I am diving into the connection between Realflow and Krakatoa and hit an issue. I loaded an preexisting RF project into MAX via Krakatoa’s PRT loader. the particles are a Hybrido RPC sequence and they load with all their detailed complexity intact. however, the whole particle cloud is reversed - mirrored so that fluids that flowed to the right are now flowing to the left and visa versa. applying MAX’s Mirror modifier only seems to reverse the PRT icon. has anyone run into this before? importing the same sequence via RF’s RFRK loader does not produce this error so the issue seems to be with how Krakatoa is interpreting the data. thanks in advance for any feedback that can be offered.
It is possible that Realflow is writing files in a coordinate system that we are not expecting. That would cause the PRT Loader to interpret the axes incorrectly.
For a work-around… The “Mirror” modifier isn’t going to cut it, but you can do it very simply:
Give your object’s transform a “-100” scale in the X axis, instead of the default “100” scale.
-Right-click the ‘scale’ icon and replace 100 with -100.
That should flip the particles.
Would you be able to send me a sample RPC file that is displaying the transform problem? We can try to see what the issue is and see if we can fix it internally.
thanks for the fast reply. I wanted to reproduce the problem with a different particle sequence before writing back. I have confirmed that all of the instances of translating RF RPC sequences into 3ds MAX via Frost and Krakatoa cause this result on my system. I see now that the particles are both mirrored and rotated 90 degrees. here is a link to a specific RPC file and a screen cap of three instances of import. the orange mesh is an imported mesh object from RF and sits correctly in the scene. the pink mesh is a Frost Mesher object, mirrored and rotated during import. the green particles are Krakatoa PRT import of the same RPC file. I moved this one away from the world center for comparison. it arrived in the same exact location as the Frost object. (although the top view of the Krakatoa object looks slightly different than the other meshes, a side view confirms the exact same profile.)
Thank you for posting the rpc file. It is very helpful.
Based on this screenshot, I’m not 100% sure what transformation you are applying to the PRT Loader to get it properly match Realflow’s loader. If you select your PRT Loader, open up the maxscript listener, and type:
$.transform
What does it say?
It does seem like a bug, we will attempt to have it fixed in the next release.
I tested this in 3ds Max 2014, using Krakatoa 2.4.3 and the RealFlow ParticleLoader from RFConnect 2015.0.0.8. It worked as expected: the particles from RealFlow (pink) and Krakatoa (green) were aligned the same way:
[attachment=0]RealFlow-and-Krakatoa.png[/attachment]
Are you using the same software versions as me?
Could you please try repeating your test using the RealFlow ParticleLoader, if you have not done so already?
I setup a new scene using the RF particle loader and the Krakatoa PRT loader. because there was a different result, I revisited the original scene and can report that the 90 degree rotation was specific to the RF mesh object only. this is consistent with my initial issue - which used a different particle sequence and saw only mirroring. so throw the 90 degree detail out of the discussion.
however, in my setup I still see a mirror image of the sequence when loading via Krakatoa. in this new screen cap you can see the correct orientation in the blue RF sequence and a mirror of it in the gold Krakatoa sequence. the maxscript listener reports no difference in the transforms of either. I am using Krakatoa loaded via the KrakatoaMX_2.4.3.59385_x64.msi and rfconnect-max-2015.0.0.8.
in the meantime I’ve come across a mention of a patch for 3ds max 2014 that RF folks had to do an update to deal with. I am unsure if I’ve ever installed the patch. and autodesk’s 3ds support site is on the fritz so I can’t download it right now. are you working with a patched 2014? perhaps if one of ours is patched the other not, this might explain the difference we are seeing (?).
I patched my 2014 3ds MAX with the final SP5 patch. the results did not change. however, I revisited the mirror command and unlike the last time I attempted this (and only the PRT icon flipped), the command worked on the particles themselves. by flipping along the Y axis I was able to get the particles to sit where they were expected. it’s an easy workaround. but the cause is still a mystery. I’ll continue to explore this but at least now I can get back to work. thanks again for looking into this. let us know if you ever discover what is going on here.
