22 Jan
2021
22 Jan
'21
3:10 p.m.
The Raspberry Pi that I have been using for at least 10 years to run the
hecnet bridge has finally died so I need to setup a new one.
In the meantime my area will be offline as far as hecnet is concerned.
Regards, Mark.
--
https://www.qrz.com/db/M0NOM/P
22 Jan
22 Jan
4:17 p.m.
On Jan 22, 2021, at 9:10 AM, Mark Wickens <mark at
wickensonline.co.uk> wrote:
The Raspberry Pi that I have been using for at least 10 years to run the hecnet bridge
has finally died so I need to setup a new one.
The entire Pi died and not just the microSD?
In the meantime my area will be offline as far as hecnet is concerned.
Regards, Mark.
--
https://www.qrz.com/db/M0NOM/P
5:41 p.m.
Yeah, I would be interested in hearing about the failure mode, too if
you know it, Mark.? I was interested in getting one of these to handle
some odd things that need getting done, probably the new 8 GB model.
10 years used to be nearly unheard of for retail machines with moving
parts.? When nothing is moving, then supposedly there is nothing to burn
out.? I have a laptop from 2000 that I replaced the disk with an SSD and
it still boots fine (great for regression testing.? Another two date
from before 2010 and have SSD's; no problems there, either.
------------------------------------------------------------------------
On 1/22/21 10:17 AM, Supratim Sanyal wrote:
The entire Pi died and not just the microSD?
> ------------------------------------------------------------------------
> On Jan 22, 2021, at 9:10 AM, Mark Wickens <mark at wickensonline.co.uk
> <mailto:mark at wickensonline.co.uk>> wrote:
>
> The Raspberry Pi that I have been using for at least 10 years to run
> the hecnet bridge has finally died so I need to setup a new one.
>
> In the meantime my area will be offline as far as hecnet is concerned.
>
> Regards, Mark.
>
> --
> https://www.qrz.com/db/M0NOM/P <https://www.qrz.com/db/M0NOM/P>
>
5:53 p.m.
On 1/22/21 11:41 AM, Thomas DeBellis wrote:
10 years used to be nearly unheard of for retail
machines with moving
parts.? When nothing is moving, then supposedly there is nothing to burn
out.
I don't know who told you that, but he/she knows jack point squat
about electronics.
Materials migration and diffusion across junctions causes
semiconductor components to fail, tin whiskers cause shorts, some types
of capacitors dry out and/or have their electrolyte deteriorate or
crystallize, resistors drift, heat/cool cycles cause PCB flexure
resulting in cracked solder joints, corrosion in air creeps into
connector pin interfaces and forces pins apart, the list goes on and on
and on.
To be fair, some of the above-listed failure modes do in fact involve
things moving, though imperceptibly so, my point stands.
-Dave
--
Dave McGuire, AK4HZ
New Kensington, PA
6:03 p.m.
Well, I know of PDP-11 systems out in production who have been running
since the early 80s. The only thing that fails is the damn RD53...
Replaced those about 20 years ago now. The rest of the machine was
running just fine. At that at a steel mill, so not exactly a friendly
environment. Last I heard, things were still running, but it might have
been replaced by now. But anyway, it's the moving parts that cause
problems. The rest usually just keep running...
Johnny
On 2021-01-22 17:53, Dave McGuire wrote:
On 1/22/21 11:41 AM, Thomas DeBellis wrote:
--
Johnny Billquist || "I'm on a bus
|| on a psychedelic trip
email: bqt at softjar.se || Reading murder books
pdp is alive! || tryin' to stay hip" - B. Idol
10 years used to be nearly unheard of for retail
machines with moving
parts.? When nothing is moving, then supposedly there is nothing to
burn out.
? I don't know who told you that, but he/she knows jack point squat
about electronics.
? Materials migration and diffusion across junctions causes
semiconductor components to fail, tin whiskers cause shorts, some types
of capacitors dry out and/or have their electrolyte deteriorate or
crystallize, resistors drift, heat/cool cycles cause PCB flexure
resulting in cracked solder joints, corrosion in air creeps into
connector pin interfaces and forces pins apart, the list goes on and on
and on.
? To be fair, some of the above-listed failure modes do in fact involve
things moving, though imperceptibly so, my point stands.
????????????? -Dave
6:15 p.m.
Sigh, yes, we all know of exceptions. But even then, I doubt it can
be proven that those PDP-11 systems in production since the early 80s
have never had components replaced.
Not everything is built to DEC standards. That said, all my life
I've been replacing chips, diodes, transistors, capacitors, and
sometimes even resistors in DEC gear. It does happen, and it's not
uncommon.
The regulator bricks in the H742/H7420 power supplies tend to blow
Zener diodes. That's fresh in my mind because we just lost another one
at the museum recently. Those are failing with such frequency now that
I've purchased stocks of the components that tend to fail. (we have a
lot of those at the museum) And those ARE built to DEC standards.
Speaking of not being built to DEC standards, I've got a Heathkit H8
on the bench right now, I'm repairing it for a new exhibit at the
museum. Four failed (non-moving!) ICs replaced so far, it's very nearly
done.
-Dave
On 1/22/21 12:03 PM, Johnny Billquist wrote:
Well, I know of PDP-11 systems out in production who
have been running
since the early 80s. The only thing that fails is the damn RD53...
Replaced those about 20 years ago now. The rest of the machine was
running just fine. At that at a steel mill, so not exactly a friendly
environment. Last I heard, things were still running, but it might have
been replaced by now. But anyway, it's the moving parts that cause
problems. The rest usually just keep running...
? Johnny
On 2021-01-22 17:53, Dave McGuire wrote:
--
Dave McGuire, AK4HZ
New Kensington, PA
On 1/22/21 11:41 AM, Thomas DeBellis wrote:
10 years used to be nearly unheard of for retail
machines with moving
parts.? When nothing is moving, then supposedly there is nothing to
burn out.
?? I don't know who told you that, but he/she knows jack point squat
about electronics.
?? Materials migration and diffusion across junctions causes
semiconductor components to fail, tin whiskers cause shorts, some
types of capacitors dry out and/or have their electrolyte deteriorate
or crystallize, resistors drift, heat/cool cycles cause PCB flexure
resulting in cracked solder joints, corrosion in air creeps into
connector pin interfaces and forces pins apart, the list goes on and
on and on.
?? To be fair, some of the above-listed failure modes do in fact
involve things moving, though imperceptibly so, my point stands.
?????????????? -Dave
6:29 p.m.
Two words... RIFA Caps.? :-D
--
John H. Reinhardt
On 1/22/2021 11:15 AM, Dave McGuire wrote:
? Sigh, yes, we all know of exceptions.? But even then, I doubt it can be proven that
those PDP-11 systems in production since the early 80s have never had components
replaced.
? Not everything is built to DEC standards.? That said, all my life I've been
replacing chips, diodes, transistors, capacitors, and sometimes even resistors in DEC
gear.? It does happen, and it's not uncommon.
? The regulator bricks in the H742/H7420 power supplies tend to blow Zener diodes.?
That's fresh in my mind because we just lost another one at the museum recently.?
Those are failing with such frequency now that I've purchased stocks of the components
that tend to fail. (we have a lot of those at the museum)? And those ARE built to DEC
standards.
? Speaking of not being built to DEC standards, I've got a Heathkit H8 on the bench
right now, I'm repairing it for a new exhibit at the museum.? Four failed
(non-moving!) ICs replaced so far, it's very nearly done.
??????????? -Dave
On 1/22/21 12:03 PM, Johnny Billquist wrote:
> Well, I know of PDP-11 systems out in production who have been running since the
early 80s. The only thing that fails is the damn RD53...
> Replaced those about 20 years ago now. The rest of the machine was running just fine.
At that at a steel mill, so not exactly a friendly environment. Last I heard, things were
still running, but it might have been replaced by now. But anyway, it's the moving
parts that cause problems. The rest usually just keep running...
>
> ?? Johnny
>
> On 2021-01-22 17:53, Dave McGuire wrote:
>> On 1/22/21 11:41 AM, Thomas DeBellis wrote:
>>> 10 years used to be nearly unheard of for retail machines with moving parts.?
When nothing is moving, then supposedly there is nothing to burn out.
>>
>> ?? I don't know who told you that, but he/she knows jack point squat about
electronics.
>>
>> ?? Materials migration and diffusion across junctions causes semiconductor
components to fail, tin whiskers cause shorts, some types of capacitors dry out and/or
have their electrolyte deteriorate or crystallize, resistors drift, heat/cool cycles cause
PCB flexure resulting in cracked solder joints, corrosion in air creeps into connector pin
interfaces and forces pins apart, the list goes on and on and on.
>>
>> ?? To be fair, some of the above-listed failure modes do in fact involve things
moving, though imperceptibly so, my point stands.
>>
>> ?????????????? -Dave
7:48 p.m.
On January 22, 2021 12:29:36 PM "John H. Reinhardt"
<johnhreinhardt at thereinhardts.org> wrote:
Two words... RIFA Caps. :-D
Here is a pic of my stock of safety capacitors, of which the primary
consumer is the museum:
http://www.neurotica.com/misc/safetycapacitors.jpg
FYI, the BA23 has three Rifa capacitors in its power supply. I've
replaced them in about ten BA23s thus far.
-Dave
--
Dave McGuire, AK4HZ
New Kensington, PA
6:31 p.m.
The KA10 built in 1966 had ONE broken memory chip (in the fast AC) after
being in storage for 30Y and traveled by sea across half of the world....
(and some bad contacts on two cards that had to be cleaned...)
-P
----- Original Message -----
From: "Dave McGuire" <mcguire at
neurotica.com>
To: "hecnet" <hecnet at Update.UU.SE>
Sent: Friday, January 22, 2021 12:15:20 PM
Subject: Re: [HECnet] No connectivity to arsgea 4
Sigh, yes, we all know of exceptions. But even then,
I doubt it can
be proven that those PDP-11 systems in production since the early 80s
have never had components replaced.
