WIP RigControl for Dawn 2.0

[Ken1171]

Ok, problem solved. The memory dots are now fully functional, with custom tooltip captions and thumbnail display when hovered. Mind you, Poser will capture the thumbs from whatever the camera is seeing when the preset was saved. You know how that works. Zoom your camera to the face when saving Face Rig memory dots to get decent thumbs!

With this I believe the dots are now complete on the code side. Of course, I will have to redo the face mapping when I have a more finished D2F version, and I will also make better images for them on the tabbed pages. The current ones are just placeholders. I also want to try adding more controls to the hands and feet, but that will have to wait, since the current D2F beta has no hand or feet controls on the figure at the moment. Maybe that will have to wait as well.

So now the tool can edit the face posing, and then store/retrieve face poses using the familiar memory dots dedicated to the face rig. I believe these are some of the most turbo-charged memory dots I have ever used. It's like having a mini Poser library built-in right into the tool.

In addition, all your saved presets are PORTABLE, meaning you can easily copy them to your Poser library if you want to. They are stored at a folder called "DotPresets" on the script folder, as seen below. Just copy them to your face library, and you can use, distribute, or sell your presets straight from the tool. You just have to rename them to your liking first. Either that, or you can save the preset to the library instead of copying them from the tool folder. It's up to you.

I believe I have done everything I could with the current D2F beta state, so I will work on other tools until we have a more complete version available. ^_____^

 

[Ken1171]

A question to the community: switching the saving to FC2 face poses has solved the missing thumbnails when I manually created my own library files, but I think it can also store head morphs, right? I first thought this should only save the facial rig rotations, but I loose control of what is included when saving as FC2. I was originally creating my own PZ2 pose files that only contained the face rig rotations, but then I don't know how to get thumbnails from Poser Python. I am still investigating this, because if there is a way to get thumbs, I am going back to manually creating my own PZ2 files that only have the data I want.

The idea is that saved Face Rig memory dots should only contain face rig rotations, right? If we apply them over a character with whatever morphs applied, we probably don't want those to be affected. I have opened a ticket with the Rendo dev team to see how we can solve this. As it stands, I couldn't find a way to generate Poser standard 91x91 thumbnails on their own with Poser Python. Until then, it will keep saving memory dot presets as FC2.

 

[unreal]

See, this is the kind of craziness that happens when a programmer is heads down with code like this.

Everyone, stand back. And keep your hands and feet well clear or you could be injured.

 

[Ken1171]

I can now generate thumbnails with Python, but it's kind of clumsy, and there are some side-effects. You know whenever we render, even in preview mode, Poser auto-switches to the Render Tab? I have to render to get a thumb, so we end up in the Render Tab, and there is no Python command to return to the Preview Tab. Hence, another workaround - we switch to the Pose Room, which also resets tabs back to Preview.

So whenever we save a dot preset, we see Poser switching to the Render Tab, then back to Preview Tab. Ugh! It doesn't break anything - it's just annoying. That doesn't happen when we save things to the library, and Poser auto-creates thumbs for us. I have added this extra info to my ticket with the Poser dev team, with hopes they will offer something better for just creating thumbs without the wackiness. Fingers crossed.

I think the tool is now complete. I will add more control points when D2F is more finalized, but for now, the functionality is already in place. ^____^

 

[unreal]

How about a checkbox that allows it to switch to a pre-defined camera view to take the snapshot, then switch back? So there's a kind of "standard view" for saving thumbnails. But a person can turn that off and the save just users the current camera view.

Maybe you could add some set lighting, some render presets, even a grey texture to the figure, then flip it all back, after the thumbnail.

For an entire "render thumbnail with 'predefined scene settings".

And a pony.

 

[Ken1171]

How about a checkbox that allows it to switch to a pre-defined camera view to take the snapshot, then switch back? So there's a kind of "standard view" for saving thumbnails. But a person can turn that off and the save just users the current camera view.

Maybe you could add some set lighting, some render presets, even a grey texture to the figure, then flip it all back, after the thumbnail.

For an entire "render thumbnail with 'predefined scene settings".

And a pony.

I could offer an option to auto-frame the face, but I'd rather not touch people's materials and lighting. I have dedicated tools just for those areas, like SkinEdit and Light Dots.

 

[Ken1171]

If it CAN be automated, then why not? Here's the new "Auto-frame Thumbnails" option for FaceRig memory dots, and it is enabled by default. This means the camera will auto-frame the face to take the thumbnail snapshot, and then automatically restore your camera to where it was before. This way the thumbs will always be shot from the same angle, and you don't have to worry about camera framing when you make a dot preset. The option can be disabled if you prefer to adjust the camera manually.

