G4 vs. G5
  • Jim
  • Yokemate of Keyboards
    Yokemate of Keyboards
    Jim
    Posts: 4977 from 2009/1/28
    From: Delaware, USA
    Quote:

    takemehomegrandma wrote:
    Quote:

    Spectre660 wrote:
    Jim wanted one.


    Hardly as an Amigan. Hardly as a Linux user as well (ARM and x86/x64 offers plenty of better options, whatever you need). He is a PPC fetishist, and it still remains to be seen if he will actually put his money where his mouth is in this matter.

    Personally, I doubt it.

    It's a lot of money.


    Actually, it's not that much money, particularly in light of the prices originally commanded by legacy Amigas.

    And I'd never claim to be an Amigan, as that cult is populated by eccentric people who frequently exhibit signs of mental illness.

    My original interest was the 68K family of processor, and Motorola CPUs in general, as they were superior to the Intel processors that were available at their introduction, and yes I like RISC, no apologies there.

    As to what I will or won't do, I'll continue to support MorphOS, as it's a nice, compact OS with a micro kernel. It's relation to Amiga OS actually isn't that important to me.
    SO...I'm a MorphOS user (registered MorphOS user), NOT an Amigan.

    Further, when it comes to 'Putting my money where my mouth is'...I do that regularly with bounty donations, contributions of hardware, and contributions to campaigns like the T2080 laptop project.
    I don't just offer my opinion, grandma, like some of us.

    SO, it might be time to remind you all of how easy it is to be a critic, without also offering anything positive as a counterpoint.

    I know a few of you have taken offense to some of the things Andreas has posted, but you know, I think HE is always balanced, fair, and most importantly factual.
    So, using that as an example, I've tried to follow his lead (not always succeeding).

    Finally, I'm not here to offer anyone advice or act as a model for your emulation. I would not recommend you buy (or not buy) hardware based on my own preferences. My interests in our pursuits is primarily from a hobbyist's perspective, and I don't feel constrained by practicality.
    "Never attribute to malice what can more readily explained by incompetence"
  • »14.09.17 - 10:34
    Profile
  • Butterfly
    Butterfly
    Posts: 80 from 2017/9/10
    Quote:

    And I'd never claim to be an Amigan, as that cult is populated by eccentric people who frequently exhibit signs of mental illness.

    i quote they are made negation of evidences...
  • »14.09.17 - 11:22
    Profile
  • Butterfly
    Butterfly
    Posts: 80 from 2017/9/10
    Quote:

    tlosmx wrote:
    Quote:

    koszer wrote:
    Doesn't dnetc use Altivec when possible?


    yes it use altivec too.
    if some one know how to made a benchmark without i will do it


    i benchmarked all cores with -bench option
    i dont know if it is good . i will do the same on 5040 when i will have the time to go in my mancave





    Quote:

    dnetc v2.9112-521-CFR-16021317 for Linux (Linux 4.10.1).
    Please provide the *entire* version descriptor when submitting bug reports.
    The distributed.net bug report pages are at http://bugs.distributed.net/