Not everything is built to DEC standards. That said, all my life
I've been replacing chips, diodes, transistors, capacitors, and
sometimes even resistors in DEC gear. It does happen, and it's not
uncommon.
The regulator bricks in the H742/H7420 power supplies tend to blow
Zener diodes. That's fresh in my mind because we just lost another one
at the museum recently. Those are failing with such frequency now that
I've purchased stocks of the components that tend to fail. (we have a
lot of those at the museum) And those ARE built to DEC standards.
Speaking of not being built to DEC standards, I've got a Heathkit H8
on the bench right now, I'm repairing it for a new exhibit at the
museum. Four failed (non-moving!) ICs replaced so far, it's very nearly
done.
-Dave
On 1/22/21 12:03 PM, Johnny Billquist wrote:
Well, I know of PDP-11 systems out in production
who have been running
since the early 80s. The only thing that fails is the damn RD53...
Replaced those about 20 years ago now. The rest of the machine was
running just fine. At that at a steel mill, so not exactly a friendly
environment. Last I heard, things were still running, but it might have
been replaced by now. But anyway, it's the moving parts that cause
problems. The rest usually just keep running...
? Johnny
On 2021-01-22 17:53, Dave McGuire wrote:
--
Dave McGuire, AK4HZ
New Kensington, PA On 1/22/21 11:41 AM, Thomas DeBellis wrote:
10 years used to be nearly unheard of for retail
machines with moving
parts.? When nothing is moving, then supposedly there is nothing to
burn out.
?? I don't know who told you that, but he/she knows jack point squat
about electronics.
?? Materials migration and diffusion across junctions causes
semiconductor components to fail, tin whiskers cause shorts, some
types of capacitors dry out and/or have their electrolyte deteriorate
or crystallize, resistors drift, heat/cool cycles cause PCB flexure
resulting in cracked solder joints, corrosion in air creeps into
connector pin interfaces and forces pins apart, the list goes on and
on and on.
?? To be fair, some of the above-listed failure modes do in fact
involve things moving, though imperceptibly so, my point stands.
?????????????? -Dave
6:36 p.m.
Not really trying to claim there is anything particular about DEC
equipment here. While they were built properly, I don't think there
necessarily is anything exceptional about them.
But no, I do not believe any components have been replaced. One of the
biggest issue is when equipment are left turned off for longer times.
That usually seems to be the time when problems appear. These machines
are (or were) running 24/7, and have always done so. There might have
been one or another card replaced from time to time, since they were on
service contracts for a long time. But just by looking into them, if
something had been replaced, it was a long time ago. Based just on the
general look of the cards.
By the way, we're talking 11/23 machines here. So BA23 box. Not that
much loaded in them, so no big strain on the power supply.
My experience with 11/70 machines are the same. If they are kept
running, they usually just continue to work. If you have them off for a
couple of years, you'll have a lot of work before they are stable again.
But disk drives get into trouble with age, even if continually running.
And some are worse than others. The RD53 is sortof infamous here...
Johnny
On 2021-01-22 18:15, Dave McGuire wrote:
? Sigh, yes, we all know of exceptions.? But even then, I doubt it can
be proven that those PDP-11 systems in production since the early 80s
have never had components replaced.
? Not everything is built to DEC standards.? That said, all my life
I've been replacing chips, diodes, transistors, capacitors, and
sometimes even resistors in DEC gear.? It does happen, and it's not
uncommon.
? The regulator bricks in the H742/H7420 power supplies tend to blow
Zener diodes.? That's fresh in my mind because we just lost another one
at the museum recently.? Those are failing with such frequency now that
I've purchased stocks of the components that tend to fail. (we have a
lot of those at the museum)? And those ARE built to DEC standards.
? Speaking of not being built to DEC standards, I've got a Heathkit H8
on the bench right now, I'm repairing it for a new exhibit at the
museum.? Four failed (non-moving!) ICs replaced so far, it's very nearly
done.
??????????? -Dave
On 1/22/21 12:03 PM, Johnny Billquist wrote:
--
Johnny Billquist || "I'm on a bus
|| on a psychedelic trip
email: bqt at softjar.se || Reading murder books
pdp is alive! || tryin' to stay hip" - B. Idol
Well, I know of PDP-11 systems out in production
who have been running
since the early 80s. The only thing that fails is the damn RD53...
Replaced those about 20 years ago now. The rest of the machine was
running just fine. At that at a steel mill, so not exactly a friendly
environment. Last I heard, things were still running, but it might
have been replaced by now. But anyway, it's the moving parts that
cause problems. The rest usually just keep running...
?? Johnny
On 2021-01-22 17:53, Dave McGuire wrote:
On 1/22/21 11:41 AM, Thomas DeBellis wrote:
10 years used to be nearly unheard of for retail
machines with
moving parts.? When nothing is moving, then supposedly there is
nothing to burn out.
?? I don't know who told you that, but he/she knows jack point squat
about electronics.
?? Materials migration and diffusion across junctions causes
semiconductor components to fail, tin whiskers cause shorts, some
types of capacitors dry out and/or have their electrolyte deteriorate
or crystallize, resistors drift, heat/cool cycles cause PCB flexure
resulting in cracked solder joints, corrosion in air creeps into
connector pin interfaces and forces pins apart, the list goes on and
on and on.
?? To be fair, some of the above-listed failure modes do in fact
involve things moving, though imperceptibly so, my point stands.
?????????????? -Dave
6:39 p.m.
And speaking of power - the one thing that I have experienced a lot on
11/70 machines are the H7440 regulators. They tend to blow up. And
especially if just kept in storage for longer periods, and then starting
to be used again...
Don't leave these machines off. That's all I say. That's when you get
trouble.
Johnny
On 2021-01-22 18:36, Johnny Billquist wrote:
Not really trying to claim there is anything
particular about DEC
equipment here. While they were built properly, I don't think there
necessarily is anything exceptional about them.
But no, I do not believe any components have been replaced. One of the
biggest issue is when equipment are left turned off for longer times.
That usually seems to be the time when problems appear. These machines
are (or were) running 24/7, and have always done so. There might have
been one or another card replaced from time to time, since they were on
service contracts for a long time. But just by looking into them, if
something had been replaced, it was a long time ago. Based just on the
general look of the cards.
By the way, we're talking 11/23 machines here. So BA23 box. Not that
much loaded in them, so no big strain on the power supply.
My experience with 11/70 machines are the same. If they are kept
running, they usually just continue to work. If you have them off for a
couple of years, you'll have a lot of work before they are stable again.
But disk drives get into trouble with age, even if continually running.
And some are worse than others. The RD53 is sortof infamous here...
? Johnny
On 2021-01-22 18:15, Dave McGuire wrote:
--
Johnny Billquist || "I'm on a bus
|| on a psychedelic trip
email: bqt at softjar.se || Reading murder books
pdp is alive! || tryin' to stay hip" - B. Idol
?? Sigh, yes, we all know of exceptions.? But even then, I doubt it
can be proven that those PDP-11 systems in production since the early
80s have never had components replaced.
?? Not everything is built to DEC standards.? That said, all my life
I've been replacing chips, diodes, transistors, capacitors, and
sometimes even resistors in DEC gear.? It does happen, and it's not
uncommon.
?? The regulator bricks in the H742/H7420 power supplies tend to blow
Zener diodes.? That's fresh in my mind because we just lost another
one at the museum recently.? Those are failing with such frequency now
that I've purchased stocks of the components that tend to fail. (we
have a lot of those at the museum)? And those ARE built to DEC standards.
?? Speaking of not being built to DEC standards, I've got a Heathkit
H8 on the bench right now, I'm repairing it for a new exhibit at the
museum.? Four failed (non-moving!) ICs replaced so far, it's very
nearly done.
???????????? -Dave
On 1/22/21 12:03 PM, Johnny Billquist wrote:
Well, I know of PDP-11 systems out in production
who have been
running since the early 80s. The only thing that fails is the damn
RD53...
Replaced those about 20 years ago now. The rest of the machine was
running just fine. At that at a steel mill, so not exactly a friendly
environment. Last I heard, things were still running, but it might
have been replaced by now. But anyway, it's the moving parts that
cause problems. The rest usually just keep running...
?? Johnny
On 2021-01-22 17:53, Dave McGuire wrote:
On 1/22/21 11:41 AM, Thomas DeBellis wrote:
> 10 years used to be nearly unheard of for retail machines with
> moving parts.? When nothing is moving, then supposedly there is
> nothing to burn out.
?? I don't know who told you that, but he/she knows jack point squat
about electronics.
?? Materials migration and diffusion across junctions causes
semiconductor components to fail, tin whiskers cause shorts, some
types of capacitors dry out and/or have their electrolyte
deteriorate or crystallize, resistors drift, heat/cool cycles cause
PCB flexure resulting in cracked solder joints, corrosion in air
creeps into connector pin interfaces and forces pins apart, the list
goes on and on and on.
?? To be fair, some of the above-listed failure modes do in fact
involve things moving, though imperceptibly so, my point stands.
?????????????? -Dave
6:43 p.m.
...despite Peter's experience with a KA10. I can't really comment on the
PDP-10s, since I've never worked much on them. But since I suspect they
might not have had switched power supplies, I'm not surprised if it
worked just fine, even after longer storage.
The only 10s I have worked a bit more on is KL10. And replacing a power
supply for the MF10 was quite a job, since it was a linear supply with a
big trafo in it. *Very* heavy... :-)
Johnny
On 2021-01-22 18:39, Johnny Billquist wrote:
And speaking of power - the one thing that I have
experienced a lot on
11/70 machines are the H7440 regulators. They tend to blow up. And
especially if just kept in storage for longer periods, and then starting
to be used again...
Don't leave these machines off. That's all I say. That's when you get
trouble.
? Johnny
On 2021-01-22 18:36, Johnny Billquist wrote:
--
Johnny Billquist || "I'm on a bus
|| on a psychedelic trip
email: bqt at softjar.se || Reading murder books
pdp is alive! || tryin' to stay hip" - B. Idol
Not really trying to claim there is anything
particular about DEC
equipment here. While they were built properly, I don't think there
necessarily is anything exceptional about them.
