Discussion Apple Silicon SoC thread

[Eug]

M1
5 nm
Unified memory architecture - LP-DDR4
16 billion transistors

8-core CPU

4 high-performance cores
192 KB instruction cache
128 KB data cache
Shared 12 MB L2 cache

4 high-efficiency cores
128 KB instruction cache
64 KB data cache
Shared 4 MB L2 cache
(Apple claims the 4 high-effiency cores alone perform like a dual-core Intel MacBook Air)

8-core iGPU (but there is a 7-core variant, likely with one inactive core)
128 execution units
Up to 24576 concurrent threads
2.6 Teraflops
82 Gigatexels/s
41 gigapixels/s

16-core neural engine
Secure Enclave
USB 4

Products:
$999 ($899 edu) 13" MacBook Air (fanless) - 18 hour video playback battery life
$699 Mac mini (with fan)
$1299 ($1199 edu) 13" MacBook Pro (with fan) - 20 hour video playback battery life

Memory options 8 GB and 16 GB. No 32 GB option (unless you go Intel).

It should be noted that the M1 chip in these three Macs is the same (aside from GPU core number). Basically, Apple is taking the same approach which these chips as they do the iPhones and iPads. Just one SKU (excluding the X variants), which is the same across all iDevices (aside from maybe slight clock speed differences occasionally).

EDIT:



M1 Pro 8-core CPU (6+2), 14-core GPU
M1 Pro 10-core CPU (8+2), 14-core GPU
M1 Pro 10-core CPU (8+2), 16-core GPU
M1 Max 10-core CPU (8+2), 24-core GPU
M1 Max 10-core CPU (8+2), 32-core GPU

M1 Pro and M1 Max discussion here:

Page 78 - Discussion - Apple Silicon SoC thread

Page 78 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

forums.anandtech.com

M1 Ultra discussion here:

Page 109 - Discussion - Apple Silicon SoC thread

Page 109 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

forums.anandtech.com

M2 discussion here:

Page 127 - Discussion - Apple Silicon SoC thread

Page 127 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

forums.anandtech.com

Second Generation 5 nm
Unified memory architecture - LPDDR5, up to 24 GB and 100 GB/s
20 billion transistors

8-core CPU

4 high-performance cores
192 KB instruction cache
128 KB data cache
Shared 16 MB L2 cache

4 high-efficiency cores
128 KB instruction cache
64 KB data cache
Shared 4 MB L2 cache

10-core iGPU (but there is an 8-core variant)
3.6 Teraflops

16-core neural engine
Secure Enclave
USB 4

Hardware acceleration for 8K h.264, h.264, ProRes

M3 Family discussion here:

Page 215 - Discussion - Apple Silicon SoC thread

Page 215 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

forums.anandtech.com

M4 Family discussion here:

Page 263 - Discussion - Apple Silicon SoC thread

Page 263 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

forums.anandtech.com

 

[GC2:CS]

Looks like TrendForce / DRAMeXchange (part of TrendForce) published slightly ahead of AnandTech, so maybe that's where Ryan got the number from? But they also said the memory clock speed of M4 only reaches about or approximately 7700 MT/s. Which is odd because 7500 is a bog standard bin for LPDDR5X and it exactly matches Apple's memory bandwidth figure.

I think it is obvous that this is due to Apple rounding things.

For 100 Gbps you need 6250 MTs for 102,4 you need 6400 MTs.

For 120 Gbps you need 7500 MTs for 122,88 = 102.4x1.2 Gbps you need 7680 MTs = 7,7 Gts

Pick your preference.
I personally thik the went from 6400 to 7500.

 

[Doug S]

Right now, apple is selling
M1 (5nm) : None
M2 (5nm) : Macbook Air, Mac Mini, Mac Studio, Mac Pro, ipad Air.
M3 (3nm) : Macbook Air, Macbook Pro, iMac.
M4 (3nm) : ipad Pro.

My guess is that M2 and M4 will stay for a bit longer, while M3 will get phased out.

Apple often keeps two generations of a product around so they can have a lower price on the older model, but with M3 being supplanted by M4 so quickly they didn't get many M3 based products out the door.

But if they wanted to quickly drop M3 why did they introduce the M3 Macbook Air only two months ago?

 

[Eug]

Apple often keeps two generations of a product around so they can have a lower price on the older model, but with M3 being supplanted by M4 so quickly they didn't get many M3 based products out the door.

But if they wanted to quickly drop M3 why did they introduce the M3 Macbook Air only two months ago?

I wonder if the MBA will skip M4. The MBA is Apple’s best selling laptop, but nonetheless Apple isn’t afraid to skip chip generations with their Macs.

I also wonder if the MB Pro will finally get 12 GB base. I’m not optimistic for this M4 generation though.

 

[StinkyPinky]

Putting these chips on IpadOS devices is like putting lipstick on a baboon and claiming it's your hot wife.

Waste of time. They need to improve the multi tasking on the operating system first.

