10th March 2010

All change again...

Well the first thing is that further to my report on the main page about Auran's plan to scupper support for TRS2006 or earlier versions, you can pretty much count me out of any further route building in Trainz for the forseeable future. Not that TS2010 isn't a good platform to build on, I just can't use it in Surveyor mode on my development laptop. So that, as they say, is that - as well as being a protest on behalf of their end users. They've also lost a FCT (first class ticket) renewal out of me which was due in the next month or so. Pah, big deal you say but multiply that by several 1000 customers feeling the same way and the pain may go a bit deeper.

I'm still feeling a bit burned out after the Tanigumi route (as of today up to a staggering 80 downloads on UKTS) and I think that has spilled over into indecision about what to do next. Suffice to say I've got a couple of potential Railworks projects on the pot boiler though the Japanese ones have dropped back to the rear echelons. A crowd pleaser might have been Weymouth up to Maiden Newton and down the Bridport branch (including the famous Quay line) but reading the forums on UKTS seems that has already been bagged by someone else. So the favourite at the moment could be a little bit of Canada, specifically the former BC Rail line out of Lillooet and up the Fraser gorge, for about 40 miles to Kelly Lake. Initial marker and DEM viability assessement currently under way and I should know in a day or two if it's a go-no go for launch.

7th March 2010

Pleased to say the Tanigumi route is now live on UKTS and has attracted a reasonable number of downloads.

Now I know I said yesterday my next project was the Joetsu line but, as tends to happen with Railworks, this hit several major snags today.

The first is that my attempt to create a "loft" to represent a concrete tunnel interior met with miserable failure. Even consulting the Railworks Wiki (which deals mostly with 3DS Max, nice if you can get it) and various threads on UKTS there just isn't enough coherent step by step information how to start from scratch and produce a lofted tunnel lining. Looks like that's one for the superior beings of train simming only. Second up, I found the 90m DEM a tad too averaged for the very hilly terrain that the Joetsu line runs through. It's one thing making a few alterations to the terrain by hand, it's another thing finding what you've done looks all wrong. The final check, on reaching the section with the long tunnel, is that it's virtually impossible to lay these when markers are on top of the terrain 500m or more above and Railworks will only allow you to lay 500m maximum length of track at one time.

So work on the Joetsu line has been shelved, at least for the time being. I'm now looking at alternative less onerous projects, at least one possibility being the route that featured in the Tomy SL de Ikou steam sim, from Otaru to Niseko on Hokkaido. Will report back in a few days whether or not that's a go...

6th March 2010

Well the Tanigumi route has been uploaded to the UKTS file library and just waiting approval before available to publically download.

Now despite 'er indoors noticing I'd finished the route and nodding rather pointedly at various paint pots, wallpaper and associated decorating accessories I've been looking at what to do for the next project. I've decided to see how the Tanigumi route does download wise before embarking on another Japanese tram route - so Minomachi is on ice for now. I mentioned a few days ago about the Joetsu line, this continues to be a favourite of mine so I decided to check out the data availability. Google Earth offered a mixture of high and low resolution data so not so good. However, just as I was about to give up I managed to find a plug in to a Japanese web map server which overlaid the GE imagery with arguably more useful raster topographic mapping data and I was able to get the primary rail markers done this afternoon. So on to the DEM, the only stuff available is the SRTM 90m data and after setting up the route in Railworks (sorry TRS guys, you'll have to wait a while for the next Trainz route now) went through the course of the line with the DEM. Good news 97% of the data is sound though there are a few holes which will need to be patched if visible from the track. No biggie, should be possible with a few road sections and use of the terrain tools. However can I appeal again to RS.com, much of the DEM data is flawed in this way and Railworks needs some kind of built in facility to heal holey DEM. None of the proprietary external DEM handling programmes which can fix the holes is capable of saving back to SRTM format, RW doesn't recognise any other format and the one programme I found that purports to turn BIL data back to SRTM creates vast terrain spikes and towers - worse than the original holes.

What this route also has is tunnels - lots of them. Now the Japanese tunnels have a particularly distinct orangey/brown concrete texture, nothing like any of the default tunnel sections in Railworks. So the next challenge is using various 3D modelling programmes to see if I can create my own tunnel "loft" to go over the buried sections. I'm also going to need to build some cave stations, but that's for another time.

5th March 2010

Good news is that my Tanigumi Tram route is now in the final stages of testing and I'm planning to release it (via UKTS) at some point over the weekend.

Bad news is that RW's lack of decent AI/signalling/pathing where you have opposing movements using the same section of track seems to have struck again. Despite, to the best of my knowledge, having signalled the route correctly and gone back to double check, I cannot set up a scenario that uses any AI trams as it constantly generates cannot find path type error messages. So I'm afraid the included scenarios are going to be a bit lonely like, you and the solitary tram. The performance (or lack of) RW AI where single track routes are concerned has been an ongoing issue even back to the days of RS. Put simply, AI trains do not like to interact with the player train.

All of which creates a bit of a dilemna as the next tram route I had planned - Minomachi - if anything has even more in the way of single track/crossing loop running. What is the point in building in Railworks if you can't recreate the busy hustle and bustle of a real urban transit route. Could build it in Trainz but I'm not sure how adaptable TRS is to tram style routes (plus of course the issue of having to self change points along the route). Still it is tempting to shun Railworks for anything other than double track lines where the AI can be confined to its own track seperate from the player and ne'er the twain shall meet. It's certainly annoying having spent the best part of a month working on the route and, though I say so myself being quite please at how it looks, to be frustrated at the last stage of the creative process.

FFS, RS.com, sort it out... You can begin to understand why in some quarters Railworks is now being referred to as RailFail!

28th February 2010

Been a little while since I posted in here, that's largely because I've been the victim of another distraction. However this one's good - trams. The release of several tram models for Railworks stirred me into thinking about creating a short route to go with them. That, combined with my ongoing exploration of OpenBVE, led me to try and recreate at least a resemblance of the Japanese Tanigumi line in Railworks. Fan's of BVE may recall Tanigumi was one of several routes built by the talented Mr Gaku and is a scenic trolley run north of Gifu in Central Japan. Only 12km in length but starting through the suburbs then climbs through mountainous and hilly terrain. It's probably all going to look a bit incongruous with double deck Blackpool trams and various RW default scenery masquerading as a very anglicised Japan, but hey if it gives a good tram run - that's the main thing. Good news is that construction is well on, most of the 3D placement has been carried out just the signalling and testing etc. to do so hoping for a mid-March release.

Once bitten by the tram bug, it's hard to shake it off. I've already pencilled in a possible next project which is the Minomachi tram route - also in Gifu. Those with a broader view of train simming may recall this was the tram route which featured in Taito's Densha de Go Nagoya collection. If you haven't got/can't play the DDG version, there's a BVE version (also very good) which works well in OpenBVE. Sadly the Nagoya Railroad closed the real tramway in 2005 which means both the DDG and BVE versions now represent a little bit of history.

Looking even further afield, I suspect my route building efforts may remain in Japan for some time to come. In addition to the trams, there are several Densha de Go routes which it would be nice to take a crack at in RW or Trainz. There's also a couple of crackers I've found in BVE, at the moment my focus being on the Joetsu Border line. This is an incredibly scenic mountain line, with the Up and Down lines running on separate alignments, featuring a heady mix of long tunnels, spirals and the trademark Japanese "cave" stations. More to come on all of that.

12th February 2010

The issue of not being able to build my Taieri route in TS2010 started to rankle a bit during the week...aka I'm a sore loser. Even though it meant starting over, I did a fresh Transdem extraction of the terrain and mapping into TRS2006. I managed to get track etc. laid out during the week so effectively back to where I was with the original version. TRS2006 may not have quite the amount of built in content but at least I can apply terrain texturing without slowing the laptop to a crawl. Also I've noticed routes I built in TRS2006 open and scroll okay in TS2010 so there's definitely something performance sapping how TS2010 applies textures. To twist the famous quote from "Jaws", if I ever want to route build in TS2010, "We're going to need a bigger laptop..."

Having got things sorted out in TRS2006 it seems that's where I'm planning to stay for the time being. In fact I've been trying to set out a campaign plan, or road map if you like, of projects for the coming year. Taieri Gorge is the #1 priority. Following that I definitely want to return to Scotland and do a bit more with my West Highland route. Biggest blow of all at the moment, is that last year I had fairly well advanced a US route in TRS2006 based on the Adirondack Line in NY State. Unfortunately I "lost" the route when I uninstalled TRS2006 in favour of the distraction(Railworks) but was confident I had backed up the cdp file as it stood if I wanted to return to it. However the file is nowhere to be found on the laptop or any backup media. Gutted. Ten weeks work down the drain for a moments impatience. Fellow route builders, if you take any lesson from this (as I have done) is never, ever delete or uninstall anything unless you're 110% you are absolutely finished with it. So at some stage I will have to put part of the Adirondack back on the agenda, even if it's just Lake Placid to Tupper Lake. The lost version had track and a fair proportion of scenery all the way up to Moran Jn. Muppet.

4th February 2010

They say a week is a long time in politics, the same can be said of train sim route building. I wrote last week of my mild disillusionment with Railworks and that I had sought solace in TS2010. However these things have a habit of biting you back. The TS2010 Taieri project started so well with the terrain extraction. I got most of the track laid over the last week and, like an artist with his blank canvas, poised to start terrain painting and 3D placement. However after placing textures and trees on one linear kilometre, Surveyor has slowed to a crawl. I had this before in TS2010 when starting a route, at the time I blamed that on using the 5m terrain grid feature but this route was laid down to the 10m standard. My suspicions are pointing at the Speedtrees of which I placed about 12 in total so far, or maybe it's just my laptop (which I hasten to add has no problems with TRS2006, MSTS or Railworks). Further experimentation needed there but to all practical purposes as this is the second instance, I don't think TS2010 is going to be suitable for route production on my current hardware.

Anyhow the upshot of that is I've gone back into Railworks (which now seems to be behaving) and dusted off the northern ECML project. Well it's only been dormant for a week so only a little dust. I think it's a question of plodding on with that for the time being and placing any other distractions firmly to one side. Taieri Gorge will come at some point, even if I have to downgrade to TRS2006 or build it in Railworks... Next week is going to be a bit hitty, missy for route building due to a particular point on my roster at work which combines two 12 hour weekend day shifts with afternoon shifts Monday and Tuesday, which is not conducive to getting much done. However I'll hopefully squeeze another km or two of placement on the ECML, to keep things ticking over.

28th January 2010

On distractions, frustrations and other sundry matters...

One of the biggest problems we face as route builders is embarking on a project, with the sudden realisation it is going to occupy most of your spare time for several months. Strange to relate this is not an issue I encountered that much with MSTS but with Railworks in particular it has become a major challenge. Despite best attempts to remain focused, it is all too easy to get distracted and decide something else might offer a better use of that precious spare time. Well folks, I regret to say I have sucuumbed again. The first distraction has been the release of the freeware Chinese locomotives and coaches for Railworks. These are so much better than even the best of the payware UK and US items produced thus far, that thoughts turned to perhaps creating a route on which to run them. Unfortunately Railworks, has an Achilles heel with its reliance on Google Earth to produce markers in that much of China is only covered at very low resolution, nowhere near good enough to plan a decent route. Distraction #2... Loaded up TS2010 the other night to find that not only are the Chinese diesel locos also in there as default items, there's also a bit more rolling stock - all to standards which surpass much of the other content. However same problem again. There's no decent topographic maps of China to download into Transdem and provide an overlay for a Chinese route. A brief struggle with imagery pasted together from Google Maps (also not very detailed) proved fruitless.

