There is an undo button — it's on the bottom right of each user message in the chat. That said, sounds like it wasn't obvious enough, so I'll rethink the UX there for sure!
- On Remotion, yeah, not sure it's the right fit, but honestly the sheer capability of models at writing code these days has surprised me. Funnily enough, this is how I used to make small graphics for videos 2-3 years back when I knew nothing about After Effects.
We've been eager to experiment with this for a while, just have to prioritize other user requests for now. Will definitely try a few approaches and see what sticks. (Also noticed they have an experimental client-side rendering version built on mediabunny, haven't tried it yet: https://www.remotion.dev/docs/client-side-rendering/)
- On WebCodecs, there are a fair set of challenges, but we wanted to take the bet. The reason we're browser-based is the same reason I love Figma and Google Docs: no install, no waiting, just open and start. That said, for broader codec support (ProRes, RAW, etc.) we'll rely on server-side transcoding with proxies where needed.
> On Remotion, yeah, not sure it's the right fit, but honestly the sheer capability of models at writing code these days has surprised me.
Just to clarify I still think code-driven graphics is the correct approach, but in my case I opted for a different library with a more powerful imperative API.
> Also noticed they have an experimental client-side rendering version built on mediabunny
Yes, I've tried it out, it was a non-starter for me because it only supports canvas-based components, and Remotion didn't seem to have good support for text on canvas because they rely on HTML for most of that.
> On WebCodecs, there are a fair set of challenges, but we wanted to take the bet
Totally understand the appeal and immediacy of a browser app, I was lured in by that too. For what it's worth I've reported showstopping WebCodecs issues in Chromium and there's basically no indication they'll get fixed on a predictable timeline.
Another issue I ran into that I just remembered is animating text on canvas. It's basically impossible to get pixel-perfect anti-aliased text animation using a canvas. I would have to dig up the exact details but it was something to do with how browsers handle sub-pixel positioning for canvas text, so there was always some jitter when animating. This coupled with the aforementioned WebCodecs issues led me to conclude that professional-quality video rendering is not currently possible in the browser environment. Aliasing, jitter and artifacts are immediately perceptible and are the type of thing that users have zero tolerance for (speaking from experience).
This is not meant to be discouraging in any way, I've just been very deep into this rabbithole and there are some very nasty well-hidden pitfalls.
> Totally understand the appeal and immediacy of a browser app, I was lured in by that too. For what it's worth I've reported showstopping WebCodecs issues in Chromium and there's basically no indication they'll get fixed on a predictable timeline.
Interestingly I have the exact opposite experience, I've reported issues both in the WebCodecs specification and the Chromium implementation, in all cases they were fixed within weeks. Simply though reports on public bug trackers and it wasn't really a major issue in any instance.
> Another issue I ran into that I just remembered is animating text on canvas. It's basically impossible to get pixel-perfect anti-aliased text animation using a canvas. I would have to dig up the exact details but it was something to do with how browsers handle sub-pixel positioning for canvas text, so there was always some jitter when animating. This coupled with the aforementioned WebCodecs issues led me to conclude that professional-quality video rendering is not currently possible in the browser environment. Aliasing, jitter and artifacts are immediately perceptible and are the type of thing that users have zero tolerance for (speaking from experience).
We're doing SOTA quality video rendering with WebCodecs + Chromium with millions of videos produced daily, or near SOTA if you consider subpixel AA a requirement for text. In general for pixel perfection of text, especially across different browsers and operating systems, you can't just use text elements in DOM or in canvas context, instead text needs to be rasterized to vector shapes and rendered as such.
Honestly not sure about potential jittering when animating text, but we've never had any complaints about anything regarding text animations and users are very often comparing our video exports with videos produced in Adobe AE or similar.
> Interestingly I have the exact opposite experience, I've reported issues both in the WebCodecs specification and the Chromium implementation, in all cases they were fixed within weeks. Simply though reports on public bug trackers and it wasn't really a major issue in any instance.
That's fair, they are responsive most of the time. I do have one major rendering issue in particular I've been waiting on with no movement for months, so I might be biased.
> We're doing SOTA quality video rendering with WebCodecs + Chromium with millions of videos produced daily, or near SOTA if you consider subpixel AA a requirement for text. In general for pixel perfection of text, especially across different browsers and operating systems, you can't just use text elements in DOM or in canvas context, instead text needs to be rasterized to vector shapes and rendered as such. Honestly not sure about potential jittering when animating text, but we've never had any complaints about anything regarding text animations and users are very often comparing our video exports with videos produced in Adobe AE or similar.
So you use a library that takes in text and vectorizes it to canvas shapes? That could work in theory, do you have a demo of this?
> So you use a library that takes in text and vectorizes it to canvas shapes? That could work in theory, do you have a demo of this?
Yea, it's harfbuzz compiled to WASM: https://harfbuzz.github.io/harfbuzzjs/
Then all text layout features must be implemented on top of it, like linebreaking, text align, line spacing, kerning, text direction, decoration etc.
We've played around with this and honestly have a lot of respect for what the Remotion team has built. Fun fact, I tinkered with it back in 2021 when they made those GitHub Wrapped videos, it was one of those projects that made me think differently about video on the web :)
Cardboard is a bit different though, aimed at non-developers who want to edit raw footage through natural language without writing any code. Motion graphics is on the roadmap and Remotion would hopefully be a natural fit when we get there.
Cool to see the space evolving from so many directions! :)
I totally resonate with you. Craft takes time, and that's completely valid. We're not focused on filmmakers right now, though we'd love to have them eventually.
That's also why we built a full editor alongside the agentic experience. Use AI where it helps, like finding the right shot or removing silences, and do the rest manually. And if you'd rather finish in your editor of choice, we support XML export for Premiere, DaVinci, etc.
And agreed, there's really no substitute for the kind of intentionality Herzog brings to his work :)
Fair point. What we mean by collaboration is closer to how Figma works. From our user interviews, video creation almost always involves multiple people but in different ways: screenwriters, marketers, designers, directors reviewing the edit and sharing feedback.
The value might not be co-editing the timeline, it's making the feedback / iteration loops faster.
My co-founder and I met in high school, and we wanted the name to carry a sense of craft. Cardboard was always that material in school projects that was firm enough to hold structure but malleable enough to build almost anything out of. That balance of structure and flexibility felt like a good metaphor for what we're building.
Also we just thought it was a cool name and bought a bunch of domains... https://cardboard.mov is one of my favorites :)
Totally fair reaction! Here's our honest thinking behind it.
We deliberately avoided credits/usage-based pricing because as founders using this in our own creative workflow, we hate the cognitive load that comes with it.
If I don't like a voiceover/variation, I should have the freedom to regenerate it until I'm happy without thinking about whether it's "worth" a credit.
That said, we could be wrong! Genuinely curious what you think would feel fair?
Sure! Will share the raw material for all the videos.
For some of the examples we shared though, we've created sample projects right within the product itself. They contain the raw assets and the exact prompts used to create the videos. You can try them out directly at https://demo.usecardboard.com and see the whole process!
We made a deliberate decision to go client-first. Video editing happens entirely in your browser without us uploading your entire footage on our end. No bandwidth costs for you, no storing your raw video on our servers. The File System Access API is what makes that possible, and unfortunately Firefox just doesn't have it yet.
It's not a forever thing though. For cloud-based projects where files live on our end anyway, Firefox support is very much on the roadmap. But for the local-first editing flow, our hands are a bit tied until Mozilla ships it.
Hope that makes sense, and fingers crossed Firefox adds support soon!
I think you should consider putting this information in your site. I always read "we don't support Firefox" as "we are lazy", but that's not always the case.
