.. in every young geeks life where he resolves to locate the designer of  
the motherboard they've just been installing and string them up from the  
nearest tall object.

Thank you unnamed designer for the entire 5mm clearance between the CPU  
heatsink clamps and Northbrige+heatsink on one side and the other 5mm  
clearance between the clamps and a set of capacitors. Also taking into  
account standard case designs putting the screw hole between the serial  
port block and where the PSU normally sits was just as much fun.

Nah it wasn't too bad I just wonder if they ever have to actually install  
one of their creations.

-- 
Phil Cook

--
I once tried to be apathetic, but I just couldn't be bothered
http://flipc.blogspot.com

Post a reply to this message

"Phil Cook" <phi### [at] nospamrocainfreeservecouk> wrote:
> .. in every young geeks life where he resolves to locate the designer of
> the motherboard they've just been installing and string them up from the
> nearest tall object.
>
> Thank you unnamed designer for the entire 5mm clearance between the CPU
> heatsink clamps and Northbrige+heatsink on one side and the other 5mm
> clearance between the clamps and a set of capacitors. Also taking into
> account standard case designs putting the screw hole between the serial
> port block and where the PSU normally sits was just as much fun.
>
> Nah it wasn't too bad I just wonder if they ever have to actually install
> one of their creations.
>
> --
> Phil Cook
>
> --
> I once tried to be apathetic, but I just couldn't be bothered
> http://flipc.blogspot.com

LOL
No. They are completely satisfied with building it. ;-)

Regards
         bluetree

Post a reply to this message

And lo on Fri, 01 Feb 2008 18:37:10 -0000, bluetree <nomail@nomail> did  
spake, saying:

> "Phil Cook" <phi### [at] nospamrocainfreeservecouk> wrote:
>> .. in every young geeks life where he resolves to locate the designer of
>> the motherboard they've just been installing and string them up from the
>> nearest tall object.
>>
>> Thank you unnamed designer for the entire 5mm clearance between the CPU
>> heatsink clamps and Northbrige+heatsink on one side and the other 5mm
>> clearance between the clamps and a set of capacitors. Also taking into
>> account standard case designs putting the screw hole between the serial
>> port block and where the PSU normally sits was just as much fun.
>>
>> Nah it wasn't too bad I just wonder if they ever have to actually  
>> install
>> one of their creations.
>>
>> --
>> Phil Cook
>>
>> --
>> I once tried to be apathetic, but I just couldn't be bothered
>> http://flipc.blogspot.com
>
> LOL
> No. They are completely satisfied with building it. ;-)

Yeah, I wouldn't have minded too much but if you think of it as being  
orientated NS then ES was a vast unsullied plain of nothing-ness.


-- 
Phil Cook

--
I once tried to be apathetic, but I just couldn't be bothered
http://flipc.blogspot.com

Post a reply to this message

"Phil Cook" <phi### [at] nospamrocainfreeservecouk> wrote:
> Yeah, I wouldn't have minded too much but if you think of it as being
> orientated NS then ES was a vast unsullied plain of nothing-ness.

^^
I know, why I am not a geek and actually not able to build a computer from piece
parts all on my own. :-D
You should see the wire-clutter, my computer has inside... (would probably be
better, if things would have a different orientation or I would have another
tower.) ;-)
But I don't complain, because I'm soooo happy, it works and I can do quite
enough with that old thing. :-)

Regards
          bluetree

Post a reply to this message

And lo on Mon, 04 Feb 2008 11:05:27 -0000, bluetree <nomail@nomail> did  
spake, saying:

> "Phil Cook" <phi### [at] nospamrocainfreeservecouk> wrote:
>> Yeah, I wouldn't have minded too much but if you think of it as being
>> orientated NS then ES was a vast unsullied plain of nothing-ness.
>
> ^^
> I know, why I am not a geek and actually not able to build a computer  
> from piece parts all on my own. :-D
> You should see the wire-clutter, my computer has inside... (would  
> probably be
> better, if things would have a different orientation or I would have  
> another tower.) ;-)
> But I don't complain, because I'm soooo happy, it works and I can do  
> quite enough with that old thing. :-)

In a older thread I pointed out how installing things in terms of hardware  
has barely changed compared to software. Depending on what you're doing  
you're still expected to pop the case, screw/unscrew things, fiddle with  
wires and jumper settings while ensuring you're not statically charged and  
shorting anything out.

Can you imagine if the only way to install anying in Windows was to locate  
the correct installation file, move some of the files over manually, and  
then set the right commands in the registry while at the same time  
ensuring it doesn't mess with any of your other programmes?

I mean sure things have got better superficially. I opened the case by  
hand rather then by screwdriver, and... um yeah that's about the only  
difference I noted.

-- 
Phil Cook

--
I once tried to be apathetic, but I just couldn't be bothered
http://flipc.blogspot.com

Post a reply to this message

Phil Cook wrote:

> In a older thread I pointed out how installing things in terms of 
> hardware has barely changed compared to software. Depending on what 
> you're doing you're still expected to pop the case, screw/unscrew 
> things, fiddle with wires and jumper settings while ensuring you're not 
> statically charged and shorting anything out.
> 
> Can you imagine if the only way to install anying in Windows was to 
> locate the correct installation file, move some of the files over 
> manually, and then set the right commands in the registry while at the 
> same time ensuring it doesn't mess with any of your other programmes?
> 
> I mean sure things have got better superficially. I opened the case by 
> hand rather then by screwdriver, and... um yeah that's about the only 
> difference I noted.

Dude. Jumpers.

Seriously. Jumpers.

It used to be the case that getting any new bit of hardware to work 
involved advanced jumper settings, twiddling dip switches and chanting 
over voodoo dolls. Today you can pretty much just plug in a PCI card and 
*expect* it to work immediately - or at least as soon as you put the CD 
in. No IRQ conflicts, no driver mismatches, IT JUST WORKS!!

Anybody who remembers the old skool ways will tell you just how much of 
an improvement that is.

