TOPS

[Michael Hodgson]

On 28/11/2023 at 15:33, Neil Urquhart said:

 

Who will be the first to add an OO gauge model of a mainframe (complete with blinking lights) to the interior of a railway office?  

 

 

It needs whirring tape drives too.

[uax6]

On 29/11/2023 at 18:59, jamie92208 said:

 

 

The thing  that amazed me then in the 1990's was that the front end system was almost identical to the Police National Computer (PNC), that is another very old system that is still running at the heart of modern peripherals and feeding many other systems.  It dates from the same era..  I believe that programmers for both systems have to be specially trained so that they can write acceptable code. 

 

 

 

Jamie

 

Also very similar to BT's CSS (customer service system, I think it stands for) system, and being familiar with that was a help when I first came into the signalbox where I got to play with TRUST. 

From a signalmans point of view, there is a woeful lack of training provided for the use of TRUST (actually its worse than that, you aren't given any at all!), so the only computer system that you have for seeing what trains are coming (there's no printed Working Timetable now), you have no training on at all, you have to pick up the bits that others in the box have learnt over time. So in effect you only really get to know a tiny amount of what is available to you.

 

Andy G

 

Edit: Don't forget that BT didn't come into existance until 1980, it being the GPO telephone department until 1969, then Post Office Telephones.

 

Edited December 2, 2023 by uax6

[Jeremy Cumberland]

8 hours ago, rodent279 said:

Something troubles me slightly.

 

We're told that TOPS, contrary to popular belief, had no 5 digit limit on loco numbers, and that Western and Hymek 4 digit numbers were entered in TOPS as they were.

 

We're also told that TOPS couldn't handle loco classes starting at 0, hence there was no 40000 for example,  D200 becoming 40122.

 

How then did TOPS handle D1000 & D7000?

I never worked with TOPS, but "number" appears to be an ordinary text field, so it can be anything you like within a particular character set, and subject to a maximum length.

 

I very much doubt there was any bar to beginning series with 0, just that it is not a natural thing to do. I don't know of any railway that began a number sequence from 0, and when class prefixes were introduced, it was obvious to begin the class number sequence from 1.

 

The first member of a class having a number ending in 0 only happens when the number sequences doesn't start from 1, but from something like 2000. People who number railway locomotives are no different from the general population, most of whom would probably say that "2000" was the first number of the two-thousands rather than the last number of the one-thousands. Nevertheless, even here you find sequences beginning with 1, such as the AC electric locomotives (E2001, E3001 and E3301, not E2000, E3000 and E3300).

[Oldddudders]

There was definitely a sudden aversion to loco numbers ending in three noughts, but I can't recall exactly when. So E5000 became E5024, I think. 

[rodent279]

None of which explains how (if?) TOPS handled D1000 & D7000? Or in fact class 08 D3000 had it lasted beyond 1972.

Edit: D7000 withdrawn 07/73, D1000 02/74, (both according to wnxx) so were they ever entered in TOPS?

Edited December 2, 2023 by rodent279

[rodent279]

29 minutes ago, Oldddudders said:

So E5000 became E5024, I think. 

According to Wackypedia, fwiw, E5000 was renumbered E5024 as early as Dec 1962, long before TOPS became a thing.

[Michael Hodgson]

2 hours ago, Jeremy Cumberland said:

The first member of a class having a number ending in 0 only happens when the number sequences doesn't start from 1, but from something like 2000. People who number railway locomotives are no different from the general population, most of whom would probably say that "2000" was the first number of the two-thousands rather than the last number of the one-thousands. Nevertheless, even here you find sequences beginning with 1, such as the AC electric locomotives (E2001, E3001 and E3301, not E2000, E3000 and E3300).

As a programmer I remember having to explain to my boss what was wrong with his specification of the validation of a Time field of 1-24 hours and 1-60 minutes!

[rodent279]

If you are simply numbering as a block, i.e. Westerns were from 1000 up, Warships from 800 up, then starting at 0 works. It's when you consider the first 2 digits as a class designator, the rest as unique identifiers, then ending in 0 doesn't make sense, as that makes the first one the zero'th, 001 the 2nd etc.

