Addendum:

> the most recent chip they sell with PPC A2 cores (PowerEN)
> has been introduced in early 2010.

This was unexpected:

"the "wire-speed processor" [...] (PowerEN) SoC [...] was not released."
https://github.com/openpower-cores/a2i

By the way will ARM be big endian again?

> Oh, and there's this.

https://www.arm.com/company/news/2020/09/nvidia-to-acquire-arm
https://nvidianews.nvidia.com/news/nvidia-to-acquire-arm-for-40-billion-creating-worlds-premier-computing-company-for-the-age-of-ai
https://www.savearm.co.uk
https://arm.nvidia.com


Edit: added another link

[ Edited by Andreas_Wolf 28.05.2021 - 13:44 ]

> Intel going ARM instead of Atom!

...or Atom with ARM companion core :-)

"The Intel Programmable Services Engine (Intel PSE) is a dedicated offload engine for IoT functions powered by an ARM Cortex-M7 microcontroller. It provides independent, low-DMIPS computing and low-speed I/Os for IoT applications, plus dedicated services for real-time computing and time-sensitive synchronization. The Intel PSE hosts new functions, including remote out-of-band device management, network proxy, embedded controller lite, and sensor hub. Flexible programming, using open source code or prebuilt firmware binaries, lets you configure the Intel PSE to meet your application requirements and run real-time, ARM-based applications."
https://newsroom.intel.com/wp-content/uploads/sites/11/2020/09/Intel-Atom-x6000E-Pentium-Celeron-platform-brief.pdf

"Leverage a dedicated offload engine with the Intel Programmable Services Engine (Intel PSE). Powered by an Arm Cortex-M7 microcontroller, it provides independent, low-DMIPs computing and low-speed I/Os for IoT applications. It also hosts new functions like remote, out-of-band device management, network proxy, embedded controller, and sensor hub."
https://www.intel.com/content/www/us/en/design/products-and-solutions/processors-and-chipsets/elkhart-lake/overview.html

> https://en.wikipedia.org/wiki/Apple_A13
> A13 is hexa-core (2 "big" cores, 4 "small" cores) at 2.65 GHz. The new "big"
> core is named Lightning, while the new "small" core is named Thunder.

New:

https://en.wikipedia.org/wiki/Apple_A14

A14 is hexa-core (2 "big" cores, 4 "small" cores) at 3.0 GHz. The new "big" core is named Firestorm, while the new "small" core is named Icestorm.


Edit: https://en.wikipedia.org/wiki/Apple_M1 is octa-core (4 "big" cores, 4 "small" cores) at 3.2 GHz.

[ Edited by Andreas_Wolf 24.11.2020 - 14:00 ]

The perfect MorphOS NG platform RPi 400

Quote:

Yes, of course. But I am afraid that Bigfoot talked about about ARM64 port, not ARM

> Bigfoot talked about about ARM64 port

Did he?

> not ARM :-(

Cortex-A72 is ARM64.

Quote:

Sorry, my mistake. Blonde was too quick with writing...

correct sentence is:
Yes, of course. But I am afraid that Bigfoot talked about about AMD64 port, not ARM

Quote:
Indeed. I don't know why we have to repeat Apple's mistakes.

> I wasn’t aware of any 32-bit limitation until I learned
> about it from Andreas above.

https://blogs.windows.com/windows-insider/2020/12/10/introducing-x64-emulation-in-preview-for-windows-10-on-arm-pcs-to-the-windows-insider-program/

Update:

>> In terms of 64-bit ARMv8 (AArch64) cores/microarchitectures,
>> there exists or has been announced so far: [...]

This is the most recent list:

- Cortex-A34, Cortex-A35, Cortex-A53, Cortex-A55, Cortex-A57, Cortex-A65(AE), Cortex-A72, Cortex-A73, Cortex-A75, Cortex-A76(AE), Cortex-A77, Cortex-A78(AE|C), Cortex-R82, Cortex-X1, Neoverse E1, Neoverse N1 and Neoverse V1 from Arm Holdings
- Cyclone, Typhoon, Twister, Hurricane, Zephyr, Monsoon, Mistral, Vortex, Tempest, Lightning, Thunder, Firestorm, Icestorm, Avalanche, Blizzard, Everest and Sawtooth from Apple
- Mongoose M1, Mongoose M2, Meerkat M3, Cheetah M4 and Lion M5 from Samsung
- Kryo 100, Falkor, Saphira and Oryon (Phoenix) from Qualcomm
- Denver, Denver2 and Carmel from Nvidia
- ThunderX, ThunderX2, ThunderX2T99 (Vulcan) and ThunderX3 (Triton) from Marvell
- Potenza, Potenza+, Potenza++ and AmpereOne (Ampere-1A/Ampere-1B) from Ampere
- Xiaomi FTC310, Xiaomi FTC660, Xiaomi FTC661, Xiaomi FTC662, Xiaomi FTC663, Xiaomi FTC664 and Xiaomi FTC862 from Phytium
- A64FX from Fujitsu
- TaiShan v110 and TaiShan v120 from HiSilicon/Huawei

(Note that announced but scrapped cores (like AMD's K12) have been removed from the list.)

In terms of ARMv9 cores/microarchitectures, there have been announced so far:

- Neoverse N2, Neoverse N3, Neoverse V2, Neoverse V3, Neoverse E2, Neoverse E3, Cortex-X2, Cortex-X3, Cortex-X4, Cortex-X925, Cortex-A510, Cortex-A520, Cortex-A710, Cortex-A715, Cortex-A720 and Cortex-A725 from Arm Holdings


Edit: added Avalanche and Blizzard
Edit2: added Cortex-X3 and Cortex-A715
Edit3: added Everest and Sawtooth
Edit4: added Neoverse V2 and E2
Edit5: added Oryon
Edit6: added Xiaomi FTC310 and FTC664
Edit7: added AmpereOne
Edit8: added Cortex-X4, Cortex-A720 and Cortex-A520
Edit9: added TaiShan v120
Edit10: added Neoverse N3, V3 and E3
Edit11: added Cortex-X925 and Cortex-A725
Edit12: added Xiaomi FTC862

[ Edited by Andreas_Wolf 16.03.2025 - 21:03 ]

Mac Rumors Forum has a topic about MorphOS. And so I post the questions and comments about it: Why not continue supporting Apple Computers? Particularly ARM Apple Computers. I would buy a newer Apple Computer if they are ARM ... I think highly of ARM ... but I do not think highly of Modern Mac OS. When Apple discontinues support for their computers over the years, it would be nice if MorphOS could be available on them.

Also I agree with the other post about Pi Devices.

It seems that ARM support is increasing even from Microsoft.

[ Edited by In_Correct 05.06.2021 - 02:49 ]

> Apple confirmed their switch of their computers to the ARM ISA now.

Seems that with this switch, Apple computers finally got something from the CPU vainly made especially for them in the first place (PA6T):

"This series adds support for the I2C controller found on Apple Silicon Macs which has quite a bit of history: Apple bought P.A. Semi in 2008 and it looks like a part of its legacy continues to live on in the M1. This controller has actually been used since at least the iPhone 4S and hasn't changed much since then."
https://lists.ozlabs.org/pipermail/linuxppc-dev/2021-September/234633.html

Zombie thread comes back from the dead...

Anything ever happen with this (or x86-64)?

Looks like the PC world could be about to change. Previously the Arm based laptops were all using phone chips so they were pretty weak ...and then they charged a lot of money for them and there were software limitations etc...
That said, there seems to be some reasonable cheap Arm based Chromebooks around.

Then the Apple M1 came along...
Now you've got the Qualcomm X-Elite arriving very shortly which have M series level performance and efficiency. Nvidia and AMD are now both working on Arm processors plus a Mediatek / Nvidia joint design is on the way. I think this time x86 could have a real fight on it's hands.

