Historical computers...

Title says it all
Message
Author
G-CPTN
Chief Pilot
Chief Pilot
Posts: 7593
Joined: Sun Aug 05, 2018 11:22 pm
Location: Tynedale
Gender:
Age: 79

Re: Historical computers...

#41 Post by G-CPTN » Sun Oct 18, 2020 4:49 pm

ian16th wrote:
Sun Oct 18, 2020 3:54 pm
Decca were the 1st people to offer me a job.
They tested my honesty by checking the price of a train ticket from Kings Lynn to their place in Surrey, and I claimed at a lower rate 'cos I travelled at the 'forces rate'.
The girl doing the checking let the cat out by saying 'But BR told us............' and as they had done the paper work, they gave me the full rate!!!
Sounds like they were paying expenses for your interview.

When I was long-term unemployed some potential employers didn't offer interview expenses - difficult . . .

Especially as a recent immigrant (albeit British and a taxpayer for 17 years) as I was 'absent' during the qualifying period the previous year I got nothing - zilch - not even family allowance.

User avatar
llondel
Chief Pilot
Chief Pilot
Posts: 5909
Joined: Wed Oct 03, 2018 3:17 am
Location: San Jose

Re: Historical computers...

#42 Post by llondel » Fri Oct 30, 2020 8:46 pm

There's a computer museum in Mountain View with a lot of history there. You can tell the way they must have ground their teeth when Colossus was declassified because it's given a small space that looks very much like an afterthought seeing as it wasn't a US invention but it was the first computer. They'd thought it was one of their own, ENIAC.

To be fair, ENIAC was the first general-purpose computer, but Colossus was definitely programmable, albeit for a limited range of tasks.

User avatar
Dushan
Capt
Capt
Posts: 1479
Joined: Sat Aug 22, 2015 8:23 pm
Location: Right wing
Gender:
Age: 71

Re: Historical computers...

#43 Post by Dushan » Thu Jan 14, 2021 10:08 pm

ExSp33db1rd wrote:
Fri Oct 16, 2020 11:25 pm
Abacus.

On a Hong Kong posting had a flight engineer who learned to use a Japanese style abacus, and used it to keep his flt. log going. We had to stop him, the constant clicking of the keys kept us awake on long night sectors !

Mrs. ExS. was an early programmer, using something called Cobol (?) She said that because of the small amount of memory available they deliberately reduced the depiction of the year date to just the last two figures, knowing that at 00.01 on 01/01/2000 the world would end in confusion, but assumed that before then "someone" would have worked out how to avoid this. We can all remember the panic that set in as that date and time approached, and the eventual non-event. She was contacted to go back and help old colleagues re-draft all the programmes that she had helped create, but refused on the grounds that she had forgotten most of by the algorithms by then, probably not, and a pity that she didn't, she could have commended a small fortune to assist.
Except that it was a no-event event is because logic was added to treat all years greater than 39 as 1940 - 1999 and all years 00 to 39 as 2000 - 2039. Guess when we are going to be called again to fix the mess?

Yeah, I'll hopefully still remember how to write the loop instruction

PERFORM 0000-MAIN VARYING SUB FORM 1 BY 1 UNTIL SUB > 100.
Because they stand on the wall and say "nothing's gonna hurt you tonight, not on my watch".

User avatar
Dushan
Capt
Capt
Posts: 1479
Joined: Sat Aug 22, 2015 8:23 pm
Location: Right wing
Gender:
Age: 71

Re: Historical computers...

#44 Post by Dushan » Thu Jan 14, 2021 10:15 pm

ian16th wrote:
Sat Oct 17, 2020 4:02 pm
unifoxos wrote:
Sat Oct 17, 2020 3:55 pm
When the IBM S/360 was new in 1965 core memory in the UK was £1 a byte.

I remember being sent to Perkins Engines to add some core memory to their S/360 mainframe about 1967. They freed up the machine in prime shift for this, treated me like royalty and dined me in the directors canteen after I'd done the job. It was a major upgrade for them - IIRC it doubled the memory capacity - 8K to 16K!
I think it must have been a Mod 20 as I think a 30 was at least 16K.
Our S/360-30 had 64K and an additional 32 by Massachusetts Core which was a separate cabinet, in the corner of the raised floor computer room. The cabinet, the size of a refrigerator, was connected to the CPU with a thick cable plugged in where the original IBM core (I don't know what this 'memory thing is you speak of :-)) would have been at 10X the cost.
Because they stand on the wall and say "nothing's gonna hurt you tonight, not on my watch".

