetrical GPU processing levels/abilities and properly load balance the rendering loads according to each GPU’s processing abilities.
I hope that PcPer will research DX12’s multi-adaptor abilities to its fullest, and compare it to AMD’s driver based dual GPU mode for gaming laptops, as AMD appears to be artificially restricting its dual mobile with Carrizo parings more to its weaker mobile discrete offerings probably as to not unnecessarily compete with AMDs top mobile discrete offerings. In this respect Microsoft’s multi-adaptor support in DX12 appears to be the more functional in allowing pairings of any GPUs from any manufacturer, and that includes properly load balancing weak integrated GPUs with stronger discrete GPU of any make and model. DX12 allowing more varied paring of whatever GPU resources are available on a computing system, than either AMD or Nvidia, appears to be doing something that could be considered very HSA friendly, in allowing all GPU resources of what ever power level/manufacturer’s make or model to work together for graphics workloads. AMD needs to make sure that Mantle/Vulkan has just as much flexibility in allowing for any make and model GPU to be utilized, and also properly load balanced according to its GPU processing ability like DX12’s Multi-Adaptor API is allowing. I see no reason that rendering loads can not be 40/60, or 20/80, integrated to discrete or any load balance between 0 and 100, integrated to discrete for any 2 GPUs, or more than 2 GPUs if they are present on a computing platform. This having to have evenly matched GPUs of any variety is just and artificial restriction, that can be abstracted away with proper software on both graphics APIs, and drivers that work together to properly load balance to whatever GPU execution resources are available.
I similarly would love it if
I similarly would love it if pcper did this, it would be an interesting topic.
Why have three different
Why have three different über-brands, when these are already differentiated by their numbers alone? And “A10” is now a salvaged die, too?
Then again, if history is any indication, none of these will come to the market in any appreciable numbers anyway, and the dies will get their functionality cut down further to be sold off for $6 a pop to one of the HPs.
Id like to see an affordable
Id like to see an affordable 17″ 100hz IPS laptop with one of these… and NO discrete videocard.
That would be a dream portable work/light gaming machine.
But i bet reality is gona be the same as before – some small 60hz TN laptops, full with all kinds of bloat and expensive. And not much options really.
With all the new chip
With all the new chip announcements lately, I’d really like to see some convertible style laptops come out that have good battery life and can still do some low demand gaming.
Being able to play Heroes of the Storm on such a device would be ideal, but might still be too much to ask. Games like Hearthstone and League of Legends would also be welcome.
The for factor of the Lenovo Flex 3 14″ looks interesting, but I’m not sure how well that performs.
Lenovo already announced that
Lenovo already announced that they will use Carrizo parts in their new Flex 3. It would also be the natural step, since the Flex 1 and Flex 2 had AMD models too (with the letter ‘D’ attached). Moreover, there are no other recent AMD parts Lenovo could use. The Flex 2 14D and 15D used Beema parts (I own a 14d with the top A8-6410) but the successor to that series isnt ready yet. Besides, Beema is designed for smaller devices, Carrizo would be more proper in 14 and 15 inch formfactors.
I’d really wish for a 14 inch Lenovo Flex 3 with the FX8800p chained to a dedicated R7 M360. For prices around 500 dollar, it would be a great upgrade to my current notebook. An IPS FHD screen would be good, but I learned to hope for that with OEMS. Also, I’m a bit frightened they will not use the full 35 watt potential of the Carrizo’s, because they will have to use more expensive cooling solutions..
So wheres the FX
So wheres the FX steamrollers?