But no, I do not believe any components have been replaced. One of the
biggest issue is when equipment are left turned off for longer times.
That usually seems to be the time when problems appear. These machines
are (or were) running 24/7, and have always done so. There might have
been one or another card replaced from time to time, since they were
on service contracts for a long time. But just by looking into them,
if something had been replaced, it was a long time ago. Based just on
the general look of the cards.
By the way, we're talking 11/23 machines here. So BA23 box. Not that
much loaded in them, so no big strain on the power supply.
My experience with 11/70 machines are the same. If they are kept
running, they usually just continue to work. If you have them off for
a couple of years, you'll have a lot of work before they are stable
again.
But disk drives get into trouble with age, even if continually
running. And some are worse than others. The RD53 is sortof infamous
here...
?? Johnny
On 2021-01-22 18:15, Dave McGuire wrote:
?? Sigh, yes, we all know of exceptions.? But even then, I doubt it
can be proven that those PDP-11 systems in production since the early
80s have never had components replaced.
?? Not everything is built to DEC standards.? That said, all my life
I've been replacing chips, diodes, transistors, capacitors, and
sometimes even resistors in DEC gear.? It does happen, and it's not
uncommon.
?? The regulator bricks in the H742/H7420 power supplies tend to blow
Zener diodes.? That's fresh in my mind because we just lost another
one at the museum recently.? Those are failing with such frequency
now that I've purchased stocks of the components that tend to fail.
(we have a lot of those at the museum)? And those ARE built to DEC
standards.
?? Speaking of not being built to DEC standards, I've got a Heathkit
H8 on the bench right now, I'm repairing it for a new exhibit at the
museum.? Four failed (non-moving!) ICs replaced so far, it's very
nearly done.
???????????? -Dave
On 1/22/21 12:03 PM, Johnny Billquist wrote:
Well, I know of PDP-11 systems out in production
who have been
running since the early 80s. The only thing that fails is the damn
RD53...
Replaced those about 20 years ago now. The rest of the machine was
running just fine. At that at a steel mill, so not exactly a
friendly environment. Last I heard, things were still running, but
it might have been replaced by now. But anyway, it's the moving
parts that cause problems. The rest usually just keep running...
?? Johnny
On 2021-01-22 17:53, Dave McGuire wrote:
> On 1/22/21 11:41 AM, Thomas DeBellis wrote:
>> 10 years used to be nearly unheard of for retail machines with
>> moving parts.? When nothing is moving, then supposedly there is
>> nothing to burn out.
>
> ?? I don't know who told you that, but he/she knows jack point
> squat about electronics.
>
> ?? Materials migration and diffusion across junctions causes
> semiconductor components to fail, tin whiskers cause shorts, some
> types of capacitors dry out and/or have their electrolyte
> deteriorate or crystallize, resistors drift, heat/cool cycles cause
> PCB flexure resulting in cracked solder joints, corrosion in air
> creeps into connector pin interfaces and forces pins apart, the
> list goes on and on and on.
>
> ?? To be fair, some of the above-listed failure modes do in fact
> involve things moving, though imperceptibly so, my point stands.
>
> ?????????????? -Dave
>
7:01 p.m.
I'll let you know. I thought it was the entire pi but it could well just be
the SD card, they don't seem to do much without an os.
It has pretty much sat doing its' thing in the same position for 10 years,
could well be more, with only the occasional power outage. In the early
days when I had real vintage hardware in the basement and a UPS I think I
got to nearly 2 years up time.
Regards, Mark.
On Fri, 22 Jan 2021 at 17:43, Johnny Billquist <bqt at softjar.se> wrote:
...despite Peter's experience with a KA10. I
can't really comment on the
PDP-10s, since I've never worked much on them. But since I suspect they
might not have had switched power supplies, I'm not surprised if it
worked just fine, even after longer storage.
The only 10s I have worked a bit more on is KL10. And replacing a power
supply for the MF10 was quite a job, since it was a linear supply with a
big trafo in it. *Very* heavy... :-)
Johnny
On 2021-01-22 18:39, Johnny Billquist wrote:
--
https://www.qrz.com/db/M0NOM/P
And speaking of power - the one thing that I have
experienced a lot on
11/70 machines are the H7440 regulators. They tend to blow up. And
especially if just kept in storage for longer periods, and then starting
to be used again...
Don't leave these machines off. That's all I say. That's when you get
trouble.
Johnny
On 2021-01-22 18:36, Johnny Billquist wrote:
--
Johnny Billquist || "I'm on a bus
|| on a psychedelic trip
email: bqt at softjar.se || Reading murder books
pdp is alive! || tryin' to stay hip" - B. Idol
Not really trying to claim there is anything
particular about DEC
equipment here. While they were built properly, I don't think there
necessarily is anything exceptional about them.
But no, I do not believe any components have been replaced. One of the
biggest issue is when equipment are left turned off for longer times.
That usually seems to be the time when problems appear. These machines
are (or were) running 24/7, and have always done so. There might have
been one or another card replaced from time to time, since they were
on service contracts for a long time. But just by looking into them,
if something had been replaced, it was a long time ago. Based just on
the general look of the cards.
By the way, we're talking 11/23 machines here. So BA23 box. Not that
much loaded in them, so no big strain on the power supply.
My experience with 11/70 machines are the same. If they are kept
running, they usually just continue to work. If you have them off for
a couple of years, you'll have a lot of work before they are stable
again.
But disk drives get into trouble with age, even if continually
running. And some are worse than others. The RD53 is sortof infamous
here...
Johnny
On 2021-01-22 18:15, Dave McGuire wrote:
Sigh, yes, we all know of exceptions. But even then, I doubt it
can be proven that those PDP-11 systems in production since the early
80s have never had components replaced.
Not everything is built to DEC standards. That said, all my life
I've been replacing chips, diodes, transistors, capacitors, and
sometimes even resistors in DEC gear. It does happen, and it's not
uncommon.
The regulator bricks in the H742/H7420 power supplies tend to blow
Zener diodes. That's fresh in my mind because we just lost another
one at the museum recently. Those are failing with such frequency
now that I've purchased stocks of the components that tend to fail.
(we have a lot of those at the museum) And those ARE built to DEC
standards.
Speaking of not being built to DEC standards, I've got a Heathkit
H8 on the bench right now, I'm repairing it for a new exhibit at the
museum. Four failed (non-moving!) ICs replaced so far, it's very
nearly done.
-Dave
On 1/22/21 12:03 PM, Johnny Billquist wrote:
> Well, I know of PDP-11 systems out in production who have been
> running since the early 80s. The only thing that fails is the damn
> RD53...
> Replaced those about 20 years ago now. The rest of the machine was
> running just fine. At that at a steel mill, so not exactly a
> friendly environment. Last I heard, things were still running, but
> it might have been replaced by now. But anyway, it's the moving
> parts that cause problems. The rest usually just keep running...
>
> Johnny
>
> On 2021-01-22 17:53, Dave McGuire wrote:
>> On 1/22/21 11:41 AM, Thomas DeBellis wrote:
>>> 10 years used to be nearly unheard of for retail machines with
>>> moving parts. When nothing is moving, then supposedly there is
>>> nothing to burn out.
>>
>> I don't know who told you that, but he/she knows jack point
>> squat about electronics.
>>
>> Materials migration and diffusion across junctions causes
>> semiconductor components to fail, tin whiskers cause shorts, some
>> types of capacitors dry out and/or have their electrolyte
>> deteriorate or crystallize, resistors drift, heat/cool cycles cause
>> PCB flexure resulting in cracked solder joints, corrosion in air
>> creeps into connector pin interfaces and forces pins apart, the
>> list goes on and on and on.
>>
>> To be fair, some of the above-listed failure modes do in fact
>> involve things moving, though imperceptibly so, my point stands.
>>
>> -Dave
>>
>
7:03 p.m.
The KA has no switching power supplies.....
-P
----- Original Message -----
From: "bqt" <bqt at softjar.se>
To: "hecnet" <hecnet at Update.UU.SE>
Sent: Friday, January 22, 2021 12:43:40 PM
Subject: Re: [HECnet] No connectivity to arsgea 4
...despite Peter's experience with a KA10. I
can't really comment on the
PDP-10s, since I've never worked much on them. But since I suspect they
might not have had switched power supplies, I'm not surprised if it
worked just fine, even after longer storage.
The only 10s I have worked a bit more on is KL10. And replacing a power
supply for the MF10 was quite a job, since it was a linear supply with a
big trafo in it. *Very* heavy... :-)
Johnny
On 2021-01-22 18:39, Johnny Billquist wrote:
And speaking of power - the one thing that I have
experienced a lot on
11/70 machines are the H7440 regulators. They tend to blow up. And
especially if just kept in storage for longer periods, and then starting
to be used again...
Don't leave these machines off. That's all I say. That's when you get
trouble.
? Johnny
On 2021-01-22 18:36, Johnny Billquist wrote:
--
Johnny Billquist || "I'm on a bus
|| on a psychedelic trip
email: bqt at softjar.se || Reading murder books
pdp is alive! || tryin' to stay hip" - B. Idol Not really trying to claim there is anything
particular about DEC
equipment here. While they were built properly, I don't think there
necessarily is anything exceptional about them.
But no, I do not believe any components have been replaced. One of the
biggest issue is when equipment are left turned off for longer times.
That usually seems to be the time when problems appear. These machines
are (or were) running 24/7, and have always done so. There might have
been one or another card replaced from time to time, since they were
on service contracts for a long time. But just by looking into them,
if something had been replaced, it was a long time ago. Based just on
the general look of the cards.
By the way, we're talking 11/23 machines here. So BA23 box. Not that
much loaded in them, so no big strain on the power supply.
My experience with 11/70 machines are the same. If they are kept
running, they usually just continue to work. If you have them off for
a couple of years, you'll have a lot of work before they are stable
again.
But disk drives get into trouble with age, even if continually
running. And some are worse than others. The RD53 is sortof infamous
here...
