P.S.
Having tremendous problems with my connection tonight! :cursing:

Okay, there's not only the problem with the missing y-translations in the one on the right, but also the y-rotation doesn't seem to be fully applied to the hip. Except for the jumping knee, I'd say the one on the left looks just fine.

So I guess what Cage suggested could be a workable solution: keep Antonia 1.0.0 (or a cut-down version of it) around for use with Walk Designer and post-process the animations it produces.

Quote - Except for the jumping knee, I'd say the one on the left looks just fine.

We really could use a special walk animation set for Antonia.  If we had the right walk pose files, presumably we'd get better results.

Maybe one of Poser's notorious Twelve Animators will become interested in Antonia.  :laugh:

Quote - > Quote - Except for the jumping knee, I'd say the one on the left looks just fine.

We really could use a special walk animation set for Antonia.  If we had the right walk pose files, presumably we'd get better results.

Maybe one of Poser's notorious Twelve Animators will become interested in Antonia.  :laugh:

I was wondering if that might be an IK related glitch in the animation(s) Les showed.

@Les:

How did your efforts with that pose mixing script we wrote work out?  Would it be worth trying to correct some existing walk cycles for Antonia using that apporach, or did it turn out to be flawed?    I have a version which can derive the bias settings from a pose, if the process is any good.  That could possibly help correct files for any of the existing walk poses.

@Cage,

Quote - If we use Antonia 1.0 as a proxy figure for the Walk Designer, everything we need is already in place.  The walk cycle would need to be saved as an animated pose, then converted like other poses, using the existing script.

Perhaps, but it may not be that simple. As I have noted previously, what you put into the WD as a walk blend (animated pose applied via the WD), is usually not exactly the same as what you get out as a walk cycle (actual animation). It may be close enough, it may not.

@odf,

Quote - Except for the jumping knee, I'd say the one on the left looks just fine.

The jerking knee could probably be fixed by a custom walk blend.

@Cage,

Quote - We really could use a special walk animation set for Antonia.  If we had the right walk pose files, presumably we'd get better results.

I had a go at making a custom walk cycle for Antonia, but was not very successful. It's harder than it sounds, at least I found it so. I have a short attention span, so as other issues came up I moved on the other things, and forgot about the walk cycle. Perhaps I will have another go, now that it is the hot topic again. That's only a "perhaps" so don't let that stop anyone who wants to have a go. Besides two heads are better than one, and out of two partially functioning poses, we may be able to make one good one.

@odf,

Quote - I was wondering if that [jumping knee] might be an IK related glitch in the animation(s) Les showed.

I had IK turned off when I made the animation, but I'm fairly certain that the WD uses some sort of IK, irrespective of the state of the figure's IK switch. So yes, it probably is IK related, I think it is also related to the default walk that the WD uses(which I assume is "walk.pz2").  The jerk only happens in one knee, and the Walk Designer poses seem to be asymmetrical. My advice is to start with a symmetrical pose, by which I mean that the left side should be doing exactly the same as the right, but offset by the required number of frames. I think it's 15 frames. I find that trying to work with an asymmetrical pose is a good way to bring on a nervous breakdown. But perhaps others could have success with a freehand approach.

@Cage,

Quote - How did your efforts with that pose mixing script we wrote work out?  Would it be worth trying to correct some existing walk cycles for Antonia using that apporach, or did it turn out to be flawed?

It's so long ago now that I forget the details. Your script did help a lot, but I think I asked for the wrong script, and made the work flow much harder as a result. It was my innitial conception of the work flow [edit] and how the script should fit into it, that was the problem, not your script itself, which did exactly what I had asked.

It's interesting to note the asymmetry in the way the feet move in my animations. The left foot seems to go a long way below the ground, and also flaps a lot. Again I suspect this may be due to asymmetries in the default walk blend.

Antonia - V4

I'm testing the pose mixer script.  It does do a very good job of correcting the positioning of the limbs, but there's trouble with the feet sliding.  I think Les noted that before, when working on it.  I think that might have something to do with the differing twist angles between the P4 source figure and Antonia, but I'm not sure.    Of course, there seems to be some foot sliding in Posette, too....

