TS2009/10 standards or not?

[PEV]

I just downloaded and tried a brand new asset for build 2.9 (TS2009) and it sort looked good but details were missing and colours were mixed poorly.

On checking the asset I found that it was constructed to the standards we would have used for UTC. For example:-

--- It has 23 textures of various sizes (mostly small like 32x32) where with careful thought probably 2 or 3 would have been enough.

--- It uses no env_metal texture to get directional shine on wheel rims (for example).. I know that this can be done other ways.

--- There are no normal maps or TS2009 material names..

--- and finally the smoke is rubbish by modern latest standards.

The sound is the latest type (yay!!!) got one right....

It's a shame.. it's a good model of an iconic loco.. but it could have been so much better.

Content creators........Please, if you are building for latest versions, use the new features.. They really make a difference.. and they are not that hard to learn.

 

[Euphod]

Could it have been an item made for an earlier version and just polished enough to work in 2009/10?

 

[Paulsw2]

Now I'm trying to guess what this loco is.....

Paul

 

[johnwhelan]

I just downloaded and tried a brand new asset for build 2.9 (TS2009) and it sort looked good but details were missing and colours were mixed poorly.

On checking the asset I found that it was constructed to the standards we would have used for UTC. For example:-

--- It has 23 textures of various sizes (mostly small like 32x32) where with careful thought probably 2 or 3 would have been enough.

--- It uses no env_metal texture to get directional shine on wheel rims (for example).. I know that this can be done other ways.

--- There are no normal maps or TS2009 material names..

--- and finally the smoke is rubbish by modern latest standards.

The sound is the latest type (yay!!!) got one right....

It's a shame.. it's a good model of an iconic loco.. but it could have been so much better.

Content creators........Please, if you are building for latest versions, use the new features.. They really make a difference.. and they are not that hard to learn.

Content on the DLS has always been a mixed lot. If you create in TS2010 it tends to drop a TS2010 version in even though the content doesn't require it.

By the way it took me best part of a year to get env_metal texture working in Blender.

Cheerio John

 

[Dinorius_Redundicus]

If it took a genius like John a year to figure out how to do a metal texture, then what hope for the rest of us? Perhaps a comprehensive, easily understood Content Creators Guide would encourage higher standards on a wider scale.

 

[PEV]

Could it have been an item made for an earlier version and just polished enough to work in 2009/10?

Not this time, Ed.

@Deane,
The supposed finer points of normal mapping are just a small part of the whole exercise. The info on how to do it unfortunately is scattered over the Wiki and the forum, as well as in the CCGTC. You are right about needing a consolidated source of info.

And getting env_metal to work should take five minutes if you use GMax or 3DSMax. Those two programs do all the thinking for you to set up reflective textures. The only user variable you need to think about is the amount of reflective texture.

I still use GMax with the Max4 exporter.. It works fine for all content that I have made. If the latest mesh/animation version is deemed essential I can just run the mesh through PM2IM. I know my meshes work doing this because I have bypassed the INFL chunk bug in my tools by having the correct type of animation construction in the first instance. And, of course, they work correctly in the game.

 

[Billegulla]

If it took a genius like John a year to figure out how to do a metal texture, then what hope for the rest of us? Perhaps a comprehensive, easily understood Content Creators Guide would encourage higher standards on a wider scale.

My sentiments, precisely!

Bill.

 

[mcguirel]

Not this time, Ed.

@Deane,
The supposed finer points of normal mapping are just a small part of the whole exercise. The info on how to do it unfortunately is scattered over the Wiki and the forum, as well as in the CCGTC. You are right about needing a consolidated source of info.

And getting env_metal to work should take five minutes if you use GMax or 3DSMax. Those two programs do all the thinking for you to set up reflective textures. The only user variable you need to think about is the amount of reflective texture.

I still use GMax with the Max4 exporter.. It works fine for all content that I have made. If the latest mesh/animation version is deemed essential I can just run the mesh through PM2IM. I know my meshes work doing this because I have bypassed the INFL chunk bug in my tools by having the correct type of animation construction in the first instance. And, of course, they work correctly in the game.

Reverse Engineer it and make it the way you want and call it your own. PEV, you are a very gifted person and the software you have made to help the community is outstanding. I am sure you could tackle this with ease. Not trying to be rude so do not take it that way, cause I value what yo have done for the community

 

[johnwhelan]

If it took a genius like John a year to figure out how to do a metal texture, then what hope for the rest of us? Perhaps a comprehensive, easily understood Content Creators Guide would encourage higher standards on a wider scale.

It is now documented and there is a sample on the DLS so you should now be able to incorporate it in your models fairly easily.

A lot of this stuff is knowing where to look and finding an appropriate tutorial.

Cheerio John

 

[PEV]

My sentiments, precisely!

Bill.

By the way Bill, I have been using your fantastic spreadsheet calculations to try to make the loco I am talking about run in AI without continuously venting the safeties.

Rather than change your "R" factor I have altered the three efficiency variables with very pleasing results. The loco's creator used 0.57 for the "R" factor, which seems to make all of the numbers, including the horsepower the same as the real world..The loco now behaves exactly as I would expect with the steam pressure moving around by about 10psi and only venting when you stop after running quickly for some reasonable period.

That spreadsheet is one very fine piece of work.

Reverse Engineer it and make it the way you want and call it your own.