Those M series chips are pretty impressive:
https://www.youtube.com/shorts/B1MJ1pl5zyA

> Zombie thread comes back from the dead...

I'm still trying to keep comment #992 up to date :-)

> Nvidia and AMD are now both working on Arm processors
> plus a Mediatek / Nvidia joint design is on the way.

Any new custom Arm core involved in these? I mean, Nvidia, AMD and Mediatek already have/had Arm processors released (Nvidia even with custom cores).

Quote:

I'm not sure about AMD but Nvidia use Arm's Neoverse cores in some designs and the rumour is the 2025 laptop chip will be based on X925 and built by ...Intel.

To be slightly on-topic for this forum, modern CPUs are getting so fast that I'm wondering if would it make sense to run MorphOS hosted? A native OS is faster and more efficient, but at what point does the difference become so small that you can't notice it?

> You need to Add Cortex X925 and Cortex A725 and A520.

A520 was already there (Edit8). A725 and X925 now added, thanks.

> modern CPUs are getting so fast that I'm wondering if would
> it make sense to run MorphOS hosted?

QEMU has been able to run MorphOS for 6 years. It cannot be registered for the emulated hardware, though, so the 30 minutes reboot cycle applies.

Quote:

That's more of a host on a host.
I'm thinking more like running native on another OS and using its drivers, a bit like like running a game.
It adds a layer between you and the hardware, but as I said, CPUs are getting so fast now would the average user even notice it? Run on Arm64 Linux and you'll get everything from the Raspberry Pi 3 up to the new Qualcomm laptops and potentially even the M series Macs.

For some perspective, in post 976 I was complaining about video stressing my laptop. I've upgraded twice since then, I can now load that video into RAM!

The Snapdragon X-Elite Geekbench scores are now appearing, there's a few different versions so scores are all over the place, but the top ones are getting similar scores to the M2 Pro/Max.

Good article on the X-Elite at Anandtech.
Not surprised the benchmarks are M2-like, some of the CPU core specs are pretty similar to the M1/2.

Arm says there's a bunch of other companies going to be chasing the PC market so looks like Intel and AMD could be about to get a whole heap of competition. I expect many will come in below the X-Elite's performance so they'll have to compete on price.

Addendum:

>> if the previous generations were, why change for the '060?

> I don't know, but I read that after the massively microcoded 68000,
> the following m68k generations used less and less microcode, so that
> the 68040 had relatively little microcode and the 68060 none at all.

A good decade later, confirmation from Motorola/Freescale/NXP's Joe Circello, lead designer of the 68060:
"[...] machines like the 060 that don't have microcode [...]. [...] the 060 [...] was basically fully hardcoded, no microcode at all. [...] The 060, just to be clear, does not have any microcode per se in it. Everything is hardwired."
https://www.youtube.com/watch?v=1takr2k7Yfo (at 57:09, 1:03:48 and 1:17:12)

>> the '060's design really derives from the '060
>> (really a kind of super scalar version of the '40).

> There are significant opcode differences between the two. And as you said,
> it's a black box, so ISA design of the 060 may be derived from the 040's,
> but this isn't necessarily true for the microarchitectural design.

In the above-linked interview, Circello also confirms that the 060 microarchitecture is a from-scratch design started in 1988 (so during 030 era), i.e. not derived from any previous m68k microarchitecture. He furthermore tells the story how he and his team at his former company (Edge?) designed what would become the 060 unbeknown to Motorola and then knocked on Motorola's door to sell the (preliminary) design and team (and whole company?) to them, which Motorola agreed to. This means that Motorola didn't have any own plan to add an 040 successor (apart from trimmed-down ColdFire released for low-power customers in the same year as the 060), which would have ended the legacy with just that, and just took opportunity of the offer, which also explains why the 060 was released by Motorola after they had started selling PowerPC to their high-power customers, which is a thing many have been curious about.
Really interesting story I'd never heard or read about before.