PDA

View Full Version : Missing attachment point?



nfitzsimmons
December 18th, 2017, 12:48 PM
I'm back in asset-fixing mode in T:ANE, and I've running across a number of older assets that show up as faulty because of a missing attachment called a.limback. Problem is, I can't find any reference to that attachment point anywhere in the confix.txt for any of these assets. Can someone tell me what might be causing this?

These assets are error-free in 2012.

clam1952
December 18th, 2017, 01:22 PM
a.limback and a.limfront are necessary to couple to the next or previous vehicle, they won't work without them, they are never referenced in the config.

Use pevs mesh viewer and check the attachment names, probably an error in the mesh as in a typo in which case not a lot you can do without editing the actual mesh.

Got a kuid?

SailorDan
December 18th, 2017, 02:50 PM
Use pevs mesh viewer and check the attachment names, probably an error in the mesh as in a typo in which case not a lot you can do without editing the actual mesh.
FWIW a.limback etc can be added as attached meshes - they don't have to be in the default mesh. Very handy when creating a traincar from a scenery item.

clam1952
December 18th, 2017, 03:15 PM
FWIW a.limback etc can be added as attached meshes - they don't have to be in the default mesh. Very handy when creating a traincar from a scenery item.

That's one I never knew about! I'm wondering though how it worked without one if it's actually missing and not a glitch.

nfitzsimmons
December 18th, 2017, 06:26 PM
Looked at Ben's work car 21B. It's a pile driver and doesn't have the attachment on the end with to driver rig. Looks fine in 2012, but apparently will never work in T:ANE if there's an error check that requires a coupler on each end. That's awkward. Looks like the same issue with everything that reports this error.

SailorDan
December 18th, 2017, 06:55 PM
Looked at Ben's work car 21B. It's a pile driver and doesn't have the attachment on the end with to driver rig. Looks fine in 2012, but apparently will never work in T:ANE if there's an error check that requires a coupler on each end. That's awkward. Looks like the same issue with everything that reports this error.

I think you can get it to work in TANE as a static vehicle, which might be OK for a pile driver. There are some cranes that are set up the same way. But adding the required attachments using Attachment Maker is easy - just be sure to get everything in the correct sequence and pointing the correct way - use any existing traincar as an example to see what
the layout needs to be.
http://trainz.shaneturner.co.uk/tutorials/index.php/home/utilities/pevsoft-trainz-tools/attachment-maker

One problem you might run into is shared textures between the attachment mesh and the shadow. If they are both PEV creations then they both use a texture that isn't actually used, but still throws a shared texture error in T:ANE. You need to either make the two texture properties the same, or change the name of one of the textures. Really annoying, considering that the textures don't actually exist.

nfitzsimmons
December 18th, 2017, 07:12 PM
Also annoying in that while Ben's asset could be made into a static object, some of the others I would like to use shouldn't be. I wonder why T:ANE requires two coupler points.

PWeiser
December 18th, 2017, 10:07 PM
Also annoying in that while Ben's asset could be made into a static object, some of the others I would like to use shouldn't be. I wonder why T:ANE requires two coupler points.

I'd guess it's to determine the space/distance it takes along the track.

If you ever have an object which must not (i.e. cannot) couple at one end, there's a slightly involved script solution: in effect, as soon as the script notices that it's coupled on that end it uncouples. Happens so fast the car seems to simply bounce away...

ianwoodmore
December 18th, 2017, 10:53 PM
I've noticed that missing a.limfront and a.limback problems occur more frequently when the meshes are not local eg aliased meshes. In these instances the attachment points are not actually missing, just not found by the validation.

p-dehnert
December 19th, 2017, 01:51 AM
Ben's work car 21B has both attachment points named a.limfront.

Peter

SailorDan
December 19th, 2017, 01:57 AM
Ben's work car 21B has both attachment points named a.limfront.
That's probably an easy fix. Just rename one of them to a.limback. Be sure to choose the right one, and if required reverse its orientation. Use a known-good traincar as your guide. The two attachment points must be correct with respect to the direction of the Y-axis of the mesh, and must point in the right direction.

p-dehnert
December 19th, 2017, 03:52 AM
Changing the name of attachment points is easy using TrainzUp IM Editor (https://yadi.sk/d/Yp1zqNucmcdWd), but the tool could be difficult to install.
Using an hex editor, the following change might work:
part_1.im at offset 0x10AE28
original: 0C 00 00 00 61 2E 6C 69 6D 66 72 6F 6E 74 00
new.....: 0C 00 00 00 &1 2E 6C 69 6D 62 61 63 6B 00 00


Peter

PWeiser
December 19th, 2017, 09:26 AM
Changing the name of attachment points is easy using TrainzUp IM Editor (https://yadi.sk/d/Yp1zqNucmcdWd), but the tool could be difficult to install.
Using an hex editor, the following change might work:
part_1.im at offset 0x10AE28
original: 0C 00 00 00 61 2E 6C 69 6D 66 72 6F 6E 74 00
new.....: 0C 00 00 00 &1 2E 6C 69 6D 62 61 63 6B 00 00


Peter

I try not to do things like this in public... it's like cowboys who used to brand all their cattle with a running iron: those who restrict themselves to a regular branding iron look at them askance. :hehe: