Topic: Sleeping cars and their relative values and liveries.

[Spenk009]

Hi all, I've had a bit of time on my hands and decided to try to get a better understanding of the workflow and improve my Blender skills. The result is a set of Mk1 and Mk3 sleeping carriages, where the Mk3s are available in all liveries present for its base. The Mk3 is based on the Mk3A first class open carriage (FO), with the window sill raised, all windows moved and original colours applied in their respective places. I'm quite happy with the result. For the Mk1, I've modified the second class corridor carriage (SK), moving the central door, adding the window pairs. This carriage looks too short in my opinion, due the bogies sitting far outside underneath.

I have updated my Blends repo found here. In due course I'll add some .dat files and create a fork for people to try out in the source repo.

I'd like to hear your thoughts on the carriages, probably best seen in the .blend files. What carriages aside from the Midland 12 wheel carriages were used prior to these, and were the exteriors vastly different to the other carriages they were like? Also, input on where to find good data aside from Wikipedia is very much appreciated.

[jamespetts]

Interesting work. I did wonder about adding sleeping carriages; they would, I think, have to be rated for a very long journey to be comfortable to make sense, and also have a higher first cost and higher maintenance/staff cost. One issue with adding modern sleeping carriages is that the pakset then becomes unbalanced if these are not available in earlier eras, and adding accurate sleeping cars for all earlier (and later) eras will be a very substantial undertaking.

[Spenk009]

If implemented it would be a significant change to the balance of large maps. I would suggest we make them unwieldy (boarding times 60-180, can only be coupled to restaurant and brake vans, 1.6x purchase and 3x fixed costs) and see how they compare. One of the main issues would be very long journeys before the 20th century, which could see a surge in popularity as the comfort values increase.

My main question was whether the requirements for adding accurate sleeping cars in the cases of the Mk1s (updated to all liveries) and Mk3s are satisfied. These liveries can't be correct, but which ones should be kept? I've opened a branch on my repo, called sleeping-carriages, for anyone to check out.

[jamespetts]

I should note that, at present, comfort does not affect the popularity of a route, only the revenue, as this would require a very major code change.

As to the liveries, I do not really understand the question, as both of your links just point to your base repository. For reference, Mk. 1 sleepers appeared, I believe, in (possibly) crimson/cream, maroon and blue/grey (we might also add Inter-City livery for players who want to use these longer than British Rail did). Mk. 3 sleepers appeared in blue/grey, Inter-City executive, Intercity swallow livery, Great Western Trains livery, the various First Great Western liveries, the (21st century) Great Western Railway livery, and various Caledonian sleeper liveries.
As noted, adding sleeping cars to some but not all eras would be fantastically confusing, and the history of sleeping carriages dates to the 1830s and is highly complex, so would take a great amount of work to implement fully.

[Spenk009]

As noted, adding sleeping cars to some but not all eras would be fantastically confusing, and the history of sleeping carriages dates to the 1830s and is highly complex, so would take a great amount of work to implement fully.

I'll just see about adding things as I go along then. It becomes very difficult to find information online.

I am hoping to keep working on further conversions of vehicles into the transparent workflow instead. Pull requests should surface in time. I'm getting the hang of the workflow, it's straightforward and well documented.

[jamespetts]

I'll just see about adding things as I go along then. It becomes very difficult to find information online.

I do have a number of books with references for these if you are interested in going back through time.

I am hoping to keep working on further conversions of vehicles into the transparent workflow instead. Pull requests should surface in time. I'm getting the hang of the workflow, it's straightforward and well documented.

That would be extremely helpful - thank you.

[Spenk009]

For the North British Railway sleeping carriage unveiled in 1873, I have created a carriage from drawings found here and here. To suit liveries, I based the colour scheme on carriage pictured at the (url=https://nbrstudygroup.co.uk/nbr/carriages.php)NBR study group(/url). Does the following statement apply? "East Coast Joint Stock was varnished teak, whilst Midland & North British Joint Stock followed the Midland's coach livery of crimson lake." ((url=https://www.lner.info/co/NBR/livery.php)link(/url))  I'd say that they would try to stick in line with Midland on this one for the long trip to London, but input would be appreciated.

[jamespetts]

This is interesting.

I would suggest that the North-Eastern Railway had a darker shade of red than did the Midland. Compare this North Eastern Railway carriage:

Victorian/Edwardian Railway Carriages of NER by Terry Pinnegar, on Flickr

with these Midland railway carriages:

Midland Railway composite carriage by Burnie van Hilst, on Flickr

Midland 6-wheel composite by James Petts, on Flickr

See also this picture.