    [Sep 14 11:28:23 UTC] Automatic processor type detection found
    a PowerPC 970MP (G5) processor.
    [Sep 14 11:28:23 UTC] OGR-NG: using core #0 (KOGE 3.1 Scalar).
    [Sep 14 11:28:42 UTC] OGR-NG: Benchmark for core #0 (KOGE 3.1 Scalar)
    0.00:00:16.17 [22,497,333 nodes/sec]
    [Sep 14 11:28:42 UTC] OGR-NG: using core #1 (KOGE 3.1 Hybrid).
    [Sep 14 11:29:01 UTC] OGR-NG: Benchmark for core #1 (KOGE 3.1 Hybrid)
    0.00:00:16.17 [42,424,383 nodes/sec]
    [Sep 14 11:29:01 UTC] OGR-NG benchmark summary :
    Default core : #1 (KOGE 3.1 Hybrid) 42,424,383 nodes/sec
    Fastest core : #1 (KOGE 3.1 Hybrid) 42,424,383 nodes/sec
    [Sep 14 11:29:01 UTC] RC5-72: using core #0 (MH 2-pipe).
    [Sep 14 11:29:20 UTC] RC5-72: Benchmark for core #0 (MH 2-pipe)
    0.00:00:16.34 [5,648,705 keys/sec]
    [Sep 14 11:29:20 UTC] RC5-72: using core #1 (KKS 2-pipe).
    [Sep 14 11:29:39 UTC] RC5-72: Benchmark for core #1 (KKS 2-pipe)
    0.00:00:17.09 [6,132,322 keys/sec]
    [Sep 14 11:29:39 UTC] RC5-72: using core #2 (KKS 604e).
    [Sep 14 11:29:59 UTC] RC5-72: Benchmark for core #2 (KKS 604e)
    0.00:00:17.07 [5,593,755 keys/sec]
    [Sep 14 11:29:59 UTC] RC5-72: using core #3 (KKS 7400).
    [Sep 14 11:30:19 UTC] RC5-72: Benchmark for core #3 (KKS 7400)
    0.00:00:17.07 [15,767,442 keys/sec]
    [Sep 14 11:30:19 UTC] RC5-72: using core #4 (KKS 7450).
    [Sep 14 11:30:39 UTC] RC5-72: Benchmark for core #4 (KKS 7450)
    0.00:00:17.06 [18,846,672 keys/sec]
    [Sep 14 11:30:39 UTC] RC5-72: using core #5 (MH 1-pipe).
    [Sep 14 11:30:58 UTC] RC5-72: Benchmark for core #5 (MH 1-pipe)
    0.00:00:17.06 [4,759,204 keys/sec]
    [Sep 14 11:30:58 UTC] RC5-72: using core #6 (MH 1-pipe 604e).
    [Sep 14 11:31:17 UTC] RC5-72: Benchmark for core #6 (MH 1-pipe 604e)
    0.00:00:16.10 [4,743,695 keys/sec]
    [Sep 14 11:31:17 UTC] RC5-72 benchmark summary :
    Default core : #4 (KKS 7450) 18,846,672 keys/sec
    Fastest core : #4 (KKS 7450) 18,846,672 keys/sec
    [Sep 14 11:31:17 UTC] Compare and share your rates in the speeds database at
    http://www.distributed.net/speed/
    (benchmark rates are for a single processor core)

  • »14.09.17 - 11:32
    Profile
  • Jim
  • Yokemate of Keyboards
    Yokemate of Keyboards
    Jim
    Posts: 4977 from 2009/1/28
    From: Delaware, USA
    Quote:

    tlosmx wrote:
    Quote:

    And I'd never claim to be an Amigan, as that cult is populated by eccentric people who frequently exhibit signs of mental illness.

    i quote they are made negation of evidences...



    I'm not sure I get that.
    But I stand by my contention that the Amiga community at large has an unusually large percentage of what can most kindly be called 'odd personalities'.

    As far as 'negation of evidences', so far the only 'evidence' I've seen are some low benchmarks for memory bandwidth.
    Other than that, you've posted a great deal of subjective impressions.

    Further, this isn't really a contest between the X5000 and the quad G5, as the only OS they have in common is Linux.
    And I will agree with grandma here that I am not a big Linux fan.
    Linux uses a monolythic kernel, and I don't find that ideal.
    If I'm going to compromise and use an OS based on one of those, I'll just use Windows, as it's SO much less of a PITA.

    So, again, I'll use Amiga OS, and Linux, BUT I'm NOT a fan.
    I'm a MorphOS user.
    "Never attribute to malice what can more readily explained by incompetence"
  • »14.09.17 - 11:56
    Profile
  • Butterfly
    Butterfly
    Posts: 80 from 2017/9/10
    Quote:

    Jim wrote:
    Quote:

    tlosmx wrote:
    Quote:

    And I'd never claim to be an Amigan, as that cult is populated by eccentric people who frequently exhibit signs of mental illness.

    i quote they are made negation of evidences...



