beginners question about building a desktop - Hardware

This is a discussion on beginners question about building a desktop - Hardware ; I am thinking of assembling a computer for debian linux. I was going to use it for music composing, and coding, and various things. I was hoping to load a 64bit system on it. I have not assembled a computer ...

+ Reply to Thread
Results 1 to 16 of 16

Thread: beginners question about building a desktop

  1. beginners question about building a desktop

    I am thinking of assembling a computer for debian linux. I was going
    to use it
    for music composing, and coding, and various things. I was hoping to
    load a
    64bit system on it. I have not assembled a computer before, though I
    have
    replaced drives, etc. I have installed linux (debian, RedHat, ubuntu)
    many
    times before. I was thinking of buying the following.:

    Shuttle SG31G2

    mushkin 4GB (2 x 2GB) 240-Pin DDR2 SDRAM DDR2 1066 (PC2 8500)

    Seagate Barracuda 7200.11 ST3320613AS 320GB 7200 RPM 16MB Cache SATA
    3.0Gb/s
    Hard Drive $70

    Intel Q6600 Processor

    LG Black 20X DVD+R 8X DVD+RW 10X DVD+R DL 20X DVD-R 6X DVD-RW 12X DVD-
    RAM 16X
    DVD-ROM 48X CD-R 32X CD-RW 48X CD-ROM 2MB Cache ATAPI / E-IDE Super
    Multi DVD
    Burner

    -----------------------------------------------------------------------------

    After I assemble this I hoping to hook up a floppy disk to it, and
    then start the debian
    install from the floppy disks (I have the debian floppy disks
    required).

    I was hoping to get feedback on what people felt about the above,
    especially
    people's criticisms as to my (beginners) choice of hardware. Also:

    Will this model of Shuttle work, will debian recognize all the
    hardware?

    Is the cooling sufficient?

    Will I need any BIOS updates before the thing will work?

    Is the choice of IDE vs SATA for the DVD burner a wise one?

    Can it boot off of an IDE drive?

    Any sort of gotchas would help.

    Thanks.

  2. Re: beginners question about building a desktop

    Rico wrote:

    > I am thinking of assembling a computer for debian linux. I was going
    > to use it for music composing, and coding, and various things. I was
    > hoping to load a 64bit system on it. I have not assembled a computer
    > before, though I have replaced drives, etc. I have installed linux
    > (debian, RedHat, ubuntu) many times before. I was thinking of buying the
    > following.:
    >
    > Shuttle SG31G2


    I can't make up the exact details of the chipsets used in that particular
    barebone, but why don't you opt for one of their GNU/Linux-certified
    barebones instead?

    The */SD30G2/* is fully supported by Shuttle for use with GNU/Linux and can
    even be bought pre-assembled with SuSE Enterprise Linux pre-installed.

    > mushkin 4GB (2 x 2GB) 240-Pin DDR2 SDRAM DDR2 1066 (PC2 8500)


    That should be good enough for a 64-bit distribution.

    > Seagate Barracuda 7200.11 ST3320613AS 320GB 7200 RPM 16MB Cache SATA
    > 3.0Gb/s Hard Drive $70


    Seagate is good stuff.

    > Intel Q6600 Processor
    >
    > LG Black 20X DVD+R 8X DVD+RW 10X DVD+R DL 20X DVD-R 6X DVD-RW 12X DVD-
    > RAM 16X
    > DVD-ROM 48X CD-R 32X CD-RW 48X CD-ROM 2MB Cache ATAPI / E-IDE Super
    > Multi DVD
    > Burner


    I would personally recommend an SATA unit over an E-IDE unit. In addition
    to SATA having a faster bus, you could then also disable the IDE controller
    in the BIOS and free up some IRQs. ;-)

    > After I assemble this I hoping to hook up a floppy disk to it, and
    > then start the debian install from the floppy disks (I have the debian
    > floppy disks required).


    You /could/ do that, but why not use a bootable CD/DVD? Just download
    the /.iso/ file, check the /md5sums/ or /sha1sums/ - so that you can rule
    out download corruption - burn it to a quality brand blank CD/DVD at low
    speed - so that you can rule out burning errors - and then boot up from it.

    > I was hoping to get feedback on what people felt about the above,
    > especially people's criticisms as to my (beginners) choice of hardware.
    > Also:
    >
    > Will this model of Shuttle work, will debian recognize all the
    > hardware?


    As I wrote higher up, this is very difficult to ascertain without having
    tested this myself. The type of barebone you are listing is not advertised
    on Shuttle's website as validated for GNU/Linux, while they _do_ offer
    barebones that are, as I've mentioned higher up as well.