User avatar
llondel
Chief Pilot
Chief Pilot
Posts: 5909
Joined: Wed Oct 03, 2018 3:17 am
Location: San Jose

Re: Historical computers...

#45 Post by llondel » Fri Jan 15, 2021 1:16 am

Dushan wrote:
Thu Jan 14, 2021 10:08 pm
ExSp33db1rd wrote:
Fri Oct 16, 2020 11:25 pm
Abacus.

On a Hong Kong posting had a flight engineer who learned to use a Japanese style abacus, and used it to keep his flt. log going. We had to stop him, the constant clicking of the keys kept us awake on long night sectors !

Mrs. ExS. was an early programmer, using something called Cobol (?) She said that because of the small amount of memory available they deliberately reduced the depiction of the year date to just the last two figures, knowing that at 00.01 on 01/01/2000 the world would end in confusion, but assumed that before then "someone" would have worked out how to avoid this. We can all remember the panic that set in as that date and time approached, and the eventual non-event. She was contacted to go back and help old colleagues re-draft all the programmes that she had helped create, but refused on the grounds that she had forgotten most of by the algorithms by then, probably not, and a pity that she didn't, she could have commended a small fortune to assist.
Except that it was a no-event event is because logic was added to treat all years greater than 39 as 1940 - 1999 and all years 00 to 39 as 2000 - 2039. Guess when we are going to be called again to fix the mess?

Yeah, I'll hopefully still remember how to write the loop instruction

PERFORM 0000-MAIN VARYING SUB FORM 1 BY 1 UNTIL SUB > 100.
I would hope that it was done by adding a couple of constants for base century and transition offset, so that any time between now and the end of 2039, someone just needs to adjust them so that they'll be good for the rest of the 21st century. We've been in this one for 20 years now, so they could at least bump the transition offset up to 60 from 40, push the problem out to 2059.

User avatar
ian16th
Chief Pilot
Chief Pilot
Posts: 10029
Joined: Fri Aug 28, 2015 9:35 am
Location: KZN South Coast with the bananas
Gender:
Age: 87

Re: Historical computers...

#46 Post by ian16th » Fri Jan 15, 2021 10:26 am

Dushan wrote:
Thu Jan 14, 2021 10:15 pm
ian16th wrote:
Sat Oct 17, 2020 4:02 pm
unifoxos wrote:
Sat Oct 17, 2020 3:55 pm
When the IBM S/360 was new in 1965 core memory in the UK was £1 a byte.

I remember being sent to Perkins Engines to add some core memory to their S/360 mainframe about 1967. They freed up the machine in prime shift for this, treated me like royalty and dined me in the directors canteen after I'd done the job. It was a major upgrade for them - IIRC it doubled the memory capacity - 8K to 16K!
I think it must have been a Mod 20 as I think a 30 was at least 16K.
Our S/360-30 had 64K and an additional 32 by Massachusetts Core which was a separate cabinet, in the corner of the raised floor computer room. The cabinet, the size of a refrigerator, was connected to the CPU with a thick cable plugged in where the original IBM core (I don't know what this 'memory thing is you speak of :-)) would have been at 10X the cost.
Seeing as the S360/30 only had 16 bit addressing, I'd love to know how this 'extra memory' was addressed!

Maybe it was an external device operating on the Selector Channel and addressed as a fast virtual disk.

The IBM 1800 DACS used the same generation technology and had 32K of 16bit words, with one bit being a sign bit and another was used as a Storage Protect bit.
This was 'frigged' worked around to address 64K by having 32K of negative addresses.
Cynicism improves with age

User avatar
Dushan
Capt
Capt
Posts: 1479
Joined: Sat Aug 22, 2015 8:23 pm
Location: Right wing
Gender:
Age: 71

Re: Historical computers...

#47 Post by Dushan » Fri Jan 15, 2021 2:42 pm

It was definitely additional core, but I don't know how it was handled. I was the operator, at the time.
Later we got a new CPU, model 40, with 128k of core so we could have bigger programs.
Because they stand on the wall and say "nothing's gonna hurt you tonight, not on my watch".

User avatar
ian16th
Chief Pilot
Chief Pilot
Posts: 10029
Joined: Fri Aug 28, 2015 9:35 am
Location: KZN South Coast with the bananas
Gender:
Age: 87

Re: Historical computers...

#48 Post by ian16th » Fri Jan 15, 2021 4:30 pm

Dushan wrote:
Fri Jan 15, 2021 2:42 pm
It was definitely additional core, but I don't know how it was handled. I was the operator, at the time.
Later we got a new CPU, model 40, with 128k of core so we could have bigger programs.
On the /40 you would see that instead of 4 rotary switches, there was 18 toggle switches for addressing storage.

The 18 switches gave the ability to address 256KB of memory. The main frame could only hold 128k and if the customer wanted 192KB or 256KB it went in a large box at right angles to the main frame.

The four rotary switches on the /30 could only address up to 'FFFF' or 64KB.

The biggest operational difference was that the /40 addressed/accessed 2 bytes at a time, where the /30 only addressed/accessed 1 byte at a time.
Cynicism improves with age

User avatar
probes
Chief Pilot
Chief Pilot
Posts: 2843
Joined: Sun Aug 23, 2015 9:01 pm
Location: 'Urop
Gender:

Re: Historical computers...

#49 Post by probes » Fri Jan 15, 2021 4:37 pm

Mis-read: Hysterical computers...

User avatar
Dushan
Capt
Capt
Posts: 1479
Joined: Sat Aug 22, 2015 8:23 pm
Location: Right wing
Gender:
Age: 71

Re: Historical computers...

#50 Post by Dushan » Fri Jan 15, 2021 9:23 pm

ian16th wrote:
Fri Jan 15, 2021 4:30 pm
Dushan wrote:
Fri Jan 15, 2021 2:42 pm
It was definitely additional core, but I don't know how it was handled. I was the operator, at the time.
Later we got a new CPU, model 40, with 128k of core so we could have bigger programs.
On the /40 you would see that instead of 4 rotary switches, there was 18 toggle switches for addressing storage.

The 18 switches gave the ability to address 256KB of memory. The main frame could only hold 128k and if the customer wanted 192KB or 256KB it went in a large box at right angles to the main frame.

The four rotary switches on the /30 could only address up to 'FFFF' or 64KB.

The biggest operational difference was that the /40 addressed/accessed 2 bytes at a time, where the /30 only addressed/accessed 1 byte at a time.
But, but, but...
I Googled it and Wiki says https://en.wikipedia.org/wiki/IBM_System/360_Model_30

"96K upgrade

In response to competitive pressures, IBM introduced a memory upgrade option, allowing 96K on a 360/30.[6] It seems, based on the system's front panel, that a provision for supporting more than 64K had been pre-planned.[7] "


So it looks like what we had was indeed possible, and supplied by a 3rd party supplier.
I definitely remember a box about 2' x 2' x 2' sitting in the corner of the computer room (not next to the CPU) and being connected by a thick cable to the CPU near the place where the transparent boxes holding the core were.
Because they stand on the wall and say "nothing's gonna hurt you tonight, not on my watch".

User avatar
TheGreenGoblin
Chief Pilot
Chief Pilot
Posts: 17596
Joined: Thu Aug 08, 2019 11:02 pm
Location: With the Water People near Trappist-1

Re: Historical computers...

#51 Post by TheGreenGoblin » Sun Jan 17, 2021 11:15 am

Dushan wrote:
Fri Jan 15, 2021 9:23 pm
ian16th wrote:
Fri Jan 15, 2021 4:30 pm
Dushan wrote:
Fri Jan 15, 2021 2:42 pm
It was definitely additional core, but I don't know how it was handled. I was the operator, at the time.
Later we got a new CPU, model 40, with 128k of core so we could have bigger programs.

On the /40 you would see that instead of 4 rotary switches, there was 18 toggle switches for addressing storage.

The 18 switches gave the ability to address 256KB of memory. The main frame could only hold 128k and if the customer wanted 192KB or 256KB it went in a large box at right angles to the main frame.

The four rotary switches on the /30 could only address up to 'FFFF' or 64KB.

The biggest operational difference was that the /40 addressed/accessed 2 bytes at a time, where the /30 only addressed/accessed 1 byte at a time.
But, but, but...
I Googled it and Wiki says https://en.wikipedia.org/wiki/IBM_System/360_Model_30

"96K upgrade

In response to competitive pressures, IBM introduced a memory upgrade option, allowing 96K on a 360/30.[6] It seems, based on the system's front panel, that a provision for supporting more than 64K had been pre-planned.[7] "


So it looks like what we had was indeed possible, and supplied by a 3rd party supplier.
I definitely remember a box about 2' x 2' x 2' sitting in the corner of the computer room (not next to the CPU) and being connected by a thick cable to the CPU near the place where the transparent boxes holding the core were.
Or a Model 67...

https://en.wikipedia.org/wiki/IBM_System/360_Model_67
The IBM System/360 Model 67 (S/360-67) was an important IBM mainframe model in the late 1960s. Unlike the rest of the S/360 series, it included features to facilitate time-sharing applications, notably a Dynamic Address Translation unit, the "DAT box", to support virtual memory, 32-bit addressing and the 2846 Channel Controller to allow sharing channels between processors. The S/360-67 was otherwise compatible with the rest of the S/360 series.

The S/360-67 design included a radical new component for implementing virtual memory, the "DAT box" (Dynamic Address Translation box). DAT on the 360/67 was based on the architecture outlined in a 1966 JACM paper by Arden, Galler, Westervelt, and O'Brien and included both segment and page tables. The Model 67's virtual memory support was very similar to the virtual memory support that eventually became standard on the entire System/370 line.

The S/360-67 provided a 24- or 32-bit address space – unlike the strictly 24-bit address space of other S/360 and early S/370 systems, and the 31-bit address space of S/370-XA available on later S/370s. The S/360-67 virtual address space was divided into pages (of 4096 bytes) grouped into segments (of 1 million bytes); pages were dynamically mapped onto the processor's real memory. These S/360-67 features plus reference and change bits as part of the storage key enabled operating systems to implement demand paging: referencing a page that was not in memory caused a page fault, which in turn could be intercepted and processed by an operating system interrupt handler.

The S/360-67's virtual memory system was capable of meeting three distinct goals:

Large address space. It mapped physical memory onto a larger pool of virtual memory, which could be dynamically swapped in and out of real memory as needed from random-access storage (typically: disk or drum storage).

Isolated OS components. It made it possible to remove most of the operating system's memory footprint from the user's environment, thereby increasing the memory available for application use, and reducing the risk of applications intruding into or corrupting operating system data and programs.

Multiple address spaces. By implementing multiple virtual address spaces, each for a different user, each user could potentially have a private virtual machine.

The first goal removed (for decades, at least) a crushing limitation of earlier machines: running out of physical storage. The second enabled substantial improvements in security and reliability. The third enabled the implementation of true virtual machines. Contemporary documents make it clear that full hardware virtualization and virtual machines were not original design goals for the S/360-67.
Though you remain
Convinced
"To be alive
You must have somewhere
To go
Your destination remains
Elusive."

User avatar
TheGreenGoblin
Chief Pilot
Chief Pilot
Posts: 17596
Joined: Thu Aug 08, 2019 11:02 pm
Location: With the Water People near Trappist-1

Re: Historical computers...

#52 Post by TheGreenGoblin » Sun Jan 17, 2021 11:32 am

Does anybody here remember the MS-DOS Extended Memory tricks you could play using the the HIMEM.SYS driver when using the the 80286 Intel CPU that allowed access of memory above the DOS 640 KB limit giving 2^20/1024, or 1024 KB addressable space? The hack allowed the mapping of a 4 bit 'virtual' register to the usual 16 bits to effectively give one a 20 bit address space although some executables would complain if one tried to push executable code into high memory..

Oh how the time simply flew by when having such fun! =))
Though you remain
Convinced
"To be alive
You must have somewhere
To go
Your destination remains
Elusive."

User avatar
unifoxos
Capt
Capt
Posts: 959
Joined: Mon Aug 31, 2015 10:36 am
Location: Twycross Zoo, or thereabouts
Gender:
Age: 78

Re: Historical computers...

#53 Post by unifoxos » Sun Jan 17, 2021 11:55 am

Seem to remember trying to get drivers to operate in that extended memory space to leave more space for user programs in the space below 640K. It was rarely successful.
Sent from my tatty old Windoze PC.

User avatar
TheGreenGoblin
Chief Pilot
Chief Pilot
Posts: 17596
Joined: Thu Aug 08, 2019 11:02 pm
Location: With the Water People near Trappist-1

Re: Historical computers...

#54 Post by TheGreenGoblin » Sun Jan 17, 2021 11:58 am

unifoxos wrote:
Sun Jan 17, 2021 11:55 am
Seem to remember trying to get drivers to operate in that extended memory space to leave more space for user programs in the space below 640K. It was rarely successful.
Yes, exactly, it was a bit hit and miss but, as you say, good if you could free the space below 640 KB. We were so short of addressable space in those days although tell that to the kids these days and they won't believe you. ;)))
Though you remain
Convinced
"To be alive
You must have somewhere
To go
Your destination remains
Elusive."

k3k3
Capt
Capt
Posts: 1492
Joined: Thu Sep 10, 2015 9:44 pm
Location: In the Transit Lounge

Re: Historical computers...

#55 Post by k3k3 » Sun Jan 17, 2021 12:40 pm

The mere thought of the 640Kb limits brings on nightmares and headaches, just not sure in which order.

User avatar
ian16th
Chief Pilot
Chief Pilot
Posts: 10029
Joined: Fri Aug 28, 2015 9:35 am
Location: KZN South Coast with the bananas
Gender:
Age: 87

Re: Historical computers...

#56 Post by ian16th » Sun Jan 17, 2021 2:07 pm

16 bit mini-computers used 'Segmentation Registers' to address blocks/partitions of memory above the directly addressable memory above 64KB.

The 'DAT Box' and introduction of virtual storage is a different thing. It appears to the user that you have unlimited real memory at the performance cost of the time to page data to disk and back again.

Addressing storage you don't have, caused an exception check, then an 'error routine' did what was necessary to page data out of memory to disk, and page the required data into real memory.
Cynicism improves with age

User avatar
TheGreenGoblin
Chief Pilot
Chief Pilot
Posts: 17596
Joined: Thu Aug 08, 2019 11:02 pm
Location: With the Water People near Trappist-1

Re: Historical computers...

#57 Post by TheGreenGoblin » Sun Jan 17, 2021 2:11 pm

ian16th wrote:
Sun Jan 17, 2021 2:07 pm
16 bit mini-computers used 'Segmentation Registers' to address blocks/partitions of memory above the directly addressable memory above 64KB.

The 'DAT Box' and introduction of virtual storage is a different thing. It appears to the user that you have unlimited real memory at the performance cost of the time to page data to disk and back again.

Addressing storage you don't have, caused an exception check, then an 'error routine' did what was necessary to page data out of memory to disk, and page the required data into real memory.
Ah, an invitation to memory paging, dish thrashing and glacial performance! ;)))

And here I sit today with a 64 bit operating system and 32 GB of RAM on a laptop! More than we had on the IBM System 38 mini computer at the electronics factory I worked for that had 25 terminals!
Though you remain
Convinced
"To be alive
You must have somewhere
To go
Your destination remains
Elusive."

User avatar
ian16th
Chief Pilot
Chief Pilot
Posts: 10029
Joined: Fri Aug 28, 2015 9:35 am
Location: KZN South Coast with the bananas
Gender:
Age: 87

Re: Historical computers...

#58 Post by ian16th » Sun Jan 17, 2021 2:58 pm

TheGreenGoblin wrote:
Sun Jan 17, 2021 2:11 pm
ian16th wrote:
Sun Jan 17, 2021 2:07 pm
16 bit mini-computers used 'Segmentation Registers' to address blocks/partitions of memory above the directly addressable memory above 64KB.

The 'DAT Box' and introduction of virtual storage is a different thing. It appears to the user that you have unlimited real memory at the performance cost of the time to page data to disk and back again.

Addressing storage you don't have, caused an exception check, then an 'error routine' did what was necessary to page data out of memory to disk, and page the required data into real memory.
Ah, an invitation to memory paging, dish thrashing and glacial performance! ;)))

And here I sit today with a 64 bit operating system and 32 GB of RAM on a laptop! More than we had on the IBM System 38 mini computer at the electronics factory I worked for that had 25 terminals!
Some people didn't listen to those words I've highlighted.

Then they complained.
Cynicism improves with age

User avatar
TheGreenGoblin
Chief Pilot
Chief Pilot
Posts: 17596
Joined: Thu Aug 08, 2019 11:02 pm
Location: With the Water People near Trappist-1

Re: Historical computers...

#59 Post by TheGreenGoblin » Sun Jan 17, 2021 6:36 pm

TheGreenGoblin wrote:
Sun Jan 17, 2021 2:11 pm

Ah, an invitation to memory paging, dish thrashing and glacial performance! ;)))
A Greek restaurant perhaps... :-?
Though you remain
Convinced
"To be alive
You must have somewhere
To go
Your destination remains
Elusive."

User avatar
ExSp33db1rd
Chief Pilot
Chief Pilot
Posts: 3229
Joined: Sat Sep 12, 2015 1:51 am
Location: Lesser Antipode
Gender:
Age: 89

Re: Historical computers...

#60 Post by ExSp33db1rd » Sun Jan 17, 2021 8:59 pm

You worry about computers ? Said it before, but I carry a 2" copy of the reverse side of the E6B Navigation Computer in my wallet, a one-time keyring novelty. Using it at the Supermarket check out recently the teenage clerk said " Wot's that " A circular slide rule I answered. " Wot's a slide rule ?" was his reply. One could weep.

Post Reply