Message Areas - "New conv3ds and the age-old broken shading problem"

Train-Sim.Com Message Areas

"New conv3ds and the age-old broken shading problem"

Email this topic to a friend
Printer-friendly version of this topic
Bookmark this topic (Registered users only)
Previous Topic | Next Topic Conferences MSTS Locomotive Design (Protected) Original message
chrisell (226 posts)
Feb-06-02, 08:33 PM (EST) "New conv3ds and the age-old broken shading problem" I just downloaded the latest patch for MSTS which I was overjoyed to see had a new version of conv3ds.exe in it. Better yet, in the release notes, it proudly tells me:* Fixed problem caused by triangles with coincident vertices breaking the face/vertex normal generationand* Fixed problem where invalid vertex normals were being generated in some cases.I thought this would mean they'd finally solved the age-old issue of co-spatial vertices in the .3ds file causing shader errors in MSTS. Sadly, they don't appear to have fixed either of the two items mentioned in the release notes. Conv3ds.exe is now v1.06 so I know I'm using the latest one.So has anyone else managed to see any change in their trains with this new conv3ds (apart from larger .s files for some reason).The odd thing is - in some cases, the shading problem actually seems to be worse now than it ever was....

Chris Longhurst. Read the mini FAQ at: http://www.christrains.com/
Alert Edit | Reply | Reply With Quote | Top
Table of Contents

Subject Author Message Date ID RE: New conv3ds and the age-old bro... Dennis Feb-06-02 1 RE: New conv3ds and the age-old bro... johnf Feb-07-02 2 RE: New conv3ds and the age-old bro... chrisell Feb-07-02 3 RE: New conv3ds and the age-old bro... SP9800 Feb-07-02 4

Lobby | Topics | Previous Topic | Next Topic
Messages in this topic
Dennis (105 posts)
Feb-06-02, 11:45 PM (EST) 1. "RE: New conv3ds and the age-old broken shading problem" Hmmmm, in that case, no patch for me then!I still have to test things, because how much does the patched conv3ds.exe influence the unpatched
simulator since not everyone will use the patch with all the problems it causes.So we will have two different type of users from now on, with patch and without patch.Did you only use the patched tools? Maybe it will only work in combination with a patched simulator?Dennis - 3DTrains
http://www.3dtrains.com Remove | Alert Edit | Reply | Reply With Quote | Top
johnf (305 posts)
Feb-07-02, 00:55 AM (EST) 2. "RE: New conv3ds and the age-old broken shading problem" I have not noticed the texture diference but it does seem to rationalise pivots better and the new makeacewin.exe with compression reduces ace size by 10 times. I made a 256x256 wheel texture tonight that with the old make ace was 220K now its 20K and with out alpha it was 3K. Thats big news for sceenery and route design types. Remove | Alert Edit | Reply | Reply With Quote | Top
chrisell (226 posts)
Feb-07-02, 11:15 AM (EST) 3. "RE: New conv3ds and the age-old broken shading problem" Re: patched vs. unpatched - the new tools make converting the models easier - you don't have to go through that nonsense of renaming short-named pantograph nodes (pantotop1 etc)- conv3ds does it for you now. Strangely, their naming convention exactly matches that in my FAQ - word for word, typo for typo. I wonder if they've been reading people's sites.....?Re: ace file sizes - I'd been using acecomp before the new makeacewin appears so I'd been having dxt-compressed textures anyway. This just removes that further step.
Chris Longhurst. Read the mini FAQ at: http://www.christrains.com/
Remove | Alert Edit | Reply | Reply With Quote | Top
SP9800 (183 posts)
Feb-07-02, 01:16 PM (EST) 4. "RE: New conv3ds and the age-old broken shading problem" I believe ACEComp uses ZLib compression, as its creator was not having luck getting DXT to work. DXT is lossy anyway, while ZLib is a lossless format. Remove | Alert Edit | Reply | Reply With Quote | Top

Lock | Archive | Remove

Lobby | Topics | Previous Topic | Next Topic