My .gif maker won't work on Vista, so I uploaded a test animation to youtube, but it's hard to get a sense of the whole cycle because it won't loop.  :unsure:  Anyway, fixed Antonia is on the left, source Posette in the middle, and non-fixed Antonia on the right.

http://www.youtube.com/watch?v=eh-sLwLuPIQ

The script might help create some basic Antonia walk cycles from which to work.  I'll tinker with it.

Cage,

You were asking about the bias files we were playing with. Here is an animation of two standard Antonia figures. Both have 50% Sexy Walk applied. The one on your left is without the bias, the one on the right is with bias applied. As you can see there is still a jerk in the left knee, but with 50% Sexy Walk there is an even worse jerk in the right knee for both figures (Les puts head in hands and sobs).

Like I said, I'm struggling to remember details. I will give you more info, when and if my memory kicks into gear. The basic process of producing the above animation was to inject some slaving code, then inject bias settings. Then run the WD, then turn the ERC on.

The ERC version would be useless as input to the WD, because the WD does not understand ERC. I think the idea was, that once I had the bias settings figured out, your script could then apply the settings to a pose file, and that pose could be used as a new walk blend. There again, I may have got that bit completely wrong. I'm still struggling to remember.

The idea shows some promise, the biased walk does look a bit better IMO, but is still a long way off being good. The real holy grail is to produce a walk blend that outputs a good walk, without the need for further tweaking.

P.S. I suspect the knee twitch may be due the the hip being too low in that frame, and IK is throwing the knee to the side because it does not understand which way to bend it. That's one of the big problems. We can't just come up with walk poses that look good when applied as animated poses, we also have to take account of how IK will handle the pose when it is used as a walk blend in the WD. I think that hip altitude, and possibly residual thigh and shin rotations may be the key factors here.

@flaviok,

Thanks for the input. I looks like you are getting better results than me, no knee twitch. I wonder if that is a version thing? I'm still on P6, they may have made improvements to the Walk Designer since then.

Above the thigh, the one the right looks pretty good.

I think I see a twitch still in flaviok's version, hard to tell with the video jump.

Quote - Above the thigh, the one the right looks pretty good.

Not just pretty good. I think it looks amazing.

I converted all of the P4 walk cycles for Antonia, and the results look pretty good. I don't see any real ugliness in it, but I'm hardly expert in this....  :unsure:

The link is a Flash .swf, exported from Poser.  Sorry.    I really need to get a new GIF maker.

http://www.the.cage.page.phantom3d.net/Antonia/test/antonia_walk.htm

Cage,

Quote - I'm testing the pose mixer script.  It does do a very good job of correcting the positioning of the limbs, but there's trouble with the feet sliding.  I think Les noted that before, when working on it.  I think that might have something to do with the differing twist angles between the P4 source figure and Antonia, but I'm not sure.    Of course, there seems to be some foot sliding in Posette, too....

Yes I think there was a problem with the feet sliding. There are two kinds of sliding, side-side is one, or the BODY moving forwards at a different rate than the feet (which I think is called "skating"), so that the foot that should be planted, is sliding forwards or backwards over the ground. I think it was this skating that was the big problem, but there may also have been a bit of side-side sliding. It may be due to twist angles, there again it may be related to differences in leg length between Antonia and P4 figures, or the proportionality between thigh length and shin length. Have you noticed these parameters in the Poser walk files?

thighLength 0.213434
feetDistance 0.045139
hipHeight     0.405027

I think the last two were added in P6, I certainly don't remember them in P4, though thighLength was there.

The Flash looks good Cage, no thigh jump there. Perhaps the arms are swinging too far out to the side though.

Quote - Have you noticed these parameters in the Poser walk files?

Yes, but I'd always just deleted them.  :lol:  Are they standard values, the same in all cases, or something we can re-set for Antonia's characteristics?

I'm not seeing any of the leg-twitching or other oddities with these converted walk poses I'm testing.

Quote - The Flash looks good Cage, no thigh jump there. Perhaps the arms are swinging too far out though.

I've altered the script so the pose bias settings can be derived from an Antonia in the scene.  Antonia is posed to align her limbs as well as possible with Posette's.  Then the script is run.  I think the alignment could be improved.  The fingers could be aligned, to correct the hand poses.

Part of the arm swing may be coming from the pose mixture.  I added in some of the power walk.  These results are coming from the Walk Designer, not straight from a converted pose.

I'm going to tinker with this a bit more tomorrow.  It's looking fairly promising.

Quote - Yes, but I'd always just deleted them.    Are they standard values, the same in all cases, or something we can re-set for Antonia's characteristics?

I seem to remember that they are figure specific.

You guys are on fire. Very happy to see it.

In that flash, the side of her head looks like it's getting a double chin, is that just the lighting?

I just checked. Antonia always gets:

thighLength 0.173937

So yes it definately is figure specific. I think the length of antonias thigh, origin to end point is 0.173 PNU.

P6 does not write out the "feetDistance" or "hipHeight", even though it is there in the WalkDesigner poses. It's probably something they were working on at the time, and did not implement the routine to write it to pz2 in P6. This may be why some of you are getting better results regarding the knee jump. If one of you gets a chanse, could you pass on the values Poser writs out for feetDistance and hipHeight. Even though P6 does not write those, it may be able to read them.

As fare as I am concerned at the moment, she can have seven chins, so long as the knees stay where they are supposed to be! :tongue2:

Well I seem to have found a way to avoid the knee jump I was getting. I set the limits in the shins to to stop them twisting or swinging, also set limits in the feet to prevent swing, then turned limits on for the figure.

This set up gives me a few glitches in the feet, but these should be easier to correct than the knee jump. It was Cage mentioning joint twisting as a possible cause that put me onthe right track. Thanks Cage.

lesbentley

Poserpro2010, carregar cr2 Antonia 1.0 - sexy 20

NOVO VIDEO YOUTUBE

http://www.youtube.com/watch?v=bjCqP2yMtUA

Wow, you came a long way in such a sort time. From what I see the arms are now more unnatural in their movement than the legs and the hips are!

And I do not mind to convert the poses again. As said I keep all versions of what I create (and buy a lot of harddrives...) to be able to go back to the original and try another version. As long as I get such sweet things like the little renamer script I will leave my fork in the locker :laugh:

About:

Oh. PS - Puppet Master Bi-Ped and Puppet Master Quadruped weren't written by PhilC. They were hosted by him but written by Kamiliche who then sold them to eFrontier who then killed them both after stealing what they could from them to create Universal Poses....

Of course you are right, it was Kamiliche's work! And about killing it after making those half-baked Universal Poses: Tue true, unfortunatly!*

@ Les:

Any idea about the feetDistance and hipHeight values for the poses?  A fully-compatible Antonia set should probably have those in place.

I guess I'll use Python to compare the values in some existing poses with the geometries to which they apply so I can try to derive the correct Antonia values, if the Antonia values aren't known.

Are you using a full set of converted walk cycle poses for Antonia?  I'm using a full set and I'm not encountering anything like the knee jump.  I think that may be present if the default walk has much influence.  If you have an Antonia pose set, presumably the Antonia Walk pose should be used instead of allowing any influence by the Poser default.

Interesting.  The feetDistance value seems to help the Walk Designer determine... how far apart to place the walking feet.  It delivers what it promises.  :lol:

I'm working with poses converted from Posette to Antonia.  For these, Antonia basically is posed to match Posette's default pose and this is mixed in with the Posette walk cycle.  So we're using the distance between Posette's feet for the purpose of this cycle, not Antonia's.  If we change the feetDistance value to an Antonia estimate (0.1332 PNU), the Walk Designer actually undoes the pose corrections for the legs, giving her that wide stance of which I've read in the news.  Erm.  :unsure:

The hipHeight is hard to understand as a precise value.  It doesn't match the hip joint center for Posette or P4 Dork.  It doesn't correspond with the top or bottom of the hip geometry for either.  It isn't an average of their two centers.  It doesn't look like the center of the hip geometry for either figure.  It's pretty close to being Antonia's hip center, though, so I assumed that was the intent and plugged Antonia's hip center into the hipHeight line.  I'm not sure whether it helps or not, but it doesn't hurt anything.

Here are the values I tested:

feetDistance 0.1332
hipHeight  0.381689

The values don't match those in the old P4 walk cycle poses, however, which is frustrating.  I think I recall a lot of fuss at one point about the length of the PNU having changed.  Possibly that's what's reflected here.

These are the values I get for Antonia and Posette, using the script:

AntoniaA
hipHeight 0.381689
feetDistance 0.079002
thighLength 0.173937

Posette
hipHeight 0.396901
feetDistance 0.043000
thighLength 0.203177

The thighLength value is the distance between the Origin and EndPoint of the Thigh actor, which makes sense.