?? Johnny
On 2021-01-22 18:15, Dave McGuire wrote:
?? Sigh, yes, we all know of exceptions.? But even then, I doubt it
can be proven that those PDP-11 systems in production since the early
80s have never had components replaced.
?? Not everything is built to DEC standards.? That said, all my life
I've been replacing chips, diodes, transistors, capacitors, and
sometimes even resistors in DEC gear.? It does happen, and it's not
uncommon.
?? The regulator bricks in the H742/H7420 power supplies tend to blow
Zener diodes.? That's fresh in my mind because we just lost another
one at the museum recently.? Those are failing with such frequency
now that I've purchased stocks of the components that tend to fail.
(we have a lot of those at the museum)? And those ARE built to DEC
standards.
?? Speaking of not being built to DEC standards, I've got a Heathkit
H8 on the bench right now, I'm repairing it for a new exhibit at the
museum.? Four failed (non-moving!) ICs replaced so far, it's very
nearly done.
???????????? -Dave
On 1/22/21 12:03 PM, Johnny Billquist wrote:
> Well, I know of PDP-11 systems out in production who have been
> running since the early 80s. The only thing that fails is the damn
> RD53...
> Replaced those about 20 years ago now. The rest of the machine was
> running just fine. At that at a steel mill, so not exactly a
> friendly environment. Last I heard, things were still running, but
> it might have been replaced by now. But anyway, it's the moving
> parts that cause problems. The rest usually just keep running...
>
> ?? Johnny
>
> On 2021-01-22 17:53, Dave McGuire wrote:
>> On 1/22/21 11:41 AM, Thomas DeBellis wrote:
>>> 10 years used to be nearly unheard of for retail machines with
>>> moving parts.? When nothing is moving, then supposedly there is
>>> nothing to burn out.
>>
>> ?? I don't know who told you that, but he/she knows jack point
>> squat about electronics.
>>
>> ?? Materials migration and diffusion across junctions causes
>> semiconductor components to fail, tin whiskers cause shorts, some
>> types of capacitors dry out and/or have their electrolyte
>> deteriorate or crystallize, resistors drift, heat/cool cycles cause
>> PCB flexure resulting in cracked solder joints, corrosion in air
>> creeps into connector pin interfaces and forces pins apart, the
>> list goes on and on and on.
>>
>> ?? To be fair, some of the above-listed failure modes do in fact
>> involve things moving, though imperceptibly so, my point stands.
>>
>> ?????????????? -Dave
>>
>
7:12 p.m.
I?ve not used SD cards in my Pi?s 3B and 4 for a while now. Most boot off USB to SSD
drives and a bunch of 8 boot off read only SD long enough to boot the rest of the way over
iSCSI.
Keith
On 22 Jan 2021, at 18:05, Peter Lothberg <roll at
stupi.com> wrote:
?The KA has no switching power supplies.....
-P
----- Original Message -----
From: "bqt" <bqt at softjar.se>
To: "hecnet" <hecnet at Update.UU.SE>
Sent: Friday, January 22, 2021 12:43:40 PM
Subject: Re: [HECnet] No connectivity to arsgea 4
> ...despite Peter's experience with a KA10. I can't really comment on the
> PDP-10s, since I've never worked much on them. But since I suspect they
> might not have had switched power supplies, I'm not surprised if it
> worked just fine, even after longer storage.
>
> The only 10s I have worked a bit more on is KL10. And replacing a power
> supply for the MF10 was quite a job, since it was a linear supply with a
> big trafo in it. *Very* heavy... :-)
>
> Johnny
>
>> On 2021-01-22 18:39, Johnny Billquist wrote:
>> And speaking of power - the one thing that I have experienced a lot on
>> 11/70 machines are the H7440 regulators. They tend to blow up. And
>> especially if just kept in storage for longer periods, and then starting
>> to be used again...
>>
>> Don't leave these machines off. That's all I say. That's when you
get
>> trouble.
>>
>> Johnny
>>
>> On 2021-01-22 18:36, Johnny Billquist wrote:
>>> Not really trying to claim there is anything particular about DEC
>>> equipment here. While they were built properly, I don't think there
>>> necessarily is anything exceptional about them.
>>>
>>> But no, I do not believe any components have been replaced. One of the
>>> biggest issue is when equipment are left turned off for longer times.
>>> That usually seems to be the time when problems appear. These machines
>>> are (or were) running 24/7, and have always done so. There might have
>>> been one or another card replaced from time to time, since they were
>>> on service contracts for a long time. But just by looking into them,
>>> if something had been replaced, it was a long time ago. Based just on
>>> the general look of the cards.
>>>
>>> By the way, we're talking 11/23 machines here. So BA23 box. Not that
>>> much loaded in them, so no big strain on the power supply.
>>>
>>> My experience with 11/70 machines are the same. If they are kept
>>> running, they usually just continue to work. If you have them off for
>>> a couple of years, you'll have a lot of work before they are stable
>>> again.
>>>
>>> But disk drives get into trouble with age, even if continually
>>> running. And some are worse than others. The RD53 is sortof infamous
>>> here...
>>>
>>> Johnny
>>>
>>> On 2021-01-22 18:15, Dave McGuire wrote:
>>>>
>>>> Sigh, yes, we all know of exceptions. But even then, I doubt it
>>>> can be proven that those PDP-11 systems in production since the early
>>>> 80s have never had components replaced.
>>>>
>>>> Not everything is built to DEC standards. That said, all my life
>>>> I've been replacing chips, diodes, transistors, capacitors, and
>>>> sometimes even resistors in DEC gear. It does happen, and it's not
>>>> uncommon.
>>>>
>>>> The regulator bricks in the H742/H7420 power supplies tend to blow
>>>> Zener diodes. That's fresh in my mind because we just lost another
>>>> one at the museum recently. Those are failing with such frequency
>>>> now that I've purchased stocks of the components that tend to fail.
>>>> (we have a lot of those at the museum) And those ARE built to DEC
>>>> standards.
>>>>
>>>> Speaking of not being built to DEC standards, I've got a Heathkit
>>>> H8 on the bench right now, I'm repairing it for a new exhibit at the
>>>> museum. Four failed (non-moving!) ICs replaced so far, it's very
>>>> nearly done.
>>>>
>>>> -Dave
>>>>
>>>> On 1/22/21 12:03 PM, Johnny Billquist wrote:
>>>>> Well, I know of PDP-11 systems out in production who have been
>>>>> running since the early 80s. The only thing that fails is the damn
>>>>> RD53...
>>>>> Replaced those about 20 years ago now. The rest of the machine was
>>>>> running just fine. At that at a steel mill, so not exactly a
>>>>> friendly environment. Last I heard, things were still running, but
>>>>> it might have been replaced by now. But anyway, it's the moving
>>>>> parts that cause problems. The rest usually just keep running...
>>>>>
>>>>> Johnny
>>>>>
>>>>> On 2021-01-22 17:53, Dave McGuire wrote:
>>>>>> On 1/22/21 11:41 AM, Thomas DeBellis wrote:
>>>>>>> 10 years used to be nearly unheard of for retail machines
with
>>>>>>> moving parts. When nothing is moving, then supposedly there
is
>>>>>>> nothing to burn out.
>>>>>>
>>>>>> I don't know who told you that, but he/she knows jack
point
>>>>>> squat about electronics.
>>>>>>
>>>>>> Materials migration and diffusion across junctions causes
>>>>>> semiconductor components to fail, tin whiskers cause shorts,
some
>>>>>> types of capacitors dry out and/or have their electrolyte
>>>>>> deteriorate or crystallize, resistors drift, heat/cool cycles
cause
>>>>>> PCB flexure resulting in cracked solder joints, corrosion in air
>>>>>> creeps into connector pin interfaces and forces pins apart, the
>>>>>> list goes on and on and on.
>>>>>>
>>>>>> To be fair, some of the above-listed failure modes do in fact
>>>>>> involve things moving, though imperceptibly so, my point stands.
>>>>>>
>>>>>> -Dave
>>>>>>
>>>>>
>>>>
>>>>
>>>
>>
>
> --
> Johnny Billquist || "I'm on a bus
> || on a psychedelic trip
> email: bqt at softjar.se || Reading murder books
> pdp is alive! || tryin' to stay hip" - B. Idol
7:57 p.m.
On 1/22/21 12:36 PM, Johnny Billquist wrote:
Not really trying to claim there is anything
particular about DEC
equipment here. While they were built properly, I don't think there
necessarily is anything exceptional about them.
I can tell you from experience that there definitely is. I've worked
on computer equipment a very wide variety of manufacturers. By far the
highest-quality by most measures has been IBM, with DEC and HP following
close behind. Nothing else even comes close. Note that I'm only
talking about minis and mainframes here.
But no, I do not believe any components have been
replaced.
That would be very unusual. Field service guys carry boards for a
reason. When I was running VAXen commercially, we had several boards
fail in our production machines, and they were replaced by DEC field
service personnel. The machines were VAX-11/750s, MicroVAX-IIs, and
MicroVAX-3s.
One of the
biggest issue is when equipment are left turned off for longer times.
That usually seems to be the time when problems appear.
When that happens, the problem usually turns out to be dried-out
electrolytic capacitors in power supplies, and occasional popcorning of
small dry tantalum capacitors.
By the way, we're talking 11/23 machines here. So
BA23 box. Not that
much loaded in them, so no big strain on the power supply.
Load, or lack thereof, makes no difference with the most common
failure mode of the BA23 power supply. The safety capacitors across the
line always see the same line voltage. Their flaws are aggravated by
humidity in the air, not load.
In fact, running the power supply under heavier load actually
improves the safety capacitor situation due to "bake-out" effects on
those capacitors. Of course such elevated temperatures then shorten the
life of the other capacitors, so..
FYI, the safety capacitors in the BA23 power supply are:
Refdes App C Class Lead Spacing
--------------------------------------------------
C101 line-line 0.47uF X 25mm
C102 line-ground 0.22uF Y 20mm
C201 line-line 0.1uF X2 20mm
My experience with 11/70 machines are the same. If
they are kept
running, they usually just continue to work. If you have them off for a
couple of years, you'll have a lot of work before they are stable again.
