That's great to hear. I am going to put up some communication channel on the site soon so I can talk back and forth to users. Let me know if you have any recommendations on what that could look like. I have tried Discord, but I think I want something more one on one.
How long can the recordings be? Recently needed to do some 1-2hr long recordings. Look works pretty well for that, QuickTime + OBS could, but generates massive files, which are hard to share
Currently, videos are recorded and uploaded directly to Amazon S3, and I haven't set an explicit limit on video length. I've personally tested uploads around 15–20 minutes without issues, and a couple of users have successfully recorded 30-minute videos, though I haven't fully rendered or extensively tested videos at that length yet.
The app uses a custom renderer inspired by ReMotion, so while the app can output MP4s, I also have a player that is dynamic HTML/CSS animation powered by JS, hosted either on my site. I could make it a player that is hosted from anywhere, and I think I will do this for a free tier.
Theoretically, this setup can handle videos of arbitrary length, but practically speaking, I've only thoroughly tested around 10–15 minutes. I'd recommend caution with significantly longer recordings (like an hour) for now, as you might encounter rendering issues.
Fyi, you can easily encode your obs recording via good old Handbreak or similar (editors usually also let you so that, but if you are not planning to edit.. )
Yes, I currently save them to Amazon S3, but they're deleted after a few days. It's still just a hobby project, so I haven't set up a privacy policy yet. You're right though, I definitely need to clarify this explicitly. Thanks for pointing it out!
I've built a 100% web-based screen recording tool aimed at product marketing managers, developers, or anyone wanting to showcase their work along with a talking head. Inspired by ScreenStudio, it allows you to effortlessly create polished recordings directly from your browser.
A couple of technical highlights I'm particularly proud of:
Initially, I used ReMotion but ran into limitations. I ended up building my own animation engine from scratch, closely inspired by ReMotion's capabilities but tailored specifically for my use-case.
Due to browser security constraints with getDisplayMedia(), I couldn't directly track cursor movements. To overcome this, I trained a custom YOLO (You-Only-Look-Once) model for cursor detection. This allowed me to implement smooth, mouse-following animations within recordings, bypassing the browser limitations entirely.
Happy to dive deeper into the tech if anyone's interested!
Nice work! I really like how simple it was to get started. I know this would make it a bit harder to signup, but I think you should consider creating a Chrome Extension that allows you to get the real mouse coordinates, if possible. I think the animations feel a bit rougher compared to ScreenStudio, so maybe having the real coordinates would help. But maybe that's fixable with more smoothing or easing.
Hey Jacob, great observation. I actually already have a chrome extension and I tried to launch it but no one would download it. That’s why I made it so easy to get started this way and had to go through such great lengths to get the cursor but I agree with you 100% the animations are actually terrible right now. I’m working on them as we speak and I’m throwing all sorts of tricks in the mix to try to get it to work better. I’d love if you follow along on my website is a link to the Twitter. If you follow me, I’ll be announcing when I update the animation changes.
Seems cool! But for privacy reasons I prefer to do the recording locally and send the file myself. I've been using the extension https://cursorful.com/ (only for Chrome), which doesn't upload the video to the cloud, it just does everything locally, and the free tier features are enough for me.
Thanks! Hosting in the cloud lets me centrally manage assets and makes it easier to add interactive features like Arcade and Supademo, which require server-side components. It also simplifies analytics and customization. Currently, I store everything in S3 but purge the videos every few days. If this gains traction, I'm already experienced with implementing standards like SOC 2.
I'd love a self-hosted version for no other reason than to be able to customize encode quality and video size. I'd like to do some video casts for my coworkers on workflows I developed/use, but sharper text would be convenient.
OP here: I'm looking for testers for my web-based screen recorder app—specifically product marketing managers or anyone regularly demonstrating software to clients. I'd love your feedback, and in return, I'll prioritize building or adjusting the tool based on your specific needs (within my technical capabilities). I'm particularly interested in exploring interactive demo features similar to Arcade or SupaDemo, so suggestions in that direction would be especially helpful.
I use screen recorders all the time. Question: what format does it capture in? Some screen recorders capture as mov files, which get janky in high zoom. A pro app like Camtasia captures in what I assume is the raw screen data, and for this reason zooming in on a clip works really well.
I use getDisplayMedia which is a JavaScript API that lets you capture recordings as large as your computer can handle, but the format is fixed to either MP4 or WebM.
There are definitely artifacts, especially in the latter, but in practice the videos I've been creating have been working well. Also, it's primarily because the zoom is so high.
In practice, I usually set the zoom much lower and it doesn't have as much artifacting. Check out the homepage video
System audio (i.e. Zoom calls) can be captured only on Chromium browsers on Windows and ChromeOS when sharing the entire screen; tab audio has wider OS support; Safari and Firefox do not support system or tab audio capture https://addpipe.com/docs/recording-client/screen-recording/#...
The MDN support table does not differentiate in this regard. (le: it actually does if you click to see the implementation notes)
Congratulations on the release, is there a plan to opensource it?
I've been using https://snapify.it/, it's opensource but I haven't been able to selfhost it.
Thank you pratio! I quit my job and this is my full time gig now. I've got about three years of runway, so if I can't get it to PMF soon--then yes and I'll move on to something else.
I'm super passionate about this though. Totally up my alley as a Dev and aspiring Growth Hacker, Product Marketing Manager.
The idea is to build a tool and use it to grow itself.
Also, anyone want to talk offline about anything at all, please contact me at john@screenrecorder.me
Hey! Great product you released. Don't feel pressured by the people asking you to release everything for free, it's a terrible strategy to create a profitable business.
Ha thanks for your vigilance owebmaster. I can see your support on some of the other comments! It's just the venue though, and I would expect nothing less. The fact that HN didn't shit all over it is a big enough bar for me for now ;-)
Would love to be able to self host this tool - I'd be fine with purchasing a license to do so but I'd really rather not have this running on hardware that isn't mine - wrapping this up in a docker container that I can deploy would be mint.
Right on! That is definitely something I am willing to explore over time! I know Basecamp has been experimenting with the Pay Once model with great success.
I don't really have a use for this particular software, but I wanted to add my +1 to pay once self-hosted Sofware. It's such a great model for the customer and I have bought software with that model even when there were free open-source alternatives. I really hope it works on the business side as I would love for it to viable.
> A tiny percentage of people might still pay for it, but the vast majority will not, and will complain loudly about any attempts to monetize.
This is true of all products though, in ecommerce 1% conversion is pretty average... generally the path is to increase the top of the funnel because it's always small at the bottom... some products get by with 0.5% or less
I would like for artisan, indie developers to be able to sustain some kind of a living off their work instead of being forced to work at a soul-sucking faang optimizing ad clicks because their hobby projects can't pay any bills.
Though I can also understand the child comment on your comment as well. I think that is a reasonable statement as well.
This is a really weird ethical dilemma that I still hold, as a hobby creator myself (just not at a scale of really earning money,but boy I sure would want to ), I want to earn money from it but I also want to distribute it freely, I am not sure how I can get both.
How much would you pay for it for lifetime value?
Context: I had built very similar tool myself & can implement a few more features to make it self host possible to sell license.
Fantastic product! I almost feel bad for Screen Studio as he's a solo creator and charges for the product, this will most certainly be bad news for him
I’ve using screen studio since it was beta release. Making alternative is a start but if you want to be better, I would suggest adding what’s missing from screen studio.
Few things I would like to see:
- Ability to add images in the time line, think of intro/outro.
- Ability to add notes (text) on frames to explain something or highlight an area.
As some comments mentioned, running locally is a must, and I have no problem paying for it.
Hey! What are your opinions on open sourcing this tool considering privacy considerations and simply I would love it if I can host my own data/ not have to worry about what proprietary thing is running!
Probably having the license AGPL or some certain restriction where you can only attract the big business users on this product.
Let's be honest, you can make money from the small guys but they are little cents as compared to giving corps self hosting guides / licenses like docker..., I know they may not come under pure open source, but the question is tricky, and so is the answer.
So this becomes "fair source" I guess.
I just want the code itself to be readable & buildable from scratch, I want to trust in the process that there are atleast some person who have searched this code and found no malicious use. I want it so that I can modify your program if I really want to, maybe even redistribute that modifications.
> I just want the code itself to be readable & buildable from scratch, I want to trust in the process that there are atleast some person who have searched this code and found no malicious use. I want it so that I can modify your program if I really want to, maybe even redistribute that modifications.
Yes. You want, want, want. But I asked what you are willing to give to get what you want and it seems to be nothing.
Yea because I am just a teenager and to be honest pretty broke right now. I can not give money, I am a pretty frugal person, the Environment and my upbringing has made me such.
Though I will contribute to the project on issues I suppose. That's the least I can/will do.
I tried capturing a portrait mode monitor, and that video came out very weird: Only the upper right quadrant of the video had a gradient, with my portrait window in the middle....
Awesome stuff.
It requires some polishing though, but I prefer it over many apps.
To me - it didn't work well with USB camera on a laptop - it could only choose one and it was the one built into my laptop.