 

[poke01]

iPad16,6 - Geekbench

Benchmark results for an iPad16,6 with an ARM processor.

browser.geekbench.com

Apple M4 crushes AMD and Intel. Remeber this is a iPad

 

[Nothingness]

iPad16,6 - Geekbench

Benchmark results for an iPad16,6 with an ARM processor.

browser.geekbench.com

Apple M4 crushes AMD and Intel. Remeber this is a iPad

That seems to be 10% clock increase plus 10% IPC increase vs M3 pro. But looking test by test there are outliers that skew the global score (object detection in particular). I'll take a closer look tomorrow.

 

[poke01]

Brendan Shanks (@bshanks@mastodon.social)

Attached: 2 images New in Xcode 15.4 RC: * CPUFAMILY_ARM_DONAN. This must be the M4? * EXC_ARM_SME_DISALLOWED. Have Apple's previous cores supported SME? @never_released @marcan@treehouse.systems @dougall @saagar@saagarjha.com

mastodon.social

M4 cpu is not the same arch as M3

 

[Nothingness]

Ha SME 😀 That alone might explain some of the tests huge speedup.

 

[poke01]

amarioguy (@amarioguy@treehouse.systems)

@bshanks@mastodon.social @never_released@mastodon.social @marcan @dougall@mastodon.social @saagar@saagarjha.com Donan is *definitely* the M4, there's even new P and E core generations to go with it in another Xcode header file.

social.treehouse.systems

Can't wait for your tests @junjie1475 via Geekerwan

 

[gdansk]

Ha SME 😀 That alone might explain some of the tests huge speedup.

Someone will run the old version of GB6 which didn't support it, for science, right?

 

[repoman27]

The plot thickens! Those multicore scores are pretty competitive with GB6 multicore scores for Snapdragon X Elite under Windows (Linux are still higher): https://browser.geekbench.com/v6/cpu/search?dir=desc&page=1&q=Qualcomm+Oryon&sort=multicore_score

Doesn't look like SMT came into play with the GB6 score though. Here's the .gb6:

{
  "date": "2024-05-08 18:02:10 -0500",
  "document_type": 0,
  "document_version": 6,
  "branch": "rosedale-main-build",
  "build": 603408,
  "checksum": "c2b1d93d0cc1b1bcc08d7ee5a15a864f0a2aa404",
  "commit": "bca065a7d9",
  "options": {
    "iterations": 5,
    "cpu_workers": 0,
    "memory_workers": 0,
    "workload_gap": 5000
  },
  "platform": {
    "os": "iOS",
    "architecture": "aarch64",
    "bits": 64
  },
  "processor_frequency": {
    "frequencies": [
      4232,
      4259,
      4311,
      4273,
      4275,
      4282,
      4269,
      4305,
      4361,
      4355,
      4364,
      4361,
      4363,
      4363,
      4348,
      4359,
      4365,
      4362,
      4363,
      4359,
      4353,
      4363,
      4288,
      4240,
      4245,
      4264,
      4310,
      4296,
      4221,
      4335,
      4352,
      4359,
      4361,
      4326,
      4377,
      4382,
      4349,
      4354,
      4357,
      4360,
      4355,
      4360,
      4370,
      4358
    ]
  },
  "runtime": 234.54420154166669,
  "system_uuid": "E6ACAB03-3511-48FE-BEB4-83AF51AE24D3",
  "clock": 9.4164847500000004,
  "uuid": "2de2a5d5d81a4669a4ca2991342cecda",
  "version": "Geekbench 6.3.0",
  "metrics": [
    {
      "id": 1,
      "value": "iOS AArch64",
      "ivalue": 0
    },
    {
      "id": 2,
      "value": "Clang 16.0.0 ",
      "ivalue": 0
    },
    {
      "id": 3,
      "value": "iOS 17.5",
      "ivalue": 0
    },
    {
      "id": 4,
      "value": "iPad16,6",
      "ivalue": 0
    },
    {
      "id": 5,
      "value": "iPad16,6",
      "ivalue": 0
    },
    {
      "id": 6,
      "value": "J721AP",
      "ivalue": 0
    },
    {
      "id": 7,
      "value": "ARM",
      "ivalue": 0
    },
    {
      "id": 8,
      "value": "ARM",
      "ivalue": 0
    },
    {
      "id": 9,
      "value": "ARM",
      "ivalue": 0
    },
    {
      "id": 12,
      "value": "10",
      "ivalue": 10
    },
    {
      "id": 13,
      "value": "10",
      "ivalue": 10
    },
    {
      "id": 14,
      "value": "1",
      "ivalue": 1
    },
    {
      "id": 15,
      "value": "4.40 GHz",
      "ivalue": 4397000000
    },
    {
      "id": 17,
      "value": "128 KB",
      "ivalue": 131072
    },
    {
      "id": 18,
      "value": "1",
      "ivalue": 1
    },
    {
      "id": 19,
      "value": "64.0 KB",
      "ivalue": 65536
    },
    {
      "id": 20,
      "value": "1",
      "ivalue": 1
    },
    {
      "id": 21,
      "value": "4.00 MB",
      "ivalue": 4194304
    },
    {
      "id": 22,
      "value": "1",
      "ivalue": 1
    },
    {
      "id": 23,
      "value": "0.00 B",
      "ivalue": 0
    },
    {
      "id": 24,
      "value": "1",
      "ivalue": 1
    },
    {
      "id": 25,
      "value": "0.00 B",
      "ivalue": 0
    },
    {
      "id": 26,
      "value": "1",
      "ivalue": 1
    },
    {
      "id": 27,
      "value": "0.00 Hz",
      "ivalue": 0
    },
    {
      "id": 29,
      "value": "15.0 GB",
      "ivalue": 16159506432
    },
    {
      "id": 30,
      "value": "",
      "ivalue": 0
    },
    {
      "id": 40,
      "value": "0x1",
      "ivalue": 1
    },
    {
      "id": 45,
      "value": "2",
      "ivalue": 2
    },
    {
      "id": 46,
      "value": "4 Cores",
      "ivalue": 0
    },
    {
      "id": 47,
      "value": "4",
      "ivalue": 4
    },
    {
      "id": 51,
      "value": "6 Cores",
      "ivalue": 0
    },
    {
      "id": 52,
      "value": "6",
      "ivalue": 6
    },
    {
      "id": 71,
      "value": "2064",
      "ivalue": 2064
    },
    {
      "id": 72,
      "value": "2752",
      "ivalue": 2752
    },
    {
      "id": 73,
      "value": "2",
      "ivalue": 2
    },
    {
      "id": 74,
      "value": "",
      "ivalue": 0
    },
    {
      "id": 88,
      "value": "Clang",
      "ivalue": 0
    },
    {
      "id": 89,
      "value": "4396 MHz",
      "ivalue": 4396
    },
    {
      "id": 90,
      "value": "",
      "ivalue": 0
    },
    {
      "id": 93,
      "value": "4",
      "ivalue": 4
    },
    {
      "id": 94,
      "value": "0x2",
      "ivalue": 2
    },
    {
      "id": 99,
      "value": "6",
      "ivalue": 6
    },
    {
      "id": 100,
      "value": "0x1",
      "ivalue": 1
    },
    {
      "id": 117,
      "value": "False",
      "ivalue": 0
    },
    {
      "id": 84,
      "value": "",
      "ivalue": 234544
    },
    {
      "id": 85,
      "value": "",
      "ivalue": 234000
    },
    {
      "id": 86,
      "value": "",
      "ivalue": 0
    }
  ],
  "multicore_score": 14677,
  "score": 3767,
  "valid": 1,
  "sections": [
    {
      "name": "Single-Core",
      "id": 1,
      "score": 3767,
      "valid": 1,
      "workloads": [
        {
          "name": "File Compression",
          "id": 101,
          "units": 2,
          "threads": 1,
          "score": 3294,
          "factory_clock": 10.947298875000001,
          "factory_runtime": 0.034159541666666668,
          "workload_clock": 10.981471791666667,
          "runtime": 1.2607787291666668,
          "runtime_warmup": 1.2475877083333333,
          "runtimes": [
            1.2471049166666668,
            1.2629290416666668,
            1.2633886666666667,
            1.2696922916666666
          ],
          "valid": 1,
          "rate": 496089587.75296593,
          "work": 625459200,
          "kernels": [],
          "temperatures": [],
          "thermal_state": 0,
          "optimize_runtime": 0.0,
          "transfer_runtime": 0.0
        },
        {
          "name": "Navigation",
          "id": 102,
          "units": 26,
          "threads": 1,
          "score": 3360,
          "factory_clock": 22.277579541666668,
          "factory_runtime": 2.0916666666666669e-5,
          "workload_clock": 22.277638166666669,
          "runtime": 1.1854278958333335,
          "runtime_warmup": 1.1702116250000001,
          "runtimes": [
            1.1725696666666667,
            1.1826957083333334,
            1.1919941250000001,
