Collection Video Wildeer Studio Collection [2023-11-16] [WildeerStudio]

4.60 star(s) 15 Votes

Yustu

Member
May 22, 2018
218
237
...

About the sims, actually, being fluid sims in general can mean double precision, so, no need to dedicate or buy an rtx4090 for that.
In double precision, in theory a RTX4090 does around 1.3TFLOPS on that (with slight variations between models), and RTX2080 0.314-0.42TFLOPS (depending on exact model), a good old Titan V (ca 800-900 second hand from China) does 7.45 TFLOPS, between five and six times the speed of the 4090.
I did not measure exactly, but I have seen it even in practice, running some double precision simulations, two newer RTX A5000, despite being newer etc. cannot compare with the speed of a Titan V.
...
He is using blender and FlipFluid addon that does simulation on CPU:

Ryzen 9 3950X is good enough for that, not ideal and I personally would invest in TR, but taking into account fast peace of work that he does /s it doesn't matter at all what he uses for those.
 

cevitamin

Newbie
Apr 19, 2018
35
65
Ehm, I don't want to repeat something already discussed in another exchange, but don't forget that most probably he is also using the machines to play, since both his machines are really "gamer" machines, and his main machine (by his indication) is the one with the 4090, so that one is also the ones he uses to play games.

Estimated earning (let's keep it to the low end), 10k per month in the last month ( ), realistic, given the indication of 4100 paid subscribers.
Actually, I am being conservative, 10k with 4100 subscribers would mean each subscriber is giving a bit less than 2.5 USD per month, while most probably the amount per person per month is bigger, not difficult to see why he wants to keep it going for as long as he can, Lara is his cash cow (evil pun on some "size", intended :p :-D).

Would not be suprised if it is also a reason why, despite bestiality, Patreon has not bothered him, a creator that makes at least 10k per month and has 4100+ paid subscribers and 9100+ total subscribers (potentially meaning additiona paid ones), interest the credit card companies and Patreon as well, for the transactions and commissions fees ;-).

About the sims, actually, being fluid sims in general can mean double precision, so, no need to dedicate or buy an rtx4090 for that.
In double precision, in theory a RTX4090 does around 1.3TFLOPS on that (with slight variations between models), and RTX2080 0.314-0.42TFLOPS (depending on exact model), a good old Titan V (ca 800-900 second hand from China) does 7.45 TFLOPS, between five and six times the speed of the 4090.
I did not measure exactly, but I have seen it even in practice, running some double precision simulations, two newer RTX A5000, despite being newer etc. cannot compare with the speed of a Titan V.

If like someone had indicated, he is from former or actual (it can be a sensitive subject, that is why I put it this way) Eastern Europe, that means his earnings from Patreon put him in "criminal-level" income, but doing something that is legal (well, excluding the possibility of fiscal evasion, and unless turns out his country has a law against NSFW animations).

Though, it is not my kink, but still, since he is "milking" Lara, maybe he could add up a scene on that ? :p :-D :-D

Fully agree,

i have put to parallel render task 4 titan v cards and it is a monster rig. FP64 does like HBM2 memory and insanely wide memmory bus. Same went for calculation speed in ETH mining, although the latter is no longer possible.
 

ouch2020

Well-Known Member
Aug 11, 2020
1,811
2,451
Fully agree,

i have put to parallel render task 4 titan v cards and it is a monster rig. FP64 does like HBM2 memory and insanely wide memmory bus. Same went for calculation speed in ETH mining, although the latter is no longer possible.
Never done mining, bought my cards for other purposes, though I ended up doing also some rendering with Daz Studio and some video editing.
I expect to put in line two GV100 (for now), then add up one Titan V I have already, in one workstation (no rig, really a workstation), and keep the other cards I have in another workstation, maybe try to make it all in a rendering farm, though the first purpose for me was ML (but for that, I will need to finish getting some old 40-56Gb network stuff and go back to study some stuff).
 

ouch2020

Well-Known Member
Aug 11, 2020
1,811
2,451
He is using blender and FlipFluid addon that does simulation on CPU:

Ryzen 9 3950X is good enough for that, not ideal and I personally would invest in TR, but taking into account fast peace of work that he does /s it doesn't matter at all what he uses for those.
Hmm, thanks, I have looked at the page, and indeed they say they do not use GPU.
Did not look at the details of the method they use, but I am not an expert in the matter, and too long from my studies, so if they say so, we have to trust them - though actually, they don't really say that they would not get any advantage from GPU, just that they are afraid the effort to manage the switch CPU-GPU and back would be more than the gain, most probably, the algorithm they use is not so easy to parallelise.
At the same time, using the CPU increases the compatibility/portability, but there is not doubt that a simulator using parallelisation and any decent GPU would beat the performance of any desktop CPU, even the Ryzen 9, the mathematical performances are just not comparable.
But it becomes a bit of a philosophical point - that is what Wildeer is using, so, that is it.
 

Donnie Brasco

Active Member
Dec 15, 2018
797
2,956
I didn't get anything spectacular during the month of subscription. A few pictures and as many updates on his progress. But I took the opportunity to download some of his previous works, which I really like. I still think his works are among the best ever. And now that the long wait for the next work apparently doesn't really bother his subscribers. When it eventually comes out, which has been announced for a long time, maybe I will subscribe again for a month. However, I believe that some authors should be given a financial boost, but it should not be exaggerated. So in the end they always get lazy.
 
  • Like
  • Thinking Face
Reactions: rammby and oz1

sivertop

Active Member
Jun 9, 2017
545
2,601
he once said it's gonna be half an hour or even more for this part.
right...to be honest with the slowmo stuff he did in the last few episodes i "almost" believe it..but then im thinking..its wildeer there is no way he wont use a excuse for why its not that long (if he truly said what you wrote, i dont know).

Anyway i can wait, since im not a patreon/sub i have no right to cry about it...i gladly accept it as long as i get my Lara dose, especially now that there are pretty much no animators left doing anything lara related.
 

3Dsmutter

Member
Apr 23, 2021
129
502
Recent update:


"September plans
Hey guys!
The summer has passed and Halloween will soon be here... Now I plan to work on the 2024 Halloween animation at the beginning of October. So I should finish Gatekeeper before that. Nice goal. It looks realistic.
So I want to cut the GK P4 video in September. Publish wip full animation. And in October we are working on the sound. Meanwhile, I can start the Halloween project.
We will be voting on the GK P4 ATWT pose as soon as the video comes out and you see all the scenes.
I have already rendered the pussy scene and now I am simulating the cum at the end.
I will add one or two seconds of transition short scenes, but they will take a day or two at most.
Thank you very much for your great interest and lots of support."
 
4.60 star(s) 15 Votes