And let's not even get started on all those USB goodies that YOU JUST 
PLUG IN AND THEY GO! It wasn't like that before. Had to turn off the 
whole PC to connect or disconnect anything...

-- 
http://blog.orphi.me.uk/
http://www.zazzle.com/MathematicalOrchid*

Post a reply to this message

And lo on Mon, 04 Feb 2008 14:53:06 -0000, Invisible <voi### [at] devnull> did  
spake, saying:

> Phil Cook wrote:
>
>> In a older thread I pointed out how installing things in terms of  
>> hardware has barely changed compared to software. Depending on what  
>> you're doing you're still expected to pop the case, screw/unscrew  
>> things, fiddle with wires and jumper settings while ensuring you're not  
>> statically charged and shorting anything out.
>>  Can you imagine if the only way to install anying in Windows was to  
>> locate the correct installation file, move some of the files over  
>> manually, and then set the right commands in the registry while at the  
>> same time ensuring it doesn't mess with any of your other programmes?
>>  I mean sure things have got better superficially. I opened the case by  
>> hand rather then by screwdriver, and... um yeah that's about the only  
>> difference I noted.
>
> Dude. Jumpers.
>
> Seriously. Jumpers.

Which I had to change to set the FSB speed; not counting the ones on the  
HDs for master/slave which I didn't need to look at. You're right though I  
recall racks of jumpers which had to be set to arcane precsion to get  
things working, but it's still a testament that's the only advance you can  
spot. Unless you want to go off on SCSI, but that's always been dark juju.

> It used to be the case that getting any new bit of hardware to work  
> involved advanced jumper settings, twiddling dip switches and chanting  
> over voodoo dolls. Today you can pretty much just plug in a PCI card and  
> *expect* it to work immediately -

> or at least as soon as you put the CD in.

So the CD counts as part of the hardware then?

> No IRQ conflicts, no driver mismatches, IT JUST WORKS!!

Describe the process for installing a CPU five/ten years ago compared to  
doing so today. Repeat with a video card leaving out bits you're doing in  
software.

> Anybody who remembers the old skool ways will tell you just how much of  
> an improvement that is.

But how much is that down to an improvement in software. Windows 95  
automatically added new hardware and sought out the drivers for it.

> And let's not even get started on all those USB goodies that YOU JUST  
> PLUG IN AND THEY GO! It wasn't like that before. Had to turn off the  
> whole PC to connect or disconnect anything...

Again though why did the older computers need to reboot - the software  
unable to dynamically load/unload drivers or a physical hardware aspect?

-- 
Phil Cook

--
I once tried to be apathetic, but I just couldn't be bothered
http://flipc.blogspot.com

Post a reply to this message

Phil Cook wrote:

>> Dude. Jumpers.
>>
>> Seriously. Jumpers.
> 
> Which I had to change to set the FSB speed;

Really??

On my motherboard, I just plugged everything in, turned it on, and it 
automatically configured itself and worked. Not one single jumper to set.

> not counting the ones on the 
> HDs for master/slave which I didn't need to look at.

SATA HDs don't have these either.

> You're right though 
> I recall racks of jumpers which had to be set to arcane precsion to get 
> things working, but it's still a testament that's the only advance you 
> can spot.

Can you think of anything else that *requires* improvement then?

>> It used to be the case that getting any new bit of hardware to work 
>> involved advanced jumper settings, twiddling dip switches and chanting 
>> over voodoo dolls. Today you can pretty much just plug in a PCI card 
>> and *expect* it to work immediately -
> 
>> or at least as soon as you put the CD in.
> 
> So the CD counts as part of the hardware then?

Depends on your definition of "works".

If you want to be able to control it, you'll need some software to do 
that. But sure, a couple of things that can be operated without software 
will work straight off...

> Describe the process for installing a CPU five/ten years ago compared to 
> doing so today.

What, you'd like the CPU to teleport itself into the socket instead? :-D

> Repeat with a video card leaving out bits you're doing 
> in software.

Remember when we all had 2 video cards and a pass-through cable to allow 
you to switch between 2D and 3D modes? ;-)

>> Anybody who remembers the old skool ways will tell you just how much 
>> of an improvement that is.
> 
> But how much is that down to an improvement in software. Windows 95 
> automatically added new hardware and sought out the drivers for it.

In the old days, things like IRQ numbers could *only* be adjusted by 
physically moving DIP switches on the circuit board. The point is that 
they CHANGED THE HARDWARE so you could change the IRQ number in 
software. THAT is what has allowed "the software to get better".

>> And let's not even get started on all those USB goodies that YOU JUST 
>> PLUG IN AND THEY GO! It wasn't like that before. Had to turn off the 
>> whole PC to connect or disconnect anything...
> 
> Again though why did the older computers need to reboot - the software 
> unable to dynamically load/unload drivers or a physical hardware aspect?

In the Old Days, if you suddenly plugged or unplugged a device, you 
would likely fry the interface circuitry. Today we use interfaces that 
are expressedly designed to support such actions. That one is purely a 
hardware thing. (Although obviously the software needs to support it too.)

-- 
http://blog.orphi.me.uk/
http://www.zazzle.com/MathematicalOrchid*

Post a reply to this message

And lo on Mon, 04 Feb 2008 15:36:29 -0000, Invisible <voi### [at] devnull> did  
spake, saying:

> Phil Cook wrote:
>
>>> Dude. Jumpers.
>>>
>>> Seriously. Jumpers.
>>  Which I had to change to set the FSB speed;
>
> Really??

<major snip> I'm not saying things haven't gotten better, just a) cmparing  
the pace compared to software and b) would you expect someone who had no  
problem installing Acme Racing Game II to fit a new CPU/Motherboard/Video  
Card/Memory? 'Open up the case and look at all those wires, now just  
remember don't touch any of them else sharks will eat your granny'. You're  
still opening the case being confronted with a PCB and a bunch of wires  
and cables hanging about it's still more a preserve of your geek.

-- 
Phil Cook

--
I once tried to be apathetic, but I just couldn't be bothered
http://flipc.blogspot.com

Post a reply to this message

Phil Cook wrote:

> <major snip> I'm not saying things haven't gotten better, just a) 
> cmparing the pace compared to software and b) would you expect someone 
> who had no problem installing Acme Racing Game II to fit a new 
> CPU/Motherboard/Video Card/Memory? 'Open up the case and look at all 
> those wires, now just remember don't touch any of them else sharks will 
> eat your granny'. You're still opening the case being confronted with a 
> PCB and a bunch of wires and cables hanging about it's still more a 
> preserve of your geek.

Well... have you opened the bonnet of your car recently? I notice newer 
cars have a big sheet of plastic under there, but beyond that it's still 
pretty much a case of "if you don't know what a distributor is, you 
probably shouldn't touch this - otherwise you might die". At least 
computers don't physically kill you if you make a mistake. ;-)

I don't really see how the situation can realistically be improved much. 
You could make all the components less fragile by using some sort of 
casing, and make it to things slot together more easily, but that's 
about it.

-- 
http://blog.orphi.me.uk/
http://www.zazzle.com/MathematicalOrchid*

Post a reply to this message

And lo on Mon, 04 Feb 2008 15:52:02 -0000, Invisible <voi### [at] devnull> did  
spake, saying:

> Phil Cook wrote:
>
>> <major snip> I'm not saying things haven't gotten better, just a)  
>> cmparing the pace compared to software and b) would you expect someone  
>> who had no problem installing Acme Racing Game II to fit a new  
>> CPU/Motherboard/Video Card/Memory? 'Open up the case and look at all  
>> those wires, now just remember don't touch any of them else sharks will  
>> eat your granny'. You're still opening the case being confronted with a  
>> PCB and a bunch of wires and cables hanging about it's still more a  
>> preserve of your geek.
>
> Well... have you opened the bonnet of your car recently? I notice newer  
> cars have a big sheet of plastic under there, but beyond that it's still  
> pretty much a case of "if you don't know what a distributor is, you  
> probably shouldn't touch this - otherwise you might die". At least  
> computers don't physically kill you if you make a mistake. ;-)

Except vehicle manufacturers don't want their customers fiddling under the  
hood, computer hardware manufacturers should at least video  
game/processor/etc retailers should.

> I don't really see how the situation can realistically be improved much.  
> You could make all the components less fragile by using some sort of  
> casing, and make it to things slot together more easily, but that's  
> about it.

Well last time we had this thread I mentioned the motherboard  
daughterboard setup combined with a hot-swap facility. Simple slots on the  
back, press the button and the daughterboard disengages and you pull it  
out and slot the new one in place. Have a dedicated slot for the CPU, one  
for the memory, and one for the video card, with the rest being generic  
PCI-type slots. Similar generic front slots (with a dedicated boot slot)  
for extra hard drives, floppy drives, CD, DVD, tape, etc.

Think how it stands at the moment someone has a computer with one DVD  
player and a floppy drive, he's never used the floppy drive in his life  
and wants to replace it with another DVD rewriter or Blu-ray player.  
What's he got to do to make that happen?

-- 
Phil Cook

--
I once tried to be apathetic, but I just couldn't be bothered
http://flipc.blogspot.com

Post a reply to this message

Invisible escribió:
> At least computers don't physically kill you if you make a mistake.
> ;-)

Open your PSU, or your CRT monitor. See you at hospital if you touch the 
wrong thing.

Post a reply to this message

>> Well... have you opened the bonnet of your car recently? I notice 
>> newer cars have a big sheet of plastic under there, but beyond that 
>> it's still pretty much a case of "if you don't know what a distributor 
>> is, you probably shouldn't touch this - otherwise you might die". At 
>> least computers don't physically kill you if you make a mistake. ;-)
> 
> Except vehicle manufacturers don't want their customers fiddling under 
> the hood, computer hardware manufacturers should at least video 
> game/processor/etc retailers should.

Mmm, true...

> Well last time we had this thread I mentioned the motherboard 
> daughterboard setup combined with a hot-swap facility. Simple slots on 
> the back, press the button and the daughterboard disengages and you pull 
> it out and slot the new one in place.

We have some 10 year old Dell PCs that do exactly this. Press a button 
and the case falls apart. (I think it's meant to unclip, but actually it 
just falls apart.) Press another button and the daughterboard with all 
the PCI stuff on it unclips and slides out of the case. (Don't ever try 
to put it back though...)

> Have a dedicated slot for the CPU, 
> one for the memory, and one for the video card, with the rest being 
> generic PCI-type slots. Similar generic front slots (with a dedicated 
> boot slot) for extra hard drives, floppy drives, CD, DVD, tape, etc.

Well now, if you buy a big expensive server, they actually come with 
hot-swappable drive bays. Take an empty draw out. Put the HD inside it. 
Slot it into the front of the case. Done.

A computer case like that should only set you back... oh, I don't know, 
a few thousand pounds.

Actually, try this:

http://www.apple.com/xserve/raid/

"Starting at just £5,999."

> Think how it stands at the moment someone has a computer with one DVD 
> player and a floppy drive, he's never used the floppy drive in his life 
> and wants to replace it with another DVD rewriter or Blu-ray player. 
> What's he got to do to make that happen?

The problem with all of this is that somebody has to design a set of 
nice slots that everybody will manufacture to. (Otherwise you won't be 
able to buy parts from other suppliers and slot them into your machine!)

If you buy something like a HP BladeCenter you can slot more CPUs, RAM, 
HDs, etc. into and out of it without even powering it off. But you can 
expect to pay at least £40,000 for that kind of thing.

So, in summary, the technology exists, it's just wickedly expensive 
right now. Because the only people who "need it" are the 
high-availability guys who want to hot-swap mission-critical components 
in some big datacenter somewhere. I guess if the industry decides that 
normal people might "want" this kind of thing they *might* get their act 
together... but don't hold your breath.

-- 
http://blog.orphi.me.uk/
http://www.zazzle.com/MathematicalOrchid*

Post a reply to this message

Nicolas Alvarez wrote:
> Invisible escribió:
>> At least computers don't physically kill you if you make a mistake.
>> ;-)
> 
> Open your PSU, or your CRT monitor. See you at hospital if you touch the 
> wrong thing.

Number of screws to open the case: 4.

Number of screws to open the PSU: 17 on average.

Touch the wrong thing and they will scatter your ashes over the place of 
your choice - because you will in fact be fried to a human crisp long 
before you reach any hospital. ;-)

-- 
http://blog.orphi.me.uk/
http://www.zazzle.com/MathematicalOrchid*

Post a reply to this message

On Mon, 04 Feb 2008 16:43:42 +0000, Invisible <voi### [at] devnull> wrote:

>
>Touch the wrong thing and they will scatter your ashes over the place of 
>your choice - because you will in fact be fried to a human crisp long 
>before you reach any hospital. ;-)

Wimp! :) I think that you've got to be quite determined to kill yourself if you
want to use a PSU for that.
Ouch! That woke me up is much more common than You deaded me.
Having said that you do need some common sense.


Regards
	Stephen

Post a reply to this message

And lo on Mon, 04 Feb 2008 16:42:11 -0000, Invisible <voi### [at] devnull> did  
spake, saying:

>>> Well... have you opened the bonnet of your car recently? I notice  
>>> newer cars have a big sheet of plastic under there, but beyond that  
>>> it's still pretty much a case of "if you don't know what a distributor  
>>> is, you probably shouldn't touch this - otherwise you might die". At  
>>> least computers don't physically kill you if you make a mistake. ;-)
>>  Except vehicle manufacturers don't want their customers fiddling under  
>> the hood, computer hardware manufacturers should at least video  
>> game/processor/etc retailers should.
>
> Mmm, true...
>
>> Well last time we had this thread I mentioned the motherboard  
>> daughterboard setup combined with a hot-swap facility. Simple slots on  
>> the back, press the button and the daughterboard disengages and you  
>> pull it out and slot the new one in place.
>
> We have some 10 year old Dell PCs that do exactly this. Press a button  
> and the case falls apart. (I think it's meant to unclip, but actually it  
> just falls apart.) Press another button and the daughterboard with all  
> the PCI stuff on it unclips and slides out of the case. (Don't ever try  
> to put it back though...)

And how much easier is that to work with, or to put it another way how  
much more comforting to a newbie do you think that would be?

>> Have a dedicated slot for the CPU, one for the memory, and one for the  
>> video card, with the rest being generic PCI-type slots. Similar generic  
>> front slots (with a dedicated boot slot) for extra hard drives, floppy  
>> drives, CD, DVD, tape, etc.
>
> Well now, if you buy a big expensive server, they actually come with  
> hot-swappable drive bays. Take an empty draw out. Put the HD inside it.  
> Slot it into the front of the case. Done.
>
> A computer case like that should only set you back... oh, I don't know,  
> a few thousand pounds.
>
> Actually, try this:
>
> http://www.apple.com/xserve/raid/
>
> "Starting at just £5,999."
>
>> Think how it stands at the moment someone has a computer with one DVD  
>> player and a floppy drive, he's never used the floppy drive in his life  
>> and wants to replace it with another DVD rewriter or Blu-ray player.  
>> What's he got to do to make that happen?
>
> The problem with all of this is that somebody has to design a set of  
> nice slots that everybody will manufacture to. (Otherwise you won't be  
> able to buy parts from other suppliers and slot them into your machine!)

You mean like PCI and SATA and 3.5" drive bays :-) The problems that  
others pointed out was the obvious limitation that these slots and  
connections would be fixed; you'd be stuck at the equivalent of AGP or  
ISA. Except when you buy a mother board you're already stuck at these  
limits.

There should be nothing stopping you from popping the components, sliding  
the side of the case open like a drawer, pulling the old motherboard,  
sliding in the new one closing it up (I'm thinking combs from hive type  
ease, minus the bees) and then clocking the components back in assuming  
backwards compatability like USB or slotting in new ones.

> If you buy something like a HP BladeCenter you can slot more CPUs, RAM,  
> HDs, etc. into and out of it without even powering it off. But you can  
> expect to pay at least £40,000 for that kind of thing.
>
> So, in summary, the technology exists, it's just wickedly expensive  
> right now. Because the only people who "need it" are the  
> high-availability guys who want to hot-swap mission-critical components  
> in some big datacenter somewhere. I guess if the industry decides that  
> normal people might "want" this kind of thing they *might* get their act  
> together... but don't hold your breath.

It's wickedly expensive because they're wrapping it up in servers and high  
quality checks and the manufacturers think that the only people who need  
it are people who can afford servers with high-quality checks.

Remember - they charge what they think you'll pay.

-- 
Phil Cook

--
I once tried to be apathetic, but I just couldn't be bothered
http://flipc.blogspot.com

Post a reply to this message

Stephen wrote:

> Wimp! :) I think that you've got to be quite determined to kill yourself if you
> want to use a PSU for that.
> Ouch! That woke me up is much more common than You deaded me.
> Having said that you do need some common sense.

http://www.xkcd.com/242/

-- 
http://blog.orphi.me.uk/
http://www.zazzle.com/MathematicalOrchid*

Post a reply to this message

On Mon, 04 Feb 2008 21:51:18 +0000, Orchid XP v7 <voi### [at] devnull> wrote:

>Stephen wrote:
>
>> Wimp! :) I think that you've got to be quite determined to kill yourself if you
>> want to use a PSU for that.
>> Ouch! That woke me up is much more common than You deaded me.
>> Having said that you do need some common sense.
>
>http://www.xkcd.com/242/

LOL
The third branch and the correct one. It when he says "Yes, that's right". -
True.

Regards
	Stephen

Post a reply to this message

Phil Cook wrote:
> I mean sure things have got better superficially. I opened the case by 
> hand rather then by screwdriver, and... um yeah that's about the only 
> difference I noted.

That's because you haven't been around long enough to solder your own 
components on the board. Two words: wire wrap.


-- 
   Darren New / San Diego, CA, USA (PST)
     On what day did God create the body thetans?

Post a reply to this message

Invisible wrote:
> http://www.apple.com/xserve/raid/
> "Starting at just £5,999."

You can get a Dell with stuff like that for $3000. It's not *that* 
expensive. You can get a 4T hard drive case for like $250 or something 
(without the 4T of drives, of course).

> If you buy something like a HP BladeCenter you can slot more CPUs, RAM, 
> HDs, etc. into and out of it without even powering it off. But you can 
> expect to pay at least £40,000 for that kind of thing.

Or buy a Tandem computer, where when a part fails, it dials up Tandem, 
and they ship you a new one with instructions on where to put it, before 
you even know it failed.

The ones that boggle my mind are the ones where you can upgrade the OS 
without stopping any of the programs.  I'm not quite sure how that works.

-- 
   Darren New / San Diego, CA, USA (PST)
     On what day did God create the body thetans?

Post a reply to this message

And lo on Tue, 05 Feb 2008 03:47:20 -0000, Darren New <dne### [at] sanrrcom>  
did spake, saying:

> Phil Cook wrote:
>> I mean sure things have got better superficially. I opened the case by  
>> hand rather then by screwdriver, and... um yeah that's about the only  
>> difference I noted.
>
> That's because you haven't been around long enough to solder your own  
> components on the board. Two words: wire wrap.

Yes I'm not that old :-) Even with that looking at it from a non-geek  
viewpoint compare the hardware then to the hardware now, repeat for the  
'software'.

-- 
Phil Cook

--
I once tried to be apathetic, but I just couldn't be bothered
http://flipc.blogspot.com

Post a reply to this message

Darren New wrote:
> Invisible wrote:
>> http://www.apple.com/xserve/raid/
>> "Starting at just £5,999."
> 
> You can get a Dell with stuff like that for $3000. It's not *that* 
> expensive. You can get a 4T hard drive case for like $250 or something 
> (without the 4T of drives, of course).

Oh, "only" $3,000. Bargin. ;-)

>> If you buy something like a HP BladeCenter you can slot more CPUs, 
>> RAM, HDs, etc. into and out of it without even powering it off. But 
>> you can expect to pay at least £40,000 for that kind of thing.
> 
> Or buy a Tandem computer, where when a part fails, it dials up Tandem, 
> and they ship you a new one with instructions on where to put it, before 
> you even know it failed.

Oh, that's cute. (Are they still going?)

> The ones that boggle my mind are the ones where you can upgrade the OS 
> without stopping any of the programs.  I'm not quite sure how that works.

Neither do I - but people claim that functional programming is going to 
revolutionise the industry by making this kind of thing standard. 
[insert sarcasm here]

-- 
http://blog.orphi.me.uk/
http://www.zazzle.com/MathematicalOrchid*

Post a reply to this message

Invisible wrote:
]
> over voodoo dolls. Today you can pretty much just plug in a PCI card and 
> *expect* it to work immediately - or at least as soon as you put the CD 
> in. No IRQ conflicts, no driver mismatches, IT JUST WORKS!!
>

Unless you're the dumbass that used to work where I currently work and 
plug in the card without first shutting down the computer. Yeah, that 
magic smoke ... that's what used to make it work, you let it out. 
Thankfully that person quit, saving the management the difficult task of 
canning them. Not a single line of their code exists here, anymore, 
either. :)

Post a reply to this message

Invisible wrote:

> And let's not even get started on all those USB goodies that YOU JUST 
> PLUG IN AND THEY GO! It wasn't like that before. Had to turn off the 
> whole PC to connect or disconnect anything...

If I had a nickel for every time that was patently false.

Sure, flash drives work easily enough, but any other piece of hardware 
usually requires you follow a specific sequence before you even plug the 
device in. Otherwise, Windows (not sure how this works on other 
platforms) will do magically bad things, such as install a persistent 
incorrect driver that takes several steps and a few sacrificed chickens 
to remove.

Post a reply to this message

Invisible wrote:

> 
> What, you'd like the CPU to teleport itself into the socket instead? :-D
> 

Remember the days before ZIF sockets? Seat the processor, jam it down 
with enough pressure to potentially crack the main board, and hope to 
hell you didn't just bend half a dozen pins. Ohh, and there were also 
fun memory upgrades, using individual DIP chips crammed into sockets. 
And god forbid if you had to remove one... Then there was SIP modules, 
with their delicate pins which you had to align precisely or you'd bend 
one. SIMMs and DIMMs are a much better improvement, especially with the 
lift and snap-in insertion that many have today. The straight down syle 
still required some pressure to get them to click in, but removal is a 
breeze.

> 
> In the Old Days, if you suddenly plugged or unplugged a device, you 
> would likely fry the interface circuitry. Today we use interfaces that 
> are expressedly designed to support such actions. That one is purely a 
> hardware thing. (Although obviously the software needs to support it too.)
> 

Oh, you still can, today .. see above.  That is the nice thing about USB 
is that devices are hot-swappable. Of course, so are devices that hang 
off of the serial port, or parallel port. (and they have been for a very 
long time ...)

Post a reply to this message

Stephen wrote:

> Ouch! That woke me up is much more common than You deaded me.
> Having said that you do need some common sense.

Popping open a camera can have a similar effect. Most cameras have an 
onboard flash, which means they have a rather sizable capacitor. Touch 
the wrong thing ...

Post a reply to this message

Mike Raiford escribió:
> Stephen wrote:
> 
>> Ouch! That woke me up is much more common than You deaded me.
>> Having said that you do need some common sense.
> 
> Popping open a camera can have a similar effect. Most cameras have an 
> onboard flash, which means they have a rather sizable capacitor. Touch 
> the wrong thing ...

I did that!!!

I had a very old Polaroid camera, and its flash (looked like an optional 
you had to plug). Bad idea touching things with the flash box open, 
while it had batteries. Me: it has batteries, it's only 3v. My dad: no, 
it also has a big capacitor with more energy. I clearly didn't listen. 
The it actually did a big *flash*, in the middle of the air, a few cms 
above where I touched. Legs shaking for around 10 minutes :D

Post a reply to this message

Mike Raiford wrote:

> Sure, flash drives work easily enough, but any other piece of hardware 
> usually requires you follow a specific sequence before you even plug the 
> device in. Otherwise, Windows (not sure how this works on other 
> platforms) will do magically bad things, such as install a persistent 
> incorrect driver that takes several steps and a few sacrificed chickens 
> to remove.

Clearly I don't have enough USB gear - I haven't had that problem. 
(Although something similar happened with a SCSI tape drive. But then, 
it's a SCSI tape drive...)

Not that anybody cares, but I have an outboard USB sound card, USB MIDI 
interface, USB printer and a USB mouse. The mouse keeps not working... 
everything else works flawlessly.

-- 
http://blog.orphi.me.uk/
http://www.zazzle.com/MathematicalOrchid*

Post a reply to this message

Invisible wrote:
> 
> Touch the wrong thing and they will scatter your ashes over the place of
> your choice - because you will in fact be fried to a human crisp long
> before you reach any hospital. ;-)
> 

Mostly not. You'll need some ultimate stupidity to get your hands around
the PSU so that the shock won't throw them off from the PSU. Possibility
of getting some current trough your heart is the real beast - 50mA can
be enough to kill a healthy man.

-- 
Eero "Aero" Ahonen
   http://www.zbxt.net
      aer### [at] removethiszbxtnetinvalid

Post a reply to this message

On Tue, 05 Feb 2008 12:12:29 -0600, Mike Raiford <mra### [at] hotmailcom> wrote:

>Stephen wrote:
>
>> Ouch! That woke me up is much more common than You deaded me.
>> Having said that you do need some common sense.
>
>Popping open a camera can have a similar effect. Most cameras have an 
>onboard flash, which means they have a rather sizable capacitor. Touch 
>the wrong thing ...

Caps should be shorted out before working. Use a resistor or a bucket of water
:)
Drat I can't remember if I lie or tell the truth :)

Regards
	Stephen

Post a reply to this message

Mike Raiford wrote:
> Sure, flash drives work easily enough, but any other piece of hardware
> usually requires you follow a specific sequence before you even plug the
> device in. Otherwise, Windows (not sure how this works on other
> platforms) will do magically bad things, such as install a persistent
> incorrect driver that takes several steps and a few sacrificed chickens
> to remove.

Yes, that IS a Windows-feature. And yes, the hw-support is still much
better in Windows (more drivers), even though it has its problems.

-- 
Eero "Aero" Ahonen
   http://www.zbxt.net
      aer### [at] removethiszbxtnetinvalid

Post a reply to this message

Invisible wrote:
> Darren New wrote:
>> Invisible wrote:
>>> http://www.apple.com/xserve/raid/
>>> "Starting at just £5,999."
>>
>> You can get a Dell with stuff like that for $3000. It's not *that* 
>> expensive. You can get a 4T hard drive case for like $250 or something 
>> (without the 4T of drives, of course).
> 
> Oh, "only" $3,000. Bargin. ;-)

Well, the $3000 covers dual power supplies, hardware raid, dual giabit 
ethernet, rack-mount, a year of Linux support from Novel, dual SATA 500G 
hard drives, 16G of RAM, four dual-core 2.<mumble> CPUs...

Yeah, not too bad.

It's not your desktop machine, no, but $3000 isn't a whole lot for a 
company. You can apparently lease them from Dell for <$50/month. Which 
is like one good person's salary for an hour, counting taxes and 
expenses and such the employer has to pay.

You can't look at stuff you run a business on and say "Wow, that's 
expensive for a desktop", any more than you can look at a bus and say 
"Wow, that costs way too much for something to commute to work in."

-- 
   Darren New / San Diego, CA, USA (PST)
     On what day did God create the body thetans?

Post a reply to this message

Eero Ahonen wrote:
> Yes, that IS a Windows-feature. 

Yeah. It sucks to give MS a driver for your hardware to be distributed 
on the OS install disks, only to find out it doesn't actually work. :-)

-- 
   Darren New / San Diego, CA, USA (PST)
     On what day did God create the body thetans?

Post a reply to this message

On Tue, 05 Feb 2008 09:32:02 +0000, Invisible wrote:

>> You can get a Dell with stuff like that for $3000. It's not *that*
>> expensive. You can get a 4T hard drive case for like $250 or something
>> (without the 4T of drives, of course).
> 
> Oh, "only" $3,000. Bargin. ;-)

I remember spending $12,000 on two 486 systems.  Compaq Prosignia servers.

I still have them in the basement, in fact.  No, they weren't bought by a 
business - I spent my own money on them (in fairness, an advance on a 
book I wrote back in the early 90's, plus a little of my own money).

Jim

Post a reply to this message

In article <47a8a92a$1@news.povray.org>, voi### [at] devnull says...
> Mike Raiford wrote:
> 
> > Sure, flash drives work easily enough, but any other piece of hardware
 
> > usually requires you follow a specific sequence before you even plug th
e 
> > device in. Otherwise, Windows (not sure how this works on other 
> > platforms) will do magically bad things, such as install a persistent
 
> > incorrect driver that takes several steps and a few sacrificed chickens
 
> > to remove.
> 
> Clearly I don't have enough USB gear - I haven't had that problem. 
> (Although something similar happened with a SCSI tape drive. But then, 
> it's a SCSI tape drive...)
> 
> Not that anybody cares, but I have an outboard USB sound card, USB MIDI
 
> interface, USB printer and a USB mouse. The mouse keeps not working... 
> everything else works flawlessly.
> 
Its mostly low end, 'cheapy', stuff in my experience. Like the 
inexpensive camera which a) didn't support **any** method to read its 
memory as a HD, under *any* OS, and b) would suffer a massive memory 
wipe if you *didn't* first install the drivers for it. Frankly, I wasn't 
real impressed with the HDSD reader I got free with the 4GB card... 
Stupid thing *acts* like its running through like USB v0.01 alpha or 
something.. It doesn't read anything, including non-SD at a normal speed 
and when trying to transfer large files (you know, anything bigger than 
about 100MB...) it actually crashes. Most stuff works fine. Most stuff 
designed by companies that have a clue work more or less OK without 
extra drivers, but the drivers are needed for more than just reading the 
internal memory. Then there are things like the v262 Motorola phone, 
which their driver/transfer software didn't support out of the box, 
needed two major patches to work, can't be accessed as a HD at all, and 
tended to frell the XP machine you installed the driver on for *all* 
phones, if something went wrong with XP SP2 and the driver, who didn't 
like each other much.