[Spenk009]

I see, it's far darker than I took from the NBR example found. Please excuse my lack of knowledge, but were coaches of the joint East Coast Stock (generally) painted to match each other? Also, the pakset contains no carriages by the North British Railway, have these been amalgated into other coaches for simplicity?

I've used the red (6A0705) found on the gresley express. It's darker than the original, but blends in better.
I currently have limited the lms period 1 sleeping coaches to couple to other period 1 and catering vehicles. Even if these vehicles won't find their way into the pakset, would you agree with the practice? Also done on later models, it's intended to force the player into using catered vehicles to complete the rake.

[Freahk]

Don't know if I got it right.
I don't think we should force players to put catering into their trains.
If a real world railway company wants to start a long distance sleeping train without catering, they will just do it. Passengers will complain about it for sure, however, they are not forced to use this train.

In extended catering will add some profit and comfort to long distance trains, so players will use catering for sleeping car trains. If the comfort level is not high enough due to the mising catering, passengers will still use the train but won't pay that much due to a lack in comfort.

If someone is about to say "I need a little more capacity and the additional profit and comfort is not that important to me", let them do it!

Also, I would not even contraint sleeping cars to be only coupled to other sleeping cars.
Again, if a real world railway company decided it would be a good idea to add carriages with normal seats to the train, they will just do it.
In fact most night train in Germany are served by ÖBB (Austrian Federal Railways) under the nightjet label. These do always have three different main classes: sleepers, couchettes and seated, sometimes even bikes, cars or motorcycles can be transported, which is, at least currently, not of interest for simutrans extended.

So why should we artificially constraint coupling of sleepers, not allowing coupling to seated cars?
On most very long lines trains will have intermediate stops so it's just fine to catch some lower class passengers at these stops that would otherwise need at line on their own or could not travel at all. Additionally, some of the higher class pessengers will then use these seats, so we get more free capacity in sleepers for passengers that need the high comfort for their long route.

One could also argue the other way round: If we have a very long route with relatively small towns at the ends, I would not start a sleeping car service in between these cities because it is very unlikely to operate profitably. However, extending existing seated lines to these cities can be profitable due to lower operational costs and thus lower passenger classes being profitable.
However, attaching a single sleeper to such trains could be profitable for the few higher class passengers travelling the long line comfortably.

At least in extended, we should only constraint coupling if it is "technically impossible". At least this is my point of view.
However, I am aware that "technically impossible" is not pretty precise. One can have endlsess discussions about this, especially in later years where not only the the coupling itself but also software plays a huge role.

[Spenk009]

Don't know if I got it right.

Yes you did. I've amended the other coaches that allow coupling to Period 1 coaches and hope to find time soon to do the same to the later coaches.

If it is possible, it should be doable. The economic simulation of the game has to be the driving force for realistic rakes.

[jamespetts]

I am just looking over these again. The idea of having sleeping carriages is most interesting, as noted above, and may become more important if/when comfort based routing is implemented as is eventually planned. These carriages do look good (although I have darkened the wood shade in the .blend for the NBR carriage as the existing wood was too light).

All other things being equal, it is probably a better thing to have sleeping carriages in the game than not; but one question that we have to consider is at what stage of completeness of the work of adding sleeping carriages is it worthwhile to add this work to the pakset? We have, at present, a carriage from the 1870s, some from the LMS (1920s) and some from BR (the 1950s-1970s). Assuming each type of carriage to last circa 30 years, then the period covered would be (to the nearest decade year) 1870-1900, 1920-1950, 1950-1980, 1970-2000. If we want minimal coverage of all eras, we are therefore missing 21st century sleeping carriages (the mk. 5 design for the Caledonian sleeper, introduced this year) and early 20th century sleeping carriages. If we want a somewhat more realistic coverage to take into account the fact that new designs would have emerged somewhat faster than the slowest replacement cycle, we would probably need to have designs from the 1880s/1890s and possibly 1910s and 1930s, and it would also be good to go back in time to the 1830s when sleeping carriages first started. We would not, I think, need coverage of more than one railway company's sleeping carriages for there to be a minimum to warrant integration, but it would of course be good to have this in time.

Thus, in order for it to be sensible for me to implement these, the absolute minimum further work would, I think, be two further sets of sleeping carriages: one from the early 20th century (at your discretion) and the latest mk. 5 sleeping carriages (and if you are doing the sleeping carriages, you might want to do the mk. 5 day carriages, too).

You might then want to fill out the sleeping car time line by filling in gaps in the late 19th century and the 1930s and then go back earlier in time to the Grand Junction Railway's bed carriage of the 1830s. I can supply research material for these earlier times if that would help.

Turning to some of your more specific questions: it is correct that there are currently no North British Railway vehicles of any sort in the pakset. This is because my policy in filling in details has been to try to have a coherent set of a few railway companies' vehicles rather than to have a smattering of content from a large number of different railway companies. So far, I have not embarked upon any work involving the North British Railway. It would certainly be good to have more North British Railway content as Scotland is very under-represented in railway terms at present. If you would like to take on the work of providing North British Railway carriages and locomotives, that would be splendid.

As to your question:

Does the following statement apply? "East Coast Joint Stock was varnished teak, whilst Midland & North British Joint Stock followed the Midland's coach livery of crimson lake." ((url=https://www.lner.info/co/NBR/livery.php)link(/url))  I'd say that they would try to stick in line with Midland on this one for the long trip to London, but input would be appreciated.

As I have not researched the North British Railway, I do not have any independent sources to confirm or contradict that, but I do know that, in general, carraiges that ran through on long distance journeys between different railway companies tended to be painted in a livery that matched that of their parent companies, and their parent companies thus tended to synchronise their carriage liveries somewhat; so East Coast Joint Stock was always varnished teak (even though the North Eastern Railway's own internal carriages were painted dark red); and the LNWR and Caledonian railways had very similar carriage liveries, no doubt on purpose. Thus, it would be reasonable, on the basis of that research, to use the Midland Railway colour for the NBR sleepers that ran through to the Midland.

My apologies, incidentally, for not having had the chance to get to this topic sooner; I am slowly ploughing through the large backlog of very helpful contributions in amongst my own Christmas project which this year comprises enhancing the GWR offerings. Thank you again for your work on this: it is much appreciated.

[Spenk009]

All other things being equal, it is probably a better thing to have sleeping carriages in the game than not; but one question that we have to consider is at what stage of completeness of the work of adding sleeping carriages is it worthwhile to add this work to the pakset?

I think you're right. One type should be sufficient initially, as these carriages will prove pointless on maps smaller than ~500km across. I have implemented the created ones on my large savegame and they see reasonable patronage. Never full, never empty, making profit but not throwing the balance of the network.

You might then want to fill out the sleeping car time line by filling in gaps in the late 19th century and the 1930s and then go back earlier in time to the Grand Junction Railway's bed carriage of the 1830s. I can supply research material for these earlier times if that would help.

That would be great, as I'm having trouble finding good data to base vehicles off. I do not own many books on the subjects, which are the best source for data.

I agree with adding MkVs. There are Mk3 Sleepers in the works and LMS Period 1, each found as -sle-livery in the Blends. NBR Sleeper too.

have a coherent set of a few railway companies' vehicles rather than to have a smattering of content from a large number of different railway companies

I believe this is vital for inexperienced players looking to find their footing. How about leaving the NBR carriage as NER to simplify things? As to further NBR content, I'll need to undertake some research as to what I can find on the internet. There seems to be a good lot on Wikipedia, usually the case for locomotives, with several good links that I'll see to look into further.

I would like to continue my focus on converting to transparent vehicles for the time being. There is still quite a bit of learning to do, especially adding lost liveries and recreating vehicles. If you would like to create some vehicles and/or improve my work you're welcome to! I'm afraid I won't be productive until returning home by the 10th of Jan. I can sort things out properlythen as I still have some time only working before the university lectures continue.

[jamespetts]

Do not worry about not being able to work until you get home on the 10th - there is much to be said for having a relaxing festive season! I have spent some time to-day looking at gaps in the transparentisation (if that is a word) of the steam locomotives, and have filled in one or two (including locomotives whose .blend files had been erroneously placed with the carriages and locomotives with .blend files not matching the locomotive's name). I have now returned to my main focus of the Christmas season, which is a comprehensive review of GWR vehicles, including adding a full range of different classes of GWR carriages.

In relation to the North British Railway sleeping carriage, do not worry about converting this to a North Eastern Railway carriage: we do not have a comprehensive coverage of North Eastern Railway vehicles in any event (I think that we only have the series of electric locomotives), and, although having coherent sets of vehicles from single companies is my policy for my own work, I do not regard this as compulsory for others contributing to the pakset, though it is to be encouraged. A North British Railway carriage might be a good encouragement to have a whole series of North British Railway vehicles in good time.

As to research material, please let me know what you would like when you need it. I do not have any books specifically on the North British Railway, although I do have some general books on carriages that might give some clues. I am a member of a model railway club that has a good library, which might be a place where I can find research material, although, being a London club, it may lack specialist material on NBR carriages.

