Some general ChronoSculpt questions

szabolcs

New member
Hi,

While testing ChronoSculpt I had a few issues that I couldn't solve or find the answer to in the documentation.

- How can I play the animation? I see the FPS setting but haven't find a play button or hotkey or any info about it in the PDF.

- How can I remove the affect of a Setup-Mode sculpt? At various times the erase brush in setup mode does nothing or translates the surface to weird places.

- How can I tone down the effect of a whole morph to see what it does to the mesh? Can I 'mute' or disable it somehow?

- Is it possible to give an explicit name to a morph? Would be nice to communicate what it does.

- Would be useful to visualize the area that was affected by a morph. Its hard to find what was changed while sculpting here and there on the mesh.

- Is it possible setting a viewport display color for an object? I saw in the youtube video that in the cloth sim fixing part the cloth was blue but I don't know how it was set.
It would be extremely useful setting contrasty colors for objects when fixing object intersections with simulations.

- Is it possible to display more objects at once with normal shading?

- Transformation from Alembic is not working just deformation. Is it a known issue or is it intended?

- I found some lua script files in the install directory, that suggests that CS can execute script files / commands. How does that work? There is nothing about it in the documentation.

Cheers,
Szabolcs Horvatth
 
Hi there, apologies for the delays lately! We're currently working on porting the software which has been quite time-consuming, but is hopefully a cost that will not accumulate much once we get the initial pre-releases out the door for OSX and Linux.

How can I play the animation? I see the FPS setting but haven't find a play button or hotkey or any info about it in the PDF.

ATM the only way is to use the arrow keys or manually scrub the timeline by dragging on it, but transport controls are the absolute next thing on the list once we get the ports out the door!

How can I remove the affect of a Setup-Mode sculpt? At various times the erase brush in setup mode does nothing or translates the surface to weird places.

The erase tool ATM is not the way to do this. The erase tool is actually very simplistic in function: it does not actually remove deformations but creates a new deformation to compensate which moves points back towards their base cache positions (moving them back to the original cache level). This is not the ideal but is currently the only technically-feasible way we've found to implement it so far.

However, your question does suggest a very nice feature, and that would be to have erase functioning differently in edit/global contexts. In those contexts, it may be more sensible to move the global positions of the base mesh back to what they were originally. We'll consider doing that -- it would be quite practical to do.

How can I tone down the effect of a whole morph to see what it does to the mesh? Can I 'mute' or disable it somehow?

Strength controls were something we've discussed before. It would be a field that lets you adjust the strength (or opacity if you like) of the deformer. That would allow you to increase/reduce the effect it has on the mesh as well as setting it to 0% to disable the effect without having to remove it. What do you think about this idea?

Is it possible to give an explicit name to a morph? Would be nice to communicate what it does.

I think that's a great idea. ATM there is actually a name stored for each morph (doesn't have to be unique) and you can actually edit those names by editing the CS projects you save out (they're plain old text files). However, we lack the UI controls to do this ATM, but can provide them instantly. Basically it's something we anticipated might be necessary but wasn't sure if people wanted (hence we do name morphs but don't provide the UI controls to rename), but this question/suggestion to me is a good indication that it would be worthwhile to provide it. Providing a string field to rename the selected deformer would be no problem at all.

Would be useful to visualize the area that was affected by a morph. Its hard to find what was changed while sculpting here and there on the mesh.

The only difficult part about visualizing the points is that it's a bit difficult to do efficiently. We do this for the pin deformer since the pin deformer simply stores a list of indices of the points being affected, but morph deformers, for example, use a much more complex representation to allow them to be applied efficiently and with variable strengths depending on the point in question. Basically we want to do it, but have some optimization issues to overcome to do it without slowing down the overall program. As a possibility, we thought about showing the effect only when you hover highlight over a morph and not when selecting it. This would avoid slowing down framerates doing normal operation since the points being affected would only be drawn on a hover event.

Is it possible setting a viewport display color for an object? I saw in the youtube video that in the cloth sim fixing part the cloth was blue but I don't know how it was set.
It would be extremely useful setting contrasty colors for objects when fixing object intersections with simulations.

In those cases an LWO was used. We recognize basic surface information from LW object files but currently lack this for Alembic and OBJs. We also lack the ability to change the surface materials after loading which could be a useful feature for visual indicators, but presents a bit of a challenge if we want to preserve the materials across on export. This is a question best answered by our interchange developer, David Forstenlechner since the main challenges relate to import and export.

Is it possible to display more objects at once with normal shading?

At the moment we're using the ghosted shading (screen door transparency) for non-selected, background-type objects, but we were thinking of making it so when we get the play button with the transport controls we're doing next, everything would be shown during playback in their normal, fully-shaded form. If that does not suffice to simply show the full shading on playback, then we could look into more options to allow you control how unselected things are displayed.

Transformation from Alembic is not working just deformation. Is it a known issue or is it intended?

I believe it is a known issue but this is also best answered by our interchange developer. ATM ChronoSculpt lacks a full-fledged motion system which would be necessary to import Alembic objects with transformations without baking the transformations into the vertex positions themselves. A motion system would be the ideal answer to this problem but baking might be a start.

I found some lua script files in the install directory, that suggests that CS can execute script files / commands. How does that work? There is nothing about it in the documentation.

ATM we use Lua as an internal development tool to streamline some of the internal development, but have no immediate plans to make this a generalistic scripting solution. We do have longer term plans to provide scripting to users in some form, but whether we expand on this internal Lua solution or use Python or both (my personal favorite) is currently undecided. For now we use Lua in the same way we use C++ in various areas to develop different parts of the product internally.
 
Last edited:
I have a couple of questions.

Are the scultped corrections simply a vector offset in world space or is there some kind of taking into consideration of tangent and normal.
For example, if you scultped the puffer fish to be fatter. If that fish turned around 180 degrees to face the opposite direction of travel, would it stay fat, or would it collapse inward as the offset vectors would do if in world space?

secondly, If you make multiple correctives on, say, a face in different areas of the face, such as brows, cheeks, lips etc, and these corrections overlap which of course they will, Is there a way of visualising these as layers on the timeline, or is there one timeline that they all have to fit into and overlap onto?
 
Are the scultped corrections simply a vector offset in world space or is there some kind of taking into consideration of tangent and normal.

They're the former: similar to a relative endomorph in LW. However...

For example, if you scultped the puffer fish to be fatter. If that fish turned around 180 degrees to face the opposite direction of travel, would it stay fat, or would it collapse inward as the offset vectors would do if in world space?

For this I think we do best by having an actual motion system in CS. It would solve some interchange issues and you'd be allowed to rotate and transform the fish to your heart's desire without any disruption of the sculpted morph deformers being applied.

We did play with a normal space implementation originally, but it's far more computationally expensive; the software would inevitably take a non-trivial performance hit. If we find though that there are cases that cannot be solved by having a combination of motion systems + relative morphs and need the full normal-space morph solution, I think we could add that in as an option. Bones actually don't need this if CS ever supports bone deformations.

secondly, If you make multiple correctives on, say, a face in different areas of the face, such as brows, cheeks, lips etc, and these corrections overlap which of course they will, Is there a way of visualising these as layers on the timeline, or is there one timeline that they all have to fit into and overlap onto?

Currently they would all be displayed on the same morph track in the timeline. We want to have more ways for the artist to organize his work starting with more modest features like color correction and the ability to custom name your morphs, but expanding towards displaying each and individual deformer on a separate track and/or having a master list to find and select them by name seems like a natural next step in regard to project organization.
 
Strength controls were something we've discussed before. It would be a field that lets you adjust the strength (or opacity if you like) of the deformer. That would allow you to increase/reduce the effect it has on the mesh as well as setting it to 0% to disable the effect without having to remove it. What do you think about this idea?

Muting a deformation should be a single toggle. Having to slide a slider is too much work for something that a user expects to be a boolean choice. In AfterEffects there's a fx toggle that'll turn off all the fx associated with that layer. Your strength control suggestion would be as if I had to dial down the opacity of an fx to not see it. That's too much manual work. Plus what if I animated the opacity of my fx. I can't use the opacity to both animate and temporarily disable the fx. Also if need be, I can toggle individual fx in addition to the global toggle for that layer. All of this is independent of any opacity channels I have on the fx for the layer.
 
Hi,

Thanks for the detailed answer!

We're currently working on porting the software which has been quite time-consuming, but is hopefully a cost that will not accumulate much once we get the initial pre-releases out the door for OSX and Linux.
Sounds great, getting a linux version out the door will widen your user base considerably.


ATM the only way is to use the arrow keys or manually scrub the timeline by dragging on it, but transport controls are the absolute next thing on the list once we get the ports out the door!
Sounds good. I was pretty sure there was a way to play the animation because of the FPS setting in the GUI...


The erase tool ATM is not the way to do this. The erase tool is actually very simplistic in function: it does not actually remove deformations but creates a new deformation to compensate which moves points back towards their base cache positions (moving them back to the original cache level). This is not the ideal but is currently the only technically-feasible way we've found to implement it so far.
Well, if the counter-deformation puts back the vertices to the original, undeformed place than its fine by me. :)


However, your question does suggest a very nice feature, and that would be to have erase functioning differently in edit/global contexts. In those contexts, it may be more sensible to move the global positions of the base mesh back to what they were originally. We'll consider doing that -- it would be quite practical to do.
I'm glad you like the idea! It was kind of intuitive (and kind of unintuitive that it did not work that way... :))


Strength controls were something we've discussed before. It would be a field that lets you adjust the strength (or opacity if you like) of the deformer. That would allow you to increase/reduce the effect it has on the mesh as well as setting it to 0% to disable the effect without having to remove it. What do you think about this idea?
I'd prefer having both an intensity control as a slider, probably between 0-1 (to be able to tone down the deformation afterwards), and a toggle to enable and disable the morph (to check the difference and to be able to do different sculpt tests, for example to present variations for dailies). The disble checkbox can be easily mapped to a hotkey (like "d" in Nuke) and modifying it wouldn't affect the already set morph intensity setting.


ATM there is actually a name stored for each morph (doesn't have to be unique) and you can actually edit those names by editing the CS projects you save out (they're plain old text files). However, we lack the UI controls to do this ATM, but can provide them instantly. Basically it's something we anticipated might be necessary but wasn't sure if people wanted (hence we do name morphs but don't provide the UI controls to rename), but this question/suggestion to me is a good indication that it would be worthwhile to provide it. Providing a string field to rename the selected deformer would be no problem at all.
That would be nice to have in the GUI.


The only difficult part about visualizing the points is that it's a bit difficult to do efficiently. We do this for the pin deformer since the pin deformer simply stores a list of indices of the points being affected, but morph deformers, for example, use a much more complex representation to allow them to be applied efficiently and with variable strengths depending on the point in question. Basically we want to do it, but have some optimization issues to overcome to do it without slowing down the overall program. As a possibility, we thought about showing the effect only when you hover highlight over a morph and not when selecting it. This would avoid slowing down framerates doing normal operation since the points being affected would only be drawn on a hover event.
I like the 'showing the effect when hovering over' solution. You could even use the colors of the various morphs to be able to differentiate if its contrasty enough.


We also lack the ability to change the surface materials after loading which could be a useful feature for visual indicators, but presents a bit of a challenge if we want to preserve the materials across on export. This is a question best answered by our interchange developer, David Forstenlechner since the main challenges relate to import and export.
I see... Well, it might be a feature for a later release. But there is a lot to be done in the interchange field, at least the Alembic side, and this is a less important feature.


At the moment we're using the ghosted shading (screen door transparency) for non-selected, background-type objects, but we were thinking of making it so when we get the play button with the transport controls we're doing next, everything would be shown during playback in their normal, fully-shaded form. If that does not suffice to simply show the full shading on playback, then we could look into more options to allow you control how unselected things are displayed.
Showing things shaded when playing sounds good to me.


ATM ChronoSculpt lacks a full-fledged motion system which would be necessary to import Alembic objects with transformations without baking the transformations into the vertex positions themselves. A motion system would be the ideal answer to this problem but baking might be a start.
Baking is a suitable workaround until a more efficient solution is done. But for things like rigid body sims baking the transformation into deformation expands the datasize quite a bit.


ATM we use Lua as an internal development tool to streamline some of the internal development, but have no immediate plans to make this a generalistic scripting solution. We do have longer term plans to provide scripting to users in some form, but whether we expand on this internal Lua solution or use Python or both (my personal favorite) is currently undecided. For now we use Lua in the same way we use C++ in various areas to develop different parts of the product internally.
From a production and pipeline point of view I'd vote for Python: its really widespread, used in lots of packages and most studios have some asset management / pipeline stuff written in python that could be used in SC too.

Cheers,
Szabolcs
 
This is GREAT NEWS that you all are working on the pre-release for OSX!!! I purchased ChronoSculpt (and NevronMotion) during the Siggraph deal period, both for Mac OSX - the only system I've used Lightwave3D on since the end of the Lightwave on the Amiga days. ;-). So, I haven't yet used ChronoSculpt even though I actually purchased it over a month ago. But that's okay with me since I'm a big Newtek supporter and have been for 20 years! :-D. I completely trust and have great faith in the dev team and the new Lightwave Group. Hope the coding for the port to the Mac is going well. Looking forward to doing some cool, .MDD cache tests in CS out of LW.
:thumbsup:
 
I have a few UI/UX-related questions about Chronosculpt:

1. Can elements of the main window be separated into their own floating window and moved to another monitor, for example the timeline and/or other panels?

2. Is it possible to have multiple "document views" open at the same time? In other words, multiple objects/object sets in their own sculpting environments, either as tabbed panes within the main window, or ideally, as separate windows (potentially on different monitors)?

