Army Rumour Service

This is a sample guest message. Register a free account today to become a member! Once signed in, you'll be able to participate on this site by adding your own topics and posts, as well as connect with other members through your own private inbox!

Something From History You Probably Never Knew...

Helm

MIA
Moderator
Book Reviewer
As were some British ships If I recall. Victory herself was dis-masted. I'm not exactly sure how that would have played out viz insurance and lost prize money, but there was money somewhere I'll be bound. But it's also worth bearing in mind that in the ruckus after the battle, people would have looked after themselves- probably not the official version. Also long servers would have either deferred or stashed their prize money. Look on it as a sort of OAP.

where there any? Even coach builts had a record of a chassis number. I mean you could be going back to pre Great war
Vin replaced the old system of chassis numbers with a standardised format.
 
As were some British ships If I recall. Victory herself was dis-masted. I'm not exactly sure how that would have played out viz insurance and lost prize money, but there was money somewhere I'll be bound. But it's also worth bearing in mind that in the ruckus after the battle, people would have looked after themselves- probably not the official version. Also long servers would have either deferred or stashed their prize money. Look on it as a sort of OAP.

where there any? Even coach builts had a record of a chassis number. I mean you could be going back to pre Great war
The loss of the prizes must have been a sore disappointment to the sailors, that after all being what they fought for. Whatever Horatio might have said to inspire them, I doubt it was thoughts of duty or England that fired them up (about one third of the crews were Irish so that would hardly have done the trick) it was more probably the thought of all that juicy money at the end of it, and having fought so valiantly and with such phenomenal courage, to see it all sink before they got it home must have been heartbreaking.
 

AlienFTM

MIA
Book Reviewer
Amazing system design, the Police National Computer or PNC, for criminal and DVLA records. Dunno when it started but at least since 1976 to date it is still a black screen with green text and a blinking cursor, and was designed to use the Julian calendar which had to be converted by the operator to the conventional.
Not unlike the RAPC Computer Centre, Worthy Down, late 80s. We converted to decimal currency in 1971, but I was told that certain aspects of the Pay system didn't get converted from £SD until shortly before I was posted there. Pay Statement print runs were converted to output decimal immediately before printing. Being in Manning myself, I might call bullsh¡t, but frankly, it wouldn't be entirely unbelievable.

I've told before about the negative half-word date issue. "32 binary bits (half a mainframe system word) can hold a number in excess of 4 billion. More than enough to store every day from 1 January 1900 (before any date required on the system) until the dim and distant future.

What didn't occur to the designers in the early 60s was that the leading but indicated a +/- number and 31 bits only held 32,768 (32Kibi - 1) days, and the leading bit would increment to 1 in Sep 1989. Think Millennium bug).

As people's run-out dates started to appear beyond September 89, the Pay Statement and Record of Service prints produced garbage, or dates going backwards from 1 January 1900. The value held was valid on the record without the negative indicator. Solution was that whenever a date was calculated, call a module that loaded the date into a full-word field (64 bits), do the sum, store the result locally.

They were still finding seldom-used sub-routines to correct when I joined in 86.

Didn't help that when designing the Date system, it was assumed that 1900 was a leap year. Any date between 1 January and 28 February 1900 was a day out. Mercifully next to zero instances, and ignored.
 
Last edited:
Not unlike the RAPC Computer Centre, Worthy Down, late 80s. We converted to decimal currency in 1971, but I was told that certain aspects of the Pay system didn't get converted from £SD until shortly before I was posted there. Pay Statement print runs were converted to output decimal immediately before printing. Being in Manning myself, I might call bullsh¡t, but frankly, it wouldn't be entirely unbelievable.

I've told before about the negative half-word date issue. "16 binary bits (half a mainframe system word) can hold a number in excess of 4 billion. More than enough to store every day from 1 January 1900 (before any date required on the system) until the dim and distant future.

What didn't occur to the designers in the early 60s was that the leading but indicated a +/- number and 15 bits only held 32,768 (32Kibi - 1) days, and the leading but would increment to 1 in Sep 1989. Think Millennium bug).

As people's run-out dates started to appear beyond September 89, the Pay Statement and Record of Service prints produced garbage, or dates going back from 1 January 1900. The value held was valid on the record without the negative indicator. Solution was that whenever a date was calculated, call a module that loaded the date into a full-word field (32 bits), do the sum, store the result locally.

They were still finding seldom-used sub-routines to correct when I joined in 86.

Didn't help that when designing the Date system, it was assumed that 1900 was a leap year. Any date between 1 January and 28 February 1900 was a day out. Mercifully next to zero instances, and ignored.
How can 16 bits hold a number of value in excess of 4 billion? You even say yourself that 15 bits only has 32K states.
 

