G4 vs. G5
  • Yokemate of Keyboards
    Yokemate of Keyboards
    Andreas_Wolf
    Posts: 9293 from 2003/5/22
    From: Germany
    > I already have Linux running on a Quad 2.5 GHz G5, [...]
    > and the video card support isn't quite as good as the X5000

    Video card support of Linux on X5000 is better than video card support of Linux on PowerMac G5?
  • »08.09.17 - 16:41
    Profile
  • Yokemate of Keyboards
    Yokemate of Keyboards
    takemehomegrandma
    Posts: 2413 from 2003/2/24
    Quote:

    Andreas_Wolf wrote:
    > wrong hardware topographic

    What does that mean?


    No idea. Maybe something Italian? ;-)


    Quote:

    > pcie 16x with real performances of a 4x

    Considering that only 4 lanes are connected, this shouldn't come as a surprise.


    I know this, since I have read the product description of the chip. I think the confusion (with the following disappointment) comes from the fact that all AeonKit marketing material and product description speaks about PCIe 16x -period-, not at all mentioning the fact that the only "PCIe 16x"-thing about it, is the shape and size of the socket. This might mislead (surprise, surprise) people to believe that they will actually get a PCIe 16x enabled motherboard/system.

    [ Edited by takemehomegrandma 08.09.2017 - 21:02 ]
    MorphOS is Amiga done right! :-)
    MorphOS NG will be AROS done right! :-)
  • »08.09.17 - 19:54
    Profile
  • Jim
  • Yokemate of Keyboards
    Yokemate of Keyboards
    Jim
    Posts: 4251 from 2009/1/28
    From: Delaware, USA
    Quote:

    Andreas_Wolf wrote:
    > I already have Linux running on a Quad 2.5 GHz G5, [...]
    > and the video card support isn't quite as good as the X5000

    Video card support of Linux on X5000 is better than video card support of Linux on PowerMac G5?


    Yes, getting Radeon HD cards working on PCI-E G5 has been a real struggle. I got some pointers from Spectre660 and sorted it out, but one thing that Aeon has done is encourage the development of Linux distros specific to their products.
    Of course, neither the X5000 or the G5 support GCN cards under Linux (at least not at the moment), but the 5000 and 6000 series cards work well.
    Looking at our future support, I might downgrade to a Radeon HD 4890 if I buy an X5000, in order to have overlay support.
    I'd rather not, as the 5000s are a significant improvement, but it is what it is.
    "Magnetic was troubled by my avatars and 'satanic' references" - Jim Igou

    "Never attribute to malice what can more readily explained by incompetence"
  • »08.09.17 - 20:52
    Profile
  • Yokemate of Keyboards
    Yokemate of Keyboards
    Andreas_Wolf
    Posts: 9293 from 2003/5/22
    From: Germany
    > all AeonKit marketing material and product description speaks about PCIe 16x -period-,
    > not at all mentioning the fact that the only "PCIe 16x"-thing about it, is the shape and
    > size of the socket.

    There's been at least one product description in an A-Eon press release (11/2016, linked to from there) that mentions the fact.
  • »08.09.17 - 23:27
    Profile
  • Yokemate of Keyboards
    Yokemate of Keyboards
    Andreas_Wolf
    Posts: 9293 from 2003/5/22
    From: Germany
    > neither the X5000 or the G5 support GCN cards under Linux

    I think they are supported in everything but 3D.
  • »09.09.17 - 00:01
    Profile
  • Just looking around
    marmotta
    Posts: 20 from 2012/6/11
    This is the x5000 topographic, from tlosm. He cant register here because website registration issue

    X5000 Topographic
  • »09.09.17 - 09:07
    Profile
  • Yokemate of Keyboards
    Yokemate of Keyboards
    Andreas_Wolf
    Posts: 9293 from 2003/5/22
    From: Germany
    >>> wrong hardware topographic

    >> What does that mean?

    > X5000 Topographic

    Thanks. And what's wrong with this topology?
  • »09.09.17 - 14:16
    Profile
  • Cocoon
    Cocoon
    Posts: 46 from 2017/9/10
    finally here! i dont know why this website dont like hotmail or yahoo.com

    ok letz explain the topo.
    first the two socs (?) i hope are the sosc not sure have the same id and float around someware
    the first controller is is the 16x aka 4x pcie is where is the gpu
    the second controller have all the bridge put on it ... with bridge over a bridge over a bridge over a bridge...
    the third controller pcie controller they forget to use.