3. How programmable are the key, mouse, and tablet mappings (esp. w.r.t. pressure sensitivity, etc.)? Can UI/UX buttons/menus be reprogrammed as well?

4. 3DCoat and ZBrush both allow for user-assignable surfacing/shading of objects being sculpted (among other reasons in order to better visualize sculpting given the object). What (if any) control does Chronosculpt give the user over object shading? From some of the earlier comments, it appears it might automatically load shading from the LWO (for example), but is it possible to override existing shading just while in Chronosculpt with more sculpting-friendly shading if needed? Any chance we can program or design our own shaders (as both 3DCoat and ZBrush allow)? Maybe CgFx custom shader support?

5. Does Chronosculpt support any kind of GoZ/Applink-like "automated transfer" of assets to/from other 3D pkgs? Or is it currently limited to manual load/save?

6. What level of 3DConnexion support is currently present?

7. Can scripting-versant users/devs easily add support for UI devices like LeapMotion?

I'm just trying to get a sense of how customizable the UI/UX infrastructure is currently, as I didn't see much discussion of such features in the various demos (if they were, please point me to where they were discussed?). I definitely make use of the UI/UX customization features in 3DCoat and ZBrush, so that kind of functionality is important to me.
 
Just to clarify, while I ask about "current" in several places, I guess I really mean "by release" -- as in, I understand it is currently incomplete, but w.r.t. features like 3DConnexion support, would be okay waiting for it as long as it shows up by the formal release.
 
Nobody here with Chronosculpt is willing to try and answer my UI/UX questions?

I'm gathering the shading options are pretty slim, though I'd still like details on exactly how it handles shading currently? The one that concerns me most is the keyboard/mouse/etc. reprogrammability, as I use that a LOT in both 3DCoat and ZBrush to keep my commonly used tools available on my little sidecar keyboard (a Logitech G13) with common keystrokes across all such apps.

Absence of cmd reprogrammability and/or lack of 3DConnexion support would still likely be big deal-breakers for me (presuming Mac port shows up), which is why I'm so interested in level of support present in those areas. The others I could probably live without, though I feel pretty strongly the inability to have multiple docs open at once would be a serious UI/UX mistake.
 
"...getting a linux version out the door will widen your user base considerably."

I seriously doubt this. But I'm willing to be convinced: what percentage of LW users run LW under Linux?
 
Why LightWave ? It's a full-fledged application.

And then it's a step for those who wish to see LW on Linux. Personally I'm not interested but we must recognize that it would be a good thing. It would also bring new users to no doubt :)
 
I have a few UI/UX-related questions about Chronosculpt:

1. Can elements of the main window be separated into their own floating window and moved to another monitor, for example the timeline and/or other panels?

no

2. Is it possible to have multiple "document views" open at the same time? In other words, multiple objects/object sets in their own sculpting environments, either as tabbed panes within the main window, or ideally, as separate windows (potentially on different monitors)?

dunno

3. How programmable are the key, mouse, and tablet mappings (esp. w.r.t. pressure sensitivity, etc.)? Can UI/UX buttons/menus be reprogrammed as well? ot

It does work great with a wacom. I don't know about any other device

4. 3DCoat and ZBrush both allow for user-assignable surfacing/shading of objects being sculpted (among other reasons in order to better visualize sculpting given the object). What (if any) control does Chronosculpt give the user over object shading? From some of the earlier comments, it appears it might automatically load shading from the LWO (for example), but is it possible to override existing shading just while in Chronosculpt with more sculpting-friendly shading if needed? Any chance we can program or design our own shaders (as both 3DCoat and ZBrush allow)? Maybe CgFx custom shader support?

nadda

5. Does Chronosculpt support any kind of GoZ/Applink-like "automated transfer" of assets to/from other 3D pkgs? Or is it currently limited to manual load/save?

no, it uses Alembic, obj and lwo. Full Alembic support for multiple mesh objects would be just as good imo, but it doesn't have that capability at the moment.

6. What level of 3DConnexion support is currently present?

dunno

7. Can scripting-versant users/devs easily add support for UI devices like LeapMotion?

I seriously doubt it, I see no scripting capabilities

I'm just trying to get a sense of how customizable the UI/UX infrastructure is currently, as I didn't see much discussion of such features in the various demos (if they were, please point me to where they were discussed?). I definitely make use of the UI/UX customization features in 3DCoat and ZBrush, so that kind of functionality is important to me.

I wouldn't really have reason to compare this app. to 3dcoat, zbrush or mudbox, it's definitely unique and has it's own way of working. Customizable UI and shortcutz aren't its forte AT ALL.
 
Last edited:
Back
Top Bottom