I can already do that in 5 minutes too, but as a software writer I am very concious of hacking.. so I don't use my arsenal of "own-use" tools for anything other than data recovery..end of argument...

The point of my first post above is to highlight the fact that some content creators are still working to the standards of UTC or TRS2004 but presenting their work as TS2009 content.. The results may work by they look "dated" as you would expect.

I suppose people like me who were involved in the Beta testing of TS2009 went down many blind alleys trying to create new content at that time and learned lots in the process..

 

[zatovisualworks]

Hi guys,

Now I'm trying to guess what this loco is.....

Paul

:hehe: Please, Paul, if finally guessed, I claim it as mine!!!!!!!!!!!! :hehe:

On the less light side...

A couple of weeks ago someone, a good-will trainzer no doubt, typed something about his hope that in the near future after the DLS winter clean-up he could use the 'native' mode... and that's the beginning of the confusion and misunderstanding basically among users and among some absent-minded creators.

By simply having assets not showing errors or nuisance warnings, users are not getting but assets for compatible mode -which I proudly defend and labelled it as aboriginal mode :hehe: - disguised with the trainz-build 2πr (or trainz-build pi2 if you like circles best).

Making assets the T9/T10 way and standards is another matter. And maybe it's going to take some more than good will and hope to have a couple of thousands of new Native-costumed assets.

Not really hard to learn, once you have gone through the desert in a horse with no name :hehe: , but new lessons to revise and a new matter to pass.. or fail. No problem, we are all at N3VRF4IL home.

Anyway, being for either aboriginal or native, doing things properly is not costly, just putting some effort and care on the disguise (it's Carnival time, guys) you're sewing.

The thread and needle side of Alberte :wave:

 

[Beattie]

The reason its not very detail and not up to trainz 2010 standards is because most people used to or still do make things compatible for trainz 2004 so thats why most content doesn't use the new advanced details hope this help.

 

[PWeiser]

What He Said!

The reason its not very detail and not up to trainz 2010 standards is because most people used to or still do make things compatible for trainz 2004 so thats why most content doesn't use the new advanced details hope this help.

Wholeheartedly agree with the above. To some, latest-and-greatest is a virtue in itself; to others, easy backward compatibility is an equal or greater virtue. Neither point of view should be disparaged... even by Auran and its dedicated fans, who might stand to make more money if only latest-and-greatest is permitted. Their question (if I may put words into their minds) is not, "How do we make them buy our latest stuff?" but "How do we make our latest stuff so attractive that they'll want to pawn the family silver to buy it, and crawl over broken Wiki-glass to learn to make assets for it?"

It's all intangible: aesthetic judgment of what looks really, really nice vs. goodwill and social solidarity of the ultra-hip vs. social solidarity of the tried-and-true. (IMHO)

 

[zatovisualworks]

It's all intangible: aesthetic judgment of what looks really, really nice vs. goodwill and social solidarity of the ultra-hip vs. social solidarity of the tried-and-true. (IMHO)

On this statement:

We are two of the same IMHO -I know, I know, there are more, many more-, but of late it seems we are are some kind of outlaws over here.

On the whole:

Whatever you do, disguised aboriginally or natively, try to do it the best way. The 23 different textures Peter encountered aren't really a nice picture :hehe: at any standards! Recently on another forum another topic we've been talking about texture template sheets for unwrapping method.

The Alberte The Kid

side of himself :wave::hehe:

 

[paulhobbs]

And getting env_metal to work should take five minutes if you use GMax or 3DSMax. Those two programs do all the thinking for you to set up reflective textures. The only user variable you need to think about is the amount of reflective texture.

It takes five minutes in Blender as well, it just takes a year to find out what you have to do in those five minutes (it took me about as long to find a technique to get nice shiney metal, but it doesn't use reflective textures)

The reason its not very detail and not up to trainz 2010 standards is because most people used to or still do make things compatible for trainz 2004 so thats why most content doesn't use the new advanced details hope this help.

Not really, because a loco that uses 23 materials is inefficient in TRS2004 as well.

Paul

 

[JoshEH]

The point of my first post above is to highlight the fact that some content creators are still working to the standards of UTC or TRS2004 but presenting their work as TS2009 content.. The results may work by they look "dated" as you would expect.

I suppose people like me who were involved in the Beta testing of TS2009 went down many blind alleys trying to create new content at that time and learned lots in the process..

I remember those days. It seemed to me like a lot of content creators resisted the push for better graphics. To me it seems most modern TS2009/10 content I'd still consider dated. There are a few exceptions though. Even the way I've seen people construct their normal maps is slightly outdated compared to the game industry at large.

However, constructing a modern next-gen asset is a lot more complex than it use to be. Even compared to 4 years ago a lot has changed and the bar has been pushed higher and higher.

I think it's unrealistic to except a lot of assets that are much past the Trainz 2004 mark, as far as graphics are concerned. A lot of the guys around here are hobbyist. Even the guys that have been doing it awhile lack the sort of experience they'd need if they ever wanted to get into the video game industry. I don't mean to knock anyone down, but there's a lot you learn and pick up once you're in the heat of production that you just don't learn outside of that. I've seen at least a few tutorials and videos on creating next-gen assets where I thought "wow, that wouldn't fly in production..." Under certain art directors the demand for perfection is great.