    The L3 cache is a fake cache it is used only for net packet and not for data... this was explained by the mantainer of the uboot
    it is called corenet.

    by the way the cpu have good integer compared the G5 970MP but overall deskop performances are choppy, slow expecially after some times. i think becasue havy ram granulation.

    plus steam report only max 2gb/s for 5020 and 4 gb/s for 5040 where the ddr3 1600 is capable on other hardware to run at 22 Gb/s

    fastest gpu dont help too much compared slowed one i test various NI board with not good big differences in result.

    NOTE:All is about linux ... i dont consider amigaos because i cant compare amigaos with faster powerpc hw.
    for sure now the faster cpu is the e5500 for amigaos.

    only mos guys can compare morphos under X5000 and G5 if sow differences in better or worst performances.

    sorry my english is not perfect.
    Im waithing the day when mos will be available for G5 Quad and RadeonHD :P

    [ Edited by tlosmx 10.09.2017 - 12:53 ]
  • »10.09.17 - 12:08
    Profile
  • ASiegel
    Posts: 885 from 2003/2/15
    From: Central Europe
    Quote:

    tlosmx wrote:
    finally here! i dont know why this website dont like hotmail or yahoo.com

    Because before blocking those, 90+ percent of all spam registrations used either. (Yahoo was especially bad and heaviliy abused by what appeared to be Chinese spam bots.)

    Considering that well below one percent of valid MorphZone users used Yahoo! or Hotmail email accounts, this seemed like a no-brainer.
  • »10.09.17 - 12:43
    Profile
  • Cocoon
    Cocoon
    Posts: 46 from 2017/9/10
    now lets explain why i buy the x5000/40
    before i had a pegasos2 1266mhz ... for me the best machine made for an amigan because linux compatible with mol-kvm, morphos, amigaos.

    take the x5000/40, (the 3 october 2016 was to me) because on paper it specs will made my g5 quad goes in pension and will gave me linux, amigaos and morphos togheter again on the same machine.
    but this was not true ... from very beginning i report to aeon the bed video gpu performances ... result they change the document specs of x5000.

    g5 quad continue is the best powerpc desktop machine made .


    about
    Quote:

    Video card support of Linux on X5000 is better than video card support of Linux on PowerMac G5?

    on linux ppc you can have all the board available on linux x86 ... only mesa are problematic because endianess in SI sourcess.
    the same gpus work on G5 and on x5000 but better on g5.
    i tested on my quad radeonhd 4.xxx,5.xxx,6.xxx an 7.xxxx, r230 all are working . only r7 is not running but can be only a firmware issue not a pcie compatibility issue.

    [ Edited by tlosmx 10.09.2017 - 13:07 ]
  • »10.09.17 - 13:00
    Profile
  • Jim
  • Yokemate of Keyboards
    Yokemate of Keyboards
    Jim
    Posts: 4251 from 2009/1/28
    From: Delaware, USA
    Quote:

    tlosmx wrote:..Im waiting the day when mos will be available for G5 Quad and RadeonHD :P


    A lot of people have asked for that. I have one, the cooling system is a bit of a PITA, but it makes a good Linux platform.
    As Kronos pointed out in the TalosII thread, the quad G5 would make a good base for the next version of MorphOS.

    BUT, many of the developers think we are supporting too many systems already.
    "Magnetic was troubled by my avatars and 'satanic' references" - Jim Igou

    "Never attribute to malice what can more readily explained by incompetence"
  • »10.09.17 - 13:34
    Profile
  • Cocoon
    Cocoon
    Posts: 46 from 2017/9/10
    Quote:

    BUT, many of the developers think we are supporting too many systems already.

    true ... but the quad will the best one ;-)
  • »10.09.17 - 14:00
    Profile
  • Jim
  • Yokemate of Keyboards
    Yokemate of Keyboards
    Jim
    Posts: 4251 from 2009/1/28
    From: Delaware, USA
    That is what everyone said when then 2.7 GHz AGP models we're supported, and the slower G5s became more popular because they were air cooled.
    I'd feel more comfortable with the quad if there was an easy way to convert it to air cooling.

