Womp 3D – The New Way to 3D
womp.com
womp.com
Anyways, I actually tried it out and... I don't really get it. I would find it frustrating to use when blender is so much snappier and more powerful. I do like the process of using metaballs for design and a lot of whatever the tech is. But I am also surprised at how much effort went into this when this seems really really niche. Someone else linked a video [2] where there's a project with so many layers... bro just use an actual 3d app at that point.
Also that landing page is horrible.
[1] https://www.producthunt.com/posts/womp-2?utm_source=badge-fe...
Right now Blender work still involves a lot of tedium, mostly related to topology. A lot of upcoming 3D ML applications also work considerably better when using SDF instead of mesh representations. I wouldn't be surprised to see this form of 3D modeling take off to a significant degree because of those two factors.
There was the original metaballs. But more recently there's also been sdf addons using geometry nodes [1] that mimic the same workflow - with my guess being that it uses voxels to generate the final polygon mesh that blender needs since it's not a fully sdf editor. Although, while I was googling this, I did find someone that managed to do it by using pure shaders [2] which is pretty cool.
Also, thanks for actually explaining that. I've seen a few examples of this kind of "clay like" sculpting approach that tries to make it easier for artists. Adobe's Modeler uses sdfs for example.
[1] https://blenderartists.org/t/geometry-nodes-in-3-3-sdf-prese...
Interestingly most folks think of 3d modeling as quad modeling/subdivision surfaces, but Toy Story 1 was done completely with NURBs (also supported by blender)
Ideally - the end result of an SDF pipeline is pixels. Going back to polygons throws away much of the advantage of SDFs. Raymarching is costly and rarely used in realtime engines but Blender isn't realtime so rendering SDFs directly would probably be viable.
The viewport still needs to update in real time for it to be a viable modeling workflow, though, right?
I don't think that's true. Most games are optimized to limit the number of draw calls and textures. The viewport of a 3D software package has no such limits. As a result, my viewport rarely runs as well as the games I play.
In any case, that's besides the original point, which was that it wouldn't need to run in realtime. My argument was that it would.
It's not a completely impossible goal. Look into what some artists are making with Dreams on the PS4. It uses raymarched SDFs for modeling.
Small correction - Dreams isn't purely Raymarched SDFs. From recollection they ended up with a fairly hybrid approach.
Why would anyone use Canva when there are plenty of other more powerful options?
I believe the answer is that it is simple and not scary for people to learn, and that there is a large number of people who could use this and create something that looks pretty good but that would be intimdated with Blender and have no clue where to start. Blender isn't really designed for a casual user to be able to pick it up and understand it in 10 minutes.
The fact one has to sign up to see the product in action says to me they either lack confidence in their product, or are desperate to monetise everyone who wants to check it out.
I've been waiting for some product to invent a painless way to sketch 3d objects with a touch interface for years.
"But if they had confidence in their product, they would just make it open and people would try it out. And if they like it, they might buy it later!" - Right? Wrong!
People have increasingly short memories. Even if they like it now, by the time it makes it out of alpha, they'll long be chasing the next, shiny thing. So you have to keep the link alive to follow up on it. That's just good business.
So how about you have some sympathy for the talented people who put this together and give them the chance to follow up in a few months on the off chance that this is something you might be willing to use properly?
I do get the desire for them to want an audience to give updates to and try to have a consistent user base. The thing that sucks is that not only do they want to email me, so does everyone else. I've never been very protective of my inbox but the volume has gotten crazy recently so that's had to change. If they let me in and then asked for my email to continue after a couple minutes I'd be more inclined to provide it.
Cool. You’re not in the target, you bounce. You save time, they save time and bandwidth and get a higher quality user funnel. Everyone is happy.
Problem solved
Not free of friction, that's for sure.
Do you want every service out there, free or not free, to ask you to sign up to access everything?
The thing that really bothers me is the normalisation of 'hand over your credentials' which inevitably maps to data leaks and identity theft. For what, trying out a product that you may or may not use?
If this free 3D product only required a DNA sample, the group who considers it free might change. Or if it required your signature. Or your corporate username.
It clearly is not free. It has a price. The question is how much different people value that price.
It should be:
- I got you to click my url, here's some screenshots and a description.
- Looks interesting? Try it yourself.
