Penguinisto posted at 3:25PM Sun, 23 June 2019 - #4354241

erogenesis posted at 6:29PM Sat, 22 June 2019 - #4354233

Rendo have a fantastic opportunity here and they CAN do it. They do need to be careful not to promise too much, but if they just focus on Poser's qualities, refine them, they should be on the right track.

Agreed, but they don't have a whole lot of runway to work with... Poser is barely earning enough to keep a very small dev team going (3-4, tops), so they're going to have to either donate some extra cash from the store profits (which will be a bit tight for now, given that they probably have a bit of debt to pay down from purchasing Poser... unless they saved their pennies and paid cash?), or put the feature requests through a very tight-meshed priority filter and pick what will have the most positive impact.

Yesm I'm pretty sure its going to be a big risk for Rendo, so they must not beat around the bush. The absolute first thing they need to do is address the figure support issue. Without a good figure and tools to support it, there's nothing to Pose, therefore Poser is kinda pointless. I obviously recommend PE, go big or go home. But ok if they prefer LaFemme, then that will have to do, as long as they start refining tools to support figures. I would almost suggest they take the DAZ route, and absolutely tailor Poser to support at least one set of figures with autofit that actually works. My mannequin idea really does work and Verbaas thought up a coockiecutter compartment-like system to autogroup the figure. This cookiecutter system will minimize stray polygons and with a simple algorithm to absorb other stray polygons. Also, using philC's wrap technique, you can temporarily clingfilm the garment to the figure to transfer weight maps and JCMs to the garment, then replace the mesh with the original shape again. Or, use the tech behind the cloth room (or ask the creator of VWD to help) to help make JCMs in the garment, but then forcing it to stick very closely to the host figure to assure proper vertex alignment. Morph and WM transfer needs to be refined and they could use subdivision tech to do this (Nerd3D probably knows what I mean). There are many little tricks available to Poser right now, they just need automating.

Whatever they do they must absolutely get guys like Nerd3D, Larry, ColorCurvature and perhaps even shvrDavid on the team, since they really do know what is going on in Poser. I would also be willing to give advice.

You do kick-ass work (though the blonde up there looking like a very tall version of my ex-wife is creepy as hell), and your concepts look cool on the surface (and I like how you think),BUT... you do realize that world+dog is sniffing around trying to push Rendo to do X or Y (and yeah, often to prefer their products or favored products, but c'est la vie.)

Hehe thanks. I've kindof withdrawn from the Poser scene, completely focusing on comics, and I definitely don't expect SM or Rendo to listen to me and do as I say. My pov can be very limited and this is made harder with the lack of communication and feedback from SM. So essentially I have no clue of the state of things behind the scenes regarding Poser. But I do expect them to want to turn Poser into a killer app and make them lots of money. And so I figure since I do use poser quite intensely, on a daily basis, made a complex Poser figure, write scripts to streamline Poser's workflow, and thereby kinda proving that everything I suggest is probably achievable, I do still make an effort to make my opinion heard, because at the end of the day it will benefit the thing I like the most: making comics. It in my own interest to want to urge these ideas, and probably in their interest too, since I would assume that they would recognize some of what I do as something of value for Poser, and perhaps Poser's community... even if I am not the most reliable vendor, and kindof an impulsive hippie :D

So, yeah - kinda agree with what I think you're shooting for. BUT (you knew there was a 'but', right?) You can have the best of both worlds - draping and collision detection off to the side that can be turned on and refined by those who want it (they're already there and can be altered/incorporated as advanced features that can be turned on), but maybe by default lean more towards a hybrid 'make-art-button' sort of happy medium that doesn't turn CPUs into butter.

Yes that's kind of what I was saying. Using cloth room tech to make morphs, or even add a cloth room gadget into the morph editor, much like the sag brush, have the drape brush, or just drape on frame. In any case, I think the cloth room can provide a useful solution to automatically building and baking JCMs.

To be fair, they would have had to refine what's there... now the day will come where the underlying engine is gonna need an overhaul, but don't expect miracles. Real-life example? Outside of CG-land, I worked with updating/upgrading a product that rakes in $1.5bn/yr. It had comments dating back from 1997 (I wish I was kidding.) There had to be at least six different languages involved, two differing databases built-in (PICK in a VM, and a bastardized fork of Postgres). Most of it was Spaghettified all to hell. Took 20 of us nearly two years just to overhaul the underlying packages, while a handful of others re-worked the UI into something useful. Both QA and C-level armchair eyeballing were omnipresent and brutal (it always is when you're modernizing the company's main money-printer.) The three months of migrating all of the customer installs (each of whom wrote rather massive checks for the thing) was even more brutal. In spite of that, there were a number of things I (and the others) hated, but simply could not touch, let alone rip out and replace... not without a radical change in what the customer(s) expected and demanded (in performance --see also PICK-- in responses, in UI layout...)

The point is this - even on Poser's scale, upgrading/improving/evolving is going to be a long and painful process. DAZ has the advantage (well, a few) in that it was built from the ground up to be free from 1990s tech (especially Kai), to be sufficiently modular that pieces and parts can be ripped out, reworked, and improved, without any degradation or compromise elsewhere. Poser is a bit more monolithic, I believe (though I hope they changed that over the years), so it's likely not as simple as you think it is.

Then again, I think we both might end up pleasantly surprised.

You are totally right. I expect it would be a drama, but if they take it step by step with relevant updates and refinements, eventually Poser can become killer again. Poser's advantage over other apps is her UI, her ability to load scenes that are absolutely stupid in size, render using the GPU, and all the essential tools to make it all happen. They all just need refining for now, especially the figure support system. With clever people like Nerd and co, without the legislation BS of SM, Poser might have a future again.

"The fool is not the one that does something foolish, but the one that does nothing at all."