            1.1944520833333334
          ],
          "valid": 1,
          "rate": 20.245853910100919,
          "work": 24,
          "kernels": [],
          "temperatures": [],
          "thermal_state": 0,
          "optimize_runtime": 0.0,
          "transfer_runtime": 0.0
        },
        {
          "name": "HTML5 Browser",
          "id": 103,
          "units": 25,
          "threads": 1,
          "score": 3693,
          "factory_clock": 33.368762833333335,
          "factory_runtime": 0.016506791666666666,
          "workload_clock": 33.385288500000001,
          "runtime": 0.10582540625,
          "runtime_warmup": 0.160593875,
          "runtimes": [
            0.10477266666666667,
            0.10565237500000001,
            0.106200875,
            0.10667570833333334
          ],
          "valid": 1,
          "rate": 75.596213456539402,
          "work": 8,
          "kernels": [],
          "temperatures": [],
          "thermal_state": 0,
          "optimize_runtime": 0.0,
          "transfer_runtime": 0.0
        },
        {
          "name": "PDF Renderer",
          "id": 104,
          "units": 9,
          "threads": 1,
          "score": 3450,
          "factory_clock": 38.997110583333338,
          "factory_runtime": 0.0044421249999999999,
          "workload_clock": 39.00157191666667,
          "runtime": 0.43899573958333338,
          "runtime_warmup": 0.46248875,
          "runtimes": [
            0.43692416666666667,
            0.43783645833333334,
            0.43932479166666671,
            0.44189754166666667
          ],
          "valid": 1,
          "rate": 79552571.587931365,
          "work": 34923240,
          "kernels": [],
          "temperatures": [],
          "thermal_state": 0,
          "optimize_runtime": 0.0,
          "transfer_runtime": 0.0
        },
        {
          "name": "Photo Library",
          "id": 105,
          "units": 20,
          "threads": 1,
          "score": 3547,
          "factory_clock": 46.225447958333334,
          "factory_runtime": 0.046808833333333334,
          "workload_clock": 46.272265333333337,
          "runtime": 0.33240289583333332,
          "runtime_warmup": 0.33507800000000004,
          "runtimes": [
            0.32995145833333334,
            0.33227125000000002,
            0.33385645833333333,
            0.33353241666666666
          ],
          "valid": 1,
          "rate": 48.134358035263304,
          "work": 16,
          "kernels": [],
          "temperatures": [],
          "thermal_state": 0,
          "optimize_runtime": 0.0,
          "transfer_runtime": 0.0
        },
        {
          "name": "Clang",
          "id": 201,
          "units": 23,
          "threads": 1,
          "score": 4246,
          "factory_clock": 53.639585125000004,
          "factory_runtime": 0.0083458750000000009,
          "workload_clock": 53.647961208333335,
          "runtime": 0.37392343750000001,
          "runtime_warmup": 0.41151675000000004,
          "runtimes": [
            0.37050875,
            0.37344858333333336,
            0.37480620833333333,
            0.37693020833333335
          ],
          "valid": 1,
          "rate": 20913.372139182902,
          "work": 7820,
          "kernels": [],
          "temperatures": [],
          "thermal_state": 0,
          "optimize_runtime": 0.0,
          "transfer_runtime": 0.0
        },
        {
          "name": "Text Processing",
          "id": 202,
          "units": 25,
          "threads": 1,
          "score": 3356,
          "factory_clock": 60.560789875000005,
          "factory_runtime": 0.047860833333333339,
          "workload_clock": 60.608660750000006,
          "runtime": 0.71076600000000001,
          "runtime_warmup": 0.69609391666666676,
          "runtimes": [
            0.70752708333333336,
            0.70860670833333339,
            0.71186454166666668,
            0.71506566666666671
          ],
          "valid": 1,
          "rate": 268.72416519642189,
          "work": 191,
          "kernels": [],
          "temperatures": [],
          "thermal_state": 0,
          "optimize_runtime": 0.0,
          "transfer_runtime": 0.0
        },
        {
          "name": "Asset Compression",
          "id": 203,
          "units": 2,
          "threads": 1,
          "score": 3253,
          "factory_clock": 69.253849958333333,
          "factory_runtime": 0.032741958333333335,
          "workload_clock": 69.286601125000004,
          "runtime": 1.3274794583333334,
          "runtime_warmup": 1.3255355833333333,
          "runtimes": [
            1.3178962083333334,
            1.3263416250000002,
            1.3307782916666668,
            1.3349017083333334
          ],
          "valid": 1,
          "rate": 105702788.18187614,
          "work": 140318280,
          "kernels": [],
          "temperatures": [],
          "thermal_state": 0,
          "optimize_runtime": 0.0,
          "transfer_runtime": 0.0
        },
        {
          "name": "Object Detection",
          "id": 301,
          "units": 20,
          "threads": 1,
          "score": 5604,
          "factory_clock": 80.928489208333332,
          "factory_runtime": 0.073898708333333341,
          "workload_clock": 81.002397500000001,
          "runtime": 0.09541008333333334,
          "runtime_warmup": 0.10833554166666667,
          "runtimes": [
            0.095031625000000008,
            0.095357666666666674,
            0.09546558333333334,
            0.095785458333333337
          ],
          "valid": 1,
          "rate": 167.69715989138115,
          "work": 16,
          "kernels": [],
          "temperatures": [],
          "thermal_state": 0,
          "optimize_runtime": 0.0,
          "transfer_runtime": 0.0
        },
        {
          "name": "Background Blur",
          "id": 302,
          "units": 20,
          "threads": 1,
          "score": 3802,
          "factory_clock": 86.503218333333336,
          "factory_runtime": 0.059878791666666667,
