AMD Mobile Kaveri SKUs

The previous generation Trinity/Richland APUs came in four variants: at the top were the highest performance 35W TDP standard voltage (SV) parts, and then we dropped into the ultra-low voltage (ULV) range with several variations: 25W, 19W, and 17W TDPs were all available. With Kaveri, AMD is mostly consolidating their lineup into two classes: SV 35W TDP parts, and ULV 19W TDP parts. AMD did mention at one point that the 19W parts can be configured to target different TDPs, however, allowing the laptop OEM to target anywhere from 15W to 25W if desired. There's also at least one 17W TDP part, which we'll get to in a moment.

One other item that AMD emphasized was their new idea of "Compute Cores", which is a way for them to compare performance potential of CPU and GPU cores. GPUs typically have hundreds of cores that are good for very specific tasks whereas CPUs have a few cores that are general purpose, but the GPU functionality is becoming increasingly complex so AMD wanted a way to compare them. For the purposes of Kaveri (and GCN graphics cards), AMD has defined a Compute Core as being "any core capable of running at least one process in its own context and virtual memory space, independently from other cores." What that means is each GCN Compute Unit counts as a Compute Core, and each CPU thread (two per Steamroller module) counts as a Compute Core.

AMD 35W Standard Voltage Mobile APUs
 
Trinity
Richland
Kaveri
Model A10-4600M A8-5557M A10-5757M A8-7200P A10-7400P FX-7600P
Core Name Trinity Richland Richland Kaveri Kaveri Kaveri
Microarch Piledriver Piledriver Piledriver Steamroller Steamroller Steamroller
Modules/Cores 2/4 2/4 2/4 2/4 2/4 2/4
CPU Base Freq 2300 2100 2500 2400 2500 2700
Max Turbo 3200 3100 3500 3300 3400 3600
TDP 35W 35W 35W 35W 35W 35W
L1 Cache 128KB I$
64 KB D$
128KB I$
64 KB D$
128KB I$
64 KB D$
192 KB I$
64 KB D$
192 KB I$
64 KB D$
192 KB I$
64 KB D$
L2 Cache 2x2MB 2x2MB 2x2MB 2x2MB 2x2MB 2x2MB
Graphics HD 7660G HD 8550G HD 8650G R5 R6 R7
GPU Cores 384 256 384 256 384 512
GPU Clock 685 720 720 626 654 686
Max DDR3 1600 1600 1600 1866 1866 2133

Starting at the top with the 35W APUs, these will be the highest performance mobile parts. At launch there will be three 35W APUs: the "entry" A8-7200P, "mainstream" A10-7400P, and "enthusiast" FX-7600P. All three APUs use the P suffix to indicate that they’re 35W parts. And right away, we see some interesting changes from the Trinity/Richland lineup.

First, you're sure to notice the use of the FX branding. Make no mistake: this is the same APU as the other Kaveri parts and it has no relation to the desktop FX processors; AMD marketing simply feels the FX brand has a good reputation among enthusiasts and consumers and they wanted to carry that over into the mobile world. Of course this also solves the question of what to call the new highest-end APU; Llano had A4/A6/A8 while Trinity used A6/A8/A10; Kaveri will use A6/A8/A10/FX (so no A12 for now; sorry).

Looking at the core clocks, the base clocks haven't changed too much (2.4-2.7GHz with Kaveri compared to 2.1-2.5GHz on Richland), but Turbo Core clocks have gone up quite a bit. AMD noted that a lot of work went into tuning the mobile Kaveri APUs for power, with one of only four Corporate Fellows at AMD being in charge of that area. The result is that Kaveri should run closer to its Turbo clock in many situations, and the maximum clock speeds have increased from 3.0-3.5GHz on Trinity/Richland to 3.3-3.6GHz on Kaveri. That might not seem like a big deal at first, but keep in mind a couple of points. First, at the same clock speed the Steamroller cores in Kaveri should be about 15-20% faster than the Piledriver cores in Trinity/Richland, thanks to architectural improvements. Second, on the desktop Richland actually topped out at 4.1/4.4GHz while Kaveri only reaches 3.7/4.0GHz, so while Kaveri still enjoys architectural improvements it had to be clocked slower; we don’t see that with the mobile parts.

