Statistics for alt.os.linux.mandrake - Mandrake

This is a discussion on Statistics for alt.os.linux.mandrake - Mandrake ; Christopher Hunter wrote: > I've built prototype preamplifiers with all-discrete components and with > op-amps, and even the "golden-eared" brigade can't hear the difference! > Modern op-amps (and I /don't/ mean expensive, esoteric ones) > are /perfectly/ "good enough". If ...

+ Reply to Thread
Page 20 of 20 FirstFirst ... 10 18 19 20
Results 381 to 400 of 400

Thread: Statistics for alt.os.linux.mandrake

  1. Re: House Wiring

    Christopher Hunter wrote:
    > I've built prototype preamplifiers with all-discrete components and with
    > op-amps, and even the "golden-eared" brigade can't hear the difference!
    > Modern op-amps (and I /don't/ mean expensive, esoteric ones)
    > are /perfectly/ "good enough". If you want pointers, try the venerable
    > NE5532, NE5534 or LM833 - they /have/ /not/ been bettered!


    Somehow I never think of op-amps for audio applications, because there
    are already so many chips specifically labelled "for audio use."

    My audio system is "mass-fi," which is good enough for me. About the
    only possibly interesting thing I have is a DNR (Dynamic Noise
    Reduction) processor, which a friend built for me using the circuit and
    PC board layout in the application notes for National Semiconductor's
    LM1894. I don't even know if they still make that chip. There are a
    few times I think it's an actual improvement, such as 78s with a lot of
    surface noise.

    > You can largely ignore the "digital tone controls" and the like - they offer
    > nothing in terms of sonic performance (they probably introduce noise and
    > distortion). Stick with the tried and tested simple analogue designs.


    In my '99 car, the sound system has knobs for bass, treble, balance and
    fade, but it's not a simple analog design. There's a delay of a second
    or so between the turning of the knob and the change in the sound. That
    really surprised me the first time it happened.

    > It is perfectly possible to build a preamplifier for ~40 that
    > will /outperform/ all the exotic Krell and similar rubbish! All it takes
    > is simple, solid design principles and (sometimes) selection of matched
    > resistor values to guarantee equal channel gain.


    But how many audiophiles would believe that? Many seem to consider
    costliness and scarcity as indicators of quality. I recently read where
    "The Amazing Randi," a magician known for debunking the paranormal, has
    extended his challenge to include some $7000 speaker cables, claiming
    they're indistinguishable from cables costing "only" a few hundred.

    Adam

  2. Re: [OT] Power, etc.

    Moe Trin wrote:
    > The rather
    > interesting problem is that many UPS are not a sine-wave output, but
    > are (at best) a stepped square wave approximation, or a trapezoidal
    > wave,


    I've never heard of a trapezoidal wave. Is it the same thing as a
    sawtooth wave?

    > and the average switching power supply is going to behave a
    > bit differently.


    I've noticed that the more expensive UPSs specify sine wave, and the
    less expensive ones say stepped square wave or something like that. A
    home system would only be powered by the [non-sine-] wave for the few
    minutes necessary to do a graceful shutdown when the AC goes out. Would
    that really have much effect on performance?

    >> Would any of the other solutions be any better? Or just different?

    >
    > Better in the eye of the beholder. I don't like to use DHCP (especially
    > with static addresses) as it's one more thing to go wrong.

    [snip]

    I couldn't quite follow everything you said (I know almost nothing about
    networking), but I *think* you meant that you would have hooked up my
    printer differently. Unless some other setup would make the printer
    faster or something, it works, so I'll leave it. I did create "another"
    printer at the same address but set to default to "draft grayscale,"
    because with that setting it prints at about the speed I'd expect from a
    laser printer. Somehow just "draft" had the vertical spacing about half
    of normal.

    >> I don't think I did anything that would cause the MBR to get rewritten.

    >
    > That is simply not supposed to happen. I'd be investigating this a LOT
    > further. I don't like unknowns or surprises.


    I can't remember exactly what I did that resulted in that happening (I
    don't /think/ I'd done anything unusual), so I don't really have
    anything to go on. I'm not really sure what to do about it, except to
    keep a closer watch on anything I do that might cause the MBR to get
    rewritten.

    Adam

  3. Re: House Wiring

    Adam wrote:

    > But how many audiophiles would believe that? Many seem to consider
    > costliness and scarcity as indicators of quality. I recently read where
    > "The Amazing Randi," a magician known for debunking the paranormal, has
    > extended his challenge to include some $7000 speaker cables, claiming
    > they're indistinguishable from cables costing "only" a few hundred.


    I caused a hi-fi salesman to cry when his 150/m "speaker cables" were no
    different (in a double-blind test) to cheap 5 Amp mains flex! I allowed
    him to choose his favourite hardware, and set up two identical systems, one
    with his expensive "speaker cable" and one with mine. He couldn't identify
    which was which, and neither could any of his staff!

    I saw a preamplifier the other day that had the PCB tracks /gold/ plated!
    It was said to "improve sonic performance", but all it really did was
    increase the price of the hardware - and there wasn't a gold component lead
    in sight!

    Apparently, Randi had the manufacturers of this "speaker cable" reply to his
    challenge. Initially they agreed to testing, but when Randi demanded that
    it was "blind" testing, they backed out...

    C.

  4. Re: House Wiring

    Adam writes:

    >Christopher Hunter wrote:
    >> I've built prototype preamplifiers with all-discrete components and with
    >> op-amps, and even the "golden-eared" brigade can't hear the difference!
    >> Modern op-amps (and I /don't/ mean expensive, esoteric ones)
    >> are /perfectly/ "good enough". If you want pointers, try the venerable
    >> NE5532, NE5534 or LM833 - they /have/ /not/ been bettered!


    >Somehow I never think of op-amps for audio applications, because there
    >are already so many chips specifically labelled "for audio use."


    >My audio system is "mass-fi," which is good enough for me. About the
    >only possibly interesting thing I have is a DNR (Dynamic Noise
    >Reduction) processor, which a friend built for me using the circuit and
    >PC board layout in the application notes for National Semiconductor's
    >LM1894. I don't even know if they still make that chip. There are a
    >few times I think it's an actual improvement, such as 78s with a lot of
    >surface noise.


    One thing I found did make a difference was high slew rate in a phono
    preamp I built.


    >> You can largely ignore the "digital tone controls" and the like - they offer
    >> nothing in terms of sonic performance (they probably introduce noise and
    >> distortion). Stick with the tried and tested simple analogue designs.




    >In my '99 car, the sound system has knobs for bass, treble, balance and
    >fade, but it's not a simple analog design. There's a delay of a second
    >or so between the turning of the knob and the change in the sound. That
    >really surprised me the first time it happened.


    >> It is perfectly possible to build a preamplifier for ~40 that
    >> will /outperform/ all the exotic Krell and similar rubbish! All it takes
    >> is simple, solid design principles and (sometimes) selection of matched
    >> resistor values to guarantee equal channel gain.


    >But how many audiophiles would believe that? Many seem to consider
    >costliness and scarcity as indicators of quality. I recently read where
    >"The Amazing Randi," a magician known for debunking the paranormal, has
    >extended his challenge to include some $7000 speaker cables, claiming
    >they're indistinguishable from cables costing "only" a few hundred.


    Or lamp cord costing 20 cents a foot.



  5. Re: House Wiring

    Unruh wrote:

    > One thing I found did make a difference was high slew rate in a phono
    > preamp I built.


    Op-amp bandwidth isn't an issue with modern parts. My favourite op-amps
    (NE5532 and NE5534) haven't really been bettered for slew rate, distortion
    and noise specifications. There are some more expensive, exotic parts, but
    I haven't been able to either hear or measure any significant difference
    when I've tried them.

    Chris

  6. Re: [OT] Power, etc.

    On Wed, 24 Oct 2007, in the Usenet newsgroup alt.os.linux.mandrake, in article
    , Adam wrote:

    >Moe Trin wrote:


    >> The rather interesting problem is that many UPS are not a sine-wave
    >> output, but are (at best) a stepped square wave approximation, or a
    >> trapezoidal wave,

    >
    >I've never heard of a trapezoidal wave. Is it the same thing as a
    >sawtooth wave?


    Flat top/bottom. Think of a "square' (or at least rectangular) wave
    run through a channel with insufficient bandwidth, such that the
    leading and trailing edge has a substantial width. The triangular
    or sawtooth wave would be one "extreme" case, where the top/bottom
    have a minimal width, while the square/rectangular wave would be at
    the other extreme ("no", or at least minimal rise/fall times).
    _ _ _ _
    /\/\/\/\ / \_/ \ | |_| |_|

    >I've noticed that the more expensive UPSs specify sine wave, and the
    >less expensive ones say stepped square wave or something like that.
    >A home system would only be powered by the [non-sine-] wave for the
    >few minutes necessary to do a graceful shutdown when the AC goes out.
    >Would that really have much effect on performance?


