4 ms·

For the OBS fans who have been wanting a builtin equalizer for years, you get the 3-band one and that's it. I submitted a decent 8-band, asked for feedback, and

For the OBS fans who have been wanting a builtin equalizer for years, you get the 3-band one and that's it. I submitted a decent 8-band, asked for feedback, and after a few months it was rejected. They want you to use plug-ins. I'd be happy to improve the one I submitted (more bands, steeper rolloff,etc) but the project does not want such a feature. I'm not butthurt, I just find it odd but half expected the rejection going in based on what I read.

moritonal 4y ago
Could you expand on the context? Why would an open source project not want what sounds like an obvious improvement?
- andreareina 4y ago
  Not the GP or associated with OBS but they could conceivably want to not handle anything that can be (and are already being) done via plugins. What's an obvious improvement to you is new code (always a risk) doing something that's already solved.
- phkahler 4y ago
  I suspect it's because the project is primarily focused on video. Another reason could be that audio effects can be an endless rabbit hole of constantly striving for some elusive perfection. Best to leave it to plugings. On the other hand, I think basic functionality out of the box is still a good idea and the 3-band doesn't quite cut it. But thats just me - I just used it once and EQ was the one thing I found wanting. The Cynic in me wonders if the project is funded by plug-in vendors and this might be a conflict! I lean towards the rabbit hole theory though.
  - corobo 4y ago
    My quick assumption is that they don't want to maintain code you wrote forever. If it's done as a plugin they don't need to worry about it. I don't know how complex the code is though, of course. E: Their response seems to be that users who would find this useful already use a plugin. https://github.com/obsproject/obs-studio/pull/7961 https://github.com/obsproject/obs-studio/pull/7961
    - phkahler 4y ago
      >> Their response seems to be that users who would find this useful already use a plugin. But nee users will have to figure that out. That is one of the worst excuses they could have offered. BTW the code is completely self contained. Nothing to maintain. That's due to a clean API on their part.
      - wsgeorge 4y ago
        > Nothing to maintain. Never say never...
  - ilyt 4y ago
    Audio-wise it should probably support LADSPA/VST plugins and that's all, there is EVERYTHING audio wise available between those 2 formats. I think it supports VST2 currently ?
  - throw0101b 4y ago
    > I suspect it's because the project is primarily focused on video. Audio is 80% of video. Most people will watch to a potato-quality video if the person talking can be understood fairly well, but will probably not bother with something where the audio is echo-y, with wind, and sounds like it was recorded over a wet string, even if the image is 4K/HDR. How advanced the included audio controls is certainly up for debate, but it's not possible to focus on 'just' video.
    - bigtex 4y ago
      This is very true, as someone helping a church do there weekly services, audio is king. Even the best soundbar at maxxed volume can't overcome a service broadcast using the built-in microphone on an iPhone. But get the microphone audio plugged directly into the stream and that changes everything.
  - someguydave 4y ago
    why not let streaming software just stream? equalizing audio should have happened before it got to the streaming software
    - VanTheBrand 4y ago
      Have you used OBS? It’s not just “streaming software” it’s a software audio video mixer.
- onion2k 4y ago
  Why would an open source project not want what sounds like an obvious improvement? Every additional feature is another thing to maintain and test. Saying no leaves more time for everything else so there needs to be an exceptional reason to say yes. This is true for all projects, not just open source.
- PaulDavisThe1st 4y ago
  OBS is not a DAW. On every (?) platform where it runs, there are ways to connect capable audio processing software to OBS, and that software already has all the capabilities you might want (and more). Yes, it's a bit more convenient to do it all in the same program, but the complexity and maintainance expansion has its own costs.
albertzeyer 4y ago
Can you link the PR? But also, what's the problem with having this as a plugin? Isn't it better to keep the core OBS small, and provide extra functionality via plugins?
- phkahler 4y ago
  >> Can you link the PR? https://github.com/obsproject/obs-studio/pull/7961 https://github.com/obsproject/obs-studio/pull/7961 I guess most frustrating to me was the request to format the code and fix the commit message as if it was on track, only to be rejected for "we don't want this in the first place". Plugins suck. There are a lot of people who will not bother to go that route. It immediately raises questions about what OS do they work on, do they cost money, how do I install them? They include compression and noise gate among others. Enough to clean up my microphone input. But not much in the way of tone control.
  - mikece 4y ago
    > Plugins suck. There are a lot of people who will not bother to go that route. Serious people WILL explore the plugins route, but one has to go from "What is this thing I just downloaded and how do I use it?" to "Okay, I've got the hang of this, I'm pretty good at it, but the stock tools are holing me back from making something better." I would venture to say that the majority using OBS will never make it to that second phase. For those who do, have a well SEOed blog post and YouTube/TikTok video ready to show them how to take it to 11.
  - capableweb 4y ago
    > I guess most frustrating to me was the request to format the code and fix the commit message as if it was on track, only to be rejected for "we don't want this in the first place". Usually, those things are easy to check, almost a part of just triage where you try to cover lots of ground but just the surface. Takes small amount of time for the maintainer unless there is a lot of ground to cover obviously. But unless they explicitly signal the change is wanted, I wouldn't (in the future) take that as a signal that the change in general is wanted. As a tip, before trying to contribute any larger patches to FOSS, create a issue to ask if this is something they'd actually be willing to merge, if you chose to build it. Then you know for sure up front if it's worth spending time on. > Plugins suck. There are a lot of people who will not bother to go that route. It immediately raises questions about what OS do they work on, do they cost money, how do I install them? Personally, I appreciate the architecture of "small core - allow extensions" so I'm happy they don't expand OBS in that way. As well, I'd appreciate it a lot if you made what you did into a plugin, better audio control is wanted by a large set of OBS and I know many do the piping outside of OBS to get a proper equalizer. So if you spend time on making it into an extension, I'm sure you'd find lots of people grateful for that.

moritonal 4y ago

Could you expand on the context? Why would an open source project not want what sounds like an obvious improvement?

andreareina 4y ago
Not the GP or associated with OBS but they could conceivably want to not handle anything that can be (and are already being) done via plugins. What's an obvious improvement to you is new code (always a risk) doing something that's already solved.
phkahler 4y ago
I suspect it's because the project is primarily focused on video. Another reason could be that audio effects can be an endless rabbit hole of constantly striving for some elusive perfection. Best to leave it to plugings. On the other hand, I think basic functionality out of the box is still a good idea and the 3-band doesn't quite cut it. But thats just me - I just used it once and EQ was the one thing I found wanting. The Cynic in me wonders if the project is funded by plug-in vendors and this might be a conflict! I lean towards the rabbit hole theory though.
- corobo 4y ago
  My quick assumption is that they don't want to maintain code you wrote forever. If it's done as a plugin they don't need to worry about it. I don't know how complex the code is though, of course. E: Their response seems to be that users who would find this useful already use a plugin. https://github.com/obsproject/obs-studio/pull/7961 https://github.com/obsproject/obs-studio/pull/7961
  - phkahler 4y ago
    >> Their response seems to be that users who would find this useful already use a plugin. But nee users will have to figure that out. That is one of the worst excuses they could have offered. BTW the code is completely self contained. Nothing to maintain. That's due to a clean API on their part.
    - wsgeorge 4y ago
      > Nothing to maintain. Never say never...
- ilyt 4y ago
  Audio-wise it should probably support LADSPA/VST plugins and that's all, there is EVERYTHING audio wise available between those 2 formats. I think it supports VST2 currently ?
- throw0101b 4y ago
  > I suspect it's because the project is primarily focused on video. Audio is 80% of video. Most people will watch to a potato-quality video if the person talking can be understood fairly well, but will probably not bother with something where the audio is echo-y, with wind, and sounds like it was recorded over a wet string, even if the image is 4K/HDR. How advanced the included audio controls is certainly up for debate, but it's not possible to focus on 'just' video.
  - bigtex 4y ago
    This is very true, as someone helping a church do there weekly services, audio is king. Even the best soundbar at maxxed volume can't overcome a service broadcast using the built-in microphone on an iPhone. But get the microphone audio plugged directly into the stream and that changes everything.
- someguydave 4y ago
  why not let streaming software just stream? equalizing audio should have happened before it got to the streaming software
  - VanTheBrand 4y ago
    Have you used OBS? It’s not just “streaming software” it’s a software audio video mixer.
onion2k 4y ago
Why would an open source project not want what sounds like an obvious improvement? Every additional feature is another thing to maintain and test. Saying no leaves more time for everything else so there needs to be an exceptional reason to say yes. This is true for all projects, not just open source.
PaulDavisThe1st 4y ago
OBS is not a DAW. On every (?) platform where it runs, there are ways to connect capable audio processing software to OBS, and that software already has all the capabilities you might want (and more). Yes, it's a bit more convenient to do it all in the same program, but the complexity and maintainance expansion has its own costs.

albertzeyer 4y ago

Can you link the PR? But also, what's the problem with having this as a plugin? Isn't it better to keep the core OBS small, and provide extra functionality via plugins?

phkahler 4y ago
>> Can you link the PR? https://github.com/obsproject/obs-studio/pull/7961 https://github.com/obsproject/obs-studio/pull/7961 I guess most frustrating to me was the request to format the code and fix the commit message as if it was on track, only to be rejected for "we don't want this in the first place". Plugins suck. There are a lot of people who will not bother to go that route. It immediately raises questions about what OS do they work on, do they cost money, how do I install them? They include compression and noise gate among others. Enough to clean up my microphone input. But not much in the way of tone control.
- mikece 4y ago
  > Plugins suck. There are a lot of people who will not bother to go that route. Serious people WILL explore the plugins route, but one has to go from "What is this thing I just downloaded and how do I use it?" to "Okay, I've got the hang of this, I'm pretty good at it, but the stock tools are holing me back from making something better." I would venture to say that the majority using OBS will never make it to that second phase. For those who do, have a well SEOed blog post and YouTube/TikTok video ready to show them how to take it to 11.
- capableweb 4y ago
  > I guess most frustrating to me was the request to format the code and fix the commit message as if it was on track, only to be rejected for "we don't want this in the first place". Usually, those things are easy to check, almost a part of just triage where you try to cover lots of ground but just the surface. Takes small amount of time for the maintainer unless there is a lot of ground to cover obviously. But unless they explicitly signal the change is wanted, I wouldn't (in the future) take that as a signal that the change in general is wanted. As a tip, before trying to contribute any larger patches to FOSS, create a issue to ask if this is something they'd actually be willing to merge, if you chose to build it. Then you know for sure up front if it's worth spending time on. > Plugins suck. There are a lot of people who will not bother to go that route. It immediately raises questions about what OS do they work on, do they cost money, how do I install them? Personally, I appreciate the architecture of "small core - allow extensions" so I'm happy they don't expand OBS in that way. As well, I'd appreciate it a lot if you made what you did into a plugin, better audio control is wanted by a large set of OBS and I know many do the piping outside of OBS to get a proper equalizer. So if you spend time on making it into an extension, I'm sure you'd find lots of people grateful for that.