However, what I do have is a book specifically about Highland Railway carriages and another specifically about Highland Railway locomotives. We have one Highland Railway locomotive in the pakset so far (the famous Jones Goods, shown in its preserved yellow livery which is quite unlike anything that it actually wore in service, the livery when it was introduced actually being identical to the LBSCR ochre ("improved engine green") livery), but it would be splendid to have a comprehensive set of HR locomotives and carriages. This is not a high priority, but if you would like to go back in time and improve Scottish content, I do have the research material available for this.

As to adding transparent vehicles, adding lost liveries and recreating lost vehicles (such as the tenders for the SR Merchant Navy and West Country and the SR Q1) is probably a good way of learning about creating new vehicles, although your efforts with the sleeping carriages suggest that you have made good progress with this in any event.

In any event, do let me know if/when you need some specific research material in the near future. I can often be busy and not able to answer straight away, but I will get around to answering eventually!

Thank you again for all your work on this, and best wishes for the new decade.

[Spenk009]

I have spent some time to-day looking at gaps in the transparentisation (if that is a word) of the steam locomotives, and have filled in one or two (including locomotives whose .blend files had been erroneously placed with the carriages and locomotives with .blend files not matching the locomotive's name).

This was exactly my intention with the relocation project and work on locomotives! I believe it is vital for another person to look over work like this and I'm just not that acquainted just yet.

good encouragement to have a whole series of North British Railway vehicles in good time.

This sounds like a good future goal. The Highlands railway book sounds great, I'll send out a heads up probably later into the month (depending on my own off-semester work's time requirements).
The NBR study group has several pages on carriages and some statistics about them, here. They reference a book of diagrams, where an edition produced in 1921 could prove very handy. I have found a site that sells a copy, but for 60£ (the ISBN appears to be 0953399516). This could be a handy book, maybe someone knows someone with a copy. I wouldn't mind contributing to its purchase, but can not store the book in the foreseeable future.

As to adding transparent vehicles, adding lost liveries and recreating lost vehicles (such as the tenders for the SR Merchant Navy and West Country and the SR Q1) is probably a good way of learning about creating new vehicles, although your efforts with the sleeping carriages suggest that you have made good progress with this in any event.

Thank you, I'm getting to know what to look for where and the techniques applied to create the models. One of the issues I regularly face is a lack of knowledge and few images online. Over time this has improved and will likely do so further with more attention. I'll look into the tenders in due course.

Are there any specific carriages you would recommend to try replicating for the time before the LMS Period style carriages? LMS Period 1 (1) and LMS Period 2 (1 <- I could possibly misuse as Period 3) I have good sources for, but earlier history is difficult to find.

In any event, do let me know if/when you need some specific research material in the near future. I can often be busy and not able to answer straight away, but I will get around to answering eventually!

It's the same for me. You're very proactive and I doubt anyone here worries about being forgotten. All the best for the new year/decade!

[jamespetts]

Happy new year to you, too!

I will see (if I remember - I do not go home until the week-end) whether the Model Railway Club has the NBR book to which you refer in its library. I might then be able to send you excerpts electronically. Let me know when you need specific information and I will endeavour to look into this.

The lack of online information gets worse as one goes into the past, especially if one goes earlier than the 20th century. For the earlier period, I have found printed books to be a better source, and have acquired quite a number of books on locomotives and carriages over the years, as well as books on buses, which is another somewhat neglected topic. However, some railway companies have better online sources than others; the London, Brighton and South Coast Railway has a good enthusiast following and thus has good online sources, as does the Great Western Railway (on which I am working now). The Great Eastern Railway Society has a good website, too: I learnt useful information about time interval signalling to write some of the signalling code based on this website.

As to styles of carriage to replicate, I have generally omitted the "period 2" LMS carriages as, so far as I can tell, these were only superficially different to period 1 carriages; the period 3 carriages were more significantly different in construction and layout, and so I have added these separately.

If you want to add multiple classes to existing series of carriages, the London and South Western Railway (LSWR) and/or London, Brighton and South Coast Railway (LBSCR) are probably good places to start: both of those are companies whose carriages I was able to add without having to buy books as there are good resources online. Many of the resources are aimed at railway modellers; but you can often infer the capacities of carriages by looking at the window arrangements and knowing how many passengers per side in a compartment can be accommodated (these are often marked in the .dat files). Things like the weights you can then just re-use from the existing carriages. The LBSCR has carriages going back to the very earliest days and thus its timeline is complete; the LSWR has carriages (and locomotives) going back to the 1890s so far. For adding multiple classes to the carriages, you can make relatively minor adjustments to the existing .blend files, and thus ease the workload.