HSF

LE
That's entirely possible. Ships captured at Trafalgar, were sold as prize and strange though it may seem the entire crews had shares in their prizemoney. That saved the treasury the cost and time of having a ship built as did the provision of either the cannon or the metal for recasting of new cannon. Their rigging cost a few bob too if they could be salvaged. I have no doubt that the treasury were rubbing their hand with glee upon news of the victory because indirectly that led to Napoleon's blockade of 1808- he didn't have the ships. Ships at prize money and their accoutrements were free of any tariffs according to the 1791 tariffs. That's probably why Nelson is so remembered, he added to the kings fleet massively. But as the the French say "C'est la Guerre" :D
I thought that the 20 or so prize ships were retaken & or lost during the massive post Trafalgar stom?Some of the older buidings along that coastline have been found to contain old ships timbers probably from the battle
 
No it was a certain item used it, it'll come to me eventually probably about 3am

JMH (Joint Message Handling). “Julian Filing Time”. If I recall correctly, it was the sequence number in the year. So February 1st would be 32. December 30th would be 364 (in a non-leap year). IIRC it was the Julian date followed by the time stamp (Z). Used to calculate processing times for the various precedences.

As I say, IF I recall correctly, that is. It’s been a long while now.
 

Helm

MIA
Moderator
Book Reviewer
JMH (Joint Message Handling). “Julian Filing Time”. If I recall correctly, it was the sequence number in the year. So February 1st would be 32. December 30th would be 364 (in a non-leap year). IIRC it was the Julian date followed by the time stamp (Z). Used to calculate processing times for the various precedences.

As I say, IF I recall correctly, that is. It’s been a long while now.
No this was a police thing.
 

AlienFTM

MIA
Book Reviewer
How can 16 bits hold a number of value in excess of 4 billion? You even say yourself that 15 bits only has 32K states.
Typo. I'm old. That's my excuse and I'm sticking to it. Good job you were paying attention to my deliberate mistake. Cough.

Edit. Corrected. Big bytes on mainframes in the day. Nowadays even Windows uses 64bit bytes (=half-word).
 
Last edited:
Not unlike the RAPC Computer Centre, Worthy Down, late 80s. We converted to decimal currency in 1971, but I was told that certain aspects of the Pay system didn't get converted from £SD until shortly before I was posted there. Pay Statement print runs were converted to output decimal immediately before printing. Being in Manning myself, I might call bullsh¡t, but frankly, it wouldn't be entirely unbelievable.

I've told before about the negative half-word date issue. "16 binary bits (half a mainframe system word) can hold a number in excess of 4 billion. More than enough to store every day from 1 January 1900 (before any date required on the system) until the dim and distant future.

What didn't occur to the designers in the early 60s was that the leading but indicated a +/- number and 15 bits only held 32,768 (32Kibi - 1) days, and the leading but would increment to 1 in Sep 1989. Think Millennium bug).

As people's run-out dates started to appear beyond September 89, the Pay Statement and Record of Service prints produced garbage, or dates going back from 1 January 1900. The value held was valid on the record without the negative indicator. Solution was that whenever a date was calculated, call a module that loaded the date into a full-word field (32 bits), do the sum, store the result locally.

They were still finding seldom-used sub-routines to correct when I joined in 86.

Didn't help that when designing the Date system, it was assumed that 1900 was a leap year. Any date between 1 January and 28 February 1900 was a day out. Mercifully next to zero instances, and ignored.
Sounds like you are just the guy for me to inquire about my 'credits'.:cool:
 
No this was a police thing.
Two types of dates for message handling.

ACP - 127

DTG (Date Time Group) - given by the originator of the message. Date time (in Zulu) month year etc 261550Z Sep 20. This was his reference for the message and allowed the recipients to see when the message was drafted.

Julian date was, indeed, used for internal processing times. JFT was given to the message to show the time of receipt at the message processing centre. That was then used to show it's progress through the message handling system, depending on precedence (RR, PP, OO, ZZ) and time of receipt.

Strict ITT (Internal transit times) were laid down for how long it could take for a message to go from originating commcen to destination commcen.

Of course, these went out of the window during exercises such as Wintex, but they were there and adhered to, on pain of getting a right bollocking if things were delayed. Indeed, the one thing you never wanted was an "inordinate delay" investigation. Tech charge time!

 

AlienFTM

MIA
Book Reviewer
Sounds like you are just the guy for me to inquire about my 'credits'.:cool:
30 years ago. A colleague (we did the same Programmer course) followed me to IBM, fronted up one day, we chatted, RAPC binned the mainframes right after I left.

But credits. I used to do the sums the Worthy Down Way ( "every other way is wrong"). But I used the tables in my training literature, not whatever the latest chancellor has changed Tax, National Insurance Disincrements* etc to.

It didn't matter what I did, or how I did it, the answer was always ⅔ gross pay. So I did that.
_____
* RAPC theorized that you didn't contribute NI, they just took it from you. And because your money went down, not up, they were the opposite of increments, so it appeared on your pay statement as Nat Ins D.
 
Yes, sorry for my distraction and that's enough of the Julian calendar, can we go back to Viking slave raids on the south coast pre yr-1066 now? Got some interesting stuff about one of their raiding ship's crews being found in a mass grave in Dorset.
 

Latest Threads

Top