Note, however, that if we're converting the pose using the pose mixing script, the feetDistance value for the original figure from which the pose was saved should be used.  A more accurate value can be derived with Python by getting the bounding boxes for the worldspace geometry of one of the feet and doubling the distance of the inner box boundary from the world origin.

Ah.  No, that's wrong, about the feetDistance.  It has nothing to do with actor geometries, which again makes sense.  It's the X offset of the Origin of the left foot.  Getting posed Origin positions is not as simple a coding task.  :crying:  But the approximation through use of the source figure's feetDistance seems to work decently.

It looks like some problems in an animation crop up when the special settings of the Walk Designer are used.  The "Align head to..." options and the settings for stride length, arm swing, head bob, etc.  The "Align to" settings caused the neck geometry to explode in animation, for a few frames.  The other settings worsen any animation I've tested so far, but with less disastrous results.

Leg-popping: I find that the knees will buckle in an animation which follows a walk path, but not in the same animation if the figure walks in place.  This suggests that the buckling may have something to do with how the Walk Designer is handling the legs when rotating the walking figure's body, rather than necessarily being flaws in the animated poses themselves.

On the other hand, even these converted walk poses are not specifically designed for Antonia.  Differences in default orientation of limbs may be having more of an effect than I, at least, would like to hope.  :unsure:

There's also the possibility that the Walk Designer may be relying on joint rotation limits being set, to prevent unrealistic effects. 

Hoping I haven't chased all of you away....  :scared:

Cage,

Quote - Any idea about the feetDistance and hipHeight values for the poses?  A fully-compatible Antonia set should probably have those in place.

I would guess that the vaslue "hipHeight" is equal to the origin of the hip, and that feetDistance is equal disisance between the origins of the feet. But that's just guess, it needs confirming. I'll check if it if I find time, but a lot is happening, and I'm being pulled in several diffrent directions at once.

Oh! I just read one of your later posts, so "hipHeight" is not the hip origin! Quote: "It's pretty close to being Antonia's hip center, though". Perhaps it is the center of the bounding box?

Quote - I think I recall a lot of fuss at one point about the length of the PNU having changed.  Possibly that's what's reflected here.

No, PNU never changed, I think that would be theoretically impossible. What changerd was the conversion factor used to convert PNU units into real world units, and vice versa. The conversion factor is just a convention, it has no inpact on the internal workings of Poser. In other words, if your Poser interface uses inches, a Poser box scaled at 100% might be reported as a diffrent number of inches high in diffrent versions, but it would always be the same number of PNU high.

Quote - Oh! I just read one of your later posts, so "hipHeight" is not the hip origin! Quote: "It's pretty close to being Antonia's hip center, though". Perhaps it is the center of the bounding box?

The hipHeight value is the origin of the hip.  In that earlier post, I was checking Posette's hip origin against the hipHeight listing in the P4 walk cycles.  The values in those walk cycles are strange, not matching anything. 

But Poser Python can return a figure's hipHeight and feetDistance using one of the built-in methods.  I used this command to get the correct data for both Antonia and Posette, as listed above.  This verified hipHeight as the Y component of the hip origin.

Quote - No, PNU never changed, I think that would be theoretically impossible. What changerd was the conversion factor used to convert PNU units into real world units, and vice versa.

Ah.  I understand.  I still don't see where they got the values for hipHeight and feetDistance in the P4 walk poses, but oh well.  :shruggy guy:  :lol:

More on the "popping knees" problem.  I'm wondering if this could relate to the default handling of the leg IK chains.  There's a trick which allows you to define which way the joint will bend with IK, but I don't remember how it works.    I think it has something to do with memorizing the figure state in the correct pose, rather than setting joint bending limits.

Does anyone recall how to do that?

Quote - Leg-popping: I find that the knees will buckle in an animation which follows a walk path, but not in the same animation if the figure walks in place.  This suggests that the buckling may have something to do with how the Walk Designer is handling the legs when rotating the walking figure's body, rather than necessarily being flaws in the animated poses themselves.

Yes, the animated pose can be OK, as a pose, but not OK when used as a walk blend. I think that frame 15 (k 14 in the pose) usually seems to be a bad frame for knee bukle. Is it just a coincidence that this is exactly half way through the walk cycle?

