OBS Dark Style concept

[Kharay]

A proper skinning system could be coded very effectively, not using all that many resources or cycles, actually. However, as Jim implied, the current UI of OBS is mostly hardcoded. Which means such a skinning engine would need to be coded from the ground up and would involve recoding large portions of OBS.

Which is a lot of work. So, for this to happen... it might take a while. As it stands, I feel that even though OBS is already in a good place as far as its functionality and performance go, it still has work to be done in areas that are of a greater priority than how it actually looks.

In my humble opinion, how the tool looks is infinitely less important than how well it performs and what it can do.

 

[Bensam123]

A GUI, where you just change colors isn't not resource intensive. You could even add spinning gifs to it and it still wouldn't matter. This isn't like running a Crysis 3 in the background and it's not like adding bloat to a instant messenger.

Using 'more resources' is very vague. Of course it will use more resources, but when you're talking about a couple KB of memory that's piss in the pool.

No sacrifices need to be made. You don't need two different applications running to simply change the color of a skin. Why would you need to sacrifice the native optimized UI to use a skin? It would look identical to what you're using now!

People are throwing around a lot of vague terms here. Skinning isn't rocket science. I mentioned I don't know exactly how it works because I haven't seen Jims implementation, which he said is hard coded (meaning it's probably compiled along with the program and there is no way of changing it without changing it for everyone).

You are definitely exaggerating. Skinning isn't super resource intensive, which is also very vague. Using a couple more KB of memory, sure, or maybe even a few MB, but it's not going to start leaking memory like a sieve and you'll end up with the performance of Xsplit. Changing colors or reorganizing the buttons, will not cause massive performance loss. I'm pretty darn sure you can't optimize for 'grey' instead of 'black', maybe Jim will have to comment on that one...

I don't know why you'd assume that Jim would have to rip OBS apart to add skinning? Hard coded can mean a lot of different things, which is also a 'vague' way of saying things. OBS will still be OBS, whether it's grey or black. I mean I almost feel as though Jim should change the grey background to black just to troll you guys.


I really have no idea why some of you are so against this, it's like you think the world is going to end by changing the skin from grey to black. Features are constantly added to OBS, tweaked, changed, and adjusted. Jim does a great job at all of this, but the moment someone suggests adding functionality to allow the gui to (optionally) be changed, you guys throw a fit.

 

[FerretBomb]

Coding time sacrifices. Not using the built-in chrome, and handling the skins in the program. Yes, it's not MUCH additional overhead on the execution side, but it IS additional overhead.

I'd rather see Jim fixing bugs and adding new features that make the actual broadcasts better and squeeze out every ounce of performance, than making the interface slightly more 'lickable'. Function over form.

 

[Muf]

http://www.stardock.com/corporatesolutions/DirectSkin/

Now, who wants to pony up $18,000?

(Also: am I the only one who hopes Andrew Dodd has since been fired from ATI/AMD?)

 

[Bensam123]

That makes everything pretty... Too bad I don't have 18k. :(

It is sorta interesting how Stardock has been releasing lots of helpful side gadgets over the last few years, not just games. Start 8 for instance there is no way I'd touch W8 without.

 

[Deleted member 3892]

Even this is non-profit organization, i dont get why you still have to pay for it, seriously how can you get 18k by doing something for free, if you understand?

 

[IamKillax]

v.0541b and you want a new skin. Adds zero functionality and reduces optimization.

PS- what's wrong with how it looks now

 

[Bensam123]

There is no reduction to optimizations, stop it. Reorganizing the position of buttons or what they look like a long with colors doesn't reduce performance. Grey is not faster then black. By definition 'skins' is a new function, it's just not utilitarian.

Jim has spent all this time putting new features into OBS and optimizing current ones, why do most of you believe he can't put in a skin system that changes the color of buttons without totally destroying everything?

 

[Kharay]

Adds zero functionality and reduces optimization.

Actually, the topic of skinning has been researched and developed upon for quite a while by now and there are many different freeware/open source solutions to implementing a skinning engine that would not reduce the streaming performance of OBS in any way, shape or form.

You're not giving Jim and his fellow coders enough credit as far as optimizing that element of OBS is concerned, actually. The UI has very little bearing on the streaming performance. If you're concerned about the UI slowing OBS down, do you go about closing every single window other than OBS and the game you're streaming? Because, well.... gotta free up those CPU cycles! Oh wait, Windows actually hardware accelerates the windows these days. ;)

Which is not to say I am begging for a skinning solution in OBS. I don't care much how it looks, just as long as it functions the way it does. Which it does just fine. All I am saying is that the argument of performance is a moot one that people should drop.

