[Anand] Chipworks opens up the Apple A8 SOC - finds interesting new designs

[LukaP]

http://www.anandtech.com/show/8562/chipworks-a8

 

 

 

A close analysis of the die shot makes it clear that there are only 4 GPU cores available and not 6, which immediately rules out the 6 core GX6650 we were previously expecting. Instead with 4 cores present this is conclusive proof that Apple is using the smaller 4 core GX6450 on A8, the direct successor to the G6430 used on the A7. GX6450 induces some performance optimizations along with some feature updates – including ASTC support, which Apple’s documentation has already confirmed is present – so its inclusion here is a natural progression for Apple.

So they do use a new GPU, but still only a 4core model.

 

 

 

 

Meanwhile opposite the GPU we have the CPU block. Unlike the GPU the CPU block has seen some significant shrinking, which Chipworks estimates is down from 17.1mm2 in A7 to 12.2mm2 in A8. In A7 Cyclone did not lend itself to easily picking apart the individual CPU cores, and neither does the CPU here in A8. We’ll be looking at the new CPU’s architecture in-depth in our iPhone 6 review, but for now it’s safe to say that while this is definitely derived from Cyclone, Apple has added a few tweaks over the last year that make it an even more potent CPU than the first Cyclone. Meanwhile based on this die shot Chipworks believes that the L2 cache has been reorganized to a per-core design, as there is no obvious single block of L2 on A8 like there was A7.

The CPU area is indeed very different to what we are used to, with no real boundaries between cores, or any distinct indicator of a shared cache at all. Though looking at it, it is obviously mirrored horisontally, so there are 2 cores, but i cant make out any specific logic inside.

 

There is still 4MB of static ram cache, acting as a joint L3 between teh GPU and the CPU cores.

 

 

 

Overall, Chipworks’ analysis points to A8 being fabbed on TSMC’s 20nm process. This makes A8 among the first SoCs to receive the 20nm treatment. Thanks to this smaller node Apple has been able to build in additional features to the SoC while simultaneously shaving off around 15% of their die size. Chipworks estimates the final die size of A8 to stand at 89mm2, versus the 104mm2 for the Samsung 28nm based A7. Chipworks notes that if this were a straight shrink that one would expect the A8 to be closer to 50% the size of A7 (though not all logic can shrink quite that well), which indicates that Apple has spent quite a bit of die size on improving performance through more complex CPU and GPU architectures and miscellaneous feature additions.

 

This is a very interesting chip, especially how they redesigned their CPU cores, with L2 being divided amongst them. we shall see how that fares in power efficiency, since now, when a core shuts down, half of the cache also shuts down.

 

 

Also yaay, 3k posts, time to go do my thing ^^

[JAKEBAB]

I did not know apple had anything to do with amd this is somewhat interesting.

[ionbasa]

I did not know apple had anything to do with amd this is somewhat interesting.

They dont... Chipworks just did the die shots a little bit of reverse engineering.

[LukaP]

I did not know apple had anything to do with amd this is somewhat interesting.

Nono, this is Apples SOC. based on ARM, called A8

[JAKEBAB]

They dont... Chipworks just did the die shots a little bit of reverse engineering.

 

 

Nono, this is Apples SOC. based on ARM, called A8

you guys are clearly smart lol can u explain this whole arm thing to me please

[LukaP]

you guys are clearly smart lol can u explain this whole arm thing to me please

Its just a different way to handle processing. basically the cpu gets instructions on how to work, and arm uses different ones from x86. they are shorter in lenght, and less complex

[JAKEBAB]

Its just a different way to handle processing. basically the cpu gets instructions on how to work, and arm uses different ones from x86. they are shorter in lenght, and less complex

ok so instead of using x86 instructions it uses arm? why is it always x86 and not x64? and what are the advanatages of using arm over x86?

[LukaP]

ok so instead of using x86 instructions it uses arm? why is it always x86 and not x64? and what are the advanatages of using arm over x86?

yes it uses arm IS. x86 was the original instruction set that is still somewhat in use. then we got x86-64 extention(commonly known as x64) and then we started getting better, more efficient instruction set extentions, for more specific uses (MMX-media related, SSE-floating point and 3d operations, AVX-vector math,...).

 

well x86 uses longer instructions, which also makes the cpu executing them, inherently more powerful to even be able to work them. in theory, ARM (using a RISC concept) is just as powerful as x86, if correctly coded for, but with arm cpus getting faster and faster, the IS they use is getting extended into longer and longer sets, so it starts to resemble x86 alot now.