19 May
2025
19 May
'25
1:42 p.m.
From: Steve Jenkin
An unanswered question about Silicon Valley is:
Why did it happen in California and not be successfully cloned
elsewhere?
One good attempt at answering this is in "Making Silicon Valley: Innovation
and the Growth of High Tech, 1930-1970", by Christophe Lecuyer; it's also a
very good history of the early Silicon Valley (before the mid-1960's).
Most of it's available online, at Google:
https://books.google.com/books?id=5TgKinNy5p8C
I have neither the time nor energy to comment in detail on your very detailed
post, but I think Lecuyer would mostly agree with your points.
It wasn't just AT&T, IBM & DEC that got
run over by commodity DRAM &
CPU's, it was the entire Minicomputer Industry, effectively extinct by
1995.
Same thing for the work-station industry (with Sun being merely the most
notable example). I have a tiny bit of second-hand personal knowldge in this
area; my wife works for NASA, as a structural engineer, and they run a lot of
large computerized mathematical models. In the 70's, they were using CDC
7600's; they moved along through various things as technology changed (IIRC,
at one point they had SGI machines). These days, they seem to mostly be using
high-end personal computers for this.
Some specialized uses (various forms of CAD) I guess still use things that
look like work-stations, but I expect they are stock personal computers
with special I/O (very large displays, etc).
So I guess now there are just supercomputers (themselves mostly built out of
large numbers of commodity CPUs), and laptops. Well, there is also cloud
computing, which is huge, but that also just uses lots of commodity CPUs.
Noel
19 May
19 May
4:18 p.m.
New subject: Demise of AT&T
On Mon, May 19, 2025 at 9:43 AM Noel Chiappa <jnc(a)mercury.lcs.mit.edu>
wrote:
Indeed. Extinction of existing practice by new technology happens all the
time and not just in the computer field. Before sound recording and
playback was invented, fancy restaurants, hotels, etc. hired orchestras to
play background music. The phonograph put a huge number of orchestras out
of business.
There's a knee-jerk tendency in our industry for companies to respond to
new technology by retreating to the high end of the market in an attempt to
protect profit margins. IBM did this when minicomputers came along; Apollo
did it when confronted with cheap workstations (Sun); DEC did it when the
PC appeared; Intel right now is attempting to do it to stave off
competition from ARM. In all cases the effort has been unsuccessful and
usually has resulted in the death of the company (IBM has survived, but as
a shadow of its former self).
I think it was Scott McNealy who said (regarding protecting existing
products and profit margins) that a tech company must be prepared to eat
its own children. If they don't, the competition will.
-Paul W.
It wasn't
just AT&T, IBM & DEC that got run over by commodity DRAM &
CPU's, it was the entire Minicomputer Industry, effectively extinct
by
1995.
Same thing for the work-station industry (with Sun being merely the most
notable example).
7:17 p.m.
New subject: Demise of AT&T
I had a talk with a DEC salesman in Sep 1990. They had just loaned me a
DECStation for eval.
me: "What happens if computers become commodities like sneakers"
DEC salesman: "we go out of business"
They were right.
On Mon, May 19, 2025 at 9:18 AM Paul Winalski <paul.winalski(a)gmail.com>
wrote:
On Mon, May 19, 2025 at 9:43 AM Noel Chiappa
<jnc(a)mercury.lcs.mit.edu>
wrote:
Indeed. Extinction of existing practice by new technology happens all the
time and not just in the computer field. Before sound recording and
playback was invented, fancy restaurants, hotels, etc. hired orchestras to
play background music. The phonograph put a huge number of orchestras out
of business.
There's a knee-jerk tendency in our industry for companies to respond to
new technology by retreating to the high end of the market in an attempt to
protect profit margins. IBM did this when minicomputers came along; Apollo
did it when confronted with cheap workstations (Sun); DEC did it when the
PC appeared; Intel right now is attempting to do it to stave off
competition from ARM. In all cases the effort has been unsuccessful and
usually has resulted in the death of the company (IBM has survived, but as
a shadow of its former self).
I think it was Scott McNealy who said (regarding protecting existing
products and profit margins) that a tech company must be prepared to eat
its own children. If they don't, the competition will.
-Paul W.
It
wasn't just AT&T, IBM & DEC that got run over by commodity DRAM &
CPU's, it was the entire Minicomputer Industry, effectively extinct
by
1995.
Same thing for the work-station industry (with Sun being merely the most
notable example).
7:44 p.m.
New subject: Whither Workstations? (Was Re: Demise of AT&T)
On Mon, May 19, 2025 at 12:36 PM Noel Chiappa <jnc(a)mercury.lcs.mit.edu> wrote:
The CPU ISAs may be _largely_ shared, but computing has bifurcated
into two distinct camps: those machines intended for use in
datacenters, and those intended for consumer use by end-users.
CPUs intended for datacenters tend to be characterized by large
caches, lower average clock speeds, wider IO and memory bandwidth.
Those intended for consumer use tend to have high clock speeds, a bit
less cache, and support for comparatively fewer IO devices/less
memory. On the end-user side, you've got a further split between
laptops/desktop machines and devices like phones, tablets, and so on.
In both cases, the dominant IO buses used are PCIe and its variants
(e.g., on the data center side you've got CXL), NVMe for storage is
common in lots of places, everything supports Ethernet more or less
(even WiFi uses the ethernet frame format), and so on. USB seems
ubiquitous for peripherals on the end-user side.
