Why use High Poly - Sketch Up ... instead of learnig GMax & Blender

What I'd personally like to see is a utility which can test a Trainz asset in a somewhat realistic usage, and report back on the real-world performance. For many objects, "how well does this one object perform?" is the wrong question. A much better question is "how well does this asset perform when spread across the map in a realistic scenario?" It would need to take into account LOD, loading time, buffer usage, memory usage, effect on frame rate, etc.

Of course, that only answers the performance question. Whether the object works as expected and looks nice are completely separate questions.

chris
 
Perhaps a mechanism to "hide" your own content from the listings- so it can still be downloaded where required, but won't be trivially downloaded by new users.

chris

That system used to exist back in 04 days - it was used a lot by creators of stuff like enginespecs and the like where an asset was really only useful as a dependency, but useless as a 'freestanding' asset.

I know the purpose of this proposal is different, but back then as an upload option it caused nothing but trouble and copped endless criticism...
 
Just a thought/suggestion here.

In RubyTMIX I make use of the N3V-provided Trainz Mesh Importer to create the IM file for an exported asset, from an XML source file that contains all the geometry and information about all of the materials. It is my understanding that the Mesh Importer is also used as the final step in the pipeline for other exporters from other tools; therefore this is (I think) a 'common' part of the process for assets created in every tool.

If that is true, then I am wondering if there is any way some type of evaluation could be done by this tool, since it has knowledge of the entire asset as it is being processed. For example, if a particular (new) command line switch is used, it could then emit an extra file that is imported into Trainz, and which is recognized by Content Manager, as being an indicator of 'asset weight' or whatever you would like to call it. This in turn could then be available as a sortable column and/or a search parameter in Content Manager, and could also be something that is displayed on the DLS.

Obviously I do not know the technical details of the inner workings of the Mesh Importer utiility, so what I am suggesting may be completely out of line. It just seems to me that the Mesh Importer is a logical place to try to perform some kind of analysis since it would appear to be a common 'entry point into Trainz' for content created in all tools. If I'm wrong about this, no problem... just thinking.

Regards,
-Mike
 
Last edited:
Just a thought/suggestion here.

In RubyTMIX I make use of the N3V-provided Trainz Mesh Importer to create the IM file for an exported asset, from an XML source file that contains all the geometry and information about all of the materials. It is my understanding that the Mesh Importer is also used as the final step in the pipeline for other exporters from other tools; therefore this is (I think) a 'common' part of the process for assets created in every tool.

If that is true, then I am wondering if there is any way some type of evaluation could be done by this tool, since it has knowledge of the entire asset as it is being processed. For example, if a particular (new) command line switch is used, it could then emit an extra file that is imported into Trainz, and which is recognized by Content Manager, as being an indicator of 'asset weight' or whatever you would like to call it. This in turn could then be available as a sortable column and/or a search parameter in Content Manager, and could also be something that is displayed on the DLS.

Obviously I do not know the technical details of the inner workings of the Mesh Importer utiility, so what I am suggesting may be completely out of line. It just seems to me that the Mesh Importer is a logical place to try to perform some kind of analysis since it would appear to be a common 'entry point into Trainz' for content created in all tools. If I'm wrong about this, no problem... just thinking.

Regards,
-Mike

Logic and Trainz don't mix well, especially when it requires programming from N3V. Actually does it? I wonder how many cases of beer slung perhaps Pev's way it would take to create a tool that created a file that contained the relevant information in a specific format? That could always be merged with the description in the config.txt file.

Cheerio John
 
Logic and Trainz don't mix well, especially when it requires programming from N3V. Actually does it? I wonder how many cases of beer slung perhaps Pev's way it would take to create a tool that created a file that contained the relevant information in a specific format? That could always be merged with the description in the config.txt file.

John,

You cast out a line and I had a nibble....?? Thanks for the thought in any event. A six pack would be enough for this one..

It's a quite a simple task to analyse a mesh to count polys and textures. The results could easily be automatically appended on to the description. A command line utility (rather than a windows one ) would be appropriate, and hopefully nice and compact. (I'm rapidly running out of space on my website.)
 
.... Simply adding a kuid to a thread that lists kuids of performance hobbling assets (for whatever reason) is not an attack on a specific creator. A creator wants feedback, and too often doesn't get any, except for compliments. While those are fine, they don't address issues the creator may not be aware of.....

For my part, I'm in complete agreement with Euphod's suggestion to create a list of problem KUID's. My thought, though, is that it should not be in a thread on the forums, but on the WIKI pages. But it seems to me that it need not be limited to items needing work. The opposite list, those items that are particularly well done, should also be established.

Finally, it seems to me that perhaps the key to the issue raised by the original poster is to develop some sort of program to grant credentials to content creators, perhaps in different bits of discipline. Anyone could still call himself or herself a content creator. But those who wish could submit materials to the community for the evaluation, and after evaluation the community would use some mechanism to award an appropriate credential. I would propose that a variety of credentials might be established: freight cars; passenger cars; diesel locomotives; steam locomotives; bridges; buildings; reskins; & any others deemed desirable.

ns
 
John,

You cast out a line and I had a nibble....?? Thanks for the thought in any event. A six pack would be enough for this one..

