Designing a DAW from Scratch with JUCE

I’m toying with the idea of hiring a programmer to design a basic DAW from Scratch, to which I can add my own unique features.
The basic DAW would have a Tracks view, VST inputs, piano roll, MIDI editor.
No fancy extras.

So what does JUCE offer to make this process easier and faster?
(eg. Relevant tutorials, predesigned modules and templates)

How many hours of work (rough estimate) would it take for someone who knows what they are doing?

How much would such a person typically get paid per hour?

I have looked for this information on Freelancer.com and they don’t even have JUCE in their search menu.
And the Jobs page here seems to keep negotiations private.

Any help would be appreciated.

It’s not part of JUCE, but the Tracktion engine might be of interest to you: Tracktion Engine - open source audio software daw

A lot. Even a basic project will take hundreds of hours to do properly. You can rush it, but you’ll just be adding technical dept to deal with later.

Even a basic DAW will likely require thousands of hours. A full team of experienced developers would take several years to build a DAW.

Depends on the individual, but minimum would be somewhere around £30/hour. For someone with more experience, and more expertise in JUCE and audio programming, you’d be looking more at £50/hour. If you want them to only work on your project until it’s done, spending months dedicated to your project, you’d probably be paying even more still.

4 Likes

Thanks for the info.

The traction option looks interesting.

I will investigate it further.

I worked on a few, both commercial and as a hobby (sometimes I wonder why I keep writing code even on weekends, must be some sort of illness), both with JUCE and with native APIs only.
It’s a lot of work.

In my opinion, JUCE gives you the following big advantages:

  • It’s a crossplatform library. You write the code once, you run it on multiple platforms
  • It’s mainly audio focused, which helps you a lot when you want to host plugins and “talk” to audio drivers easily
  • It’s a well maintained library, which helps a lot when a new OS version comes out
  • It’s a mature library

and a few disadvantages:

  • It’s a crossplatform library, but you don’t have easy access to native widgets and APIs
  • If you want to do some heavy drawing, you need to know the ins and outs of the library to determine what bottlenecks you may encounter during development

As already mentioned, the Tracktion Engine can be a great head start.
When you say

The basic DAW would have a Tracks view, VST inputs, piano roll, MIDI editor.
No fancy extras.

This requires years of work, because all these things are just what the user is going to see.
But these features are built on top of an engine. I wrote a few engines in the last ten years or so, they can be a pain in the ass if you want to achieve scalability. You may want to add a few tracks, but what happens if you want to add 100? Will the engine be able to play audio, MIDI, plugins and whatnot without glitching? So yeah, the Tracktion Engine should be great for that.

And then there’s the UI/UX side of things. This can go wrong very fast, especially nowadays with large screens and a tremendous amount of pixels to handle.

Here’s one I made as a pet project, I’ve been working on it on and off for the past eight years and I use it to produce my music. It doesn’t use the Tracktion Engine, everything is written from scratch and it’s made with JUCE (for the UI, audio drivers attachment and plugin hosting).

It’s a fun project because there’s no product manager that tells me what to do :slight_smile:

Good luck!

6 Likes

Rather than designing a new DAW from the ground-up, you could consider adding the extra features you want to an open-source project, like Ardour.

You could still hire a freelancer if you don’t want to do the work yourself, but have them create a fork of the project and add the new features for you. Would take significantly less time.

2 Likes

Just pointing out the obligations the GPLv2 License of Ardour would bring.

I just tell some ballpark figures from my experience, for a MVP I would at least plan with 3-5 months full time. For a proper product I would not work with a freelancer but try to get somebody on board, since when a project is handed over you usually pay a heavy extra price. Very often with the new freelancer replacing stuff with their own mind model, making you pay twice for the same feature…

And I agree with the previous posters, that this could easily (and most likely will) become a lifelong journey.

I know it is an annoying question, but ask yourself why you need yet another DAW. The users will be asking the same question, and you need a compelling argument for them to switch. A DAW comes with the users knowledge of how to do things and sometimes with a whole ecosystem that the users have to abandon for your new DAW.

It is much easier to get them to use it if it can be somehow integrate with their existing tools, templates, knowledge, workflows…

3 Likes

There is a reason why ‘teams’ of developers are required for such a mammoth task.