Edited December 2, 2023 by rodent279

[rodent279]

2 hours ago, Jeremy Cumberland said:

The first member of a class having a number ending in 0 only happens when the number sequences doesn't start from 1, but from something like 2000. People who number railway locomotives are no different from the general population, most of whom would probably say that "2000" was the first number of the two-thousands rather than the last number of the one-thousands. Nevertheless, even here you find sequences beginning with 1, such as the AC electric locomotives (E2001, E3001 and E3301, not E2000, E3000 and E3300).

I think that is why 47500 works, but 47000 doesn't. Why though was there no 47300 or 47400, but there was 31400?

[Michael Hodgson]

6 minutes ago, rodent279 said:

If you are simply numbering as a block, i.e. Westerns were from 1000 up, Warships from 800 up, then starting at 0 works. It's when you consider the first 2 digits as a class designator, the rest as unique identifiers, then ending in 0 doesn't make sense, as that makes the first one the zero'th, 001 the 2nd etc.

Why didn't BR have a loco no. D0 or a class zero?  Or was there a prototype that escaped my notice?

[iands]

1 hour ago, rodent279 said:

None of which explains how (if?) TOPS handled D1000 & D7000? Or in fact class 08 D3000 had it lasted beyond 1972.

Edit: D7000 withdrawn 07/73, D1000 02/74, (both according to wnxx) so were they ever entered in TOPS?

But weren't the Westerns and Hymeks designated class 52 and class 35 respectively? So, if they were ever in TOPS, they could have been 5200x/3500x with the first of class being renumbered to the last. 

It is strange that TOPS couldn't (allegedly) handle 5-digit loco numbers ending '000', yet it could quite happily cope with 5-digit TOPS location codes ending '000'. One example being 15000 Seal Sands. 

[jamie92208]

25 minutes ago, iands said:

But weren't the Westerns and Hymeks designated class 52 and class 35 respectively? So, if they were ever in TOPS, they could have been 5200x/3500x with the first of class being renumbered to the last. 

It is strange that TOPS couldn't (allegedly) handle 5-digit loco numbers ending '000', yet it could quite happily cope with 5-digit TOPS location codes ending '000'. One example being 15000 Seal Sands. 

As a spotter who typed up numbers of new Tops numbers for a friend, as far as I know the Westerns and Hymeks were never given TOPS numbers but IIRC D9000 became 55022. I suspect that in the original coding there were 2 separate fields, 2 numerals for the class eg.  47' and then 3 numerals for the serial number.  000 Wouldn't work but 300 would.  I can't remember what D200 became but it certainly wasn't 40000.  I have a vague memory that the 000 loco often got a number from the first member of the class that had been withdrawn. 

 

Jamie

Edited December 2, 2023 by jamie92208

[rodent279]

26 minutes ago, jamie92208 said:

As a spotter who typed up numbers of new Tops numbers for a friend, as far as I know the Westerns and Hymeks were never given TOPS numbers but IIRC D9000 became 55022. I suspect that in the original coding there were 2 separate fields, 2 numerals for the class eg.  47' and then 3 numerals for the serial number.  000 Wouldn't work but 300 would.  I can't remember what D200 became but it certainly wasn't 40000.  I have a vague memory that the 000 loco often got a number from the first member of the class that had been withdrawn. 

 

Jamie

D9000 did indeed become 55022, similarly D200 became 40122, to replace D322 that was written off before renumbering. Had that not happened, D200 would have been 40200.

There was a 47500, and a 47200, also a 31400, but not 47300. Which clearly shows that TOPS could handle 0 being the last digit, but not for the first of a class. Hence no 47000, 55000.

Edited December 2, 2023 by rodent279

[Peter Kazmierczak]

When BR started their second diesel numbering system (the first was the 5-digit series starting with10000) with the D prefix, they tended to start with even numbers, except for D1 of course. So for the Type 4s we got:

D1
D200
D400

D600
D800
D1000


 

[rodent279]

23 minutes ago, jamie92208 said:

Westerns and Hymeks were never given TOPS numbers

They were in TOPS, as 1000-1073, 7000-7100, but the numbers were not applied to the locos 

[rodent279]

50 minutes ago, iands said:

But weren't the Westerns and Hymeks designated class 52 and class 35 respectively? So, if they were ever in TOPS, they could have been 5200x/3500x with the first of class being renumbered to the last. 