    Yet, there is a great chance that most if not all of the hardware will work.
    Things that /could/ be problematic - again, I don't have all the data on
    this configuration - would in that case be the wireless networking chipset
    - if it's an Intel, Orinoco or Atheros chipset, it'll most certainly work.

    I also don't know what kind of video chipset is used in these barebones. As
    such, there might be a chance that you need to separately download a
    proprietary videodriver in order to get hardware 3D support, but 2D will
    never be a problem.

    If all else fails and if you can live without the 3D acceleration, you can
    get any video adapter to work in VESA mode.

    > Is the cooling sufficient?


    That's pretty hard to say, but I reckon it will be, or they wouldn't be
    selling it. :-)

    As an extra precaution, you can always scout for a Zalman, CoolerMaster or
    Thermaltake heatsink and extra fans, but of course, the more fans you have,
    the higher the noise level will be. I don't like noisy fans either, but if
    I have to choose between adequate cooling and lower noise, I always go for
    adequate cooling.

    > Will I need any BIOS updates before the thing will work?


    I don't think so. Most motherboards come with a recent-enough BIOS version.

    > Is the choice of IDE vs SATA for the DVD burner a wise one?


    In my humble opinion, I would go for SATA, for the reasons I've stated
    higher up.

    > Can it boot off of an IDE drive?


    Yes, of course. They can all do that. :-)

    > Any sort of gotchas would help.


    The only real /gotcha/ I see is that there is no way of knowing in advance
    as to whether everything on the motherboard is fully supported for use with
    the Linux kernel (while a certified Linux-compatible model does exist). It
    would have been easier if you could have told us what chipsets the
    motherboard uses. :-/

    Another possible /gotcha/ is that you have to be careful not to zap the
    motherboard, so make sure that you've earthed yourself while installing the
    stuff inside that little chassis.

    When applying heat-conductive contact paste on the processor before mounting
    the heatsink, don't use too much of it and spread it out evenly with your
    finger. Then make sure that you mount the heatsink tightly to the
    processor, but without using too much force - you will bend the motherboard
    and possibly rupture some of the solder or the thin electric pathways if
    you do.

    Lastly, a 64-bit distribution usually requires installing some libraries in
    a 32-bit version (alongside their 64-bit siblings) in order to get some
    plugins or codecs to work properly, mainly from proprietary software
    developers who refuse to either open up the source code or develop 64-bit
    versions of their software.

    Normally however, your distribution should take care of this for you, but it
    may be handy to remember that when upgrading certain parts of your
    installation with newer packages.

    Hope this was helpful... ;-)

    --
    *Aragorn*
    (registered GNU/Linux user #223157)

  3. Re: beginners question about building a desktop


    The chipset for the SG31G2 is Intel G31 + ICH7. Zareason builds a
    Ubuntu linux computer
    based on the SG31G2, though I don't know what they've added, possibly
    nothing.

  4. Re: beginners question about building a desktop

    Rico wrote:

    > The chipset for the SG31G2 is Intel G31 + ICH7. Zareason builds a
    > Ubuntu linux computer based on the SG31G2, though I don't know what
    > they've added, possibly nothing.


    Then I don't see any objections as to why you couldn't use this barebone for
    a GNU/Linux installation, given that you keep my other recommendations into
    account. ;-)

    P.S: Might I give you some advice with regard to Usenet? :-)

    (1) Try using a real newsreader, connected to a real newsserver - there are
    plenty of free newsservers if your ISP does not provide for one. Many
    people are filtering out Google Groups posts because of the fact that
    Google Groups posters tend to be quite clueless and think that Usenet *is*
    Google Groups, or some webbased "forum", and due to the fact that many
    trolls post through Google Groups in order to mask their true whereabouts
    and identity.


    (2) Always include some content of the post you are replying to, so that
    other posters immediately know what you're talking about without having to
    read the previous posts in the thread.


    (3) When following my advice in (2), do not top-post. Top-posting is when
    you write your reply at the top of the message and leave the post you are
    replying to dangling at the bottom of your message. That way, people who
    are new to the thread or whose newsservers have dropped the original post
    for some technical reason - e.g. a temporary network glitch - won't have to
    scroll down to below your reply, then read from there to the bottom, and
    then scroll back up all the way to the top and read from there again to
    know what the debate is about and what you are replying to.

    The proper way to reply is to use an interleaved style, i.e. you reply to
    the individual paragraphs or even individual sentences by writing your
    reply underneath the quoted paragraph/sentence (and before the next quoted
    paragraph/sentence). And while doing this, you should trim out the
    irrelevant (or no longer relevant) sections of the post you are replying
    to; there's no need to quote everything, only just enough to allow anyone
    to get in on the conversation and know what is being discussed. :-)


    (4) If you feel that you might have better chances at posting to multiple
    newsgroups, then use crossposting instead of multiposting. Crossposting is
    when you write one post and put multiple groups in the "Group" field.
    Multiposting is when you write an article and send it to a newsgroup, and
    then send an identical copy of it to another newsgroup.

