Train describers (DAISY) and 3 unit

[Deleted]

While the DLR continues to run a mixture of 2 and 3 unit stock, why don't they show on the DAISY which is expected? I got a 3 unit trait out of East India yesterday - not common, but it meant it went further forward on the platform.

OK, perhaps it hasn't been designed for this - but being advised somehow would make things easier.

[rincew1nd]

MerseyRail do this, they also include train lenghts in their online timetables.

[mrfs42]

Could someone expand DAISY for a bear of little brain, please? I'm guessing the D is possibly Digital meaning it could be Digital And Illuminated Signs Yah!

[Deleted]

Docklands Arrivals Information SYstem

[mrfs42]

Aha! TVM. Is that an acronym that has evolved or was it used from introduction?

[Deleted]

Berlin U-Bahn (and now some of the bus and tram stops) also have DAISY - Dynamic Advice and Information System (DAISY). I think it translates to something different in German.

[davidp]

Here in Amsterdam, my Metro route (50) is supposed to be 3-car during the peak and daytime but quite often is short-formed. On platform signs, after the train has left the preceding stop, the estimated arrival time of the train is replaced by a pictogram of a two- or three-car train as appropriate, allowing people to move to the appropriate place on the platforms. I've often thought that this would be useful on the DLR.

[Deleted]

Oct 19, 2010 9:21:28 GMT mrfs42 said:

Is that an acronym that has evolved or was it used from introduction?

I wish I knew - I just googled "DLR DAISY" ;D

[mrfs42]

Oct 19, 2010 10:41:32 GMT futurix said:

Oct 19, 2010 9:21:28 GMT mrfs42 said:

Is that an acronym that has evolved or was it used from introduction?

I wish I knew - I just googled "DLR DAISY" ;D

Hmm, that's cheating! I think Reg's answer fits best, the Docklands part is a back-formation IYSWIM.

[Deleted]

Found it - the BVG's "DAISY" stands for -

Dynamisches Auskunfts- und Informationssystem

Although there are many people who tend to be sceptical about "how long is a BVG minute", I have always found to be most reliable. It also goes beyond the normal next times often found here, in that on low-ebb services (e.g. 20 min intervals) it would show the next two trains as, for example, 17 mins and 37 mins.

Daisy's a cute little lady ;D ;D

[Deleted]

Oslo's system for its Metro (called SIS - translates to Real Time Information System) manages to do different numbers of cars - trains are coupled to permanent three car sets but can be run up to four three car sets I believe. Not that there are sufficiently long platforms anywhere though!

farm4.static.flickr.com/3458/3264881596_cb5ea51216.jpg?v=0

This image is of a six car train - shown as two cars on the SIS screen. The next one due (over 15 minutes shows as a clock time based on time table, traffic level and normal running time all put together in one number I believe!) is a three car train.

(I fail to see how this is relevant though!)

[Tomcakes]

I do remember that the DLR system has been called DAISY for many years - about 10 years ago they put the system online so you could view from your PC when the next train was. Novel in its time!

[Deleted]

I do believe the DLR also had something called WANDA however i'm not sure if this was another information system.

[mrfs42]

Oct 19, 2010 21:47:07 GMT @jimar53 said:

I do believe the DLR also had something called WANDA however i'm not sure if this was another information system.

ISTR that was a solar heating project in Stuttgart. Different DLR, perhaps.

WANDA was first-generation WAP information for the Docklands.

[jswallow]

WANDA was the system that provided similar arrival and departure data by station, but to WAP-enabled mobile phones.

[thirstquensher]

Yes, and it never worked, because all the mobile operators involved with enabling WAP content cached all pages, meaning you only ever saw a static snapshot of each station's data from several hours prior to callup - i.e. at 2pm, 3pm and 4pm you'd see the same page stating "Time is now 08:17".

And no, that wasn't my phone's browser doing the caching, because each station's page would be an old one, for the first time I had ever viewed it.

Thankfully, WAP was quickly seen as the useless white elephant it really was, and smartphones with proper browsers that deliver mobile-centric content in standard HTML rather than WML (the WAP equivalent) renders it obsolete.

In general it seems that DLR have no ability to fundamentally amend the behaviour of the DAISY displays, it seems. I'm of the solid opinion that it can be done, but I suspect programmers who know how to make such changes can't be found. I think it would require flashing EPROMs on board each display unit to give them updated decoding algorithms for the control data sent to each one.

This is why, for instance, you've never seen any variations on the 'top line, big font, bottom 2 lines in small font' vs. 'two lines in
big font' arrangements. If there was more flexibility, we may well have seen scenarios where one line of big text is used for Next Train 1, one line of small text is used for Next Train 2, and a third line, again in small text, is used for the type of scrolling messages that currently take up two big lines, blanking out all Next Train details.

[bowchurch]

Oct 19, 2010 10:34:51 GMT davidp said:

Here in Amsterdam, my Metro route (50) is supposed to be 3-car during the peak and daytime but quite often is short-formed. On platform signs, after the train has left the preceding stop, the estimated arrival time of the train is replaced by a pictogram of a two- or three-car train as appropriate, allowing people to move to the appropriate place on the platforms. I've often thought that this would be useful on the DLR.

Thanks, always did wonder why a pictures of trains appeared on the departure boards in Amsterdam. Obviously not spent long enough standing at Waterlooplein and Station RAI to work it out for myself!