If you want to move beyond trains, road vehicles may be a good next step.

[jamespetts]

Alternatively, you might focus on the LNER and its constituents, as none of these carriages have been updated with class capacities yet, and there are also holes in the timeline that you can fill in (as, currently, the LNER timeline has only express passenger carriages and locomotives and mixed traffic locomotives, and the only constituent represented in anything other than electric locomotives is the GNR, aside from a very small selection of GER locomotives and one set of carriages). You might eventually progress through the NER and NBR; but, if you start with the LNER, you should find it fairly easy to find research material on this website.

Since it is you who will be doing the work, it is, of course, ultimately for you to decide what to do; but you have at least now some ideas of what would be both helpful and reasonably easy to get started with.

[Spenk009]

Thank you for the suggestions to important work on the pakset.  I really like the suggestion to create further LNER carriages.

I have commenced work on the Quad Arts. Here is the blend file, could you or anyone else look over the file and let me know how it compares? I have adapted a LNER Triplet and changed most of the exterior significantly. Materials are sourced from the teak gresley express carriages, but they look a little bright in my opinion.

If this is satisfactory, I will continue creating the other three carriages to match.

[jamespetts]

Interesting. Have you checked the correct scale length using the scale ruler in the pakset? The quad-art carriages were quite short compared to other carriages in reality. It is really quite important to get the dimensions right otherwise things can look very anomalous.

As to the colours, use this post, which gives definitive colours for all liveries used in the game.

[Spenk009]

I have checked the scale length by comparing it to the 15m length ruler, see the image below. The carriage should be 15.5m (51') long, so this means it is 0.5m too short. I've based the coaches on this diagram and this diagram, using pictures to supplement further details. I take it, that while shorter than express coaches, the first two coaches are still considerably longer than the second two. Here is a formation that suggests their use, but no coach length is indicated there. The plan was to recreate all four coaches on the diagram and put them together as a formation. How far off the mark am I?

As to the colours, I imported the material from the other LNER Coaches (Shift + F1), so that they would certainly match. I applied the "Grouping
LNER teak: 0.06924,0.798,0.600 Edit: New for 3 January 2017", but am getting a very wrong result:

[Rollmaterial]

The colour codes in the thread are in HSV, not in RGB.

[jamespetts]

Splendid, thank you for checking that. Of course, I recall now that the quad-arts were of different lengths; it was the quint-arts that were of consistent length and shorter than what you have so far.

The diagrams are a good source, and the Blender screenshot shows that you have done a good job of realising this (although the model is floating somewhat; but this is easy to fix). Your plan seems entirely sensible. You should know that these carriages were generally regarded as quite uncomfortable (the third class compartments were especially cramped), so this should be reflected in the .dat files.

Rollmaterial is right about the colours: they are all expressed in HSV rather than RGB.

Thank you for your work on this: this is most splendid.

[Spenk009]

The colour codes in the thread are in HSV, not in RGB.

Oh my, thank you! I should have checked this ahead of posting. Here's the correct colour, it looks much better:


I just thought the colours were based on the original shade of teak, not the weathered/darkened look. Also the carriage appeared to be floating due to some objects being hidden.

Your plan seems entirely sensible. You should know that these carriages were generally regarded as quite uncomfortable (the third class compartments were especially cramped), so this should be reflected in the .dat files.

I am especially grateful for these snippets of information, they are very helpful to the less knowledgable. In any event, I'd still like to have you look over the .dat files and make improvements. The last coach has even smaller compartments, this must have been quite the unpopular coach! Should I differentiate in this case? It's unreasonable, but do passengers still enter the train at the foremost seat available? I have completed the 3D models, now progressing into creating the dat file.

...the quad-arts were of different lengths; it was the quint-arts that were of consistent length and shorter than what you have so far.

This is a good coach set to continue onwards with. I've found a few sources that hopefully will provide sufficient data to build a reasonable model, but which liveries would the quints be available in? The LNER.info site is very good for further coaches and I'll look into them further as I progress.

[jamespetts]

Splendid, that is a more sensible colour! This does look good now.

By floating, I meant that the carriage's wheels were positioned above the base line in Blender; I have made sure that the models that I have rendered are always placed on the base line for consistency.

I should be happy to look over the .dat files when you have completed this. As to the quint-arts, they would have been only in three liveries: teak, BR carmine (early) and BR maroon (revised).