It is strange that TOPS couldn't (allegedly) handle 5-digit loco numbers ending '000', yet it could quite happily cope with 5-digit TOPS location codes ending '000'. One example being 15000 Seal Sands. 

And, as I say above, it could handle the likes of 47500 & 31400.

[Peter Kazmierczak]

I'm sure we've had this TOPS discussion before, but I can't recaIl the outcome...

As a totally non-IT person, it does kind of imply that there were two fields for the loco number; the class and the individual vehicle identifer. 000 in a field would suggest it was empty. Hence 45000 woul'nt be possible, but 45001 or 45100 would. Also rembember that most shunters began their class number with a 0, which also sugests that all digits in the field had to be filled. So 00 001 wouldn't be possible, but 01 001 would be.

Or am I barking up the wrong tree?

[rodent279]

1 hour ago, Peter Kazmierczak said:

I'm sure we've had this TOPS discussion before, but I can't recaIl the outcome...

As a totally non-IT person, it does kind of imply that there were two fields for the loco number; the class and the individual vehicle identifer. 000 in a field would suggest it was empty. Hence 45000 woul'nt be possible, but 45001 or 45100 would. Also rembember that most shunters began their class number with a 0, which also sugests that all digits in the field had to be filled. So 00 001 wouldn't be possible, but 01 001 would be.

Or am I barking up the wrong tree?

I think you're not far wrong.

That or sometime made an arbitrary decision that the numerically first of each class (as opposed to sub-class) would start with 001.

[Simon Lee]

The loco line in TOPS was / is  known as the 06 line, for the simple reason it starts 06.

 

The original 06 line was "split" in the first field to cater for non renumbered locos, thus imputs by the TOPS clerk would be for example as follows 

 

06 D 1000

 

06 ; 47101 the skip character replaced the D

 

once the 4 digit locos were withdrawn the field was amended/updated removing the need for the skip during one of the regular Saturday night update outages. One regular feature was the updates/amendments to the system were promogulated by a "ZZ" all terminals message on the previous Thursday.

 

There is, in, one of the late Adrian Curtis's books a photo of a TOPS printout featuring a Western a former colleague on the DOver office, had dealt with Westerns during his time at Hither Green TOPS. 

[Jeremy Cumberland]

4 hours ago, rodent279 said:

Why though was there no 47300 or 47400, but there was 31400?

31400 was used because the 31/4s had two different types of ETH equipment. One type was fitted to 31401-25 and another type was fitted to 31426 onwards. When 31401 sustained accident damage in 1988, 31161 received its early-type ETH equipment, and it was thought better to number it in the "up to 31425" sequence rather than 31470 in the "31426 onwards" sequence. Since it was not the practice to re-use previous numbers, and there was no gap between the two sequences, 31161 received the number 31400.

 

There was a 47300. Again it was a late addition, converted from the withdrawn 47468 in 1992 to replace the accident-damaged 47343. I have no idea why 47300 was chosen rather than 47382.

 

 

Does anyone have any evidence that TOPS couldn't handle 01000, for example? From what I've seen of how TOPS manages loco numbers, I can't see why such a number couldn't be used. Computers, in general, don't care where a number sequence begins, but beginning from 1 makes life a little easier for humans, and I expect it was a human rather than a computer that decided number sequences within a class should begin with 001.

[jamie92208]

Maybe someone who understands how TOPS is programmed can help.  I suspect that there are two sets of rules in play.

 

The first is what data is allowed to be entered to create a record.  I suspect that when a loco record is created a 2 digit class number has to be entered in the range 01 to 99.  This is followed by a separate field with a serial number in the range 001 to 999.   That gets the data in. Those rules will be hard coded into tne database and

The record will be validated before the record can be created.

 

When a search is made both fields will be combined and displayed together.   The search field would probably have different rules. 

 

When Loco TOPS numbers were applied there was often a gap between the class number and the serial number. 