    [ Edited by Jim 10.09.2017 - 12:17 ]
    "Magnetic was troubled by my avatars and 'satanic' references" - Jim Igou

    "Never attribute to malice what can more readily explained by incompetence"
  • »10.09.17 - 14:23
    Profile
  • Yokemate of Keyboards
    Yokemate of Keyboards
    Andreas_Wolf
    Posts: 9293 from 2003/5/22
    From: Germany
    >>> X5000 Topographic

    >> Thanks. And what's wrong with this topology?

    > first the two socs (?) i hope are the sosc not sure have the same id and float around
    > someware

    There's only one SoC on the Cyrus/X5000.

    > the second controller have all the bridge put on it ...
    > with bridge over a bridge over a bridge over a bridge...

    Actually, it's one PCIe-PCIe bridge with another PCIe-PCI bridge (for the PCI slots) attached to it.

    http://morph.zone/modules/newbb_plus/viewtopic.php?forum=11&topic_id=11137&start=712

    > the third controller pcie controller they forget to use.

    I doubt Varisys simply forgot to use more than two PCIe controllers, but believe there is a good reason for using only two of them. It might be connected to the fact that the board has been designed for three different SoCs (P3041, P5020, P5040) which are not 100% pin-compatible, so the actual board design represents the lowest common denominator between the three SoCs.

    http://morph.zone/modules/newbb_plus/viewtopic.php?forum=3&topic_id=7001&start=893

    Considering the fact that the P3041 version was abandoned anyway, they might as well have selected the P5021 instead of the P5020, as the P5021 and the P5040 are 100% pin-compatible.

    > The L3 cache is a fake cache it is used only for net packet and not for data...

    Interesting. I didn't know that the CoreNet L3 cache can't be used also like a normal L3 cache. But even then this has nothing to do with the board design, don't you think?

    > overall deskop performances are choppy, slow expecially after some times.
    > i think becasue havy ram granulation.

    Do you think this is a board design issue?

    > steam report only max 2gb/s for 5020 and 4 gb/s for 5040

    Do you think this is a board design issue?
  • »10.09.17 - 14:34
    Profile
  • Yokemate of Keyboards
    Yokemate of Keyboards
    Andreas_Wolf
    Posts: 9293 from 2003/5/22
    From: Germany
    > from very beginning i report to aeon the bed video gpu performances ...
    > result they change the document specs of x5000.

    Interesting. Which specs in particular did they change after your reporting?

    > g5 quad continue is the best powerpc desktop machine made .

    Together with the TerraSoft/Fixstars/YDL PowerStation, I guess :-)

    >> Video card support of Linux on X5000 is better than video card support of
    >> Linux on PowerMac G5?

    > the same gpus work on G5 and on x5000 but better on g5.

    Isn't that a contradiction to what Jim said in comments #50 and #53? :-)

    > on my quad [...] only r7 is not running

    Isn't that a contradiction to what you just said yourself? :-)
  • »10.09.17 - 14:51
    Profile
  • Cocoon
    Cocoon
    Posts: 46 from 2017/9/10
    Quote:

    Andreas_Wolf wrote:
    >>> X5000 Topographic

    >> Thanks. And what's wrong with this topology?

    > first the two socs (?) i hope are the sosc not sure have the same id and float around
    > someware

    There's only one SoC on the Cyrus/X5000.

    > the second controller have all the bridge put on it ...
    > with bridge over a bridge over a bridge over a bridge...

    Actually, it's one PCIe-PCIe bridge with another PCIe-PCI bridge (for the PCI slots) attached to it.

    http://morph.zone/modules/newbb_plus/viewtopic.php?forum=11&topic_id=11137&start=712

    > the third controller pcie controller they forget to use.

    I doubt Varisys simply forgot to use more than two PCIe controllers, but believe there is a good reason for using only two of them. It might be connected to the fact that the board has been designed for three different SoCs (P3041, P5020, P5040) which are not 100% pin-compatible, so the actual board design represents the lowest common denominator between the three SoCs.

    http://morph.zone/modules/newbb_plus/viewtopic.php?forum=3&topic_id=7001&start=893

    Considering the fact that the P3041 version was abandoned anyway, they might as well have selected the P5021 instead of the P5020, as the P5021 and the P5040 are 100% pin-compatible.

    > The L3 cache is a fake cache it is used only for net packet and not for data...

    Interesting. I didn't know that the CoreNet L3 cache can't be used also like a normal L3 cache. But even then this has nothing to do with the board design, don't you think?

    > overall deskop performances are choppy, slow expecially after some times.
    > i think becasue havy ram granulation.

    Do you think this is a board design issue?

    > steam report only max 2gb/s for 5020 and 4 gb/s for 5040

    Do you think this is a board design issue?


    oh mama to much question.
    >There's only one SoC on the Cyrus/X5000.
    ... bad because it made the topo more strange than before,

    >one PCIe-PCIe bridge with another PCIe-PCI bridge
    ok that can be normal
    ... but on the topo you see more bridges in the schemas are the squares.
    one is the extra bridge of the cmi board (and is normal) many others none.
    im try to reproduce the board scheme on qemu and it refuse to do something like that.
    it continue make a "normal topo" shot

    >Interesting. I didn't know that the CoreNet L3 cache can't be used also like a normal L3 cache. But even then this has nothing to do with the board design, don't you think?
    yes it can be used as fast sdram too. but i was referring to the topo scheme.

    >Do you think this is a board design issue?
    can be it or bad kernel integrations, low ram speed bottle neck.
    ram speed so slow... i hope is only bad uboot integration... bur my fear is the e5500 true speed is ddr 200 at true 400MTs data rate.
  • »10.09.17 - 15:34
    Profile
  • Cocoon
    Cocoon
    Posts: 46 from 2017/9/10
    > result they change the document specs of x5000.

    Interesting. Which specs in particular did they change after your reporting?
    made more better the pcie 16x description ;-)


    > g5 quad continue is the best powerpc desktop machine made .

    Together with the TerraSoft/Fixstars/YDL PowerStation, I guess :-)
    I was thinking this machine are more for server use :P


    >Isn't that a contradiction to what you just said yourself? :-)

    >>only r7 is not running but can be only a firmware issue not a pcie compatibility issue.
    where?
  • »10.09.17 - 15:39
    Profile
  • Yokemate of Keyboards
    Yokemate of Keyboards
    Andreas_Wolf
    Posts: 9293 from 2003/5/22
    From: Germany
    >> There's only one SoC on the Cyrus/X5000.

    > ... bad because it made the topo more strange than before

    Why? I don't see where the lstopo output shows more than one SoC.

    > im try to reproduce the board scheme on qemu and it refuse to do something like that.

    So you say that the Cyrus board topology is wrong because QEMU can't recreate it? Wouldn't it rather mean that there's something wrong with QEMU if it can't recreate the topology of an existing board?

    >>> The L3 cache is a fake cache it is used only for net packet and not for data...

    >> I didn't know that the CoreNet L3 cache can't be used also like a normal L3 cache.

    > yes it can be used as fast sdram too.

    Just like it's supposed to be, right? Everything fine then.

    >> this has nothing to do with the board design, don't you think?

    > i was referring to the topo scheme.

    What's wrong with the L3 cache in the lstopo output?

    > my fear is the e5500 true speed is ddr 200 at true 400MTs data rate.

    The memory controller is not part of the e5500 core. It's part of the SoC's 'uncore'. Also, 400 MHz and 800 MT/s is minimum spec for DDR3.

    https://en.wikipedia.org/wiki/DDR3_SDRAM#JEDEC_standard_modules
  • »10.09.17 - 21:35
    Profile
  • Yokemate of Keyboards
    Yokemate of Keyboards
    Andreas_Wolf
    Posts: 9293 from 2003/5/22
    From: Germany
    >>> from very beginning i report to aeon the bed video gpu performances ...
    >>> result they change the document specs of x5000.

    >> Which specs in particular did they change after your reporting?

    > made more better the pcie 16x description ;-)

    You mean when A-Eon wrote "PCIe x16 slot (Gen2 x4)" in their November 2016 press release? Or do you mean they omitted the fact in the original Cyrus Plus technical reference manual (version 1.1.1 from July 2014 makes no secret of the fact)?

    >> Together with the TerraSoft/Fixstars/YDL PowerStation, I guess :-)

    > I was thinking this machine are more for server use :P

    It was advertised for both server and workstation use.

    >>> on my quad [...] only r7 is not running

    >> Isn't that a contradiction to what you just said yourself? :-)

    > where?

    There: "the same gpus work on G5 and on x5000" :-)
  • »10.09.17 - 21:53
    Profile
  • Cocoon
    Cocoon
    Posts: 46 from 2017/9/10
    Quote:

    slot (Gen2 x4)" in their November 2016 press

    this ... i need to make a correction my x5000 come 29/9/2016 .. i was sure it come in october but i was wrong... just find my video of unboxing.

