'Dress' at least says what it is. A cheongsam may turn out to be the ideal sort of dress I need for my project, but if it does not show up when I search for a dress I may not find it.

The thing being a dress is just one aspect. It also being chinese (style) is just as important. Then for need of variation or standing out in origing the maker may name it 'qipao' or, in other spelling 'chipao'. Then this would only work for users relying on a Latin character set. Not sure if a user working in chinese would see those items turning up in a search.

Of course this holds not only for chinese dresses for 'Kleid' (German), 'Robe' or 'Jupon' (French), 'Vestido'(Spanish) 'Jurk'(Dutch), or what you want.   

The name of a thing can be a unique identifier, but anything unique by definition is not suitable as a searchable term unless one knows the answer to the search already. John Wilkins Real Character was a nice try but it did not work.  

The name of the maker in the product name is just a waste of space. 

Really more effort should be put in providing searchable Metadata. Search terms should be standardized ('clothing, female, skirt typelower part'), and stores should not accept items without a fully filled metadata file and add a unique identifier of the item. 

However, this would reduce the waste of content by users (who will buy new things when they already own a similar item but failed to find it) so it would be bad for business and will not materialize.

In older Poser versions, the default ground prop was set up as a shadow catcher. So set up it gives the shadows the figure would have on flat ground  under the scene light. You can then layer the render over an image in Photoshop with more or less realistic ground shadow.  

You may try the cloth room. If you assign the verties of the zipper as  a series of groups which you can assign to decorated or a group with very low stretch stiffness and make the facets invisible, you may get the effect of a zipper.    

Indeed they appear to have some startup problems with the new site. MD3 works fine for me, but I agree in some aspects my expereince is a step back from MD2. Also, some of the new functions do not work as promised. One strong point: Simulation seems to work better and more smoothly.

Try to make your clothes in Marvelous designer. UV's are perfectly flat and surface deforms as real cloth would deform. Consequently you can use a procedural shader like from the 'loom' of BB's mathmatic.

My Calf Pants for Antonia use a more simple procedural shader with a striped effect.

Ah, that explains. I was talking about making conforming clothing. 

Well of course in MD you can pose the avatar in T-pose if that is easier for you to work with. Then, depending on the type of clothing, after swing you can change the sleeves to make fit. If there is a 'bellow' at the shoulder as in some medieval dresses of course that takes care of the lengthening.   

What problem do you have there draping sleeves then? Inclined arms works great for me. 

In fact my problem is not the draping in MD, but the application to Poser figures. 

Workflow:

• export .obj with lowered arms shape from MD (I do not want wrinkled shoulders and normal proportions of textureson the shoulder)

• load .obj in Poser, in setuproom inject the rigging rig made to suit the MD avatar (or lowered arm version of Poser figure if I used that in MD. Carrier figures are just body and main limbs,hand.feet. No fingers, toes, head features.

• in Pose room pose the cloth-figure in a pose to suit my target figure (T-pose, arms bend, legs position), tweak joints when necessary for good shape. Texture on shoulders of clothing look horrible of course, but UV's are preserved

•  in fitting room convert clothing to target figure.

• when posed to natural 'arms down' position the texture on the shoulders returns to more or less original proportions. 

So with T-posed figures I need to do an extra setup room session and can only hope the deformations induced in the bending to T-pose are reversed well when the clothing is bend to a pose in which the fabric should be neither stretched or wrinkled. 

If the 'arms low' pose would be the default, the .obj could go tot he fitting room directly.

Chothing in its natural position is almost always cut with arms down 45 degrees. If I would have say a jacket cur with arms in T-pose, I would be unable to lower my arms for the ball of fabric under my armpits. also, the fabric on my shoulders would be stretched to breakage.

Marvelous Desiger uses Arms lowered pose as a default, and all goes brilliant. If I want to fit my generated clothing with perfect flat and undistorted UV's to a T-Posed poser figure, disaster strikes. The sides of my jacket spread like wings.

I absolutely hate the default T-pose for this reason. 

I fully understand modeling arms and hands of base models is much easier in the T-pose. Textures however are much easier to make for the poses in which the skin is relaxed.

We need a tool to freeze a pose with all rotations etc applied as a new zero.  

Congratulations!!

(and thank you!)

I made it past the subscription hurdle of PoserClub. (with Google Chrome!) I must say  I like Ayame.

Nice detail is that her teeth and inner mouth load as a separate prop, so you can esaily delete those if she keeps het mouth closed and save polygons. With about 20K polies she is not a big consumer.

She has a limited set of morphs. 

She looks good and personal 'out of the box'. She comes with a hair prop. Bending is not great, in particular in the hip area at large angles, but she can do nicely for standing poses. 

I have a section in my Runtimes where she will find a place. 

My Japanese is somewhat rusty, so from this place: Z3D, thank you. 

Just beware that Have used sometimes Japanese limb names etc., which non-Japanese Poser had diffikulties to handle. For me the figures at the time caused problems with Poser 7 'hanging'.

This may be solved in present Poser versions or new OS systems.

Eroko's size and rigging resembled Victoria2's enough to make clothing items fit for poses not too far away from neutral.

I so recognize the felling and the fun. There is so much good stuff on the old trunks! It is fairly low-poly compared with today's standards, and if necesary Snarley's Subdivider is a great help. Then use EZSkin to add SSS, and you will not beleve your eyes.  Reality/Lux too provide subdivision and a new lore to old stuff. 

For me it works. Pro 2012, Reality 3 and Lux, all on an Athlon 64X2 running WinXP Pro 32 bits. Not a miracle of speed, but stable and good enough to see where results it goes as far as light and tone are concerned. Then if it looks good I close lux, and transfer to another (Unix) machine on which I have not Poser/reality of course, but 64 bit Lux and much much more capacity.  

Exactly, if you type 0.4 in the input box, and your input level is say 0.5, the result is 0.5*0.4 = 0.2.

The displacement channel is an ideal method to visualise the results. Use a unit square, horizontal, plug your equation (with U and/or V dependency) in the displacement channel, render, and watch your own 3-D graphs. 

Just make sure your work is so good people do not notice the are Poser renders and ride withthe rest of CG art an photography. 

The p-node retuns the co-ordinates of the point in universe space in 1/10th inches units: x-co-ordinate in red channel, y-co-ordinate in green channel, etc. so to have say output 1 at 1 foot use multiplier 1/120 = 0.00825

note Poser units are 8.6 foot. 

input values for the transparancy channel are scalar.

color inputs are converted to scalar by averaging the rgb channel values, so your output will be -1*(0+0+-2)/3 = 0.666667, your transparncy therefore should be 0.5*0.6667 = 0.333333

Inputs outside 0-1 are clamped to the nearest limit.