The knee bukle seems to me to be one of the bigest problems, because we don't understand it properly. If we understood how and why it is happening, it might make a solution easier.

I think it is the IK solver trying to reconsile the distance between the foot and the hip by rotations of the shin and/or thigh, that is at the root of the problen. Setting limits on the twist and swing of the shin seems to go a long way towarsd resolving the problem, but is not without concequences for the feet, and perhaps other actors. A more elegant solution may be possible by adjusting the hip ytran in the pose, for problem fremes, but that is just a theory, I have not tested it. The 'hipHight' and 'thighLength' may also have a bearing on this. If the WD thinks the hip is higher or lower than it actually is, or that the thighs are longer or shorter that they actually are, that could contribute to the problem.

But all the above is just the idle rambelings of my mind, I may be completely on the wrong track.

I Kind of cross-posted there.

Quote - There's a trick which allows you to define which way the joint will bend with IK, but I don't remember how it works.

That would be a handy trick to know!

Quote - Yes, the animated pose can be OK, as a pose, but not OK when used as a walk blend. I think that frame 15 (k 14 in the pose) usually seems to be a bad frame for knee bukle. Is it just a coincidence that this is exactly half way through the walk cycle?

I'm only noticing this problem when a walk path is involved.  When walking in place, I don't see a problem with any blend of the poses (other than some trouble with the fingers).  I'll run a test and see if I'm overlooking something. 

Apparently the trick for setting the "favored IK angles" is discussed in Bloodsong's book, *Secrets of Figure Creation with Poser 5. * Google books previews the opening paragraphs of that section, but omits the part where the secret is revealed.  :lol:

What limits are you setting on the joints?  That does sound like a move in the right direction. 

Okay, I see it, now that I'm looking for it.  Sort of looks like she hesitates a bit before placing her left foot, in the middle of the animation.  :cursing:

Here is what Frame 11 of your AP4 walk.pz2 looks like in P6, when implimented in different ways. The green outline is the pose used as an animated pose, with no involvement of the WD. White is the figure following a straight path in the WD. The solid figure is using Walk in Place in the WD.

All look quite diffrent. The animated pose and walk in place look identical above the thighs, but the legs are brought together a lot using Walk in Place.

I'm still geting some knee bukle with your poses when following a path. Do you get that in P8, or what ever you are using?

Quote - I'm only noticing this problem when a walk path is involved.  When walking in place, I don't see a problem with any blend of the poses (other than some trouble with the fingers).  I'll run a test and see if I'm overlooking something.

Yes I'm talking about following a path. I have not really tested Walk in Place to see if it happens there.

Note tht the hip seems to be lower, and her right foot higher in the walk path than in the animated pose!

Cage, you asked about the limits I was using in the shins. I forced yrot (twist) to zero, and right and left zrot (swing) to -1.5 and 1.5 respectively. The zrot setting is just personal taste, zero should work just as well. I also forced the foot swing to zero, though I should probably have given it a bit of an outwards bias. I left the limits un-forced so I could turn them on/off via the Figure menu.

I am using Poser 8, yes.  I have noticed the knee buckling problem in the middle of a mixed animation using walk in place.  So it isn't wholly the path-following function, but I think that complicates matters.  I think part of it, with paths, is that it's trying to calculate the stride lengths to fit the characteristics of the path.  When the figure turns, it puts down a foot and pivots on the weight.  Which is as it should be, but I think in order to accomplish that many other things get snafu'd.  :lol:

Basically, our problem is whatever automatic posing is being applied by the WD, rather than the input poses.  And that's where we really can't control things.  :sad:

I'd be interested in seeing a control test, using Posette or another figure, to see if the peculiarities with which we're struggling show up for other figures.    That should help us understand what may be able to be corrected and what we just have to accept.  I was planning to try this today, but I've been distracted.  :lol:

What can we vary, to try to control results?  I have:

• thighLength
• hipHeight
• feetDistance
• "Favored" IK angles
• Joint rotation limits

The first three are apparently now correct for the figure.  I'm not sure how to set the IK angles (drat the search functions in this place :lol:).  Setting up joint limits, at least in the legs, may help us, but I wouldn't want to force unnatural limits.

Is there anything else you can think of?  Perhaps it really is the source poses and we have euler problems of some sort, due to the differing joint defaults?

Quote - Setting up joint limits, at least in the legs, may help us, but I wouldn't want to force unnatural limits.

