A possible future for MorphOS
  • Just looking around
    Posts: 15 from 2016/1/22
    > the next Nintendo will use X64
    We don't know, it may just as well be ARM based.

    > Manufacturers of lightweight devices and manufacturers of servers hardly overlap.
    > Lightweight devices and servers are distinct markets. Even if the same ISA was prevalent
    > in both, it would be completely different SoCs/chips and even microarchitectures.
    I think you're both right and wrong. While it if obvious the same IC is not going to be used for both servers and lightweight devices, VHDL code is being reused. When they design a new IC, they reuse part of the existing code, modify it and combine it with new code, so it is part reuse and part new. And they are trying to make servers containing Intel Atom processors, which is officially a mobility-oriented processor, which means they are trying to put low-power processors in servers. Companies will reuse when it doesn't compromise another aspect.

    > A single-core SoC would run worse with the popular operating systems,
    Not necessarly, a single core p6600 running at 2ghz would have similar performance to a triple-core i6400 running at 1,33ghz and probably wouldn't cost more. There would be no reason for a board designer intending to run linux on a board to prefer the triple core i6400 or the p6600 before carefully looking at his specific application, at which time, he may find that one of the other is more suitable for him. Also, the p6600 supports hyperthreading so it can be run as multiple slower "virtual cores" if one needs to do so. It is worth noting that enabling SMP makes it more difficult to optimize an operating system's setting for low latency operation.

    > What is the legal structure of this alleged registered "MorphOS Development" company?
    Well, I'm not sure.

    > and more and more tablets use it. Sure Intel and X64 in not dead for long time.
    I made an error and should have said that Windows based computers will remain x86 based. The only reason the number of x86 based tablets is growing is poeple are stupid enough to want to run that windows garbadge operating system. Apple based tablets are ARM based and so are most Android based tablets. Finally, never said x86 will be dead, the office oriented, windows running computer will remain x86 based, there is no reason for this to change.

    > Except in phones Intel chips are efficient for the calculating power they deliver.
    They are not efficient, they are inexpensive, they offered the best performance per dollar, this is the reason why they are popular. They are inexpensive because they are made in huge numbers and they are made in huge numbers because only x86 runs Windows and people are dumb enough to want to run that garbage operating system by the hundreds of millions. ARM is also made in huge numbers and as such has become a contender. However, performance per watt is more important than performance per dollar in mobile applications, it will soon be true also for servers. This means non-x86. Once some non-x86 architecture(s) becomes popular for both servers and mobile devices, other non-windows machines will switch over. It fact it is already happening with the ARM architecture.
  • »19.04.16 - 22:45
    Profile
  • Jim
  • Yokemate of Keyboards
    Yokemate of Keyboards
    Jim
    Posts: 4714 from 2009/1/28
    From: Delaware, USA
    @ Amiga believer

    Personally, I am rather ambivalent about what ISA is running the software I use as long as that software runs well.
    As such MIPS, ARM, whatever.
    We ARE going X64 and we currently support PPCs.

    I see no massive advantage in changing that.
    "Never attribute to malice what can more readily explained by incompetence"
  • »19.04.16 - 23:20
    Profile
  • Yokemate of Keyboards
    Yokemate of Keyboards
    Andreas_Wolf
    Posts: 10096 from 2003/5/22
    From: Germany
    >> A single-core SoC would run worse with the popular operating systems

    > Not necessarly, a single core p6600 running at 2ghz would have similar performance
    > to a triple-core i6400 running at 1,33ghz

    I was comparing single-core SoC vs. multicore SoC using the very same core(s), of course.

    > There would be no reason for a board designer intending to run linux on a board
    > to prefer the triple core i6400 or the p6600

    In most cases, it would be reasonable for a board designer intending to run Linux on his board to prefer a multicore SoC to a single-core SoC, with both using the very same core(s). Usually, the decision for one or the other wouldn't even have an impact on the actual board design as the SoC variants with different core counts are interchangeably usable as pin-compatible drop-in replacements.

    > the p6600 supports hyperthreading

    How "to make sure no one develops software depending on unofficial capabilities which would be removed in successor machines" in this case? :-)
  • »11.12.16 - 13:06
    Profile
  • Jim
  • Yokemate of Keyboards
    Yokemate of Keyboards
    Jim
    Posts: 4714 from 2009/1/28
    From: Delaware, USA
    >> Imagination has stated that the MIPS market is growing.

    yeah...growing legs apparently ;-)
    "Never attribute to malice what can more readily explained by incompetence"
  • »07.05.17 - 00:52
    Profile
  • Yokemate of Keyboards
    Yokemate of Keyboards
    Andreas_Wolf
    Posts: 10096 from 2003/5/22
    From: Germany
    Update:

    > MIPS Tech Inc. sold to Wave Computing: [...]

    "Wave Computing [...] announced it will open source its MIPS instruction set architecture (ISA) to accelerate the ability for semiconductor companies, developers and universities to adopt and innovate using MIPS for next-generation system-on-chip (SoC) designs. Under the MIPS Open program, participants will have full access to the most recent versions of the 32-bit and 64-bit MIPS ISA free of charge – with no licensing or royalty fees."
    https://wavecomp.ai/wave-computing-launches-the-mips-open-initiative/

    More info:
    https://wavecomp.ai/mipsopen/
    https://www.mips.com/mipsopen/
  • »17.12.18 - 22:22
    Profile
  • vox
  • Order of the Butterfly
    Order of the Butterfly
    vox
    Posts: 428 from 2003/11/24
    From: Belgrade
    MorphOS 4.x for selected x64 boards and PPC64 (G5, x5000, Tabor maybe?) with real SMP, 64-bit memory map and modern Radeon drivers and FAT32/NTFS/EXT4 support would be awesome. Yes, OS3 and WOS/PUP/MOS 1-3.x apps could be boxed and Crysalid pack ported to be native. Some AROS x64 backporting only if needed fore some more Amiga x64 apps and one can have stylish and modern OS. Some "dual" application format could be used for installs, like Macs did once or mixed m68k/PPC apps were.

    E-mail client indicate what kind of applications could be developed and userbase could make some OS4 developers "do the blue too" (as some did, Vampire could be m68k flavour where applicable) and some x64 developers might come with proper development tools Vamp and OS4 currently lack of.

    Thats a nice wish, and alternatives are nice too exist, but I dont see where e.g.
    MorphOS 4 for ARM64 would lead without tablet scaling and extensive touch screen support. And would I or you use it then? MIPS would even fare way worse.

    [ Edited by vox 18.12.2018 - 20:41 ]
    ------------------------------------------
    x1000 user, ASAP Vampire Standalone user, future MOS user
  • »18.12.18 - 19:39
    Profile
  • Yokemate of Keyboards
    Yokemate of Keyboards
    Andreas_Wolf
    Posts: 10096 from 2003/5/22
    From: Germany
    > PPC64 (G5, x5000, Tabor maybe?)

    The e500v2 cores inside Tabor's QorIQ P1022 CPU are 32-bit.

    > Crysalid pack [could be] ported to be native.

    ...as long as sources are available.

    > I dont see where e.g. MorphOS 4 for ARM64 would lead without tablet scaling
    > and extensive touch screen support.

    http://linuxgizmos.com/catalog-of-116-open-spec-hacker-boards/

    40 of those are 64-bit ARMv8 (AArch64, "ARM64").


    Edit: new list (43 of those are 64-bit ARMv8)

    [ Edited by Andreas_Wolf 04.01.2019 - 12:26 ]
  • »18.12.18 - 22:27
    Profile
  • vox
  • Order of the Butterfly
    Order of the Butterfly
    vox
    Posts: 428 from 2003/11/24
    From: Belgrade
    Quote:

    Andreas_Wolf wrote:
    > PPC64 (G5, x5000, Tabor maybe?)

    The e500v2 cores inside Tabor's QorIQ P1022 CPU are 32-bit.


    Really? AmigaOS 4 anyway has nothing to do with 64 bit beside patching file sys.

    Quote:


    > Crysalid pack [could be] ported to be native.

    ...as long as sources are available.



    True. Sources and people willing to recompile, bugfix and continue development will be critical.

    Quote:


    > I dont see where e.g. MorphOS 4 for ARM64 would lead without tablet scaling
    > and extensive touch screen support.

    http://linuxgizmos.com/catalog-of-116-open-spec-hacker-boards/

    40 of those are 64-bit ARMv8 (AArch64, "ARM64").



    I do understand ARM64 exists, as well as Linux and Android for it. I just dont see MOS with no touchscreen support there yet. Or even with it. Kind of backward retro enthusiasm - would love to see MOS advance to convenient desktop/laptop system, but not that far.
    Its up to MOS team to decide.

    [ Edited by vox 19.12.2018 - 07:38 ]
    ------------------------------------------
    x1000 user, ASAP Vampire Standalone user, future MOS user
  • »19.12.18 - 06:37
    Profile
  • Yokemate of Keyboards
    Yokemate of Keyboards
    Andreas_Wolf
    Posts: 10096 from 2003/5/22
    From: Germany
    >>> MorphOS 4.x for [...] PPC64 (G5, x5000, Tabor maybe?)

    >> The e500v2 cores inside Tabor's QorIQ P1022 CPU are 32-bit.

    > Really?

    Yes, see https://en.wikipedia.org/wiki/PowerPC_e500.

    > AmigaOS 4 anyway has nothing to do with 64 bit

    You were talking about "MorphOS 4.x" at that point, not about OS4.

    >>> Crysalid pack [could be] ported to be native.

    >> ...as long as sources are available.

    > [...] and people willing to recompile

    It's not even clear if a simple recompile will suffice when porting current MorphOS programs to a future MorphOS with modernized API.

    >>> I dont see where e.g. MorphOS 4 for ARM64 would lead without
    >>> tablet scaling and extensive touch screen support.

    >> http://linuxgizmos.com/catalog-of-116-open-spec-hacker-boards/
    >> 40 of those are 64-bit ARMv8 (AArch64, "ARM64").

    > I do understand ARM64 exists, as well as Linux and Android for it.
    > I just dont see MOS with no touchscreen support there yet.

    What you don't seem to understand is that those 40 "ARM64" boards do not need any touchscreen support, and that anybody is free to port whatever OS to them.

    > Or even with it.

    Why even bring the matter of touchscreen support up then?

    > Kind of backward retro enthusiasm

    I see nothing retro about multiple 2+ GHz Cortex-A72 or newer cores.
  • »19.12.18 - 09:58
    Profile
  • Paladin of the Pegasos
    Paladin of the Pegasos
    Zylesea
    Posts: 1872 from 2003/6/4
    I am still all for the x64 switch as long taerm goal. But as a stop gap I would welcome power9 support (i.e. Raptor blackbird) in the meantime. Would bring us one up to date hardware at a still rather reasonable price tag.
    But x64 is the key for future hardware supply (well designed laptops, powerful desktops).
    And a Power9 machine would make a nice development system for a 64 bit little endian MorphOS. To me the backbird seem like a good step between the generations (from current 32 bit big endian PPC to future 64 bit little endian/endianess indipendant x64/multi ISA system).
    --
    http://www.via-altera.de

    Whenever you're sad just remember the world is 4.543 billion years old and you somehow managed to exist at the same time as David Bowie.
  • »19.12.18 - 13:37
    Profile Visit Website