dem

[rudder]

Hi all, How real is the dem results ie. If the true elevation is 700 meters what can we expect to see in the sim. thanks.

[zbobg]

It is certainly accurate enough for RailSim.

My experience is that most DEM data (whether from USGS or SRTM) is something like plus or minus 3 meters off the real world value.

In populated areas where there is long standing survey data (USGS data for example) the accuracy can be expected to be plus or minus 1 meter or less.

[landnrailroader]

Colleagues,

Absolute accurace of DEM data varies considerably, and the SRTM data that is preferred by RS may be off by many meters in populated areas, since it is measured by RADAR and thus tops of tall buildings will appear as high elevations even though the ground may be flat.

I prefer to use MicroDem and generate my .hgt data from seamless USGS files in as fine a resolution as possible, preferably 1/3 arc second, though in some areas, we now have 1/9 arc second. Even so, I ran into a situation at Clifton Forge where the ground seems to suddenly fall away at a point in the railroad yard, where several parallel tracks continue on a slightly risiing grade for another 2 miles or so. Now I know that at this point, the mainline and the yard are on sort of a shelf along the ridgeline, so I set my track elevations to be within 3 meters or so of the countour interval value - RS measures elevation at the top of the rail. I've probably got my tracks a little high but it works.

As an aside, I created a scenario using default UP stock - not too far off real since we see a lot of UP engines on CSX and the track laid so far is very smooth except for one kink that I think I can fix. Sculpting scenery works pretty well, but the result is a sawtooth profile along the side of the track, where it ought to be smooth, so I need to study this a bit. I am liking this sim a lot but I cannot use it for my historic representations unless a few things are improved on. Laying track is trickier than one might think but fairly easy. I tend to sometimes generate a really bad wiggle as I start to lay a new segment, and have to delete the itty bitty pieces. This may be a mouse problem, or a nervous finger problem.
I have listed the improvements that I see are needed in the route design area of these forums.

landnrailroader

[Paddington bear]

Hi,
Wiggle track has nothing to do with your mouse or your twitchy finger.
Its a post patch BUG which is becoming a pain in the arse, so take heart from the good bits.

edit:- its official............ http://www.railsimulator.com/en/node/2469

Ross

[Craig H]

Want built in markers? That's not a mound of dirt in the picture, it's a very tall rural grain elevator. This is from SRTM .hgt data. Sidescanned areas with mountains or skyscrapers could be subject to a lot of inaccuracies.

Might actually be better to use converted DEM data if it's available to you.

[rudder]

Hi all thanks for your replies, I dont know If this Is the way to go, I recorded the alt. in google earth at the time I set my markers, I then recorded the alt. at the base of my markers in RS. this Is all done in the country not towns. Without to much checking I can see a diff. of -3m to 10m, this is not very good for my route. I don't know much about converted dems. may'b this is the time to learn.thanks.

[landnrailroader]

I gave up on using the downloaded .hgt files simply because they are inaccurate. MicroDem is not hard at all to master, at least the necessary parts. I would recommend that you go that route and wherever it is available use either 1/3 or 1/9 arcsecond seamless data. 1/3 is available over most of the US, while the least detailed data is available virtually world wide.

In the forums there is a link to an excellent tutorial on how to use it. I print some of these out on 8.5x11 paper and put them in a 3-ring binder. Don't do all of them, but just the ones I really need to reference a lot. SRTM data, derived from seamless DEM data is considerably more accurate than the downloaded .hgt files, and you will use it and other tools, also free, to produce the desired .hgt data.

Another issue that I would hope Kuju (lissen up Adam) would consider is a constant readout of altitude at the tip of the mouse end of a track, when it is laid. The window at the bottom tells us gradient and radius, but you have to stop laying track, and then double click to get a elevation readout ("y") in the side panel. Much more accurate grades and routes could be created if this was a constant readout.

landnrailroader

[rudder]

Hi all, I have not had much luck in finding the dem files,I found and dl, ned files, can they be used? if so what do I need to convert them? thanks.

[Craig H]

Try http://seamless.usgs.gov/

[rudder]

Hi ,I crawled all over this site, unless I am blind I donot see any DEM files, thanks

[rudder]

Hi folks, looks like I found how to get dem files, I think It was a long way around . hope It was worth It. thanks