 

Jamie

[kevinlms]

On 02/12/2023 at 22:22, Peter Kazmierczak said:

When BR started their second diesel numbering system (the first was the 5-digit series starting with10000) with the D prefix, they tended to start with even numbers, except for D1 of course. So for the Type 4s we got:

D1
D200
D400

D600
D800
D1000


 

But that was always the case with steam locos too, belonging to companies. No one had a loco numbered 0, but most had a 1.

Other classes started at ##0 etc and were class leaders. The GWR classes were often referenced like 57XX, but the numbers started at 5700.

[Neil Urquhart]

As far as I know, the 5 digit loco numbers (known as TOPs numbers) were not acually required by TOPS, but were part of a related system (RAVERS?) which was implemented to deal with loco maintainance at around the same time as TOPS was rolled out.

 

I think that TOPS could cope with a loco number being anthing, as long it was unique (not sure if it could cope with letters for the D & E prefixes).  As far as I am aware TOPS does have the concept of grouping locos into classes (with the same technical info for braking and horsepower). I think (and I will try and verify this) that locos were allocated to classes (e.g. 37,47,52) in TOPS irrespective of their actual running number.

 

Neil

[jamie92208]

23 minutes ago, Neil Urquhart said:

As far as I know, the 5 digit loco numbers (known as TOPs numbers) were not acually required by TOPS, but were part of a related system (RAVERS?) which was implemented to deal with loco maintainance at around the same time as TOPS was rolled out.

 

I think that TOPS could cope with a loco number being anthing, as long it was unique (not sure if it could cope with letters for the D & E prefixes).  As far as I am aware TOPS does have the concept of grouping locos into classes (with the same technical info for braking and horsepower). I think (and I will try and verify this) that locos were allocated to classes (e.g. 37,47,52) in TOPS irrespective of their actual running number.

 

Neil

All class 47's for instance had 47 as the prefix and then a serial number between 001 and 999.  However they were sorted so that 001 to 299 were IIRC, Seam heat fitted Locos.  300 to 399 had no heating and we're freight only.  400 upwards had Electric heating.  601 and later 901 were the same loco used to test the engines for the 56 and 58's. From memory a loco record showed the maintenance intervals,when the next exam was due and kept track of the engine hours. 

 

Jamie

Edited December 4, 2023 by jamie92208

[The Stationmaster]

On 02/12/2023 at 07:00, Jeremy Cumberland said:

I never worked with TOPS, but "number" appears to be an ordinary text field, so it can be anything you like within a particular character set, and subject to a maximum length.

 

I very much doubt there was any bar to beginning series with 0, just that it is not a natural thing to do. I don't know of any railway that began a number sequence from 0, and when class prefixes were introduced, it was obvious to begin the class number sequence from 1.

 

The first member of a class having a number ending in 0 only happens when the number sequences doesn't start from 1, but from something like 2000. People who number railway locomotives are no different from the general population, most of whom would probably say that "2000" was the first number of the two-thousands rather than the last number of the one-thousands. Nevertheless, even here you find sequences beginning with 1, such as the AC electric locomotives (E2001, E3001 and E3301, not E2000, E3000 and E3300).

TOPS was of course prgrammed for US operation so basically it could accept 4 (or fewer) digit loco numbers although it might have been able to accept a n umber preceded by an X if a loco was running an Extra (train).  Initially the loco number field was completely free format with only the limit of the number of characters the field would accept

 

It had to be re-programed to deal with BR class identifier numbers so at first the data identified running numbers to their relevant class identifier number so that it produced the correct information for Brake Force.  The system then required further programme changes to achieve two things - firstly to deal with the renumbering of locos to a class identifier led running number and secondly to be able to carry out the various load and brake force calculations required to save a human having to do that job (it could add up train weights from inception as that had been needed in the SP application.

 

I'm not sure what was done at which time for the latter two but I think that what was done on the system at some stage was a 'translation' in each loco file between the existing running number and the class identifier led running number so that the system would recognise the loco whichever of the two numbers was input.  Certainly right from the start it could say what class a loco was when you input its existing running number and that facility was obviously left in the system for as long as it was needed.

Edited December 4, 2023 by The Stationmaster