Take the feedback, do some improvements. I will use it even if you don't - love it :-)
Thank you! If you're willing to provide feedback over time, I'll implement features you're looking for if you have legit use cases for this tool and are willing to use it. I'll give you free service in exchange once I begin charging.
The aspect ratio issue you mention has not been well tested, and your feedback is already very valuable. I will fix this.
130 comments
[ 2.7 ms ] story [ 191 ms ] threadI'd love to make it do what you need.
The app uses a custom renderer inspired by ReMotion, so while the app can output MP4s, I also have a player that is dynamic HTML/CSS animation powered by JS, hosted either on my site. I could make it a player that is hosted from anywhere, and I think I will do this for a free tier.
Theoretically, this setup can handle videos of arbitrary length, but practically speaking, I've only thoroughly tested around 10–15 minutes. I'd recommend caution with significantly longer recordings (like an hour) for now, as you might encounter rendering issues.
https://handbrake.fr/
Might be worth calling that out either way.
I've built a 100% web-based screen recording tool aimed at product marketing managers, developers, or anyone wanting to showcase their work along with a talking head. Inspired by ScreenStudio, it allows you to effortlessly create polished recordings directly from your browser.
A couple of technical highlights I'm particularly proud of:
Initially, I used ReMotion but ran into limitations. I ended up building my own animation engine from scratch, closely inspired by ReMotion's capabilities but tailored specifically for my use-case.
Due to browser security constraints with getDisplayMedia(), I couldn't directly track cursor movements. To overcome this, I trained a custom YOLO (You-Only-Look-Once) model for cursor detection. This allowed me to implement smooth, mouse-following animations within recordings, bypassing the browser limitations entirely.
Happy to dive deeper into the tech if anyone's interested!
I really want a free version from commercial use tbh, not that cursorful is expensive
Here is the JS-based player: https://screenrecorder.me/projects/s3hx9YRgXNsrN1s/player
Here is a video that I rendered: https://screenrecorder.me/player/s3hx9YRgXNsrN1s
There are definitely artifacts, especially in the latter, but in practice the videos I've been creating have been working well. Also, it's primarily because the zoom is so high.
In practice, I usually set the zoom much lower and it doesn't have as much artifacting. Check out the homepage video
https://screenrecorder.me/
The MDN support table does not differentiate in this regard. (le: it actually does if you click to see the implementation notes)
I'm super passionate about this though. Totally up my alley as a Dev and aspiring Growth Hacker, Product Marketing Manager.
The idea is to build a tool and use it to grow itself.
Also, anyone want to talk offline about anything at all, please contact me at john@screenrecorder.me
First we really need the Source code of this tool though if we really want this tool to stay for posterity!
"If I make something so great then people will still pay me even if I give it away for free!"
A tiny percentage of people might still pay for it, but the vast majority will not, and will complain loudly about any attempts to monetize.
This is true of all products though, in ecommerce 1% conversion is pretty average... generally the path is to increase the top of the funnel because it's always small at the bottom... some products get by with 0.5% or less
You said it yourself, this is a hobby FOSS developer. Key word hobby.
Have people stopped understanding that others do things simply because they want to? Not everything is profit-driven, even on this site.
Though I can also understand the child comment on your comment as well. I think that is a reasonable statement as well.
This is a really weird ethical dilemma that I still hold, as a hobby creator myself (just not at a scale of really earning money,but boy I sure would want to ), I want to earn money from it but I also want to distribute it freely, I am not sure how I can get both.
What is the cultural significance of this tool that would require it?
People are a nightmare, I can already see someone leaking secrets with this.
Few things I would like to see:
- Ability to add images in the time line, think of intro/outro.
- Ability to add notes (text) on frames to explain something or highlight an area.
As some comments mentioned, running locally is a must, and I have no problem paying for it.
Good luck
Thanks in advance
What are your opinions on the best way for you to pay for it as an open source tool?
Let's be honest, you can make money from the small guys but they are little cents as compared to giving corps self hosting guides / licenses like docker..., I know they may not come under pure open source, but the question is tricky, and so is the answer.
So this becomes "fair source" I guess.
I just want the code itself to be readable & buildable from scratch, I want to trust in the process that there are atleast some person who have searched this code and found no malicious use. I want it so that I can modify your program if I really want to, maybe even redistribute that modifications.
Yes. You want, want, want. But I asked what you are willing to give to get what you want and it seems to be nothing.
Though I will contribute to the project on issues I suppose. That's the least I can/will do.
Take the feedback, do some improvements. I will use it even if you don't - love it :-)
The aspect ratio issue you mention has not been well tested, and your feedback is already very valuable. I will fix this.