    I'm not sure I get that.
    But I stand by my contention that the Amiga community at large has an unusually large percentage of what can most kindly be called 'odd personalities'.

    As far as 'negation of evidences', so far the only 'evidence' I've seen are some low benchmarks for memory bandwidth.
    Other than that, you've posted a great deal of subjective impressions.

    Further, this isn't really a contest between the X5000 and the quad G5, as the only OS they have in common is Linux.
    And I will agree with grandma here that I am not a big Linux fan.
    Linux uses a monolythic kernel, and I don't find that ideal.
    If I'm going to compromise and use an OS based on one of those, I'll just use Windows, as it's SO much less of a PITA.

    So, again, I'll use Amiga OS, and Linux, BUT I'm NOT a fan.
    I'm a MorphOS user.


    Quote:

    I'm a MorphOS user.


    im not morphos big fan, i appreciate it optimization and how much it is responsive.i like it but for sure i will love it when will be on quad with smp ;-)
    i start appreciate linux when i found a great alternative of osx on my Quad and from that time i start studing it.

    this are couples of my morphos videos hope you will like it. was made from two of my machines.
    the best machine for me was and continue be the Pegasos2 ...
    Im my life i try to be more objective possible.

    video 1
    video 2


    edit: about my x5000 consideration be sure of one thing this machine cant be paragonable in nothing to the performance of g5 quad. and what made this machine a really not good desktop experience is the slow ram. with a faster ram speed im sure everything can be much more better every ware. And ram speed cant be fixed . belive me

    [ Edited by tlosmx 14.09.2017 - 13:49 ]
  • »14.09.17 - 12:41
    Profile
  • Jim
  • Yokemate of Keyboards
    Yokemate of Keyboards
    Jim
    Posts: 4977 from 2009/1/28
    From: Delaware, USA
    Well, the X5000 ought to outperform a Peg2. Even the low memory bandwidth figures you've quoted for the X5000 compare favorably to a G4.
    And the integer performance would definitely be higher, floating point would not have AltiVec, but at almost twice the speed that would also be better.
    Then there is a video slot differences.
    Even limited to four lanes a Gen2 PCI-E connection is going to provide better bandwidth than the modified PCI slot that the Peg2 uses for AGP.
    And the X5000's PCI-E slot can handle modern Radeon HD video cards which the Peg2 can not.
    In fact, TBMK the Peg2 tops out with the Radeon 9800 as it cannot use cards intended for 4x/8x AGP slots (it provides too high a voltage to the AGP slot).
    So while MorphOS support AGP variants of the R400 and R500 video cards (and may eventually provide support for R600 and R700 AGP cards), these can never be used in a Pegasos.

    So, while I have a strong affinity for the Peg2 as a native MorphOS platform, my G4 system is a PowerMac MDD.
    It's faster, has an AGP slot I can force 8x AGP card to work in (by masking two pins), and they are dirt cheap (as well as commonly available).

    I agree with you that the Quad G5 would make a good MorphOS platform, but there has been no commitment to port to that system.
    And a port is being prepared for the X5000.
    So, while an X5000 isn't a G5, it is better than a G4, it has PCI-E expansion, and it's available new.

    And the price isn't that unreasonable, considering the price of a new Mac or the Lenovo laptops I buy.