1 Like

Exactly. I wouldn’t approach it with the notion to create it once and done and dusted…

You need someone permanently, or at least long term available. And if it’s one person and that one drops out for any natural reason, you are basically screwed.

1 Like

Wow, I am impressed by your effort here.
You must have learned many things along the way!

I will certainly remember you, if I decide to go down this path.

Do you know if I chose the Tracktion option, would my work belong to them, or can I develop my own proprietary DAW while using the Tracktion Engine?

Hi Daniel,
Can you explain more about the licensing of Ardour?
If I built a DAW using Ardour would it necessarily become open source and I could never sell it?

Also, my ideas for a new DAW are not currently being implemented anywhere, and the current design of the VST3 architecture may prevent me from making a plugin that can do what I want.

My DAW would be by far the best in the world for a certain subset of users, and I’m sure they would gladly abandon their current DAWs for the features I would offer.
Of course I would make it as user-friendly as possible. (I find Reaper and FL Studio to be incredibly awkward to use).
Most DAWs seem geared towards genres like hip hop, trance, house style music and those users would not benefit much from what I would have to offer.

Ardour is open source under the GPL license.
This license allows everybody to use and to modify the code, which is good for you.
But on the other hand it demands that you publish every modification you make in source code, so that everybody can use it for free.

Creating a commercial product is very hard, since everybody could use this source code and give your software away for free.

But IANAL, please check GNU General Public License v2.0 - GNU Project - Free Software Foundation and talk to a lawyer if in doubt.

I am sure of that, but I won’t be the judge of that, your users will be :wink:

Good luck!

1 Like

Do you want to make a DAW, or make plugins, or both?

As I’m sure you know, many DAWS include built-in effects and instruments that are part of the DAW itself, and so do not have to conform to any of the plugin formats. This is one advantage of building your own DAW from scratch - you can create custom devices with very few limitations to what they can do or how they function.

However, as others have pointed out, building a DAW from scratch is a LOT of work. If you’re bankrolling this yourself, I’d guess you’ll need at least $100,000+ to pay your team of developers for all the time it will take (and that is probably a low estimate). I also wouldn’t expect to have even a working beta version for a year or two (again, a conservative estimate). This is an absolutely enormous undertaking, I cannot stress this enough.

1 Like
  • what subset of users, exactly?
  • I think you’ll find that it is a very tough sell to get someone to switch DAWs, especially someone experienced with music production. Especially a brand new product with zero name recognition, personally I would not give it a second look. Just my two cents.
1 Like

I appreciate your input, because as a user of many DAWs, it isn’t obvious that building one would be that difficult.

And as I don’t have $100,000 to splash on this, I will need to find a simpler solution.

Perhaps a standalone Windows App, but ideally a plugin.
I’m still making enquiries as to whether this is possible.

If it works, it will be more useful than Autotune.
But I’m obviously not going to blab what exactly my idea is.
There is no way to patent an idea.
I just have to attempt to execute it successfully.

If you want to make something like Auto tune, that’s a plugin, not a DAW.

Obviously yes.
But due to current limitations, my idea of a plugin may not be possible.
Hence, worst case scenario, I may have to build a DAW from the ground up.

Personally, I would only embark on DAW-building if the DAW itself was your big idea and passion. Building an entire DAW just to be able to make one special effect seems… inefficient?

I can only speculate what exact limitations you’re encountering, but based on your posts it seems that you are not a programmer yourself – in which case, I think the best course of action is to actually hire a programmer for a consultation, tell them your exact idea and potential problems, and let that person advise you with their specific and in-depth technical knowledge.

1 Like

I have done programming in the past, but not recently.
So I know how painful it is.
But fortunately there are millions of freelance programmers I can hire to do the tedious work of coding.
I’m an ideas person and don’t want to get bogged down in small details.
I just have to know enough detail to know which direction to take.

And a DAW would be an ideal solution in the long run, as I don’t just have one idea but many related ideas that feed into my core idea.
So far, no one has attempted to make what I have in mind, but once I release the product, I can be sure to have competitors following, just as Autotune does.
At this point I’m not sure I want to go down this path, but I can see it would pay off if I was successful.

Sure. I’m just suggesting that us here on the forum shooting in the dark because we don’t actually know what you’re trying to do will not get you very far.

I’m sure if you DM anyone who replied to this thread, they would be happy to offer you a one hour consultation to discuss the specifics of your idea, which is what I think you need :wink:

1 Like