Hacker News new | past | comments | ask | show | jobs | submit login
Atomontage Engine (atomontage.com)
73 points by Stahll on July 9, 2016 | hide | past | favorite | 21 comments



I've been following this one-man project for several years now, and its amazing to see how far it has come. The engine can render giga-voxel scenes with one voxel per pixel detail, simulate voxel-voxel physics, and can even render dynamically deformed voxels. Check out his dev blog for more details!


This is a nice illustration of "you can simulate anything as long as it's made of Jello." It's not hard to do all-soft-body animation. (Anybody remember Mathengine?)[1] But if you crank up the stiffness, the integration tends to go unstable.

[1] https://en.wikipedia.org/wiki/Vortex_(software)


I thought this was particularly impressive: https://youtu.be/J62z_7JaYMw?t=64


Does anyone know if the voxels are axis-aligned? It looks like they're actually rotated, which should make collision detection expensive...


Is this actually performing independent soft body deformation, or just applying a demo-scene sine wave shader in the renderer?


Did you watch the whole thing? The car is bobbing as the surface is moving.


Tires of car are also deforming.


Any update on this project since mid 2015?


I see this objection clearly stated a lot or inferred a lot on HN. But I've never heard the reason behind this apparently quite popular objection. May I ask why? It (the technology) does what you need, or it doesn't. Is the rate and frequency of github commits actually mean anything in particular?

I understand that if I'm choosing some framework and can see the need to hire 20 people in the future to work on it, it might be nice to pick one popular enough that I might have some non-zero chance of hiring someone already familiar with it, but otherwise -- I've never really understood why someone would or would not choose a technology that may or may not fit their needs based on how frenetic the developer's commit history looks.


Not an objection at all sorry for not clarifying. I'm totally fine with older things and reposts, I was hoping to find out if there were updates that were, for example, not on this page, like a blog or Twitter.

I've enjoyed revisiting this project every time it's been posted to HN. Asking for updates shouldn't imply I'm somehow being contrarian but I know why you'd think so and wish HN would be more supportive in general. I'll do my best to be more clear in the future :)


Yeah - everything on FB and Twitter is 2016 mid-May or so. A year ahead of the on-site blog, but not a heck of a lot of detail about what's going on.

I wasn't trying to call you out for what I perceived to be an objection, but I see people say things like: "developer hasn't posted a commit for 9 whole months! no thank you! I'm out!". Which isn't very comment worthy; ranks right up there with "site doesn't display on my iphone".

You weren't that direct (which should have been a clue that you were actually, you know, asking a real question), but you didn't seem very hostile, so I thought I would ask someone who was, you know, not hostile! :-) Because I seriously don't understand the objection when I see it and I've never taken time to ask.


"I wasn't trying to call you out for what I perceived to be an objection"

Except that you literally did...


It doesn't matter now, the question has been answered.


I'm always happy to have these kinds of exchanges on HN :)

To your earlier point where you mentioned github, I do think there is some reasonable doubt that people have for unmaintained repos.

A maintained repo implies (pretty strongly) an active community. This is good for new and veteran developers. New developers are helped by good community and documentation, and veteran devs need the community when they hit a wall – using some library/framework to its limit, to the point that they need to make changes, or they think they need to make changes. Having an active community makes that process easier.


That makes sense.

I've never had much of an "active" community of anything -- just a lot of very demanding ticket (issue) submitters. Well, actually, maybe only 10% of the ticket submitters? They still irked me beyond reason. Maybe I need to grow a skin.

Back in the bad old days of FTP, you could put up a gzipped tar file, and it worked for people, or it didn't. Maybe there was a mailing list, or even something on usenet. It seemed a lot more cordial and give-and-take, and less: "When are you going to support XYZ? I've been waiting for 3 whole weeks!"

I just gave up on the new world and pulled the repositories. I guess you could say that I got off the damn kids' lawn! :-)

Thanks!


Yep, I am similar to you I think – I've avoided doing much because supporting a community is hard and unless you have a really solid vision about what you will or will not do, incoming requests often seem overwhelming. I do think having a thick skin helps, as well as having distinct contributor guidelines.

Trying to please everyone will always wear you down :)


Off topic at this point, but I can at least try and answer your question from my perspective. Let's say you're building some software and you're looking for libraries to ease the burden of how much you have to write yourself.

In a vacuum, it would be fine to just look at features and call it a day. But software is aiming at a moving target. Whether you're building a web app and the browsers update out from under you, or a desktop program that needs to keep working after an OS update, or whatever your outside dependencies may be... unless you control literally everything (down to the hardware) there is going to be some level of bit rot, and you're going to have to invest in keeping your software up-to-date.

So, if $LIBRARY looks okay, but hasn't had obvious signs of maintenance, it could mean one of two things: either it doesn't need any, or the maintainer isn't active. If it's the latter, then when maintenance time comes, it'll be you that has to do the maintainer's work. Taking on that ownership might be okay, but it's something you need to be aware of when deciding to use that code for something nontrivial.

I haven't left any such comments on HN, and sometimes they might be phrased uncharitably, but I find them helpful to take into consideration.


Thank you for answering. My clumsy intro up-thread appears to have gotten in the way of getting this answer.

And, that makes perfect sense. I could imagine that for the shifting quicksands underlying front-end web development that this could be (almost certainly would be!) a real issue.

In most of the projects I end up being engaged on, for various reasons, we have to take control of the maintenance path ourselves and we get to treat the web interface as slim 'shim-to-the-world' slapped onto the core of our 'real product'. I have had blinders on, I guess.

Thanks!


Is there a github? I thought this was closed source, and the lack of updates made me think it might never get released even as a commercial engine or just a game. I'd love to just try these demos.


I've seen some semi-recent activity on both Twitter and Facebook, but haven't seen a github repo linked anywhere.

The reference to github was a (probably faulty) mental consolidation of people mentioning their concerns about a project that appears to not have many recent commits or releases. And then I foolishly mis-attributed the same concern to the poster top-of-thread.


The cross section rendering would be useful for oil exploration visualizers.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: