The iPhone XS & XS Max Review: Unveiling the Silicon Secrets
by Andrei Frumusanu on October 5, 2018 8:00 AM EST- Posted in
- Mobile
- Apple
- Smartphones
- iPhone XS
- iPhone XS Max
The A12 Vortex CPU µarch
When talking about the Vortex microarchitecture, we first need to talk about exactly what kind of frequencies we’re seeing on Apple’s new SoC. Over the last few generations Apple has been steadily raising frequencies of its big cores, all while also raising the microarchitecture’s IPC. I did a quick test of the frequency behaviour of the A12 versus the A11, and came up with the following table:
Maximum Frequency vs Loaded Threads Per-Core Maximum MHz |
||||||
Apple A11 | 1 | 2 | 3 | 4 | 5 | 6 |
Big 1 | 2380 | 2325 | 2083 | 2083 | 2083 | 2083 |
Big 2 | 2325 | 2083 | 2083 | 2083 | 2083 | |
Little 1 | 1694 | 1587 | 1587 | 1587 | ||
Little 2 | 1587 | 1587 | 1587 | |||
Little 3 | 1587 | 1587 | ||||
Little 4 | 1587 | |||||
Apple A12 | 1 | 2 | 3 | 4 | 5 | 6 |
Big 1 | 2500 | 2380 | 2380 | 2380 | 2380 | 2380 |
Big 2 | 2380 | 2380 | 2380 | 2380 | 2380 | |
Little 1 | 1587 | 1562 | 1562 | 1538 | ||
Little 2 | 1562 | 1562 | 1538 | |||
Little 3 | 1562 | 1538 | ||||
Little 4 | 1538 |
Both the A11 and A12’s maximum frequency is actually a single-thread boost clock – 2380MHz for the A11’s Monsoon cores and 2500MHz for the new Vortex cores in the A12. This is just a 5% boost in frequency in ST applications. When adding a second big thread, both the A11 and A12 clock down to respectively 2325 and 2380MHz. It’s when we are also concurrently running threads onto the small cores that things between the two SoCs diverge: while the A11 further clocks down to 2083MHz, the A12 retains the same 2380 until it hits thermal limits and eventually throttles down.
On the small core side of things, the new Tempest cores are actually clocked more conservatively compared to the Mistral predecessors. When the system just had one small core running on the A11, this would boost up to 1694MHz. This behaviour is now gone on the A12, and the clock maximum clock is 1587MHz. The frequency further slightly reduces to down to 1538MHz when there’s four small cores fully loaded.
Much improved memory latency
As mentioned in the previous page, it’s evident that Apple has put a significant amount of work into the cache hierarchy as well as memory subsystem of the A12. Going back to a linear latency graph, we see the following behaviours for full random latencies, for both big and small cores:
The Vortex cores have only a 5% boost in frequency over the Monsoon cores, yet the absolute L2 memory latency has improved by 29% from ~11.5ns down to ~8.8ns. Meaning the new Vortex cores’ L2 cache now completes its operations in a significantly fewer number of cycles. On the Tempest side, the L2 cycle latency seems to have remained the same, but again there’s been a large change in terms of the L2 partitioning and power management, allowing access to a larger chunk of the physical L2.
I only had the test depth test up until 64MB and it’s evident that the latency curves don’t flatten out yet in this data set, but it’s visible that latency to DRAM has seen some improvements. The larger difference of the DRAM access of the Tempest cores could be explained by a raising of the maximum memory controller DVFS frequency when just small cores are active – their performance will look better when there’s also a big thread on the big cores running.
The system cache of the A12 has seen some dramatic changes in its behaviour. While bandwidth is this part of the cache hierarchy has seen a reduction compared to the A11, the latency has been much improved. One significant effect here which can be either attributed to the L2 prefetcher, or what I also see a possibility, prefetchers on the system cache side: The latency performance as well as the amount of streaming prefetchers has gone up.
Instruction throughput and latency
Backend Execution Throughput and Latency | ||||||||
Cortex-A75 | Cortex-A76 | Exynos-M3 | Monsoon | Vortex | |||||
Exec | Lat | Exec | Lat | Exec | Lat | Exec | Lat | |
Integer Arithmetic ADD |
2 | 1 | 3 | 1 | 4 | 1 | 6 | 1 |
Integer Multiply 32b MUL |
1 | 3 | 1 | 2 | 2 | 3 | 2 | 4 |
Integer Multiply 64b MUL |
1 | 3 | 1 | 2 | 1 (2x 0.5) |
4 | 2 | 4 |
Integer Division 32b SDIV |
0.25 | 12 | 0.2 | < 12 | 1/12 - 1 | < 12 | 0.2 | 10 | 8 |
Integer Division 64b SDIV |
0.25 | 12 | 0.2 | < 12 | 1/21 - 1 | < 21 | 0.2 | 10 | 8 |
Move MOV |
2 | 1 | 3 | 1 | 3 | 1 | 3 | 1 |
Shift ops LSL |
2 | 1 | 3 | 1 | 3 | 1 | 6 | 1 |
Load instructions | 2 | 4 | 2 | 4 | 2 | 4 | 2 | |
Store instructions | 2 | 1 | 2 | 1 | 1 | 1 | 2 | |
FP Arithmetic FADD |
2 | 3 | 2 | 2 | 3 | 2 | 3 | 3 |
FP Multiply FMUL |
2 | 3 | 2 | 3 | 3 | 4 | 3 | 4 |
Multiply Accumulate MLA |
2 | 5 | 2 | 4 | 3 | 4 | 3 | 4 |
FP Division (S-form) | 0.2-0.33 | 6-10 | 0.66 | 7 | >0.16 | 12 | 0.5 | 1 | 10 | 8 |
FP Load | 2 | 5 | 2 | 5 | 2 | 5 | ||
FP Store | 2 | 1-N | 2 | 2 | 2 | 1 | ||
Vector Arithmetic | 2 | 3 | 2 | 2 | 3 | 1 | 3 | 2 |
Vector Multiply | 1 | 4 | 1 | 4 | 1 | 3 | 3 | 3 |
Vector Multiply Accumulate | 1 | 4 | 1 | 4 | 1 | 3 | 3 | 3 |
Vector FP Arithmetic | 2 | 3 | 2 | 2 | 3 | 2 | 3 | 3 |
Vector FP Multiply | 2 | 3 | 2 | 3 | 1 | 3 | 3 | 4 |
Vector Chained MAC (VMLA) |
2 | 6 | 2 | 5 | 3 | 5 | 3 | 3 |
Vector FP Fused MAC (VFMA) |
2 | 5 | 2 | 4 | 3 | 4 | 3 | 3 |
To compare the backend characteristics of Vortex, we’ve tested the instruction throughput. The backend performance is determined by the amount of execution units and the latency is dictated by the quality of their design.