    In theory (and this really is affected by the design of the PC power
    supply), the regulation of the DC outputs is less because the primary
    switching function is done by SCRs. The SCR starts conducting at a RC
    delay verses voltage rise time (lower _load_ voltage causes the SCR
    to turn on earlier on the sine-wave), but the typical SCR circuit
    lacks means to turn off the SCR _other_than_ the input voltage
    becoming less that the output - reverse biasing the SCR.

    >> I don't like to use DHCP (especially with static addresses) as it's
    >> one more thing to go wrong.


    >I couldn't quite follow everything you said (I know almost nothing
    >about networking), but I *think* you meant that you would have hooked
    >up my printer differently.


    I'd have set things up using static addresses directly, and not
    depending on the DHCP server.

    >Unless some other setup would make the printer faster or something,
    >it works, so I'll leave it.


    That's fine.

    >I did create "another" printer at the same address but set to default
    >to "draft grayscale," because with that setting it prints at about the
    >speed I'd expect from a laser printer. Somehow just "draft" had the
    >vertical spacing about half of normal.


    That's saying the printer definition is wrong, but I'm not the printer
    guy so I can't specify what. Recall that the stuff going over the wire
    isn't raw text, but is data from a control program that translates
    characters/fonts into commands the printer uses. Your 'draft' mode
    has some mis-match in that translation.

    >> That is simply not supposed to happen. I'd be investigating this a
    >> LOT further. I don't like unknowns or surprises.

    >
    >I can't remember exactly what I did that resulted in that happening (I
    >don't /think/ I'd done anything unusual), so I don't really have
    >anything to go on. I'm not really sure what to do about it, except to
    >keep a closer watch on anything I do that might cause the MBR to get
    >rewritten.


    The only way to write to the MBR is a privileged call (meaning something
    done by root). In this case, it's something that wrote to /dev/hda
    (as opposed to /dev/hda1, or /dev/hda5 or what-ever). There aren't
    that many things that write directly to the device. What is supposed
    to be there? LILO or GRUB boot data?

    Old guy

  7. Re: House Wiring

    Unruh wrote:
    >> $7000 speaker cables, claiming
    >> they're indistinguishable from cables costing "only" a few hundred.

    >
    > Or lamp cord costing 20 cents a foot.


    Yep, that's about the going price for 18- and 16-gauge wire. I splurge
    and use 16-gauge, myself.

    'Way back in '83-85 I was a retail salesperson for Radio Shack (two
    words then), known as Tandy overseas. Back then, their speaker wire ran
    from 16- to 24-gauge. If anyone asked my advice, I usually suggested
    18-gauge, or 16-gauge if it was a longer run or the customer wanted
    something better. The more astute customers realized that, given the
    quality of equipment that RS sold, anything more would be overkill IMHO.

    Adam

  8. Re: [OT] Power, etc.

    Moe Trin wrote:
    >> I've never heard of a trapezoidal wave. Is it the same thing as a
    >> sawtooth wave?

    >
    > Flat top/bottom. Think of a "square' (or at least rectangular) wave
    > run through a channel with insufficient bandwidth, such that the
    > leading and trailing edge has a substantial width.

    [snip]
    > _ _ _ _
    > /\/\/\/\ / \_/ \ | |_| |_|


    Oh, okay. I assume those rises and falls are basically straight lines,
    not portions of a sine wave. I'd do a Fourier transform on the
    trapezoidal wave, except that I've forgotten how.

    >> A home system would only be powered by the [non-sine-] wave for the
    >> few minutes necessary to do a graceful shutdown when the AC goes out.
    >> Would that really have much effect on performance?

    >
    > In theory (and this really is affected by the design of the PC power
    > supply), the regulation of the DC outputs is less because the primary
    > switching function is done by SCRs.


    So it's a small problem, but not one to worry about?

    >> I did create "another" printer at the same address but set to default
    >> to "draft grayscale," because with that setting it prints at about the
    >> speed I'd expect from a laser printer. Somehow just "draft" had the
    >> vertical spacing about half of normal.

    >
    > That's saying the printer definition is wrong

    [snip]
    > Your 'draft' mode has some mis-match in that translation.


