Also note that when you use poser chars you are running smack into what is called the "uncanny curve"...google the word for details, but basically what it means is that people are much more forgiving of say a toonish mouse with human traits then a human doll with human traits. The closer the doll/figure gets to looking 'real' the less likely a viewer is able to accept any flaws in its look/motion. In fact it even starts looking very creepy simply because it makes us uncomfortable to see an 'almost' real 3d object that suddenly moves as if it had broken its knee or elbow (all too common in poser).

Quote - As for animatoin, I agree wholeheartedly.  But if only animation support for Poser stuff was more ubiquitous!   BodyStudio is one approach - but an expensive one - and you still need to do the animation in Poser.  Vue appears to be the best bet - but I haven't had a chance to test that yet (Vue 6 I).  My solution is to let you do animation in the target application (Cinema 4D).  Now you don't have to fret about doing the animatoin in Poser.

Agreed, which is why i would love to see them start supporting proper open file formats like COLLADA or FBX. That alone will make it a lot easier to use poser alongside other apps for animation and rendering. Well 3 'reasons' left to go, so one can hope...

Quote - Infinite undos?  Name one application (in existence) that allows 'infinite undos'.  Even the PROFESSIONAL, HIGH_PRICED (as in, I spent several thousand dollars) applications don't deliver that dream-state of a vision.  Good luck.  Have fun.  Snap out of it!

Well if you define 'infinite' as 'to the limit of available RAM' (which is what i think the OP meant) then many apps do have this feature. Not that im complaining of course, cause anything more then single undos is all for the better IMVHO :p

Anyway what i find strange is why some people are so gung ho about animation in poser... The fact is poser is not and has never been an animation app (no matter what they market it as) It has always been an artist's 'dummy'/ref app, sorta the virtual equivalent of the wooden ones you can buy in art shops. Its primary market, as EF well know, is tills/illustration, and you dont have to take my word for this either... A quick look around will show you that for every anim clip done in poser, there are 1000s of stills. So to expect them to abandon their userbase and suddenly switch to RnD in animation is kinda naive IMO. Remember that after all EF are a relatively minor player in the 3D industry, and they sell poser at a very low price to a conservative userbase. So they will think twice before taking any kind of major risk in the features they add/change.

Plus its not like you cant get awesome animation apps for cheap these days. Blender, a very decent little animation tool is free to download. XSI foundation at $499 has animation tools that make even Maya and MAX break out in cold sweats. For a little more you can get Lightwave, another very well rounded app that doesnt cost the earth. And if you only wanna do the anims for fun or learning, you can get free or educational versions of every major pro app for pennies like XSI mod tool, Maya PLE, Houdini apprentice etc (all 3 of which can be freely downloaded from the respective sites). If anyone is serious about animation, then these are the apps they should be aiming for cause not only are they quite capable, but they make animation fun and intuitive. There is no point in struggling with poser for this.

I thought it was pretty clear from my post that I was comparing single core to single core, and multi core to multi core... and the only single core archi from Intel in recent years has been the P4s. I guess maybe xeons could be considered a different CPU, but they are also clobbered by their opteron equivalents... and I did say Intel were killing them in the multicore deptt didnt I ?

As for 4 cores, Of course 4 cores arent as big a jump as 2 cores, that's just how SMP works. But 4 cores with 2 pairs of cores sharing a cache is still a lot better then 4 individual CPUs, which is all that was possible before they came up with the core 2 quadro. And who knows they may make a fully shared cache version in the near future, they are just getting started with multi cores after all...So i for one am quite excited by it, but i admit that its partly because I use other apps besides Poser that would indeed make good use of them.

Its not entirely marketing though, AMDs A64s actually do give the same performance as an equivalent intel of the same rating. ie an athlon 64 3500 gives you close to the performance an intel running at 3.5 GHz would. Even better the damn things overclock like crazy, my 3500+ is supposed to run at 2.2 GHz, but i use it at 2.8 GHz on air cooling without any problem. Thats an AMD 4400+ for the price of a 3500+ !

Too bad intel's core 2 duo/quadros tromp them in the multicore version...

Windows isn't the most reliable source for this so use something like :

http://www.cpuid.com/cpuz.php

to see what your CPU is doing (or any other freeware bench/info tool like Everest or Sandra)

Note i don't know what exactly they are doing 'under the hood' in the P7 engine but based on my experience with similar things in other apps,

1. kd-tree Accelerator: This is basically a way of organizing the data of your scene in a way that makes it faster to render raytracing. How raytracing (reflections, refractions, IBL, AO etc) works is that the camera shoots rays into a scene and then searches through the scene to see if the ray 'hit' any triangle/object. But if the scene is dense then this could mean a slow search through a large number of triangles to see just which one the ray actually 'hit' Kd-trees are a way of organizing the scene such that this process of search becomes much faster. You can find the tech details on google or something, but in general it should speed up everything that uses raytracing in any form, like the previously mentioned reflection, refraction, image based lighting, occlusion etc. But note that for best results the kdtree has to be 'tuned' for a scene by the user (if poser 7 allows this) and you have to find an optimal value for 'depth' and 'size' of the tree, which are the 2 values that control how the scene gets split into the kd-tree.