          "workload_clock": 86.56310533333334,
          "runtime": 0.63549967708333344,
          "runtime_warmup": 0.62988533333333341,
          "runtimes": [
            0.63246050000000009,
            0.63512062499999999,
            0.63685091666666671,
            0.63756666666666673
          ],
          "valid": 1,
          "rate": 15.735649223136733,
          "work": 10,
          "kernels": [],
          "temperatures": [],
          "thermal_state": 0,
          "optimize_runtime": 0.0,
          "transfer_runtime": 0.0
        },
        {
          "name": "Horizon Detection",
          "id": 401,
          "units": 9,
          "threads": 1,
          "score": 3917,
          "factory_clock": 94.742204541666666,
          "factory_runtime": 0.13646791666666669,
          "workload_clock": 94.878682833333343,
          "runtime": 0.39379790625,
          "runtime_warmup": 0.38985045833333337,
          "runtimes": [
            0.39184508333333334,
            0.39468404166666671,
            0.39262904166666668,
            0.39603345833333337
          ],
          "valid": 1,
          "rate": 121889931.96303974,
          "work": 48000000,
          "kernels": [],
          "temperatures": [],
          "thermal_state": 0,
          "optimize_runtime": 0.0,
          "transfer_runtime": 0.0
        },
        {
          "name": "Object Remover",
          "id": 402,
          "units": 9,
          "threads": 1,
          "score": 4448,
          "factory_clock": 101.85607366666667,
          "factory_runtime": 0.1338955,
          "workload_clock": 101.98997779166667,
          "runtime": 1.0552571875000001,
          "runtime_warmup": 1.0356735833333335,
          "runtimes": [
            1.0493253333333334,
            1.0555175833333335,
            1.0584402916666666,
            1.0577455416666668
          ],
          "valid": 1,
          "rate": 342023939.06935596,
          "work": 360923220,
          "kernels": [],
          "temperatures": [],
          "thermal_state": 0,
          "optimize_runtime": 0.0,
          "transfer_runtime": 0.0
        },
        {
          "name": "HDR",
          "id": 403,
          "units": 9,
          "threads": 1,
          "score": 4285,
          "factory_clock": 112.25456150000001,
          "factory_runtime": 0.0072381250000000006,
          "workload_clock": 112.26182887500001,
          "runtime": 0.52854480208333343,
          "runtime_warmup": 0.57262304166666667,
          "runtimes": [
            0.52555450000000004,
            0.52872833333333336,
            0.53010170833333337,
            0.52979466666666675
          ],
          "valid": 1,
          "rate": 125729178.94200113,
          "work": 66453504,
          "kernels": [],
          "temperatures": [],
          "thermal_state": 0,
          "optimize_runtime": 0.0,
          "transfer_runtime": 0.0
        },
        {
          "name": "Photo Filter",
          "id": 404,
          "units": 20,
          "threads": 1,
          "score": 4687,
          "factory_clock": 119.95787695833334,
          "factory_runtime": 8.2625000000000003e-5,
          "workload_clock": 119.95799708333334,
          "runtime": 0.2150187291666667,
          "runtime_warmup": 0.23759770833333335,
          "runtimes": [
            0.21838754166666668,
            0.21352554166666668,
            0.2138565,
            0.21430533333333335
          ],
          "valid": 1,
          "rate": 46.507576520223665,
          "work": 10,
          "kernels": [],
          "temperatures": [],
          "thermal_state": 0,
          "optimize_runtime": 0.0,
          "transfer_runtime": 0.0
        },
        {
          "name": "Ray Tracer",
          "id": 501,
          "units": 9,
          "threads": 1,
          "score": 2960,
          "factory_clock": 126.420134125,
          "factory_runtime": 0.49180608333333337,
          "workload_clock": 126.91195408333334,
          "runtime": 0.34916373958333335,
          "runtime_warmup": 0.34533708333333335,
          "runtimes": [
            0.34892858333333338,
            0.34788608333333337,
            0.34961345833333335,
            0.35022683333333338
          ],
          "valid": 1,
          "rate": 2863985.8227928462,
          "work": 1000000,
          "kernels": [],
          "temperatures": [],
          "thermal_state": 0,
          "optimize_runtime": 0.0,
          "transfer_runtime": 0.0
        },
        {
          "name": "Structure from Motion",
          "id": 502,
          "units": 9,
          "threads": 1,
          "score": 3458,
          "factory_clock": 133.73663445833333,
          "factory_runtime": 0.12021612500000001,
          "workload_clock": 133.85685925000001,
          "runtime": 0.58457931250000006,
          "runtime_warmup": 0.57271575000000008,
          "runtimes": [
            0.57953516666666671,
            0.58321870833333334,
            0.58481037499999999,
            0.59075300000000008
          ],
          "valid": 1,
          "rate": 109480.43940573075,
          "work": 64000,
          "kernels": [],
          "temperatures": [],
          "thermal_state": 0,
          "optimize_runtime": 0.0,
          "transfer_runtime": 0.0
        }
      ]
    },
    {
      "name": "Multi-Core",
      "id": 2,
      "score": 14677,
      "valid": 1,
      "workloads": [
        {
          "name": "File Compression",
          "id": 101,
          "units": 2,
          "threads": 10,
          "score": 10782,
          "factory_clock": 141.77582216666667,
          "factory_runtime": 0.056126708333333337,
          "workload_clock": 141.83195733333335,
          "runtime": 0.3852227604166667,
          "runtime_warmup": 0.388772375,
          "runtimes": [
            0.38629625000000001,
            0.38224429166666668,
            0.38669616666666667,
            0.38565433333333338
          ],
          "valid": 1,
          "rate": 1623629920.8371994,
          "work": 625459200,