It's a quite a simple task to analyse a mesh to count polys and textures. The results could easily be automatically appended on to the description. A command line utility (rather than a windows one ) would be appropriate, and hopefully nice and compact. (I'm rapidly running out of space on my website.)

Could it include a rough size so as to differentiate between a small scenery object of 1,000 polys and a very large one?

Thanks John
 
Could it include a rough size so as to differentiate between a small scenery object of 1,000 polys and a very large one?

Some one was talking about a measure of polys in a given volume. Or, say, polys per cubic metre.. that may work.. Again dead easy to extract and calculate.

All meshes made with TrainzMeshImporter (any version) and later versions of the 3DSMax exporters have the bounding box specified in the mesh. Not sure about the TACS.

Just thinking of details a bit, I will look at the config to get the highest LOD mesh only. I'll have to cover both types of LOD but that's also easy enough.

I'll add the 300 polys per texture as well. At the start I won't try to compensate for texture image size.

I'm busy with AssetX updates at present, and I have another tool I'm doing finishing touches on as well. Meaning this little exercise may be quite a few weeks away.
 
MJmolnir

issue raised by the original poster is to develop some sort of program to grant credentials to content creators
If this happens I will stop creating, this is a hobby not a competition and if it changes I have no desire to compete for arbitrarily assessed credentials.

I have no objections to people criticizing my work 'poly-wise'

Peter
 
With regard to textures, and their effect on performance, is it the number of images or the number of materials that matters? (eg. from just 1 image file, I can create a single-sided diffuse material, a two-sided diffuse material, an opacity material, a two-sided opacity material, etc). Or is it some combination of both images and materials that counts?

Also I would question the validity of a polygon-per-volume index. Maybe I'm being too technically simplistic here, but I suspect if the game engine sees an object, it doesn't care how big or small it is, it still has to render whatever number of polys/vertices are in it. So I don't think poly density is such a helpful metric.
 
logoswg.jpg


NOT authorised by N3V. JUST for fun. A jpg so there'll be no mucking about wiv it ok ^^
:o
Hexagon is currently free from Daz 3D. Some might like to give it a try.
I have NO idea how compatible it would be with TRSXX mmmk?
 
Last edited:
Some one was talking about a measure of polys in a given volume. Or, say, polys per cubic metre.. that may work.. Again dead easy to extract and calculate.

All meshes made with TrainzMeshImporter (any version) and later versions of the 3DSMax exporters have the bounding box specified in the mesh. Not sure about the TACS.

Just thinking of details a bit, I will look at the config to get the highest LOD mesh only. I'll have to cover both types of LOD but that's also easy enough.

I'll add the 300 polys per texture as well. At the start I won't try to compensate for texture image size.

I'm busy with AssetX updates at present, and I have another tool I'm doing finishing touches on as well. Meaning this little exercise may be quite a few weeks away.

I think Chris's figures were 300 poly overhead per mesh and 200 per texture.

Cheerio John
 
With regard to textures, and their effect on performance, is it the number of images or the number of materials that matters? (eg. from just 1 image file, I can create a single-sided diffuse material, a two-sided diffuse material, an opacity material, a two-sided opacity material, etc). Or is it some combination of both images and materials that counts?

Also I would question the validity of a polygon-per-volume index. Maybe I'm being too technically simplistic here, but I suspect if the game engine sees an object, it doesn't care how big or small it is, it still has to render whatever number of polys/vertices are in it. So I don't think poly density is such a helpful metric.

The size comes into it as the overhead per asset is roughly 500 poly equivalents so if you create an asset of four houses then you only get one overhead but the poly count will be four times that of a single house. KUID: 86627:100029 is of this type.

Cheerio John
 
logoswg.jpg


NOT authorised by N3V. JUST for fun. A jpg so there'll be no mucking about wiv it ok ^^
:o
Hexagon is currently free from Daz 3D. Some might like to give it a try.
I have NO idea how compatible it would be with TRSXX mmmk?

Hexagon makes great models. You then take them and run them through Blender or Sketch-up to make a Trainz (Jet) compatible file. I did that with the small strip mall my bro and I worked on a few Trainz versions ago. Back then we made the model in Amapi Pro which became Hexagon. Once the model was made, we exported it to a 3ds and ran it quickly through GMax. It's the same idea.

John
 
MJmolnir


If this happens I will stop creating, this is a hobby not a competition and if it changes I have no desire to compete for arbitrarily assessed credentials.

I have no objections to people criticizing my work 'poly-wise'

Peter

I agree Peter. Designing and approving by committee sounds too much like work to me. This is supposed to be a hobby and for fun.

John
 
Back
Top