Mon, Dec 23, 4:46 AM CST

Renderosity Forums / Poser - OFFICIAL



Welcome to the Poser - OFFICIAL Forum

Forum Coordinators: RedPhantom

Poser - OFFICIAL F.A.Q (Last Updated: 2024 Dec 22 10:18 pm)



Subject: Rotations and translations on hip vs.body


ElZagna ( ) posted Thu, 01 September 2011 at 7:30 PM · edited Mon, 23 December 2024 at 4:46 AM

Any rules on when you should use hip vs. body for rotations and translations?



OS: Windows 10 64-bit, Poser: 10


Acadia ( ) posted Thu, 01 September 2011 at 8:17 PM

I'm not sure what you mean?

Are you referring to making figure poses?  If you are the rule of thumb is to take your clothing off and look in a full length mirror as to how the body moves.  The hip is the centre of gravity and tends to remain stationary. The turning movement actually comes from above the hip area at the waist, or the joints of the hip. But the pelvis itself remains in place.

"It is good to see ourselves as others see us. Try as we may, we are never
able to know ourselves fully as we are, especially the evil side of us.
This we can do only if we are not angry with our critics but will take in good
heart whatever they might have to say." - Ghandi



wdupre ( ) posted Thu, 01 September 2011 at 8:41 PM

the original reason that this was important was for creating poses, hip translations will be saved in poses so unless you want the pose to move from the center position you dont want to use the hip, but if say you are creating poses where two figures need to be interacting or the figure needs to be interacting with an object such as sitting in a chair, using the hip to rotate and translate would be a good thing because it will provide absolute positioning for the figure. if you are just moving the figure around for a scene you are setting up for your own scene and not for a pose to save than its not an issue really.



ElZagna ( ) posted Thu, 01 September 2011 at 9:37 PM

Thanks, wdupre, that's what I was getting at. Sorry I wasn't clear, Acadia.

Looking through a sampling of my pose files I see that about a third have a Body actor, but all the Body parameters are set to zero. It seems to me that a "proper" pose file should never have a Body actor at all. Furthermore, except for the cases mentioned above, the x and z translations for the hip should always be zero. In other words, the creator of a pose file needs a darn good reason to move the character off center.

Also , as a matter of pratice, users should move their characters with the Body actor and not the hip unless they are planning to save the pose. Does that seem right?



OS: Windows 10 64-bit, Poser: 10


markschum ( ) posted Thu, 01 September 2011 at 10:26 PM

The hip works differently with IK on than with IK off.

Moving the BODY actor moves the entire figure regardless of IK setting.

With Poser 5 the Body actor did not get saved, with Poser 7 its an option when saving poses.  If the body actor has translations the figure will move to those coordinates when th pose is applied.

General pose rules is leave the body actor alone , zeroed, and use Hip instead.


Miss Nancy ( ) posted Thu, 01 September 2011 at 11:09 PM

bvh files don't save body params. hence, importing bvh to poser, hip is often zeroed into foor plane.



lesbentley ( ) posted Fri, 02 September 2011 at 3:40 PM · edited Fri, 02 September 2011 at 3:50 PM

Quote - Any rules on when you should use hip vs. body for rotations and translations?

Yes there are some principles. Below is the system I use when saving poses. If you don't intend to save a pose you can be a little more relaxed about translating the hip.

Before starting to create a pose, make sure that yTran for the Body and x and z rotate are zero, you may want to set all the translations and rotations of the Body to zero before you start, but that is not essential.

Don't use xTran or zTran for the hip, keep them at zero (at least in the final pose). If you use IK, you may need to translate the hip to get the desired effect, but afterwards, tun IK off and zero x and z tran. On the other hand do use yTran in the hip to set the feet - or the lowest part of the figure in the particular pose - on the ground.

The advantage of this system is that you can apply saved poses without the figure jumping to a different location in the scene, as would happen if you used x or z translations of the hip.

Pose data for the Body is by default not saved in a pose file, Poser 7 introduced an option to save Body data, but you should not use that unless you have a specific reason for doing so, and understand the effect it will have.

Use the Body translations to position your figure as appropriate in a particular  scene. Use the BODY yRotate to set the orientation of the figure for the particular scene. You should not normally use any of the other rotations in the Body.

To sum up.
 
In the hip:
Use only yTran and any rotation.

In the Body:
Use only yRotate and any translation.
(and remember, Body data does not get saved to a pose)
These are rules of thumb, not laws of nature. They make saving any applying poses easier, and prevent some potential problems. There can be exceptions. For example, if you make poses for two figures, where the figures need to stand in a particular relationship to each other, use hip translations in one of the the figures to achieve the desired separation.

Always turn IK off before saving a pose. Check the pose after turning IK off, because things can change. In particular check the altitude of the feet in relation to the ground.


ElZagna ( ) posted Fri, 02 September 2011 at 6:32 PM

Wow, we seem to have a rare consensus here. In essence:

If you are creating a pose to save and use later, position with the Hip and only rarely use the X and Z translations.

If you are using a saved pose in a scene, position with the Body and only rarely use the X and Z rotations.



OS: Windows 10 64-bit, Poser: 10


lesbentley ( ) posted Sat, 03 September 2011 at 3:29 PM · edited Sat, 03 September 2011 at 3:36 PM

file_472529.png

 

Yes. With the hip, you want to be able to apply the pose no matter where the figure is located in the scene, without the pose changing the location of the figure (x or z translation), but you do want the pose to put the figure's feet (or bum, or what ever) on the floor, so you set the yTran of the hip in the pose.

With the Body, either before or after you apply the pose, you use the translations to set the figure's location in the scene (only using yTran if the figure is to stand on a chair). You use the Body's yRotate if it is not facing in the direction you want. Normally the other rotations of the Body should not be used, except in special circumstances, eg to orientate the figure to the rolling deck of a ship at sea, or a weightless scene in outer space.

Body data does not (or should not) get saved to a pose, so when saving a pose you want the Body's yTran and x and z rotations to be zero, so that you can see that the figures feet are in the correct relationship to the ground. Normally you would construct a pose with the Body completely zeroed, but it does not matter if it is x or z translated, so long as you realize that the translation will not be saved in the pose.

Some other considerations with poses. I think that putting the figures feet (or what ever) accuracy on the ground is important to the quality of a pose.  You can use "Drop to Floor" to, well ... drop the figure to the floor obviously. However the "Floor" is not quite the same thing as the Poser "GROUND" plain. The Floor (as in "Drop to Floor") is at zero altitude, the GROUND on the other hand is slightly below the Floor at -0.001 PNU (Poser Native Units), or -2.62128 mm, or -0.1032 inch. Also, real flesh tends to compress when it is baring weight. For these reasons, you may want to set the altitude of the figure slightly lower than Drop to Floor would put it. There is a bit of a dilemma here, because some people may be using the GROUND plain, and others may be using a room prop where the floor is at zero altitude. So which do you set the figure altitude for? I've had different theories on this at different times, at the moment I tend to compromise, and set feet so they are just touching the GROUND, this gives the illusion of a some compression of the feet when used with a floor at zero altitude, but the feet look a little too high when used with the GROUND plain. That's just me, you have to decide for yourself. There is also an argument for using the DTF altitude as a standard, and letting the user set their own correction by using a slight yTran for the Body.

To set the figure's altitude, I view through one of the orthogonal cameras (eg Front or Right Camera), with the Element Display Style of the GROUND set to Outline Display Style (see image above). The image is of V4's left foot with the toes bent 5°. The altitude of the figure has been dropped by -0.001 PNU over and beyond the altitude set by Drop to Floor. The top line is at zero altitude, the bottom line is the GROUND plain.

If you want a guide to zero altitude that will not render, load a Walk Path, set its ScaleX to zero, and its yRotate as appropriate to your view.

Morphs are another thing to be considered when saving a pose. If someone has spent the last hour setting up character morphs, they don't want to discover that applying your pose has reset all those morphs. Normally poses for distribution should be saved without morphs, unless they are specifically meant as morph setting (MOR) poses.

I recently made some lollipop sucking poses for the Antonia Polygon figure. The figure uses morph (targetGeom) channels for ERC based fake hand grasp. I wanted to include the Hand_Grasp and Mouth_Open channels, but not any of the PBMCC, or other morph channels. So I saved two versions of each pose, one with and one without morphs. I then used a cr2 editor to copy only the morph channels I wanted to the pose that had been saved without morphs (a somewhat tedious process).


ElZagna ( ) posted Sun, 04 September 2011 at 2:47 PM

lesbentley - Thank you for your very thorough and very clear comments. It really helps me a lot - as well as many other users - to have something so lucidly laid out.

RE: Floor vs. Ground –Experimenting with 4 floors in my collection I wasn’t able to find much consistency in them.  Most were loaded with yTran = 0 but one was set to -0.042; some had a yScale that you could adjust, others had zero depth; “Drop to Ground” set the yTrans to 0 for some and to some negative setting for others. Props were also all over the place when it came to the yTrans setting after “Drop to Ground”. This inconsistency is a very… well… consistent feature of Poser files.

Quote - There is also an argument for using the DTF altitude as a standard, and letting the user set their own correction by using a slight yTran for the Body.

That would be my preference. Most of us Poser users are hobbyists, and the subtle differences you mention will be lost on the great unwashed masses of us who are just trying to keep Vicky from looking like she was hit by a truck.

Quote - Morphs are another thing to be considered when saving a pose. If someone has spent the last hour setting up character morphs, they don't want to discover that applying your pose has reset all those morphs. Normally poses for distribution should be saved without morphs, unless they are specifically meant as morph setting (MOR) poses.

Absolutely. One of the things I’m trying to accomplish with this is to come up with a set of guidelines for what should and shouldn’t be included in pose files.



OS: Windows 10 64-bit, Poser: 10


Privacy Notice

This site uses cookies to deliver the best experience. Our own cookies make user accounts and other features possible. Third-party cookies are used to display relevant ads and to analyze how Renderosity is used. By using our site, you acknowledge that you have read and understood our Terms of Service, including our Cookie Policy and our Privacy Policy.