A look at the new MorphOS
  • Order of the Butterfly
    Order of the Butterfly
    zukow
    Posts: 466 from 2005/2/9
    From: Poland
    The question is on which hemisphere? :)
  • »19.03.18 - 08:42
    Profile Visit Website
  • Priest of the Order of the Butterfly
    Priest of the Order of the Butterfly
    koszer
    Posts: 813 from 2004/2/8
    From: Poland
    Quote:

    zukow wrote:
    The question is on which hemisphere? :)


    Not much time left if we're talking about the southern one then, eh?
  • »19.03.18 - 09:29
    Profile
  • Acolyte of the Butterfly
    Acolyte of the Butterfly
    Tom01
    Posts: 146 from 2009/9/20
    Quote:

    Andreas_Wolf wrote:
    > IMHO there is little practical point in comparing non-SIMD performance,
    > since the 970 *does* have Altivec.

    Most tasks can‘t be vectorized, so comparing CPUs by their non-SIMD performance does make sense.


    Tasks, which are calculating something can be vectorized.
    If a Program mostly consists of the User-Interface, then Vectorization makes no Sense.
  • »19.03.18 - 15:44
    Profile Visit Website
  • Yokemate of Keyboards
    Yokemate of Keyboards
    Andreas_Wolf
    Posts: 9838 from 2003/5/22
    From: Germany
    >> Most tasks can‘t be vectorized

    > Tasks, which are calculating something can be vectorized. If a Program mostly
    > consists of the User-Interface, then Vectorization makes no Sense.

    Of course, 1+2 can be computed by the AltiVec unit. But vectorization is for SIMD use cases, which means the very same computation is applied to a sufficiently large number of elements. Too small a number may mean that the overhead exceeds the gain. And often, the bottleneck is not in the computation itself (but in I/O speed etc.). So I should better say that most tasks, even if vectorizable in theory, wouldn't benefit from vectorization in practice.
  • »19.03.18 - 16:23
    Profile
  • JKD
  • Order of the Butterfly
    Order of the Butterfly
    JKD
    Posts: 448 from 2003/4/4
    From: South of heaven
    Quote:

    Acill wrote:
    Cant wait for this to be released! The beta on my X5000 was fantastic to use during Amiwest. I will be keeping my G5 as well for it I think, just much easier to boot into a single machine and not worry about corruption from the other OS reading the drives.


    How far away are we from release Paul?
  • »23.03.18 - 16:50
    Profile
  • Order of the Butterfly
    Order of the Butterfly
    DanicaTalos
    Posts: 361 from 2010/10/15
    From: 01101110011100...
    A PM G5 PCI-e Version is better. A x5000 costs horrible! A PM G5 PCI-e costs ca. 150,- - 200,-.

    [ Editiert durch DanicaTalos 25.03.2018 - 17:15 ]
    Mac Mini G4 -1,50 GHz - 64MB VRAM - 1GB RAM - 60GB SSD - MorphOS 3.8
    PowerBook G4 - 1,67 GHz - 128MB VRAM - 2GB RAM - 64GB SSD - Kubuntu 12.04
  • »25.03.18 - 10:49
    Profile
  • Moderator
    Kronos
    Posts: 1766 from 2003/2/24
    Quote:

    JKD wrote:

    How far away are we from release Paul?




    I'd say about 20km, but that number may differ for you ;)
    --------------------- May the 4th be with you ------------------
    Mother Russia dance of the Zar, don't you know how lucky you are
  • »25.03.18 - 11:32
    Profile
  • Moderator
    Kronos
    Posts: 1766 from 2003/2/24
    Quote:

    Kronos wrote:
    Quote:

    JKD wrote:

    How far away are we from release Paul?




    I'd say about 20km, but that number may differ for you ;)


    So now we know that MorphOS was traveling at about 2.5km/h !!
    --------------------- May the 4th be with you ------------------
    Mother Russia dance of the Zar, don't you know how lucky you are
  • »25.03.18 - 19:44
    Profile