Okay! Whatever the problem was, it got cleaned up in the latest version of the complete patch. Patch installed, program comes up fine. Yay.

I won't deny it -- lordbyron got an impressive likeness there. So under some circumstances, it can be done -- and he says he did it in an hour and a half. But I have to think he was fortunate in having a target that fell close to some morph norms or something. If the red contours won't snap, they won't snap -- plenty of others have posted the same results. Nor will I take back a word about the "pins." I've worked as much as five hours on a project and given up in disgust. But after seeing lordbyron's post -- which I thank you for, Tirjasdyn -- I will try a bunch of different targets, and see if I have just had the bad luck to pick hard ones up till now. (I've been searching the forum, though, and I don't see any other success stories like lordbyron's.) Bill And I urge you: try it yourself. See what you think.

I don't think it works very well, Artist3D. Putting the green control dots into place does not get the red contour lines (actual predictors of mesh shape) into the right place, and dragging the green dots around until the red lines come closer is a fishing expedition with no guarantee the red lines will ever match the face contour. Then when you try to correct the bad stuff with the Face Shaping tool you find out that yes, you can move an individual vertex but no, the "pins" will not hold the surrounding vertices in place so that you can't even correct a bump in the nose without shoving the nose into the face.

It looks like a good start, a good interface for the job, but so far, no, it doesn't work well.

I read posters who fell they're getting better results than I am, though none of them argue you can get real likenesses. It's true that I don't work with "plain vanilla" faces. But if I did, what would I need the Face Room for?

Bill Adams

Although, let me make it clear, my position is the same as Little Dragon's -- there's plenty to like in Poser 5, and I expect most of its negative aspects to be cleared up over time; meanwhile, I use both Poser 4 and 5 -- but if you task me, Tirjasdyn, to make specific what I don't like about the Face Room, I'll do it, at the risk of sounding like one of those whining naysayers.

You say "Just cause it doen't work they way you want it too doesn't mean it is doesn't work." That would certainly apply if I went on and on about the complexity of the Materials room. It is harder to use than I would like, but that's because it's very powerful; it certainly works, and I expect to get more and more out of it the more I explore it.

But I don't think my expectations or my desired way of working is the problem with the Face Room. It's just not delivering. You say you've seen some good posts, but are they good likenesses? I've seen some good texture merges out of the Face Room, and if your target shape looks enough like Judy, a texture can go a long way to impersonating a shape likeness. But if somebody is getting great shape likenesses out of the Face Room, I'd like to hear from them and find out what I'm doing wrong.

What I'm seeing is this: You can move all the green control dots into the right places, but the red contour lines don't necessarily follow -- at all! And they are a good predictor of the shape you actually get. In other words, this feature doesn't work. And when you try to fix the shape problems with the Face Shaping Tool, you find out that all that stuff you heard about manipulating individual vertices is semi-fraudulent. Yes, you can move an individual vertex -- P5 will manipulate all necessary morph dials behind the scenes to get that vertex into the desired place. Trouble is, all the other vertices affected by those morph dials also move. The "pins" that are supposed to hold them in place will not do so, period. Maybe the pins exert some kind of measurable constraint in some sense of the word, but they are of no practical use. You cannot do something as simple as reduce the bump in a nose without shoving the whole nose into the face. And if you can't do that, then the pins don't work and the tool is just another way of applying blunt-instrument morph dials.

I wish it were otherwise; I'm not demanding my money back because I think it will get better. But I'm not simply complaining about something that doesn't work the way I wish it would; I'm complaining about something that doesn't work. Try it yourself and see if I'm so wrong.

P.S. On-target reference, Phantast. Safire definitely invented the phrase, but you're right, it harks back strongly to the Krazy Kat era. You could imagine W.C. Fields saying it, too.

Here's another wishy-washy middle position: I like some of the new features of P5 a lot, haven't had the worst experiences of some other people (i.e., lots of crashes). But it is slow, the Materials room seems fantastically complicated and the manual gives no examples or tutorials to help you through it, and a few things (e.g., the calibration of the "real-world" measurements) are just wrong. But I continue to learn it, because I know bugs will be corrected, Renderosity whizzes will write us the tutorials, and the whole thing will run a lot faster the next time I upgrade my hardware. Meanwhile, my P4 remains installed, and I still use it most of the time for speed, without feeling like a big sucker; I choose the best of both worlds.

Of course, if the face room front-and-side photograph thing actually worked, I'd have cheerfully bought P5 for that alone. But in my heart I knew that was too good to be true, and don't hold it against CL. Besides, now that they've made the commitment and have a start on an engine for it, they may make it come true in time.