- Like it? Sign up for an account to save your work, or maybe submit an email for announcements.
It's a reasonable response to the environment we live in, not necessarily a criticism of this specific app. Unfortunately it seems like the OP has wanted something like this for a long time, but their knee-jerk reaction to its outward appearance will prevent them from even trying it.
Like how an allergic reaction is our body's immune system over-reacting to something that may actually be benign.
But I'm on the same "mildly annoyed" page, because with the large majority of "it looks cool" new products or services I try out, I know within 5 minutes that it doesn't actually meet my needs and I know I'm _not_ interested in it, but I still have to trust that this organization both today and in perpetuity (or as long as I keep that address) will be well-behaved and technically competent in using my information.
Recently on vacation I bought tickets to a museum in a country I was visiting, and in the process shared my email in order to receive a QR code for entry. They have a legitimate use case which involves sending something to me; fine. To my knowledge the ticket sales service involved does not operate in my region, and I would really prefer never to hear from them again after this one useful interaction. Only a couple of weeks later, in the space of 30 minutes I receive _dozens_ of marketing messages from them, each in a different language I do not speak, all promoting the same 10% off sale _for something I cannot make use of_. I assume some bug didn't associate a specific language with my record so they decided to send me messages in all of them. Their "unsubscribe" link was broken, perhaps for related lack-of-competence reasons. There's literally no way for me to tell them I am not in a position to buy anything further from them. But we've totally normalized online interactions that require (potential) customers to trust that an unfamiliar organization with be both well-intentioned and skillful in their handling of your contact info.
Imagine if receiving a free sample at a grocery store of a new product that might be appealing to you but might be entirely not to your taste required you to first give the company your contact info. Or if trying on a garment in a department store required the same trust. In a non-online context, this behavior would be creepy and intrusive. And you know that you might realize in seconds that you hate the mouthfeel of their product, or that the cut of those pants just doesn't work for your body, and no you won't be buying their product -- but you can't claw back your email.
I think in online contexts we should normalize at least one of two other patterns:
- Let people try it for k minutes/actions/new objects without a sign-up, after which the UI is blocked by a modal asking them to create an account. The customer gets to establish whether they actually want to use the product, and the company still gets info about people who are actually interested.
- Rather than relying on _emails_ specifically, account setup could rely on _revocable_ access to any other service that allows sending/receiving messages.
Do you know https://stephaneginier.com/sculptgl/? It is open source [1] so you could adjust it to your needs.
> For a B2B or B2B2C product
I don't need a company email to extract the information I care about.
Never did research for competition when replying to RFPs?
How could you even consider such a thing? /s
I highly recommend Blender Guru on YouTube as one of several great places to start for anyone interested in learning Blender. He's got a couple beginner tutorials there that'll get one up and rolling in no time at all. It's well worth the effort to learn for those who have an interest in 3D graphics or game development. Having said all that, I'm always in favor of more new tools / toys appearing in this space, and I thank you for adding to the mix.
Someone saying "use" instead of "learn", to people who've never used blender, is fine. IF you've never used blender, everything in blender is hard (except maybe deleting that cube) because it's good software, with a rock solid elaborate UI that follows (and shaped) industry-conventions. This tool doesn't, it's much easier to use if you come in knowing nothing.
As-is, I can't appraise Womp3D as a Blender alternative since it requires me to create a user account. If you ever drop this requirement (or let people self-host the app) let me know and I'd be happy to compare the two and potentially even recommend it as a Blender alternative.
Overall I don't see it replacing any serious tool at the moment, its quite slow in modeling (because of the rendering) and precision.
What it can be useful for is if someone doesn't have the necessary tools/skills and they just want to make a basic and quick 3d shape like a smiley face, have it bounce/animate and export that to a little gif or whatever.
Signed Distance Function modeling has different strengths to other approaches. It's one more tool in your toolbox. One could just as easily say "I don't see [insert other paradigm here] replacing implicit SDFs".
In terms of "serious" usage Inigo Quilez did a ton of incredible work on Pixar's Brave using SDF techniques.
https://www.cgw.com/Publications/CGW/2012/Volume-35-Issue-4-...
But as a webapp, it is cool looking. I'm curious how he managed to get it to work in a browser.
