Charlie Gibbs <cgibbs@kltpzyxm.invalid> wrote:
> On 2021-09-23, Rich <rich@example.invalid> wrote:
>
>> In comp.os.linux.misc Peter Flass <peter_flass@yahoo.com> wrote:
>>
>>> Carlos E. R. <robin_listas@es.invalid> wrote:
>>>
>>>> On 22/09/2021 21.57, Andreas Kohlbach wrote:
>>>>
>>>>> On Wed, 22 Sep 2021 09:13:12 +0200, Carlos E. R. wrote:
>>>>>
>>>>>> Might not work (a capture). Just a guess. If you do try,
>>>>>> make sure to not use mp3. Now that I think, I would try the
>>>>>> experiment, to find out.
>>>>>
>>>>> There was a short lived [1] UK computer show "4 Computer Buffs"
>>>>> <https://www.imdb.com/title/tt1178499/> I never heard of before.
>>>>> They sent program data via audio the audience record. When I found
>>>>> the show on Youtube I tested that (extracted the audio at that
>>>>> position) and ran the resulting WAV in an emulator fir that
>>>>> particular machine on my PC. It my amazement it worked.
>>>>
>>>> Wow.
>>>
>>> Systems that saved programs on cassette used an audio format. Due to
>>> the sloppiness of the media, I think the recording format had to be
>>> pretty robust.
>>
>> At least in the case of the Atari 400/800 cassette format it was a very
>> simple format:
>>
>> Format details are here: https://www.atariarchives.org/dere/chaptC.php
>>
>> 132 byte records, two start bytes for 'speed detection', a control
>> byte, 128 data bytes, and a single checksum byte (and the checksum is
>> just a simple endaround carry sum of the 131 other bytes in the record).
>>
>> The physical byte encoding on the tape was frequency shift keying, with
>> 5327 Hz for a mark and 3995 Hz for a space.
>>
>> So it at least it had a simple checksum, but the packet format was
>> hardly "robust". Workable, but memory of those days was that the
>> cassette was quite flakey as a data storage format, sometimes it
>> worked, sometimes it did not. And when it did not rereading things all
>> over again sometimes magically had them work.
>
> Ah yes, I remember the good old days with my IMSAI. I didn't have
> cassette decks, but I had a couple of reel-to-reel decks, so I broke
> into their motor circuits and built a control box that would use the
> cassette motor control circuits to activate relays to switch 110-volt
> motor power on and off.

I was astonished when I got my first home computer with a cassette drive
that it didn’t do this!

>
> I didn't have a real cassette interface in the beginning, but I did
> have a Bell 202 modem (1200 bps async) that I picked up somewhere.
> I recorded its output to tape and played it back in - it worked well
> enough, although when I finally scraped up the bucks for a CUTS board
> it was more reliable (but not much faster).
>

--
Pete

On 2021-09-23, Eli the Bearded <*@eli.users.panix.com> wrote:

> In comp.os.linux.misc, Charlie Gibbs <cgibbs@kltpzyxm.invalid> wrote:
>
>> 1 7/8 ips, actually. But even a consumer-grade 1/4-inch 4-track
>> reel-to-reel deck running at 7 1/2 ips has an inherent 9dB advantage
>> in signal-to-noise ratio over a cassette - not to mention much better
>> high-frequency response. That's why I stuck with reel-to-reel when
>> the world went to cassettes.
>
> https://en.wikipedia.org/wiki/NT_(cassette)
>
> Tiny mag tape can work, but you have to be careful.

Interesting. I hadn't heard about that device.

I thought Sony gave up on proprietary media when the Elcaset flopped.

>> I've read that CDs - including remixes of good vinyl recordings -
>> are often compressed down to something like a 12dB dynamic range.
>
> People who know more than me about audio have explained that as
> "a make the CDs sound 'good' on car stereos" effort. In better
> environments, you can hear how much worse it is, but competing with the
> engine sounds and traffic sounds, the compressed audio seems better.
>
> The proper fix would have been better car stereos, but...

I doubt even that would work. Engine and road noise set a pretty high
noise floor - you probably couldn't get a car stereo loud enough to
get 80 dB above that. Nor would you want to - unless you like boom cars.
If you set the volume at a bearable level, the quiet bits on a recording
with good dynamic range will be well below the noise floor. So you're
stuck with the choice of compression if you want to hear those quiet bits
in your car, or no compression if you want a more natural sound at home.
The closest we could come to a fix would be better soundproofing in cars.

In the meantime, a recording that sounds good at home will sound lousy
in your car - and vice versa. I suppose we could have two mixes of
the same album - one compressed and one not - but I doubt enough people
would care enough to make it worthwhile for record companies to issue
two different versions.

However, a friend has a car that offers a great work-around -
switchable compression in the stereo system. That way you can
hear everything on the road, but if you're parked somewhere and
have no background noise to worry about, you can switch off the
compression and hear the recording in all its glory. Of course,
you have to have a recording with good dynamic range in the first
place.

>> I got my hands on an MP3 of a wonderful Joni Mitchell track that
>> I cherish on vinyl. It sounded so horrible that I deleted it -
>> and I don't delete _anything_.
>
> Vinyl also has the quirk that outer edge tracks have better quality than
> center tracks. Most other media doesn't suffer that restriction.

Sounds like the different zones on a hard disk...

--
/~\ Charlie Gibbs | Life is perverse.
\ / <cgibbs@kltpzyxm.invalid> | It can be beautiful -
X I'm really at ac.dekanfrus | but it won't.
/ \ if you read it the right way. | -- Lily Tomlin

In comp.os.linux.misc, Charlie Gibbs <cgibbs@kltpzyxm.invalid> wrote:
> However, a friend has a car that offers a great work-around -
> switchable compression in the stereo system. That way you can
> hear everything on the road, but if you're parked somewhere and
> have no background noise to worry about, you can switch off the
> compression and hear the recording in all its glory.

That sort of feature is what I meant by "better car stereos". I'm not an
audio guy and will be quite imprecise on such matters.

(Quoting me)

>> Vinyl also has the quirk that outer edge tracks have better quality than
>> center tracks. Most other media doesn't suffer that restriction.
> Sounds like the different zones on a hard disk...

Back to vintage computer reminiscence: I can remember when "where to put
your swap partition?" was one of those zones discussion points.

Elijah
------
slowly computers become less machines and more rocks

On 23/09/2021 21.55, Peter Flass wrote:
> Charlie Gibbs <cgibbs@kltpzyxm.invalid> wrote:
>> On 2021-09-23, Rich <rich@example.invalid> wrote:
>>
>>> In comp.os.linux.misc Peter Flass <peter_flass@yahoo.com> wrote:
>>>
>>>> Carlos E. R. <robin_listas@es.invalid> wrote:
>>>>
>>>>> On 22/09/2021 21.57, Andreas Kohlbach wrote:
>>>>>
>>>>>> On Wed, 22 Sep 2021 09:13:12 +0200, Carlos E. R. wrote:
>>>>>>
>>>>>>> Might not work (a capture). Just a guess. If you do try,
>>>>>>> make sure to not use mp3. Now that I think, I would try the
>>>>>>> experiment, to find out.
>>>>>>
>>>>>> There was a short lived [1] UK computer show "4 Computer Buffs"
>>>>>> <https://www.imdb.com/title/tt1178499/> I never heard of before.
>>>>>> They sent program data via audio the audience record. When I found
>>>>>> the show on Youtube I tested that (extracted the audio at that
>>>>>> position) and ran the resulting WAV in an emulator fir that
>>>>>> particular machine on my PC. It my amazement it worked.
>>>>>
>>>>> Wow.
>>>>
>>>> Systems that saved programs on cassette used an audio format. Due to
>>>> the sloppiness of the media, I think the recording format had to be
>>>> pretty robust.
>>>
>>> At least in the case of the Atari 400/800 cassette format it was a very
>>> simple format:
>>>
>>> Format details are here: https://www.atariarchives.org/dere/chaptC.php
>>>
>>> 132 byte records, two start bytes for 'speed detection', a control
>>> byte, 128 data bytes, and a single checksum byte (and the checksum is
>>> just a simple endaround carry sum of the 131 other bytes in the record).
>>>
>>> The physical byte encoding on the tape was frequency shift keying, with
>>> 5327 Hz for a mark and 3995 Hz for a space.
>>>
>>> So it at least it had a simple checksum, but the packet format was
>>> hardly "robust". Workable, but memory of those days was that the
>>> cassette was quite flakey as a data storage format, sometimes it
>>> worked, sometimes it did not. And when it did not rereading things all
>>> over again sometimes magically had them work.
>>
>> Ah yes, I remember the good old days with my IMSAI. I didn't have
>> cassette decks, but I had a couple of reel-to-reel decks, so I broke
>> into their motor circuits and built a control box that would use the
>> cassette motor control circuits to activate relays to switch 110-volt
>> motor power on and off.
>
> I was astonished when I got my first home computer with a cassette drive
> that it didn’t do this!

I have a foggy memory that it used the switch on/off wires of the
microphone, which in some/all tape machines stopped the motor.

That would be the Sinclair Spectrum if any, but can't vouch for it.

--
Cheers,
Carlos E.R.

On 23/09/2021 21.06, Eli the Bearded wrote:
> In comp.os.linux.misc, Charlie Gibbs <cgibbs@kltpzyxm.invalid> wrote:

...

>> I got my hands on an MP3 of a wonderful Joni Mitchell track that
>> I cherish on vinyl. It sounded so horrible that I deleted it -
>> and I don't delete _anything_.
>
> Vinyl also has the quirk that outer edge tracks have better quality than
> center tracks. Most other media doesn't suffer that restriction.
>
> Elijah
> ------
> most other media doesn't let you drill an off-center hole for fun

Ow :-D

--
Cheers,
Carlos E.R.

On 23/09/2021 20.31, Charlie Gibbs wrote:
> On 2021-09-23, The Natural Philosopher <tnp@invalid.invalid> wrote:
>
>> On 23/09/2021 16:32, SevenOverSix wrote:
>>
>>> MP3 isn't THAT awful - think of how much music is encoded MP3.
>>>   In this case it just has to nail a couple of mid-range audio
>>>   tones, the 'warble' of old-tyme acoustic modems.
>>>
>>>   I'll give it a try and see. Meanwhile, I think you can still
>>>   buy brand-new compact cassette recorders. There are still a few
>>>   market niches.
>>
>> Ive got a barely used ttwin deck hifi cassette recorder going back to
>> the 1980s that hasn't been used in at least 30 years...heads in great
>> condition. Only the rubber bands might need replacing :-)
>>
>> Offers on a postcard.,..
>>
>> I think even reel-2-reel tape has made something
>>>   of a come-back lately, just like vinyl. R2R, run at a decent
>>>   speed, was GOOD. Most commercial music up until near the 90s
>>>   was still made on magtape, sometimes physically cut-n-pasted
>>>   to achieve the final result.
>>
>> 15 or 30 ips on 1/2" tape was as good as you need.
>>
>> 1 15/16" ips and 1/8" tape - not so much...
>
> 1 7/8 ips, actually. But even a consumer-grade 1/4-inch 4-track
> reel-to-reel deck running at 7 1/2 ips has an inherent 9dB advantage
> in signal-to-noise ratio over a cassette - not to mention much better
> high-frequency response. That's why I stuck with reel-to-reel when
> the world went to cassettes.
I remember an electronics magazine describing a method to use video
tapes with computers to achieve greater data speed and capacity.