Yup, and this is something we deal with all the time at the museum.
We obviously cannot keep all of that hardware running all the time.
During times of sparse tours (pandemic..), we rotate through the
machines and run them for a little while every few weeks.
But disk drives get into trouble with age, even if
continually running.
And some are worse than others. The RD53 is sortof infamous here...
Yeah, they started dying 20+ years ago. Awful.
-Dave
--
Dave McGuire, AK4HZ
New Kensington, PA
7:18 p.m.
Moving parts are trouble, absolutely. That's why I built one of David Gesswein's
MFM emulators recently to replace the original hard drives in my Pro.
Then again, other parts can fail. The power inlet filter in my Pro failed spectacularly
(smoke, melting, stink). Fortunately replacing that is easier than replacing an ancient
hard drive. And that is a common point: many electronic things can be replaced by modern
parts, while mechanical things are far more problematic.
SSDs can fail, though, especially at modern densities. I recently saw a 256 GB microSD
card; that boggles the mind. I wonder what the life expectancy of one of those is. SD
cards certainly have write limits, so any that are written to a lot are likely to fail at
some point. More quickly if they are subjected to short (one block) random access writes
due to "write amplification".
Modern politically correct solder can be problematic too. "Lead free" processes
are less reliable than real soldering. For years, maybe still, lead free solder was not
acceptable for space applications. Perhaps that has been cured by now, I don't know
and don't care. (I was told quite directly by a professional "if you are
required to use lead free solder, do so. If you aren't required, avoid it.")
I do like SSDs. My home firewall, DECnet router, and Subversion server is a fanless (heat
sink cooled industrial type) SSD based PC running Linux. Unlike its plain consumer PC
predecessor, it has no objections at all to spending its life in a dusty basement.
paul
On Jan 22, 2021, at 12:03 PM, Johnny Billquist <bqt
at softjar.se> wrote:
Well, I know of PDP-11 systems out in production who have been running since the early
80s. The only thing that fails is the damn RD53...
Replaced those about 20 years ago now. The rest of the machine was running just fine. At
that at a steel mill, so not exactly a friendly environment. Last I heard, things were
still running, but it might have been replaced by now. But anyway, it's the moving
parts that cause problems. The rest usually just keep running...
Johnny
On 2021-01-22 17:53, Dave McGuire wrote:
On 1/22/21 11:41 AM, Thomas DeBellis wrote:
--
Johnny Billquist || "I'm on a bus
|| on a psychedelic trip
email: bqt at softjar.se || Reading murder books
pdp is alive! || tryin' to stay hip" - B. Idol 10 years used to be nearly unheard of for retail
machines with moving parts. When nothing is moving, then supposedly there is nothing to
burn out.
I don't know who told you that, but he/she knows jack point squat
about electronics.
Materials migration and diffusion across junctions causes semiconductor components to
fail, tin whiskers cause shorts, some types of capacitors dry out and/or have their
electrolyte deteriorate or crystallize, resistors drift, heat/cool cycles cause PCB
flexure resulting in cracked solder joints, corrosion in air creeps into connector pin
interfaces and forces pins apart, the list goes on and on and on.
To be fair, some of the above-listed failure modes do in fact involve things moving,
though imperceptibly so, my point stands.
-Dave 
7:35 p.m.
You do know microSD cards come in 1TB these days, right? :-D
-brian
On 1/22/21 1:18 PM, Paul Koning wrote:
Moving parts are trouble, absolutely. That's why
I built one of David Gesswein's MFM emulators recently to replace the original hard
drives in my Pro.
Then again, other parts can fail. The power inlet filter in my Pro failed spectacularly
(smoke, melting, stink). Fortunately replacing that is easier than replacing an ancient
hard drive. And that is a common point: many electronic things can be replaced by modern
parts, while mechanical things are far more problematic.
SSDs can fail, though, especially at modern densities. I recently saw a 256 GB microSD
card; that boggles the mind. I wonder what the life expectancy of one of those is. SD
cards certainly have write limits, so any that are written to a lot are likely to fail at
some point. More quickly if they are subjected to short (one block) random access writes
due to "write amplification".
Modern politically correct solder can be problematic too. "Lead free"
processes are less reliable than real soldering. For years, maybe still, lead free solder
was not acceptable for space applications. Perhaps that has been cured by now, I
don't know and don't care. (I was told quite directly by a professional "if
you are required to use lead free solder, do so. If you aren't required, avoid
it.")
I do like SSDs. My home firewall, DECnet router, and Subversion server is a fanless
(heat sink cooled industrial type) SSD based PC running Linux. Unlike its plain consumer
PC predecessor, it has no objections at all to spending its life in a dusty basement.
paul
> On Jan 22, 2021, at 12:03 PM, Johnny Billquist <bqt at softjar.se> wrote:
>
> Well, I know of PDP-11 systems out in production who have been running since the
early 80s. The only thing that fails is the damn RD53...
> Replaced those about 20 years ago now. The rest of the machine was running just fine.
At that at a steel mill, so not exactly a friendly environment. Last I heard, things were
still running, but it might have been replaced by now. But anyway, it's the moving
parts that cause problems. The rest usually just keep running...
>
> Johnny
>
> On 2021-01-22 17:53, Dave McGuire wrote:
>> On 1/22/21 11:41 AM, Thomas DeBellis wrote:
>>> 10 years used to be nearly unheard of for retail machines with moving parts.
When nothing is moving, then supposedly there is nothing to burn out.
>> I don't know who told you that, but he/she knows jack point squat about
electronics.
>> Materials migration and diffusion across junctions causes semiconductor
components to fail, tin whiskers cause shorts, some types of capacitors dry out and/or
have their electrolyte deteriorate or crystallize, resistors drift, heat/cool cycles cause
PCB flexure resulting in cracked solder joints, corrosion in air creeps into connector pin
interfaces and forces pins apart, the list goes on and on and on.
>> To be fair, some of the above-listed failure modes do in fact involve things
moving, though imperceptibly so, my point stands.
>> -Dave
> --
> Johnny Billquist || "I'm on a bus
> || on a psychedelic trip
> email: bqt at softjar.se || Reading murder books
> pdp is alive! || tryin' to stay hip" - B. Idol
7:57 p.m.
On 1/22/21 1:18 PM, Paul Koning wrote:
SSDs can fail, though, especially at modern densities.
I recently saw a 256 GB microSD card; that boggles the mind. I wonder what the life
expectancy of one of those is.
Not long. They get away with it because they're primarily intended
for the consumer market, and with the throwaway mentality that salesmen
have cultivated, combined with today's short attention spans, nobody
even seems to care when they fail.
Modern politically correct solder can be problematic
too. "Lead free" processes are less reliable than real soldering. For years,
maybe still, lead free solder was not acceptable for space applications. Perhaps that has
been cured by now, I don't know and don't care. (I was told quite directly by a
professional "if you are required to use lead free solder, do so. If you aren't
required, avoid it.")
I myself do care, as I do commercial design and manufacturing. Lead
free solder is still awful by every meaningful metric, and that
situation is not likely to improve. Fortunately, "lead free", when
applied to an assembly, only means "contains less than some percentage
of lead by weight", so most PCBs can actually be manufactured with
lead-bearing solder. I assemble mine using lead-bearing solder.
Remember the Toyota Prius accelerator failure that killed all those
people? That was caused by lead-free solder.
I do like SSDs. My home firewall, DECnet router, and
Subversion server is a fanless (heat sink cooled industrial type) SSD based PC running
Linux. Unlike its plain consumer PC predecessor, it has no objections at all to spending
its life in a dusty basement.
I've found SSDs to be a lot less reliable than spinning disks. I
still use them for reasons of speed and power consumption, but I do more
frequent backups.
-Dave
--
Dave McGuire, AK4HZ
New Kensington, PA
8:46 p.m.
Raspberry Pi is fine, it was the SD card that had failed.
However, the difference between the original Pi and the 2B in terms of boot
speed is quite noticeable!
Regards, Mark.
On Fri, 22 Jan 2021 at 18:57, Dave McGuire <mcguire at neurotica.com> wrote:
On 1/22/21 1:18 PM, Paul Koning wrote:
--
https://www.qrz.com/db/M0NOM/P
SSDs can fail, though, especially at modern
densities. I recently saw a
256 GB microSD card; that boggles the mind. I wonder
what the life
expectancy of one of those is.
Not long. They get away with it because they're primarily intended
for the consumer market, and with the throwaway mentality that salesmen
have cultivated, combined with today's short attention spans, nobody
even seems to care when they fail.
Modern politically correct solder can be
problematic too. "Lead free"
processes are less reliable than real
soldering. For years, maybe still,
lead free solder was not acceptable for space applications. Perhaps that
has been cured by now, I don't know and don't care. (I was told quite
directly by a professional "if you are required to use lead free solder, do
so. If you aren't required, avoid it.")
I myself do care, as I do commercial design and manufacturing. Lead
free solder is still awful by every meaningful metric, and that
situation is not likely to improve. Fortunately, "lead free", when
applied to an assembly, only means "contains less than some percentage
of lead by weight", so most PCBs can actually be manufactured with
lead-bearing solder. I assemble mine using lead-bearing solder.
Remember the Toyota Prius accelerator failure that killed all those
people? That was caused by lead-free solder.
I do like SSDs. My home firewall, DECnet router,
and Subversion server
is a fanless (heat sink cooled industrial type) SSD based PC
running
Linux. Unlike its plain consumer PC predecessor, it has no objections at
all to spending its life in a dusty basement.
I've found SSDs to be a lot less reliable than spinning disks. I
still use them for reasons of speed and power consumption, but I do more
frequent backups.
-Dave
--
Dave McGuire, AK4HZ
New Kensington, PA
10:29 p.m.
Then from 2b to 4 is like warp speed.
Aug.
Sent from my iPad
On Jan 22, 2021, at 1:47 PM, Mark Wickens <mark at
wickensonline.co.uk> wrote:
?
