Anybody willing to pay for this?

[Approach_Medium]

Hi;
I have been collecting DEM data for Trainz routes, creating GeoReferenced Raster maps using Google Earth, and drawing routes on the raster maps, to create Trainz routes which are as close to prototype as possible.

So far, I have done this only in the USA, but so long as DEM data is available for the entire world, I would expect I could do it for any place.
Google Earth maps vary in detail, so it is not always possible to get enough detail for a rail route to lay track directly from the raster maps, but here in the USA, eastern New Jersey and New York, I have been able to get such detailed GE maps to see crossovers and sidings, and even signals sometimes.

This process is very time consuming, so I was wondering whether there are any Trainzers out there who would love to do this, but just don't have the time.
Since I have been unemployed for a while, I have a lot of time on my hands, and might be interested in producing prototype Trainz routes (basic track, road, water, etc), for a small fee.
At this point I do not know how much anyone is willing to pay for such work, but of course the fee would have to be based on the amount of time required to collect, reference, and create the route.

Currently, I have TS2006, 2009, and 2010, so I could produce routes for any of these versions.

Please let me know if you would be interested in something like this, and what you might be willing to pay for it.

Thanks

FW

 

[jadebullet]

Yes, I am sure that people would pay for it. I might be interested in some point if the need arose.

 

[cascaderailroad]

Low detailed DEMs are hard to use, as treetops and buildings deform the true height of the landform.

I believe that quality 1/3 arc second data makes all the difference.

(Similar to: Fishlipsatwork data-1/10 arc data)-Horseshoe Curve-Very low detailed cutting of terrain-Mountains are rounded off

Jerker-Horseshoe Curve-High Detail-Waterways and RR cuts are shown-terrain is exact

Fishlipsatwork DEM-Railroads up on the hillside ... lack of cutting data

Everybody wants a DEM ... But a DEM is not really needed at all ... and turns a otherwise simple route, into an absolute obsessive compulsive PITA chore.

DEMS make all the mountains and valleys for you ... but they make laying gradients a nightmare.

I would have been done my route if I had used flat baseboards.

 

[geophil]

Low detailed DEMs are hard to use, as treetops and buildings deform the true height of the landform.

I believe that quality 1/3 arc second data makes all the difference.

Fishlipsatwork-Horseshoe Curve-Very low detailed cutting of terrain-Mountains are rounded off
[image]

Jerker-Horseshoe Curve-High Detail-Waterways and RR cuts are shown-terrain is exact
[image]

Are you sure the first screenshot is "Fishlipsatwork"? Apparently both "Fishlipsatwork" and "Jerker" show the same set of map ground textures. As Fishlipsatwork used to work with MicroDEM/HOG, textures on maps created by him would look different. And he wouldn't have used topo maps. His data source was TIGER, IIRC.

Concerning DEM resolution outside the US and Canada, most people do not have access to higher resolution DEMs at all. Yes, you have to create cuttings and embankments yourself over here. And you may have to adjust terrain height at places. But I don't think it has stopped anybody from going prototypical, neither in Europe, nor in Australia or New Zealand.

 

[cascaderailroad]

Yes you are right, I was mistaken ... correction ... Fishlipsatwork DEM's look "like" "similar to" the detail shown in photo #1 lacking detailed topography cutting.

Rivers and RR's are mistakingly slightly "off register", and are instead displayed up on hillsides, and are off by +/- 20m in all dirrections and heights.

Here is a detailed DEM:

One of Fishlipsatwok DEM's has major height flaws, where terrain absolutely does not match up. In several locations huge rifts and cliffs are at DEM seams

 

[geophil]

Don't blame it all to the rather lo-res DEM data. Keep in mind that TIGER vector data was not meant to come close to a topographic map. TIGER was/is a map platform for the US census authority and does not compete with USGS. If I remember correctly, the TIGER specs say something like 1:160,000 whereas the standard USGS topo map is 1:24,000. This will explain quite a few rivers flowing uphill. Furthermore, the MicroDEM/HOG process was never bulletproof, mainly due a number of manual steps involved, all subject to human error. But also to technical shortcomings in the process which have been discussed often enough.

It may not even be the DEM, though. When I started with USGS DEMs about 9 years ago (before the advent of Trainz), I already could use 30m DEMs (now 1 arc sec) and they did indeed show the major earthworks. It may have to do with the way the DEMs were transferred to HOG as an image and the subsequent low-pass filter in HOG itself.

 

[Approach_Medium]

I agree. The 1/3arc are the only way to go, if the area you are prototyping is available. I tried to use 1arc SRTM data, also from USGS, but it has "black holes" all over the place, and would be very difficult to fix.
TransDem works so well with the USGS DEMs and doing GeoReference from Google Earth, it is possible to create a route that is ready for grading.

I don't particularly enjoy the grading process, but have it down to a science, so it goes pretty quickly now. I may not end up with the exact cuts and fills, but it works, and looks great.

The one thing I would love would be if TransDem would allow the use of more textures, so that I could actually use the ground textures it generates.
I think in TS2004, that can be done easily, but in 2006 and later, the textures are compressed/encoded in the texture files, so it's not such a simple process to create them.

So, I will continue to use the method I have been.

As for the original subject of this thread, I am going to upload all of my new creations to DLS and offer for free. After all, I have DL so much great content from others, it is time I give something back.

What I would be creating is basic proto routes with proper terrain, track, road and water in the right places, but I cannot guarantee that I will be able to get switches and signals placed, since all that depends on the quality of the GE maps I am using.

Also, there is the issue with scenery. It is very difficult to find just the right buildings for the proto route. So, I may just create the bare-bones route without scenery, and let those who DL create it for themselves.

I would like to try creating some sessions though. I haven't played with AI all that much.

One more thing: I have to learn to limit my route size to that which can be uploaded. I don't recall offhand what the max size for upload is, but I will search for that info.

I am focusing mainly on Eastern US railroads now, but hope to eventually build some others.

FW

 

[Dap]

FW

In reply to your original question, yes. I would be willing to pay a nominal fee for the creation of the 1/3arc high resolution terrain of a prototype route.

David

 

[Vern]

Why pay when you can buy a copy of Transdem for a reasonable price and be able to create limitless routes/layouts yourself?

The learning curve to get up to speed with basic functionality in Transdem (terrain - route outline - maps - export) is around 20 minutes, following the tutorials.

As regards placing gradients, embankments - you are never going to get these off the DEM. You either need to follow the gradient profile/chart (in conjunction with the other data) or if you don't have this available take logical and sensible measurements to calculate the gradient required over a fixed distance - depending on the type of terrain I tend to use 500, 1000 or 1500 metre intervals.

I agree the map data can sometimes be offset a little, depending on the source of the data (the Poehali Russian maps probably worst as Roland and I have discussed separately) but in hilly terrain this can easily be compensated for and on the plains not that critical.

Back in the early days of Trainz/MSTS, before the advent of DEM handling utilities (and decent DEM) all our routes had hand carved terrain and it was a PITA in terms of time and effort to get anything vaguely resembling the prototype.

 

[geophil]

The learning curve to get up to speed with basic functionality in Transdem (terrain - route outline - maps - export) is around 20 minutes, following the tutorials.

There are also the new video tutorials, kindly provided by Gisa, for a Canadian route: http://forum.transdem.de/viewforum.php?f=6&sid=f92cc3d1998028da6e75685f1f3ad3f1

As regards placing gradients, embankments - you are never going to get these off the DEM. You either need to follow the gradient profile/chart (in conjunction with the other data) or if you don't have this available take logical and sensible measurements to calculate the gradient required over a fixed distance - depending on the type of terrain I tend to use 500, 1000 or 1500 metre intervals.

I agree, over here in Europe the DEM won't suffice and you will need additional sources. However, the USGS 1/3 arc sec and - in particular - the 1/9 arc sec NED DEMs deliver astounding detail. 1/9 arc sec translates to less than 3m (which TransDEM will round to 5m)! They will probably be called "HD" DEMs in the near future.

1/9 arc sec NED (shown before)

 

[Approach_Medium]

Why pay when you can buy a copy of Transdem for a reasonable price and be able to create limitless routes/layouts yourself?

The learning curve to get up to speed with basic functionality in Transdem (terrain - route outline - maps - export) is around 20 minutes, following the tutorials.

As regards placing gradients, embankments - you are never going to get these off the DEM. You either need to follow the gradient profile/chart (in conjunction with the other data) or if you don't have this available take logical and sensible measurements to calculate the gradient required over a fixed distance - depending on the type of terrain I tend to use 500, 1000 or 1500 metre intervals.

I agree the map data can sometimes be offset a little, depending on the source of the data (the Poehali Russian maps probably worst as Roland and I have discussed separately) but in hilly terrain this can easily be compensated for and on the plains not that critical.

Back in the early days of Trainz/MSTS, before the advent of DEM handling utilities (and decent DEM) all our routes had hand carved terrain and it was a PITA in terms of time and effort to get anything vaguely resembling the prototype.

I agree with you on TransDem. It is very easy to learn, with the great tutorials provided. I am just thinking that perhaps others don't have the time, or do not particularly care for the task of creating a route, but enjoy placing scenery, putting the finishing touches on, and running trains.

I am usually able to get the cuts/fills correct by knowing which road crossings are grade, and which are bridges or underpasses. I am usually able to get this data from Google Earth.
As you stated, the DEM (and GE) data varies in accuracy/detail from place to place.
Where I live, in northern NJ, USA, both are excellent. But when I looked at GE for the state of Vermont, New Hampshire, and Maine, the data is not nearly as detailed, except for major city areas. I guess that makes sense, since the vast majority of the users of the data are looking at population centers.

Perhaps if I am trying to make a little money with what I am doing, I could put my work on a website I own, then put Google Adsense ads on the site. They don't pay much, but it might actually be a better alternative to DLS, since I could place much larger routes than I could upload.

Happy Demming

FW

 

[Approach_Medium]

There are also the new video tutorials, kindly provided by Gisa, for a Canadian route: http://forum.transdem.de/viewforum.php?f=6&sid=f92cc3d1998028da6e75685f1f3ad3f1

I agree, over here in Europe the DEM won't suffice and you will need additional sources. However, the USGS 1/3 arc sec and - in particular - the 1/9 arc sec NED DEMs deliver astounding detail. 1/9 arc sec translates to less than 3m (which TransDEM will round to 5m)! They will probably be called "HD" DEMs in the near future.

So far, the 1/9arc DEMs are available only for a few areas in the US. I am sure that over time more there will be more of them.

FW

 

[Approach_Medium]

On the subject of the USGS DEMs:
When I go to the Seamless Viewer, the map first comes up green, but immediately the US goes to grayscale. Is this normal, or is there an issue with my browser (Firefox 3.6.13).

Thanks

FW

Edit: On the viewer Display tab, I have the GTOPO60 ticked, and 1/3arc ticked. If I un-tick all of the arc selections, the map appears in green again.
I suppose this is simply to make the map load faster, or provide better detail.

Edit 2: OK, I think I see what is going on. The detail selected (1arc, 1/3arc, 1/9arc) is shown in grayscale. Areas that are not available for the selected detail are shown in green.

 

[ApproachMedium]

I would be interested in paying for a high quality DEM or whatever for the Maine Central. Learning the software may be easy, but I sure don't have time to figure it out and find the files etc, etc.

I'm not even sure if a high quality DEM could be created of the Maine Central system, but if someone could do it, I'd buy it!

I'd like to see a sample before purchase of course. And I'd expect a reasonable price tag.

I'd lay the track and roads, etc. All I'm interested in is the geography, and a high quality image overlay for reference. If this is possible, let me know.

-Joe