A bunch of JS on a CDN doesn't cost that much to serve, if it's running on a remote server Stadia style than yes, but then what is the goal here? User/Customer acquisition always costs a lot of money...
At the very least, let me test the editor quickly, and prompt me to register upon saving a project.
edit: I always thought it was stupid that Stadia didn't provide tools like Blender, Gimp and co, it would have been a huge avenue for user growth. Then strike deals with Maxon for instance and run Cinema 4D in the cloud, that's a missed opportunity.
Colab allows that, you can run Blender in the cloud and basically use the service as a rendering farm, and it's damn fast. No GUI obviously...
Even on Android, it took them 10 years to provide what is basic tooling for game development on other platforms.
Games succeed on Android because of the game development community standing in for the lack of motivation and tooling support from Android team, not thanks to them.
I really don't like the idea and the practice of helping a Google monopoly, and I am guilty of it at times. But on the other hand, using such features is a way to encourage Apple to upgrade their browser... which they have little incentive because they want apps.
- including proprietary code (open source browsers are reluctant to do that (widevine is usually the only exception and it's opt in on Firefox and some Chromium forks)) (Google does this for other parts of Chrome at least, idk if it's used for WebSpeech or just live captions)
- interfacing with OS-provided proprietary code (doesn't exist on all platforms)
- streaming everything to a cloud server (likely prohibitively expensive unless you're Google) (Google does/did this too)
Safari not allowing requests to localhost from HTTPS is weird and I'm not sure why it's like that but it works fine in Firefox. Same for OffscreenCanvas which is the only canvas feature I've missed, Firefox finally added it 2 months ago.
Anyway, just don't block specific browsers, feature detect and if possible use a polyfill or ponyfill.
How the world has changed.
That ain't their arrogance. Their arrogance is declaring all browsers other than the single one they tested to be "unsupported" and not even worth attempting to use with the app. It's one thing to warn against untested browsers; it's another thing to outright prohibit them and kick you back out to the homepage (after collecting my email address with which to spam me, no less).
If it's more a matter of some specific thing(s) that Chrome has implemented and other browsers haven't, then it'd be nice to be upfront about what that thing is so that those of us on non-Chrome browsers can bug e.g. Mozilla and Apple about it - and it would further be nice to actually test against the existence of that thing rather than gatekeeping the user agent string like it's 1999.
May be we should just got back to native apps? Hell, native apps that do not connect to the internet. That'd be glorious actually. Problem with browser based apps is that you have zero control over versioning. Server serves, you get the fix. No option to request a particular version of the app. Don't like the new version? Tough luck. Server issues? Go outside.
You know there is something to be said about code/binaries residing on your computer. Call me old fashioned.
What we need is a new app runtime with a unified permission-based API across platforms. All the OS-agnostic parts of that runtime would be a huge pain to develop, so all platforms would probably settle on a shared open source implementation and only implement OS-specific "adapters" for the native stuff.
What I've done there, is reinvented Electron...fuck....
I assume it uses SDFs, though there isn't really any technical explanation for how it works.
It seems to be 3D object/character modeling? And maybe a little bit of rigging?
Because there are plenty of other things people want to do ‘in 3D’ that are excluded from that very narrow field - like scene modeling, CAD, motion graphics, Sfx, game development, visualization, capture, etc…
That being said, I believe in conservation of complexity in these highly technical tasks. Complexity cannot be removed, only moved around. 3DCG is an incredibly deep field that necessitates 1000 button apps for good reason.
Not to say you can't get cool stuff done in a toy metaball modeller, but it's also not an alternative to blender in the exact same way that you can make glorious paintings in mspaint and it's still not a substitute for Photoshop.
Womp looks like https://www.vectary.com with few additional features.
Anyways switching to chrome and disabling all security it worked.
Why do models take many seconds to load?
Interesting rendering algorithm: You see it converge from low res to high res within a few seconds after rotating:
https://alpha.womp3d.com/preview/45050
Pretty good end results I guess? No expert on ray tracing or whatever they use.
Signed Distance Functions
Nice explanation (with some code) : https://www.alanzucconi.com/2016/07/01/signed-distance-funct...
Before clicking I though it would be about these new AIs that convert text to 3D models.
Example : https://dreamfusion3d.github.io/
At least support something like "Sign in with Apple" so I can forget about you later.