Raspberry Pi is fine, it was the SD card that had failed.
However, the difference between the original Pi and the 2B in terms of boot speed is
quite noticeable!
Regards, Mark.
On Fri, 22 Jan 2021 at 18:57, Dave McGuire
<mcguire at neurotica.com> wrote:
On 1/22/21 1:18 PM, Paul Koning wrote:
--
https://www.qrz.com/db/M0NOM/P
SSDs can fail, though, especially at modern
densities. I recently saw a 256 GB microSD card; that boggles the mind. I wonder what
the life expectancy of one of those is.
Not long. They get away with it because they're primarily intended
for the consumer market, and with the throwaway mentality that salesmen
have cultivated, combined with today's short attention spans, nobody
even seems to care when they fail.
Modern politically correct solder can be
problematic too. "Lead free" processes are less reliable than real soldering.
For years, maybe still, lead free solder was not acceptable for space applications.
Perhaps that has been cured by now, I don't know and don't care. (I was told
quite directly by a professional "if you are required to use lead free solder, do so.
If you aren't required, avoid it.")
I myself do care, as I do commercial design and manufacturing. Lead
free solder is still awful by every meaningful metric, and that
situation is not likely to improve. Fortunately, "lead free", when
applied to an assembly, only means "contains less than some percentage
of lead by weight", so most PCBs can actually be manufactured with
lead-bearing solder. I assemble mine using lead-bearing solder.
Remember the Toyota Prius accelerator failure that killed all those
people? That was caused by lead-free solder.
I do like SSDs. My home firewall, DECnet router,
and Subversion server is a fanless (heat sink cooled industrial type) SSD based PC running
Linux. Unlike its plain consumer PC predecessor, it has no objections at all to spending
its life in a dusty basement.
I've found SSDs to be a lot less reliable than spinning disks. I
still use them for reasons of speed and power consumption, but I do more
frequent backups.
-Dave
--
Dave McGuire, AK4HZ
New Kensington, PA
10:44 p.m.
USB 3 and true gigabit Ethernet on the 4 makes it more like trans warp, compared to 2b :)
On 22 Jan 2021, at 21:29, August Treubig <atreubig49 at charter.net> wrote:
? Then from 2b to 4 is like warp speed.
Aug.
Sent from my iPad
On Jan 22, 2021, at 1:47 PM, Mark Wickens <mark at wickensonline.co.uk> wrote:
?
Raspberry Pi is fine, it was the SD card that had failed.
However, the difference between the original Pi and the 2B in terms of boot speed is quite
noticeable!
Regards, Mark.
On Fri, 22 Jan 2021 at 18:57, Dave McGuire <mcguire at neurotica.com<mailto:mcguire
at neurotica.com>> wrote:
On 1/22/21 1:18 PM, Paul Koning wrote:
SSDs can fail, though, especially at modern densities.
I recently saw a 256 GB microSD card; that boggles the mind. I wonder what the life
expectancy of one of those is.
Not long. They get away with it because they're primarily intended
for the consumer market, and with the throwaway mentality that salesmen
have cultivated, combined with today's short attention spans, nobody
even seems to care when they fail.
Modern politically correct solder can be problematic
too. "Lead free" processes are less reliable than real soldering. For years,
maybe still, lead free solder was not acceptable for space applications. Perhaps that has
been cured by now, I don't know and don't care. (I was told quite directly by a
professional "if you are required to use lead free solder, do so. If you aren't
required, avoid it.")
I myself do care, as I do commercial design and manufacturing. Lead
free solder is still awful by every meaningful metric, and that
situation is not likely to improve. Fortunately, "lead free", when
applied to an assembly, only means "contains less than some percentage
of lead by weight", so most PCBs can actually be manufactured with
lead-bearing solder. I assemble mine using lead-bearing solder.
Remember the Toyota Prius accelerator failure that killed all those
people? That was caused by lead-free solder.
I do like SSDs. My home firewall, DECnet router, and
Subversion server is a fanless (heat sink cooled industrial type) SSD based PC running
Linux. Unlike its plain consumer PC predecessor, it has no objections at all to spending
its life in a dusty basement.
I've found SSDs to be a lot less reliable than spinning disks. I
still use them for reasons of speed and power consumption, but I do more
frequent backups.
-Dave
--
Dave McGuire, AK4HZ
New Kensington, PA
--
https://www.qrz.com/db/M0NOM/P
10:52 p.m.
Has anyone tried the Pi Keyboard?
I like the idea of it, but not sure if it would be good?
Regards, Mark.
On Fri, 22 Jan 2021 at 21:44, Keith Halewood <Keith.Halewood at pitbulluk.org>
wrote:
USB 3 and true gigabit Ethernet on the 4 makes it more
like trans warp,
compared to 2b :)
On 22 Jan 2021, at 21:29, August Treubig <atreubig49 at charter.net> wrote:
? Then from 2b to 4 is like warp speed.
Aug.
Sent from my iPad
On Jan 22, 2021, at 1:47 PM, Mark Wickens <mark at wickensonline.co.uk>
wrote:
?
Raspberry Pi is fine, it was the SD card that had failed.
However, the difference between the original Pi and the 2B in terms of
boot speed is quite noticeable!
Regards, Mark.
On Fri, 22 Jan 2021 at 18:57, Dave McGuire <mcguire at neurotica.com> wrote:
--
https://www.qrz.com/db/M0NOM/P
On 1/22/21 1:18 PM, Paul Koning wrote:
--
https://www.qrz.com/db/M0NOM/P
SSDs can fail, though, especially at modern
densities. I recently saw
a 256 GB microSD card; that boggles the mind. I wonder
what the life
expectancy of one of those is.
Not long. They get away with it because they're primarily intended
for the consumer market, and with the throwaway mentality that salesmen
have cultivated, combined with today's short attention spans, nobody
even seems to care when they fail.
Modern politically correct solder can be
problematic too. "Lead free"
processes are less reliable than real
soldering. For years, maybe still,
lead free solder was not acceptable for space applications. Perhaps that
has been cured by now, I don't know and don't care. (I was told quite
directly by a professional "if you are required to use lead free solder, do
so. If you aren't required, avoid it.")
I myself do care, as I do commercial design and manufacturing. Lead
free solder is still awful by every meaningful metric, and that
situation is not likely to improve. Fortunately, "lead free", when
applied to an assembly, only means "contains less than some percentage
of lead by weight", so most PCBs can actually be manufactured with
lead-bearing solder. I assemble mine using lead-bearing solder.
Remember the Toyota Prius accelerator failure that killed all those
people? That was caused by lead-free solder.
I do like SSDs. My home firewall, DECnet router,
and Subversion server
is a fanless (heat sink cooled industrial type) SSD based PC
running
Linux. Unlike its plain consumer PC predecessor, it has no objections at
all to spending its life in a dusty basement.
I've found SSDs to be a lot less reliable than spinning disks. I
still use them for reasons of speed and power consumption, but I do more
frequent backups.
-Dave
--
Dave McGuire, AK4HZ
New Kensington, PA
11:06 p.m.
I have one thanks to Christmas - I enjoy playing with it. I?m more used to ?headless? Pi
though with serial consoles.
I?m having difficulty getting the HDMI setting right for the Samsung quad hd ultra wide
monitor I have.
On 22 Jan 2021, at 21:53, Mark Wickens <mark at wickensonline.co.uk> wrote:
?
Has anyone tried the Pi Keyboard?
I like the idea of it, but not sure if it would be good?
Regards, Mark.
On Fri, 22 Jan 2021 at 21:44, Keith Halewood <Keith.Halewood at
pitbulluk.org<mailto:Keith.Halewood at pitbulluk.org>> wrote:
USB 3 and true gigabit Ethernet on the 4 makes it more like trans warp, compared to 2b :)
On 22 Jan 2021, at 21:29, August Treubig <atreubig49 at
charter.net<mailto:atreubig49 at charter.net>> wrote:
? Then from 2b to 4 is like warp speed.
Aug.
Sent from my iPad
On Jan 22, 2021, at 1:47 PM, Mark Wickens <mark at wickensonline.co.uk<mailto:mark
at wickensonline.co.uk>> wrote:
?
Raspberry Pi is fine, it was the SD card that had failed.
However, the difference between the original Pi and the 2B in terms of boot speed is quite
noticeable!
Regards, Mark.
On Fri, 22 Jan 2021 at 18:57, Dave McGuire <mcguire at neurotica.com<mailto:mcguire
at neurotica.com>> wrote:
On 1/22/21 1:18 PM, Paul Koning wrote:
SSDs can fail, though, especially at modern densities.
I recently saw a 256 GB microSD card; that boggles the mind. I wonder what the life
expectancy of one of those is.
Not long. They get away with it because they're primarily intended
for the consumer market, and with the throwaway mentality that salesmen
have cultivated, combined with today's short attention spans, nobody
even seems to care when they fail.
Modern politically correct solder can be problematic
too. "Lead free" processes are less reliable than real soldering. For years,
maybe still, lead free solder was not acceptable for space applications. Perhaps that has
been cured by now, I don't know and don't care. (I was told quite directly by a
professional "if you are required to use lead free solder, do so. If you aren't
required, avoid it.")
I myself do care, as I do commercial design and manufacturing. Lead
free solder is still awful by every meaningful metric, and that
situation is not likely to improve. Fortunately, "lead free", when
applied to an assembly, only means "contains less than some percentage
of lead by weight", so most PCBs can actually be manufactured with
lead-bearing solder. I assemble mine using lead-bearing solder.
Remember the Toyota Prius accelerator failure that killed all those
people? That was caused by lead-free solder.
I do like SSDs. My home firewall, DECnet router, and
Subversion server is a fanless (heat sink cooled industrial type) SSD based PC running
Linux. Unlike its plain consumer PC predecessor, it has no objections at all to spending
its life in a dusty basement.
I've found SSDs to be a lot less reliable than spinning disks. I
still use them for reasons of speed and power consumption, but I do more
frequent backups.