So, back to the ECML in Railworks. First check yesterday on loading Railworks is it seems Steam wants me to re-enter my password every time I start the wretched thing, even though I opt to remember it. At least half the time it tells me that password is invalid, even though I typed it correctly. I hate Steam. Next thing on going to load the fledgling route in Railworks, the programme locked up and eventually exited with the dreaded "Railworksproc" error. Why? I haven't done anything to the programme. However it does appear Steam interferes with files on your PC. The other day I also had to reinstall the IOW route (the RS version off DVD) as it had disappeared from my Railworks installation. There has been confirmation that Steam will delete files it thinks shouldn't be on your PC, after it inadvertently downloaded the Tornado payware to some users which was promptly snatched back by Steam a few days later. A post regarding this on UKTS, to gauge whether anyone else had the problem and to discuss the implications was promptly snuffled by the mods over there which has also increased my generally pissed off state with RW (and UKTS) again - though don't worry the teddy is staying in the pram at present as regards my file library uploads. Finally, on going to start RW today, after once again going through the Steam log in process and getting the route to load this time, the whole thing was running like a slideshow.

At the moment then, Railworks is back on the virtual shelf. Thoughts have now turned to creating a route in TS2010 on which the Chinese stuff will look good. I've picked out a prototype on which to base a project, it's the Taieri Gorge line on the South Island of New Zealand. (Wikipedia entry here). One plus of a NZ project is that the amazing Transdem programme will connect to a web mapping service for the country. Consequently it is a doddle to overlay the course of the line on the DEM and have a fully extracted and mapped out set of tiles in TS2010 in just over an hour. Eat that, Railworks... Now I'm not saying at some point I won't return and finish the ECML in Railworks but frankly it was already beginning to bore me. Not that many people seem to be downloading the routes anyway (the count for Heartbeat Moor and the updated Sulitjelma are not good), in fact I wouldn't be surprised if less than 1000 people in the UK are actively using Railworks. Trainz has a huge target audience and while things can get a bit flakey with the DLS and uploading files etc., for starters there's vastly more scenery items and textures available in the default items alone. I love Trainz!!

25th January 2010

Well I can report that things are off to a pretty good start. The first part of the process was marker creation in Google Earth. The railway itself has been traced from Alnmouth to Berwick along with the coastline and roads in the Beal to Tweedmouth area. In creating such a potentially long route as this I was very conscious of past experience, it is comparatively easy to lay lengthy track mileage quite quickly but that can create a huge mental block when it comes to the process of decoration, which takes much longer. So I've decided on this project not to race ahead and lay 50 miles of track in one go, but to work in ten mile or so sections. That is to say, lay 10 miles of track then break off to do the terrain painting and primary 3D placement. Rinse and repeat. That way, I don't get faced with a whole lot of 3D placement to do and there's always something else to look at when I get fed up of decorating yet another section. Signalling/speed limits and other interactives will of necessity go in right at the end. This is an old MSTS habit, probably not strictly necessary in Railworks but always in the back of my mind, this thing was programmed by the same outfit and there are similarities in other areas, so let's not take any chances.

Anyhow the first section is approx 9 miles from Beal to Berwick. Track laying took no time at all, though having driven the OpenBVE version a couple of times since I probably need to tweak the DGL at Berwick. The default DE viaduct in Railworks was an almost perfect match for the Royal Border bridge. I've started 3D placement working north from Beal, I normally work in 1km strips across the line, for non-urban areas generally speaking it's about 3 to 4 hours work to paint the terrain and plant the 3D objects. Terrain painting is one of the main chores in Railworks. As I've commented on the forums in the past the tools and techniques are quite different from the "pre-printed" tile patches we used in MSTS or the relatively easy swirl and mix of TRS. You can't copy and paste from one area to another either which makes it probably the most time consuming task. 3D placement is about the same as the other sims, slow in the country and painstakingly slow in towns. I do wish they would fix the snap to terrain function for lofted objects so they properly follow the lie of the land. Nothing more annoying than laying road sections and finding a piece pops up in the air rather than flush with the terrain. Trainz splines are better in that respect.

So progress to date is 4 x 1km strips with primary placement and the fifth underway. Wow, that's about 5% of the total... I reckon this 9 miles (or around 15km) will take at least another two weeks to work through with the 3D stuff plus I need to dive back in Google Earth and finish off the markers for roads etc. around Berwick. At some point I will need to think about 3D modelling. There's all the station signs to be made plus structures like Tweedmouth signalbox which can't really be represented generically. However that's in the future, for now the key is to get a solid base to build on and ensure I have enough time invested and confidence in the project to carry on.

More news soon then!