Don't worry, before I posted it, I proclaimed to all readers:

"Thou shalt not spit..."

:ping:

lo and behold, it worketh:o I dideth not spit and I wonderest why I dideth not spit. Now I knoweth:yep:

Exiteth stage left (before I'm accused of being a really bad imitator of John Cleese.):arrgh!:

Hi, Skwas.
Good work.
But not so convenient control for camera ... small (low) sensitivity for linear moving (and not convenient buttons for mouse ... maybe for rot. - left but., for linear moving - right but., zoom - wheel)...
And in "Show All" mode i would like to have functions "Hide" (and "Unhide All") for selected object ...
and maybe add smooth ...

Some people are never happy! :rotfl:
Me, I'll be toasting to skwasjer every time I'll be on an outing! Oh, and if I ever happen to be at the beach during sunset, to Dave as well.

Vacation time is upon me, I'll see you guys in September! :up:

http://www.longam.net/sh4/skwasjer/zeus.jpg

I think it's time to bump it up to coolness with some kelp sunglasses. :lol:

Bug report
 

S3D v9 bug report.



When importing a .dat file with a nod that has an embedded 3D model, SH4 will CTD when the .dat file with the new 3D model comes into visual range.


Background

I have used S3D to edit / change / import .dat files for both RSRDC / OM for a while now. I have always been able to import nodes with 3D models in the past without problem. With v9 I’m have a “re-creatable on demand” CTD when I attempt to add nodes with 3D models in them.

The following test case is the procedures I use for the importation of nodes with 3D models and have use these steps with older versions of S3D without problems.

• create working directory
• copy NDD_Clemson.dat & NTR.dat into the working directory
• launch S3D and open the NTR.dat file
• select and expand (7) Node – NTR then select and expand (12) Node – NTR_B
• select (21) NTR_barca_03 and remove “parent id” so that the node is now is a “standalone”
• Repeat step last step with NTR_barca_06.
• select (7) Node – NTR and delete
• Keep “empty” nodes 3 & 5 as they are required “material” nodes for NTR_barca_03 and NTR_barca_06 and delete the remaining “empty” node.
• Select and execute “Remap ID’s” on the file.
• save NTR.dat then use the “save as” option to save the file with the name zzz.dat
• delete “AuthorInfo”, Start, End, Unknown chunk, Index and Eof from zzz.dat
• save file zzz.dat
• open NDD_Clemson.dat
• select (7) Node – NDD_Clemson
• use the right mouse button to select “import”, then dat.file
• Select zzz.dat and select yes when ask to import 10 chunks.
• Save file.
• backup original NDD_ Clemson.dat in SH4, the copy in new file
• create test mission under “Single Missions” with the Clemson Class DD. Ensure own sub is at least 75km from Clemson Class DD.
• launch SH4 and then test mission
• head sub to the Clemson Class DD location.
• When own subs is with in range of the Clemson to be seen, SH4 will CTD.

Thanks for the detailed description. Can you also send/upload the .dat file. If I for one reason do not do it exactly like you do (or you missed a step), I may be hunting for a bug for ages. I'd like to compare it with an actual file. But the walkthrough is excellent for sure, people, take note!

Currently unable but will be able to tomorrow. I'm not currently in the same location as my development computer.

Well, no hurry. Saturday, I'm on shore leave so I can't really do much until I'm back.

I have no idea what may be the cause though. I haven't really touched that part of S3D, and while working on the fans I've used similar functionality. But who knows what odd bug you've run in to...

It is sarcasm, not to be taken literally .

Anvart, this (:rotfl: ) always means a joke.

S3D v9 bug report. Part 2
 

S3D v9 bug report. Part 2

This morning I decided to reload S3D v82 to generate a test case to show that the importation of nodes with 3D models was working correctly prior to v9. To may surprise, using the same steps that I reported earlier, SH4 CTD at the same spot.

I continue additional testing until I could get it to work as I had done with other 3D models in the pass. Once that was compete, I de-installed S3D v82 and loaded v9 and repeated the “new” set of steps that worked with v82. When run SH4 there were “no” CTD using the v9 version of the test case.

