Mirari ppc board - *when/if* available, a new hardware for MorphOS
  • Priest of the Order of the Butterfly
    Priest of the Order of the Butterfly
    matt3
    Posts: 747 from 2004/2/10
    Quote:

    Cego wrote:
    E6500 looks interested with Altivec 2.0 support. With accelerated video decoding this could be a very nice MorphOS setup. With a price tag at <1000 Euro i would consider buying one.


    Hmmm, I thought it didn't have Altivec. If that is the case, would be a great box for MorphOS.

    I definitely will be buying one, when MorphOS is available for it.
  • »02.05.25 - 20:58
    Profile
  • Butterfly
    Butterfly
    Posts: 72 from 2023/12/1
    The higher end CPU option appears to have Altivec per their description.

    -T10x2 (2-4 e5500 cores) up to 1.5GHz / T2081 (4 dual-threaded e6500 cores with Altivec 2.0) up to 1.8GHz

    TJ


    Quote:

    matt3 wrote:
    Quote:

    Cego wrote:
    E6500 looks interested with Altivec 2.0 support. With accelerated video decoding this could be a very nice MorphOS setup. With a price tag at <1000 Euro i would consider buying one.


    Hmmm, I thought it didn't have Altivec. If that is the case, would be a great box for MorphOS.

    I definitely will be buying one, when MorphOS is available for it.




    [ Edited by amigasociety 02.05.2025 - 13:30 ]
  • »02.05.25 - 21:29
    Profile Visit Website
  • Cocoon
    Cocoon
    Posts: 45 from 2018/6/9
    Definitely keeping an eye on this one. Obviously it needs software support, but looks interesting for sure.
  • »02.05.25 - 21:54
    Profile
  • Yokemate of Keyboards
    Yokemate of Keyboards
    Andreas_Wolf
    Posts: 12385 from 2003/5/22
    From: Germany
    > how will it perform in comparison with current supported hardware.

    At least the T1042 version can be directly compared to the X5000's P50x0 based on the clock rate as it's the same e5500 core (in 28nm rather than 45nm process node, though).
    And while the e6500 is slightly faster per clock than the e5500 in scalar code and has AltiVec SIMD in addition, it's maxing out at 1.8 GHz compared to the 2.2 GHz of the X5000/40's e5500. I wouldn't be too surprised if 2.2 GHz e5500 proves faster in real world usage than 1.8 GHz e6500.
  • »03.05.25 - 03:25
    Profile
  • Paladin of the Pegasos
    Paladin of the Pegasos
    ThePlayer
    Posts: 1074 from 2003/3/24
    From: Hamburg/Germany
    Sounds interesting. But how much will the Board cost?
    And how fast will it be compared to an G5 with 2.5Ghz?
    Found a comparison between G5 and e6500 but don't understand if lower or higher values are better.
    Here are the Benchmarks
    PowerMac G5 Quad 2.5 running UWQHD Resolution
  • »03.05.25 - 13:34
    Profile
  • Moderator
    Kronos
    Posts: 2439 from 2003/2/24
    Quote:

    ThePlayer wrote:

    Found a comparison between G5 and e6500 but don't understand if lower or higher values are better.
    Here are the Benchmarks


    Some so some so...

    Some are scores (higher == better) other are times (lower == better) but you can't really gain much of this as it is 4 double threaded cores vs. 2 single threaded one.
    So would the performance of 1 core running 1 thread 1/8, 1/4 or maybe something else?

    It's pretty much the same core as used in the X5000 so disregarding Altivec about the same as a G5 running at the same clock.
  • »03.05.25 - 13:52
    Profile
  • Paladin of the Pegasos
    Paladin of the Pegasos
    koszer
    Posts: 1333 from 2004/2/8
    From: Poland
    Quote:

    ThePlayer wrote:
    Found a comparison between G5 and e6500 but don't understand if lower or higher values are better.
    Here are the Benchmarks


    The results are puzzling indeed.
    From what I've found it's

    CPU Blowfish: higher = better (but G5 beating both e6500 and i7-4600U to the ground? Hmm...)
    CPU CryptoHash: higher = better (both e6500 and i7 way faster than G5)
    CPU Fibonacci (s): lower = better (both PPCs almost on par, e6500 slightly faster. i7 much faster)
    CPU N-Queens (s): lower = better (e6500 comes first, then i7. G5 1,8 far behind)
    CPU Zlib: higher = better (G5 slowest, e6500 better, i7 wins)
    FPU FFT: higher = better (i7 worst, then e6500, and G5 wins)
    FPU Raytracing (s): lower = better (G5 fails miserably, then e6500, and i7 just a bit better)
    GPU Drawing: higher = better (Radeon 9600 is the slowest, then RadeonHD 5450 and Intel HD Graphics 4400 wins)
  • »03.05.25 - 13:57
    Profile
  • Paladin of the Pegasos
    Paladin of the Pegasos
    ThePlayer
    Posts: 1074 from 2003/3/24
    From: Hamburg/Germany
    Thx Kronos and koszer.
    So the E6500 won't be much faster as the G5 in general. Just in some tasks.
    I hope that at least the power efficiency will be much better.
    What i need is a machine that is capable of running MorphOS and being much faster when it comes to browsing the internet then the G5.
    PowerMac G5 Quad 2.5 running UWQHD Resolution
  • »03.05.25 - 19:48
    Profile
  • Priest of the Order of the Butterfly
    Priest of the Order of the Butterfly
    eliot
    Posts: 633 from 2004/4/15
    I would l like to have a new modern PPC board for a effordable prize.
    My Mac Mini is dying and has too less vram, the Mac MDD is too loud and the Power Mac G5 consumes
    too much electricity.

    So having a new board, which provides modern mass storage devices in (m)itx case with a slightly newer gfx card
    and with the speed of a G4 would be very cool.

    Go 4 it!!!!
    regards
    eliot
  • »04.05.25 - 09:41
    Profile
  • Yokemate of Keyboards
    Yokemate of Keyboards
    Andreas_Wolf
    Posts: 12385 from 2003/5/22
    From: Germany
    > I hope that at least the power efficiency will be much better.

    Yes, e6500 is much more power efficient than PPC970.

    > What i need is a machine that is [...] much faster when it comes
    > to browsing the internet then the G5.

    Won't happen with e6500.
  • »04.05.25 - 14:49
    Profile
  • Yokemate of Keyboards
    Yokemate of Keyboards
    Andreas_Wolf
    Posts: 12385 from 2003/5/22
    From: Germany
    > a new board [...] in (m)itx case

    This microATX board won't fit in such case.
  • »04.05.25 - 15:04
    Profile
  • Priest of the Order of the Butterfly
    Priest of the Order of the Butterfly
    Cego
    Posts: 738 from 2006/5/28
    From: Germany
    hmm, Mirari board arriving in Dortmund maybe a hint towards a MorphOS port :)
    Pegasos II G4 @1.0GHz, 1GB DDR Ram, Radeon 9200Pro, 240GB SSD+160GB HD, MorphOS 3.18, AmigaOS4.1 FE, Debian 8
  • »17.05.25 - 16:17
    Profile
  • Priest of the Order of the Butterfly
    Priest of the Order of the Butterfly
    matt3
    Posts: 747 from 2004/2/10
    Quote:

    Cego wrote:
    hmm, Mirari board arriving in Dortmund maybe a hint towards a MorphOS port :)


    Well there are a few more pictures on the MorphOS Discord that confirms that.

    MorphOS is being ported!
  • »17.05.25 - 19:14
    Profile
  • Priest of the Order of the Butterfly
    Priest of the Order of the Butterfly
    Cego
    Posts: 738 from 2006/5/28
    From: Germany
    any invitation link? :)
    Pegasos II G4 @1.0GHz, 1GB DDR Ram, Radeon 9200Pro, 240GB SSD+160GB HD, MorphOS 3.18, AmigaOS4.1 FE, Debian 8
  • »17.05.25 - 20:45
    Profile
  • Yokemate of Keyboards
    Yokemate of Keyboards
    takemehomegrandma
    Posts: 2725 from 2003/2/24
    Quote:

    matt3 wrote:
    Well there are a few more pictures on the MorphOS Discord that confirms that.

    MorphOS is being ported!


    Well…


    (Click!)


    :-)
    MorphOS is Amiga done right! :-)
    MorphOS NG will be AROS done right! :-)
  • »23.05.25 - 10:06
    Profile
  • Butterfly
    Butterfly
    SkulleateR
    Posts: 69 from 2023/9/19
    Now that was fast :-D
  • »23.05.25 - 11:41
    Profile
  • Yokemate of Keyboards
    Yokemate of Keyboards
    takemehomegrandma
    Posts: 2725 from 2003/2/24
    This is not completely uninteresting. ;-) Cool that MorphOS runs on it already. Increadible job by Frank as always. :-)

    Anyone knows if the T2081 variant will be actually be realized? And the price of it? I might actually be a little interested in that.

    Anyway, it's cool to see how this board has progressed. And that there is no pre-sale going on with coupons or empty promises or such. This is exactly how things *should be done*; the devs creates a product and when it's ready it's being put on the market.

    Look at that never ending story about that laptop motherboard. The entire idea with that seems to be raising money without making any products. The only things delivered are new excuses and new fund raising goals. Like so many things in the Amiga history. And the Tabor is the biggest joke.

    But these guys just develops and delivers. Well, not to end-users yet, but at least the thing seems to work and MorphOS is up and running! :-) Only remaining issue is the economy thing... ;-)
    MorphOS is Amiga done right! :-)
    MorphOS NG will be AROS done right! :-)
  • »23.05.25 - 14:57
    Profile
  • rob
  • Acolyte of the Butterfly
    Acolyte of the Butterfly
    rob
    Posts: 140 from 2008/7/22
    Quote:

    Anyone knows if the T2081 variant will be actually be realized?


    It should just be cases of soldering a T2081 to the board instead of the T1042. No idea about the board in general, the T2081NSN8TTB is the least expensive 1.8Ghz part at just over $200, so it should be about a $100 difference between the two variants.
  • »23.05.25 - 21:25
    Profile
  • Yokemate of Keyboards
    Yokemate of Keyboards
    takemehomegrandma
    Posts: 2725 from 2003/2/24
    Quote:

    rob wrote:
    Quote:

    Anyone knows if the T2081 variant will be actually be realized?


    It should just be cases of soldering a T2081 to the board instead of the T1042. No idea about the board in general, the T2081NSN8TTB is the least expensive 1.8Ghz part at just over $200, so it should be about a $100 difference between the two variants.




    Yes I understand that the possibility exists, at least in theory. But it seems that none of the prototypes has this CPU yet though, so it still hasn’t been tested in practice? And maybe I’m marked by passed experiences but haven’t we read so many times about specs of upcoming projects where it says “up to” this and that performance, clock speed etc, only to fall short in the end, only offering a less exciting option for economical reasons or whatever?

    But the point is that all PPC options that still exists are rather weak by todays standards, as we notice in CPU intensive applications like web browsing, so there is absolutely no reason to choose anything less than the most powerful option. We will need the highest possible clockspeed, we will need Altivec. So in this case, anything less than the 1.8GHz T2081 would be uninteresting IMHO…
    MorphOS is Amiga done right! :-)
    MorphOS NG will be AROS done right! :-)
  • »24.05.25 - 10:32
    Profile