    Yep, another item to add to my "to do" list, towards the top. Gotta sit
    down and actually write out (or type) my computer to-do list.

    > The only way to write to the MBR is a privileged call (meaning something
    > done by root). In this case, it's something that wrote to /dev/hda
    > (as opposed to /dev/hda1, or /dev/hda5 or what-ever). There aren't
    > that many things that write directly to the device. What is supposed
    > to be there? LILO or GRUB boot data?


    Actually, hda's MBR is W2K. One, I'm tech support for my parents, so I
    need some version of Windows, and Windows really wants control of the
    MBR, and Linux doesn't mind that. Two, it's not going to change, so all
    I have to do when I install a new distro is copy its boot record to C:\
    and modify boot.ini -- no need to mess with a new MBR. (I can find a
    link to a "howto" for this if anybody cares.) I don't even remember
    which OS I was using before the MBR vanished, though probably Mandriva
    2007.0. It may have even been the "slot glide" that I sprayed hda's
    contacts with. I don't think it was physical damage, though, because I
    had no problem writing the new MBR.

    Adam, watching the leaves turn pretty colors

  9. Re: [OT] Power, etc.

    On Fri, 26 Oct 2007, in the Usenet newsgroup alt.os.linux.mandrake, in article
    , Adam wrote:

    Moe Trin wrote:

    >> Flat top/bottom. Think of a "square' (or at least rectangular) wave
    >> run through a channel with insufficient bandwidth, such that the
    >> leading and trailing edge has a substantial width.

    >[snip]
    >> _ _ _ _
    >> /\/\/\/\ / \_/ \ | |_| |_|

    >
    >Oh, okay. I assume those rises and falls are basically straight lines,
    >not portions of a sine wave.


    Simple description - yes. This could be a 'slew rate limit' in the
    device somewhere, or a genuine bandwidth function.

    >> In theory (and this really is affected by the design of the PC power
    >> supply), the regulation of the DC outputs is less because the primary
    >> switching function is done by SCRs.

    >
    >So it's a small problem, but not one to worry about?


    I haven't seen much of a problem from it. The advantage of the square
    wave or stepped square wave output from the UPS side is that being
    switches to discrete voltage levels, the efficiency is higher than
    having a (nominally) linear analog output. That results in a longer
    operating time on the battery, as well as less heat.

    There are several fundamental designs of switching power supplies.
    These can range from the least complicated with SCRs doing the
    switching on the input voltage on through more complex designs with
    a rectifier feeding switching/decomutating SCRs, to designs with the
    DC feeding a 'center tapped transformer through switches using a high
    fixed frequency (20-100 KHz) that is pulse-width modulated. These
    supplies can be extra fancy with foldback current limiting and soft
    start. The high switching frequency makes the transformer much smaller.
    I had a system that had multiple +5, +/-20 and 28 volt supplies, all
    using a common design, all rated at 275 Watts, and the transformer
    weighed right around a pound. The regulation of the output voltage was
    done by the fixed frequency pulse width modulation of the switching -
    wider width for more energy, etc. They were a _very_ efficient design.

    >> That's saying the printer definition is wrong

    >[snip]
    >> Your 'draft' mode has some mis-match in that translation.

    >
    >Yep, another item to add to my "to do" list, towards the top. Gotta
    >sit down and actually write out (or type) my computer to-do list.


    Don't do that - it might be demoralizing ;-)

    >> The only way to write to the MBR is a privileged call (meaning
    >> something done by root). In this case, it's something that wrote to
    >> /dev/hda (as opposed to /dev/hda1, or /dev/hda5 or what-ever).


    >Actually, hda's MBR is W2K. One, I'm tech support for my parents, so
    >I need some version of Windows, and Windows really wants control of the
    >MBR, and Linux doesn't mind that.


    Ah, OK! Look at the BIOS setup, and see if there is an 'Anti-Virus'
    mode. In a number of different versions I have access to (mainly
    Award, AMI, and MR), there is a selection to block writing to the
    boot sector. This is only of use to the windoze crowd that often gets
    a boot-sector virus. With the boot sector protection running, any
    attempt to write there causes the BIOS to halt the computer. When
    you actually NEED to do something there, you can go into the BIOS
    setup screen, disable the protection, boot and do your thing, and
    then reboot into the BIOS setup and re-enable the protection. A
    neighbor has some grand-children who were notorious for infecting
    his computer. I couldn't convince him to either switch to *nix or
    at least get an up-to-date anti-malware package, but switching on
    this BIOS protection eliminated the boot-sector viruses. (Didn't
    help the rest of the viruses, but at least it was one less problem to
    deal with.)

