5m vs 10m Grid

[rrfoose]

Are there any negatives of converting from the standard 10m grid to the 5m grid? It seems that Trainz defaults to 10m, and I wanted to make sure that before I converted it to the 5m grid for smoother terrain, that I wasn't walking into a "trap" - no one really ever mentions the 5m capability! Thanks!

Chase

 

[mp202]

Very good question. I really never noticed a difference. I just thought it was better for track laying (which it is).

 

[chrisaw]

I was hoping it would make it possible to have narrower embankments, but it doesn't seem to make any difference.....unless I'm doing something wrong.

Chris.

 

[2995Valliant]

Chris - you can. I'll send you some screenies.

There's a higher load on the PC using the 5m grid (more vertices) so only use it on the baseboards under to the track.

Anthony

 

[Dermmy]

When 5m came along i though WOW, tried it, and went back to 10m!

5m looks marginally better if you just click 'Smooth Spline' and leave it at that. However even with 5m I found that the terrain still needed manipulated manually to really smooth things out. The problem is that with 5m there is significantly more manipulation needed. The other issue is that the processing load is significantly larger with the smaller grid.

With a 10 grid, the first 10m square has four reference points, but each additional square in the first row adds only two more. Subsequent rows after the first square add only one more reference point per 10m grid square.

With 5 m grid, the first 10m square now has nine reference points, but each additional first-row sqaure adds six more, additional rows add four more for each 10m square. Work that out over a route! And every one of those reference points needs processed and - worse - manipulated. OK - you're not going to manipulate a whole board, but even the trackside manipulation is significantly more time-consuming.

After applying ground texture there is so little visual improvement with 5m over the traditional 10m I just don't think the pain is worth the gain....

Andy

 

[stagecoach]

Not noticed any difference in the computer, but for smoother hills and steeper embankments it works fine. Dont mix 5 and 10m boards on the same route as this does effect the computer. Dont go from 5m back to 10m boards because you will see a big difference.

 

[rrfoose]

Thanks for the input everyone! Considering the higher load on the computer, I think I'll stick with 10m. I spend more time in surveyor than I do in driver simply because of the choppy animation from today's graphics intense items. I think that for fairly flat terrain, 10m is certainly the way to go. I'm working on a route in the Appalachians, and with all of the rock cuts and fills, I thought 5m might work better and give me a more realistic appearance. I'll have to check out some of those rock-cliff splines I've seen photos of - if those things work out, I won't have to worry about how my rock cuts look! Thanks again!

 

[2995Valliant]

Two pics of a single track cutting and embankment using the 5m grid:

@rrfoose - I doubt you'd notice the performance difference running a 5m grid on just the baseboards you need.

Regards,

Anthony

 

[rrfoose]

Thanks for the demo pics Anthony! It sounds as though mixing the boards may be bad? Have you done this with no ill effects? Thanks!

 

[JCitron]

If you do use the 5m grid, don't convert a baseboard that is already textured. It will make the textures turn into weird blocks as they'll lose their blending. I found out the hard way with this, and made an awful mess with a section of my big route.

Mixing the baseboards doesn't seem to affect the performance so much as the terrain its self. I noticed weird clear spots coming through the baseboard edges where the two kinds meet up. No matter how much you'll try to blend them in, that space will always be there.

John

 

[schweitzerdude]

I agree with JCitron that changing a 10m grid baseboard to 5m will cause a texture mess, but the mess is limited to the baseboard(s) you converted. My rule of thumb is, when adding baseboards to a route:

1. baseboards with track and/or detailed texturing are added as 5m grids.
2. baseboards with little going on (background plains/hills/ocean) are added as 10m to save time when saving and decrease file size.
3. if an existing 10m baseboard incurs any subsequent development, I try to remember to convert to 5m grid before doing anything else to it.

 

[2995Valliant]

The route in my two screenies is about 25 miles miles long, 3 miles wide and started in in TRS2004 (so 10m grid) which now has 5m grid boards under all the track. Losing the texture blending is a pain, but not the end of the world to do if you have to upgrade already textured boards. Fortunately I have never experienced the issue mentioned by John of clear spots where the two types of baseboard meet.

It's each to their own as to whether or not it's worth it, and depends a lot on the terrain being modelled I suspect. For my routes, which tend to be set in the west of England with a lot of fairly small scale earthworks the extra work is well worth it for the effect.

hth,

Anthony

 

[steamboateng]

I wouldn't dismiss the 5m grid so readily. Sure the goundfile is larger. But the groundfile is only a fraction of the finished route, allowing for terrain textures, buildings, track, and rolling stock
When building roadbed cuts and fills, that smaller grid allows more realisic terrain features close-up. Also rivers, streams, creeks, marshland, etc. can be more realistically modeled.
It all depends how much detail you want to put into the route.

 

[chrisaw]

On thinking a bit further on the query I have with the 5m grid, I now realise that my frustration isn't with the grid size, but that the smooth spline tool parameters haven't changed to match i.e. there's no appreciable change in the embankment angle (and dig hole is also still 10m).

Chris.

 

[Dermmy]

On thinking a bit further on the query I have with the 5m grid, I now realise that my frustration isn't with the grid size, but that the smooth spline tool parameters haven't changed to match i.e. there's no appreciable change in the embankment angle (and dig hole is also still 10m).

Chris.

Exactly what I found, which is why I went back - the terrain still needs manipulated manually, but the job is significantly harder....

 

[WindWalkr]

The upside of the 5m grid is that you have improved precision in forming and texturing the terrain. In some cases this doesn't matter, but in some cases (eg. angled cuttings) this can make the terrain look substantially more realistic.

The downside of the 5m grid is that it takes 4x the disk space, 4x the RAM usage, and 4x the number of polygons when rendering. This doesn't matter much for occasional usage, but you probably don't want to construct an entire route at the 5m detail. Best to use 10m as filler away from the track and only increase to 5m near the track where additional detail is actually required.

For example, you could start building at 10m, then increase some areas to 5m when you realise that you don't have sufficient control at 10m. As others have noted, it's best to make this decision before you get serious about texturing the route.

chris

 

[rrfoose]

On thinking a bit further on the query I have with the 5m grid, I now realise that my frustration isn't with the grid size, but that the smooth spline tool parameters haven't changed to match i.e. there's no appreciable change in the embankment angle (and dig hole is also still 10m).

Chris.

Well, that really takes some "umph" out of the 5m's case. If you have to adjust it all by hand anyway, it might be easier to just stick with the 10m...

 

[steamboateng]

Just a simple observation.................it seems the closer one gets to reality, the bigger pia it is.
Tsung Zsu; The Art of Trainz; Chapter IV; unpublished notes............

 

[clam1952]

10, compared to 5m? 5m every time, much better control over fine detail to the terrain and no excessively wide bits when using the smooth tool, I'm voting for an even smaller grid.

 

[Vern]

As Chris said, 5m grid occupies 4 x the hard drive space for the same area. It is invaluable if you're using overlaid maps (e.g. with Transdem) to get sharp detail and on a small route (say up to 20 miles) you will probably get away with all 5m grid. However anything over that and you are looking at some seriously large sizes for the *.gnd file and that's before you start placing objects or painting textures. In which case go for 5m to maybe 1500m either side of the track and 10m for the rest. If you use Transdem to set up your routes, you can specify this at the extraction stage when Transdem builds the route tiles.