Algorand Ident

[fixeighted]

https://www.behance.net/gallery/1586...Algorand-Ident

A fun little project finally completed and included initially at their daily events in Dubai over the last week.
We produced this as the first completed draft, expecting feedback and potential changes, however it was
approved on first viewing, which rarely occurs, so that's nice

There are not too many elements in this, though they did produce some significant challenges as I iterated, mainly with the sheer volume of
instances and managing their art-directed growth, as well as the landscape, which was the first time I attempted such a task, especially on that scale.
Initially I was testing Tyflow for the foliage but that proved impossible, oddly, though it looked great.
I used Forest ultimately, which was in comparison was very simple and controllable with ease.

Great thanks to damaggio for initial help and inspiration stemming from his Lidar-enabled Monument Valley project post.
In the event, the landscape is fairly obscured but its detail and scale was necessary in any case I think.

As usual, we used Vantage so that my collaborator (the cameraman/editor) could work up the camera moves separately, as he has no knowledge of Max.
Vantage is a great tool for him, given its relative simplicity and he is in his virtual element using it to do these things.
It was a little painful trying to get the mesh proxy exports working for him, but it all worked out in the end.

Anyway, feel free to nitpick and critique/suggest. Although it's all done it'd be good to hear thoughts

[damaggio]

It came out excellent! the entire animation was really well done , congratulations on the result and all the hard work you put into it fixeighted , new tricks learned too.
The sunset opening looked gorgeous and that foliage animation must have been quite complicated to do.
It's nice to have a colleague to help in such a complex task.Team work.
Kudos to you both.

[fixeighted]

Thanks. It's a bit of a shame the landscape wasn't more fully seen, after the work it took to get there, but it was a particularly fun part of the process anyway, plus yeah,
those opening shots came out really well, so I'm glad I pushed for that location.

The foliage was eventually really quite simple, luckily, just driven by a quite random image sequence.

Now I just have get them to pay up for it

[Joelaff]

This was the project we were discussing Tyflow for on the forum, yes? Nicely done. Yeah, that landscape is pretty slick" Would have liked to have seen more of it.

[fixeighted]

Yes indeed, Tyflow was my first call for this but it proved useless in this case.
I actually got so far down that particular rabbit hole that I bought loads more ram to be able to
do it. There was eventually so much grass and flower geometry that it was near impossible to iterate
and the control I thought it would offer came at the price of way too much complexity.
The Forest option was just so simple that I wish I'd explored that first

I spent a load of time on the landscape, though much of that was figuring out the exact workflow using World Machine
and quite a lot of time looking at other options like World Creator etc., so actually from scratch I could now do it in a fraction of
the time. I'll admit that it is not as well realised as damaggio 's version, though that is mostly as I realised it would mostly
only be seen as is, so any real appreciable detail would be lost.

The fortunate thing about all of it is that just by chance much of the work had been done for personal r&d before this job even arrived, so
I actually made a healthy profit for once

[OPEN_RANGE]

Not sure why but when I click on the link to behance, nothing loads. maybe temporarily down?

[super gnu]

ha,, ive heard bits and bobs about various aspects of this job for ages (3d landscape, planet atmospheres, tyflow planting) , never actually realised all of it was a single job! very nice work. .. the rendertimes on the refractive section must have been chunky.

[fixeighted]

Thanks super gnu
Hehe yeah that was the great thing about this one, all of those tests
were done 'before' this job even appeared, so most of the hard work had been done already.

I wish I could have leveraged the gpu for these frames but sadly again that idea was scuppered by
one limitation or another, making it just a headache.
CPU times were actually pretty good, at a max of 15 mins I think, with denoiser on some but not all.
The earth is actually VC Orb, simply because of the relative fiddliness with getting the zoom speed correctly
matched along with directorial post control of it, so a hack in a way but it appears to work

The best bit is that I didn't need to do any fixes, no post cc (he did the grading) or any other stuff...just model it/shade it/hit render/collect reward

OPEN_RANGE the link definitely works so yes maybe it was glitching...