The Vortex core looks pretty much the same as the predecessor Monsoon (A11) – with the exception that we’re seemingly looking at new division units, as the execution latency has seen a shaving of 2 cycles both on the integer and FP side. On the FP side the division throughput has seen a doubling.
Monsoon (A11) was a major microarchitectural update in terms of the mid-core and backend. It’s there that Apple had shifted the microarchitecture in Hurricane (A10) from a 6-wide decode from to a 7-wide decode. The most significant change in the backend here was the addition of two integer ALU units, upping them from 4 to 6 units.
Monsoon (A11) and Vortex (A12) are extremely wide machines – with 6 integer execution pipelines among which two are complex units, two load/store units, two branch ports, and three FP/vector pipelines this gives an estimated 13 execution ports, far wider than Arm’s upcoming Cortex A76 and also wider than Samsung’s M3. In fact, assuming we're not looking at an atypical shared port situation, Apple’s microarchitecture seems to far surpass anything else in terms of width, including desktop CPUs.
253 Comments
View All Comments
Constructor - Monday, October 8, 2018 - link
I don't see that happening at all because Apple has explicitly maintained a clear distinction between Macs and iOS exactly along the lines of different interaction paradigms (point-based vs. touch).Windows with touch continues to be a mess and I don't see Apple following Microsoft into that dead end.
Constructor - Monday, October 8, 2018 - link
If they'd have a replacement offering noticeably higher performance than any Intel Mac Pro and if legacy software at least ran decently until new ARM recompiles were available, I don't think most users would mind all that much.Going from PowerMacs to Mac Pros was also not entirely painless, but most users still thought it was worth it overall.
id4andrei - Sunday, October 7, 2018 - link
Andrei, I remember you mentioning in the comment sections of an article - maybe the S9 review - that the A11 cannot maintain it's freq and drops by as much as 40% while the Snapdragon drops only 10% on sustained workloads.You made your testing on a bench fan. You tested the potential of the A12, and it is incredible, but not the real life performance of the iphone. When used for prolonged sessions the A12 might reach its threshold faster than the Snapdragon and drop performance. What are your musings on this? Throttling matters and identifying it is very important, especially considering Apple's recent history. The CPU is great but is that top performance sustainable and for how long?
Andrei Frumusanu - Sunday, October 7, 2018 - link
What you mention is in regards to the GPU performance, it's addressed in that section in this piece.And of course it's the real performance of a phone. The duration of a benchmark doesn't change the fact that the CPU is capable of that throughput. Real world workloads are transactional and are a few seconds at best in the majority of use-cases. In such scenarios, the performance is well exhibited.
id4andrei - Sunday, October 7, 2018 - link
That makes perfect sense. No one does folding on smartphones. Thanks for the prompt reply.eastcoast_pete - Sunday, October 7, 2018 - link
Also, folding your smartphone is really hard, and doesn't end well for the phone (:FunBunny2 - Sunday, October 7, 2018 - link
"folding your smartphone is really hard, and doesn't end well for the phone"I don't recall (too lazy to confirm :) ) which company, but a patent was awarded a couple or so years ago for a flexible display, such that it would (according to the drawing I saw) make a cylindrical bend the hinge when closed. still hasn't appeared, so far as I know. let's see... looks like Samsung and LG have some, and more recently than when I first saw..
here: https://www.androidauthority.com/lg-foldable-phone...
eastcoast_pete - Monday, October 8, 2018 - link
Yes, that comment was in jest. I believe both Samsung, LG and Huawei have folding smartphones with folding screens underdevelopment. If those work out and aren't too pricey, I'd be interested. Nice to be able to fold a phone with a 7 inch display to a little more then half its full size.varase - Tuesday, October 23, 2018 - link
While this would probably be neat to see in the short run, I can't imagine that would yield a long lasting display over the long haul.Javert89 - Sunday, October 7, 2018 - link
Hi, do you think the current power draw of the CPU (in watt) is sustainable for the battery, expecially in the long term? In this review you cite a case where a GPU benchmark made crash the phone because the power required is too high.. Any chance to see this behavior on real life scenario? Moreover do you think that the power draw (watt) is sustainable in smartphone envelope? Or other aspects like overall power consumption or leakage count?