-Dave
--
Dave McGuire, AK4HZ
New Kensington, PA
--
https://www.qrz.com/db/M0NOM/P
--
https://www.qrz.com/db/M0NOM/P
23 Jan
23 Jan
5:54 p.m.
Well, as a matter of fact, I myself told myself this, based on my
anecdotal experience of 45 years in the field.? Since it's an anecdotal
observation, it can't be taken as scientific.? Moreover, your point is,
of course, correct.? And as soon as I sent the below, I wanted to
qualify it both in terms of manufacturing purpose and duty cycle.
Even with moving parts, 'professional' equipment will typically last
longer than consumer.? I have a highly venerable IBM X8668 server from
about 2000 that is fine, yet it has moving parts; a six drive RAID 5 and
the fans.? Those drives have never broken; I think we've had to blow out
a fan.
However, I should qualify this with the fact that the unit was almost
never shut off ever for over 12 years and was on a triply redundant
conditioned power supply during all that time.? So, no power up flexing,
Etc.? As a matter of fact, any machine that I care about here is on a
conditioned power supply (at least an APC Smart UPS).? My remark might
have been better put in that context; the power up surge flexing is what
eventually will do anything in.
To be fair, you're not the first person who has retorted to me about my
squatness with regards to electronics, my brother (who does the hardware
support) being very high on that particular list.? I'm quite happy
(often delighted) to blithely reply, "Yeah...? They don't do much
without me programming them, do they?"
------------------------------------------------------------------------
On 1/22/21 11:53 AM, Dave McGuire wrote:
? I don't know who told you that, but he/she knows jack point squat
about electronics.
? Materials migration and diffusion across junctions causes
semiconductor components to fail, tin whiskers cause shorts, some
types of capacitors dry out and/or have their electrolyte deteriorate
or crystallize, resistors drift, heat/cool cycles cause PCB flexure
resulting in cracked solder joints, corrosion in air creeps into
connector pin interfaces and forces pins apart, the list goes on and
on and on.
? To be fair, some of the above-listed failure modes do in fact
involve things moving, though imperceptibly so, my point stands.
????????????? -Dave
> ------------------------------------------------------------------------
> On 1/22/21 11:41 AM, Thomas DeBellis wrote:
>
> 10 years used to be nearly unheard of for retail machines with moving
> parts.? When nothing is moving, then supposedly there is nothing to
> burn out.
6:40 p.m.
On 1/23/21 11:54 AM, Thomas DeBellis wrote:
Well, as a matter of fact, I myself told myself this,
based on my
anecdotal experience of 45 years in the field.
In that case, please stay far, far away from soldering irons. ;)
Even with moving parts, 'professional'
equipment will typically last
longer than consumer.? I have a highly venerable IBM X8668 server from
about 2000 that is fine, yet it has moving parts; a six drive RAID 5 and
the fans.? Those drives have never broken; I think we've had to blow out
a fan.
Yes, that makes a lot of sense.
A funny story about that. In a large ISP environment in the
mid-1990s, suits smelled money and started infiltrating the company. Of
course, the predictable thing happened, everything went to shit. Part
of that process, though, was amusing. The suits whined and whined about
"all of these overpriced Sun Microsystems computers that we've never
heard of". (we had over two thousand of them at that point) They whined
that a PC costs $300, so why should we be buying $15,000 Sun computers?
So they started buying PCs, just a few, maybe a dozen. The $300
price they'd mentioned before was for eMachines garbage at Best Buy, but
what they actually ended up buying were monstrous $20,000 Compaq
ProLiant machines. About 25% of the computing power in 5x the rack
space at a higher price, wow what a great decision! Idiot suits. At
least they didn't break very often...only about twice as often as the
SPARCs.
So they quoted what they COULD spend on PC hardware, but not what
they DID spend. That's suit logic for you. The company is not long for
this world when processor architecture becomes a business decision.
Of course the new carpet-bagger VP's whole family suddenly went to
Tahiti for two weeks after the Compaq sales guy landed the sale, so we
knew what had transpired.
However, I should qualify this with the fact that the
unit was almost
never shut off ever for over 12 years and was on a triply redundant
conditioned power supply during all that time.? So, no power up flexing,
Etc.? As a matter of fact, any machine that I care about here is on a
conditioned power supply (at least an APC Smart UPS).? My remark might
have been better put in that context; the power up surge flexing is what
eventually will do anything in.
That's a good approach. The power-up surge flexing, as you aptly put
it, is damaging to this type of hardware. However, just plain age, when
it comes to components, is the larger problem as things start to get
very old. I know this from learning about how the components actually
work, but my repair area's trash can, chock full of failed capacitors
and 7400-series ICs, is literally a bucket of data in support of it. ;)
We have quite a lot of very old computer hardware at the museum, and
we repair and maintain nearly all of it. I've been repairing stuff all
my life (I bought myself a new bike as a kid by fishing broken TVs out
of dumpsters, then fixing and selling them), but that really kicked into
high gear with the construction of the museum. As far as failure modes
in electronics, I think at this point we've pretty much seen it all over
there.
To be fair, you're not the first person who has
retorted to me about my
squatness with regards to electronics, my brother (who does the hardware
support) being very high on that particular list.? I'm quite happy
(often delighted) to blithely reply, "Yeah...? They don't do much
without me programming them, do they?"
ROFL!
-Dave
--
Dave McGuire, AK4HZ
New Kensington, PA
6:48 p.m.
Speaking of things like 74-series stuff. Based on various restoration
projects I'm aware of, there seems to be a huge difference in quality of
those ICs. Depending on manufacturing batch, and manufacturer...
Some never fail, while with some others, you can almost predict that
they will be non-functional now.
So I guess it also do depend a bit on who you picked as supplier back in
the day.
Johnny
On 2021-01-23 18:40, Dave McGuire wrote:
On 1/23/21 11:54 AM, Thomas DeBellis wrote:
--
Johnny Billquist || "I'm on a bus
|| on a psychedelic trip
email: bqt at softjar.se || Reading murder books
pdp is alive! || tryin' to stay hip" - B. Idol
Well, as a matter of fact, I myself told myself
this, based on my
anecdotal experience of 45 years in the field.
? In that case, please stay far, far away from soldering irons. ;)
Even with moving parts, 'professional'
equipment will typically last
longer than consumer.? I have a highly venerable IBM X8668 server from
about 2000 that is fine, yet it has moving parts; a six drive RAID 5
and the fans.? Those drives have never broken; I think we've had to
blow out a fan.
? Yes, that makes a lot of sense.
? A funny story about that.? In a large ISP environment in the
mid-1990s, suits smelled money and started infiltrating the company.? Of
course, the predictable thing happened, everything went to shit.? Part
of that process, though, was amusing.? The suits whined and whined about
"all of these overpriced Sun Microsystems computers that we've never
heard of". (we had over two thousand of them at that point)? They whined
that a PC costs $300, so why should we be buying $15,000 Sun computers?
? So they started buying PCs, just a few, maybe a dozen.? The $300
price they'd mentioned before was for eMachines garbage at Best Buy, but
what they actually ended up buying were monstrous $20,000 Compaq
ProLiant machines.? About 25% of the computing power in 5x the rack
space at a higher price, wow what a great decision!? Idiot suits.? At
least they didn't break very often...only about twice as often as the
SPARCs.
? So they quoted what they COULD spend on PC hardware, but not what
they DID spend.? That's suit logic for you.? The company is not long for
this world when processor architecture becomes a business decision.
? Of course the new carpet-bagger VP's whole family suddenly went to
Tahiti for two weeks after the Compaq sales guy landed the sale, so we
knew what had transpired.
However, I should qualify this with the fact that
the unit was almost
never shut off ever for over 12 years and was on a triply redundant
conditioned power supply during all that time.? So, no power up
flexing, Etc.? As a matter of fact, any machine that I care about here
is on a conditioned power supply (at least an APC Smart UPS).? My
remark might have been better put in that context; the power up surge
flexing is what eventually will do anything in.
? That's a good approach.? The power-up surge flexing, as you aptly put
it, is damaging to this type of hardware.? However, just plain age, when
it comes to components, is the larger problem as things start to get
very old.? I know this from learning about how the components actually
work, but my repair area's trash can, chock full of failed capacitors
and 7400-series ICs, is literally a bucket of data in support of it. ;)
? We have quite a lot of very old computer hardware at the museum, and
we repair and maintain nearly all of it.? I've been repairing stuff all
my life (I bought myself a new bike as a kid by fishing broken TVs out
of dumpsters, then fixing and selling them), but that really kicked into
high gear with the construction of the museum.? As far as failure modes
in electronics, I think at this point we've pretty much seen it all over
there.
To be fair, you're not the first person who
has retorted to me about
my squatness with regards to electronics, my brother (who does the
hardware support) being very high on that particular list.? I'm quite
happy (often delighted) to blithely reply, "Yeah...? They don't do
much without me programming them, do they?"
? ROFL!
??????????? -Dave
7:22 p.m.
On 1/23/21 12:48 PM, Johnny Billquist wrote:
Speaking of things like 74-series stuff. Based on
various restoration
projects I'm aware of, there seems to be a huge difference in quality of
those ICs. Depending on manufacturing batch, and manufacturer...
Some never fail, while with some others, you can almost predict that
they will be non-functional now.
So I guess it also do depend a bit on who you picked as supplier back in
the day.
Yes, very much so. We've noticed that National Semiconductor is the
worst offender for actual IC failures, and Texas Instruments is the
worst for corrosion on IC leads.
-Dave
--
Dave McGuire, AK4HZ
New Kensington, PA
24 Jan
24 Jan
12:40 a.m.
Reminds me a little of the hardware development of the BBC Micro. The original design used
an NS 81LS95 multiplexer for the memory, and for reasons that nobody could explain, only
the NS component worked in their design - the same component from other vendors failed.
The core reason was due to the bus speed and how the original infrastructure design was
done, but the choice of many components was critical to the stability of the machine.
I'm not implying NS is better - just saying that it's interesting to compare
stability and long term failure rates in the field and how much they can vary.