    The advantage of crossposting is that there's less overhead, and that people
    subscribed to both (or all) groups you've posted your article to will only
    have to read it once, because a newsreader recognizes crossposted articles
    - Usenet posts all have a unique message ID, and if an article has been
    read already, the newsreader will not download it a second time if the
    header shows up in another newsgroup.

    It is however advisable to keep the total amount of groups you crosspost to
    limited to around four max. Many people filter out posts that are being
    sent to more than four groups because of the trolls and spammers, who do
    that sort of thing.


    There are plenty of other such guidelines, and there are also plenty of
    people who just do as they please or who would even start debating over the
    right policy for posting to Usenet, but the above guidelines are the ones
    accepted to be the most efficient ones throughout most of the Usenet
    community. ;-)

    If you're interested, you can find more information here...:

    http://www.faqs.org/docs/Linux-HOWTO...ews-HOWTO.html

    I personally use /KNode/ as a newsreader. It comes with the KDE desktop
    suite and can be used either as a standalone application or embedded
    into /Kontact./ Other popular GUI newsreaders are /Pan/ and /Thunderbird./

    One advantage of the above newsreaders for instance is that the words that I
    spell delimited with forward slashes will show up in italic font, the words
    delimited by underscores will be underlined, and words delimited by
    asterisks will be in a bold font. Google Groups does not interpret those
    formatting characters. :-)

    Among the non-GUI (character mode) newsreaders, /slrn/ is a very powerful
    and very popular one, and it allows you to pick an editor of your choice to
    compose your messages with, from /mcedit/ or /nano/ over to /emacs/
    or /vi(m)./ ;-)

    Hope this was helpful... :-)

    --
    *Aragorn*
    (registered GNU/Linux user #223157)

  5. Image processing

    Stefan Patric wrote:

    > If you're doing serious music/sound work, I'd want to have the option for
    > more RAM, at least 8GB. I'm a commercial photographer, and image
    > processing like sound processing uses lots of RAM. Right now I'm at 4GB
    > on my 64-bit AMD system, that I built in Jan 2007, and I consider that a
    > workable minimum. Initially, I started with 2GB, thinking that would be
    > more than enough for "light" editing, but it quickly became apparent I
    > needed more. My motherboard max's out at 8, and I'm looking to upgrade
    > to that later this year.


    May I ask, out of curiosity, what you're using for photo editing? Gimp
    is of course well-known, but I personally would need (for several
    reasons, longer story) something that can handle pixel precisions beyond
    8 bit - gimp is a bit limited.

    Greetings,
    Thomas

  6. Re: Image processing

    I was also wondering about memory matters. It says on Shuttle's site
    for the SG31G2 that the memory it expects is
    Dual Channel DDR2 667/800, which I take it means that the most I can
    expect for it is to put in DDR2 800 chips. Aside
    from this specification, do the timings, cas latency matter or not?

  7. Re: Image processing

    Does the voltage advertised on these modules make a difference? I
    read in wikipedia that 1.8V is the standard, and they must
    be able to withstand 2.3v without suffering damage but they may not
    work properly at this voltage. Some specifications say
    that their module works at 1.8v, while other (different) brands say
    that their module works at 2.1v. Will the motherboard adapt
    to this or what, or will some memory modules fail to work with certain
    boards?

  8. Re: Image processing

    On Tue, 10 Jun 2008 00:36:47 -0700, Rico wrote:

    > I was also wondering about memory matters. It says on Shuttle's site
    > for the SG31G2 that the memory it expects is Dual Channel DDR2 667/800,
    > which I take it means that the most I can expect for it is to put in
    > DDR2 800 chips. Aside from this specification, do the timings, cas
    > latency matter or not?


    It has been my experience that as far as image processing (and sound and
    video) that RAM chip timings, latencies, etc. make little difference in
    overall processing speed. The amount of RAM is significantly more
    important. So, a machine with "slower" chips and lots of RAM will
    process your images faster than a machine with "faster" chips but minimal
    RAM.

    FWIW: I would want a system with a minimum of 4 GB of RAM. I'm a
    commercial photographer, and sometimes work with 100MB images, but
    average is 20 to 30, and 4GB RAM is what I have and I'm satisfied with
    the speed of my 2.0 GHz Athlon 64, Fedora 9 (64-bit) system. Max RAM
    supported is 8GB.

    Stef

  9. Re: Image processing

    On Fri, 30 May 2008 21:32:51 -0700, Rico wrote:
    > I am thinking of assembling a computer for debian linux. I was going to
    > use it
    > for music composing, and coding, and various things. I was hoping to
    > load a
    > 64bit system on it. I have not assembled a computer before, though I
    > have
    > replaced drives, etc. I have installed linux (debian, RedHat, ubuntu)
    > many
    > times before. I was thinking of buying the following.:


    > Shuttle SG31G2


    ....

    > Intel Q6600 Processor


    Now I have been assured that this will work, due to Zareason marketing
    a similar
    combination. But now after doing more research, I think an Intel 8400
    would be more appropriate, which they don't market in combination with
    the SG31G2. The question is If I
    can get the motherboard to recognize it, which Shuttle says it can is
    there any problem
    with getting linux to recognize it? Does the linux system depend on
    the processor
    directly or the motherboard having the right kind of interaction with
    the CPU?

  10. Re: Image processing

    On Monday 16 June 2008 18:21, someone who identifies as *Rico* wrote
    in /comp.os.linux.hardware:/

    > On Fri, 30 May 2008 21:32:51 -0700, Rico wrote:
    >
    >> [...]
    >> Shuttle SG31G2

    >
    > ...
    >
    >> Intel Q6600 Processor

    >
    > Now I have been assured that this will work, due to Zareason marketing
    > a similar combination. But now after doing more research, I think an
    > Intel 8400 would be more appropriate, which they don't market in
    > combination with the SG31G2. The question is If I can get the motherboard
    > to recognize it, which Shuttle says it can is there any problem
    > with getting linux to recognize it? Does the linux system depend on
    > the processor directly or the motherboard having the right kind of
    > interaction with the CPU?


    All hardware is important, but I haven't heard of a processor before -
    especially in the /x86/ range - that doesn't support Linux. In fact, one
    of the strengths of GNU/Linux is that it runs on practically every
    platform.

    But either way, what makes you think it wouldn't support that processor?
    It's an /x86-64,/ isn't it? Of course, don't think that your distribution
    will support _all_ of your processor's functions. Distributions build
    their kernel with a lot of generic code so that it will run on a whole
    variety of processors in that range, from Intel and AMD alike. If you do
    want your kernel to be optimized for your particular processor, then you'll
    have to compile your own kernel, and for good measure, you should then even
    rebuild /glibc,/ which is the most used library in the system.

    Compatibility issues between the Linux kernel and certain motherboards
    usually have to do with the chipsets on the motherboard for sound,
    PATA/SATA, (wireless) networking and graphics. Most motherboard
    manufacturers do however use PATA/SATA chipsets that are supported, as well
    as commonly supported ethernet chipsets, a sound chip that's typically
    supported via /hda-intel/ or /ac97/ and a graphics adapter that at the
    least provides you 2D via an Open Source driver.

    Wireless ethernet is another matter, because most of those chipsets do not
    have Open Source drivers. Only a few do, and some of those that don't can
    be made to work using the Windows-specific driver, loaded into the Linux
    kernel via /ndiswrapper./

    --
    *Aragorn*
    (registered GNU/Linux user #223157)

  11. Re: Image processing

    > > ... But now after doing more research, I think an
    > > Intel 8400 would be more appropriate, which they don't market in
    > > combination with theSG31G2. The question is If I can get the motherboard
    > > to recognize it, which Shuttle says it can is there any problem
    > > with getting linux to recognize it? Does the linux system depend on
    > > the processor directly or the motherboard having the right kind of
    > > interaction with the CPU?

    >
    > All hardware is important, but I haven't heard of a processor before -
    > especially in the /x86/ range - that doesn't support Linux. In fact, one
    > of the strengths of GNU/Linux is that it runs on practically every
    > platform.
    >


    I have built my computer, thus far with no hard disk (I have it but
    don't have it installed),
    and have been running a Ubuntu 8.04 64-bit live disk on it. However I
    have some doubts
    as to whether the motherboard is recognizing the e8400 correctly.
    That is look at the l2
    cache:

    ubuntu@ubuntu:~$ sudo lshw
    ubuntu
    description: Desktop Computer
    product: SG31
    vendor: Shuttle Inc
    version: V10
    serial: 0
    width: 32 bits
    capabilities: smbios-2.5 dmi-2.5 smp-1.4 smp
    configuration: boot=normal chassis=desktop cpus=0
    uuid=01020907-0301-0103-08 07-060504030201
    *-core
    description: Motherboard
    product: FG31
    vendor: Shuttle Inc
    physical id: 0
    version: V10
    serial: 0
    *-firmware
    description: BIOS
    vendor: Phoenix Technologies, LTD
    physical id: 0
    version: 6.00 PG (01/18/2008)
    size: 128KiB
    capacity: 960KiB
    capabilities: isa pci pnp apm upgrade shadowing cdboot
    bootselect sock etedrom edd int13floppy360 int13floppy1200
    int13floppy720 int13floppy2880 int5pr intscreen int9keyboard
    int14serial int17printer int10video acpi usb ls120boot zi pboot
    biosbootspecification
    *-cpu
    description: CPU
    product: Intel(R) Core(TM)2 Duo CPU E8400 @ 3.00GHz
    vendor: Intel Corp.
    physical id: 5
    bus info: cpu@0
    version: Intel(R)
    slot: Socket 775
    size: 2997MHz
    capacity: 4GHz
    width: 64 bits
    clock: 333MHz
    capabilities: fpu fpu_exception wp vme de pse tsc msr pae
    mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx
    fxsr sse sse2 ss ht tm pbe syscall nx x86-64 constant_tsc
    arch_perfmon pebs bts rep_good pni monitor ds_cp l vmx smx est tm2
    ssse3 cx16 xtpr sse4_1 lahf_lm
    *-cache:0
    description: L1 cache
    physical id: b
    slot: Internal Cache
    size: 32KiB
    capacity: 32KiB
    capabilities: synchronous internal write-back
    *-cache:1 DISABLED
    description: L2 cache
    physical id: c
    slot: External Cache
    capabilities: synchronous external write-back
    *-memory
    description: System Memory
    physical id: 18
    slot: System board or motherboard
    size: 4GiB
    *-bank:0
    description: DIMM Synchronous 800 MHz (1.2 ns)
    physical id: 0
    slot: A0
    size: 2GiB
    clock: 800MHz (1.2ns)
    *-bank:1
    description: DIMM [empty]
    physical id: 1
    slot: A1
    *-bank:2
    description: DIMM Synchronous 800 MHz (1.2 ns)
    physical id: 2
    slot: A2
    size: 2GiB
    clock: 800MHz (1.2ns)
    *-bank:3
    description: DIMM [empty]
    physical id: 3
    slot: A3
    *-pci
    description: Host bridge
    product: 82G33/G31/P35/P31 Express DRAM Controller
    vendor: Intel Corporation
    physical id: 100
    bus info: pci@0000:00:00.0
    version: 02
    width: 32 bits
    clock: 33MHz
    configuration: driver=agpgart-intel module=intel_agp
    *-display:0 UNCLAIMED
    description: VGA compatible controller
    product: 82G33/G31 Express Integrated Graphics Controller
    vendor: Intel Corporation
    physical id: 2
    bus info: pci@0000:00:02.0
    version: 02
    width: 32 bits
    clock: 33MHz
    capabilities: msi pm vga_controller bus_master cap_list
    configuration: latency=0
    *-display:1 UNCLAIMED
    description: Display controller
    product: 82G33/G31 Express Integrated Graphics Controller
    vendor: Intel Corporation
    physical id: 2.1
    bus info: pci@0000:00:02.1
    version: 02
    width: 32 bits
    clock: 33MHz
    capabilities: pm bus_master cap_list
    configuration: latency=0
    *-multimedia
    description: Audio device
    product: 82801G (ICH7 Family) High Definition Audio
    Controller
    vendor: Intel Corporation
    physical id: 1b
    bus info: pci@0000:00:1b.0
    version: 01
    width: 64 bits
    clock: 33MHz
    capabilities: pm msi pciexpress bus_master cap_list
    configuration: driver=HDA Intel latency=0
    module=snd_hda_intel
    *-pci:0
    description: PCI bridge
    product: 82801G (ICH7 Family) PCI Express Port 1
    vendor: Intel Corporation
    physical id: 1c
    bus info: pci@0000:00:1c.0
    version: 01
    width: 32 bits
    clock: 33MHz
    capabilities: pci pciexpress msi pm normal_decode
    bus_master cap_li st
    configuration: driver=pcieport-driver
    *-pci:1
    description: PCI bridge
    product: 82801G (ICH7 Family) PCI Express Port 2
    vendor: Intel Corporation
    physical id: 1c.1
    bus info: pci@0000:00:1c.1
    version: 01
    width: 32 bits
    clock: 33MHz
    capabilities: pci pciexpress msi pm normal_decode
    bus_master cap_li st
    configuration: driver=pcieport-driver
    *-network
    description: Ethernet interface
    product: 88E8056 PCI-E Gigabit Ethernet Controller
    vendor: Marvell Technology Group Ltd.
    physical id: 0
    bus info: pci@0000:02:00.0
    logical name: eth0
    version: 12
    serial: 00:30:1b:45:db:99
    size: 100MB/s
    capacity: 1GB/s
