Important Alt Crop Upgrade.

RayJay

New Member
After the tedious pixel by pixel setting of recording area was relieved by ALT drag rectangle that selects your desire area we just need 1 more thing.

Make a way to use this selection for video output size.
Its really nice that you can just drag a rectangle to make your own 354 x 673 selection for the thing you want.
But then that gets half defeated when you actual recorded video is still 1920 x 1080 with black bars filling everything outside the selection, and the only way to change this is again the pixel by pixel type box...

You can do this in a number of ways, modify the alt key, ALT + SHIFT modifies both the selection and the output resolution.

Or you can drag and resize the canvas independently.

Or you can just put a checkbox somewhere "output video size to equal first source selection size"

Whatever your method this will be a time saver as significant as the ALT drag crop feature itself.
 

Fenrir

Forum Admin
Just a heads up you can access the crop information from the transform menu which might make your life a little easier for this kind of task. Click the source and press ctrl+e.
 

RayJay

New Member
Do we get to see this happen in the near future or? Usability for this should be pretty high, and complexity pretty low. I dare you to try and use any visual edition software only use canvas resizing the way you do it now. I mean seriously you can drag crop you canvas in MS paint.
 

Fenrir

Forum Admin
I think you're vastly underestimating the "complexity" aspect of something like this. It needs to be implemented in a way that means people won't be shooting themselves in the foot with it, outside the actual code to make sure that all the proper updates happen in real-time when the canvas is resized, locking when output is active, etc. etc. etc.

"Doesn't seem that hard" is a massive fallacy when dealing with a program on the complexity scale that OBS is.

Since it's pretty trivial to get the current size of a source and change the canvas size as needed, this is a very low priority feature that will not likely be added any time soon.
 

RayJay

New Member
Don't you love when people try to bullshit a programmer about programming.

Let me elaborate your programming mysticism for people.

You have a block of code that works - it does stuff.

Right now it takes numbers from a typebox in a popup. People give it the numbers and the programming magic happens. These numbers are handed to an already written function that has a lot of friends and together they know what, where and how to cut.

Lets call the typebox BOB and the function Joe.

Bob is now the only thing allowed to give the Joe the numbers (after Bob got it from the user).

To make this happen you need to create a rectangle (of which you already have at least one - it can be made slightly smarter or CTRL+C, CTRL+V)
This rectangle after being dragged around KNOWS its size a.k.a the numbers we need - we can call it Molly.

If Molly doesn't like to build a relationship with Joe herself to do cutting. She can just pass the numbers to Bob. Bob knows what to do with them and Joe knows what to do with them when given numbers by Bob. This iz just amaze, amIrite?

And for the question of "How to access molly?" I can go speak to some phd rocket scientists on your behalf and we'll solve this unearthly mystery for you?

I have no doubt making a recording software is lots of hard work. But hey I can perform some mysticism of my own and post what code you need to make a rectangle and call it Molly if you want to reveal the whole godly complexity to the other forum users?

But please go and explain in full detail the complexity so our coding department can have a little fun in the coffee breaks.

Saying that you are busy and don't have the desire for this minor UI improvement is completely fine.
Bullshitting people about it - is laughable.
 
Last edited:

Osiris

Active Member
If you are a programmer, why don't you add it to OBS? It is open source after all.
OBS has 1 dev, he can't do everything, that's why a lot of features in the latest version were added by contributors.

Also Fenrir put complexity between double quotes, implying it's not that complex, but that there are some things to consider when creating it. He didn't say it was some extremely complex thing to do.
 
Last edited:

dodgepong

Administrator
Community Helper
I feel like things are getting a bit out of hand with this thread, so let me step in here and try to alleviate some things.

RayJay, thanks for your suggestion. The ability to easily resize the canvas to the size of a specific source has been requested before, usually for people trying to make a simple recording of just a portion of a screen. Personally, I recommend using something like ShareX for this purpose, as using OBS seems like overkill. That said, it's still something that some people want, and that is certainly acknowledged.

OBS Classic actually used to have a feature like this, where you could click a button and resize your canvas to the size of a display capture or window capture. While it was handy for the purpose that you're describing, it actually sort of turned into a support nightmare, because users would press that button without understanding what it did, and then would complain about their recordings coming out at weird sizes. So any feature that allows for changing the canvas size needs to be treated with care so that users understand what is happening when they are doing it so they don't accidentally screw up their settings. This is a UX challenge more than anything else.

I will add that the concept of applying a crop to a source and resizing the canvas are fundamentally different concepts within the program. Even though on the UI side, dragging a box around a thing to resize it feels the same, they are technically wildly different under the hood. There are a lot of concerns to think about when resizing the canvas -- for example, you can't resize the canvas while outputs are active (recording, streaming, NDI, etc.), but you can resize and re-crop sources. If the UI for resizing the canvas and sources were the same, it wouldn't be immediately clear to users why one behaves differently than the other.

Obviously that doesn't mean any of this is impossible, but it need to be designed with a lot of care for the user to make it clear what is going on. I've been volunteering support for OBS for five years, and if there's one thing I've learned, it's the importance of UI that is clear and easy to understand.

So, where does that leave us? Well, there's already a FOSS tool that is specifically designed to do what you're asking for (ShareX), whereas OBS's focus is primarily on video production intended for viewing in a standard-sized viewport (on a computer monitor, potentially in fullscreen, or on a TV). Does that mean it will never get done? Not at all, but it does mean it will be prioritized lower than many other things. Valuable developer time will go to those things first unless a volunteer comes along and decides to take up this project on their own.
 
Last edited:
Top