P.S. to Mark: the guy who wrote that speech for Agnew was William Safire, who has often regretted getting carried away by the alliteration. Bill Adams

Thanks, JeffH. You might have included

Attached Link: http://www.renderosity.com/messages.ez?Form.ShowMessage=787553

for that matter. That's where I first hashed it out with a bunch of you Renderosity habitues that the Poser unit was 8 feet. (Though I sympathize with geep, whose slightly shorter version, though without authority, brings all the figures even closer to U.S. average height.) However, whether you accept geep's measure or the 8 foot measure, the new P5 measure of 8.6 is surely a mistake. Posette a six-footer? Don 6'5". Come now. Anyway, thanks Jeff. This does confirm that others are seeing the same new measure in P5. Bill Adams

Hunh. Why aren't my pictures showing? Here are the links: http://wadams9.home.netcom.com/01Posercap.jpg http://wadams9.home.netcom.com/02Posercap.jpg http://wadams9.home.netcom.com/03Posercap.jpg

Bless you, Kalypso! At last an "How do I --?" for the Create Perspective UV button. (My own dumb fault for not searching these forums.) And yes, it is supposed to be in Poser 5.

It's the old Poser Manual problem -- lots of good information, but all on some higher, more abstract plane than that of actual use. Users ask, "How do I -- ?" and in some places, like the Material Room, it's almost impossible to find out.

My pet peeve, by the way, has always been the Create Perspective UVs Button in the Group Editing section. It sounds like such a great thing, in the manual. Does it just not work? Or is a Perspective UV actually created every time you press it -- somewhere? If so, where? What is it named? Can I have one?

Bill Adams

p.s. I see that same ridiculously high-pointed nose no matter what photograph I put in. B.A. again

Does it look like the originals? So far, I've had no luck with photographs. Then I thought of making the photographs gray-scale, brightening and lowering contrast until they were just barely there, and tracing in -- with heavy lines -- the contours they're looking for. So that the green points would be obvious and the red lines would know what to cling to. No dice. (Though actually, a little superior to what I got from the original photographs!) But there's obviously potential there. If they would just give us a LOT of green dots and a clear relationship to the red lines, like Bezier curves in a drawing program.

Bill Adams

Okay, I continue to research head-height proportionality, and here's what I learn: Metacreations uses heads as a base proportion for constructing figures because that's what most art teachers have traditionally done. And although Posette's "ideal height" of 8 heads doesn't have much to do with women in the real world, it's not out of line with women in illustrations. Here's a quote from John Adkins Richardson's COMPLETE BOOK OF CARTOONING (Prentice-Hall, 1977, and a terrific book of this kind, by the way): "The average man or woman is about 7 1/2 heads tall, in a proportion first noted by the Greek sculptor Polycleitus (active circa 450-420 B.C.) who worked out a canon of ideal proportions of which only rudiments have come down to us. Later Greek sculptors, notably Praxiteles (active circa 370-330 B.C.) and his near contemporary Lysippus, lightened the proportion to 8 heads. By the time of Michaelangelo (1475-1564) the ideal proportion for a male figure was 8 1/2 heads, a proportion rare among living men but extremely impressive in statuary. Because of its applications it came to be known as the 'heroic proportion.' . . . [M]ost adventure heroes in the comic pages are 8 or 8 1/2 heads tall." (The big exception is Prince Valiant, at 7 1/2 heads tall, and if you look at those classic strips -- based on the more realistic illustration-style of, say, Howard Pyle, rather than on comic-strip norms -- you'll see how different the proportions of all Hal Foster's characters look compared to those in a contemporary comic.) So . . . what does it all mean? I'd say Metacreations is not so far out of line using 8 heads for Posette's ideal proportion, now that we know their "ideal height" is really a "heroic proportion" for a figure a few inches taller than average height. We are all used to seeing this proportion in illustrations, so much so that real proportions can seem wrong on the page. (I have to admit that I never saw anything out of proportion in Posette until we started this discussion.) Most of us use Poser for illustration purposes, so maybe Metacreations was right on the money with 8 heads. On the other hand, everything we've developed so far in this thread confirms me in my long-held gut opinion that the "fashion-model" version of Posette is really a "basketball-player" or "glandular-case" version. It is Vickie who is the real fashion-model height, and a pretty tall model at that. I'll have one more comment on this when I finish collating all the "real-world" heights of the Poser characters. Which I can finally do now -- thanks, once again, to the great responses I got to my original question.

Okay, Mac is certainly right that Posette is 8 heads high. But I was overstating the case (ur, uh, or to put it another way, completely wrong) when I said actual humans were more like 6 heads high. The actual average is apparently closer to 7. Here's a link to what figure-drawing teachers think about these proportions: http://www2.evansville.edu/drawinglab/body.html (But note: although the actual figure is supposed to be 7ish, elsewhere on the page it says that 8 heads "looks better in a drawing". I guess Poser feels the same way! --------------------------------------------------------------------------------

That is strange, mac. Actual human adults tend to fall closer to six than seven heads tall, much less eight. Why do you associate the words CanonType with heads-height? (Just asking; I'm very ignorant about cr2s, myself.)

Thank you all for these excellent answers! I now know how to proceed.