Comment by rfgplk
21 hours ago
I don't see how companies like this can survive going forwards. DaVinci Resolve is effectively ~200 Astra/Fable prompts, maybe even fewer if you're willing to forgo polish.
21 hours ago
I don't see how companies like this can survive going forwards. DaVinci Resolve is effectively ~200 Astra/Fable prompts, maybe even fewer if you're willing to forgo polish.
I'd give you $100K no questions asked if you manage to get something even close to Resolve with 2K Astra/Fable prompts, polish included. You vastly underestimate how much work goes into "polish" when software with huge surface area is decades old.
I think there's a really interesting tension between what you're saying and what the parent post is saying, and it's something that's been discussed for a long time but LLMs are throwing a new wrench into it: a given user probably only uses 5-10% of the features of a piece of software, but across the whole body of users you get close to full coverage.
I use DaVinci Resolve for editing GoPro, Canon DSLR, iPhone, and Drone FPV footage. I import clips, use the razor to cut them and rearrange them, occasionally crop and zoom, and then encode them down to h264 or whatever. There are a bazillion features that I sort of know exist but haven't ever had any need to use. Could I make a tool using an LLM that covers my DaVinci workflow? Almost certainly, and I don't think it would take anywhere near 200 prompts. I've made significantly more complex bespoke data analysis pipelines for drone flight telemetry in like... a dozen prompts. But I have no illusions that what I would make as my "DaVinci Clone" would have anywhere near the functionality of DaVinci, just the tiny piece of that functionality that I use.
I'm pretty sure your editing workflow can be achieved in the Preview app on macOS. If your editing workflow is just basic chopping clips out of a longer video then yes, you could vibe code it.
> a given user probably only uses 5-10% of the features of a piece of software
Then vibe your own 5% editor, but saying that a software so mature and complete as Resolve can be done with some AI and time is something I expect coming from the mouth of a teenager, not an adult.
regardless of whether it’s possible, there is enormous value in the community, ecosystem, and learning materials centred around Reaolve as it is right now. the workflow has been refined over years. no one shot slop effort will have that network effect behind it, and if the workflow isn’t a drop-in replacement it’s dead in the water anyway.
You can say anything, why stop at claiming 200 prompts, why not say you think Astra/Fable can one shot Davinci Resolve! And you can EASILY make the BM cinema cameras and other hardware by prompting Fable to run a 3D printer. (≧▽≦)
The free tier is very generous and it’s only $300, no subscription. Those Fable credits would cost a lot more than that. You’d spend thousands of dollars to save $300.
Moreover, not everyone wants to have the hassle of building and maintaining their own version of a thing that isn’t compatible with anyone else’s version of the same thing. And if you hire anyone to help you, now you have to train them on how to use your app. If they leave, those skills are useless for their next job.
(I’m not even covering the idea that you could do it in 200 prompts. I got Claude to create a video player and that alone took over 300 prompts.)
Black Magic Design's primary business is selling cinematography hardware. Anyways, those ~200 prompts are going to cost you more money than a license once you factor in the cost of your own time. Value proposition of AI still ain't there, and it probably won't be until the hardware bubble pops, and it very well might never pop without market intervention. This is assuming your 200 prompts is a realistic figure, which I'm frankly skeptical of.
you do know this company also sells myriad of hardware , can't Fable those just yet. and I am sure majority of their income comes from that part of their business, selling equipment to studios etc.
LOL. I wish people could make good software that easy!