Home Artists Posts Import Register
Patreon importer is back online! Tell your friends ✅

Content

This will be a little ranty. Going and ahead and typing it up stream-of-consciousness.

The main bottleneck in my creation of these sets are two things:

  • Render times
  • Good predator characters

Render Times

For render times, I've basically had to spend an extra half hour at the start of every set getting lighting set up. Indoor scenes take forever, generally, my lighting has improved but it sucks compared to where it could be. Consider some of my sets are 70+ images, and the high end ones take 20 minutes a pop to render. That's a 23hr time investment (since computer requires manual intervention to render the next thing). You've probably noticed I seem to rely on either outdoor, or incredibly simple backgrounds.

On environments, this really comes into play in two ways:

  • restaurant scenes (I've wanted a restaurant forever)
  • internals (digestion requires glistening surfaces with increase render times by 3x)

Good Predator Characters

When I first started on Patreon, I thought, "Great! I'll be able to have scenes with new characters because I'll be able to justify purchasing them!" Unfortunately, I was wrong. There are basically only two good tongues:

  • the wormbeast
  • the toon alligator

Very few of the models play well conforming to other characters, and I have yet to find a way to do things with tentacles and so forth that's fast. I've got a litany of predator characters which have never seen the light of day because I can't get 'em to work. Outside of these I've listed, some can make the cut. However, if you can't rely on the tongue, then this leaves you with:

  • size differences
  • camera angles & bulges for swallows

The biggest thing is, if you're going to just have a gal get swallowed without tongue play, you need digestion. But - due to above - I can't do digestion. So this has locked me out of a lot of areas and storylines.

Optimization Attempts

I've spent a ludicrous amount of time trying to find iray cloud providers, pricing render setups, investigating other software to create sets (blender, sfm, etc), figuring out lighting techniques, etc. I've spent so much that - believe it or not - I haven't actually made a single set this month. I made them all last month, scheduled them, and have devoted this entire time to try and figure this out.

I'm in this valley between "student" and "professional animator with company budget." So this means - even with the current monthly amount of support I get (thank you!)- after taxes, and after projected costs - I don't really end up with anything. This isn't about me posting on DA anymore, this is about you all, and me being able to consistently produce. It's not that this isn't a personal pursuit (it is), but it's not totally a personal pursuit, either. Given that this is essentially a second full time job in terms of time commitment and time is limited, I either need to scale up or scale down.

Although I posted about Octane earlier, I didn't get into it further. A lot of my lighting techniques didn't translate, and skin texture looked weird. Then, I saw this. Just watch the first couple video samples:

https://youtu.be/6xE3J56pabk?t=301

You see those 3-4 phases of clearing out fuzziness to arrive at an image? That'd be two hours of effort on my part, done there in 5 seconds.

That's 0.07% of the time it'd normally take.

That 23 hour set? Yeah, go ahead and cut that down to 45 minutes.

This opens the pathway to actually rendering out movies. (I have one I've wanted to do, "Tarisa Worm Dance," that's 1600+ frames long. At 10 seconds/render, that's still ~270 hours. Which is to say, a 25% of a typical month straight in a process where failure means starting over.)

Next Steps

I'm going to be moving to octane inside Daz3d, up front. I'll need to learn a bunch new things, so on, so on. Ok cool.

But SimpleGreenBag, what about 3d animation?

Well I'm glad you asked. I'm gonna then move onto Maya3d, at some point, and give it a go. If it lets me do animation in a fraction of the time, then I'll start being able to roll out both.

This will bump up monthly costs to basically breakeven again. Hopefully the trade is, I'll be able to produce a lot more a lot more quickly. So, sometime soon, my goal will be:

  • weekly content (continued)
  • "sfw" monthly promo set (high quality - I've noticed whenever I release a public, iray set, there's an influx of people)
  • video (tbd)

Reward Structure

Something that I, as a Patreon person, have been absolutely terrible at is making the different reward tiers worth it. I've got the same ones I set up initially, pretty much, and each experiment differently has resulted in a similar outcome: the amount of time it takes to coordinate and track it all increases the time commitment past what I can possibly do. And all of this gets in the way of the most important thing: consistently producing good content.

Moving forward, I'm gonna need to update the tiers to allow for upward growth. I'm thinking this will look like:

$5/$10/$20 stay the same. Although, possibly, moving forward $20 unlocks all content ever, and $5/$10 become access-to-that-month.

$50-tier: Access to video (tbd)

$Current 50 tier reduced to 1 slot (Since that's what you signed up for, Punny, grandather rate long as you go for it)

Past that, I'd love to think of other ways to go at this. Other kinds of rewards which are easy to coordinate, not sure. The issue is renders have historically taken so long to do that I can't both conduct a poll everyone sees, and get the render done in the month people paid for it. Maybe that will change.

Something I don't want to do  is lock down ideas.

Anyway, open to ideas here. If you've though "really wish that SimpleGreenBag would x, y z..." I'd love to hear it! I suppose I'll also need to think about marketing strategies.

Anyway, I attached a teaser image from an upcoming series. :) 

Files

Comments

Anonymous

Being helpful: Looking on their site, Octane Render is compatible with Geforce Cards and can load balance across multiple cards. Right now on the market there's a huge glut of used cryptominer cards as everyone shifts from CISC (Nvidia) to RISC Processor nodes. If you google Brutalus, you can see an example of a password cracking box based on 8x1080GTX Cards. Point being, find yourself a used quarter-rack off craigs list for a few hundred, pair that with a rosewill 4U Case and a high wattage power supply for the main carrier. You will find 2nd and 3rd Gen 12-core Xeons are going for peanuts right now and brand new supermicro mainboards are available on newegg; check out cpubenchmark.com and look at e.g. an E5-2696 v2. DDR3 ECC memory is cheap as well and above 32GB of total system memory becomes a requirement. Used processors never fail, I've repaired thousands of PC's, I've had 2 failures. If you play your cards right, you can likely do 24 cores of processing power, 64GB of ECC Memory on a Dual Processor Supermicro Board plus the aformentioned quarter rack and PSU for $1500. From there you hit up E-bay and find a seller willing to get rid of 4-6 Geforce 1080's for under $1k. You now have an obsene render farm box and the best part is, compared to your existing setup, you have the capacity to piecemeal it together over the next year. Keep in mind 1000W-1500W Power supply is a necessity.