Posted: June 20th, 2007, 5:53 pmIf everyone wants a skinnable playlist, that is going to delay the release of the Component SDK as it will need support for the skinning engine. At the moment, the skinning support comes from a single component; Shimmer (the interface component).
I would have to abstract this back into the core in order to support skinning. After this I will have to add the ability to the skinning engine, to allow for tiling, stretching and resizing (As everyone wants a resizing library/playlist). Thereafter I can start rebuilding the playlist control and building a custom list control that uses the skin engine from the core. There may or may not be problems doing it this way.
To create something like Kev-O has mocked up, is going to be a nightmare using a PSD file, as far as I can tell. It really depends on how customized/skinned it needs to go. A lot of users have already expressed that using Windowblinds already allows them the level of customization that they desire, while others seem to want more out of a playlist/library.
I've had some thoughts about perhaps combining the playlist into the player skin, so that you can define a region where the list will get generated, but I haven't create a prototype yet to see how feasible it is.
Currently I do all of the skin rendering in Software. Hence the reason that users with slower CPU's will be experiencing slow skins. Xion's current bottleneck for skin rendering is being held back by CPU. For most users this isn't a problem as I've optimized it quite heavily.
I have been toying with the idea of Hardware acceleration through the use of DirectX, however I am as yet, unsure whether this will even work in the way that I need it to. I do hope to create a prototype of this in the future. I do plan on still supporting Software rendering if Xion is unable to create a DirectX surface, however these days, almost all systems should be able to create a DirectX surface.
A lot of people are requesting that Xion have a Library/Playlist similar to Foobar. They also want it skinnable. One of the reasons why Foobar can make such a powerful Library/Playlist, is that it isn't skinnable, or only supports a very simple skinnable format.
So basically it boils down to the following:
If everyone wants an awesome skinnable playlist/library/any component, then that is going to take some time before anything, including the SDK is available. (I work on this in my spare time. Xion provides no income for me and people aren't willing to pay for it, so I have no other choice).
If on the other hand, the skinnable playlist/library/any component isn't that important, then I can release the SDK in its current form and let people loose on creating custom Components.
So, after that run down, which road should I take? Feel free to ask any more questions about anything, and I'll do my best to answer them.
Cliff
I would have to abstract this back into the core in order to support skinning. After this I will have to add the ability to the skinning engine, to allow for tiling, stretching and resizing (As everyone wants a resizing library/playlist). Thereafter I can start rebuilding the playlist control and building a custom list control that uses the skin engine from the core. There may or may not be problems doing it this way.
To create something like Kev-O has mocked up, is going to be a nightmare using a PSD file, as far as I can tell. It really depends on how customized/skinned it needs to go. A lot of users have already expressed that using Windowblinds already allows them the level of customization that they desire, while others seem to want more out of a playlist/library.
I've had some thoughts about perhaps combining the playlist into the player skin, so that you can define a region where the list will get generated, but I haven't create a prototype yet to see how feasible it is.
Currently I do all of the skin rendering in Software. Hence the reason that users with slower CPU's will be experiencing slow skins. Xion's current bottleneck for skin rendering is being held back by CPU. For most users this isn't a problem as I've optimized it quite heavily.
I have been toying with the idea of Hardware acceleration through the use of DirectX, however I am as yet, unsure whether this will even work in the way that I need it to. I do hope to create a prototype of this in the future. I do plan on still supporting Software rendering if Xion is unable to create a DirectX surface, however these days, almost all systems should be able to create a DirectX surface.
A lot of people are requesting that Xion have a Library/Playlist similar to Foobar. They also want it skinnable. One of the reasons why Foobar can make such a powerful Library/Playlist, is that it isn't skinnable, or only supports a very simple skinnable format.
So basically it boils down to the following:
If everyone wants an awesome skinnable playlist/library/any component, then that is going to take some time before anything, including the SDK is available. (I work on this in my spare time. Xion provides no income for me and people aren't willing to pay for it, so I have no other choice).
If on the other hand, the skinnable playlist/library/any component isn't that important, then I can release the SDK in its current form and let people loose on creating custom Components.
So, after that run down, which road should I take? Feel free to ask any more questions about anything, and I'll do my best to answer them.
Cliff
