Loading...
> >

Please Help - "It's (not so) fun to play with the M-J-M-C!". (Multiple Joint Controlled Morph)

DaveEdDaveEd Posts: 2
edited December 1969 in Technical Help (nuts n bolts)

Hi guys! I've been using DS for years but this is the first time I'm posting on these forums so please be nice. ;-P

I'm having trouble trying to link corrective morphs for two joint movements (eg. Shoulder Bend Down + Shoulder Front = Corrected Morph).
Can anyone explain to me how I can achieve this?

I'm more than able to make and link single JCMs in the Property Hierarchy but when it comes multiple movements of a joint I'm completely lost. I did do a search through the forums but I only found a few that I couldn't understand and some with dead links.

Comments

  • Richard HaseltineRichard Haseltine Posts: 100,833
    edited December 1969

    What are you wanting to do? If you want the morph to kick in only as both are set you can add one of them as a Multiply controller, under Second stage, with the Scalar on the first stage adjusted to balance (or you could create a new property that goes from 0 to 1 as the rotation goes from 0 to its maximum, then add that as a multiplier for the morph).

  • Cayman StudiosCayman Studios Posts: 1,135
    edited December 1969

    I've been creating quite a lot of these two-joint MCMs recently (and a few three-joint ones!) and it took me some time to work out what to do. The main difficulty is calculating the respective scalars attributed to each of the joints. They have to be set so that the value of the MCM equals 1.

    The formula for a two-joint MCM Scalar is as follows:

    Where joint A is at an angle of x, and Joint B is at an angle of y,
    the MCM scalar for A = 1/x * 1/2 * (JCM scalar for A * JCM scalar for B)/(x * y)
    the MCM scalar for B = 1/y * 1/2 * (JCM scalar for A * JCM scalar for B)/(x * y)

    The three-joint MCM scalars would therefore be:

    For joint A (angle x), Joint B (angle y), Joint C (angle z)
    MCM scalar for A = 1/x * 1/3 * (JCM scalar for A * JCM scalar for B * JCM scalar for C)/(x * y * z)
    MCM scalar for B = 1/y * 1/3 * ditto
    MCM scalar for C = 1/z * 1/3 * ditto.

    And so on. I haven't tried any four, five or six-joint MCMs yet, but the principle would be the same.

  • RAMWolffRAMWolff Posts: 10,212
    edited December 1969

    Thanks I'll see if I can work this out... probably not but I have hope.... sort of ! lol

  • RAMWolffRAMWolff Posts: 10,212
    edited December 1969

    As stated in my post in the Commons... the Thigh bend is -105% and the shin bend is 140%.. So not sure how that translates.

  • RAMWolffRAMWolff Posts: 10,212
    edited December 1969

    I wonder if DAZ has any info on this in their data base??

  • RAMWolffRAMWolff Posts: 10,212
    edited December 1969

    AHA... I just did a Google search and found some documentation

    http://docs.daz3d.com/doku.php/public/software/dazstudio/4/userguide/creating_content/assembling/tutorials/creating_morph_controlled_morphs/start

  • RAMWolffRAMWolff Posts: 10,212
    edited December 1969

    Well, I filed a report. The information on the page above was written before DS 4.5 was released so it's in need, badly, of some updating. I went ahead and made the morph and applied it via ERC to the thigh and then attempted to link in the shin but had no luck. I'm just not technically smart enough to know how to do that. I also suggested they grab Josh Darling to do a YouTube vid for us for the channel to make it even nicer. Fingers crossed.

  • SickleYieldSickleYield Posts: 7,634
    edited December 1969

    When I've done double bend correctors, I would set up two single bend correctors first usually (e.g. right and left thigh) via the usual method of loading them with reverse deformation, posing at their max value, and doing an ERC freeze from parameters on the joint bend (e.g. on the left thigh bend dial itself). This might seem counterintuitive but it is the fastest way to set up simple single bend corrections.

    Then I would set those as the controllers of the double bend and use DeltaAdd (rather than multiply) and scalars that add up to 1. Sometimes this is sufficient. Sometimes I would change the ERC to keyed instead of DeltaAdd (that's not the full correct name, but it's part of it) and manually set up individual keys (bearing in mind that the last key number needs to be 1 if it's a 100% bend corrector and 1.10 if it's 110%, etc.).

    In the second part, setting the double corrector up with the two other JCM/MCMs as controllers is done by dragging and dropping - select the clothing item, go to Property Hierarchy, and search by the name of the JCM until you find it, then expand it, then find the Controllers section. Then go to Parameters, find those two JCM, and drag them over to the Property Hierarchy tab under the Controllers heading. Then their settings will appear in the hierarchy and their scalars can be set.

  • RAMWolffRAMWolff Posts: 10,212
    edited December 1969

    HAH... that went totally over my head. Perhaps tomorrow when I'm rested. LOL Thanks hon!

  • Cayman StudiosCayman Studios Posts: 1,135
    edited December 1969

    RAMWolff, looking at your picture, you may be able to solve the poke-through by just applying the Smoothing Modifier (Edit > Figure > Geometry > Apply Smoothing Modifier), and not bother with JCMs or MCMs at all.

    But if you want to use JCMs, as SickleYield says, you should create two separate JCMs first - one for the Shin and one for the Thigh. Even if there is no obvious poke-through you can still morph the garment so that it is slightly further from the skin surface, and so will be less likely to poke through when further joint movements are brought into play. You may then not even need the two-joint MCM.

  • RAMWolffRAMWolff Posts: 10,212
    edited December 1969

    I really don't want to rely on smoothing and I've already created corrective JCM's that work as expected. This "Knee Up" movement is causing a new set of poke throughs though that an MCM is called for.

Sign In or Register to comment.