Yeah, most stuff works OK, once you do the disc switch shuffle, but some 
stuff.... Nothing short of an act of the FSM will get working acceptably 
(or maybe a sacrificed chicken).

-- 
void main () {
  If Schrödingers_cat is alive or version > 98 {
    if version = "Vista" {
      call slow_by_half();
      call DRM_everything();
    }
    call functional_code();
  }
  else
    call crash_windows();
}

<A HREF='http://www.daz3d.com/index.php?refid=16130551'>Get 3D Models,
 
3D Content, and 3D Software at DAZ3D!</A>

Post a reply to this message

>> Oh, "only" $3,000. Bargin. ;-)
> 
> Well, the $3000 covers dual power supplies, hardware raid, dual giabit 
> ethernet, rack-mount, a year of Linux support from Novel, dual SATA 500G 
> hard drives, 16G of RAM, four dual-core 2.<mumble> CPUs...
> 
> Yeah, not too bad.

That's actually quite good.

The server I'm using was £2,000 - which is currently about $4,000 - and 
it doesn't have half that stuff.

OTOH, Phil was complaining that you can't buy desktop machines that slot 
together like lego. And I countered that you *can* buy such equipment, 
it's just far too expensive to use as a desktop. ;-)

-- 
http://blog.orphi.me.uk/
http://www.zazzle.com/MathematicalOrchid*

Post a reply to this message

And lo on Wed, 06 Feb 2008 09:57:38 -0000, Invisible <voi### [at] devnull> did  
spake, saying:

>>> Oh, "only" $3,000. Bargin. ;-)
>>  Well, the $3000 covers dual power supplies, hardware raid, dual giabit  
>> ethernet, rack-mount, a year of Linux support from Novel, dual SATA  
>> 500G hard drives, 16G of RAM, four dual-core 2.<mumble> CPUs...
>>  Yeah, not too bad.
>
> That's actually quite good.
>
> The server I'm using was £2,000 - which is currently about $4,000 - and  
> it doesn't have half that stuff.
>
> OTOH, Phil was complaining that you can't buy desktop machines that slot  
> together like lego. And I countered that you *can* buy such equipment,  
> it's just far too expensive to use as a desktop. ;-)

Succinctly put thank you. The whole thing is why hasn't the slot-together  
style reached down into the standard market, and if it did how would that  
affect 'normal' users' computer buying patterns. Instead of buying an  
entire new computer would they instead opt to upgrade the old one because  
it's a piece of cake to do.

I mean seriously I've listened to friends talking about upgrading their  
entire computer to get more speed out of a game or something when all they  
need to do is switch the video card (or *to* a video card rather then the  
MB GPU). To them the computer is a lump like a television, opening the  
case doesn't occur to them and I don't think they'd be enthused by what  
they'd find if they did.

-- 
Phil Cook

--
I once tried to be apathetic, but I just couldn't be bothered
http://flipc.blogspot.com

Post a reply to this message

Phil Cook wrote:

> Succinctly put thank you. The whole thing is why hasn't the 
> slot-together style reached down into the standard market, and if it did 
> how would that affect 'normal' users' computer buying patterns. Instead 
> of buying an entire new computer would they instead opt to upgrade the 
> old one because it's a piece of cake to do.

Currently manufacturers assume that the only reason you could want to be 
able to replace parts "easily" is because you need to do it fast, and if 
you need to do it fast it must be because you have a "high availability" 
setup. And if that's the case, the cost of downtime dwarfs any hardware 
price you can come up with. These types of people will pay through the 
nose to avoid downtime - and so that's what gets charged.

In fairness, it's much harder to design hardware this way. Current 
hardware is "easy enough" that most people can manage it, and there's no 
real large pressure for that to change.

> I mean seriously I've listened to friends talking about upgrading their 
> entire computer to get more speed out of a game or something when all 
> they need to do is switch the video card (or *to* a video card rather 
> then the MB GPU). To them the computer is a lump like a television, 
> opening the case doesn't occur to them and I don't think they'd be 
> enthused by what they'd find if they did.

Heh. I bet the hardware guys would want to prevent that. ;-)



In fairness, if you want to replace the CPU, you will probably end up 
having to replace virtually the entire PC anyway.

(You'll almost certainly need a new motherboard and new RAM. And 
replacing a motherboard is currently *hard*. So many things to 
disconnect and reconnect. I especially enjoy playing with the header 
pins for the front plate...)

-- 
http://blog.orphi.me.uk/
http://www.zazzle.com/MathematicalOrchid*

Post a reply to this message

Stephen wrote:

> 
> Caps should be shorted out before working. Use a resistor or a bucket of water
> :)
> Drat I can't remember if I lie or tell the truth :)
> 

Yes. A resistor works. Extra fun shorting it with a screwdriver. Causes 
the whole "holy crap! I just vaporized the end of my screwdriver" 
reaction, though. Actually, don't try that one at home... It stresses 
the capacitor, and could irreparably damage it.

Post a reply to this message

Mike Raiford wrote:

> Yes. A resistor works. Extra fun shorting it with a screwdriver. Causes 
> the whole "holy crap! I just vaporized the end of my screwdriver" 
> reaction, though. Actually, don't try that one at home... It stresses 
> the capacitor, and could irreparably damage it.

