zanoni-mbdyn / blendyn

MBDyn (https://www.mbdyn.org/) graphical post-processor for blender (https://www.blender.org/)
GNU General Public License v2.0
40 stars 8 forks source link

cancel old frames when importing new results from a simulation #6

Closed louisgag closed 7 years ago

louisgag commented 7 years ago

This issue can count towards the entry exam for the Blendyn related projects for MBDyn-GSoC2017

This is more a feature request than an issue.

When importing the data into an already animated scene but using a newly created .nc or .mov file which ends before the previous import, the frames that come after the last timestep (may depend also on import frequency) of the new results will remain as they were before and this creates confusion by mixing results from two simulations. An ideal behavior would be that the nodes/elements frames that are present in the new results file get cleared before importing the new results.

Andrea may want to give his point of view on this one...

zanoni-mbdyn commented 7 years ago

I totally agree, and this should be straightforward.

Cheers Andrea

Il 14 mar 2017 17:36, "Louis Gagnon" notifications@github.com ha scritto:

This issue can count towards the entry exam for the Blendyn related projects for MBDyn-GSoC2017

This is more a feature request than an issue.

When importing the data into an already animated scene but using a newly created .nc or .mov file which ends before the previous import, the frames that come after the last timestep (may depend also on import frequency) of the new results will remain as they were before and this creates confusion by mixing results from two simulations. An ideal behavior would be that the nodes/elements frames that are present in the new results file get cleared before importing the new results.

Andrea may want to give his point of view on this one...

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/zanoni-mbdyn/blendyn/issues/6, or mute the thread https://github.com/notifications/unsubscribe-auth/AHc7yNVXdJhGChj3Sn8ctbmPraYX2-frks5rlsH-gaJpZM4Mc1ST .

janga1997 commented 7 years ago

@louisgag

An ideal behavior would be that the nodes/elements frames that are present in the new results file

Do you mean frames/nodes present in the scene?

louisgag commented 7 years ago

@janga1997 I mean that the elements that get imported from the new results file (.mov or .nc), whatever they may be (joint, node, etc) get their animation data (keyframes, etc) removed before the import

janga1997 commented 7 years ago

@louisgag I don't get it still. You want the all the motion data, (x, y, z, alpha, beta, and gamma) of the new .mov/.nc file , to be cleared before import, and just the nodes/joints/elements to be imported? But I don't understand what is the point of importing a new results file, if not updating the motion data (keyframes) ?

louisgag commented 7 years ago

Try this and you'll probably understand my issue:

  1. run mbdyn with final time of 10 seconds.
  2. import the results into Blendyn
  3. change the mbdyn input file to change the behavior (example: multiply gravity by 2)
  4. change the final time to 5 seconds
  5. re-run mbdyn
  6. import the new results in Blendyn

you'll see that the previous results still show up in the animation for period 5-10 and make it incoherent..

Regards,

On 15-Mar-2017 09:18 AM, VSN Reddy Janga wrote:

@louisgag https://github.com/louisgag I don't get it still. You want the all the motion data, (x, y, z, alpha, beta, and gamma) of the new .mov/.nc file , to be cleared before import, and just the nodes/joints/elements to be imported? But I don't understand what is the point of importing a new results file, if not updating the motion data (keyframes) ?

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/zanoni-mbdyn/blendyn/issues/6#issuecomment-286670905, or mute the thread https://github.com/notifications/unsubscribe-auth/ALuOffu80rnO4LNg0Mi8Tat61lqIH2PFks5rl57jgaJpZM4Mc1ST.

-- Louis Gagnon, Ph.D. Postdoctoral fellow Dipartimento di Scienze e Tecnologie Aerospaziali Politecnico di Milano +39 02 2399 8642 http://louisgagnon.com/research/

janga1997 commented 7 years ago

@louisgag I understand the problem of different time intervals. What i am still not clear about, is do we want to clear the new data of 5 secs, or what i think you mean, is to clear the old data of 10 secs?

louisgag commented 7 years ago

We want that no old data shows after importing new data for an element

On 15-Mar-2017 10:05 AM, VSN Reddy Janga wrote:

@louisgag https://github.com/louisgag I understand the problem of different time intervals. What i am still not clear about, is do we want to clear the new data of 5 secs, or what i think you mean, is to clear the old data of 10 secs?

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/zanoni-mbdyn/blendyn/issues/6#issuecomment-286681001, or mute the thread https://github.com/notifications/unsubscribe-auth/ALuOfcdzWGHclkyfgLElQ_h4wLjbWrS-ks5rl6nAgaJpZM4Mc1ST.

louisgag commented 7 years ago

It works, but it does break compatibility with old models. I guess I'll have to live with that ;-)

zanoni-mbdyn commented 7 years ago

How it does so, Louis? I haven't tried to run it with an old model, to be honest.

louisgag commented 7 years ago

@zanoni-mbdyn Actually, I've looked into the error message properly to reproduce the bug and realized the issue is different. With this new version I cannot change the location of the result file as it will still look at the old directory for the log file. For my personal purposes I can fix this creating a link, but the old Blendyn version did not have this behavior (not sure how far back though). Should I open a new GSoC issue?

zanoni-mbdyn commented 7 years ago

Yes, go ahead... I don't remember to have changed related to that portion of the code recently, though. I don't know why this is happening.

louisgag commented 7 years ago

I've tried reproducing the bug using a model created with the current version of Blendyn and the issue doesn't occur. It's probably something related to an old Blendyn (or more likely an old mbdyn-blender) release. I thus won't open an issue.