--
Cheers,
Carlos E.R.

On 23/09/2021 23.29, Eli the Bearded wrote:

...

>>> Vinyl also has the quirk that outer edge tracks have better quality than
>>> center tracks. Most other media doesn't suffer that restriction.
>> Sounds like the different zones on a hard disk...
>
> Back to vintage computer reminiscence: I can remember when "where to put
> your swap partition?" was one of those zones discussion points.

Once or twice, I created a bunch of partitions (say a hundred) and
measured the speed on each of them. Turned out that the disk was faster
at about 1/3 of the way.

--
Cheers,
Carlos E.R.

On 23/09/2021 19.26, Rich wrote:
> SevenOverSix <hae274c.net> wrote:
>> On 9/23/21 6:17 AM, Carlos E. R. wrote:
>>> It is not compression that matters, but that mp3 is lossy.
>>> Commercial software exploited techniques on the fringe of what an
>>> audio tape recorder could do.
>>
>> MP3 isn't THAT awful - think of how much music is encoded MP3. In
>> this case it just has to nail a couple of mid-range audio tones,
>> the 'warble' of old-tyme acoustic modems.
>
> In case MP3 does something bad with the frequency bands the interface
> relies upon, there is always Flac https://en.wikipedia.org/wiki/Flac
> which is lossless.

Certainly.

--
Cheers,
Carlos E.R.

On 9/23/21 2:31 PM, Charlie Gibbs wrote:
> On 2021-09-23, The Natural Philosopher <tnp@invalid.invalid> wrote:
>
>> On 23/09/2021 16:32, SevenOverSix wrote:
>>
>>> MP3 isn't THAT awful - think of how much music is encoded MP3.
>>>   In this case it just has to nail a couple of mid-range audio
>>>   tones, the 'warble' of old-tyme acoustic modems.
>>>
>>>   I'll give it a try and see. Meanwhile, I think you can still
>>>   buy brand-new compact cassette recorders. There are still a few
>>>   market niches.
>>
>> Ive got a barely used ttwin deck hifi cassette recorder going back to
>> the 1980s that hasn't been used in at least 30 years...heads in great
>> condition. Only the rubber bands might need replacing :-)
>>
>> Offers on a postcard.,..
>>
>> I think even reel-2-reel tape has made something
>>>   of a come-back lately, just like vinyl. R2R, run at a decent
>>>   speed, was GOOD. Most commercial music up until near the 90s
>>>   was still made on magtape, sometimes physically cut-n-pasted
>>>   to achieve the final result.
>>
>> 15 or 30 ips on 1/2" tape was as good as you need.
>>
>> 1 15/16" ips and 1/8" tape - not so much...
>
> 1 7/8 ips, actually. But even a consumer-grade 1/4-inch 4-track
> reel-to-reel deck running at 7 1/2 ips has an inherent 9dB advantage
> in signal-to-noise ratio over a cassette - not to mention muchbetter
> high-frequency response. That's why I stuck with reel-to-reel when
> the world went to cassettes.

Cassettes were the MP3s of the era ... "good enough" for
the average consumer. They didn't (have to) sound THAT
bad. Many were used in car audio systems (the car I have
actually HAS a cassette slot).

But 1/4 or especially 1/2 inch - VASTLY better.

>> Vinyl has made a comeback within the touchy feely ArtStudent brigade
>> who have no idea what hi fidelity actually meant.
>
> If it rumbles it must be good, right? At least that's the message
> I get from that guy in the boom car somewhere within 100 yards of me
> in a traffic jam.

Most 'modern' music was never made with "high fidelity" in
mind. It's meant to be played LOUD, usually on a crap system
with 1000w bargain-basement class-D amps.

>> Years ago I bought a vinyl copy of Electric Ladyland. I was always
>> unhappy about distortion on one track. Eventually I bought the CD. To
>> my surprise, the distortion was on the master tape, and not the needle
>> chattering in the groove.
>
> Wow.
>
>> A good CD will give you 0-19khz with 90dB signal to noise and no actual
>> distortion beyond digitisation noise.
>>
>> You would be lucky to do better than 65dB signal to noise on vinyl and
>> 0.5% distortion.

Note that "sound" and "hearing" are not exactly the same things.
An absolutely "perfect" system might not SOUND so good. The ear
and brain have to be pleased. Certain harmonics and frequency
biases on vinyl "sound better". The old transistor amps were
technically "more perfect" than tubes/valves - but they sounded
harsh, brash. Odd harmonics -vs- even harmonics from the vac tubes,
emphasis in the wrong places, gaps in the wrong places.

So now people blow thousands, sometimes tens of thousands, on
vac tube amps/pre-amps and vinyl. Try :
https://www.theabsolutesound.com/
to see what the "budget unlimited" crowd craves.