    width: 64 bits
    clock: 33MHz
    capabilities: pm vpd msi pciexpress bus_master
    cap_list ethernet physical tp 10bt 10bt-fd 100bt 100bt-fd 1000bt
    1000bt-fd autonegotiation
    configuration: autonegotiation=on broadcast=yes
    driver=sky2 driv erversion=1.20 duplex=full firmware=N/A
    ip=192.168.1.101 latency=0 link=yes modu le=sky2 multicast=yes
    port=twisted pair speed=100MB/s
    *-usb:0
    description: USB Controller
    product: 82801G (ICH7 Family) USB UHCI Controller #1
    vendor: Intel Corporation
    physical id: 1d
    bus info: pci@0000:00:1d.0
    version: 01
    width: 32 bits
    clock: 33MHz
    capabilities: uhci bus_master
    configuration: driver=uhci_hcd latency=0 module=uhci_hcd
    *-usb:1
    description: USB Controller
    product: 82801G (ICH7 Family) USB UHCI Controller #2
    vendor: Intel Corporation
    physical id: 1d.1
    bus info: pci@0000:00:1d.1
    version: 01
    width: 32 bits
    clock: 33MHz
    capabilities: uhci bus_master
    configuration: driver=uhci_hcd latency=0 module=uhci_hcd
    *-usb:2
    description: USB Controller
    product: 82801G (ICH7 Family) USB UHCI Controller #3
    vendor: Intel Corporation
    physical id: 1d.2
    bus info: pci@0000:00:1d.2
    version: 01
    width: 32 bits
    clock: 33MHz
    capabilities: uhci bus_master
    configuration: driver=uhci_hcd latency=0 module=uhci_hcd
    *-usb:3
    description: USB Controller
    product: 82801G (ICH7 Family) USB UHCI Controller #4
    vendor: Intel Corporation
    physical id: 1d.3
    bus info: pci@0000:00:1d.3
    version: 01
    width: 32 bits
    clock: 33MHz
    capabilities: uhci bus_master
    configuration: driver=uhci_hcd latency=0 module=uhci_hcd
    *-usb:4
    description: USB Controller
    product: 82801G (ICH7 Family) USB2 EHCI Controller
    vendor: Intel Corporation
    physical id: 1d.7
    bus info: pci@0000:00:1d.7
    version: 01
    width: 32 bits
    clock: 33MHz
    capabilities: pm debug ehci bus_master cap_list
    configuration: driver=ehci_hcd latency=0 module=ehci_hcd
    *-pci:2
    description: PCI bridge
    product: 82801 PCI Bridge
    vendor: Intel Corporation
    physical id: 1e
    bus info: pci@0000:00:1e.0
    version: e1
    width: 32 bits
    clock: 33MHz
    capabilities: pci subtractive_decode bus_master cap_list
    *-firewire
    description: FireWire (IEEE 1394)
    product: TSB43AB22/A IEEE-1394a-2000 Controller (PHY/
    Link)
    vendor: Texas Instruments
    physical id: a
    bus info: pci@0000:03:0a.0
    version: 00
    width: 32 bits
    clock: 33MHz
    capabilities: pm ohci bus_master cap_list
    configuration: driver=ohci1394 latency=32 maxlatency=4
    mingnt=2 module=ohci1394
    *-isa
    description: ISA bridge
    product: 82801GB/GR (ICH7 Family) LPC Interface Bridge
    vendor: Intel Corporation
    physical id: 1f
    bus info: pci@0000:00:1f.0
    version: 01
    width: 32 bits
    clock: 33MHz
    capabilities: isa bus_master cap_list
    configuration: latency=0
    *-ide:0
    description: IDE interface
    product: 82801G (ICH7 Family) IDE Controller
    vendor: Intel Corporation
    physical id: 1f.1
    bus info: pci@0000:00:1f.1
    version: 01
    width: 32 bits
    clock: 33MHz
    capabilities: ide bus_master
    configuration: driver=ata_piix latency=0 module=ata_piix
    *-ide:1
    description: IDE interface
    product: 82801GB/GR/GH (ICH7 Family) SATA IDE Controller
    vendor: Intel Corporation
    physical id: 1f.2
    bus info: pci@0000:00:1f.2
    logical name: scsi2
    version: 01
    width: 32 bits
    clock: 66MHz
    capabilities: ide pm bus_master cap_list emulated
    configuration: driver=ata_piix latency=0 module=ata_piix
    *-cdrom
    description: DVD-RAM writer
    product: DVDRW LH-20A1L
    vendor: LITE-ON
    physical id: 0.0.0
    bus info: scsi@2:0.0.0
    logical name: /dev/cdrom
    logical name: /dev/dvd
    logical name: /dev/scd0
    logical name: /dev/sr0
    logical name: /cdrom
    version: BL06
    capabilities: removable audio cd-r cd-rw dvd dvd-r dvd-
    ram
    configuration: ansiversion=5 mount.fstype=iso9660
    mount.options= ro,noatime,relatime state=mounted status=ready
    *-medium
    physical id: 0
    logical name: /dev/cdrom
    logical name: /cdrom
    configuration: mount.fstype=iso9660
    mount.options=ro,noatime, relatime state=mounted
    *-serial UNCLAIMED
    description: SMBus
    product: 82801G (ICH7 Family) SMBus Controller
    vendor: Intel Corporation
    physical id: 1f.3
    bus info: pci@0000:00:1f.3
    version: 01
    width: 32 bits
    clock: 33MHz
    configuration: latency=0
    ubuntu@ubuntu:~$


