RealFlow Official Forum

Just finished the first realflow project for this year.
I'm quite happy about it for the short production time (total of ~3 weeks, incl concept,re-modelling cad-geo, render and post)

But was wondering what other realflow people think (be honest)

https://vimeo.com/75213955

Hi Marnix,

Thumbs up.
I'd say that overall looks pretty cool imho.
Rendering is the low point for me, but I am not sure the exact look you or the client were after.
Simulation perhaps could benefit from a few more particles so the meshes could have a bit more detail. Having said that, it is quite a smooth movement, so it holds up pretty well as it is.

Nice job from my point of view.

Have you ever considered trying Maxwell Render? I think this will definitely take it to a different level in terms of finishing.

g-)

thanks, good to know the liquid is sorta 'approved'

The main challenge for me was the short project duration.
The 2 big sims were around 1.5/2 million particles, but I think the tension/viscosity could have been slightly lower to get that bit more detail. In the end all was simmed in a week and no time for do-overs.

At this moment the frame render times were between 10 min - 1.5 hour, No shadows, no gi, nothing basically
The full hd version took ~5 days on my "200" Ghz-farm (7 x i7 ~3Ghz)
I would have liked to have it a bit more crispy myself, the client is quite traditional with their look/feel so it fits for them.

As for the rendering, how quick is maxwell these days?
Will test it when I have a time to spare, last time I've played with it there were no plugins yet for 3rd party integration (in my case, 3ds max)

time vs. quality.. always a tricky one.. especially if the budget doesn't allow the use of external farms

Hi,

Nice job all around. The only thing that sticks out is if it's supposed to be oil, I'd bump up viscosity, sticky, pariticle friction, and roughness to make the liquid thicker and stick to the sides of the container.

Oldcode wrote:Hi,

Nice job all around. The only thing that sticks out is if it's supposed to be oil, I'd bump up viscosity, sticky, pariticle friction, and roughness to make the liquid thicker and stick to the sides of the container.

tnx!

I thought about that in the beginning, but then figured it would be quicker to do something with the wetmaps.. but in the end that didn't come out as I thought and then the time was gone

marnix@rckmn.nl wrote: At this moment the frame render times were between 10 min - 1.5 hour, No shadows, no gi, nothing basically
The full hd version took ~5 days on my "200" Ghz-farm (7 x i7 ~3Ghz)
I would have liked to have it a bit more crispy myself, the client is quite traditional with their look/feel so it fits for them.

As for the rendering, how quick is maxwell these days?
Will test it when I have a time to spare, last time I've played with it there were no plugins yet for 3rd party integration (in my case, 3ds max)

time vs. quality.. always a tricky one.. especially if the budget doesn't allow the use of external farms

I'd say you give it a go and find out yourself. My bet is you will be pleasantly surprised. Maxwell do have a 3DSMax plugin that runs also nicely.

We can put you in touch with their team for an eval if you are up for it.

Best,

g-)

I think it looks great, well done.

As far as maxwell goes, I've used it for a long time and render times are well within the timings you've already given - it's well worth a go, it can produce some stunning results with very little effort.

mmh..have been looking a bit at maxwell things online lately.
All examples look awesome, have seen some interface shots come by here and that looks pretty daunting for 'just' a renderer

I've bought a bunch of http://www.Lifx.co led lights and have an artist friend that wants to help me build custom lights to fit those in.
She's good in processing metals/plastics in any shape desirable and I'm looking to build a virtual setup where I can test various shapes and material and see how to amplify the light (mirroring surfaces) and basically test the lights before building them.

Now i'm wondering if Maxwell would be suited to simulate this properly and would be able to give me a good sense on how different translucent materials would react to the lights reflected of the mirroring surfaces. Is this something that can be achieved relatively easy with maxwell or will it have an intense learning curve?