2. Tiled texture loading: Normally when you use a texture in poser (no matter what format) poser will load the entire texture into mem and then use it as and when it is needed in the render. This means that even if say the texture is on a tiny little cube in one corner of a scene, the entire texture will still be loaded and take up RAM. What the new method should do is that it will only load parts of a texture as and when needed. Like in the cube example, it wont touch the texture until it gets to the part of the scene that actually has the cube, and even then it will only load the parts of the texture it actually needs, not the whole thing. In other apps, this sort of thing usually requires that you save your texture in a special format that the renderer can use in this fashion, I dunno how they are doing it in P7, maybe they will do it automatically for every texture or have an option to convert them etc.

3. Irradiance Caching: now this is a tough one to explain cause its quite technical. But in very simple terms, it is basically a way to speed up GI. How it works is that before the scene actually starts to render, the engine sends out virtual 'photons' from the lights in the scene. These then bounce around the scene, adding light to it and gradually becoming weaker till they finally die out. After this process, what is left behind is a 'map' of how the light is interacting with the scene (ie how the light is bouncing around and contributing to the scene) This map is then stored to disk (the cache) and when the render starts, it can be used to simulate realistic looking GI in the scene and the photon map used to calculate how each light will contribute to the scene in a more realistic fashion.

Of course what these terms actually mean in P7, is something we will find out only after we get it.

Well its different for coding of course, cause there its clear cut just how original and useful your work is. Its not like you took a bunch of premade libs and just linked to them for the plugin now is it ? (and if thats all you did, then you WILL be laughed at, no offense)

Plus TD types are always in short supply and you couldv been coding Warcraft machinama pron for the last 5 months and they would still give you a look see i bet :p

Firefly is REYES is it not ? In which case like all other REYES renderers it should be rendering internally in float (or 32 bit per channel)

As for Poser's RAM usage, i'm with kawecki on this, since i also don't know what the hell is going on in the render. Either its not freeing the RAM properly, doesn't have a proper scene BSP in the first place or has some mem leak somewhere. Cause in both Lightwave and XSI (the 2 other apps i use with poser) i can load up 20 James high rez meshes with full 3k textures and the render does not cross 700-800 MB in RAM. In poser even 4-5 such figs start running into double that.

Hope they work on this in 7, cause throwing RAM or 64 bits at the problem is no solution. Even if Vista lets you cross the 4GB barrier, most non server boards in the market have hard limits on the amount of RAM they let you add. Many wont even run in dual channel mode after 4GB or will slow the RAM down to 333 MHz or even lower.. So adding more and more RAM is not as yet a very good option IMO.

No matter what you feel about the Poser bias (which is totally unjustified BTW) this doesn't change the fact that it's quite real and pervasive in the 3D industry. There is nothing you can do about this really, if you want a job, don't mention that you use Poser unless the job is for a 3d-related field like illustration or something. And even then you should be presenting your work as illustration/2D done in Photoshop/Painter and not a 3D render in poser.

Plus keep in mind that Poser is not the best medium to show off your skills in the first place. Poser makes things too cookie cutter and easy for you. eg if your render has V3, lights by A, clothes by B, backdrops by C etc, then anyone with the same kind of props and presets will get the same result from poser. This makes it very hard for a prospective employer to evaluate just what was 'your' contribution in a work. ie why should they hire you instead of the next guy doing the same thing ? Of course you may have made a better composition and scene, but frankly that isn't as important as you may think, since all that stuff is already decided (by the design, storyboard and concept guys) way before a scene ever reaches your computer.

Sure its different if you are your own boss or are working freelance or something, cause then only the results matter and as long as they are under budget/on time no one cares what you use.

Quote - Firefox 2.0 is still in beta. It won't be considered an update until it has been officially released.

It went final on the 24th:

http://www.mozilla.com/en-US/press/mozilla-2006-10-24.html

Quote - That's interesting as I'm running 1.5, but doing a "Check for Updates..." says there are none.  Maybe they don't consider 2.0 'an update'.

Actually you are quite correct, they dont update major versions with the check for updates thing. It didnt work for when they released the 1.5 version either IIRC.

If being laughed at by the CG world is a big deal for ya, you shouldn't be using poser in the first place, cause that ship has sailed. For the rest of us this is an example of a good solid feature that will help day to day work much more then any flashy new thing they could have added.

Plus if it makes you feel any better, you can always go to the Lightwave forum and laugh at them since until recently they would have killed to have multiple undos.

BTW Firefox 2.0 is out if anyone wants to get the new version:

http://www.mozilla.com

That old saw about CG: "fast, cheap or good, pick any two..." pretty much says it all here...

Poser is awesome value, it has cloth, hair, shader nodes, IBL, displacement etc for a very affordable price. Which of course means that speed has had to be let go. Even so Firefly isnt bad at all for what it does and it works quite well for most non commercial renders if you have patience and work with it a litle. Its perfect for those only doing this poser thing as a hobby since it produces nice quality and has sufficient 'tweakability' in the form of the nodes that you can get pretty much what you want. And of course once you do get to a place where you are doing commercial stuff, there is a very well defined line starting at inexpensive rendering apps all the way to full featured 3d suites that cost thous of dollars. And you can pick any one of these depending on what your clients are willing to pay/what your budgets are like. Same with hardware.

eg any of the advanced render engines out there like PRMan, mentalray, VRay etc chew through multi million poly poser scenes in minutes/seconds and they render very high quality hair, textures, reflections, IBL etc. But you are also looking at high cost and significant amounts of work in loading the poser scene into the app. Which may be viable if you are rendering a 3000 frame anim for an ad and the client wants it yesterday, but probably not if its just a funny still for uncle fred on a weekend.