No disrespect, and no trolling but seriously, there could be no simpler object to model. 20 minutes max. Oniy assigning UVs might take twice as long as the modeling process.

Signed,

Vertex Slinger

and (candidly) I don't know anything that no one else does.

My knowledge is limited, granted.

Neon lit cars in present day "old" Hollywood have undercarriage strips fore to aft on both sides, but rarely transverse (perhaps a personal choice, granted).

The overall environment usually has greater contrast - what's going on with the bright area on the right?- with less throw (influence? distance?) from the neons, since these lights will surely respect square inverse distance.

So in LDR: not working. HDR? Possibilities open a bit. Subtlety is never a bad thing.

re: from -> to

I'm doing something very similar.

Hips too large, rib cage too small around shoulders. Nipples should be more "outboard", with a stronger line separating chest from abdomen.

Overall, very soft looking. No indication of underlying bone or muscle (iliac crest, rib cage, serratus, abs). This looks like a lo res mesh, so not sure how much detail it can support. Highly recommend ZBrush Digital Scultping Human Anatomy. Even if not using the program, is really insightful.

To reiterate what others have said, it depends on the model and its usage. All in all, there are some generally agreed best practices.

If your model will occupy very little screen space (say, less than 10%), 70K polys may be a bit excessive. On the other hand, it's not uncommon for a "hero object" to have greater than 100 - 200k polys.

I usually try to "get the shape to read properly" using the fewest polys because, ideally, the scene should be heavily populated to provide the detail we're accustomed to seeing in the real world. As an example, when I look around my workstation, each object I see can be modeled with less than 4000 polys. Some may be crude but will occupy very little screen space. In that case, it's not so much about the accuracy than the impression. A human on the other hand, I would budget 30-50K polys with extra detail provided by ZBrush.

It's all about exploiting the shortcuts and catering to the strengths of the rendering system. Personally, I'm a huge fan of balancing poly weight / smoothing /subDs with textures and displacements.

As far as your castle is concerned, low poly for long shots, medium polys + textures / displacements - perhaps broken into modular sets for interiors and backgrounds. Otherwise doubtful that the castle would be the "subject", so err on the side of less.

The thing about Teyon's / Spencer's technique is that it can also be done in Photoshop / Gimp / et all once the UVs have already been prepared. The advantage of using ZBrush (in this way) is that you don't have to worry about matching seams.

It is a fairly easy but really powerful technique that is definitely worth study.

If I understand you correctly....each node renders a complete frame? That's not necessarily a bad thing.

Forgot to mention: if you decide to use PoserFusion, I would strongly recommend rendering each figure with alphas (and the building) and comp them in post. You can still get the job done and keep your hair =)

It will work but it depends on how much patience you have.

It's been eons but PoserFusion is descended from Reiss Body Studio. RBS allows the host program to use and render Poser animations with the following caveat: each frame of the animation is one of a series of frozen / textured .objs. Said another way, using this system there is no way to "re-pose" your figure without first going back to Poser. This can be a hinderance compared to Vue for instance, which allows you to repose within it.

A less obvious issue to consider: even using a mid poly character (such as Jessie / James) will slow down your interface considerably. That is not to discourage you, but just thought you'd like to know before dropping $$$. Would probably second wolf359 and consider using biped or whatever, which I'm sure has its own obstacles.

Impressive and ambitious.

Some of the character motion seems "dreamy" ; try editing tangents on the keyframes to give them more "ease in / ease out"

Success!!!

(1) each cow receives its own command port

(2) each system receives its own static IP. Not crazy about the idea but it works.

Cerro Verde.vue, rendered internally: 13m49s

Cerro Verde.vue, one node external: 18m18s

Cerro Verde.vue, five nodes external: 03m38s

Definitely worth the investment in time.

Currently using DCHP, but Hyper Vue and the switch are reporting wildly different IP addresses. One reports 10.1.x.x while the other reports 192.168.x.x. Weird.

Interim solution: 

per original post, no firewalls enabled, and on a separate, non-internetted wired 1000 mbps network.

Within each node's render cows, there is an "auto-discovery port" 5591 and a "command port" 500X. Though the IP addresses keep changing (hmmmmm....... ; will perhaps need a different switch), giving each node a distinct command port address and corresponding Hyper Vue address - though dodgy, at the moment - has me now seeing 3/5 nodes so I'm getting closer, though not yet seeing the speed gains I'd hoped for....

But why doesn't HyperVue (8.0) open as it's own executable (for offline / pre-render confirguration), rather than as an online Vue utility? ARGH!!!!

Will try and report back.

Prior to reading your response, tried changing settings in the router - I can see the host + 1 render cow, but don't know why the other three aren't showing. No worries; the time I invest here will be well worth the time I will save, no?

Switched in 2007, Poser on last years aluminum 24" iMac, no difficulties here.

I WILL tell you that there is only one thing I miss about Windows that I haven't been able to replicate on the Mac: single click when you hover on an icon. I miss it but not a deal breaker when you look at the rest of the package. Also, you should be aware that Mac hardware tends to run a generation behind "comparable" PC hardware, but I think that's an advantage: because the hardware is "proven" it is more predictable.. 

(sigh, here we go again) the primary strength of Mac OSX is that underneath its GUI runs UNIX, an industrial strength operating system under continuous development for 40 years with an emphasis on security. As a result, it is very stable and mostly invulnerable to viruses, spyware and the like. I crash the web browser from time to time but I have never been able to crash the system - and I have tried. I can give you a link right now that will kill Windows in 15 seconds, which has no effect on a Mac.

PC users like to tinker, upgrade and customize. Nothing wrong with that - except when devices don't play nice together. Because Mac hardware and software are "co-develeloped" the experience is harmonious and hassle free. The following statement will be difficult to comprehend: I spent ZERO time doing system maintenance. I haven't restarted my system in 24 days, 7 hours and 23 minutes. There is no such thing a "defragging hard drives", and it was mind-boggling at first: what do I do will all this free time? I can do art!!!!

I contacted Smith-Micro to let them know I changed platforms, I connected to the Runtime I already had and I was up and running in a few minutes. But many of the utilities are not available for the Mac. My two thoughts on this:

(1) Since I had fewer tools to work with, I got really good with Photoshop fast. Maybe not as convenient, but now I can do anything I want.

(2) this is a long shot, but if you can convince developers to give you the source code, you can recomplie for Mac OSX using tools provided with every Mac sold (xCode).

I'm not going to tell you to get a Mac but my experience has been pleasurable so far.