    Quote:

    There: "the same gpus work on G5 and on x5000" :-)


    ah ah you right. for have the gpus run on x5000 the gpu firmware have to be build on the kernel. On g5 quad need to be installed on the lib/firmwares on the hd. i did not have the firmware in the lib when i made the test on G5 this is why im not sure.

    Quote:


    It was advertised for both server and workstation use.


    you have better memory then me :-)

    [ Edited by tlosmx 10.09.2017 - 22:02 ]
  • »10.09.17 - 22:01
    Profile
  • Cocoon
    Cocoon
    Posts: 46 from 2017/9/10
    Andreas my english is not good and for me is difficult to write .


    >Why? I don't see where the lstopo output shows more than one SoC.

    the soc if one on the topo have to be one only ... if you see the image you will see two with same id


    >So you say that the Cyrus board topology is wrong because QEMU can't recreate it? Wouldn't it rather mean that there's >something wrong with QEMU if it can't recreate the topology of an existing board?

    Nope

    >> this has nothing to do with the board design, don't you think?
    yes the cpu scheme is all ok, are the pci bridges and soc strange

    >What's wrong with the L3 cache in the lstopo output?
    Nothing strange there about L3 cache is ok in the topo . but it is used only for corenet and not as sdram
    for be sdram have to be set up in source config of uboot ... but no one setup it. and build.
    i hope now is better :P



    > my fear is the e5500 true speed is ddr 200 at true 400MTs data rate.
    by the way 1 GB/s to 4 gb/s max for 5040 are not DD3 1600 speed .. (stream and linux kernel report)
    1 GB/s to 2 gbs max for 5020

    G5 Quad is 7/8 Gb/s near 10Gb/s in Altivec

    Os4 ragemem Performances of 5020
    image.jpg
    vs x1000
    image.jpg

    Quote:

    So you say that the Cyrus board topology is wrong because QEMU can't recreate it? Wouldn't it rather mean that there's something wrong with QEMU if it can't recreate the topology of an existing board?

    nope ... the topology of the cyrrus is totally strange compared all topology in the world Informaic Tecnology.
    qemu dont manage it because strange.



    check and compare topos:


    SAM 460

    https://www.dropbox.com/s/pps1q9ropeusfh0/16196243_10208175166078464_2023629800_o.jpg?dl=0

    X1000

    https://www.dropbox.com/s/synwq2x1k081jn3/16117271_10208143633210162_1285018612_n.jpg?dl=0

    PowerMac G5 Quad

    https://www.dropbox.com/s/b9ajetsf8r1kjh0/16118530_10208141403954432_751303834_n.png?dl=0

    Tabor
    https://www.dropbox.com/s/joziw57zfoq0h3u/16237927_10208180847220489_965077643_n.jpg?dl=0

    a PC
    https://www.dropbox.com/s/fu3bwiooqg4q9as/16111653_10208132217124767_495101249_n.png?dl=0

    X5000 /20...
    https://www.dropbox.com/s/wdwuib52bii2bcn/16176754_10208155966798494_1032589834_n.png?dl=0

    X5000/40
    https://www.dropbox.com/s/8rgcvr6532fmzt5/16128954_10208172248685531_756994869_n.png?dl=0




    [ Edited by tlosmx 10.09.2017 - 23:12 ]
  • »10.09.17 - 22:30
    Profile
  • Jim
  • Yokemate of Keyboards
    Yokemate of Keyboards
    Jim
    Posts: 4251 from 2009/1/28
    From: Delaware, USA
    Hmm, those memory benches are very disappointing.
    "Magnetic was troubled by my avatars and 'satanic' references" - Jim Igou

    "Never attribute to malice what can more readily explained by incompetence"
  • »10.09.17 - 22:58
    Profile
  • Cocoon
    Cocoon
    Posts: 46 from 2017/9/10
    Quote:

    Jim wrote:
    Hmm, those memory benches are very disappointing.


    exactly for the 5020 is the same speed of P4 born in 2001 ddr 266mhz
  • »10.09.17 - 23:10
    Profile
  • Cocoon
    Cocoon
    Posts: 46 from 2017/9/10
    if you need stream result here are for compare
    https://www.cs.virginia.edu/stream/by_name/Bandwidth.html
  • »10.09.17 - 23:14
    Profile