 

[ThoNohT]

All the skinning that can be done that is hardware accelerated can be done in your control panel by changing your own theme.

If you want real skinning, like what winamp does for example, the point is far from moot. Even though modern systems really should have no problems handling them. There is just no point in introducing it, with no real benefit gained from it.
And even if cpu (or gpu) cycles are of no concern, it's going to take people writing the code, I am very sure Jim, and some other people here know how to do it. But I'd much rather see bugs being fixed, and functionality being added.

Reordering buttons, that's not skinning. That's ui optimization, and that's definitely on the list of things to do, because those are actual functional changes to the ui.

 

[Kharay]

it's going to take people writing the code, I am very sure Jim, and some other people here know how to do it. But I'd much rather see bugs being fixed, and functionality being added.

Yes, this has been my point of view has well. I just wanted to make it clear to people that OBS' streaming performance will not be affected by a skinning engine of some sort. If it would be affected then we're talking about really, really ancient hardware that would have an issue either way. With or without the skinning.

Just to prove my point even further: DOTA2, a modern title, 1600 x 900 (windowed game capture) + Winamp, 10 times, skinned with a modern skin. and... not losing a single frame over it.

 

[Bensam123]

The benefit gained is people can make OBS look like whatever they want. This includes coming out with minimalist UIs (which I'm a fan of). Sometimes there are better UI developers in the community too, who aren't quite up to par when it comes to coding, but can easily move some buttons around. I'm not a big looks or bling guy (hence the love for minimalist or utilitarian), but there is definitely more then one take on the arrangement of buttons (heck there are tons of people that do this in MMOs).

I think too many people have seen instant messengers become bloated and slow when they add more bling to them, that doesn't need to be the way with OBS.

 

[Jack0r]

Consensus seems to be:
current GUI is not perfect, but good
"Skinning/Theming" is partly possible through Windows
Bug fixes and new features are more important
GUI is currently hard coded
GUI changes are on the list for later
GUI changes will hardly be liked by everyone (some prefer this, some prefer that)
No "designer" volunteered to help so far
performance might not be affected if done correct
loading time might not be affected if done correct

 

[Bensam123]

If only such popular votes were required for development on other parts of the project neither you or I spend time on then Jim wouldn't have spent so much time getting capture cards to work!

Making the client skinnable isn't just changing the theme. It means anyone can change it easily and make their own themes.

Please don't present what other posts have said as 'fact', as if one of the developers said it (I'm sure you're doing this so more support for skinning wont happen). Jim did not say bug fixes and new features are more important (skinning is a new feature FYI).

 

[Jack0r]

If only such popular votes were required for development on other parts of the project neither you or I spend time on then Jim wouldn't have spent so much time getting capture cards to work!

Sorry if my english is too bad, but I dont understand what you want to say with this sentence.

Making the client skinnable isn't just changing the theme. It means anyone can change it easily and make their own themes.

This is just a question of how you define skins/themes. As I tried to express with "partly" its not a full skin support where you can change everything.

Please don't present what other posts have said as 'fact', as if one of the developers said it (I'm sure you're doing this so more support for skinning wont happen). Jim did not say bug fixes and new features are more important (skinning is a new feature FYI).

Sorry but I often saw Jim replying to design ideas with, "yea later, currently there are bugs to fix". I dont want to kill any of your ideas but as you can see already on these two pages that half of the people disagree with ideas of the other half. Sorry if you feel attacked by me trying to get the "consensus" of the discussion so far. I didnt say anywhere that I am posting facts.
The point of this discussions seems to go nowhere, for over a page now, you are just talking back and forth, so I wanted to list the parts that everyone can kinda agree on and make it easily reviewable for someone thats want to help on this topic. So maybe we can either get the discussion back to a state with useful new posts or lock it.

 

[tao]

upload it please

 

[Bensam123]

Hmmm... well it definitely seemed like you're proposing there was a end to this discussion. I suppose if you don't natively speak english it may not have been your intent by implying it.

Aye, that's how discussions go... they go back and forth between different points between different people. Sometimes points are repeated multiple times in different ways so people can better understand them or if they glanced over them in the first place.

 

[Muf]

Necroing this thread to direct any programmers who feel strongly about "dark styles" or other skins to this project:
https://code.google.com/p/wxskintoy/

It's woefully outdated, and would need to be patched to support the latest version of wxWidgets. The rewrite of OBS uses wxWidgets as its cross-platform GUI toolkit, so it would need an updated version of wxSkin to support skinning.