    >Adam, watching the leaves turn pretty colors


    Yeah, that's one thing I do miss.

    Old guy

  10. Re: [OT] Telephone cable vs Ethernet for DSL hookup

    I'm considering going from dialup to dsl. I currently have the
    dmarcation point, on the outside of my house, a couple of feet from
    the powerpanel.

    On the plywood where the power panel is mounted, where the telephone
    line from outside connects, there is a 4 wire connection box, that has
    a center terminal, connected to ground via the wire pipe, with lightning
    "resisters". There used to be 2 telephone lines in use here, but
    currently, there is only one.

    The current telephone box, on the plywood panel has 5 telephone lines
    connected going from there throughout the house, and undergroung
    through a sealed pvx pipe to a seperate garage.

    I'm planning on installing a new 4 pin junction box, moving the existing
    5 lines to it, with a single DSL filter connecting it to the existing
    box. I'll be running an unfiltered line from the existing box, to the
    dsl modem/router, with ethernet, from the dsl modem to my computer.

    My question is, should I use a short line from the telephone box
    to the DSL modem, with a 30 ft ethernet cable, or a 30 ft telephone
    line to the DSL, modem, with a short ethernet cable, to the computer's
    ethernet card?

    Which would be likely to give me the best throughput?

    Given the recent conversations here, this seems like a good place to
    ask.

    Regards, Dave Hodgins

    --
    Change nomail.afraid.org to ody.ca to reply by email.
    (nomail.afraid.org has been set up specifically for
    use in usenet. Feel free to use it yourself.)

  11. Re: [OT] Telephone cable vs Ethernet for DSL hookup

    On Sun, 28 Oct 2007 02:47:23 -0400, David W. Hodgins wrote:

    > My question is, should I use a short line from the telephone box to the
    > DSL modem, with a 30 ft ethernet cable, or a 30 ft telephone line to the
    > DSL, modem, with a short ethernet cable, to the computer's ethernet
    > card?


    From tel terminal box to jack for dsl modem all you need is standard
    telephone wire, although cat3 has pretty much become the wire used for
    that these days. However, I'd run cat5 for future upgrading to fios or
    something else. That way you won't have to do anything if/when they
    change to fiber in the future. The DSL modem filter adapter connects to a
    standard phone jack. You don't connect the dsl modem direct to the phone
    line. You can have all the actual phones behind one filter though if you
    like. Or you can put a filter on each end of the line if you have enough.

    --
    Want the ultimate in free OTA SD/HDTV Recorder? http://mythtv.org
    My Tivo Experience http://wesnewell.no-ip.com/tivo.htm
    Tivo HD/S3 compared http://wesnewell.no-ip.com/mythtivo.htm
    AMD cpu help http://wesnewell.no-ip.com/cpu.php

  12. Re: [OT] Power, etc.

    Moe Trin wrote:
    >> Gotta
    >> sit down and actually write out (or type) my computer to-do list.

    >
    > Don't do that - it might be demoralizing ;-)


    I like making lists -- it gives a feeling of accomplishment without
    actually having to DO anything. I tried the opposite, listing what I've
    accomplished since the start of this thread. The biggest was switching
    from dialup to DSL and beefing up security accordingly.

    >> Actually, hda's MBR is W2K.

    >
    > Ah, OK! Look at the BIOS setup, and see if there is an 'Anti-Virus'
    > mode.


    Not in this AMI BIOS, though I've had it in previous systems. No
    complaints when I rewrote the MBR. I suppose I should have copied the
    hosed MBR to something else first, just to see what happened. In fact,
    I copied the entire BIOS to a disk file (all 256K of it), and didn't see
    the text of the BIOS setup screens -- where could that be?

    Adam

  13. Re: [OT] Telephone cable vs Ethernet for DSL hookup

    David W. Hodgins wrote:
    > I'm considering going from dialup to dsl.

    [snip]
    > My question is, should I use a short line from the telephone box
    > to the DSL modem, with a 30 ft ethernet cable, or a 30 ft telephone
    > line to the DSL, modem, with a short ethernet cable, to the computer's
    > ethernet card?


    Finally, a thread where I think I can actually add something, based on
    my limited experience of a few months ago.

    Either way should work. One question that I was told was, basically,
    "unanswerable" was how much difference the last few feet of cabling
    made. After all, there are already several thousand feet of wiring, and
    some splices, between the CO and your home, of uncertain (and not easily
    changeable) quality. But you already knew that both ways would work,
    and wanted to know which was preferable.

    Now I'm not an expert in this (far from it!), but if I were in your
    situation, I'd go for a short cable from your computer to the DSL modem,
    and a long cable from the telephone box to the DSL modem. In fact,
    that's what I did. What *might* make a difference is using a
    broadband/ADSL patch cable for the run from the phone box to the DSL
    modem. This has regular RJ-11/14 connectors on each end, but is made
    with twisted-pair wiring, or at least something that's better than
    regular phone wiring. Just as an example, in the US, RadioShack (not
    the cheapest place) has # 279-108, 25 feet, $15.

    The main advantage of this setup is that this puts the DSL modem (or
    modem/router) near your computer. This way you can see the its lights
    to tell you what's going on, you can fiddle with its switches and
    connections without having to go very far, and you can plug it into the
    same power strip as the rest of your system if you want to. Make sure
    the phone line has a surge protector too.

    I never even thought about the other way, because the "starter kit" that
    Verizon DSL sent me included fairly short RJ-45 (computer to router) and
    RJ-11 (wall jack, or in my case surge protector, to router) cables,
    which clearly implied that the modem/router would be near the computer.
    I already had a dollar-store RJ-11 cable running the 40 feet from the
    duplex adapter on the wall jack to my dialup modem, but soon replaced
    that. And now I have a printer also connected to the DSL router, which
    would be much harder if the router was some distance away. BTW if you
    have a choice between a DSL modem and a modem/router, get the router.

    Regarding those DSL microfilters that you have to attach to everything
    else: You might want to put them right by each device, instead of using
    one central one. In my case, the phone line output of my surge
    protector has one of those duplex jacks. One line goes to the DSL
    router with Verizon's twisted-pair cable, the other goes through the DSL
    filter to my dialup faxmodem and a small phone that's plugged into the
    faxmodem. This means that I can still use my faxmodem to send, and
    presumably receive, faxes, and I can still use the telephone that's
    connected to it. BTW Verizon never asked me how many filters I wanted,
    but sent me a bag of five, all for desktop phones. If I'd needed more,
    or needed one for a wall-mounted phone, I suppose I would have had to
    buy my own, at $10-15 each.

    Adam
    --
    Email: adam seven zero seven at verizon dot net

  14. Re: [OT] Telephone cable vs Ethernet for DSL hookup

    On 2007-10-28, Adam wrote:
    > David W. Hodgins wrote:
    >> I'm considering going from dialup to dsl.

    > [snip]
    >> My question is, should I use a short line from the telephone box
    >> to the DSL modem, with a 30 ft ethernet cable, or a 30 ft telephone
    >> line to the DSL, modem, with a short ethernet cable, to the computer's
    >> ethernet card?

    >
    > Finally, a thread where I think I can actually add something, based on
    > my limited experience of a few months ago.
    >
    > Either way should work. One question that I was told was, basically,
    > "unanswerable" was how much difference the last few feet of cabling
    > made. After all, there are already several thousand feet of wiring, and
    > some splices, between the CO and your home, of uncertain (and not easily
    > changeable) quality. But you already knew that both ways would work,
    > and wanted to know which was preferable.
    >
    > Now I'm not an expert in this (far from it!), but if I were in your
    > situation, I'd go for a short cable from your computer to the DSL modem,
    > and a long cable from the telephone box to the DSL modem. In fact,
    > that's what I did. What *might* make a difference is using a
    > broadband/ADSL patch cable for the run from the phone box to the DSL
    > modem. This has regular RJ-11/14 connectors on each end, but is made
    > with twisted-pair wiring, or at least something that's better than
    > regular phone wiring. Just as an example, in the US, RadioShack (not
    > the cheapest place) has # 279-108, 25 feet, $15.
    >
    > The main advantage of this setup is that this puts the DSL modem (or
    > modem/router) near your computer. This way you can see the its lights
    > to tell you what's going on, you can fiddle with its switches and
    > connections without having to go very far, and you can plug it into the
    > same power strip as the rest of your system if you want to. Make sure
    > the phone line has a surge protector too.


    I'll second that: Having the modem close to the computer is
    useful, especially when troubleshooting.

    > ...
    > Regarding those DSL microfilters that you have to attach to everything
    > else: You might want to put them right by each device, instead of using
    > one central one. In my case, the phone line output of my surge
    > protector has one of those duplex jacks. One line goes to the DSL
    > router with Verizon's twisted-pair cable, the other goes through the DSL
    > filter to my dialup faxmodem and a small phone that's plugged into the
    > faxmodem. This means that I can still use my faxmodem to send, and
    > presumably receive, faxes, and I can still use the telephone that's
    > connected to it. BTW Verizon never asked me how many filters I wanted,
    > but sent me a bag of five, all for desktop phones. If I'd needed more,
    > or needed one for a wall-mounted phone, I suppose I would have had to
    > buy my own, at $10-15 each.


    Or, if you have a GoodWill or comparable thrift store
    nearby, they will probably have a bunch of DSL microfilters
    for a dollar or two each.

    --
    Robert Riches
    spamtrap42@verizon.net
    (Yes, that is one of my email addresses.)

  15. Re: [OT] Power, etc.

    On Sun, 28 Oct 2007, in the Usenet newsgroup alt.os.linux.mandrake, in article
    , Adam wrote:

    >Moe Trin wrote:


    >I like making lists -- it gives a feeling of accomplishment without
    >actually having to DO anything.


    ;-)

    >> Ah, OK! Look at the BIOS setup, and see if there is an 'Anti-Virus'
    >> mode.

    >
    >Not in this AMI BIOS, though I've had it in previous systems. No
    >complaints when I rewrote the MBR.


    Looking through several manuals, I seem some don't have it, and those
    that do have it listed in the "ADVANCED CMOS SETUP". One older
    manual shows the "AMI WINBIOS", which is a GUI. That one has an icon
    on the "Security" page.

    >In fact, I copied the entire BIOS to a disk file (all 256K of it), and
    >didn't see the text of the BIOS setup screens -- where could that be?


    I don't know, but would expect it to be in the CMOS rather than an item
    on the disk. Recall that most systems will boot far enough to run the
    BIOS checks without a hard drive even installed. That's how the hardware
    was initially set up anyway.

    Old guy

  16. Re: [OT] Telephone cable vs Ethernet for DSL hookup

    On Sun, 28 Oct 2007 18:44:55 -0400, Adam wrote:

    > The main advantage of this setup is that this puts the DSL modem (or
    > modem/router) near your computer. This way you can see the its lights
    > to tell you what's going on, you can fiddle with its switches and


    Thanks for the replies from you, Wes and Robert. I think that is the route
    I'll take, since, as you point out, it makes trouble shooting easier, and
    likely won't make any difference to the s/n level.

    > Regarding those DSL microfilters that you have to attach to everything
    > else: You might want to put them right by each device, instead of using
    > one central one. In my case, the phone line output of my surge


    Two of the phones (kitchen and garage) are wall mounts, where there just
    isn't room for the filters. The various dsl providers available in this
    area, all seem to be charging $5 per filter. I think I'll go with two.
    One central one, for the regular phone lines, and one at the computer
    for the fax modem. Thanks for the reminder of the need for that one.

    I'll be sure to get a router dsl modem.

    The power bar I'm using has a surge supressors for the phone line, which
    I'll then split, with one line for the dsl modem, and one filtered line
    for the fax modem.

    Regards, Dave Hodgins

    --
    Change nomail.afraid.org to ody.ca to reply by email.
    (nomail.afraid.org has been set up specifically for
    use in usenet. Feel free to use it yourself.)

  17. Re: [OT] Power, etc.

    Moe Trin wrote:
    >> In fact, I copied the entire BIOS to a disk file (all 256K of it), and
    >> didn't see the text of the BIOS setup screens -- where could that be?

    >
    > I don't know, but would expect it to be in the CMOS rather than an item
    > on the disk. Recall that most systems will boot far enough to run the
    > BIOS checks without a hard drive even installed. That's how the hardware
    > was initially set up anyway.


    I was assuming that all the text strings from the BIOS setup screens
    would be in the boot ROM (EEPROM), not some disk file. I copied the
    EEPROM to a disk file because I couldn't figure out how to access it
    directly, but don't see the text strings from the BIOS setup screens,
    unless they're encoded or something. (I'm one of those people who
    actually go looking for text strings in executable files. Every once in
    a while something interesting shows up.)

    Adam

  18. Re: [OT] Telephone cable vs Ethernet for DSL hookup

    David W. Hodgins wrote:
    > $5 per filter. I think I'll go with two.
    > One central one, for the regular phone lines, and one at the computer
    > for the fax modem. Thanks for the reminder of the need for that one.


    I've seen DSL filters for wall-mount phones (they mount onto wall plate,
    phone mounts onto filter), but what you're planning sounds just as good
    IMHO and is cheaper.

    Adam

  19. Re: [OT] Power, etc.

    On Wed, 31 Oct 2007, in the Usenet newsgroup alt.os.linux.mandrake, in article
    , Adam wrote:

    >Moe Trin wrote:


    >>> In fact, I copied the entire BIOS to a disk file (all 256K of it), and
    >>> didn't see the text of the BIOS setup screens -- where could that be?

    >>
    >> I don't know, but would expect it to be in the CMOS rather than an item
    >> on the disk.


    Duh

    >I was assuming that all the text strings from the BIOS setup screens
    >would be in the boot ROM (EEPROM), not some disk file. I copied the
    >EEPROM to a disk file because I couldn't figure out how to access it
    >directly, but don't see the text strings from the BIOS setup screens,
    >unless they're encoded or something.


    I don't know. You might try disassembling the code if you are that
    interested, but this hasn't been an area I've looked at since the
    mid-eighties. How "full" was the ROM? How much text is shown in
    the setup screens (all of them)? Recall that ASCII is only a seven
    bit code, and it's conceivable that they might have packed the text
    (eight characters in seven bytes) or even compressed it (though I
    wouldn't think there is enough text to make compression viable - you
    have to have space for the un-packing algorithm, and that eats into
    the space you saved by compressing things). A quick scan of the
    ..newsrc file here pulls up 'alt.comp.bios', which doesn't look very
    active (10 articles back had several spams, and went back to early
    September), but appears to have at least one individual answering
    questions.

    >(I'm one of those people who actually go looking for text strings in
    >executable files. Every once in a while something interesting shows
    >up.)


    Yeah, I've had to do that too, while trying to figure out what a
    piece of software is doing. I'll occasionally get some binary without
    the sources that I need to get working. Sometimes I have the source,
    but my C skills are so horrible... ;-)

    Old guy

  20. Re: [OT] Compressed BIOS (was: Power, etc.)

    Moe Trin wrote:
    >> I was assuming that all the text strings from the BIOS setup screens
    >> would be in the boot ROM (EEPROM), not some disk file. I copied the
    >> EEPROM to a disk file because I couldn't figure out how to access it
    >> directly, but don't see the text strings from the BIOS setup screens,
    >> unless they're encoded or something.

    >
    > I don't know. You might try disassembling the code if you are that
    > interested, but this hasn't been an area I've looked at since the
    > mid-eighties. How "full" was the ROM? How much text is shown in
    > the setup screens (all of them)? Recall that ASCII is only a seven
    > bit code, and it's conceivable that they might have packed the text
    > (eight characters in seven bytes) or even compressed it (though I
    > wouldn't think there is enough text to make compression viable - you
    > have to have space for the un-packing algorithm, and that eats into
    > the space you saved by compressing things).


    I did some online research and, sure enough, the usual practice nowadays
    seems to be compressed code in the BIOS, with a BIOS consisting of a
    number of modules on a single EEPROM. It looks like LZH or LHARC
    compression are the most common methods. I also downloaded amibcp75.exe
    (MS-DOS program) which told me which modules were stored where in my AMI
    BIOS, and that the bit from F000:8000 to the end is uncompressed, and
    presumably contains the unpacking code -- F000:C000 thru F000:F700 is
    the bulk of the executable code. A rough disassembly of that section
    came to about 7000 lines. That program also showed that my 256K BIOS
    uncompresses to almost 411K.

    As I see it, I'm faced with four options: (1) look through 7000 lines of
    disassembled code for a clue, (2) search harder for a program to
    uncompress, or at least analyze, the BIOS in more detail, (3) write a
    program that's a floppy's boot sector and uses BIOS calls to write
    0000:0000 thru FFFF:FFFF to the floppy starting with its second sector
    (then use 'dd' under Linux to write it to a file) , or (4) decide the
    whole thing isn't worth the effort. Well, that should keep me busy over
    the weekend. If it doesn't, I can always install Mandriva 2008.

    Adam

+ Reply to Thread
Page 20 of 20 FirstFirst ... 10 18 19 20