>> Good analogue tape at 15ips can net you something up around 80dB or
>> better. Which is getting near better than the microphones. Of course
>> digital tape first and then digital disk recording has made the whole
>> recording process limited only by the microphones and their pre-amps
>>
>> Sadly as the recording quality has improved immeasurably, the recorded
>> material and sound engineer quality has deteriorated to utter drivel
>>
>> Many of the best classical orchestral recordings were done with a simple
>> crossed pair of stereo mics. Many hits of the 50s and 60s were done on 4
>> or 8 track recorders
>>
>> 128 track digitally mastered 'baby shark' just doesn't cut it. :-)
>
> I've read that CDs - including remixes of good vinyl recordings -
> are often compressed down to something like a 12dB dynamic range.

That'd be an AWFUL transfer. OK for boom-boom-boom I guess.

Now I have a CD, R.E.M., where one track opens with the sound
of someone using an old manual typewriter. With a decent amp
and magneplanar speakers (inefficient but utterly 'transparent')
I was almost shocked by it - sounded REALLY RIGHT IN THE ROOM.
The rest of the CD was top quality too - you could just hear
the bassists fingers sliding a bit on the strings, WAY down
the decibel ladder.

So CDs CAN be good. Thing is, almost nobody bothers. MP3 can
be good too - but you have to use VERY little compression
and crank the sample rate up a bit. There have been some
makers trying to push "super-CD" with a much higher sample
rate and A/D depth. Never got traction though. In theory
a DVD, properly purposed, could easily fit in the double
or triple-sized audio files.

> I got my hands on an MP3 of a wonderful Joni a track that
> I cherish on vinyl. It sounded so horrible that I deleted it -
> and I don't delete _anything_.
>
> https://en.wikipedia.org/wiki/Loudness_war
>
> That's not to say that there weren't variations even on vinyl.
> The labels in the WEA triumvirate (Warner/Elektra/Atlantic)
> were arranged in decreasing order of recording quality.
> If you turned up the volume on a Warner recording, the sound
> would come out of the speakers and envelop you. If you turned
> up the volume on an Atlantic recording, it just got loud.

That's the sign of 'audiophile' quality - increasing the
volume simply makes it louder ... instead of adding that
distorted edge. But even that's not ALWAYS good. Got
some "re-mastered" Led Zep stuff. TERRIBLE sound. They
utterly ruined it. Thin bass and no distortions. Back
in the day we'd crank it up EXPECTING, WANTING, that
distortion. That's what it was SUPPOSED to sound like.
When The Levee Breaks is SUPPOSED to shake brain cells.

I may use Audacity "re-ruin" those discs :-)

On 9/23/21 5:29 PM, Eli the Bearded wrote:
> In comp.os.linux.misc, Charlie Gibbs <cgibbs@kltpzyxm.invalid> wrote:
>> However, a friend has a car that offers a great work-around -
>> switchable compression in the stereo system. That way you can
>> hear everything on the road, but if you're parked somewhere and
>> have no background noise to worry about, you can switch off the
>> compression and hear the recording in all its glory.
>
> That sort of feature is what I meant by "better car stereos". I'm not an
> audio guy and will be quite imprecise on such matters.
>
> (Quoting me)
>
>>> Vinyl also has the quirk that outer edge tracks have better quality than
>>> center tracks. Most other media doesn't suffer that restriction.
>> Sounds like the different zones on a hard disk...
>
> Back to vintage computer reminiscence: I can remember when "where to put
> your swap partition?" was one of those zones discussion points.

Disks read from the inside out.

In any case, your swap really should be right next
to your system partition so minimal head movement
is required. Most people put them at the very end
of the disk, which is wrong unless your system
partition is next-2-last as well.

SDD's, well, doesn't matter - but SDDs just do not
have the dollar/capacity or the rewrite endurance of
magnetic HDDs yet. Not something you should use in
a busy database server that's constantly indexing
indexing indexing. Magnetics will still have a solid
niche a decade from now, though the price will creep
up as more "home" units switch to SDDs. Had good luck
with WD-Golds (basically re-labeled HGSTs) so maybe
I'll buy a few 6tbs before that price creep starts.

The old big video-disks came in two flavors - Constant
Linear Velocity and Constant Angular Velocity. Linear
meant the disc actually slowed down as you read towards
the outer edge - maintained the bits/sec ratio. Angular
spread the pits on the disk further apart as you got to
the edge - again preserving the bits/sec but wasting
surface area with progressively less-packed pits. Angular
was more like a sectored HDD, easier to random-access
and the platter speed never changed. HDDs, thanks to
the newer electronics, can pack the bits in better -
the heads are under-challenged on the inner tracks and
at the bleeding edge on the outer tracks.

Still have one of those machines and about two dozen disks.
Not exactly 4k quality, but for the day they were great.

Path:
buffer2.nntp.dca1.giganews.com!border1.nntp.dca1.giganews.com!nntp.giganews.com!news.misty.com!2.eu.feeder.erje.net!feeder.erje.net!news.swapon.de!fu-berlin.de!uni-berlin.de!individual.net!not-for-mail
From: Carlos E. R. <robin_listas@es.invalid>
Newsgroups: comp.os.linux.misc
Subject: Re: Obit - Sir Clive Sinclair, Computing Pioneer
Date: Fri, 24 Sep 2021 00:30:50 +0200
Lines: 19
Message-ID: <qpbv1ixkgr.ln2@minas-tirith.valinor>
References: <POCdnc09ErazA9n8nZ2dnUU7-aHNnZ2d@earthlink.com>
<9K33J.43934$md6.4877@fx36.iad> <eli$2109231506@qaz.wtf>
<gW53J.442$2m1.423@fx26.iad> <eli$2109231728@qaz.wtf>
Mime-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
X-Trace: individual.net W5TCw2htbaqSSYHdovTbgAhPy0AUUZBiSAl/8vTR+HZwjzGWNS
X-Orig-Path: minas-tirith.valinor!not-for-mail
Cancel-Lock: sha1:gSn1+PkbZK/kYmvgF3Z1w7zXlFY=
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:78.0) Gecko/20100101
Thunderbird/78.12.0
In-Reply-To: <eli$2109231728@qaz.wtf>
Content-Language: en-GB
Bytes: 1673
Xref: number.nntp.giganews.com comp.os.linux.misc:773290

On 23/09/2021 23.29, Eli the Bearded wrote:

...