In short, these machines may be called "personal computers" and they
may be PCs in the sense of being used primarily by one user, but
contemporary data center machines have more in common with mainframes
and high-end servers than the original PCs, and consumer machines are
much closer architecturally to high end workstations than to
yesteryear's PCs.
My desktop machine is a Mac Studio with an ARM CPU; I call it a
workstation and I think that's pretty accurate. At work, one of our
EE's has a big x86 thing with some stupidly powerful graphics card
that he uses to do board layout. It's a workstation in every
recognizable sense, though it does happen to run Windows.
- Dan C.
[snip]
It wasn't just AT&T, IBM & DEC that got run over by commodity DRAM &
CPU's, it was the entire Minicomputer Industry, effectively extinct by
1995.
Same thing for the work-station industry (with Sun being merely the most
notable example). I have a tiny bit of second-hand personal knowldge in this
area; my wife works for NASA, as a structural engineer, and they run a lot of
large computerized mathematical models. In the 70's, they were using CDC
7600's; they moved along through various things as technology changed (IIRC,
at one point they had SGI machines). These days, they seem to mostly be using
high-end personal computers for this.
Some specialized uses (various forms of CAD) I guess still use things that
look like work-stations, but I expect they are stock personal computers
with special I/O (very large displays, etc).
So I guess now there are just supercomputers (themselves mostly built out of
large numbers of commodity CPUs), and laptops. Well, there is also cloud
computing, which is huge, but that also just uses lots of commodity CPUs. 
8:28 p.m.
New subject: Whither Workstations? (Was Re: Demise of AT&T)
On Monday, May 19th, 2025 at 12:45 PM, Dan Cross <crossd(a)gmail.com> wrote:
On Mon, May 19, 2025 at 12:36 PM Noel Chiappa
jnc(a)mercury.lcs.mit.edu wrote:
The CPU ISAs may be largely shared, but computing has bifurcated
into two distinct camps: those machines intended for use in
datacenters, and those intended for consumer use by end-users.
CPUs intended for datacenters tend to be characterized by large
caches, lower average clock speeds, wider IO and memory bandwidth.
Those intended for consumer use tend to have high clock speeds, a bit
less cache, and support for comparatively fewer IO devices/less
memory. On the end-user side, you've got a further split between
laptops/desktop machines and devices like phones, tablets, and so on.
In both cases, the dominant IO buses used are PCIe and its variants
(e.g., on the data center side you've got CXL), NVMe for storage is
common in lots of places, everything supports Ethernet more or less
(even WiFi uses the ethernet frame format), and so on. USB seems
ubiquitous for peripherals on the end-user side.
In short, these machines may be called "personal computers" and they
may be PCs in the sense of being used primarily by one user, but
contemporary data center machines have more in common with mainframes
and high-end servers than the original PCs, and consumer machines are
much closer architecturally to high end workstations than to
yesteryear's PCs.
My desktop machine is a Mac Studio with an ARM CPU; I call it a
workstation and I think that's pretty accurate. At work, one of our
EE's has a big x86 thing with some stupidly powerful graphics card
that he uses to do board layout. It's a workstation in every
recognizable sense, though it does happen to run Windows.
- Dan C.
This may be getting into the weeds a bit but don't forget industrial hardware, the
stuff where you approach the blurry demarcation between CPU and MCU. This for me is
always the third class when I'm discussing that sort of thing. Of course this also
means "operating systems" closer to standalone applications sitting on top of
some microkernel like (se)L4, but you did have VME-based workstations and UNIX versions
specifically for VME systems. For me these walk a line between true workstations and
minicomputers, but as usual that is from the perspective of having not been there. For
the record, one of the canonical UNIX development environments from AT&T for WE32x00
stuff was a WE32100 and support chips thrown on a VME module running System V/VME. From
what I know, VME is still quite common in industrial control. How much UNIX and
workalikes constitute the OS landscape in today's VME ecosystem eludes me.
Either way, I feel like this is a class of computers that frequently flies under the
radar, but if we're talking strictly consumer stuff, yeah VME very quickly loses
relevance.
- Matt G.
[snip]
It wasn't just AT&T, IBM & DEC that got run over by commodity DRAM &
CPU's, it was the entire Minicomputer Industry, effectively extinct by
1995.
Same thing for the work-station industry (with Sun being merely the most
notable example). I have a tiny bit of second-hand personal knowldge in this
area; my wife works for NASA, as a structural engineer, and they run a lot of
large computerized mathematical models. In the 70's, they were using CDC
7600's; they moved along through various things as technology changed (IIRC,
at one point they had SGI machines). These days, they seem to mostly be using
high-end personal computers for this.
Some specialized uses (various forms of CAD) I guess still use things that
look like work-stations, but I expect they are stock personal computers
with special I/O (very large displays, etc).
So I guess now there are just supercomputers (themselves mostly built out of
large numbers of commodity CPUs), and laptops. Well, there is also cloud
computing, which is huge, but that also just uses lots of commodity CPUs.
8:54 p.m.
New subject: Whither Workstations? (Was Re: Demise of AT&T)
On Mon, May 19, 2025 at 4:28 PM segaloco via TUHS <tuhs(a)tuhs.org> wrote:
On Monday, May 19th, 2025 at 12:45 PM, Dan Cross
<crossd(a)gmail.com> wrote:
That's fair. I also ignored SBCs like the Raspberry Pi and its
imitators and strictly embedded stuff.
On Mon, May 19, 2025 at 12:36 PM Noel Chiappa
jnc(a)mercury.lcs.mit.edu wrote:
The CPU ISAs may be largely shared, but computing has bifurcated
into two distinct camps: those machines intended for use in
datacenters, and those intended for consumer use by end-users.
