• Our booking engine at tickets.railforums.co.uk (powered by TrainSplit) helps support the running of the forum with every ticket purchase! Find out more and ask any questions/give us feedback in this thread!

Railmiles irregularities?

Status
Not open for further replies.
Joined
24 Jul 2011
Messages
442
Location
Wigan
There's a very good chance that it's me that is at fault here, but I was hoping someone could shed some light on a possible irregularity within Railmiles.

I've come across some interesting mileages when using the Metrolink in Manchester, but assumed it was as a result of it being light rail. However, I've come across a potential issue when travelling between Wigan and Kendal.

The input of today's 12:18 journey ,returns a distance of 44m 41ch. A mileage engine query returns a distance of 57m 24ch. And whilst only as far as Oxenholme, the National Rail Timetable seems to agree with the higher milage.

Would I be right in thinking that there's an error with the Railmiles calculation, and that it is the higher 57m 24ch distance?

Thanks
 

Attachments

  • Screenshot_20220109-122349.jpg
    Screenshot_20220109-122349.jpg
    245.5 KB · Views: 67
  • Screenshot_20220109-125240.jpg
    Screenshot_20220109-125240.jpg
    183.7 KB · Views: 67
  • Screenshot_20220109-125127.jpg
    Screenshot_20220109-125127.jpg
    96.5 KB · Views: 65
Sponsor Post - registered members do not see these adverts; click here to register, or click here to log in
R

RailUK Forums

JonathanH

Veteran Member
Joined
29 May 2011
Messages
18,531
There is a similar irregularity coming out of London Bridge towards East Croydon on the Thameslink route. It seems part of the route is missing when the distance is calculated.
 

Mcr Warrior

Veteran Member
Joined
8 Jan 2009
Messages
11,648
Don't forget also that Manchester Metrolink track distances are, I believe, all measured in km.
 

etr221

Member
Joined
10 Mar 2018
Messages
1,037
There is a similar irregularity coming out of London Bridge towards East Croydon on the Thameslink route. It seems part of the route is missing when the distance is calculated.
Checking myself, the Railmiles Mileage Engine is giving an excessive distance from London Bridge to Croydon, as it is using a devious indirect route...
 

D6975

Established Member
Joined
26 Nov 2009
Messages
2,864
Location
Bristol
Many large stations give you several identical options on the dropdown list when entering them into Railmiles. The different options appear to be different platforms/sections of the station, but there is no indication of which is which (different codes but no explanation). As a consequence, you can get some stupid results, for instance if you chose Clapham Junction to East Croydon and pick the option that is the lower numbered platforms at CJ you'll get a silly route (but it will be the shortest possible from the 'wrong' platform). You always need to check the route and if it's wrong pick another option from the dropdown list.

ps It also gives the shortest route from the options you put in. If this involves a reversal it will give you a route that involves reversing at a junction. And mileages via York will take you round the avoider unless you're heading up the Scarborough line. Similarly, in my neck of the woods it will give mileages via the Rhubarb in Bristol if avoiding TM is the shorter route.

pps Wigan-Kendal is indeed 57 and a bit miles. Railmiles gives me 57m 24c
 

Attachments

  • Untitled.jpg
    Untitled.jpg
    269.9 KB · Views: 22
Last edited:

etr221

Member
Joined
10 Mar 2018
Messages
1,037
I have been in contact with Railmiles Support - in particular the London Bridge-Croydon issue, and they have given an explanation: as it is essentially a free spin off from their commercial product, under certain circumstances a simple A to B enquiry does not provide expected options (and whose absence they do not provide for), and in these cases essentially their system cannot 'see' a route segment (in this case London Bridge direct to New Cross Gate) and will produce an alternative, non-sensible, route avoiding it... which makes it somewhat useless.

Over the issue of different options for some stations, they pointed me at CRS, NLC, TIPLOC and STANOX Codes (on the railwaycodes site) , which explains them (they are a subset of TIPLOCs).
 

Tom

Member
Joined
19 Jan 2008
Messages
550
Location
35,000ft
I'm told there have been eight or so queries about London Bridge and the Sussex route over the weekend. I think the above message is a bit of a simplification of the messages that were sent out over that time.

When the mileage engine was moved to the main RailMiles platform in 2020, it migrated to using the same regularly updated database as RailMiles and RTT. That gave a lot of advantages in reducing the maintenance burden and other things. In doing that, it also gained the ability to route more 'appropriately' using a weighting system. This meant that between Sheffield and Doncaster it routed via Swinton as you'd normally go rather than Thrybergh, etc. The move was a lift and shift in a lot of the logic.

The complexity of London Bridge towards New Cross Gate means that the shortest route is not the normal route and there is a minor mileage distance depending on the junctions you pass through. To prevent the system routing across the mostly illogical shortest route (but you still can go that way... which is the problem), essentially all the junctions had to be weighted. This, in essence, meant that in a normal world there was no route from Spa Road to New Cross Gate unless you specified a number of via points along the way. To get the route you'd have to do it in the way RTT was doing it, by searching between mandatory timing points, so you'd search London Bridge-Spa Road, Spa Road-Bricklayers Arms, Bricklayers Arms-East Croydon, etc.

Having read all the tickets this morning, I've decided to change the algorithm back to as advertised and it will now always route across the shortest distance regardless of whether this is typical. There's more detail on the RTT blog about it.

The input of today's 12:18 journey ,returns a distance of 44m 41ch. A mileage engine query returns a distance of 57m 24ch. And whilst only as far as Oxenholme, the National Rail Timetable seems to agree with the higher milage.
It's a glitch relating to the signal at Oxenholme, and it has now been fixed. 57 and a bit miles is the correct distance.
 

ainsworth74

Forum Staff
Staff Member
Global Moderator
Joined
16 Nov 2009
Messages
27,538
Location
Redcar
Has Railmiles not had the Battersea Power Station extension added?
 
Status
Not open for further replies.

Top