[a / b / c / d / e / f / g / gif / h / hr / k / m / o / p / r / s / t / u / v / vg / vr / w / wg] [i / ic] [r9k / s4s / vip / qa] [cm / hm / lgbt / y] [3 / aco / adv / an / asp / bant / biz / cgl / ck / co / diy / fa / fit / gd / hc / his / int / jp / lit / mlp / mu / n / news / out / po / pol / qst / sci / soc / sp / tg / toy / trv / tv / vp / wsg / wsr / x] [Settings] [Search] [Home]
Board
Settings Home
/3/ - 3DCG


Thread archived.
You cannot reply anymore.



File: 0098.png (22 KB, 100x100)
22 KB
22 KB PNG
I modeled an explosion in blender. It is just a short explosion running at normal speed.

Now I wanted to render it in slow motion. But guess what, there is now way to render smoke and fire simulations in blender in slow motion?

What the fuck, is that so hard? Just render 10 frames for every frame of animation to make it ten times slower. Fucking blender can't do that.

"Time Remap" is a piece of shit, you can't even scrobble the timeline anymore when you have it enabled and it does not affect simulations, so that's out of the question.

I suffer with this shit while those fuckers in Amsterdam are jerking off to their monochromatic icons and not fixing the real issues with this piece of shit software.

tl;dr
how to make slow motion explosion in blender without changing the physics of the explosion itself?
>>
You can't simply render in slow motion, you have to re-simulate sub-frames, or simulate more frames from the start and play at a reduced rate (say, sim 240 frames for 1 sec at 24fps). Everything else is just interpolating between simulated frames, which may give visually acceptable results but is not physically correct.
>>
>>662866
just render more fps.

If you want 2x slow motion render that much then play back at half speed lol
>>
>>662889
I don't care, after baking it should just interpolate between frames.

>>662890
This will not work with blender, since all the animations will then run faster. They all have a fixed length frame-wise. E.g. The explosion keeps a length of 24 frames, even if I render at 240fps.
>>
>>662895
Wait, did you actually bake your simulation but now want it to render in slow motion?

Holy shit are you retarded?
>>
>>662905
There is software capable of that.
>>
>>662907
There are countless solutions in Blender. I would reccomend using the questions thread, or the blender thread to ask.Also maybe google solutions or looking at the blender website.

I tell you this on the off chance you aren't here just trying to start a software wars thread. I swear fucking autodesk representatives and shills are getting lazy. I fucking know they do this stuff they are all over the fucking industry its so tiring.
>>
>>662866
Just render at a higher fps, Jesus fucking Christ.
Who cares if you already baked it, bake it again. You're not going to get more information out of something at 24 fps unless you make more information.
You can cheat and get something like Twixtor or some other interpolation program to get more frames out of it, but it won't be the same unless you go back and do it properly.
Once you have enough frames just slow it down to your source frame-rate. It's basic math and ratios. It'll work in either case.

Want a slow motion animation that lasts 4 seconds?
Take a 2s, 60fps animation (120 frames total) render it out, and half the speed.
You'll get a 4s, 30fps animation that lasts 240 frames.

It's the same principle how slow motion cameras work. They're not recording at 24fps, they're recording at crazy fps like 10,000 and you just play it at a common frame rate like 24fps which makes it look slow. If you played it at its base frame rate of 10,000 fps, it'd look the same as any normal speed video. You're just getting more information per second at 10,000fps compared to 24fps, for obvious reasons.
If you took something that runs at 30 or something and tried playing it at 60, it's gonna play faster just because of the way ratios work. You're taking half the information and playing it at double the speed. Of course it's gonna be faster.
>>
File: 1519855485002.png (117 KB, 625x773)
117 KB
117 KB PNG
>>662938
>Just render at a higher fps, Jesus fucking Christ.
Everything will then just run much faster, jesus fucking christ.

When I render at 240 fps, the explosion is 10 times faster. When I replay the 240fps at 24 fps it just normal speed.
Also You can't even render at more than 120 fps.

Time remapping is a broken piece of shit and does not affect the speed of fire simulation at all, so that's out of the question.

The time scale of the simulation domain is clamped to an arbitrary 20% of realtime. I need 1% or less.

>>662913
>There are countless solutions in Blender.
Name one.
>>
>>662913
Tune down that paranoia, my dude. I was answering to the guy saying that it's retarded to cache a simulation with the goal of retiming it later. It's perfectly okay to do when there's no need for more detail or accuracy; and that, in fact, it's possible to do in other common software. (I don't know about Blender.)
>>
>>662970
You will not get a proper result.
1)You are retarded for trying to do that in Blender.
How about using a tool which actually can do this.
Blender gives you absolutely NO CONTROL over anything fluid related. The pyro sim tools absolutely suck ass and the end result will look ugly and not behave realistically.
Houdini, FumeFX and PhoenixFD give good results, forget the rest.
2) you behave like an idiot. No you can't magically create some huge amount of data from a small sample of data.
Doesn't work like that. There is no way around simulating your Explosion in slow-motion if you want to have it play in slomo.


>>662913
>There are countless solutions in Blender.
There are none for doing good looking Explosions.


pic related, 15 mins dicking around in Houdini.
>>
>>662983
I realized too late that this was just a thinly veiled thread to start a software war.

Theres no way the guy could be stupid enough not to google his question to learn how to do what he needs. He also knows that blender isn't the best for fluid sim but it can still make some visually nice explosions if you take the time to learn how.

I was just hoping that the guy was at least getting paid to do this. People don't pretend to be this retarded for free do they?
>>
>>663006
I know it must be hard for you to accept but:
There are people here who are this retarded and there are people here who recommend tools which they think are better, without any payment in exchange.
>>
>>663006
Stop listening to Rosenthal, Autodesk doesn't give even half a fuck about Blender.
>>
>>662970
I just told you how to fucking do it you absolute braindead retard.
I even told you WHY it plays faster. I didn't say jack shit about time remapping. I said to make it according to the ratio. It's the same as in every fucking program you fucking mong.
The reason your animation is playing 10x faster is because you're playing it 10x faster. Playing it at 24fps isn't going to magically make it slow motion. You stretch the video clip to match the source frame rate. So in this case if you had a 1s 240fps animation and you want it to play at 24fps, you'd play/stretch it 10x slower. Then you'd match the fps because 240fps playing 10x slower is 24fps. You may have to change the source frame rate in the program so it knows that you're doing that, but that's how slow motion works. It doesn't matter what your high frame rate is. It'd work the same going 30 to 15 or anywhere else. Hell you could do 2fps to 1fps. Not to mention if you can't render 240fps how the fuck did you get the 240fps animation rendered in the first place?

Of course all that, is how you'd do it in post.
Want the animation to play 10x longer in Blender? Make the animation 10x fucking longer. Animation ends on frame 24? Move it to end on 240. Just drag the keyframe over in the dopesheet. Hell, you can just grab all the frames and scale it by 10. Don't even have to mess around with frame rates.
Honestly, it's not hard, you're just outright refusing to do it right to spark some shitty software war, or you're retarded.
Just do the fucking math.
Or just interpolate externally.
>>
>>663057
Dude please stop spamming my thread with such useless answers.
Why are you even talking about changing the framerate if you know that it doesn't change anything.
So your answer boils down to make the animation longer. Well duh, I already have a nice looking simulation and I don't want to fuck around with the physics parameters to make it last longer.

>keyframes
Fire+smoke simulations don't have keyframes you assfuck.

Does anyone who actually knows blender have a solution for my problem?
>>
File: 1547507396274.jpg (18 KB, 292x257)
18 KB
18 KB JPG
blender for simulations
>>
>googles the question
>this must be so hard if the guy still hasn't figured out the answer
>finds the solution immediately
>the question has been asked in a similar way before and is answered directly on the blender stackexchange website.

I considered sharing, but at this point I don't think you're actually looking for an answer and just want to shitpost or are simply too retarded to understand.
>>
>>663071
>Does anyone who actually knows blender have a solution for my problem?
I know a bit of Blender. The solution is to export the simulation in OpenVDB and then retime it in Maya/C4D/Houdini, or recreate your sim in any of those programs and time it appropriately.
>>
>>663073
Actually you are the only one here shitposting. Look at this "solution":
https://blender.stackexchange.com/questions/33407/smoke-simulation-how-to-make-a-slow-fog


>There are 2 ways and you may need to use a combination of both depending on how slow you want to slow it down:

>1) Change the Time: Scale with the CUBE DOMAIN domain

>2) Slow down the frame rate. E.G. change 29.97 to 120 fps. Unfortunately it's not a perfect solution- it will change the characteristics of your smoke and you'll basically have to tweak everything from the very beginning.

Notice something?
>>
>>663076
Under [] change the following [] should remain set to [] since the [] frame will of course be frame [] regardless of the speed; if you simulation starts at a later [], the new [] frame can be calculated like this: <new [] frame> = [](<old [] frame>[]).
[]: [] ([])
[]: [] ([])
[] shouldn't need to be changed, as it's a percentage.
[] should be fine as is, but may need some tweaking.

Under [] change [] to [] ([]).
[] shouldn't need to be changed, and you may even be able to [] it, since you now have more real []. [] are only used in calculations, and are never [] or even []. A typical use case is to prevent fast moving [] from passing through a [].

Don't forget to also change the number of [] to [] in the [] tab appropriately ([] in this example), but leave the [] unchanged ([] in this example)

That is a possible answer I got off of google with all the things that might actually help you removed. It wasn't even the most likely answer to your question on that page and it was the first result when googling your question.

Upload your file.
>>
>>663079
That answer (https://blender.stackexchange.com/a/75361/46166) doesn't concern cached simulations.
>>
>>663081
Upload your file
>>
>>663084
Wait shit, nevermind. I forgot i really don't want to download a massive simulation file.
>>
>>663081
Then re-bake it. You can't get something from nothing. Multiple anons have already told you.
You don't have enough data from your baked simulation to get slow motion from it. Either take the time to re-bake it, or use something to fake the data with interpolation. Either way it'll take time.
>>
>>663087
Interpolation would be much faster than re-simulating X times the initial amount of frames. I guess that's what OP is actually after.
>>
>>663079
This is not what I want. It only changes the emission particle system, not the smoke simulation itself, which meams the speed of burning. The process from goung from fire to smoke.

It also involves re-doing a lot of things in the particle system itself, which is shit.

>>663087
Again, re baking does not magically change the simulation to last more frames. This is some shit. This is the whole point of the thread.
>>
>>663096
You can't re simulate the thing, because it always lasts a fixed amount of frames. Re-simulating it will actually cause the sim to run faster, this is the whole problem here!
>>
you can cache out the substeps, try simming at higher fps or use the vel field to advect the fiel- oh you are using blender what are you doing nigga



Delete Post: [File Only] Style:
[Disable Mobile View / Use Desktop Site]

[Enable Mobile View / Use Mobile Site]

All trademarks and copyrights on this page are owned by their respective parties. Images uploaded are the responsibility of the Poster. Comments are owned by the Poster.