I'm sorry, it stresses *the capacitor*?? o_O

-- 
http://blog.orphi.me.uk/
http://www.zazzle.com/MathematicalOrchid*

Post a reply to this message

Invisible wrote:

> 
> I'm sorry, it stresses *the capacitor*?? o_O
> 

Yep. And any domestic animals, or small children in the vicinity. 
Seriously, though, most capacitors don't do well when discharging all of 
their energy inside of a microsecond.

Post a reply to this message

> I especially enjoy playing with the header pins for the front plate...)

Yeh, you'd think they would have come up with a standard multi-pin 
plug/socket standard for that by now...

Post a reply to this message

On Wed, 06 Feb 2008 06:28:53 -0600, Mike Raiford <mra### [at] hotmailcom> wrote:

>Stephen wrote:
>
>> 
>> Caps should be shorted out before working. Use a resistor or a bucket of water
>> :)
>> Drat I can't remember if I lie or tell the truth :)
>> 
>
>Yes. A resistor works. Extra fun shorting it with a screwdriver. Causes 
>the whole "holy crap! I just vaporized the end of my screwdriver" 
>reaction, though. Actually, don't try that one at home... It stresses 
>the capacitor, and could irreparably damage it.

He, he! I remember once shorting out a tuning capacitor on a CVT. There was only
about a quarter of a volt across the terminals. The spark threw the screwdriver
across the room and took a chunk out of it. Mind you the capacitor was 0.1 of a
farad. Gulp! Lesson learned :)


Regards
	Stephen

Post a reply to this message

scott wrote:
>> I especially enjoy playing with the header pins for the front plate...)
> 
> Yeh, you'd think they would have come up with a standard multi-pin 
> plug/socket standard for that by now...

It's better than that.

My dad can't plug in the HD light. Because on the case it's a 3-pin 
block with the middle pin not connected. But on the motherboard it's two 
adjacent pins...

-- 
http://blog.orphi.me.uk/
http://www.zazzle.com/MathematicalOrchid*

Post a reply to this message

>> I'm sorry, it stresses *the capacitor*?? o_O
> 
> Yep. And any domestic animals, or small children in the vicinity. 
> Seriously, though, most capacitors don't do well when discharging all of 
> their energy inside of a microsecond.

Hmm. I thought that was the design goal of a capacitor?

-- 
http://blog.orphi.me.uk/
http://www.zazzle.com/MathematicalOrchid*

Post a reply to this message

Stephen wrote:
> He, he! I remember once shorting out a tuning capacitor on a CVT. There was only
> about a quarter of a volt across the terminals. The spark threw the screwdriver
> across the room and took a chunk out of it.

Are you sure it wasn't you who threw the screwdriver across the room? 
;-) I know I would have!

Post a reply to this message

Invisible wrote:

> 
> Hmm. I thought that was the design goal of a capacitor?
> 

In the case of the capacitor storing the energy for a flash, flash 
discharges are actually rather finely timed. Something to the effect of 
1 millisecond for a full-power flash, this discharges the capacitor 
quickly, but not instantly. Lower power strobes only allow a partial 
discharge at the same rate, which is why recycle time is much quicker 
when using a flash at a lower power, because the cap isn't empty. 
Shorting a capacitor without a resistor allows high amounts of current 
to flow, which can cause an arc across the dielectric, thus destroying 
the capacitor. At least that's how I understand it.

In other applications, discharge is much slower, such as when they are 
used to smooth ripple current from a DC power supply.

Post a reply to this message

Mike Raiford wrote:

>> Hmm. I thought that was the design goal of a capacitor?
>>
> 
> In the case of the capacitor storing the energy for a flash, flash 
> discharges are actually rather finely timed. Something to the effect of 
> 1 millisecond for a full-power flash, this discharges the capacitor 
> quickly, but not instantly. Lower power strobes only allow a partial 
> discharge at the same rate, which is why recycle time is much quicker 
> when using a flash at a lower power, because the cap isn't empty. 
> Shorting a capacitor without a resistor allows high amounts of current 
> to flow, which can cause an arc across the dielectric, thus destroying 
> the capacitor. At least that's how I understand it.
> 
> In other applications, discharge is much slower, such as when they are 
> used to smooth ripple current from a DC power supply.

...and here I was thinking they use capacitors to tune MHz-frequency 
oscilators...

-- 
http://blog.orphi.me.uk/
http://www.zazzle.com/MathematicalOrchid*

Post a reply to this message

On Wed, 06 Feb 2008 14:12:01 +0000, Bill Pragnell <bil### [at] hotmailcom>
wrote:

>
>Are you sure it wasn't you who threw the screwdriver across the room? 
>;-) I know I would have!

It might have been, it was over 30 years go. I do remember I bounced it off the
terminals and it bounced back a lot further than I expected and left my hand. It
was lucky that no one was standing behind me at the time. Now I take into
account what my reactions might be if something like that happens. 
Reactions can be hard to control. When you are working with electricity you
expect shocks and are quick to move. But sometimes it is better not to move as
you could be putting yourself into danger say by moving it against other live
terminals or ripping the skin off on sharp objects. One rule is to always be
insulated from the ground and another rule is to keep one hand in your pocket.
Never hold onto the side of an enclosure for balance don't let the electricity
escape to ground through your body especially if the path could be across your
heart. Also be aware of the differences between AC and DC. AC throws you off DC
can make you grab hold. 
One other word of wisdom. Don't urinate on a sub-power station or transformer.
(It was been done.)

Regards
	Stephen

Post a reply to this message

On Wed, 06 Feb 2008 08:10:29 -0600, Mike Raiford <mra### [at] hotmailcom> wrote:

>
>In other applications, discharge is much slower, such as when they are 
>used to smooth ripple current from a DC power supply.

If the cap is across the supply and ground then it is acting as a short circuit
to AC but a block to DC.These caps generally have a discharging resistor > 1 meg
across them to slowly discharge it when there is no power applied.

Regards
	Stephen

Post a reply to this message