It appears that SH4 is “very sensitive” to the placement of 3D model files within the .dat file and the uses or non-uses of a “parent” node ID

Here is the “revised” set of steps I used to get the .dat file to work correctly in SH4:

( new steps are in red )

• create working directory
• copy NDD_Clemson.dat & NTR.dat into the working directory
• launch S3D and open the NTR.dat file
• select and expand (7) Node – NTR then select and expand (12) Node – NTR_B
• select (21) NTR_barca_03 and remove “parent id” so that the node is now is a “standalone”
• Repeat step last step with NTR_barca_06.
• select (7) Node – NTR and delete
• Keep “empty” nodes 3 & 5 as they are required “material” nodes for NTR_barca_03 and NTR_barca_06 and delete the remaining “empty” node.
• Select and execute “Remap ID’s” on the file.
• Open NDD_Clemson.dat file and copied the node ID of “NDD_Clemson_B”
• Added the node ID of “NDD_Clemson_B” to the parent ID slot of NTR_barca_03 and NTR_barca_06.
• save NTR.dat then use the “save as” option to save the file with the name zzz.dat
• delete “AuthorInfo”, Start, End, Unknown chunk, Index and Eof from zzz.dat
• save file zzz.dat
• open NDD_Clemson.dat
• select (71) Node cfg#W02_NDD_Clemson
• use the right mouse button to select “import”, then dat.file
• Select zzz.dat and select yes when ask to import 10 chunks.
• Noted that NTR_barca_03 and NTR_barca_06 were “after” Node cfg#W02_NDD_Clemson and where “children to “NDD_Clemson_B” node.
• Save file. backup original NDD_ Clemson.dat in SH4, the copy in new file
• create test mission under “Single Missions” with the Clemson Class DD. Ensure own sub is at least 75km from Clemson Class DD.
• launch SH4 and then test mission
• head sub to the Clemson Class DD location and arrived a location and was able o see the newly modified Clemson Class DD’s

In the following link you will find a compressed file with the following:

CTDTEST

This folder / directory has a JSGME ready test mission for recreating the problem

ORG

This folder / directory continues the original NDD_Clemson.dat & NTR.dat files

V82

This folder / directory continues the S3D v82 NDD_Clemson.dat, NTR.dat & zzz.dat files that were used to added new 3D models and works without problem.

V90

This folder / directory continues the S3D v90 NDD_Clemson.dat, NTR.dat & zzz.dat files that were used to added new 3D models and works without problem.

V90_tst2

This folder / directory continues “second” test run of the S3D v90 NDD_Clemson.dat, NTR.dat & zzz.dat files that were used to added new 3D models and works without problem.


V90_broke

This folder / directory continues the S3D v90 NDD_Clemson.dat, NTR.dat & zzz.dat files that were used using the “original” test case and “will” cause SH4 to CTD.


Link

http://files.filefront.com/s3dctdtes.../fileinfo.html

Ah I see. Yea, the import dat-file works the same as with raw chunks. The insertion point is from the node you have currently selected, so you usually just have to point to one of the nodes at the end of the file (or at the end of all child nodes of a certain node) before importing.

The ordering of individual chunks is very important. Child nodes/objects can't be in front (index wise) of a parent. Model and material chunks must also exist before a node uses them. If not done, you will be guaranteed to get a CTD.

The change of parent id's is not really necessary in between, you can also do it after importing but that's only a procedural thing. Whatever works.

Some techical yada yada:

I finally managed to debug the entire StateMachineClass controller (AnimGraphs folder). Until now, I always missed a couple of pieces of the puzzle and always missed the time to find out. It still requires some more work to integrate support into the S3D UI, but what this controller will allow you to do is 'script' all the behaviors of all 3D crew (interior), including (but not limited to) what sentence they speak, their facial expressions (normal, angry, etc), their tone of voice (normal, angry, etc), what type of (character) animation is played when specific orders are executed, where the eyes will look at during specific orders etc.

There are a whole lot of possibilities when this support is available to the modder. For instance, correcting speech/animation bugs, but also to change how the crew reacts to certain orders (ie: turn divewheel to left, play speech order 'Rudder left'), when a collision occurs, when they get wounded, when you hit/miss a torpedo, etc.

Do note that this controller does not control/contain the actual 3D (mesh) animations. It controls the state of the actual character, what animation is played when, etc, and in what sequence. Together with the already supported lipsync controller, the crew behaviors can be further improved.

Support for mesh and bone animations, which will allow full control over all the bones (parts of the body) will follow in time. Meshanim support is already in development for some time, and the character part (ie. marine.dat) is the final piece of the DAT-puzzle to integrate. Apart from a couple of minor things I've now come close to supporting the full DAT-format (currently 97-98% of all data is supported).

No release date, but I just wanted to share the good news :up: