Thoughts on Online Web Based DAW development #7494
Replies: 5 comments 5 replies
-
Web is inherently slower for audio applications and afaik, realtime and web never goes hand in hand. There's this thing with audio that every sample should be processed 44100 times every second, so that might introduce lotta lag. There is an online daw named trackton waveform that i know of, but don't know the inner workings tho. Anyway give it a shot. You'll learn a lot, even if the effort ultimately gets wasted, the learnings won't. If you get stuck, feel free to ask here for help. Or a better place would be the discord server. It's easier to have a chat over there. |
Beta Was this translation helpful? Give feedback.
-
cant we just use bandlab? |
Beta Was this translation helpful? Give feedback.
-
Shouldn't be impossible to make some kind of DAW using this: |
Beta Was this translation helpful? Give feedback.
-
This is interesting! I'm not quite keen on browser DAWS, since (like Rossmaxx says, and my experiences with web-based DAWS) they can unintentionally introduce a huge buffer that heavily limits playback. A tendency with browsers such as Google Chrome is that the processes within the web application can cause serious buffers depending on the user's device, introducing even more buffer/lag that minimizes capabilities and functions on large-scale projects. I think you should strive and attempt to develop such a project because you may know how to combat this or learn how to improve what other web-based DAWS struggle with. Softrabbit's source gives quite a unique view of processing! This is only my thoughts based on my experiences :D |
Beta Was this translation helpful? Give feedback.
-
Hi guys! I am planning to develop an online DAW, since you guys are aware that most of such products are desktop based applications and my peers are raising questions on its feasibility on the web, because most of the plugins are written using c++ and there can be performance bottlenecks for a MERN stack based DAW, What do you think, should I give it a go?? Actually, if it was a hobby project then i wouldnt have thought much and just started without having feasibility concerns but since it is gonna be my FYP, I cant take chances. Because if I get stuck in an issue in the middle of development , then it would be wastage of alot of time and starting again would make it harder for me to meet the deadlines. What techstack do you guys propose?
Beta Was this translation helpful? Give feedback.
All reactions