 

[Ken1171]

I have reverted back to creating my own PZ2 pose files, instead of using FC2 face presets from Poser. This guarantees the saved preset will only contain facial rig rotations, and not any morphs, scale or translations. So if we apply a dot preset to a figure with morphs, those will not be affected. This also means if you make an expression that mixes morphs with facial rig, only the facial rig part will be saved on dot presets. The reason being that saving FC2 face presets cannot distinguish between facial expression dials and regular head shape morphs.

My version of the D2F beta doesn't have any morphs, so this might need more testing later on, when the morphs become available. If necessary, I can include an extra option to include morphs or not, and this will only affect the memory dots part.

One new update for today is the inclusion of the last created or accessed memory dot state, which shows in blue. This state is preserved across sessions, so you always know which dot was the last created or accessed. In older Poser versions, this color used to be orange, but in P12 they had it changed to blue, so I am using the same color here to match Poser 12.

 

[Ken1171]

For today's update, the Rendo dev team has now provided me with a new command on the Python API to just create a PNG thumbnail, which has resolved the workaround where Poser would always switch to the Render Tab every time I created a new thumbnail for the facial rig memory dots. I could now finally delete all the workaround to make a thumbnail, and it no longer switches tabs every time a dot preset is saved. This removes the most annoying part that was bugging me - kudos to the Poser dev team!

 

[unreal]

OMG, what sort of madness is this??? The dev team quickly adding features to support... errr... devs. Who are trying to support their business. By supporting their product.

Rendo needs to seriously consider some "help" for their Poser product manager. Obviously, the person is nuts and probably needs a padded room and soft squishy keyboard. Something like the ol' VT102. Mushy.

New figures. New features.
What's next? Poor innocent little bugs?

Scared.

PS: Times are tough, and if proper help for the PM is too expensive, a six-pack selection of exotic beers might work. I'll chip in

 

[Ken1171]

OMG, what sort of madness is this???

It's a matter of priorities. If I am making a Poser product that will be sold at Renderosity (the Poser owners store), and there is something in Poser that is keeping me from having it done, I get high priority in my request. This is, of course, not counting the bugs I have found in the Python API that the devs have fixed already. Conversely, any other request that is not tied to a Rendo store product being created will get lower priority.

The dev team has already provided other Python API additions that were lacking in my other scripts for Poser 12, so this was not the first time. But I didn't say I got them quickly - some took months, and a lot of cute-big-anime-eyes begging. Some I am still waiting for, but I know they are working on it because the ticket status claims they are still active.

Fortunately for this script, there is nothing I need to have it finished, besides a more complete version of the D2F beta. I want to add more visual posing control automations, but that require D2F features that are not available yet. If everything goes according to plan, this tool will automatically support both D2F and D2M, which is an extra bonus, but it will depend on how compatible the rigging will be between the figures. As the rigging work progresses, I will know for sure.

 

[unreal]

I assumed "quickly" from when you started this thread about your code. Of course, if you've been working on this since not long after Da2 was a spark in someone's eye, this could be a decade or more. I further assume features in Da2 have been someone's dream since Poser was in lower single digit versions, and the only way to have V4 was to put two V2's in a scene, stacked on top of each other. Being ignorant as all heck, assumptions are all I have. Aggressively ignorant and proud!

I'm going to switch the scale for "quickly" to geological scale, just to be safe. And now I stand by "quickly" with confidence.

Your description of priorities makes perfect sense. I always thought that was the SM+Poser problem. Did they make any money from content? I suppose the deeper question is did they make any money, period. Probably not. But thankfully moot

 

[Ken1171]

That's a good reference. I started this thread about a couple of weeks after I started the script, which is now over a month old. By the time I needed the thumbs, I posted the request to the devs, so it only took a couple of weeks, which in script development time feels like an eternity to me. LOL But I can't complain - they have been helping patching the holes on the Poser Python API so far, and now I only have 1 single show-stopping missing feature (for another script, not this one), so it's a solid record.

 

[DanaTA]

Congratulations, Ken! You are in the newsletter, as Exclusive Vendor Spotlight!

Dana

 

[Ken1171]

Congratulations, Ken! You are in the newsletter, as Exclusive Vendor Spotlight!

Dana

Really cool! Can you send me a copy? I didn't get it.

 

[DanaTA]

Really cool! Can you send me a copy? I didn't get it.

Send how?

Dana

 

[Ken1171]

Send how?

Dana

Sent you a note.

 

[Ken1171]

Now that I've got Light Dots, SkinEdit and the Scripts Manager delivered to the store, D2F got more updates this week, and I have progressed this "visual posing script" for her a bit more. I have plans to add individual posing control dots to each finger, and toe, and also dots to control the entire hand, like spreading the fingers, opening and closing fists - whatever will be available when the ERC controls are in place. I have added a control dot to each of the fingers, so we can bend, side-side and twist each individually, as well as the 2 metacarpals on each hand.

A new feature added today is dragging control dots with the MIDDLE mouse button. With LEFT button dragging, we can bend and side-side, and with middle button, we can now also twist, so all 3 axes are accessible now - with or without symmetry. When the posing ERCs become available, I will also add full hand posing dots for the entire hand.

This will represent a whole new way of "visual posing" as never seen before in Poser. One of the issues I have with dragging chips is the lack of control. For this script I have added 4 levels of mouse dragging precision with the SHIFT, CTRL and ALT keys, plus the option to disable specific bending directions.

On the face tab, I have added control points to the inner mouth, like the lower jaw and tongue segments. When the D2F posing ERCs are in place, I plan on adding control points to control tongue posing as a whole as well. How it will work will depend on how Paul sets up the ERC controls. I believe this will be the most FUN part when posing with control points. Paul is already adding the ERC controls, so we are almost there.

 

[Ken1171]

Completed the initial version of the tool. It now contains all the controls for bones (red dots) and ERC dials (blue dots). I have added a new feature where hovering the mouse over the dots shows the control name on the black bar below the tabs. The interface is shown below, very minimalistic and mostly visual - but no less powerful.

The bone controls (red dots) are placed right where the actual joint is on the figure. so you always know exactly what they do. The ERC posing controls (blue dots) were placed close to the areas they affect, like arms controls are on the shoulders, the torso controls are on the torso, and the leg controls are at the bottom. The hand controls are on the hands, and the finger controls are by the fingers. The toes controls are by the toes. The thumbs controls are close to the thumbs, and so on. I guess it makes sense, visually and intuitively. Hovering the mouse over the dots shows their name, so we always know what is what.

We can filter mouse control by axis, so if we want to only use up-down, we can disable side-side. We drag with the middle mouse button to only affect twisting, which is rather rare, but it's available. The upper/lower teeth and tongue are shown separately from the head, so we can control their parts more clearly. I have included all fingers and toes bones in there. Posing those tend to be difficult in Poser because selecting a specific part of a finger when the figure is posed can be challenging. That's where I think this will come handy (no pun intended).

One new feature is that we can control either bone rotations (red dots) or ERC posing controls (blue dots) seamlessly by dragging the dots with the mouse, where the dragging direction matters for bone rotations (control 2 axes at the same time, and the 3rd with middle mouse), but not for ERC controls (only horizontal dragging, like we do with dials). So we don't have to worry that bone rotations and ERC controls work in very different ways - we just click and drag them to get things done, as if they were both the same.

This script inherits some features from the "Dial Master P12", where we have 4 levels of dialing precision using the SHIFT. CTRL and ALT keys. This overrides individual dial's sensitivity in real-time to give more control - something we cannot have with regular Poser dials.

There is also the custom Memory Dots tab with 25 slots to save your face presets, where it shows a thumbnail of the preset when we hover the mouse over it, as well as an optional text caption you can set to remind you of what it does. The thumbnails are auto-framed by default, but you can disable the feature if you like.

I have played with the script to control all the specific parts of Dawn 2 without touching any dials, and it was fun. We have the option of enabling symmetry, and also to invert it, which can save time when posing or making expressions. Some interesting things can happen, like inverted symmetry on "fist hand" ERC makes one hand open and the other close when we drag the control.

Being a visual person, I find this way easier and quicker to manipulate the figure parts using this panel, comparing to hunting for dials, or trying to select a body part from the viewport when the figure is clothed. I am happy that I was able to put all this functionality into such a small panel with tabs. People should be able to use this right away with a shallow learning curve, since it's 95% all visual. If you can drag a dot, you know how to use it. ^_____^

 

[Ken1171]

Today I have adjusted the tool to allow more control when posing bones with the mouse, when we are controlling 2 axes at the same time. I have introduced a "predominant axis" to the mouse movement that filters out other motions. For example, if we drag the mouse up-down to pose a bone (red dots), it filters out all minor side-side motion. This way, up-down (for example) becomes predominantly up-down, so dragging with the mouse becomes more forgiving when it's not precise. It makes the tool smarter when understanding the mouse motions when more than 1 axis is involved. If the mouse motion doesn't have a predominant direction, like moving in a circle, all axes are affected, which is likely what we wanted.

This also allows setting ERC dials (blue dots) using either up-down or side-side mouse dragging, which previously only worked with side-side motion. It now picks the predominant motion direction and makes that the one controlling the dial, which I find more intuitive to work with.