    Is this a bug in the lshw program or what, to not report the l2
    cache? How could I check on
    this?
    u

  12. Re: Image processing

    On Wednesday 25 June 2008 11:31, someone who identifies as *Rico* wrote
    in /comp.os.linux.hardware:/

    > I have built my computer, thus far with no hard disk (I have it but
    > don't have it installed), and have been running a Ubuntu 8.04 64-bit live
    > disk on it. However I have some doubts as to whether the motherboard is
    > recognizing the e8400 correctly. That is look at the l2 cache:
    >
    > [...]
    > *-cache:0
    > description: L1 cache
    > physical id: b
    > slot: Internal Cache
    > size: 32KiB
    > capacity: 32KiB
    > capabilities: synchronous internal write-back
    > *-cache:1 DISABLED
    > description: L2 cache
    > physical id: c
    > slot: External Cache
    > capabilities: synchronous external write-back
    > [...]
    > Is this a bug in the lshw program or what, to not report the l2
    > cache? How could I check on this?


    L2 cache can be enabled or disabled in the BIOS set-up program, so you
    should check there, and consult the motherboard manual. I doubt that there
    is a bug in /lshw/ with regard to this issue.

    --
    *Aragorn*
    (registered GNU/Linux user #223157)

  13. Re: Image processing

    > L2 cache can be enabled or disabled in the BIOS set-up program, so you
    > should check there, and consult the motherboard manual. I doubt that there
    > is a bug in /lshw/ with regard to this issue.
    >
    > --
    > *Aragorn*
    > (registered GNU/Linux user #223157)


    I have searched the bios settings and the manual, there is no explicit
    control for the
    l2 cache (unlike some of my older machines, which have also an award
    bios, but have
    some cpu cache controls). There is some cpu features control, but
    none of the options
    set forth in the manual explicitly have to do with the l2 cache. The
    ones that look
    like they might have in some remote chance something to do with the l2
    cache
    are 'PPM Mode', 'Limit CPUID MAXVal' 'Virtualization Technology' 'Core
    Multi-processing'.

  14. Re: Image processing

    On Wednesday 25 June 2008 23:32, someone who identifies as *Rico* wrote
    in /comp.os.linux.hardware:/

    >> L2 cache can be enabled or disabled in the BIOS set-up program, so you
    >> should check there, and consult the motherboard manual. I doubt that
    >> there is a bug in /lshw/ with regard to this issue.

    >
    > I have searched the bios settings and the manual, there is no explicit
    > control for the l2 cache (unlike some of my older machines, which have
    > also an award bios, but have some cpu cache controls). There is some cpu
    > features control, but none of the options set forth in the manual
    > explicitly have to do with the l2 cache.


    Still, there should be a summary screen telling you how much L2 cache is
    available.

    > The ones that look like they might have in some remote chance something to
    > do with the l2 cache are 'PPM Mode'


    Doesn't ring a bell...

    > , 'Limit CPUID MAXVal'


    Has to do with the CPUID numbers. I'm not entirely sure what it means, but
    it is either a limitation on the numbering of CPUs or a limitation on the
    number of CPUs you can use. In the latter case, this presumably because of
    licensing per CPU of some proprietary operating systems - notably the ones
    from Microsoft.

    > 'Virtualization Technology'


    Has nothing to do with cache. This is the Intel hardware virtualization
    support on your CPU(s) and memory controller. If you enable this, you will
    be able to do hardware virtualization, i.e. the running of unmodified
    operating systems inside virtual machines (as opposed to
    paravirtualization, i.e. the running of modified operating systems inside
    virtual machines).

    > 'Core Multi-processing'.


