Mt St Helens Timelapse

Started by WAS, November 01, 2021, 01:23:03 pm

Previous topic - Next topic

WAS

Just an odd experiment transition between pre-eruption and post.

Never did get the snow the way I wanted via intersections and masks, so I just ended up masking by flows and called it good.

It's interesting to note what features of the mountain remained after it's slip, and also what features were eroded quite a bit in only 41 years.

Hannes

That looks cool! Somehow like a deflating balloon. ;)

Kadri

I was just doing a test with erosion myself. I love these kind of videos very much. Sweet!

WAS

Quote from: Kadri on November 01, 2021, 04:02:38 pmI was just doing a test with erosion myself. I love these kind of videos very much. Sweet!

Thank you. It's cool that so long as you maintain the same seed for erosions, transition between them is very accurate. Could probably totally do a timelapse of a low duration erosion, and a high duration erosion to show time.

Quote from: Hannes on November 01, 2021, 02:35:35 pmThat looks cool! Somehow like a deflating balloon. ;)

It totally is. I actually expected more discrepancies between erosions while it morphed, but should have known better.

Kadri

November 01, 2021, 04:19:57 pm #4 Last Edit: November 01, 2021, 04:44:39 pm by Kadri
Quote from: WAS on November 01, 2021, 04:07:44 pm...It's cool that so long as you maintain the same seed for erosions, transition between them is very accurate. Could probably totally do a timelapse of a low duration erosion, and a high duration erosion to show time.

Interesting that you say that. My test is about accurate transition. Not sure if we mean the same thing but in my past tests the erosion parts were moving-dancing kinda. But if i understand you correctly Jordan you made this test with merging?
My old tests were just with animating the erosion node itself. Those didn't look good to me.

WAS

Yeah, I used two erosion shaders on each height map, same seeds, and settings, little PF detail over that fed into surface layers of each erosion chain, then fed into a merge and animated the mix.

It's probably easier to defeat dancing details because you just have two points to transition from. There is a soft change in preserved details in post-eruption but I figured that was actually just further erosion of the mountain itself in the last 41 years. It has been rumbling this whole time, plus massive annual snow cover and melts.

WAS

PS I think the erosion heightmap is multi-threaded? It seems to render my computer unable to perform any other even moderate tasks like a VLC player, playing a Mp3 song. It gets all choppy. So there may be discrepancies between tiles that stitch differently which is why for stuff like baking textures, heightmaps, etc, people usually do one bucket/thread the size of the output resolution.

PPS I think I've even seen this effect when playing with non-normalized erosion maps, and multiplying them extremely, with huge patch sizes, I think I was getting this weird grid on the map.

Kadri


Yes all might have an effect. Just would love to know which is the most important one.
Feels to me mostly more like trial and error while animating.
In the test the frame between 42 and 43 jumped in height even.
Maybe my error while animating...still when you wait for hours it sucks.

The merging method feels to me not as i would like mostly.
It looks more like the mountain loose heights in general only
other then things happening really all around. If this makes sense what i say.

WAS

Well it is loosing a lot of height, it is mixing to post-eruption. It is the same mountain and resolution LIDARs at same position, so not much will really happen in a 41 year span.

When doing something like thousands of years you're going to need multiple merges and phases.

Merge one mixes to A 100%, merge2 then mixes to A 100%, then merge3 mixes to A 100%, etc, setting the keys accordingly. Each erosion map you mix to should have gradual changes. More down cutting for example, with progessively more duration (though maybe only stepping by 0.05 each time).

Using the erosion shader itself seems like it maybe problematic inherently because of it being raster based. But maybe if you did less frames, and interpolated between them in post you'd get less dancing.

Maybe the plugin also responds to lowering the CPU count within TG? Maybe 1 CPU would provide a more stable map. But naturally be much slower.

PS in my example, you can see though, even with the features being mainly the same on the sides the mountain didn't slip, the tinest change in shape causes flows to totally flip sides of displacement, taking on whole new courses. So if the terrain changes any over time, this will dramatically impact the erosion result.

DocCharly65

I think meanwhile I understand 5% of how you did it  ::) ;D

But I 100% agree with Hannes, the deflating balloon and that I like it.

A little suggestion just for fun: Would it be interesting to animate a reverse scene - something like the inflating Yosemite National Park? I think I remember to have heard about such an effect of the Yellowstone Caldera.

Dune

With erosion you mean Daniil's plugin? Or the heightfield erosion shader? If the first, I've noticed that even with the same seed and settings, outcome may differ if you open the file again. That may cause dancing too, slight differences between sessions of erosion.

WAS

November 02, 2021, 01:19:55 pm #11 Last Edit: November 02, 2021, 01:32:41 pm by WAS
**Wrong Topic**

Dune

Kind of, yes. Actually replying to Kadri's comment.

WAS

Oops, sorry. I posted in the wrong topic I meant.