It is splendid to see some progress with the LNER! One useful thing to add from the LNER at some point might be the steam railmotor; I have recently added the GWR steam railmotor, but that is only available in the 1900s/1910s; the LNER version was built, I think, in the 1920s/1930s, so should allow players to build this sort of vehicle for a longer period.

[Spenk009]

I've made changes and (accidentally) pushed them to transparent-vehicles, commit is here. I've had trouble finding the carmine (BR-Early livery).


The Quad Arts in Teak and BR Maroon livery. There should be the Carmine, but I couldn't make it display. Apologies for the very blurry image if it is blurry, I am currently using a Lenovo W541 with the 3K panel which has a pixel density very unsuitable for accurate representation. This is all I can work with for the foreseeable future, as scaling apparently is terrible in Linux.

James, while researching I stumbled over your shared Quint Art 3D printable models here. They can be imported into Blender, but when scaled appropriately they just look off on the Quad Art (originally sourced from the Gresley Articulated BTO Twin Carriage), see here. Looking into dimensions, do I correctly understand the Quint Arts to be 43' 6" (13.26m)? (source) There is a great set of posts recreating them found here, but they do look a slight bit longer than 43'6".

Instead, I have started gathering information towards the LNER Standard 51' carriages. The LNER.info website is a great source. Would I presume the same liveries as the Quint Arts or the Quad Arts?

[jamespetts]

Excellent, thank you for this. A response on a number of the items below.

Livery

What do you mean that you could not make the carmine livery display? The relevant colour is:

BR carmine: 0.004177,0.957,0,417

Just search for "carmine" in the Blender colours post.

Quint arts

I am not sure what you mean by "just off on" - can you elaborate here? Remember, all the vehicles in Pak128.Britain are 1.25 times as tall/wide as they are long, whereas my 3d printable carriages are intended to be made into an accurate scale model.

These carriages are indeed 43ft6, as confirmed by a diagram that I have of one.

LNER 51ft carriages

The LNER website is indeed splendid, and, yes, these should be in teak, carmine and maroon, as with all LNER carriages save for those (relatively few) which were steel panelled and painted in green and white rather than finished in varnished teak.

[Spenk009]

What do you mean that you could not make the carmine livery display? The relevant colour is:

I couldn't get the carmine livery to display in-game. I started the game in 1949 (timeline disabled, BR-Early livery introduction is in 1948), cycled through all the liveries, but only the maroon and teak would display. You can see the images from the commit and I'll update the Blends repository too. The list is great, it's a good way to ensure consistency across the range and I've used it for the vehicles.

I am not sure what you mean by "just off on" - can you elaborate here? Remember, all the vehicles in Pak128.Britain are 1.25 times as tall/wide as they are long, whereas my 3d printable carriages are intended to be made into an accurate scale model.

I have attached an image of the depot window in 1922, where the Quad Arts are visible in the top row, fourth and third from the right, with the other possible carriage below. Do these look too small/low?

[jamespetts]

Thank you for that - I have done some testing. The carmine livery appears fine, but this is only visible when the timeline is enabled, so do not worry about the livery.

Compared to LMS corridor carriages, these do appear a little short, but it is fairly subtle; it is possible that they were slightly shorter. Have you checked the height with the scale (remembering that height and width are at a 1.25 scale compared to the length)?

In any event, other than the possible scaling issues, these do look good. Excellent work!

[Spenk009]

Compared to LMS corridor carriages, these do appear a little short, but it is fairly subtle; it is possible that they were slightly shorter. Have you checked the height with the scale (remembering that height and width are at a 1.25 scale compared to the length)?

I have now checked with the Quint Art Carriages I had in the works, by enlarging their height by 1.25. Images are attached and I do reckon they look better. I shall therefore rescale the Quad-Arts too.


The Quint Arts in their different liveries! I know it's a little confusing to have completed these rather than the announced 51' carriages.

edit:
Here is a screenshot of the revised Quad Art Carriages.

[Spenk009]

Here are the LNER Standard 51ft Carriages!



What coupling constraint would be appropriate? I can see to constrain the subset of 51ft Standard carriages or to add them to the grouping of LNER-Gresley-Express, GNR Gresley, and GNR Clerestory. And as it seems like common practice was to have the brake van central, so I don't see the usual brake - centre - brake arrangement being enforcible. "At least 1 brake van had to be included in each set, but it was not required that the van (parcels) area be located at the outer end. The brake van (or van end) could often be found towards the centre of a short set, which conveniently suited bi-directional working. "

Another question is, the Diagram 49 & 50 mention: "A later design, Diagram 244, had only 2 1st class but 5 3rd class compartments and with full-size windows to the 3rd class semi-corridor only. All 50 of these, built 1936,..." Would this vehicle add value if incorporated?

[jamespetts]

Thank you very much for that. I have now had a chance to test the quad-art carriages, and, with a number of modifications, included them into the set.

The various issues, which it will help to know for reference for producing future vehicles, were as follows:

(1) the names were not consistent with the naming convention:
  (a) there were words in capital letters which do not require capital letters (e.g. "Third");
  (b) the class of passengers carried was sometimes described in the name of the carriage rather than in brackets at the end (and sometimes both);
  (c) the composite and all third carriages had some of their names swapped;
  (d) the number of compartments in the carriages was not specified in words;
  (e) extraneous non-standard information (such as "outer" and "inner") was included; and
  (f) the term "quad art" was used, whereas it might be difficult for players not familiar with these carriages to know what that means, so I have spelled it out in full as "quad articulated";
(2) the comfort was anomalously high; you will need to use "Comfort calibration overview.ods" to calibrate the comfort of the carriages by comparing them to like vehicles and add the comfort values that you deduce as appropriate to that file;
(3) the overcrowded capacities were too low: for compartment carriages such as these, the overcrowded capacity of a carriage should be half the second and third class seating capacity of the vehicle in question combined;
(4) it is probably better to set the retirement date later as players may want to build these into the 1930s, since there are no replacements;
(5) because these are articulated carriages, the rolling resistance per carriage will be a little lower than that of ordinary bogie carriages, so I have adjusted the values in the .dat files accordingly; and
(6) the loading times are too long for a vehicle such as these which had an outside door to every compartment and where passengers would not normally have been carrying luggage; I have lowered them accordingly - generally, loading times should be copied from an equivalent type of carriage.

There are one or two other things that I have not modified but which may need consideration after further research. First of all, I notice that there is no second class accommodation; I believe that these may still have had second class accommodation when built, but I am not entirely sure of this, so this will need to be checked. The quint-art types certainly had second class when built. I have added a second class comfort value in the calibration schema. Secondly, are you sure that the composite carriages were only 5 compartment carriages even though the long thirds were 8 compartments? Finally, it would be helpful to know the distance between compartments for refining the comfort calibration.

Incidentally, I also notice that you have re-rendered the mk. 1 carriages, which is splendid. May I suggest that the next priority to be to re-render all the DMUs and EMUs that use mk. 1 carriage graphics so as to make sure that they are all consistent in size and rendering style throughout the trains?

Edit: I see that the above has already been done - excellent.

In relation to the LNER standard bogie carriages, the better thing to do with the brakes is to require one brake carriage at the rear, but not require a front brake. Because Simutrans-Extended does not have a mechanism for requiring there to be one carriage of a particular type somewhere in the rake, it is better to be unrealistic as to the exact position of the carriage in the train (which has no economic consequences) than to be unrealistic in allowing players to build a train without a brake carriage (which would have significant economic consequences, as the trains would have a higher capacity and a lower fixed cost than one with a brake carriage).

Thank you again for your work on these - this is much appreciated.

[Spenk009]

Thank you very much for that. I have now had a chance to test the quad-art carriages, and, with a number of modifications, included them into the set.

I had fixed some of the issues in the mean time, but not committed any changes thus far. I was going to summarise the three sets into a commit and pull request.


(1) name consistency: I'll apply these points to the Quint-Art and Standard carriages. I had used the names given on the diagrams.
(2) comfort: The tabs! I didn't see the tabs below. I had suggested a comfort of 46=3rd, 50=2nd, 62=1st, but will base values on the given ones.
(3) the overcrowded capacities: With the very small room for people to stand between the benches, I felt that three or four would still be reasonably able to fit in with the other legs.
(4) retirement date: I have set the Standard Carriages' retirements to 1942. Maybe that is reasonable for the articulated sets?
(5) rolling resistance per carriage: Thank you. I'll use their new values for the other carriages.
(6) loading times: I was thinking of the luggage in combination with the brake carriage allowing such stowage, but it is unrealistic to have lots of luggage on commuter lines.
I set the maximum speed to 105km/h, should this be higher? I reckoned a 60mph running speed would be reasonable, with some leeway but if you have good data this should be implemented.

I've based my information mainly on Marcmodels' and Steve Banks' pages.

2nd class: "It's often forgotten that the GN Section retained 2nd Class accommodation in London and the composite quad-arts were built with 1st, 2nd and 3rd Class compartments. The 2nd Class was abandoned in 1938 and 1st Class during the War, in 1942" There would have been three second class compartments in the long centre vehicles (C and F) and another three in the long brake vehicles (D and E). I had forgotten these and can add them too. Also, there were no 5 compartments but 7, thank you for spotting that! The Quint Articulated carriages do have the second class implemented.

In relation to the LNER standard bogie carriages, the better thing to do with the brakes is to require one brake carriage at the rear, but not require a front brake. ... it is better to be unrealistic as to the exact position of the carriage in the train ...

Then I shall constrain them within their type, not allowing coupling to Quad, Quint, Gresley Express, or foreign carriages (at least before the end brake carriage).

[jamespetts]

Thank you very much for that, and apologies for integrating prematurely: I had not realised that you had planned to work more on these. I shall look forward to your commit/pull request in due course.

To respond to various issues: the comfort that you give above seems inconsistently low with other contemporaneous vehicles, so it would be helpful if you could calibrate the others to match my revised quad-art figures. The quint-arts should have the same comfort as the quad-arts, and the standard carriages a higher comfort.

As to overcrowding, the received wisdom in railway circles by my understanding is that compartment carriages have a 150% overcrowded capacity, which is what I have used for all other carriages. If you have some better data, please let me know, but without relaible data, it is better to stick to this standard formula, I think.

As to loading times, it is important to make these consistent with other carriages of a similar type; for commuter carriages, the standard has been to disregard luggage loading times, as, for short distance journeys, people do not generally take luggage. The question to ask yourself is whether, at the vehicle's maximum comfortable journey time, a significant proportion of passengers would be likely to be carrying bulky luggage.

As to the maximum speed, the question is not what a reasonable top speed of the sort of service on which the carriage would be used would be, but rather whether there would be any technical limitations in the carriage design itself that would prevent it from going beyond a certain speed. The convention is that all carriages built before the modern era (1960s) have a maximum speed of no more than 160km/h, and that bogie carriages in that era have a maximum speed of 160km/h unless there were a specific reason that the carriage design itself limited speed to a lower amount. In the absence of data, I suggest setting the maximum speed of the carriages to 160km/h. It is not the carriage speed limit that should generally be the determining factor for the speed of commuter services.

As to second class - this should indeed be added to the quad-arts: thank you for looking into this. In relation to the retirement dates, 1942 would indeed be a reasonable retirement date for the quad-arts, so feel free to adapt this in your commit/pull request.

Likewise, do adjust the capacities and translation texts specifying the number of compartments in the latest commit to be correct for seven compartment composites rather than five compartment.

Thank you again for your work on this - it is much appreciated.

[Spenk009]

Apologies for the delay in finishing up.

The quint-arts should have the same comfort as the quad-arts, and the standard carriages a higher comfort.

I would like to disagree with you on the topic of comfort in the Quint-arts. They offer even less space per compartment than the Quad-arts, at least that was my experience during the design stage. I was really squashing the compartments together when adding the composite carriages, so I'm quite reluctant to place them at the same comfort. As for overcrowding, I've changed this to suit the other vehicles.

I've made alterations to suit and the pull request is found here. It's ugly, since I've been adding things as I've had time: should I try to make the addition a single commit?

[jamespetts]

Thank you for your work on this, and apologies for the delay. I have now had a chance to test these (see the "spenk-transparent" branch on my Github repository).

A few comments:

(1) the graphics seem good, except that the brake carriage of the quint arts seems to have its length defined as too short so that it overlaps with the following carriage;
(2) the word "Standard" should be removed from the description of the non-articulated carriages to make this consistent with all other types of carriage;
(3) the front brake in the standard carriages seems to have some issue with its coupling constraints: it appears available in the depot window even without a locomotive (it is the only carriage other than the GWR railcar to do this), and is able to be put at the rear of a train; and
(4) the comfort of the standard carriages seems not to be correctly calibrated; there is as much of a difference between second and third class comfort as there is between first and second, whereas in reality, by this time, second class was much closer to third class in comfort than first class.

The comfort numbers also seem somewhat rounded; you should really consider exactly how the compartments rank in comfort compared to other similar vehicles and give them the exact numbers that suits that rank: use the Comfort calibration overview.ods spreadsheet (look at "All Vehicles") and make sure to add your newly created vehicles to this so that we can keep track of the comfort of these in the future. In relation to the relative comfort of quint and quad arts, if you have data showing that they had a different spacing between compartments in the same class, then do reflect this in the comfort ratings (using the spreadsheet to find the comfort of similar non-lavatory vehicles).

Do not worry, incidentally, about the number of commits: this is not a problem.

Thank you again for your work on this: this is most helpful. Do let me know when you have addressed the issues above so that I can re-test ready for integration.