Well real shins don't do zrot (at least mine don't). I don't think they yrot either. When it looks like those things are happening, I think it's really thigh twist that we are seeing. Besides I not suggesting forcing the limits, just setting them, so they can be turned on and off via the File menu.

Quote - Is there anything else you can think of?

in the pose file:

hip ytran
shin and thigh yrot

As well as hipHight (the parmeter), I think actual ytran value if the hip in the pose is worth investigating. bringing it up a bit might prevent the buckling. Playing with the yRotations of the shin and thigh may also help, but sounds devilishly complicated.

i've got bloodsong's book. is this the info you were looking for? p.159, Initial pose, IK favoring, memorizing "The other reason for posing the figure is the Favored IK Angles.... To get your limbs to bend properly when IK is on, you should bend them slightly (3-5 degrees is plenty) in the direction they are supposed to bend. Your thigh, for example, should bend forward and the shin bend back. Then, when the IK is applied, the limb will tend to continue bending towards these favored angles.... The Favored IK angle works with the initial pose, or with the current state of the limbs when the user turns it on. It is not dependant on the memorized state of the figure."

Quote - Well real shins don't do zrot (at least mine don't). I don't think they yrot either. When it looks like those things are happening, I think it's really thigh twist that we are seeing. Besides I not suggesting forcing the limits, just setting them, so they can be turned on and off via the File menu.

Sorry.  Poor choice of words.  :lol:  Really, it may not be bad to set up some more restrictive joint limits on a special Antonia-WD build.  Assuming we'd use the figure solely for generating poses for the main Antonia version, restrictions which help create the best walk poses might be a good thing.  :unsure:

Quote - As well as hipHight (the parmeter), I think actual ytran value if the hip in the pose is worth investigating. bringing it up a bit might prevent the buckling. Playing with the yRotations of the shin and thigh may also help, but sounds devilishly complicated.

I'll try a version with the hipHeight raised on Y.

If we can determine consistent changes which are needed for any limb in the poses, we can apply them using a script, much like what's been done so far.  I'm afraid the worst troubles really are coming out of the default handling applied by the Walk Designer itself, however, regardless of the poses.  :sad:

Quote - i've got bloodsong's book. is this the info you were looking for?

p.159, Initial pose, IK favoring, memorizing....

Thank you, jancory!  I'm thinking about picking up the book on Amazon next month.  :laugh:

The trick is more straightforward than I thought, but unfortunately I don't see how it will actually help us here.  So I'm sad.  :sad:

Although, it would be good to create a "Favored IK" pose for Antonia.

@odf:  If you're there, what do you think about adding an IK-priming pose to the main distribution when version 1.0.1 comes out? 

Cage, here is a result I got using your AP4 walk, and setting limits before applying the walk cycle. It seems to completely cure the knee buckle. I think the legs are now a bit too close together, but perhaps that can be fixed by modifying the pose file. I like the fact that using limits has brought the legs closer together (not intentional), but think it may have gone a little too far.

Overall I'm quite pleased with the result, I think it is an improvement, but this is early days, and just one test of one walk blend.

There are problems with the feet, the toes are going below the ground, but that was also happening before limits were applied.

So from where I am now, I see three major problems remaining. Getting a bit more separation in the legs, and fixing the feet so the toes do not go below the ground, and curing any sliding that is happening for the feet. I have not checked for foot sliding so far.

I will attach a pose for the limits in my next post.

Quote - If we can determine consistent changes which are needed for any limb in the poses, we can apply them using a script, much like what's been done so far.  I'm afraid the worst troubles really are coming out of the default handling applied by the Walk Designer itself, however, regardless of the poses.

Yes, I agree, it's the default handling applied by the Walk Designer itself. The up shot of that is that it may be better in terms of results to apply corrections after running the WD, rather than applying them to the pose before running the WD. This is more inconvenient for the user, because it mean an extra step, but may not be avoidable.

What we really need is a better Walk Designer, but until that happens, we must continue to do battle with what we have.

I think the Favoured IK Angles would apply to the current pose in the current frame. I seems bloodsong is suggesting other wise, so I may be wrong. On the other hand, bloodsong might not be thinking in terms of animation, but talking about one static frame. In that case a priming pose would probably be of little use.

P.S.
In a previous post where I told you the limits I was using, I got the values for R and L shins the wrong way round.