Cheers, Wiz!!
-----Original Message-----
From: owner-hecnet at Update.UU.SE [mailto:owner-
hecnet at Update.UU.SE] On Behalf Of Dave McGuire
Sent: Sunday, 24 January 2021 5:23 AM
To: hecnet at Update.UU.SE
Subject: Re: [HECnet] No connectivity to arsgea 4
On 1/23/21 12:48 PM, Johnny Billquist wrote:
Speaking of things like 74-series stuff. Based on
various restoration
projects I'm aware of, there seems to be a huge difference in quality of
those ICs. Depending on manufacturing batch, and manufacturer...
Some never fail, while with some others, you can almost predict that
they will be non-functional now.
So I guess it also do depend a bit on who you picked as supplier back in
the day.
Yes, very much so. We've noticed that National Semiconductor is the
worst offender for actual IC failures, and Texas Instruments is the
worst for corrosion on IC leads.
-Dave
--
Dave McGuire, AK4HZ
New Kensington, PA
7:07 a.m.
On 1/23/21 6:40 PM, David Moylan wrote:
Reminds me a little of the hardware development of the
BBC Micro. The original design used an NS 81LS95 multiplexer for the memory, and for
reasons that nobody could explain, only the NS component worked in their design - the same
component from other vendors failed.
The core reason was due to the bus speed and how the original infrastructure design was
done, but the choice of many components was critical to the stability of the machine.
I'm not implying NS is better - just saying that it's interesting to compare
stability and long term failure rates in the field and how much they can vary.
Far be it for me to disparage Nanny Beeb, but this type of dependency
is indicative of poor design. Part of the design process involves
calculating out all margins and tolerances to arrive at worst-case
figures for things like logic voltage levels, timing, etc.
The datasheet for a component constitutes a sort of contract. The
manufacturer of the component guarantees that the component's
specifications will match those listed in the datasheet...but nothing
more. All manufacturers can and do to change HOW a component meets
those specifications, and subtle variations can and do occur as a result.
Datasheets usually list minimum, typical, and maximum values for
various parameters. From one fab to the next, from one production batch
to the next, and even from one chip to the next, the component's
performance on a given parameter will vary within that range. All
designers know this (or at least they're supposed to!) and they must
perform end-to-end calculations based on the min/max range of those
parameters.
A dependency on a particular manufacturer's implementation of a
standard and common design pattern like the 81LS95 is a bug. I sure
hope they fixed it in a later revision.
-Dave
--
Dave McGuire, AK4HZ
New Kensington, PA
11:05 a.m.
From: owner-hecnet at Update.UU.SE [mailto:owner-
hecnet at Update.UU.SE] On Behalf Of Dave McGuire
Far be it for me to disparage Nanny Beeb, but this
type of dependency
is indicative of poor design.
Oh yes. Poor design is certainly the case here. In order to keep everything simplistic
they wanted to avoid BUS contention while running the CPU at 2MHz. They had trouble
finding RAM which would match this design and Hitachi was the only manufacturer with a
suitable component. The NS 81LS95 multiplexer was the only component that was stable in
this configuration.
To make matters worse, they were exceeding the CPU's specifications which cause the
machine to crash. The designers worked out if they put their finger on a certain spot on
the board the machine became stable. They updated the design to include a resistor pack
across the BUS which kept the machine stable. This kludge was nick named "the
engineers fingers".
A dependency on a particular manufacturer's
implementation of a
standard and common design pattern like the 81LS95 is a bug. I sure
hope they fixed it in a later revision.
Absolutely not. They produced around 1.5 million machines with both of these kludge fixes
in place and the machine made it to end of life with no revision changes.
Cheers, Wiz!!
12:44 p.m.
I still have a perfectly functioning BBC Model B, the only concession being a floppy disc
emulator and video output improvements. It?s one of the fastest 6502 based machines we
used at school and university. The computer science department had classrooms of them with
their own VT100 ROMs for connection to the VAXen via Gandalf boxes. 24 lines for the
terminal and the remaining 12 lines for a programmable ?crib? to display the mapping of
the function keys. Other ESC sequences were eventually added to allow full access to
graphics and to allow download of binary images into memory for execution. Our VAXen could
download all the games! Those were the days.
The computer centre authored their own 3270 emulator ROM for connections to their beloved
4341 and 3083. Funnily enough, nobody wanted a great big noisy (and sometimes smelly)
Memorex
terminal in their offices.
Keith
On 24 Jan 2021, at 10:06, David Moylan <djm at
wiz.net.au> wrote:
?
From: owner-hecnet at Update.UU.SE [mailto:owner-
hecnet at Update.UU.SE] On Behalf Of Dave McGuire
Far be it for me to disparage Nanny Beeb, but
this type of dependency
is indicative of poor design.
Oh yes. Poor design is certainly the case here. In order to keep everything simplistic
they wanted to avoid BUS contention while running the CPU at 2MHz. They had trouble
finding RAM which would match this design and Hitachi was the only manufacturer with a
suitable component. The NS 81LS95 multiplexer was the only component that was stable in
this configuration.
To make matters worse, they were exceeding the CPU's specifications which cause the
machine to crash. The designers worked out if they put their finger on a certain spot on
the board the machine became stable. They updated the design to include a resistor pack
across the BUS which kept the machine stable. This kludge was nick named "the
engineers fingers".
A dependency on a particular manufacturer's
implementation of a
standard and common design pattern like the 81LS95 is a bug. I sure
hope they fixed it in a later revision.
Absolutely not. They produced around 1.5 million machines with both of these kludge fixes
in place and the machine made it to end of life with no revision changes.
Cheers, Wiz!!
5:28 p.m.
------------------------------------------------------------------------
On 1/23/21 11:54 AM, Thomas DeBellis wrote:
Well, as a matter of fact, I myself told myself this, based on my
anecdotal experience of 45 years in the field.
------------------------------------------------------------------------
On 1/23/21 12:40 PM, Dave McGuire wrote:
? In that case, please stay far, far away from soldering irons. ;) ? A funny story about that.? In a large ISP
environment in the
mid-1990s, suits smelled money and started infiltrating the company.?
Of course, the predictable thing happened, everything went to shit.?
Part of that process, though, was amusing.? The suits whined and
whined about "all of these overpriced Sun Microsystems computers that
we've never heard of". (we had over two thousand of them at that
point)? They whined that a PC costs $300, so why should we be buying
$15,000 Sun computers?
? So they started buying PCs, just a few, maybe a dozen.? The $300
price they'd mentioned before was for eMachines garbage at Best Buy,
but what they actually ended up buying were monstrous $20,000 Compaq
ProLiant machines.? About 25% of the computing power in 5x the rack
space at a higher price, wow what a great decision! Idiot suits.? At
least they didn't break very often...only about twice as often as the
SPARCs.
This jogs another memory: at a hospital, I managed support for a number
of medical practices on the campus and handled communications with the
main hospital network.
Medical practices have similar requirements and these particular ones
all ran the same scheduling, billing and records software. The initial
meeting with each ran along similar lines, "Please explain how you make
money" and after this was answered, "How much downtime can you afford?"?
Invariably, the answer to the 2^nd was a wide eyed, "None",? to which I
responded, "OK, so go buy X, Y and Z and you pretty much won't
experience any kind of extended outage or, more likely, /never/ have them"
X, Y and Z were an IBM eServer with RAID and backup, redundant
conditioned power supplies and separate HVAC.? Now, the 8668 is in no
way the answer to the world's ills, but in this case, due to other
circumstances, it really was ridiculous to even consider anything else.?
Again, unique circumstances, likely unrepeatable. Yet the practices that
did this got exactly what I told them. Zero outages.? Not a single one
...Except...
Talking to one particular owner was like talking to an open window.?
He'd nod sagely and then go do whatever he felt like.? So he went and
bought a (you guessed it) Compaq workstation, _nothing_ else and then
ran the entire office office off that, a task that it was utterly
unsuited for.? So, short term, Yes--they saved the up front costs.? On
the other hand, they were forever having capacity issues until the poor
thing finally gave up the ghost, turning the entire practice into a
paper operation for the several days it took to put everything back
together.
Yep; there is nothing like doing something you're not qualified to do,
whether or not you're a scoundrel.
22 Jan
22 Jan
6:37 p.m.
I lost my first pi (an early B rev 1) late last year when it melted during
a kernel rebuild trying to add decnet.
The other two seem to be going strong.
On Fri, 22 Jan 2021, 16:42 Thomas DeBellis, <tommytimesharing at gmail.com>
wrote:
Yeah, I would be interested in hearing about the
failure mode, too if you
know it, Mark. I was interested in getting one of these to handle some odd
things that need getting done, probably the new 8 GB model.
10 years used to be nearly unheard of for retail machines with moving
parts. When nothing is moving, then supposedly there is nothing to burn
out. I have a laptop from 2000 that I replaced the disk with an SSD and it
still boots fine (great for regression testing. Another two date from
before 2010 and have SSD's; no problems there, either.
------------------------------
On 1/22/21 10:17 AM, Supratim Sanyal wrote:
The entire Pi died and not just the microSD?
------------------------------
On Jan 22, 2021, at 9:10 AM, Mark Wickens <mark at wickensonline.co.uk>
wrote:
The Raspberry Pi that I have been using for at least 10 years to run the
hecnet bridge has finally died so I need to setup a new one.
In the meantime my area will be offline as far as hecnet is concerned.
Regards, Mark.
--
https://www.qrz.com/db/M0NOM/P
1194
days inactive
1196
days old
33 comments
13 participants
participants (13)
-
atreubig49＠charter.net -
bqt＠softjar.se -
djm＠wiz.net.au -
johnhreinhardt＠thereinhardts.org -
Keith.Halewood＠pitbulluk.org -
mark＠wickensonline.co.uk -
mcguire＠neurotica.com -
paulkoning＠comcast.net -
roll＠stupi.com -
supratim＠riseup.net -
tommytimesharing＠gmail.com -
tonyblews＠gmail.com -
wonko＠4amlunch.net