    Has nothing to do with cache either. This presumably has to do with setting
    up the APIC(s) and local APICs for symmetric multiprocessing. Sorry I
    can't be more specific, but I have little experience with the latest Intel
    offerings... :-/

    As a sidenote however, /lshw/ gets its information from */proc* and possibly
    from the kernel ring buffer - I don't have /lshw/ installed on this machine
    here so I don't know exactly, but the /man/ page should be able to tell you
    more - and then there is also /dmesg/ which you can consult to see the
    contents of the actual kernel ring buffer.

    As root, issue...

    dmesg | less

    .... at a character mode console. Use *PgUp/PgDn* or the arrow keys to page
    through the screen. Press "q" to exit and return to the commandline
    prompt.

    This said, you are doing this check from a live CD, which is not the same
    thing as an installed distribution. For one, it won't have the latest
    kernel, and secondly, to my experience live CDs often have insufficiently
    tested kernels.

    On the other hand, I would be thoroughly surprised if Linux didn't use or
    find your L2 cache, since this is something enabled or disabled at the
    hardware level, not from within the kernel, and I have certainly never
    heard of any cache problems with Linux. So in the end, it /may/ be a
    shortcoming of the /lshw/ version on your live CD, or possibly a flaw in
    the kernel's ring buffer output.

    --
    *Aragorn*
    (registered GNU/Linux user #223157)

  15. Re: Image processing

    On Jun 25, 6:12 pm, Aragorn wrote:
    ....
    > On the other hand, I would be thoroughly surprised if Linux didn't use or
    > find your L2 cache, since this is something enabled or disabled at the
    > hardware level, not from within the kernel, and I have certainly never
    > heard of any cache problems with Linux. So in the end, it /may/ be a
    > shortcoming of the /lshw/ version on your live CD, or possibly a flaw in
    > the kernel's ring buffer output.
    >
    > --
    > *Aragorn*
    > (registered GNU/Linux user #223157)


    I did as you said part of my dmesg output is:

    [ 31.644605] CPU: L1 I cache: 32K, L1 D cache: 32K
    [ 31.644606] CPU: L2 cache: 6144K
    [ 31.644607] CPU 1/1 -> Node 0
    [ 31.644608] CPU: Physical Processor ID: 0
    [ 31.644609] CPU: Processor Core ID: 1
    [ 31.644613] CPU1: Thermal monitoring enabled (TM2)
    [ 31.645229] Intel(R) Core(TM)2 Duo CPU E8400 @ 3.00GHz
    stepping 06
    [ 31.645273] checking TSC synchronization [CPU#0 -> CPU#1]: passed.
    [ 31.665283] Brought up 2 CPUs
    [ 31.665339] CPU0 attaching sched-domain:
    [ 31.665341] domain 0: span 03
    [ 31.665341] groups: 01 02
    [ 31.665343] domain 1: span 03
    [ 31.665344] groups: 03
    [ 31.665345] CPU1 attaching sched-domain:
    [ 31.665346] domain 0: span 03
    [ 31.665347] groups: 02 01
    [ 31.665348] domain 1: span 03
    [ 31.665349] groups: 03

    Now does this mean that the kernel uses the l2 cache, or that this is
    part of the message
    that the e8400 spews? I see no message in the entire dmesg output
    that the l2 output
    has been disabled or anything out of what I would expect...

  16. Re: Image processing

    On Thursday 26 June 2008 05:02, someone who identifies as *Rico* wrote
    in /comp.os.linux.hardware:/

    > On Jun 25, 6:12 pm, Aragorn wrote:
    >
    >> On the other hand, I would be thoroughly surprised if Linux didn't use or
    >> find your L2 cache, since this is something enabled or disabled at the
    >> hardware level, not from within the kernel, and I have certainly never
    >> heard of any cache problems with Linux. So in the end, it /may/ be a
    >> shortcoming of the /lshw/ version on your live CD, or possibly a flaw in
    >> the kernel's ring buffer output.

    >
    > I did as you said part of my dmesg output is:
    >
    > [ 31.644605] CPU: L1 I cache: 32K, L1 D cache: 32K
    > [ 31.644606] CPU: L2 cache: 6144K
    > [ 31.644607] CPU 1/1 -> Node 0




    > Now does this mean that the kernel uses the l2 cache, or that this is
    > part of the message that the e8400 spews? I see no message in the entire
    > dmesg output that the l2 output has been disabled or anything out of what
    > I would expect...


    Strictly speaking, it is a message from the kernel that it has found a
    usable L2 cache the size of 6144 KB. And what it has found, it uses. ;-)

    So in other words, the /lshw/ report was flawed and it may very well be a
    bug in /lshw/ then. ;-)

    --
    *Aragorn*
    (registered GNU/Linux user #223157)

+ Reply to Thread