>>> Vinyl also has the quirk that outer edge tracks have better quality
than
>>> center tracks. Most other media doesn't suffer that restriction.
>> Sounds like the different zones on a hard disk...
>
> Back to vintage computer reminiscence: I can remember when "where to put
> your swap partition?" was one of those zones discussion points.

Once or twice, I created a bunch of partitions (say a hundred) and
measured the speed on each of them. Turned out that the disk was faster
at about 1/3 of the way.

Yep.

Best compromise of data density and head-movement speed.
Takes a lot of work to fling the heads from one end of
the drive to the other - half of it wasted decellerating
and stabilizing the things as they near the outer edge.

Remember when disk drives used more mechanical means,
sometimes linear actuators or even worm screws and stepper
motors, to move the heads ? Few today have seen the old
removable-platter units, about the size of a dish washer,
where the arms (usually square tube-in-tube) moved straight
in and out. Took the better part of a second to move from
inner to outer tracks - and then there was some fine-tuning
to center perfectly over the desired track. Most could
move the heads independently, so there was an art to
placing your data on the platters so one head could pick
up for the other on the next platter and there'd be
minimal delay and head movement. Didn't always work out
in a multi-user/multi-tasking situation though. STILL
a lot faster than mag-tape reels. You'd store your
programs and bulk data on those :-)

So if you've got a dual+ boot setup, put your crappy Winders
stuff at the bottom and your Linux in the middle. Put your
swap partition right next to your system partition.

SDDs, irrelevant - but they cost WAY too much per TB and
have miserable re-write endurance at present. Great for
"home PCs/laptops" but NOT for hard-working servers.
Some DO set up their systems on a SDD and then go through
all the trouble to shift the data and such off to magnetics.
You can also RAID 1/5/6 SDDs so WHEN they burn out you can swap
in new ones without losing anything (if you don't wait too long).
Of course whatever your box is, NETWORK speed is the bottleneck
in getting said data out to the users. Even 10/g can be
"slow" in some circumstances.

Kept waiting for high-density ferroelectric memory chips, but it
never happened ... I think 512K are the biggest you can get.
A terabyte - a case bigger than a Flintstones lunch box
for sure and a LOT of wiring inside. Plus side, about four
or five times the speed of current SDDs and essentially
infinite rewrites. Static RAM also exists, with roughly
the same plusses and minuses.

On 24/09/2021 06.16, SevenOverSix wrote:
>

....

>   So if you've got a dual+ boot setup, put your crappy Winders
>   stuff at the bottom and your Linux in the middle. Put your
>   swap partition right next to your system partition.
>
>   SDDs, irrelevant - but they cost WAY too much per TB and
>   have miserable re-write endurance at present. Great for
>   "home PCs/laptops" but NOT for hard-working servers.
>   Some DO set up their systems on a SDD and then go through
>   all the trouble to shift the data and such off to magnetics.
>   You can also RAID 1/5/6 SDDs so WHEN they burn out you can swap
>   in new ones without losing anything (if you don't wait too long).
>   Of course whatever your box is, NETWORK speed is the bottleneck
>   in getting said data out to the users. Even 10/g can be
>   "slow" in some circumstances.

No, SSD are quite durable nowdays (of course, with limits), and they are
indeed used in servers, when speed is important.

For example, crypto coin mining use them a lot.

--
Cheers,
Carlos E.R.

On Fri, 24 Sep 2021 00:21:01 +0200
"Carlos E. R." <robin_listas@es.invalid> wrote:

> On 23/09/2021 21.55, Peter Flass wrote:
> > Charlie Gibbs <cgibbs@kltpzyxm.invalid> wrote:
> >> On 2021-09-23, Rich <rich@example.invalid> wrote:
> >>
> >>> In comp.os.linux.misc Peter Flass <peter_flass@yahoo.com> wrote:
> >>>
> >>>> Carlos E. R. <robin_listas@es.invalid> wrote:
> >>>>
> >>>>> On 22/09/2021 21.57, Andreas Kohlbach wrote:
> >>>>>
> >>>>>> On Wed, 22 Sep 2021 09:13:12 +0200, Carlos E. R. wrote:
> >>>>>>
> >>>>>>> Might not work (a capture). Just a guess. If you do try,
> >>>>>>> make sure to not use mp3. Now that I think, I would try the
> >>>>>>> experiment, to find out.
> >>>>>>
> >>>>>> There was a short lived [1] UK computer show "4 Computer
> >>>>>> Buffs" <https://www.imdb.com/title/tt1178499/> I never heard
> >>>>>> of before. They sent program data via audio the audience
> >>>>>> record. When I found the show on Youtube I tested that
> >>>>>> (extracted the audio at that position) and ran the resulting
> >>>>>> WAV in an emulator fir that particular machine on my PC. It
> >>>>>> my amazement it worked.
> >>>>>
> >>>>> Wow.
> >>>>
> >>>> Systems that saved programs on cassette used an audio format.
> >>>> Due to the sloppiness of the media, I think the recording format
> >>>> had to be pretty robust.
> >>>
> >>> At least in the case of the Atari 400/800 cassette format it was
> >>> a very simple format:
> >>>
> >>> Format details are here:
> >>> https://www.atariarchives.org/dere/chaptC.php
> >>>
> >>> 132 byte records, two start bytes for 'speed detection', a
> >>> control byte, 128 data bytes, and a single checksum byte (and the
> >>> checksum is just a simple endaround carry sum of the 131 other
> >>> bytes in the record).
> >>>
> >>> The physical byte encoding on the tape was frequency shift
> >>> keying, with 5327 Hz for a mark and 3995 Hz for a space.
> >>>
> >>> So it at least it had a simple checksum, but the packet format
> >>> was hardly "robust". Workable, but memory of those days was that
> >>> the cassette was quite flakey as a data storage format, sometimes
> >>> it worked, sometimes it did not. And when it did not rereading
> >>> things all over again sometimes magically had them work.
> >>
> >> Ah yes, I remember the good old days with my IMSAI. I didn't have
> >> cassette decks, but I had a couple of reel-to-reel decks, so I
> >> broke into their motor circuits and built a control box that would
> >> use the cassette motor control circuits to activate relays to
> >> switch 110-volt motor power on and off.
> >
> > I was astonished when I got my first home computer with a cassette
> > drive that it didn’t do this!
>
> I have a foggy memory that it used the switch on/off wires of the
> microphone, which in some/all tape machines stopped the motor.
>
> That would be the Sinclair Spectrum if any, but can't vouch for it.

Drifting; the Amstrad early version (z80 based) CPC464 had a builtin cassette recorder.

--
Bah, and indeed Humbug.

I had CPC6128 2x64kb and FLOPPY! z80 CPU alright and CP/M OS.
Worked in bakershop in England to save money ti buy it.
Learned to program on z80.

--
7-77-777
\|/
---
/|\

On 2021-09-24, Kerr-Mudd, John <admin@127.0.0.1> wrote:
> On Fri, 24 Sep 2021 00:21:01 +0200
> "Carlos E. R." <robin_listas@es.invalid> wrote:
>
>> On 23/09/2021 21.55, Peter Flass wrote:
>> > Charlie Gibbs <cgibbs@kltpzyxm.invalid> wrote:
>> >> On 2021-09-23, Rich <rich@example.invalid> wrote:
>> >>
>> >>> In comp.os.linux.misc Peter Flass <peter_flass@yahoo.com> wrote:
>> >>>
>> >>>> Carlos E. R. <robin_listas@es.invalid> wrote:
>> >>>>
>> >>>>> On 22/09/2021 21.57, Andreas Kohlbach wrote:
>> >>>>>
>> >>>>>> On Wed, 22 Sep 2021 09:13:12 +0200, Carlos E. R. wrote:
>> >>>>>>
>> >>>>>>> Might not work (a capture). Just a guess. If you do try,
>> >>>>>>> make sure to not use mp3. Now that I think, I would try the
>> >>>>>>> experiment, to find out.
>> >>>>>>
>> >>>>>> There was a short lived [1] UK computer show "4 Computer
>> >>>>>> Buffs" <https://www.imdb.com/title/tt1178499/> I never heard
>> >>>>>> of before. They sent program data via audio the audience
>> >>>>>> record. When I found the show on Youtube I tested that
>> >>>>>> (extracted the audio at that position) and ran the resulting
>> >>>>>> WAV in an emulator fir that particular machine on my PC. It
>> >>>>>> my amazement it worked.
>> >>>>>
>> >>>>> Wow.
>> >>>>
>> >>>> Systems that saved programs on cassette used an audio format.
>> >>>> Due to the sloppiness of the media, I think the recording format
>> >>>> had to be pretty robust.
>> >>>
>> >>> At least in the case of the Atari 400/800 cassette format it was
>> >>> a very simple format:
>> >>>
>> >>> Format details are here:
>> >>> https://www.atariarchives.org/dere/chaptC.php
>> >>>
>> >>> 132 byte records, two start bytes for 'speed detection', a
>> >>> control byte, 128 data bytes, and a single checksum byte (and the
>> >>> checksum is just a simple endaround carry sum of the 131 other
>> >>> bytes in the record).
>> >>>
>> >>> The physical byte encoding on the tape was frequency shift
>> >>> keying, with 5327 Hz for a mark and 3995 Hz for a space.
>> >>>
>> >>> So it at least it had a simple checksum, but the packet format
>> >>> was hardly "robust". Workable, but memory of those days was that
>> >>> the cassette was quite flakey as a data storage format, sometimes
>> >>> it worked, sometimes it did not. And when it did not rereading
>> >>> things all over again sometimes magically had them work.
>> >>
>> >> Ah yes, I remember the good old days with my IMSAI. I didn't have
>> >> cassette decks, but I had a couple of reel-to-reel decks, so I
>> >> broke into their motor circuits and built a control box that would
>> >> use the cassette motor control circuits to activate relays to
>> >> switch 110-volt motor power on and off.
>> >
>> > I was astonished when I got my first home computer with a cassette
>> > drive that it didn’t do this!
>>
>> I have a foggy memory that it used the switch on/off wires of the
>> microphone, which in some/all tape machines stopped the motor.
>>
>> That would be the Sinclair Spectrum if any, but can't vouch for it.
>
> Drifting; the Amstrad early version (z80 based) CPC464 had a builtin cassette recorder.
>

--
Evil Sinner!