[bowchurch]

Jan 20, 2011 21:59:09 GMT @thirstquensher said:

In general it seems that DLR have no ability to fundamentally amend the behaviour of the DAISY displays, it seems. I'm of the solid opinion that it can be done, but I suspect programmers who know how to make such changes can't be found. I think it would require flashing EPROMs on board each display unit to give them updated decoding algorithms for the control data sent to each one.

I would expect the boards to have all the effects available you could possibly want, they just do what they are told. I've been working on an Infotec LED installation recently and have been pleasently surprised by what their control protocol can do. I'm sure I could get the boards scrolling Arabic in the correct direction if I needed to.

It depends where the formatting codes get inserted as to how easy they are to change. If complete strings including commands are sent out centrally then the changes only need to happen in the one place. If any kind of local intermediate decoding happens then it gets more difficult depending on how it has been designed. If the board is working in a very dumb way, such as being driven by a binary set of GPI contacts rather than serial data, then a local progamming change will be required. But if the boards are similar to the Infotec boards then you just upload the new macros over the serial link or TCP socket remotely and the board writes them to flash, no ladders are required!

[thirstquensher]

I've uploaded a rather rudimentary 'simulation' of how such messages currently look, and how I think they ought to look:


DLR Display Options by Zone7, on Flickr

[bowchurch]

Right, see what you are trying to do now. I'd be surprised if the hardware allows only a small region of each line to be scrolled. I don't know the specifics of the hardware, but it might be possible without the clock - as long as the line controller hardware can scroll half lines, but that's also unlikely. The boards I've programmed don't have any hardware support for half lines, I'd have to fudge it using some small custom fonts and overtype mode if I ever needed it. Even then it would be restricted to static displays unless both lines scrolled at once.

You would also reduce the distance the sign is readable from. If you are standing two and a half car lengths away it's easy to spot if a half line reads an expected destination like Canary Wharf or Lewisham, an unfamiliar scrolling message would be more of a challenge to read.

[thirstquensher]

The way I understand it, the DAISY displays work by scanning from bottom-to-top, both 'big lines' together - i.e. as the lowermost pixel line of the top 'big line' is being written to the LEDs, so is the lowermost pixel line of the bottom 'big line'. I don't know about anyone else, but I can see the scanning quite clearly and I see them scanning this way, both lines in sync with each other. I have always supposed that the reasoning for the scanning working this way is so that scrolling text slopes to the left instead of sloping to the right, as it would if scanning top-to-bottom and also, dividing the scanning area into two segments is so that the required freshrate could be maintained using the technology of the day. Presumably if the whole display field was being scanned in one go, too much flicker would result in timing the scans in line with the desired scrolling speed for scrolling messages. To get the scroll speed right, remember, the scanning/refresh rate has to match, so that the message can be moved along one pixel for every refresh. Otherwise, you end up with a very jerky, jagged effect on scrolling messages, which can be seen on the first-generation JLE DMIs and indeed some of the B09 stock's internal displays which appear to be wrongly configured (and do two screen refreshes per pixel scroll).

Scrolling the second line of small font shouldn't, I would have thought, be a problem, because all that needs to happen is that the second line is nudged along by one pixel every refresh. The refresh rate would be the same, and the scan pattern would be the same - except that during the refresh of the lower 'big line' (the lower half), the refresh is drawing half of it ("NO LIFT SERVICE") in the same position as the last field, and drawing the other half of it (the scrolling part) one pixel to the left. The resultant scrolling effect would still be as smooth as if it were a big line.

That said, I understand about the readability issue, but then given the frequency of display of these messages, I'd say that partly counteracts that issue, which is the original point I'm making - they appear far too often.

Alternatively, you could have the scrolling message taking up the top line in big font, then show Next Trains 1 and 2 in small font beneath - again, not ideal in terms of readability, but better than not showing the Next Trains *at all* (as now).

Or perhaps, DLR could resist the temptation to cram so many damn scrolling messages in in the first place! Lifts out of service, granted. Closures next week, granted. Closures in two weeks' time, closures on 5 or 6 LUL lines, and the fact that CCTV is monitored by the BTP, a bit overboard I think. The latter category of information is given out over the PA system verbally almost every 90 seconds now, so I think visual confirmation is unnecessary. OK, so there's the DDA to consider, but does that create a legal requirement to turn Next Train Indicators into constant subtitling systems for every single announcement uttered?

[Chris M]

The frequent changes of display also probably create a DDA issue for those with eye-sight problems.
It seems that many people implementing measures to help one set of disabled people don't think about whether it will have any impact (adverse or otherwise) on people with a different disability. This can be seen in many modern bus designs where the large area for wheelchairs and buggies means that there are few or no seats in reach of people who can walk but only for short distances.

[pib]

Just for the sake of correctness, the platform displays are PIDs, Daisies are the monitor based displays usually on concourses.

[thirstquensher]

Well, to be truly pedantic, no term is more right or wrong than another, given that between the TfL modes over the years, all of the following terms have been used to refer to the same device:

NTI = Next Train Indicator
DMI = Dot Matrix Indicator
CIS = Customer Information System
PID = Passenger Information Display
VID = Visual Information Display
VEID = Visual Electronic Information Display

etc.