AMD's Joe Macri (Corp VP and Product CTO Global Business Unit) discussed some of the design decisions that went into Kaveri, noting that choosing the right type of transistor for an APU is different than building a pure CPU. At a high level, there are “V” shaped transistors frequently used with CPUs that can run at higher clock speeds, and “T” shaped transistors that tend to work better in the highly parallel design of GPU cores. From what I could gather, AMD used "V" transistors in Trinity/Richland but has switched to "T" transistors for Kaveri, which explains the drop in maximum clock speed. Joe also noted that 47% of the Kaveri core is dedicated to GPU, again highlighting the importance of the transistor choice.

Moving over to the GPU side of things, each of the 35W APUs comes with a different GPU configuration, using 256, 384, or 512 stream processors for the A8/A10/FX parts, respectively. GPU clock speeds top out at 686MHz with Kaveri compared to 720MHz with Richland, so this is another small step back but GCN’s architectural efficiency and increase in SPUs in the FX-7600P should more than compensate. That said, feeding these increasingly powerful GPUs becomes an increasingly difficult task, which is why maximum memory clock speeds are up to 1866MHz for the A8 and A10, and meanwhile the FX processor goes one further to 2133MHz. Given just how hard it is to feed a fully enabled APU like the FX-7600P – a problem we’ve already seen on the similarly configured desktop SKUs – the memory bandwidth increase is a welcome sight.

AMD 17W Ultra-Low Voltage Mobile APUs
 
Trinity
Richland
Kaveri
Model A4-4455M A4-4145M A6-5345M A6-7000
Core Name Trinity Richland Richland Kaveri
Microarch Piledriver Piledriver Piledriver Steamroller
Modules/Cores 1/2 1/2 1/2 1/2
CPU Base Freq 2100 2000 2200 2200
Max Turbo 2600 2600 2800 3000
TDP 17W 17W 17W 17W
L1 Cache 128KB I$
64 KB D$
128KB I$
64 KB D$
128 KB I$
64 KB D$
192 KB I$
64 KB D$
L2 Cache 2MB 1MB 1MB 1MB
Graphics HD 7500G HD 8130G HD 8410G R4
GPU Cores 256 128 192 192
GPU Clock 424 554 600 553
Max DDR3 1333 1333 1333 1600
AMD 19W Ultra-Low Voltage Mobile APUs
 
Trinity
Richland
Kaveri
Model A8-4555M A8-5545M A8-7100 A10-7300 FX-7500
Core Name Trinity Richland Kaveri Kaveri Kaveri
Microarch Piledriver Piledriver Steamroller Steamroller Steamroller
Modules/Cores 2/4 2/4 2/4 2/4 2/4
CPU Base Freq 1600 1700 1800 1900 2100
Max Turbo 2400 2700 3000 3200 3300
TDP 19W 19W 19W 19W 19W
L1 Cache 128KB I$
64 KB D$
128KB I$
64 KB D$
192 KB I$
64 KB D$
192 KB I$
64 KB D$
192 KB I$
64 KB D$
L2 Cache 2x2MB 2x2MB 2x2MB 2x2MB 2x2MB
Graphics HD 7600G HD 8510G R5 R6 R7
GPU Cores 384 384 256 384 384
GPU Clock 424 554 514 533 553
Max DDR3 1333 1333 1600 1600 1600

