Forum: Poser - OFFICIAL


Subject: Ok , it has been a year - where is Poser 7?

jpiazzo opened this issue on Mar 25, 2006 ยท 32 posts


jpiazzo posted Sat, 25 March 2006 at 4:38 PM

Ok, 6.5 would be ok - am I asking too much!

JP


Casette posted Sat, 25 March 2006 at 5:15 PM

I don't understand the hurries of people desiring a new (and expensive) Poser version... I prefer to ask for a SR3 :P


CASETTE
=======
"Poser isn't a SOFTWARE... it's a RELIGION!"


Miss Nancy posted Sat, 25 March 2006 at 5:32 PM

due to problems with vista and intel universal binaries, I sincerely hope they wait at least another 6 months before releasing P7.



jpiazzo posted Sat, 25 March 2006 at 6:54 PM

I'll pay, I'll pay... But I am expecting MAJOR new features! In particular, I expect a complete overhaul of Posers Animation features. Complete Non Linear Animation features, multiple layers, motion blending - (it should be like MotionBuilder and Endorphin) - with a built in implementation of poser physics and soft / hard body dynamics, animation paths, and an integrated lip-sync application like Daz's mimic. Then we should have a new feature called Poser AI - where you type into a text box - "walk ten steps forward, turn left, smile, wave and sit down" AND BINGO, Poser animates that automatically using a built in library of animation moves! This of-course comes with a whole new set-up room and grouping tool - we will also have a built-in UV mapping tool like UV-Mapper that will have a one button click for our textures to load into the image application of our choice for editing! The FACE ROOM is combined into a new full body MORPH CENTER with direct mesh sculpting manipulation tools (like Z-brush) along with automatic morph creation. Again a one button click takes us to a seamless integration with SHADE so we can model more complex shapes with ease. Even problems with changing or adding polygons are GONE - we just get a new figure! CLOTH, HAIR improved dramatically along with new Hy-bred clothing feature called DYNAMIC CONFORMING. Check a box and switch between the two types depending on pose situation. All existing clothing props would be automatically converted. Key frameable of-course. POSING of fully clothed figures is improved by a new VIRTUAL HANDLES feature which adds a easily selectable nodes to the figure that will manipulate body parts without having to make sure you are selecting the base figure rather than the prop when you move it. RENDERING - Poser now uses SHADE's render engine as an option, along with major 3rd party engines such as Renderman. Toon rendering is a direct link to Manga Studio's shaders and projects. Flash export creates a Motion Artist file. IMPORT EXPORT options support FBX and all major applications now supported by plug-in such as body studio. VUE integration is seamless because e-on and e-frontier have merged into one company. I know this sounds crazy BUT. Allot of this functionality already exists in a hodge podge of external apps, python scripts, and plug-ins. Hey, "E" Lets really put these talented Poser programmers to work! I just installed ADOBES Production Studio - interoperability between apps IS THE FUTURE - lets get with it! JP


dlfurman posted Sat, 25 March 2006 at 7:31 PM

Nah. Dont rush it. Let them read our comments here and pull out the good stuff.

"Few are agreeable in conversation, because each thinks more of what he intends to say than that of what others are saying, and listens no more when he himself has a chance to speak." - Francois de la Rochefoucauld

Intel Core i7 920, 24GB RAM, GeForce GTX 1050 4GB video, 6TB HDD space
Poser 12: Inches (Poser(PC) user since 1 and the floppies/manual to prove it!)


Gordon_S posted Sat, 25 March 2006 at 9:37 PM

Or maybe fix the memory management so that it doesn't fail to render large scenes.


madriver posted Sat, 25 March 2006 at 10:19 PM

Jpiazzo---love your improvements. But is it even technically feasible to do the auto animation stuff yet? Even motionbuilder can't come close to doing that. This actually sounds more like Poser 9.5 rather than 6.5 :)


svdl posted Sat, 25 March 2006 at 11:08 PM

"Interoperability between apps is the future" I agree. MS Office has had it for years, and automating Office is a delight. A powerful programming language (not scripting language, it has strong typing and will store a compiled version), callable from external 3rd party apps. So if Poser 7 (or 6.5, or SR3, whatever) would just expose itself as a Python library, so that it can be scripted from outside Poser itself.... that would create unlimited possibilities. On another note, conforming/dynamic hybrids already exist. They work in Poser 5 and up. I haven't made one yet that's good enough to post in freestuff, but I'm getting there. EnglishBob already has some hybrids out there, so has Jim Burton. Oh, I'd love a plugin render architecture. Even if a scene could just be exported to a Renderman RIB file I'd already be very happy. An overhaul of the walk designer so that figures can walk on something else besides the ground plane would be very welcome too. But most of all, a serious overhaul of the core. Bring it up to contemporary coding standards. Get rid of the Mac OS 7 artefacts. Make it rock solid stable. THAT is what all users want most, more than a load of new features.

The pen is mightier than the sword. But if you literally want to have some impact, use a typewriter

My gallery   My freestuff


3dkaya posted Sat, 25 March 2006 at 11:13 PM

I would like to have Poser 6 for at least a year before 7 comes out. I got my copy April 15, 2005 (digital download). Actually another 6 months hopefully be great but I won't buy it when right when it comes out.


ratscloset posted Sat, 25 March 2006 at 11:59 PM

It was almost 3 years between P6 and P5, but part of that was ownership changes I think. Give it time. It would be nice to have those Animation features, but I have to say that I would want to make sure I could afford it!

ratscloset
aka John


Netherworks posted Sun, 26 March 2006 at 12:17 AM

Why not true plug-in architecture? Then things can be developed upon and installed as we and the industry expands. Plugin a new renderer. Plugin a new animator. and so on. The python plugins are fun to work with but I'd like to also see things that are a bit more bolted in. Yeah, I too would like to see true undo/redo. Better standardized coding. Also choices of interfaces... I love the Kai (I know I'm in the minority) but hey why not make it skinnable where you can get a standard look with standard dials if folks want that.

.


xantor posted Sun, 26 March 2006 at 4:16 AM

What I want for poser 7 is being able to use and render double sided polygons.


dona_ferentes posted Sun, 26 March 2006 at 7:30 AM

And dynamic bones, like in Animation Master. That would be SO cool for us animators.


j_g posted Sun, 26 March 2006 at 11:20 AM

Yeah, support for ActiveX scripting would be cool. Then a programmer can write an add-on (plug-in -- whatever you want to call it) in the language of his choice.


stewer posted Sun, 26 March 2006 at 2:45 PM

"Even if a scene could just be exported to a Renderman RIB file I'd already be very happy."

Built-in since version 1. In the menu, "File/Export/Rib..."

"Plugin a new renderer."

http://www.stewreo.de/poser/poserman.html

"What I want for poser 7 is being able to use and render double sided polygons."

Check "normals_forward" on the root material node. Message edited on: 03/26/2006 14:47


jpiazzo posted Sun, 26 March 2006 at 4:02 PM

Gee I just thought talking about this would be fun - but boy have I learned alot from this thread! Already downloaded - Poserman / 3DELIGHT - off to experiment. Thanx stewer Searched out Englishbob's stuff, had not been to morphography before - good Tutorials - I'm very interested in this conforming/dynamic hybrids stuff - but my total tailoring experience is limited. Thanx svdl even found this interesting discussion link - http://www.renderosity.com/messages.ez?Form.ShowMessage=2228908 And THANX everybody here - this is the BEST forum ever! JP


vilters posted Sun, 26 March 2006 at 4:14 PM

While we are at it; the finished figure should pop out of the screen, blow a kiss, wash the car, clean the house, iron, kook, Sorry, it was too strong, could not help myself> Just, make it stable, please. Integrated UV mapper. Integrated face and body room that works by magnets or handles so that any figure can be used. New high AND LO-res figures that look like people and not like the next generation pin-up club thing. And for the High res figures, I just deleted all figures that had a higher poly count than Posette. Got it? Breasts, no balloons. This is a hard one, I still have to find the first figure with breasts. Gravity would be nice too. And, euh, stable please, Thanks

