Should new DAWs be developed in 2024?
-
- KVRian
- 1419 posts since 7 Oct, 2023 from Tokyo
I am just hoping the current DAWs don't get too far in to the AI circle jerk before deleting it all. I will be so happy when the genAI fad dies out and it just becomes another Clippy.
-
Touch The Universe Touch The Universe https://www.kvraudio.com/forum/memberlist.php?mode=viewprofile&u=190615 - KVRAF
- 5834 posts since 2 Oct, 2008
What happens when AI can create DAWS and synths 
100 High Quality Soundsets: Omnisphere 2, Dune 3, Tone 2 Synths, Pigments, Uhe Synths, Halion, Spire, and others.
TTU Youtube
TTU Youtube
-
- KVRian
- 1166 posts since 20 Oct, 2023
I'm willing to bet Reaper has this feature.
-
- KVRAF
- 2471 posts since 25 Sep, 2014 from Specific Northwest
That would be the longest script ever.
I started on Logic 5 with a PowerBook G4 550Mhz. I now have a MacBook Air M1 and it's ~165x faster! So, why is my music not proportionally better? 
-
- KVRian
- 1166 posts since 20 Oct, 2023
-
- KVRian
- 1419 posts since 7 Oct, 2023 from Tokyo
So... exactly like Reaper thenVOODOO U wrote: Sun Jul 21, 2024 2:02 amAI doesn't run on scripts. It runs on electro-magnetic frequencies which are goverened by perpetual motion satellites. Duh.
-
- KVRian
- 1166 posts since 20 Oct, 2023
I'm willing to bet Reaper has this feature.JerGoertz wrote: Sun Jul 21, 2024 2:28 am Let's just remove all the bugs from the pre-existing DAWs, then we can make new ones.
-
machinesworking machinesworking https://www.kvraudio.com/forum/memberlist.php?mode=viewprofile&u=8505 - KVRAF
- 8025 posts since 15 Aug, 2003 from seattle
IMO it's always geared in the wrong direction. AI would be great for search functions for instance, find the third beat in every fourth measure and select, or delete it. We have voice activated search functions for our phones FFS why not for logical functions in our DAWs? Instead we get AI drummers and other silliness.stoopicus wrote: Sun Jul 21, 2024 1:05 am I am just hoping the current DAWs don't get too far in to the AI circle jerk before deleting it all. I will be so happy when the genAI fad dies out and it just becomes another Clippy.
AI could also be useful for similarity searches for patches, sounds etc. but it's near worthless for making the actual music.
-
- KVRist
- 375 posts since 17 Nov, 2022
Search functions don't belong to the strengths of AI because they need to be exact.
-
- KVRAF
- 2471 posts since 25 Sep, 2014 from Specific Northwest
I think they mean AI should be used to analyze and group samples by sound qualities rather than name, like XO and Atlas do. This is something that AI can do quite well, even if simple heuristics can probably do it faster using fewer resources.WackyZoundz wrote: Sun Jul 21, 2024 7:36 am Search functions don't belong to the strengths of AI because they need to be exact.
I hate the term AI and I don't think Machine Learning is much better. It's all Markov chains and Newton-Raphson type stuff under the hood, but with dedicated hardware to make it happen. It's just another to in the toolbox, not a Universal Fix-o-matic Hammer you can hit anything and everything with,
I started on Logic 5 with a PowerBook G4 550Mhz. I now have a MacBook Air M1 and it's ~165x faster! So, why is my music not proportionally better?
-
- KVRAF
- 2140 posts since 16 Jan, 2013 from USA
It's an interesting video, and most of the argument is financial. Thinking about it, I actually agree, but for the most part, all DAWs do the same thing, which is also amply pointed out in the video.
What most people praise or complain about with a DAW is how you create the input for the engine, aka, the interface. Most of the missing features I see people ask for are interface/methodology-related. My personal favorite DAW is completely based on how well the interface and methods fit the way I work. In truth, I could accomplish my goals with almost any DAW.
If a commonly available engine were in place to begin with, you're cutting a whole lot of hours from the development process. Not that interfaces are easy as you can tell from the many mediocre ones. But if all you're doing is creating new ways to interact with an existing data format and engine, there's plenty of room for creativity.
Of course, where you find such an engine is the issue. I think Bitwig tried this to begin with, but kept running into problems with the engine whose name I can't remember and am too lazy to look up. But it did allow for a rather fast initial development.
If such an engine were readily available, relatively affordable, and easy to use, I think we might see more fresh takes. Or at least tailored ones. That of course depends on the development tools available.
But what this also means is that if an existing DAW's engine is modular enough, the company that owns it could slap a completely new interface on it for far less money. Currently, every once in a while they "revamp" them, but they always have legacy customers to keep happy and it's always modest, incremental, and never revolutionary.
One example where a company did slap a completely new interface on their existing engine (I think), then promptly put it out to pasture when it didn't take over from Garage Band, was remarkably--Steinberg with Sequel.
https://jonljacobi.com/steinberg-sequel ... great-daw/
I was fully convinced that a variation of that interface was going to be applied to Cubase in short order, but I was dead wrong. To this day, I think the company missed a golden opportunity.
This got long in a hurry... Enough said.
What most people praise or complain about with a DAW is how you create the input for the engine, aka, the interface. Most of the missing features I see people ask for are interface/methodology-related. My personal favorite DAW is completely based on how well the interface and methods fit the way I work. In truth, I could accomplish my goals with almost any DAW.
If a commonly available engine were in place to begin with, you're cutting a whole lot of hours from the development process. Not that interfaces are easy as you can tell from the many mediocre ones. But if all you're doing is creating new ways to interact with an existing data format and engine, there's plenty of room for creativity.
Of course, where you find such an engine is the issue. I think Bitwig tried this to begin with, but kept running into problems with the engine whose name I can't remember and am too lazy to look up. But it did allow for a rather fast initial development.
If such an engine were readily available, relatively affordable, and easy to use, I think we might see more fresh takes. Or at least tailored ones. That of course depends on the development tools available.
But what this also means is that if an existing DAW's engine is modular enough, the company that owns it could slap a completely new interface on it for far less money. Currently, every once in a while they "revamp" them, but they always have legacy customers to keep happy and it's always modest, incremental, and never revolutionary.
One example where a company did slap a completely new interface on their existing engine (I think), then promptly put it out to pasture when it didn't take over from Garage Band, was remarkably--Steinberg with Sequel.
https://jonljacobi.com/steinberg-sequel ... great-daw/
I was fully convinced that a variation of that interface was going to be applied to Cubase in short order, but I was dead wrong. To this day, I think the company missed a golden opportunity.
This got long in a hurry... Enough said.
-
- KVRian
- 1419 posts since 7 Oct, 2023 from Tokyo
As a DAW consumer (and not a DAW maker), wanting to have access to fewer tool options makes no sense, so I am going to go with "Keep making new DAWs please".
-
- KVRAF
- 2140 posts since 16 Jan, 2013 from USA
Methinks you are buying the AI hype. It's nowhere near that. Just look at your spell-checker...