Moving on to the 17W/19W parts, there's apparently a single 17W APU, the A6-7000, along with three 19W APUs. (AMD didn't provide details on the A6-7000 at the briefing, but we've since confirmed the above specifications, and there's also a Pro equivalent -- see below.) Richland had two 17W parts and one 19W part, so AMD has sort of flipped roles here. As expected the 17W A6-7000 is a rather lean chip; it has one Steamroller module and 192 GCN steaming processors (for a total of five Compute Cores, if you're counting), with a base/turbo CPU clock speed of 2.2GHz/3.0GHz while the GPU turbo clock stands at 553MHz. Compared to the previous generation Richland processors, the A6-7000 gains all of Kaveri’s architectural improvements along with an additional 200MHz for the maximum CPU turbo clock. GPU clock speeds on the other hand take a hit, but this is offset by GCN’s greater performance and a badly needed increase in the maximum DDR3 memory clock speed. In fact with mobile Kaveri, DDR3-1600 is now the baseline, with all processors supporting 1600 or better.

Meanwhile the 19W parts all have two Steamroller modules, and maximum base/turbo clock speeds only differ by 300MHz; most of the differentiation comes in the GPU department. The A8-7100 includes four GCN CUs, the A10-7300 has six CUs with slightly higher GPU clocks, and the FX-7500 also has six CUs with another moderate bump in clock speed. AMD has also brought along the "R-series" branding for the GPUs, so the A6-7000 gets an R4 GPU, the A8 is an R5, A10 gets an R6, and the FX has an R7 GPU. Maximum GPU clocks are again down in some cases compared with Richland, though architectural difference should more than cover any loss in clock speed.

AMD 17/19W Commercial Pro Series ULV APUs
Model A6 Pro-7050B A8 Pro-7150B A10 Pro-7350B
Core Name Kaveri Kaveri Kaveri
Microarch Steamroller Steamroller Steamroller
Modules/Cores 1/2 2/4 2/4
CPU Base Freq 2200 1900 2100
Max Turbo 3000 3200 3300
TDP 17W 19W 19W
L1 Cache 192 KB I$
64 KB D$
192 KB I$
64 KB D$
192 KB I$
64 KB D$
L2 Cache 1MB 2x2MB 2x2MB
Graphics R4 R5 R6
GPU Cores 192 384 384
GPU Clock 533 533 553
Max DDR3 1600 1600 1600

There's one final category of APUs, which appear to be AMD's equivalent of Intel's SIPP (Stable Image Platform Program) CPUs: the AMD Pro Series. There are three Pro APUs, the A10 Pro-7350B, A8 Pro-7150B, and A6 Pro-7050B. These APUs are functionally equivalent to the FX-7500, A10-7300, and A6-7000 respectively. The Pro series targets business customers with a message of commercial stability and management. AMD guarantees that these APUs will remain available for an extended period of time, so enterprise customers won't need to worry about validating new hardware for a couple years.

Besides offering different levels of performance, AMD is also differentiating their Kaveri APUs based on other features. Eyefinity and TrueAudio support will be limited to the A10 and FX APUs; the A6/A8 APUs lose this functionality. Similarly, the A6 does not have Dual Graphics functionality; I didn't have much luck with Dual Graphics on Richland/Trinity laptops, but with the iGPU and dGPU both being GCN architectures now there's at least more potential to extract additional performance through CrossFire. Finally, note that the A6 APUs get ARM TrustZone functionality, whereas the higher-end A8, A10, and FX APUs do not.

All of the above was discussed previously (and then pulled), but we have something new to add with today's launch: a performance preview.

Introducing AMD's Mobile Kaveri APUs AMD Kaveri FX-7600P System/CPU Performance Preview
Comments Locked

125 Comments

View All Comments

  • rhx123 - Wednesday, June 4, 2014 - link

    I am really confused about the Acer, as according to Ark, the 4500U has no PCIE 16x lane output, how is the 750M attached?
    http://ark.intel.com/products/75460/Intel-Core-i7-...
  • TheinsanegamerN - Wednesday, June 4, 2014 - link

    It has a single x4 link and dual x2 ink. the 750m would be connected by the x4 link.
  • coburn_c - Wednesday, June 4, 2014 - link