          "kernels": [],
          "temperatures": [],
          "thermal_state": 0,
          "optimize_runtime": 0.0,
          "transfer_runtime": 0.0
        },
        {
          "name": "Navigation",
          "id": 102,
          "units": 26,
          "threads": 10,
          "score": 15187,
          "factory_clock": 148.76689841666666,
          "factory_runtime": 6.5166666666666674e-5,
          "workload_clock": 148.76700220833334,
          "runtime": 0.26228507291666669,
          "runtime_warmup": 0.26159329166666667,
          "runtimes": [
            0.260133375,
            0.27027533333333337,
            0.25955633333333333,
            0.25917525000000002
          ],
          "valid": 1,
          "rate": 91.503491727969163,
          "work": 24,
          "kernels": [],
          "temperatures": [],
          "thermal_state": 0,
          "optimize_runtime": 0.0,
          "transfer_runtime": 0.0
        },
        {
          "name": "HTML5 Browser",
          "id": 103,
          "units": 25,
          "threads": 10,
          "score": 16675,
          "factory_clock": 155.25668683333333,
          "factory_runtime": 0.022244125,
          "workload_clock": 155.27894462500001,
          "runtime": 0.093742124999999996,
          "runtime_warmup": 0.12612929166666667,
          "runtimes": [
            0.095400125000000002,
            0.094668208333333337,
            0.093378208333333337,
            0.091521958333333334
          ],
          "valid": 1,
          "rate": 341.36200774198369,
          "work": 32,
          "kernels": [],
          "temperatures": [],
          "thermal_state": 0,
          "optimize_runtime": 0.0,
          "transfer_runtime": 0.0
        },
        {
          "name": "PDF Renderer",
          "id": 104,
          "units": 9,
          "threads": 10,
          "score": 15937,
          "factory_clock": 160.88737583333335,
          "factory_runtime": 0.0045074166666666665,
          "workload_clock": 160.89190758333334,
          "runtime": 0.38007842708333339,
          "runtime_warmup": 0.41978075000000004,
          "runtimes": [
            0.38348095833333334,
            0.39657579166666668,
            0.37511133333333335,
            0.36514562500000003
          ],
          "valid": 1,
          "rate": 367537197.70938718,
          "work": 139692960,
          "kernels": [],
          "temperatures": [],
          "thermal_state": 0,
          "optimize_runtime": 0.0,
          "transfer_runtime": 0.0
        },
        {
          "name": "Photo Library",
          "id": 105,
          "units": 20,
          "threads": 10,
          "score": 17917,
          "factory_clock": 167.84339650000001,
          "factory_runtime": 0.042060083333333338,
          "workload_clock": 167.88546733333334,
          "runtime": 0.2632020729166667,
          "runtime_warmup": 0.27533925000000004,
          "runtimes": [
            0.2661264166666667,
            0.26224395833333336,
            0.26028995833333335,
            0.26414795833333338
          ],
          "valid": 1,
          "rate": 243.15917914621917,
          "work": 64,
          "kernels": [],
          "temperatures": [],
          "thermal_state": 0,
          "optimize_runtime": 0.0,
          "transfer_runtime": 0.0
        },
        {
          "name": "Clang",
          "id": 201,
          "units": 23,
          "threads": 10,
          "score": 23357,
          "factory_clock": 174.95738829166669,
          "factory_runtime": 0.0081828750000000009,
          "workload_clock": 174.96559295833333,
          "runtime": 0.81573295833333337,
          "runtime_warmup": 0.810600875,
          "runtimes": [
            0.80257345833333338,
            0.82471570833333341,
            0.82703508333333342,
            0.80860758333333338
          ],
          "valid": 1,
          "rate": 115037.64686881036,
          "work": 93840,
          "kernels": [],
          "temperatures": [],
          "thermal_state": 0,
          "optimize_runtime": 0.0,
          "transfer_runtime": 0.0
        },
        {
          "name": "Text Processing",
          "id": 202,
          "units": 25,
          "threads": 1,
          "score": 4418,
          "factory_clock": 184.04712612500001,
          "factory_runtime": 0.042005916666666671,
          "workload_clock": 184.08913945833334,
          "runtime": 0.53984254166666668,
          "runtime_warmup": 0.54679983333333337,
          "runtimes": [
            0.5417357083333334,
            0.53720037500000006,
            0.5400200833333334,
            0.54041400000000006
          ],
          "valid": 1,
          "rate": 353.80687007422921,
          "work": 191,
          "kernels": [],
          "temperatures": [],
          "thermal_state": 0,
          "optimize_runtime": 0.0,
          "transfer_runtime": 0.0
        },
        {
          "name": "Asset Compression",
          "id": 203,
          "units": 2,
          "threads": 10,
          "score": 18629,
          "factory_clock": 191.90489304166667,
          "factory_runtime": 0.039821541666666668,
          "workload_clock": 191.94472350000001,
          "runtime": 0.23181878124999999,
          "runtime_warmup": 0.23307487500000001,
          "runtimes": [
            0.227118125,
            0.23281812500000001,
            0.23268958333333334,
            0.23464929166666668
          ],
          "valid": 1,
          "rate": 605292976.01938796,
          "work": 140318280,
          "kernels": [],
          "temperatures": [],
          "thermal_state": 0,
          "optimize_runtime": 0.0,
          "transfer_runtime": 0.0
        },
        {
          "name": "Object Detection",
          "id": 301,
          "units": 20,
          "threads": 10,
          "score": 14115,
          "factory_clock": 198.11133275,
          "factory_runtime": 0.13945879166666666,
          "workload_clock": 198.25079995833335,
          "runtime": 0.15152463541666666,
          "runtime_warmup": 0.17956029166666668,
          "runtimes": [
            0.15170283333333334,
            0.14817395833333336,
            0.15409595833333334,
            0.15212579166666668
          ],
          "valid": 1,
          "rate": 422.37356205485014,
          "work": 64,
          "kernels": [],
          "temperatures": [],
          "thermal_state": 0,
          "optimize_runtime": 0.0,
          "transfer_runtime": 0.0
        },
        {
          "name": "Background Blur",
          "id": 302,
          "units": 20,
          "threads": 10,
          "score": 13810,
          "factory_clock": 204.06141279166667,
          "factory_runtime": 0.067551541666666673,
          "workload_clock": 204.12897775000002,
          "runtime": 0.17496571875,
          "runtime_warmup": 0.17772841666666667,
          "runtimes": [
            0.170836875,
            0.17403608333333334,
            0.17908654166666668,
            0.175903375
          ],
          "valid": 1,
          "rate": 57.154053213638456,
          "work": 10,
          "kernels": [],
          "temperatures": [],
          "thermal_state": 0,
          "optimize_runtime": 0.0,
          "transfer_runtime": 0.0
        },
        {
          "name": "Horizon Detection",
          "id": 401,
          "units": 9,
          "threads": 10,
          "score": 18379,
          "factory_clock": 210.01393729166668,
          "factory_runtime": 0.13666708333333333,
          "workload_clock": 210.15061779166669,
          "runtime": 0.083927031250000006,
          "runtime_warmup": 0.088085916666666667,
          "runtimes": [
            0.085657833333333336,
            0.083279791666666672,
            0.082828083333333344,
            0.083942416666666672
          ],
          "valid": 1,
          "rate": 571925389.05634165,
          "work": 48000000,
          "kernels": [],
          "temperatures": [],
          "thermal_state": 0,
          "optimize_runtime": 0.0,
          "transfer_runtime": 0.0
        },
        {
          "name": "Object Remover",
          "id": 402,
          "units": 9,
          "threads": 10,
          "score": 13836,
          "factory_clock": 215.58618075000001,
          "factory_runtime": 0.13555562500000001,
          "workload_clock": 215.72174470833335,
          "runtime": 0.33927786458333331,
          "runtime_warmup": 0.33398262500000003,
          "runtimes": [
            0.34135862500000003,
            0.33768533333333334,
            0.33933704166666667,
            0.33873045833333337
          ],
          "valid": 1,
          "rate": 1063798313.0530761,
          "work": 360923220,
          "kernels": [],
          "temperatures": [],
          "thermal_state": 0,
          "optimize_runtime": 0.0,
          "transfer_runtime": 0.0
        },
        {
          "name": "HDR",
          "id": 403,
          "units": 9,
          "threads": 10,
          "score": 17251,
          "factory_clock": 222.42092383333335,
          "factory_runtime": 0.0066665000000000006,
          "workload_clock": 222.42761904166667,
          "runtime": 0.13127153125000002,
          "runtime_warmup": 0.16420058333333334,
          "runtimes": [
            0.13168475000000002,
            0.13226329166666667,
            0.133128,
            0.12801008333333333
          ],
          "valid": 1,
          "rate": 506229365.70643526,
          "work": 66453504,
          "kernels": [],
          "temperatures": [],
          "thermal_state": 0,
          "optimize_runtime": 0.0,
          "transfer_runtime": 0.0
        },
        {
          "name": "Photo Filter",
          "id": 404,
          "units": 20,
          "threads": 10,
          "score": 15074,
          "factory_clock": 228.12637629166667,
          "factory_runtime": 0.00012487500000000001,
          "workload_clock": 228.12653045833335,
          "runtime": 0.066860854166666678,
          "runtime_warmup": 0.076963875000000001,
          "runtimes": [
            0.067496374999999997,
            0.060260041666666674,
            0.07438575,
            0.065301250000000005
          ],
          "valid": 1,
          "rate": 149.56434710021213,
          "work": 10,
          "kernels": [],
          "temperatures": [],
          "thermal_state": 0,
          "optimize_runtime": 0.0,
          "transfer_runtime": 0.0
        },
        {
          "name": "Ray Tracer",
          "id": 501,
          "units": 9,
          "threads": 10,
          "score": 18317,
          "factory_clock": 233.84844416666667,
          "factory_runtime": 0.49969283333333336,
          "workload_clock": 234.34814837500002,
          "runtime": 0.056420697916666672,
          "runtime_warmup": 0.055812041666666673,
          "runtimes": [
            0.056277375000000004,
            0.056077916666666672,
            0.05669095833333334,
            0.056636541666666672
          ],
          "valid": 1,
          "rate": 17723992.026419085,
          "work": 1000000,
          "kernels": [],
          "temperatures": [],
          "thermal_state": 0,
          "optimize_runtime": 0.0,
          "transfer_runtime": 0.0
        },
        {
          "name": "Structure from Motion",
          "id": 502,
          "units": 9,
          "threads": 10,
          "score": 15650,
          "factory_clock": 239.7138869166667,
          "factory_runtime": 0.12189758333333334,
          "workload_clock": 239.83579225000003,
          "runtime": 0.12915627083333334,
          "runtime_warmup": 0.13177508333333335,
          "runtimes": [
            0.12783358333333333,
            0.12806558333333334,
            0.13209325,
            0.12863266666666667
          ],
          "valid": 1,
          "rate": 495523.75263751065,
          "work": 64000,
          "kernels": [],
          "temperatures": [],
          "thermal_state": 0,
          "optimize_runtime": 0.0,
          "transfer_runtime": 0.0
        }
      ]
    }
  ],
  "complete_benchmark": 1
}

 

[gdansk]

SMT

What now?

 

[repoman27]

What now?

Lol! Reading quickly through the thread I thought I saw SMT, but it was SME. Totally different, disregard.

 

[SpudLobby]

iPad16,6 - Geekbench

Benchmark results for an iPad16,6 with an ARM processor.

browser.geekbench.com

Apple M4 crushes AMD and Intel. Remeber this is a iPad

Holy. Wow

 

[poke01]

https://twitter.com/x/status/1788338408445919698

Even better

iPad16,6 - Geekbench

Benchmark results for an iPad16,6 with an ARM processor.

browser.geekbench.com

 

[poke01]

yeah, M4 is definitely NOT the same CPU architecture

 

[SpudLobby]

How much of this is just SME use?

 

[SarahKerrigan]

How much of this is just SME use?

I only looked at it briefly, but I think a lot of the gains are SME. The clock-normalized gains for the normal integer-spaghetti subtests against A17 are pretty small, often close to flat.

 

[poke01]

both GB 3.0.

So scores are comparable.

From:

https://twitter.com/x/status/1788344136661356561

 

[SarahKerrigan]

I put together a quick and dirty clock-normalized comparison spreadsheet between M4 and A17:



I'm using https://browser.geekbench.com/v6/cpu/6010114 for the iPhone data. (Also it only just occurred to me that I could have compared against M3 instead of A17. I am a very stupid lady at times. Still, broad conclusions should still apply.)

 

[poke01]

I

I put together a quick and dirty clock-normalized comparison spreadsheet between M4 and A17:

View attachment 98626

I'm using https://browser.geekbench.com/v6/cpu/6010114 for the iPhone data. (Also it only just occurred to me that I could have compared against M3 instead of A17. I am a very stupid lady at times. Still, broad conclusions should still apply.)

Interesting. I guess we will have to look at cinebench and Spec and also real world tests

 

[poke01]

Another important thing is Apple depreciated AMX and is using SME first gen.

EDIT: Not fully confirmed

 

[soresu]

I only looked at it briefly, but I think a lot of the gains are SME. The clock-normalized gains for the normal integer-spaghetti subtests against A17 are pretty small, often close to flat.

Wait wat?

Is it v9-A compatible?

Does it have SVE too?

 

[poke01]

Is it v9-A compatible?

If it has SME then its ARMv9

 

[soresu]

If it has SME then its ARMv9

I hope it has SVE2 also then.

Up until now there has been no impetus for OSS to actually start playing with SVE2, especially if they already have comprehensive NEON code paths.