Pls give us the example scenes
- Thread starter spoonetti
- Start date
In the TMT deep dive livestream, Sam did say something about this in relation to their legal team. I suspect the delay is more in dealing with releasing their copyrighted works for us to play with directly. Cutting through red tape is one of the longest, most drawn out processes in any industry. That's my guess, anyway.
Only a guess, but perhaps it's taken a backseat to work on the serious bugs in the game that popped up on the last update? There may not be that many people working on PC currently, and they may also need to make sure this next dlc pack doesn't cause even more problems. I'm not seeing it as them wilfully ignoring fulfilling the promise to release it, we don't know what all is going on in the background right now...
HeatherG
Volunteer Moderator
I’m sure they’re working hard on many things at the same time. Good things are worth waiting for [happy]
Very true!
Just refreshed the page again, still says "These will be available for download soon" -- and day before yesterday got a "very soon" from Chante, thanks! Very much looking forward to these . . . [uhh]
Apparently the Cobra engine which they use for all there games, has been used and in continual development since 1988.
I would love to have seen the code for the vector libraries back then, It was all either HEX or Mnemonics. (at least binary was long gone, That was so slow to program)
But nevertheless they should be fairly familiar with there engine after 30 years.
Last edited:
Vampiro
Volunteer Moderator
New year wish for 2019: Material example!I hope for both finished view and a view of the files used (how it looks like in the files). Also for combinations of the more rare types. We usually use simple stuff like RN, SM, EM...
Show us how to maximize Planet Coaster!
[yesnod]
Yes!!
I have to admit it brings the community together as we're daily sharing our findings on how to make things better and sharing tips and trics. Still it would be SO much easyer if we have a few examples of how Frontier is handling stuff.
Maybe with more helpful hints and tools (examples) right from the beginning the game would also run smoother. Currently all the TMT objects are results of more or less trials and errors with the effect the game need endless time to start and breaks when you're working on an important part of your park.
It's not necessarily the TMT that makes your game load slower. If you have dozens of TMT object, especially those that are done with no grasp of the importance of LODing, those eat away on performance, that's not down to the TMT.
However.. those example scenes.. i don't think anyone will find them helpful by now.
What i personally would rather see is a COMPREHENSIVE and EXPLICIT guide on the workflow steps that TMT requires. The guide is like one of those useless books that scratch on the surface of a topic but never give a true insight. And in a lot of places the info given isn't even accurate.
I've just tested (upload failed with a timeout notice) an asset that closely scratches on the max. spec. but apparently that LOD "guidance" is useless. I can now waste another day trying to fix something that SHOULD work fine according to the guidelines.
Here'S an example "LOD 5 - 10%-0%" that would be a margin between 800 and 0 tris. This is misleading, why not put an explicit max. tri-budget there? I doubt 800 tris for the farthest LOD is anywhere near useful. However, the model i'm working on does need a few tris even in the lowest LOD, but i currently don't even know whether the uipload fails due to one of the LODs being over-target OR if it's some other hickup.
Also, it would be GREAT if someone from Frontier could go out of their way to explain to the average PlanCo user how a 3D model with textures just ain't the same as a Blueprint (and therefore naturally has a bigger filesize) because i'm getting tired of being rendered stupid by people who don't even get the difference between the two. Instead it was stated how filesizes should be kept small without further specifying on the matter.
The TMT is great in its conception, but you are currently doing EVERYTHING to make it impossible to feel ok with the way things are being handled. Us modders we get all the blame and the beating and the ones who should come forward to clear up some confusion and to make it a little easier on us, they just shy away, waiting out the storm in their hidden shelter.
Who wouldn't be mad if they had to completely change over and adjust their workflow while working blindfolded basically. And what adds insult to injury, the TMT was much more stable and reliable when it was released, now it's become a coding-mess with any new upload attempt being a game of Roullette.
However.. those example scenes.. i don't think anyone will find them helpful by now.
What i personally would rather see is a COMPREHENSIVE and EXPLICIT guide on the workflow steps that TMT requires. The guide is like one of those useless books that scratch on the surface of a topic but never give a true insight. And in a lot of places the info given isn't even accurate.
I've just tested (upload failed with a timeout notice) an asset that closely scratches on the max. spec. but apparently that LOD "guidance" is useless. I can now waste another day trying to fix something that SHOULD work fine according to the guidelines.
Here'S an example "LOD 5 - 10%-0%" that would be a margin between 800 and 0 tris. This is misleading, why not put an explicit max. tri-budget there? I doubt 800 tris for the farthest LOD is anywhere near useful. However, the model i'm working on does need a few tris even in the lowest LOD, but i currently don't even know whether the uipload fails due to one of the LODs being over-target OR if it's some other hickup.
Also, it would be GREAT if someone from Frontier could go out of their way to explain to the average PlanCo user how a 3D model with textures just ain't the same as a Blueprint (and therefore naturally has a bigger filesize) because i'm getting tired of being rendered stupid by people who don't even get the difference between the two. Instead it was stated how filesizes should be kept small without further specifying on the matter.
The TMT is great in its conception, but you are currently doing EVERYTHING to make it impossible to feel ok with the way things are being handled. Us modders we get all the blame and the beating and the ones who should come forward to clear up some confusion and to make it a little easier on us, they just shy away, waiting out the storm in their hidden shelter.
Who wouldn't be mad if they had to completely change over and adjust their workflow while working blindfolded basically. And what adds insult to injury, the TMT was much more stable and reliable when it was released, now it's become a coding-mess with any new upload attempt being a game of Roullette.
The Timeout doesn't mean it failed because there is a problem with your Model. It just means the Communication timed out. If you had tried again later it would have worked probably.
I tried it 20 times.. i think that can count as trying multiple times.
I found the error, though, it has to do with the polycount in connection with the ASCII Format.
As far as i can tell the conversion tool now prefers Binary for bigger models that scratch the 8k limit for LOD0.
With ASCII (yes i had the audacity to try it again today, just to see if the same error comes up) it will time out now.
This wasn't an issue before, but my point is, what is so complicated in just making a simple thread with some simple changelog, just so we have a source of information. Even if it just said: Jan. X/X/XX code change / code tweak/optimiuzation (or) bugfixing attempt. Would at least be something more than just the usual surprises that keep happening with every change to the conversion tool without anyone knowing something has been done to the tool. I find it very tiring to find out a change has been made after i spent days on creating an asset with a given workflow, just to find out afterwards said workflow is not working anymore
I hope you and everyone at Frontier had an excellent time over the holidays! Once you've recovered from that, I'd appreciate it if someone could let us know the progress on this. And maybe an ETA?
Do not be so impatient. Finally, the Thememakers Toolkit was released just 6 weeks ago. [yesnod] Please excuse the little irony.
[heart]
And in those six weeks people have done amazing things, even through trial'n'error. The examples should still be extremely helpful:
Look at what this image alone tells us - "good" topography (all-quads, no poles more than 5) is unnecessary, multi-poles and some tris are fine. Don't see any n-gons, but "select by trait" in Blender would show that, if we had the example scenes.
I have less skin in the game since it's going to be a while before I have time/energy to get back into this. But there are plenty of people who could currently use the info that we were supposed to have, even given holiday time, over a month ago. [sulk]
N-gons seem okay too, provided it's a very simple object and the mesh doesn't need to flex for animation purposes. I've been making my signs with the front and back face as n-gons - it makes the whole build process for these much easier. I could technically add the extra edges necessary to make everything quads at the end of my build, but I've just been skipping that as it doesn't seem to make any difference in the final product whatsoever. (The tri count is the same regardless.)
But yeah, anything more complicated requires better topology. UV unwrapping an n-gon on a complicated mesh sometimes ends up with weird distortions, and trying to animate it would be even worse.
The questions if the mesh in photos are the in studio meshes, and not actual in game mesh?
I read that game engines usually read only triangles, so the export would convert all quads and ngons to triangles.
The user guide also talk about triangles as such. So it could be the case for PlanCo too.
So maybe the TMT site convert into triangles also as a part of the whole "building" process.
so ngons would be converted into "many" triangles (depending on the number of vertices), quads converted into 2 triangles.
I read that game engines usually read only triangles, so the export would convert all quads and ngons to triangles.
The user guide also talk about triangles as such. So it could be the case for PlanCo too.
So maybe the TMT site convert into triangles also as a part of the whole "building" process.
so ngons would be converted into "many" triangles (depending on the number of vertices), quads converted into 2 triangles.