On Fri, 24 Sep 2021 11:47:53 GMT
Branimir Maksimovic <branimir.maksimovic@gmail.com> wrote:

> I had CPC6128 2x64kb and FLOPPY! z80 CPU alright and CP/M OS.
> Worked in bakershop in England to save money ti buy it.
> Learned to program on z80.
>
> --
> 7-77-777
> \|/
> ---
> /|\
>
Please adopt usenet convention and post your reply text at the bottom, there's a good chap. And some judicious snipping would help too. (Yup I didn't do it last time, mea culpa).

> On 2021-09-24, Kerr-Mudd, John <admin@127.0.0.1> wrote:
> > On Fri, 24 Sep 2021 00:21:01 +0200
[]
> >
> > Drifting; the Amstrad early version (z80 based) CPC464 had a
> > builtin cassette recorder.
> >
>
>
> --
> Evil Sinner!

--
Bah, and indeed Humbug.

Ok, I'll do that when you learn how to format text properly.
Top posting is what emphaises is on, and, if lot of text, they
will not even read what is written.
YES, CPC464 was very popular, and breaking SPEEDLOCK protections
as excercize :P
--
7-77-777
\|/
/|\

On 2021-09-24, Kerr-Mudd, John <admin@127.0.0.1> wrote:
> On Fri, 24 Sep 2021 11:47:53 GMT Branimir Maksimovic
> <branimir.maksimovic@gmail.com> wrote:
>
>> I had CPC6128 2x64kb and FLOPPY! z80 CPU alright and CP/M OS. Worked in
>> bakershop in England to save money ti buy it. Learned to program on z80.
>>
>> -- 7-77-777 \|/ --- /|\
>>
> Please adopt usenet convention and post your reply text at the bottom,
> there's a good chap. And some judicious snipping would help too. (Yup I
> didn't do it last time, mea culpa).
>
>
>
>> On 2021-09-24, Kerr-Mudd, John <admin@127.0.0.1> wrote:
>> > On Fri, 24 Sep 2021 00:21:01 +0200
> []
>> >
>> > Drifting; the Amstrad early version (z80 based) CPC464 had a builtin
>> > cassette recorder.
>> >
>>
>>
>> -- Evil Sinner!
>
>

--
Evil Sinner!

Same here. I so much hate having to scroll down to read what I want to
read, that could be just there for me to see.

On 24.09.21 15:05, Branimir Maksimovic wrote:
> Ok, I'll do that when you learn how to format text properly.
> Top posting is what emphaises is on, and, if lot of text, they
> will not even read what is written.
> YES, CPC464 was very popular, and breaking SPEEDLOCK protections
> as excercize :P
> --
> 7-77-777
> \|/
> /|\
>
> On 2021-09-24, Kerr-Mudd, John <admin@127.0.0.1> wrote:
>> On Fri, 24 Sep 2021 11:47:53 GMT Branimir Maksimovic
>> <branimir.maksimovic@gmail.com> wrote:
>>
>>> I had CPC6128 2x64kb and FLOPPY! z80 CPU alright and CP/M OS. Worked in
>>> bakershop in England to save money ti buy it. Learned to program on z80.
>>>
>>> -- 7-77-777 \|/ --- /|\
>>>
>> Please adopt usenet convention and post your reply text at the bottom,
>> there's a good chap. And some judicious snipping would help too. (Yup I
>> didn't do it last time, mea culpa).
>>
>>
>>
>>> On 2021-09-24, Kerr-Mudd, John <admin@127.0.0.1> wrote:
>>>> On Fri, 24 Sep 2021 00:21:01 +0200
>> []
>>>>
>>>> Drifting; the Amstrad early version (z80 based) CPC464 had a builtin
>>>> cassette recorder.
>>>>
>>>
>>>
>>> -- Evil Sinner!
>>
>>
>
>

On 2021-09-24, Branimir Maksimovic <branimir.maksimovic@gmail.com> wrote:

> Ok, I'll do that when you learn how to format text properly.

Or when you learn to spell "OK" properly... :-)

> Top posting is what emphaises is on, and, if lot of text, they
> will not even read what is written.

That only happens when people don't trim quoted text appropriately.

If you can't be bothered taking the time to make your message
easy to read, I can't be bothered taking the time to decipher it.

Just because Outlook[1] vict^H^H^H^Husers succumb to its pressure to
top-post doesn't make it a Good Thing.

[1] Properly pronounced "Look out!"

--
/~\ Charlie Gibbs | A: It messes up the flow of the thread.
\ / <cgibbs@kltpzyxm.invalid> | Q: Why is top-posting bad?
X I'm really at ac.dekanfrus | A: Top-posting.
/ \ if you read it the right way. | Q: What is an impediment to readability?

On 2021-09-23, Carlos E. R. <robin_listas@es.invalid> wrote:

> I remember an electronics magazine describing a method to use video
> tapes with computers to achieve greater data speed and capacity.

I read an article about a system from Ampex called TBM, for
"terabit memory". It used a bank of those huge VTRs that
used 2-inch tape, which were the mainstay of TV studios in
the '60s and '70s. I don't think it caught on...

--
/~\ Charlie Gibbs | Life is perverse.
\ / <cgibbs@kltpzyxm.invalid> | It can be beautiful -
X I'm really at ac.dekanfrus | but it won't.
/ \ if you read it the right way. | -- Lily Tomlin

On 2021-09-24, SevenOverSix <hae274c.net> wrote:

> Remember when disk drives used more mechanical means,
> sometimes linear actuators or even worm screws and stepper
> motors, to move the heads ? Few today have seen the old
> removable-platter units, about the size of a dish washer,
> where the arms (usually square tube-in-tube) moved straight
> in and out. Took the better part of a second to move from
> inner to outer tracks - and then there was some fine-tuning
> to center perfectly over the desired track.

