Absolutely superb. The patience involved, that's what I can't get over about stop-motion animation – the ability to see the big picture while you're inching along frame by frame. Incredible job, it really is.

May as well put this here because it popped into my mind while watching your piece.

You know, if we do move as a society from the internal combustion engine to the electric drive train something of great beauty, mechanical logic, and engineering prowess will have been lost. Sort of humbling to think that these may only exist in the museums and enthusiast workshops of tomorrow. Kind of like steam engines today, see here: http://www.isvrally.com/ for a local example – “Innishannon Steam & Vintage Rally”

Thanks! The hardest thing is to try to remove multiple parts staggered over shots. Because the smallest distraction and you've forgotten which bolt you already turned and which needs to be turned more.

Very interesting insight. Thanks.

Beautiful! Are you planning on rebuilding it? Another shoot? Is there a reason you don't mark the parts original location or is that simply because you intend to replace them?

The honest truth: This whole car is headed for the scrapyard at the end because it's not worth the extra time ($$$ per hour) while we film to mark parts, directions, alignments etc. I mark things when there's time. We're also modeling the whole thing in 3D and I'm sure the guys doing that will be mixing the bolts and fixings up. I'm kind of interested in how well it runs after this reckless teardown and rebuild.

Ah, there is a commercial angle to this. Ok, that explains it. I've done a couple of rebuilds and marking parts is something I do religiously otherwise the rebuild won't live long.

You could probably get some $$$ back if you put the car up for sale (as a "for parts or not working/not roadworthy", to avoid any possible liability issues) --- I don't know about the market for Miatas, but I know that there are automotive collectors out there who are interested in cars that have "interesting" histories.

>We're also modeling the whole thing in 3D ...

Can you elaborate on the process for this? I'm curious what software you're using, how many modelers you have, and if you're using 3D scanning equipment.

Modeling everything by hand has to be unimaginably laborious.

Incredible work by the way, you've already sold me as a customer. :)

We're making it all by hand because it needs to be relatively low poly for the app and website :) I'll write about the process when we've figured it out completely. I'd say we are about 70% done on the modeling. Rigging and animating is a whole other animal.

This was the state of the engine a few days ago - all internals are done. https://twitter.com/howacarworks/status/895417122385481728

Very nice.

In that case, I suppose the most utility 3D scans would have is as references to enable a faster low-poly modeling workflow.

Then again, scanning would likely incur a non-trivial time cost—one that can't outweigh the very benefits the process affords.

We use photogrammetry for the block, head, dashboard and transmission. It's invaluable for these organic parts. Then retopologized.

BLENDER IS AWESOME.

Blender is amazing. Check out the new Eevee realtime renderer[1] in 2.8 as well! It's astonishing :)

[1] https://www.youtube.com/watch?v=eAVjwXEjDdo

I'd be as much interested in a 'making of' as I am in the final product you are making.

The amount of things to do on this project is incredible. When I embarked I thought we'd shoot 'making of' stuff all the way, but the reality is that even filming the film is hard enough work - to film the filming of the film would just be too much. I'd like to go back at the end and cover how we do some things - we really learned a lot. The other reason it's hard to do a 'making of' is that I actually don't know how to do this stuff, so we'd waste a lot of time explaining how to do it, only to discover a better way two weeks later.

> The other reason it's hard to do a 'making of' is that I actually don't know how to do this stuff, so we'd waste a lot of time explaining how to do it, only to discover a better way two weeks later.

That's exactly why a making of would be interesting. Here's an idea: shoot all that stuff anyway, with a single fisheye camera so that if you ever decide to do a making of you at least have some raw material to work with. Documenting this work that you are doing would have lasting value.

You are spending so much time and money tearing that poor Miata apart.

What inspired this?

It's a Miata, so it was probably inspired by hate.

I kid, Miata fans.

Honest question, if you just ran the video in reverse, could anyone tell that it was not an actual rebuild? It is stop motion after all, in a sense there is no motion but just captured moments in time, and those moments can be reversed.

Yes, of course. The parts are all still dirty, the seals still broken. After a rebuild an engine looks like new.

Alex, beautiful work. What kind equipment does one need to do stop-motion pictures? How did you learn about the process?

Just a DSLR really - the camera guy for this used a Panasonic GH5. It's more photography than video. But a couple of lights that you can position for a subject like this make a huge difference.

I love the style, which to me was reminiscent of a demoscene production. It put me in mind of this one from a few years back: http://www.pouet.net/prod.php?which=50105

Good to see I'm not the only one who thought it was very demoscene-ish. In fact if you were to say the first 30 seconds or so were CGI, I'd probably be convinced.

Really nice work!

How did you do the shots where the cassette tape and the engine parts are moving around in the air?

String. Coat hangers. Bits of wire. Sticks as fishing rods. We tried various ways and in the end I mainly used some conductive cotton thread that I bought from Adafruit ages ago. It was strong enough to take a lot of weight, but nice and thin too.

For the heavy stuff I just had to hold it and then photoshop my legs and arms out.

That is really great - I never would have guessed how you did it.

> photoshop my legs and arms out

This reminds me of the first time I saw Forrest Gump. For some reason I didn't know who Gary Sinise was, and all through the movie I thought the actor playing Lt. Dan actually was a double amputee.

And in the flashback scene that showed him walking, I marveled at how they were able to fit him with artificial legs that let him walk so naturally!

http://articles.latimes.com/1994-07-10/entertainment/ca-1404...

> And in the flashback scene that showed him walking, I marveled at how they were able to fit him with artificial legs that let him walk so naturally!

They filmed that scene first then chopped off his legs and filmed the bits that came "before".

You owns the course or you just did the stop-motion?

I own the course. As much as you can own a video these days!

Congratulations, very well done.

judging by his username and the intro the videocourse https://www.howacarworks.com/video-course I'd say the former

You could've cut filming time in half or doubled it if you had chosen a different engine.

That would have required picking a different car to start with.

Ford v10 would have about the same amount of stuff per cylinder but 2.5x as many cylinders.

SOHC I4 from a "manual everything" economy car with 1st gen EFI would probably be about the lowest part count without getting into flathead territory.

2 stroke Wartburg or Trabant.