Poser 1, 2, 3, 4, 5, 7, P8 and PPro2010, P9 and PP2012, P10 and PP2014 Game Dev
"Do not drive faster then your angel can fly"!


jpiazzo posted Sun, 26 March 2006 at 7:21 PM

I don't know - P6 has been very stable for me - much better than P5. I recently worked for over a week on a pretty intensive animation project - not one crash (although I still did do a save very chance I could get) - an AUTO SAVE feature would be nice! I would give poser an 8 out of 10 for stability (cloth hangs sometimes)... My bullet proof award goes to Adobe! Damn, I threw ever thing I had at After Effects 6.5 recently - well beyond my system specs and memory - but I could NOT crash it! Menu bars flashed white, task manager said not responding - but, after a bit - the program would recover!!! over and over again! I TOTALLY agree that I would like to see human figures that more reflected the reality of the diversity of human form rather than some Greek / roman / cosmopolitan/ ideal that has no relation to reality! I mean, does anybody look like V3?! Please, give me characters that I have to buy ADD ON morphs to make them super-models - not the other way around. The super-size me collection! JP


R_Hatch posted Mon, 27 March 2006 at 12:18 AM

ActiveX (and DirectX as well) is quite awesome, provided you're running Windows. A better fit would be a C++ SDK. Remember that there are folks using Macs for posing/3d work.


j_g posted Mon, 27 March 2006 at 2:53 AM

A C++ SDK would then limit people to writing an add-on in C++. With ActiveX, they could write an add-on in Python, Java, Visual BASIC, C++, C, Delphi, .NET, C#, Perl, etc. That's a lot more developers you're targetting. For me, a C++ SDK would be fine since I know C++. But I'll bet most of the would-be plug-in developers would be using Visual Basic, Python, and Java because those languages are geared more for hobbyist programmers (ie, the kind who tend to write plug-ins). What's the use of having a plug-in architecture for both Windows and Mac if you've got very, very few developers writing plug-ins (because they have to know both C++, and Poser's "C++ SDK" which no doubt would have a lot less docs and examples than ActiveX, if Poser's user manual is any indication)?


Hubert.Holin posted Mon, 27 March 2006 at 10:19 AM Online Now!

Python scripts we can do already... with the proviso that the version embeded within Poser has no usable Tk for the Mac. Embedding a newer version would likely mean a port to XCode (for the Mac version), which anyway is a must w.r.t. MacTels. A C++ SDK (not mentioning C++ extensions to Python, which should already be possible, though I have yet to try), would of course be welcome. Any platform-specific coding language will only wield more bitterness and userbase fragmentation... not a good thing in my book! And who in his right mind would use Java :-P ? Hubert Holin


stewer posted Mon, 27 March 2006 at 10:57 AM

Attached Link: http://www.swig.org/

You can use C or C++ libraries in Python plugins, for example PoserPhysics is doing that. There's a pretty straightforward API for communicating between the two languages, and SWIG is a very useful tool to generate Python wrappers for C/C++ code.

svdl posted Mon, 27 March 2006 at 11:16 AM

I definitely prefer Python. It's a good language with many possibilities, there are good free editors, a plethora of modules and extras, and cross-platform is very good. Tkinter is broken on Mac OSX - no fault of Curious Labs/e-frontier, the Tk code just never has been ported to OSX. Instead of clinging to the antiquated Tkinter, it would be much better when Poser exposes wxPython - a far more modern approach to writing GUIs in Python, and fully supported on just any imaginable operating system. While I really like ActiveX for its ease of use and its raw performance, it has the distinct disadvantage of not being portable. Stick to Python and expose all functionality using SWIG.

The pen is mightier than the sword. But if you literally want to have some impact, use a typewriter

My gallery   My freestuff


Hubert.Holin posted Mon, 27 March 2006 at 3:54 PM Online Now!

Yes, I am aware of SWIG (and similar), but I am partial to (and biased towards) Boost.Python which I do use... My comment was that I am not confident it would be easy to assemble the toolset needed to compile extensions for the Python embeded within Poser 6. It is a nightmare as it is trying to play catchup whith the various incompatibilities between XCode 2.1, XCode 2.2 and the various Python distributions currently available for MacOS X (situation fast improving, though) to try to cater to that embeded version in addition, and stay (reasonably) sane. Well, as you point out, it's been done (I did purchase Poser Physics, but did not yet have the time to play with it :-( ), and there's a distutils for that Python so that's doable (if perhaps not pleasant). Tk is legacy, even for Python, but wxWindows (which wxPython wraps) is not top-notch standard C++ so... and the last time we did have a discution about doing a new C++ standard abiding GUI framework on Boost, I believe things faltered out... (R.L. is a b*tch!). We did get ourselves mired into the One-Only-Look vs Look-as-your-platform-does-but-do-not-code-to-it though... So there currently is no real happy middle ground. Anybody wants to contribute? Hubert Holin


stewer posted Mon, 27 March 2006 at 4:56 PM

Xcode won't cut it for writing Poser 6 plugins - Xcode/gcc/ld don't know how to build CFM binaries. You need to use CodeWarrior, that knows how to build both MachO and CFM.


Hubert.Holin posted Tue, 28 March 2006 at 10:47 AM Online Now!

CodeWarrior is DEAD No matter what we feel about it (or how much we have invested in it), there is no bringing it back. XCode et al. is the ONLY way (currently and for the foreseable future) to build native MacTel binaries (this has been hammered pretty clearly to us by Apple). The crux of the problem (at least the main problem), indeed might be that Poser 6 still relies on CFM code. I do not know that, as I have not investigated seriously that problem, and do not have any insider information (since I am not an insider :-) ). Tcl/Tk does indeed work on MacOS X, it is simply not CFM-abiding, and thus not callable from CFM code (if that's what Poser 6 is, a very reasonable guess), unless one uses some arcane trickery (I do not recall if what had been developped was for calling CFM from MachO or the other way around; I do remember it was first said to be impossible, so someone did go and prove everybody else wrong; I was on MacOS 9 at the time so did not pay much attention to it, unfortunately). No such trickery will be available for MacTel software (unless again someone proves Apple and everbody wrong ;-) ). Perhaps we could sum up the present situation under MacOS X as: Poser 6 relies on end-of-life, or beyond, tools to extend, Poser 7 had better be ported to XCode & al. to have any chance of survival. Let us just hope that the PTB are aware of this. Hubert Holin


svdl posted Tue, 28 March 2006 at 11:29 AM

I'm not a Mac guru (or even Mac user), but Hubert.Holin presents an excellent argument for a full rewrite of the Poser core. Unless e-frontier wants to ditch Mac support, of course. wxWindows may not be the end-all of C++ programming, the framework is well accepted on multiple platforms. The practical approach would be using a framework that is widely accepted and used, and since P6 already uses parts of wxPython, extending the support would not be a really painful operation. Java? Would be nice. I prefer strong typed languages. But I haven't seen anything that plays nice with Java unless it has been written in Java itself. And I seem to remember that Sun is far more restrictive when it comes to adapting Java to the needs of the application than Python. Not a technical issue, a political one. Javascript? Maybe, but its object handling is peculiar to say the least. As for calling CFM code from MachO and vice versa: I'm fairly sure it involves sockets. As long as CFM runs on Mac OSX (and I'm pretty sure CFM will be dropped soon) it's possible to communicate using sockets. Slow, but it'll work.