The first such drives I worked with were Univac's clones of
the IBM 2311 and 2314. They had hydraulic actuators; there
was a felt-lined drip pad under the actuator.

One time we left the air conditioning on over a weekend
when the machine was shut down, and when we came back in
our breath was condensing in the machine room. The oil
in the actuators had congealed; we had to let the disks
spin for half an hour or so before it warmed up enough
to let the heads load.

> Most could
> move the heads independently, so there was an art to
> placing your data on the platters so one head could pick
> up for the other on the next platter and there'd be
> minimal delay and head movement. Didn't always work out
> in a multi-user/multi-tasking situation though. STILL
> a lot faster than mag-tape reels. You'd store your
> programs and bulk data on those :-)

I never saw a drive with multiple actuators, although I
wouldn't be surprised if such a thing existed. I did hear
about drives that had fixed heads mounted on a few tracks
for rapid access.

Some systems allowed split-cylinder allocation; if you
did it right you could access two files without moving
the head assembly. It would take so much care in setting
up properly, though, that I never saw it actually used.

--
/~\ Charlie Gibbs | Life is perverse.
\ / <cgibbs@kltpzyxm.invalid> | It can be beautiful -
X I'm really at ac.dekanfrus | but it won't.
/ \ if you read it the right way. | -- Lily Tomlin

On 2021-09-23, Carlos E. R. <robin_listas@es.invalid> wrote:

> On 23/09/2021 21.06, Eli the Bearded wrote:
>
>> In comp.os.linux.misc, Charlie Gibbs <cgibbs@kltpzyxm.invalid> wrote:
>
> ..
>
>>> I got my hands on an MP3 of a wonderful Joni Mitchell track that
>>> I cherish on vinyl. It sounded so horrible that I deleted it -
>>> and I don't delete _anything_.
>>
>> Vinyl also has the quirk that outer edge tracks have better quality than
>> center tracks. Most other media doesn't suffer that restriction.
>>
>> Elijah
>> ------
>> most other media doesn't let you drill an off-center hole for fun
>
> Ow :-D

No, wow. :-)

--
/~\ Charlie Gibbs | Life is perverse.
\ / <cgibbs@kltpzyxm.invalid> | It can be beautiful -
X I'm really at ac.dekanfrus | but it won't.
/ \ if you read it the right way. | -- Lily Tomlin

On Fri, 24 Sep 2021 13:05:37 GMT, Branimir Maksimovic wrote:
>
> Ok, I'll do that when you learn how to format text properly.
> Top posting is what emphaises is on, and, if lot of text, they
> will not even read what is written.
> YES, CPC464 was very popular, and breaking SPEEDLOCK protections
> as excercize :P

It's not only to (please) avoid top posting, regarding usenet conventions
existing since the 1980s. It's also to trim unnecessary quotes. Like I
just did. Isn't that easier to read?
--
Andreas

On Fri, 24 Sep 2021 17:16:55 GMT, Charlie Gibbs wrote:
>
> On 2021-09-24, Branimir Maksimovic <branimir.maksimovic@gmail.com> wrote:
>
>> Ok, I'll do that when you learn how to format text properly.
>
> Or when you learn to spell "OK" properly... :-)

?

>> Top posting is what emphaises is on, and, if lot of text, they
>> will not even read what is written.
>
> That only happens when people don't trim quoted text appropriately.
>
> If you can't be bothered taking the time to make your message
> easy to read, I can't be bothered taking the time to decipher it.
>
> Just because Outlook[1] vict^H^H^H^Husers succumb to its pressure to
> top-post doesn't make it a Good Thing.

Outlook (Express) can be configured to put the cursor below the
quote. Default is top posting thought. One of the countless annoying
"features" of it. Am amazed, that people today still use Outlook.

F'up2 comp.os.linux.misc, although it's off topic in either group.
--
Andreas

On 24/09/2021 in message <87wnn57siz.fsf@usenet.ankman.de> Andreas
Kohlbach wrote:

>Outlook (Express) can be configured to put the cursor below the
>quote. Default is top posting thought. One of the countless annoying
>"features" of it. Am amazed, that people today still use Outlook.

The cursor is placed at the top so you can scroll down and trim before
replying.

--
Jeff Gaines Wiltshire UK
The facts, although interesting, are irrelevant

On 2021-09-24, Andreas Kohlbach <ank@spamfence.net> wrote:

> On Fri, 24 Sep 2021 17:16:55 GMT, Charlie Gibbs wrote:
>
>> On 2021-09-24, Branimir Maksimovic <branimir.maksimovic@gmail.com> wrote:
>>
>>> Ok, I'll do that when you learn how to format text properly.
>>
>> Or when you learn to spell "OK" properly... :-)
>
> ?

Both letters should be capitalized. That way you ensure that
it's pronounced "oh kay". It's not a word, but a concatenation
of letters, like "TCP/IP". If I see someone writing it "Ok",
I make sure to pronounce it "awk" just to piss them off.

Going back to the obligatory Microsoft-bashing, the first time
I saw "Ok" was the prompt from their BASIC interpreter for CP/M,
so I assign them the blame. It was carried over into GW-BASIC
on MS-DOS, which I still use from time to time; to keep my
blood pressure down I patched the .EXE file to display "OK".

--
/~\ Charlie Gibbs | Life is perverse.
\ / <cgibbs@kltpzyxm.invalid> | It can be beautiful -
X I'm really at ac.dekanfrus | but it won't.
/ \ if you read it the right way. | -- Lily Tomlin