    [ Edited by Jim 14.09.2017 - 10:56 ]
    "Never attribute to malice what can more readily explained by incompetence"
  • »14.09.17 - 13:46
    Profile
  • Butterfly
    Butterfly
    Posts: 80 from 2017/9/10
    Quote:

    Jim wrote:
    Well, the X5000 ought to outperform a Peg2.

    this is for sure, but the peg2 for me was the best machine made in the amigang hystory.
    probably because made by bplan (phase5)?,
    i like the good choose of open firmware, the cpu book3s and the perfect compatibily.
    the cpu can be upgraded, plus was ready for a dual core module.
  • »14.09.17 - 17:02
    Profile
  • Jim
  • Yokemate of Keyboards
    Yokemate of Keyboards
    Jim
    Posts: 4977 from 2009/1/28
    From: Delaware, USA
    Quote:

    tlosmx wrote:
    Quote:

    Jim wrote:
    Well, the X5000 ought to outperform a Peg2.

    this is for sure, but the peg2 for me was the best machine made in the amigang hystory.
    probably because made by bplan (phase5)?,
    i like the good choose of open firmware, the cpu book3s and the perfect compatibily.
    the cpu can be upgraded, plus was ready for a dual core module.



    Yes, I agree that bPlan hardware had it all over the stuff that was carrying the AmigaOne banner, and the Peg2 got rid of the Artica S, the weak point of early AmigaOne and Peg1 boards.

    I was hoping I could draw Bill Buck back into the community for a final PPC board project, as he still has access to those production facilities, but lack of enthusiasm for the project on Freescale's part helped torpedo that.

    But the Pegasos2 was a high point for our community.
    Open firmware, an upgradable CPU board, and eventually the design details were released to the public.
    "Never attribute to malice what can more readily explained by incompetence"
  • »14.09.17 - 17:41
    Profile
  • Yokemate of Keyboards
    Yokemate of Keyboards
    Andreas_Wolf
    Posts: 12019 from 2003/5/22
    From: Germany
    > yes it use altivec too. if some one know how to made a benchmark without i will do it

    By default, dnetc should run both scalar and vector benchmarks on CPUs with SIMD, and just the scalar ones on CPUs without SIMD.

    dnetc comparison between pampers' X5000/20 and my PowerMac G5 2.3 GHz (=1.00), both with MorphOS (real ratio first, then scaled per-clock ratio):

    0.99 | 1.14 : KOGE 3.1 Scalar
    1.05 | 1.20 : MH 2-pipe
    0.98 | 1.13 : KKS 2-pipe
    1.10 | 1.27 : KKS 604e
    1.22 | 1.40 : MH 1-pipe
    1.20 | 1.38 : MH 1-pipe 604e

    http://morph.zone/modules/newbb_plus/viewtopic.php?forum=11&topic_id=11137&start=593
  • »14.09.17 - 21:29
    Profile
  • Yokemate of Keyboards
    Yokemate of Keyboards
    Andreas_Wolf
    Posts: 12019 from 2003/5/22
    From: Germany
    >> KOGE 3.1 Scalar [...] 19,861,876 nodes/sec
    >> KKS 604e [...] 5,574,672 keys/sec

    > KOGE 3.1 Scalar [...] 22,497,333 nodes/sec
    > KKS 604e [...] 5,593,755 keys/sec

    2.5 GHz PPC970MP (=1.00) vs. 2.0 GHz P5020 (real ratio | scaled per-clock ratio):
    0.90 | 1.12 : KOGE 3.1 Scalar
    0.95 | 1.19 : MH 2-pipe
    0.89 | 1.12 : KKS 2-pipe
    1.00 | 1.25 : KKS 604e
    1.11 | 1.38 : MH 1-pipe
    1.09 | 1.36 : MH 1-pipe 604e

    2.5 GHz PPC970MP (=1.00) vs. 2.2 GHz P5040 (real ratio | scaled per-clock ratio):
    0.99 | 1.12 : KOGE 3.1 Scalar
    1.05 | 1.20 : MH 2-pipe
    0.99 | 1.12 : KKS 2-pipe
    1.11 | 1.26 : KKS 604e
    1.22 | 1.39 : MH 1-pipe
    1.20 | 1.37 : MH 1-pipe 604e

    Unsurprisingly, your scaled per-clock ratios are similar to my results (see previous comment). Yours are slightly lower because your PPC970MP is slightly faster per clock than my PPC970FX.


    Edit: used more complete figures

    [ Edited by Andreas_Wolf 23.09.2017 - 21:47 ]
  • »14.09.17 - 21:58
    Profile
  • Yokemate of Keyboards
    Yokemate of Keyboards
    Andreas_Wolf
    Posts: 12019 from 2003/5/22
    From: Germany
    > the peg2 for me was the best machine made in the amigang hystory. [...]
    > the cpu can be upgraded

    Upgraded to what? Maybe buyers of the cheaper 600 MHz G3 model could upgrade to 1 GHz G4, but CPU modules weren't retailed, so they had to search the secondhand market and hope for G4 users with broken mainboards. There was a handful of 1.4 GHz G4 modules produced but not openly sold.

    > plus was ready for a dual core module.

    Never existed.
  • »14.09.17 - 22:40
    Profile
  • Yokemate of Keyboards
    Yokemate of Keyboards
    Andreas_Wolf
    Posts: 12019 from 2003/5/22
    From: Germany
    > a few of you have taken offense to some of the things Andreas has posted

    Huh? What have I missed? :-)
  • »14.09.17 - 22:56
    Profile
  • Jim
  • Yokemate of Keyboards
    Yokemate of Keyboards
    Jim
    Posts: 4977 from 2009/1/28
    From: Delaware, USA
    Quote:

    Andreas_Wolf wrote:
    > a few of you have taken offense to some of the things Andreas has posted

    Huh? What have I missed? :-)


    I think some people find your direct manner off-putting. 😁

    I have always found you well informed and unwilling to compromise principles.

    That is important to me.
    And you're a great resource for info.
    "Never attribute to malice what can more readily explained by incompetence"
  • »14.09.17 - 23:04
    Profile
  • Yokemate of Keyboards
    Yokemate of Keyboards
    Andreas_Wolf
    Posts: 12019 from 2003/5/22
    From: Germany
    > so far the only 'evidence' I've seen are some low benchmarks for memory
    > bandwidth. [...] this isn't really a contest between the X5000 and the quad G5,
    > as the only OS they have in common is Linux.

    As you know, the Cyrus/X5000 memory performance doesn't come off significantly better if compared to AGP-based G5 PowerMacs with benchmarks completely done on MorphOS.

    > monolythic kernel [...] I don't find that ideal. If I'm going to [...] use an OS based
    > on one of those, I'll just use Windows

    "the kernel performs almost all the tasks of a traditional microkernel; the strict distinction between Executive and Kernel is the most prominent remnant of the original microkernel design, and historical design documentation consistently refers to the kernel component as "the microkernel"."
    https://en.wikipedia.org/wiki/Architecture_of_Windows_NT#Kernel

    ;-)
  • »14.09.17 - 23:09
    Profile
  • Yokemate of Keyboards
    Yokemate of Keyboards
    Andreas_Wolf
    Posts: 12019 from 2003/5/22
    From: Germany
    > ram speed cant be fixed . belive me

    So you already know it's not an U-Boot issue?
  • »14.09.17 - 23:17
    Profile
  • Jim
  • Yokemate of Keyboards
    Yokemate of Keyboards
    Jim
    Posts: 4977 from 2009/1/28
    From: Delaware, USA
    Quote:

    Andreas_Wolf wrote:
    > so far the only 'evidence' I've seen are some low benchmarks for memory
    > bandwidth. [...] this isn't really a contest between the X5000 and the quad G5,
    > as the only OS they have in common is Linux.

    As you know, the Cyrus/X5000 memory performance doesn't come off significantly better if compared to AGP-based G5 PowerMacs with benchmarks completely done on MorphOS.

    > monolythic kernel [...] I don't find that ideal. If I'm going to [...] use an OS based
    > on one of those, I'll just use Windows

    "the kernel performs almost all the tasks of a traditional microkernel; the strict distinction between Executive and Kernel is the most prominent remnant of the original microkernel design, and historical design documentation consistently refers to the kernel component as "the microkernel"."
    https://en.wikipedia.org/wiki/Architecture_of_Windows_NT#Kernel

    ;-)


    Interesting quote.
    There was a period, about the time CDi was trying to make headway, that Microsoft was rumored to be thinking about shifting toward a micro kernel based OS. And OSX has a curious hybrid core in Mach.
    Linux actually seems the most primitive in that it's kernel was become a monster.

    IF someone took the idea serious, MorphOS (or something based on it) could actually be upgraded to compete with these operating systems.
    "Never attribute to malice what can more readily explained by incompetence"
  • »14.09.17 - 23:40
    Profile
  • Yokemate of Keyboards
    Yokemate of Keyboards
    Andreas_Wolf
    Posts: 12019 from 2003/5/22
    From: Germany
    >>> a few of you have taken offense to some of the things Andreas has posted

    >> Huh? What have I missed? :-)

    > I think some people find your direct manner off-putting.

    Honestly, I'm not quite sure what warrants making me into a subject of this thread out of nowhere. This thread has been about "G4 vs. G5" originally and quickly turned into a thread about the Cyrus/X5000. I've been trying to clear some misunderstandings and questions in this thread, in particular with regards to the Cyrus board design, by offering technical information. I can't remember anybody has taken offense to anything of that, quite to the contrary.
  • »15.09.17 - 00:25
    Profile
  • Yokemate of Keyboards
    Yokemate of Keyboards
    Andreas_Wolf
    Posts: 12019 from 2003/5/22
    From: Germany
    > OSX has a curious hybrid core in Mach.

    Yes, Mach/OSFMK is the low-level microkernel part of the hybrid XNU kernel (with the FreeBSD kernel delivering the higher-level monolithic part).
  • »15.09.17 - 00:37
    Profile
  • Butterfly
    Butterfly
    Posts: 80 from 2017/9/10
    Quote:

    Andreas_Wolf wrote:
    > ram speed cant be fixed . belive me

    So you already know it's not an U-Boot issue?


    Quote:

    So you already know it's not an U-Boot issue?

    from where the report that i have come im pratically 99.99% sure.
  • »15.09.17 - 03:13
    Profile
  • Yokemate of Keyboards
    Yokemate of Keyboards
    Andreas_Wolf
    Posts: 12019 from 2003/5/22
    From: Germany
    >> So you already know it's not an U-Boot issue?

    > from where the report that i have come im pratically 99.99% sure.

    And does your source also give away what the actual cause of the problem is? The SoC's memory controller? The board design? Anything else?
  • »15.09.17 - 08:20
    Profile
  • Order of the Butterfly
    Order of the Butterfly
    Spectre660
    Posts: 275 from 2015/6/30
    Optimally Configuring DDR for Custom Boards
    http://www.nxpic.org/document/download/output/id-16447
  • »15.09.17 - 09:53
    Profile
  • Butterfly
    Butterfly
    Posts: 80 from 2017/9/10
    Quote:

    Spectre660 wrote:
    Optimally Configuring DDR for Custom Boards
    http://www.nxpic.org/document/download/output/id-16447



    on document is explained the ddr3 on qoriq will not go faster than 3.5 gb/s like a ddr 400
    where on other architectures it run at 12,80 gb/s for channel and 25,60 in dual channel.
  • »15.09.17 - 10:35
    Profile
  • Yokemate of Keyboards
    Yokemate of Keyboards
    Andreas_Wolf
    Posts: 12019 from 2003/5/22
    From: Germany
    > on document is explained the ddr3 on qoriq will not go faster than 3.5 gb/s like a ddr 400

    On which page?
  • »15.09.17 - 10:55
    Profile