The pen is mightier than the sword. But if you literally want to have some impact, use a typewriter

My gallery   My freestuff


stewer posted Tue, 28 March 2006 at 12:25 PM

"CodeWarrior is DEAD"

Now it is. At the time Poser 6 was written, Xcode was still at version 1.5, which with all due respect, sucked.

And no, you don't have to do a "full rewrite" to get a MachO version, an Xcode version or a Universal Binary. "As for calling CFM code from MachO and vice versa: I'm fairly sure it involves sockets." It doesn't. See http://developer.apple.com/samplecode/CFM_MachO_CFM/CFM_MachO_CFM.html Message edited on: 03/28/2006 12:27


stewer posted Tue, 28 March 2006 at 1:25 PM

Oh, and one little thing: Shade on the Mac is built with Xcode and will ship as Universal Binary soon (Version 8.5). Shade has a PoserFusion plugin (also built in Xcode) and guess what the core of PoserFusion is...(hint: the plugin file is called MiniPoser.shdplugin).


stewer posted Tue, 28 March 2006 at 1:25 PM

Oh, and one little thing: Shade on the Mac is built with Xcode and will ship as Universal Binary soon (Version 8.5). Shade has a PoserFusion plugin (also built in Xcode) and guess what the core of PoserFusion is...(hint: the plugin file is called MiniPoser.shdplugin).


vilters posted Tue, 28 March 2006 at 1:55 PM

Open Poser 914.00.253, SR 1245 File - new goto - Internet Google for; ( enter name of person to be build ) Make ! Render as ; ( enter name of favorite movie ) y"ve got 3 minutes. Thank you. LOL.

Poser 1, 2, 3, 4, 5, 7, P8 and PPro2010, P9 and PP2012, P10 and PP2014 Game Dev
"Do not drive faster then your angel can fly"!


Hubert.Holin posted Tue, 28 March 2006 at 3:42 PM Online Now!

"CodeWarrior is DEAD" > Now it is. At the time Poser 6 was written, Xcode was still at version 1.5, > which with all due respect, sucked. I never stated that XCode was superior to CW. Indeed today I still lament some of it's functionality (debugger anyone?). And let's not delve into "optimisation"... However, even when XCode was at version 1.5, Metrowerks was well on its way to finishing seppuku, which it has now gloriously (and ineptly, I may say) acomplished. > And no, you don't have to do a "full rewrite" to get a MachO version, an > Xcode version or a Universal Binary. No, of course, but you do have to do a port, which is (almost) never a trivial thing (even between versions of a same toolset), whatever marketing slugs may say to the contrary. In this case, I would guess there is quite some platform-specific tweaking to make it run acceptably, and I would think that part has to be largely re-invented for the MacTel. I would not like to face the prospect of running Poser under Rosetta, for speed reasons alone, not to mention extensions building problems... > Shade on the Mac is built with Xcode and will ship as Universal Binary soon > (Version 8.5). Shade has a PoserFusion plugin (also built in Xcode) and > guess what the core of PoserFusion is...(hint: the plugin file is called > MiniPoser.shdplugin). This is a good portent, at the very least! I eagerly await my free upgrade from Shade 8 to Shade 8.5 (even though it will probably be at least a year before I can buy a MacTel, Shade 8.5 seems just great!). Well, for the other side of the fence, even though the processors will not change much for them, I would think the forever forthcomming "New" OS from redmont will throw a few curves in the development of Poser as well, so perhaps the porting to XCode could help there too (if only by forcing a clean-up of the cross-platform part of the code). And, who knows, with a clean(er) code, and bold ambitions, they might perhaps be tempted to cater to the Linux crowd as well (OK, this is a totally speculative presumption on my complete outsider part). After all, D|S does not. And Python does. Hubert Holin