    The GPU isn't good enough to game with and the CPU isn't good enough to do anything else with. You'll get half the battery life of the competition and it will get twice as hot. Abject failure.
  • frozentundra123456 - Wednesday, June 4, 2014 - link

    Yes, you hit on the crux of the matter. Problem is still that of all APUs, the igp performance is still in limbo: almost there but not quite for gaming, while cpu performance and power consumption trail intel badly. Looking at the gaming tests, looks like one will be limited to 768p, and even then COH (admittedly terribly optimized) is not playable, and we have no tests for demanding games like Watchdogs, Metro LL, or Crysis 3. So one will be stuck with either a crappy 768p screen or playing at non-native resolution for a lot of games. Not to mention, this mobile chip should have come out first instead of being so close to the broadwell mobile launch.
  • silverblue - Wednesday, June 4, 2014 - link

    You say "cpu performance and power consumption trail intel badly". There's no real data on the latter apart from TDP values (which, in themselves, aren't an accurate indicator), and the FX-7600P is certainly no slouch when viewing the CPU benchmarks in this preview. Let's not forget that large portion of the die dedicated to the GPU, either - just because it's there, doesn't mean it's being used. Toms had the FX-7600P doing well at GPGPU (it should) and PCMark, less well at physics scores in 3DMark (the competition was a high-end i7) and even worse in Sandra, but we know that doesn't always mean everything; after all, the Core 2 had low memory bandwidth due to having no IMC but was still easily the match of the Phenom II series.

    This is a preview, so quite why people are expecting numerous tests on 2013/14 titles from a very brief testing period is beyond me. Patience, grasshopper!
  • Novaguy - Thursday, June 5, 2014 - link

    But I suspect you can hit 1080p with an awful lot of 2007~2012 games, making kaveri mobile + 1080p a great humble bundle mobile gaming solution. Now, if they combined it with freesync, you could get away with lower frame rates and play even more recent games.
  • The_Assimilator - Thursday, June 5, 2014 - link

    The use case for a laptop is simple: you don't require graphics horsepower the majority of the time. Hence it makes more sense to use a CPU that uses less power (and performs better), and pair it with a discrete GPU that won't be drawing any power most of the time. Intel understands this and it's why they make money.

    AMD on the other hand, have dreamed up a mythical future reality where everyone who buys a laptop expects it to deliver decent GPU performance. But very few people buy laptops to play games on, they buy laptops to be productive while mobile - and faster CPU that use less battery power are always going to win out.
  • nemi2 - Wednesday, June 4, 2014 - link

    It's becoming more of a glaring omission as time progresses that AMD doesn't have a SSD caching option in their SATA BIOS/drivers. AMD is targeting the cheaper end of notebooks so I agree with the author: a pure SSD storage solution will not make the BOM, and because there is no AMD SSD caching solution a 8,16,32GD SSD cache can't make the BOM either - discrete chece is not supported and a hybrid HD unfortunately also seem to also command a $ premium and don't have as good integration with windows.
    Makes me wonder what the state of USB "ready boost" is with Windows these days?, things are very quiet on the that front. Also where is a cheap / free 3rd party utility to provide SSD/ NVM caching, do AMD laptop vendors have any budget / driver solution?
  • lmcd - Wednesday, June 4, 2014 - link

    The SSD caching probably explains it -- thanks for posting (not an obvious conclusion to reach!)
  • JarredWalton - Wednesday, June 4, 2014 - link

    SSD caches are really only good for two things:

    1) Inflate certain benchmark scores (PCMark being a major one).
    2) Help laptop boot/resume faster.

    Seriously, most other tasks that hit the storage hard -- installing an application, loading a bunch of apps at once, opening a browser with 30 active tabs -- are all only slightly faster than an HDD with SSD caching, where a pure SSD is substantially faster. The best SSD caching solution right now comes from Apple, and it's only good because it's 64GB or more; 24/32GB SSD caches just don't cut it in my experience.

Log in

Don't have an account? Sign up now