CPUs intended for datacenters tend to be characterized by large
caches, lower average clock speeds, wider IO and memory bandwidth.
Those intended for consumer use tend to have high clock speeds, a bit
less cache, and support for comparatively fewer IO devices/less
memory. On the end-user side, you've got a further split between
laptops/desktop machines and devices like phones, tablets, and so on.
In both cases, the dominant IO buses used are PCIe and its variants
(e.g., on the data center side you've got CXL), NVMe for storage is
common in lots of places, everything supports Ethernet more or less
(even WiFi uses the ethernet frame format), and so on. USB seems
ubiquitous for peripherals on the end-user side.
In short, these machines may be called "personal computers" and they
may be PCs in the sense of being used primarily by one user, but
contemporary data center machines have more in common with mainframes
and high-end servers than the original PCs, and consumer machines are
much closer architecturally to high end workstations than to
yesteryear's PCs.
My desktop machine is a Mac Studio with an ARM CPU; I call it a
workstation and I think that's pretty accurate. At work, one of our
EE's has a big x86 thing with some stupidly powerful graphics card
that he uses to do board layout. It's a workstation in every
recognizable sense, though it does happen to run Windows.
This may be getting into the weeds a bit but don't forget industrial hardware, the
stuff where you approach the blurry demarcation between CPU and MCU. [snip]
It wasn't just AT&T, IBM & DEC that got run over by commodity DRAM &
CPU's, it was the entire Minicomputer Industry, effectively extinct by
1995.
Same thing for the work-station industry (with Sun being merely the most
notable example). I have a tiny bit of second-hand personal knowldge in this
area; my wife works for NASA, as a structural engineer, and they run a lot of
large computerized mathematical models. In the 70's, they were using CDC
7600's; they moved along through various things as technology changed (IIRC,
at one point they had SGI machines). These days, they seem to mostly be using
high-end personal computers for this.
Some specialized uses (various forms of CAD) I guess still use things that
look like work-stations, but I expect they are stock personal computers
with special I/O (very large displays, etc).
So I guess now there are just supercomputers (themselves mostly built out of
large numbers of commodity CPUs), and laptops. Well, there is also cloud
computing, which is huge, but that also just uses lots of commodity CPUs. This for me is always the third class when I'm
discussing that sort of thing. Of course this also means "operating systems"
closer to standalone applications sitting on top of some microkernel like (se)L4, but you
did have VME-based workstations and UNIX versions specifically for VME systems. For me
these walk a line between true workstations and minicomputers, but as usual that is from
the perspective of having not been there. For the record, one of the canonical UNIX
development environments from AT&T for WE32x00 stuff was a WE32100 and support chips
thrown on a VME module running System V/VME. From what I know, VME is still quite common
in industrial control. How much UNIX and workalikes constitute the OS landscape in
today's VME ecosystem eludes me.
My sense is that a lot of industrial hardware these days follows the
same pattern as consumer devices, albeit often in a hardened chassis
or with slightly different peripherals that allow them to work in
environments with temperature extremes, poor airflow, and so on. I
don't get the sense that VME is still the dominant factor that it once
was in that space, but I'm not on a factory floor, either. Idly, I
wonder how many industrial systems are built from, say, a Raspberry Pi
compute module?
Perhaps a more interesting third dimension would be safety-critical
systems in automotive, aerospace, or medical applications. The PC in
my doctor's office is just a small desktop thing running Windows and a
bunch of software from EPIC, but the automagic blood pressure cuff
with the nifty graphical display the nurse uses to take my vitals is
something else entirely.
Either way, I feel like this is a class of computers
that frequently flies under the radar, but if we're talking strictly consumer stuff,
yeah VME very quickly loses relevance.
Interestingly, some of the earlier Sun machines were VME based. As I
recall, if you popped the hood off of a "pizzabox" Sun 3/50, there was
a VME SBC in there with some drive bays. In that regard, the
SPARCstation 1 paper is worth reading as an evolutionary marker.
- Dan C.
21 May
21 May
3:30 p.m.
New subject: Whither Workstations? (Was Re: Demise of AT&T)
On Tue, 20 May 2025 at 06:55, Dan Cross <crossd(a)gmail.com> wrote:
Interestingly, some of the earlier Sun machines were
VME based. As I
recall, if you popped the hood off of a "pizzabox" Sun 3/50, there was
a VME SBC in there with some drive bays. In that regard, the
SPARCstation 1 paper is worth reading as an evolutionary marker.
- Dan C.
I recall installing small bunches of Sun 3/50 or 3/60 workstations,
networked to a Sun 3/280 deskside server(-ish), for engineering offices and
tertiary training rooms (around 1986 or so?) -- all VME-based -- after Sun
switched from Multibus as used in the Sun 2/50 and 2/180(?) workstations.
IIRC some Sun 3's desksides may still have used Multibus, for some I/O
options (reel tape??) -- though I might be hallucinating that part...
channeling my inner AI...
- Stuart Remphrey.
3:59 p.m.
New subject: Whither Workstations? (Was Re: Demise of AT&T)
IIRC some Sun 3's desksides may still have used
Multibus, for some I/O
options (reel tape??) -- though I might be hallucinating that part...
channeling my inner AI...
Yes, there were a few Multibus devices using Multibus-to-VME adaptors;
the Xylogics 450/451 SMD controller were run this way.
27 May
27 May
2:52 p.m.
New subject: Whither Workstations? (Was Re: Demise of AT&T)
On 2025-05-19 16:54, Dan Cross wrote:
[...]
Perhaps a more interesting third dimension would be
safety-critical
systems in automotive, aerospace, or medical applications. The PC in
my doctor's office is just a small desktop thing running Windows and a
bunch of software from EPIC, but the automagic blood pressure cuff
with the nifty graphical display the nurse uses to take my vitals is
something else entirely.
Everyone forgets about embedded systems. When I was still noodling,
there were several RTOSes that were POSIX-certified (QNX and VxWorks,
amongst others). Of course, these ran on the higher end 32-bit MCUs, of
which dozens exist in modern cars. That medical stuff probably conforms
to IEC 62304, regardless of its internals.
S.
9:02 p.m.
New subject: Whither Workstations? (Was Re: Demise of AT&T)
<stuff(a)riddermarkfarm.ca> wrote:
Everyone forgets about embedded systems. When I was still noodling,
there were several RTOSes that were POSIX-certified (QNX and VxWorks,
amongst others). Of course, these ran on the higher end 32-bit MCUs, of
which dozens exist in modern cars. That medical stuff probably conforms
to IEC 62304, regardless of its internals.
S.
Ford's Sync 3 and Sync 4 infortainment systems run QNX but there's no hint of
the OS to users, just a fairly polished GUI. By contrast, some of the locomotives I work
on also run QNX but the entire boot process, starting with the command line, then briefly
a General Electric branded GUI is visible, before the end user GUI appears. QNX is
available, as a free of charge licenced edition for non-commercial use since January 2024.
I read Ford is to replace QNX with Android in coming generations of its Sync system.
Cameron Tyre
10:42 p.m.
New subject: Whither Workstations? (Was Re: Demise of AT&T)
On 28 May 2025, at 00:52, Stuff Received
<stuff(a)riddermarkfarm.ca> wrote:
Everyone forgets about embedded systems. When I was still noodling, there were several
RTOSes that were POSIX-certified (QNX and VxWorks, amongst others). Of course, these ran
on the higher end 32-bit MCUs, of which dozens exist in modern cars. That medical stuff
probably conforms to IEC 62304, regardless of its internals.
S.
related:
anyone on list know where all the ARM ‘CPUs’ (cores or multi-core chips?) get
used?
ARM, as the licenser, declared it licensed 250B “CPUs” in 2024.
We know 1-2B go into smartphones, perhaps another 250M into PC-like devices (250M is
approx PC market)
Where do the rest go?
I’ve read some HDD’s use ARM processors, so a few billion there perhaps.
--
Steve Jenkin, IT Systems and Design
0412 786 915 (+61 412 786 915)
PO Box 38, Kippax ACT 2615, AUSTRALIA
mailto:sjenkin@canb.auug.org.au http://members.tip.net.au/~sjenkin
11:03 p.m.
New subject: Whither Workstations? (Was Re: Demise of AT&T)
Am Mi., 28. Mai 2025 um 00:43 Uhr schrieb <sjenkin(a)canb.auug.org.au>:
We know 1-2B go into smartphones, perhaps another 250M
into PC-like devices (250M is approx PC market)
Where do the rest go?
Everywhere. E.g. every USB connector probably has some ARM cpu in it.
Btw: This has become OT.
Best
Martin
28 May
28 May
2:13 a.m.
New subject: Whither Workstations? (Was Re: Demise of AT&T)
There are tiny ARM processors in SD cards.
On 05/27/2025 03:42 PM, sjenkin(a)canb.auug.org.au wrote:
On 28 May 2025, at 00:52, Stuff Received
<stuff(a)riddermarkfarm.ca> wrote:
Everyone forgets about embedded systems. When I was still noodling,
there were several RTOSes that were POSIX-certified (QNX and VxWorks,
amongst others). Of course, these ran on the higher end 32-bit MCUs,
of which dozens exist in modern cars. That medical stuff probably
conforms to IEC 62304, regardless of its internals.
S.
related:
anyone on list know where all the ARM ‘CPUs’ (cores or multi-core
chips?) get used?
ARM, as the licenser, declared it licensed 250B “CPUs” in 2024.
We know 1-2B go into smartphones, perhaps another 250M into PC-like
devices (250M is approx PC market)
Where do the rest go?
I’ve read some HDD’s use ARM processors, so a few billion there perhaps.
--
Steve Jenkin, IT Systems and Design
0412 786 915 (+61 412 786 915)
PO Box 38, Kippax ACT 2615, AUSTRALIA
mailto:sjenkin@canb.auug.org.au http://members.tip.net.au/~sjenkin
2:16 a.m.
New subject: Whither Workstations? (Was Re: Demise of AT&T)
and thumb (USB) drives, and power banks, and probably anything you can
connect via USB, like keyboards
On 05/27/2025 07:13 PM, Luther Johnson wrote:
There are tiny ARM processors in SD cards.
On 05/27/2025 03:42 PM, sjenkin(a)canb.auug.org.au wrote:
On 28 May 2025, at 00:52, Stuff Received
<stuff(a)riddermarkfarm.ca>
wrote:
Everyone forgets about embedded systems. When I was still noodling,
there were several RTOSes that were POSIX-certified (QNX and
VxWorks, amongst others). Of course, these ran on the higher end
32-bit MCUs, of which dozens exist in modern cars. That medical
stuff probably conforms to IEC 62304, regardless of its internals.
S.
related:
anyone on list know where all the ARM ‘CPUs’ (cores or multi-core
chips?) get used?
ARM, as the licenser, declared it licensed 250B “CPUs” in 2024.
We know 1-2B go into smartphones, perhaps another 250M into PC-like
devices (250M is approx PC market)
Where do the rest go?
I’ve read some HDD’s use ARM processors, so a few billion there perhaps.
--
Steve Jenkin, IT Systems and Design
0412 786 915 (+61 412 786 915)
PO Box 38, Kippax ACT 2615, AUSTRALIA
mailto:sjenkin@canb.auug.org.au http://members.tip.net.au/~sjenkin
2:18 a.m.
New subject: Whither Workstations? (Was Re: Demise of AT&T)
and wireless devices, hot spots ...
On 05/27/2025 07:16 PM, Luther Johnson wrote:
and thumb (USB) drives, and power banks, and probably anything you can
connect via USB, like keyboards
On 05/27/2025 07:13 PM, Luther Johnson wrote:
There are tiny ARM processors in SD cards.
On 05/27/2025 03:42 PM, sjenkin(a)canb.auug.org.au wrote:
On 28 May 2025, at 00:52, Stuff Received
<stuff(a)riddermarkfarm.ca>
wrote:
Everyone forgets about embedded systems. When I was still
noodling, there were several RTOSes that were POSIX-certified (QNX
and VxWorks, amongst others). Of course, these ran on the higher
end 32-bit MCUs, of which dozens exist in modern cars. That
medical stuff probably conforms to IEC 62304, regardless of its
internals.
S.
related:
anyone on list know where all the ARM ‘CPUs’ (cores or multi-core
chips?) get used?
ARM, as the licenser, declared it licensed 250B “CPUs” in 2024.
We know 1-2B go into smartphones, perhaps another 250M into PC-like
devices (250M is approx PC market)
Where do the rest go?
I’ve read some HDD’s use ARM processors, so a few billion there perhaps.
--
Steve Jenkin, IT Systems and Design
0412 786 915 (+61 412 786 915)
PO Box 38, Kippax ACT 2615, AUSTRALIA
mailto:sjenkin@canb.auug.org.au http://members.tip.net.au/~sjenkin
2:19 a.m.
New subject: Whither Workstations? (Was Re: Demise of AT&T)
network switches, etc.
On 05/27/2025 07:18 PM, Luther Johnson wrote:
and wireless devices, hot spots ...
On 05/27/2025 07:16 PM, Luther Johnson wrote:
and thumb (USB) drives, and power banks, and probably anything you
can connect via USB, like keyboards
On 05/27/2025 07:13 PM, Luther Johnson wrote:
There are tiny ARM processors in SD cards.
On 05/27/2025 03:42 PM, sjenkin(a)canb.auug.org.au wrote:
> On 28 May 2025, at 00:52, Stuff Received <stuff(a)riddermarkfarm.ca>
> wrote:
>
> Everyone forgets about embedded systems. When I was still
> noodling, there were several RTOSes that were POSIX-certified (QNX
> and VxWorks, amongst others). Of course, these ran on the higher
> end 32-bit MCUs, of which dozens exist in modern cars. That
> medical stuff probably conforms to IEC 62304, regardless of its
> internals.
>
> S.
related:
anyone on list know where all the ARM ‘CPUs’ (cores or multi-core
chips?) get used?
ARM, as the licenser, declared it licensed 250B “CPUs” in 2024.
We know 1-2B go into smartphones, perhaps another 250M into PC-like
devices (250M is approx PC market)
Where do the rest go?
I’ve read some HDD’s use ARM processors, so a few billion there
perhaps.
--
Steve Jenkin, IT Systems and Design
0412 786 915 (+61 412 786 915)
PO Box 38, Kippax ACT 2615, AUSTRALIA
mailto:sjenkin@canb.auug.org.au http://members.tip.net.au/~sjenkin
2:23 a.m.
New subject: Whither Workstations? (Was Re: Demise of AT&T)
and coffee makers, microwave ovens, automobile engine controllers,
there's not much in the embedded world that is not ARM these days, of
course RISC V wants to make zero-licensing cost inroads there
On 05/27/2025 07:19 PM, Luther Johnson wrote:
network switches, etc.
On 05/27/2025 07:18 PM, Luther Johnson wrote:
and wireless devices, hot spots ...
On 05/27/2025 07:16 PM, Luther Johnson wrote:
and thumb (USB) drives, and power banks, and probably anything you
can connect via USB, like keyboards
On 05/27/2025 07:13 PM, Luther Johnson wrote:
There are tiny ARM processors in SD cards.
On 05/27/2025 03:42 PM, sjenkin(a)canb.auug.org.au wrote:
>
>
>> On 28 May 2025, at 00:52, Stuff Received
>> <stuff(a)riddermarkfarm.ca> wrote:
>>
>> Everyone forgets about embedded systems. When I was still
>> noodling, there were several RTOSes that were POSIX-certified
>> (QNX and VxWorks, amongst others). Of course, these ran on the
>> higher end 32-bit MCUs, of which dozens exist in modern cars.
>> That medical stuff probably conforms to IEC 62304, regardless of
>> its internals.
>>
>> S.
>
> related:
>
> anyone on list know where all the ARM ‘CPUs’ (cores or multi-core
> chips?) get used?
>
> ARM, as the licenser, declared it licensed 250B “CPUs” in 2024.
>
> We know 1-2B go into smartphones, perhaps another 250M into
> PC-like devices (250M is approx PC market)
>
> Where do the rest go?
>
> I’ve read some HDD’s use ARM processors, so a few billion there
> perhaps.
>
> --
> Steve Jenkin, IT Systems and Design
> 0412 786 915 (+61 412 786 915)
> PO Box 38, Kippax ACT 2615, AUSTRALIA
>
> mailto:sjenkin@canb.auug.org.au http://members.tip.net.au/~sjenkin
>
7:18 a.m.
New subject: Whither Workstations? (Was Re: Demise of AT&T)
COFF time for this as well.
Cheers, Warren
On 28 May 2025 12:23:23 pm AEST, Luther Johnson <luther.johnson(a)makerlisp.com>
wrote:
and coffee makers, microwave ovens, automobile engine
controllers, there's not much in the embedded world that is not ARM these days, of
course RISC V wants to make zero-licensing cost inroads there
On 05/27/2025 07:19 PM, Luther Johnson wrote:
--
Sent from my Android phone with K-9 Mail. Please excuse my brevity.
network switches, etc.
On 05/27/2025 07:18 PM, Luther Johnson wrote:
and wireless devices, hot spots ...
On 05/27/2025 07:16 PM, Luther Johnson wrote:
and thumb (USB) drives, and power banks, and probably anything you can connect via USB,
like keyboards
On 05/27/2025 07:13 PM, Luther Johnson wrote:
>
> There are tiny ARM processors in SD cards.
>
> On 05/27/2025 03:42 PM, sjenkin(a)canb.auug.org.au wrote:
>>
>>
>>> On 28 May 2025, at 00:52, Stuff Received <stuff(a)riddermarkfarm.ca>
wrote:
>>>
>>> Everyone forgets about embedded systems. When I was still noodling, there
were several RTOSes that were POSIX-certified (QNX and VxWorks, amongst others). Of
course, these ran on the higher end 32-bit MCUs, of which dozens exist in modern cars.
That medical stuff probably conforms to IEC 62304, regardless of its internals.
>>>
>>> S.
>>
>> related:
>>
>> anyone on list know where all the ARM ‘CPUs’ (cores or multi-core chips?) get
used?
>>
>> ARM, as the licenser, declared it licensed 250B “CPUs” in 2024.
>>
>> We know 1-2B go into smartphones, perhaps another 250M into PC-like devices (250M
is approx PC market)
>>
>> Where do the rest go?
>>
>> I’ve read some HDD’s use ARM processors, so a few billion there perhaps.
>>
>> --
>> Steve Jenkin, IT Systems and Design
>> 0412 786 915 (+61 412 786 915)
>> PO Box 38, Kippax ACT 2615, AUSTRALIA
>>
>> mailto:sjenkin@canb.auug.org.au http://members.tip.net.au/~sjenkin
>>
>
23 May
23 May
8:25 p.m.
New subject: Whither Workstations? (Was Re: Demise of AT&T)
On Mon, May 19, 2025 at 1:28 PM segaloco via TUHS <tuhs(a)tuhs.org> wrote:
On Monday, May 19th, 2025 at 12:45 PM, Dan Cross <crossd(a)gmail.com> wrote:
I would not be surprised if the VME board was used to prototype
datakit stuff. A major benefit is the eurocard form factor, which is
easy to build custom hardware around due to a shared ecosystem.
Physics labs commonly spin their own boards, you will find a lot of
VME docs online hosted by them. The VME design is relatively "clean"
and "simple" while still allowing for quite a bit of power, and it was
pumped over the years to 320MB/s while a center P0 connector could
hold any desired newer busses for unbounded interconnection. One area
it shines is shared memory multi-computers, nothing as popular
thereafter really encourages this way of systems design.
But the major "success" of the WE32k was the 3B2. Those sold in good
number. I don't think the WE VME product did from my own research.
As others have alluded to, VME was once popular in UNIX workstations.
Sun3 and early SGI are the popular examples but there were others like
Motorola and many smaller and niche players. Both Sun and SGI had
ways of bringing VME along once newer busses took over. For instance,
SGI is common in flight simulators and a VME crate would drive the
real time interfacing and control while the SGI orchestrated the
situation and graphics.
On Mon, May 19, 2025 at 12:36 PM Noel Chiappa
jnc(a)mercury.lcs.mit.edu wrote:
The CPU ISAs may be largely shared, but computing has bifurcated
into two distinct camps: those machines intended for use in
datacenters, and those intended for consumer use by end-users.
CPUs intended for datacenters tend to be characterized by large
caches, lower average clock speeds, wider IO and memory bandwidth.
Those intended for consumer use tend to have high clock speeds, a bit
less cache, and support for comparatively fewer IO devices/less
memory. On the end-user side, you've got a further split between
laptops/desktop machines and devices like phones, tablets, and so on.
In both cases, the dominant IO buses used are PCIe and its variants
(e.g., on the data center side you've got CXL), NVMe for storage is
common in lots of places, everything supports Ethernet more or less
(even WiFi uses the ethernet frame format), and so on. USB seems
ubiquitous for peripherals on the end-user side.
In short, these machines may be called "personal computers" and they
may be PCs in the sense of being used primarily by one user, but
contemporary data center machines have more in common with mainframes
and high-end servers than the original PCs, and consumer machines are
much closer architecturally to high end workstations than to
yesteryear's PCs.
My desktop machine is a Mac Studio with an ARM CPU; I call it a
workstation and I think that's pretty accurate. At work, one of our
EE's has a big x86 thing with some stupidly powerful graphics card
that he uses to do board layout. It's a workstation in every
recognizable sense, though it does happen to run Windows.
- Dan C.
This may be getting into the weeds a bit but don't forget industrial hardware, the
stuff where you approach the blurry demarcation between CPU and MCU. This for me is
always the third class when I'm discussing that sort of thing. Of course this also
means "operating systems" closer to standalone applications sitting on top of
some microkernel like (se)L4, but you did have VME-based workstations and UNIX versions
specifically for VME systems. For me these walk a line between true workstations and
minicomputers, but as usual that is from the perspective of having not been there. For
the record, one of the canonical UNIX development environments from AT&T for WE32x00
stuff was a WE32100 and support chips thrown on a VME module running System V/VME. From
what I know, VME is still quite common in industrial control. How much UNIX and
workalikes constitute the OS landscape in today's VME ecosystem eludes me. [snip]
It wasn't just AT&T, IBM & DEC that got run over by commodity DRAM &
CPU's, it was the entire Minicomputer Industry, effectively extinct by
1995.
Same thing for the work-station industry (with Sun being merely the most
notable example). I have a tiny bit of second-hand personal knowldge in this
area; my wife works for NASA, as a structural engineer, and they run a lot of
large computerized mathematical models. In the 70's, they were using CDC
7600's; they moved along through various things as technology changed (IIRC,
at one point they had SGI machines). These days, they seem to mostly be using
high-end personal computers for this.
Some specialized uses (various forms of CAD) I guess still use things that
look like work-stations, but I expect they are stock personal computers
with special I/O (very large displays, etc).
So I guess now there are just supercomputers (themselves mostly built out of
large numbers of commodity CPUs), and laptops. Well, there is also cloud
computing, which is huge, but that also just uses lots of commodity CPUs. Either way, I feel like this is a class of computers
that frequently flies under the radar, but if we're talking strictly consumer stuff,
yeah VME very quickly loses relevance.
VME is pervasive enough there are still "new" designs, although they
are dwindling as switched and high speed serial protocols have taken
over all aspects of system interconnect. VPX is the new one. But the
tail is long, any new or overhauled M1A2 Abrams tank is getting some
conduction cooled VME hardware for instance.
- Matt G.
24 May
24 May
9:32 a.m.
New subject: Whither Workstations? (Was Re: Demise of AT&T)
On Friday, May 23rd, 2025 at 1:26 PM, Kevin Bowling <kevin.bowling(a)kev009.com>
wrote:
On Mon, May 19, 2025 at 1:28 PM segaloco via TUHS
tuhs(a)tuhs.org wrote:
Well and to be frank the persistence of VME has had me taking more and more of an interest
lately as I have a long term interest of eventually finding employment in maintaining
legacy systems, and coupled with my current schemes with $dayjob, I'm hopeful
I'll be in the position to jump into something a little closer to older hardware in
the coming decades.
The WE321SB seemed to be largely marketed towards systems folks rather than as a
standalone VME solution. You wouldn't get a 321 card for day to day operations but
you would get a 321 card to throw in your VME crate and prototype stuff to get added to a
3B system of some kind. Peripheral dev, driver dev, that sort of thing. Again though I
wasn't there but that's the gist I've gotten from various studies over the
years.
Eventually I hope to secure one of the WE321 series bits, be it SB, EB, whichever, the
long term goal being that there is an electric history museum in town, I eventually want
to throw together an interactive installation of WECo computing stuff. There's an
old Bell telco office in town that has at least one 5ESS that, if I pull the installation
together, I also want to get some of the CenturyLink/Lumen folks in on it since
they're our local RBOC descendent and again, they're literally the old telco CO
in town and have at least one 5ESS going.
A guy can dream...
- Matt G.
On Monday, May 19th, 2025 at 12:45 PM, Dan Cross
crossd(a)gmail.com wrote:
I would not be surprised if the VME board was used to prototype
datakit stuff. A major benefit is the eurocard form factor, which is
easy to build custom hardware around due to a shared ecosystem.
Physics labs commonly spin their own boards, you will find a lot of
VME docs online hosted by them. The VME design is relatively "clean"
and "simple" while still allowing for quite a bit of power, and it was
pumped over the years to 320MB/s while a center P0 connector could
hold any desired newer busses for unbounded interconnection. One area
it shines is shared memory multi-computers, nothing as popular
thereafter really encourages this way of systems design.
But the major "success" of the WE32k was the 3B2. Those sold in good
number. I don't think the WE VME product did from my own research.
As others have alluded to, VME was once popular in UNIX workstations.
Sun3 and early SGI are the popular examples but there were others like
Motorola and many smaller and niche players. Both Sun and SGI had
ways of bringing VME along once newer busses took over. For instance,
SGI is common in flight simulators and a VME crate would drive the
real time interfacing and control while the SGI orchestrated the
situation and graphics.
On Mon, May 19, 2025 at 12:36 PM Noel Chiappa
jnc(a)mercury.lcs.mit.edu wrote:
This may be getting into the weeds a bit but don't forget industrial hardware, the
stuff where you approach the blurry demarcation between CPU and MCU. This for me is always
the third class when I'm discussing that sort of thing. Of course this also means
"operating systems" closer to standalone applications sitting on top of some
microkernel like (se)L4, but you did have VME-based workstations and UNIX versions
specifically for VME systems. For me these walk a line between true workstations and
minicomputers, but as usual that is from the perspective of having not been there. For the
record, one of the canonical UNIX development environments from AT&T for WE32x00 stuff
was a WE32100 and support chips thrown on a VME module running System V/VME. From what I
know, VME is still quite common in industrial control. How much UNIX and workalikes
constitute the OS landscape in today's VME ecosystem eludes me. > [snip]
> It wasn't just AT&T, IBM & DEC that got run over by commodity DRAM
&
> CPU's, it was the entire Minicomputer Industry, effectively extinct by
> 1995.
Same thing for the work-station industry (with Sun being merely the most
notable example). I have a tiny bit of second-hand personal knowldge in this
area; my wife works for NASA, as a structural engineer, and they run a lot of
large computerized mathematical models. In the 70's, they were using CDC
7600's; they moved along through various things as technology changed (IIRC,
at one point they had SGI machines). These days, they seem to mostly be using
high-end personal computers for this.
Some specialized uses (various forms of CAD) I guess still use things that
look like work-stations, but I expect they are stock personal computers
with special I/O (very large displays, etc).
So I guess now there are just supercomputers (themselves mostly built out of
large numbers of commodity CPUs), and laptops. Well, there is also cloud
computing, which is huge, but that also just uses lots of commodity CPUs.
The CPU ISAs may be largely shared, but computing has bifurcated
into two distinct camps: those machines intended for use in
datacenters, and those intended for consumer use by end-users.
CPUs intended for datacenters tend to be characterized by large
caches, lower average clock speeds, wider IO and memory bandwidth.
Those intended for consumer use tend to have high clock speeds, a bit
less cache, and support for comparatively fewer IO devices/less
memory. On the end-user side, you've got a further split between
laptops/desktop machines and devices like phones, tablets, and so on.
In both cases, the dominant IO buses used are PCIe and its variants
(e.g., on the data center side you've got CXL), NVMe for storage is
common in lots of places, everything supports Ethernet more or less
(even WiFi uses the ethernet frame format), and so on. USB seems
ubiquitous for peripherals on the end-user side.
In short, these machines may be called "personal computers" and they
may be PCs in the sense of being used primarily by one user, but
contemporary data center machines have more in common with mainframes
and high-end servers than the original PCs, and consumer machines are
much closer architecturally to high end workstations than to
yesteryear's PCs.
My desktop machine is a Mac Studio with an ARM CPU; I call it a
workstation and I think that's pretty accurate. At work, one of our
EE's has a big x86 thing with some stupidly powerful graphics card
that he uses to do board layout. It's a workstation in every
recognizable sense, though it does happen to run Windows.
- Dan C. Either way, I feel like this is a class of
computers that frequently flies under the radar, but if we're talking strictly
consumer stuff, yeah VME very quickly loses relevance.
VME is pervasive enough there are still "new" designs, although they
are dwindling as switched and high speed serial protocols have taken
over all aspects of system interconnect. VPX is the new one. But the
tail is long, any new or overhauled M1A2 Abrams tank is getting some
conduction cooled VME hardware for instance.
> - Matt G. 
26 May
26 May
11:26 p.m.
New subject: WE321 3B/VME boar (Re: Whither Workstations? (Was Re: Demise of AT&T))
At Sat, 24 May 2025 09:32:26 +0000, segaloco via TUHS <tuhs(a)tuhs.org> wrote:
Subject: [TUHS] Re: Whither Workstations? (Was Re: Demise of AT&T)
The WE321SB seemed to be largely marketed towards systems folks rather
than as a standalone VME solution. You wouldn't get a 321 card for
day to day operations but you would get a 321 card to throw in your
VME crate and prototype stuff to get added to a 3B system of some
kind. Peripheral dev, driver dev, that sort of thing. Again though I
wasn't there but that's the gist I've gotten from various studies over
the years.
I'm no expert but I don't think anyone would use the VME WE321 series
for prototyping anything targeted at AT&T 3B systems (other than maybe
applications software, but 3B2s were much cheaper and easier to manage
than a VME machine unless you already had a ton of spare VME boxes).
The smaller 3B2s (/300, /400, /500, /600, /700, /1000) were all very
PC-like motherboards with an I/O expansion bus that accommodated
"feature" cards, but that was not a VME bus. There was a design manual
that would have allowed third parties to develop I/O boards for the
3B2s, but I'm not sure any third-party boards were ever produced. (The
bigger systems also included expansion slots for more memory, cache, and
other things, but those were, I believe, entirely proprietary slots.)
VME was more suited, then and perhaps still, for industrial and
communications control systems where VME was already in use, but perhaps
with a different CPU. There was a companion manual for the WE321 cards
to help users develop UNIX System V drivers for other VME devices. I
think it was basically AT&T's attempt to sell UNIX into the VME market
directly (instead of relying on third parties to do it independently),
and because it was ABI-compatible with the 3B line it gave ISVs access
to a rather vast amount of commercial software that was already ported
to 3Bs.
--
Greg A. Woods <gwoods(a)acm.org>
Kelowna, BC +1 250 762-7675 RoboHack <woods(a)robohack.ca>
Planix, Inc. <woods(a)planix.com> Avoncote Farms <woods(a)avoncote.ca>
16
days inactive
25
days old
21 comments
16 participants
participants (16)
-
Cameron Míċeál Tyre -
Dan Cross -
Greg A. Woods -
jnc＠mercury.lcs.mit.edu -
Kevin Bowling -
Lars Brinkhoff -
Luther Johnson -
Martin Schröder -
Miod Vallat -
Paul Winalski -
ron minnich -
segaloco -
sjenkin＠canb.auug.org.au -
Stuart Remphrey -
Stuff Received -
Warren Toomey