Hi jahc,

The Audio device mode was implemented some time ago, but I never got round to releasing an upgrade before deciding to finish the skinable system. It'll be there in the next version.

Cheers,
Bean.

@nubechecorre

Quote:

I'm beginning to wish I hadn't made the skin capabilities so comprehensive! Still, I've been driven by what the testers have wanted, so it'll be worth it in the end.

In brief I've reduced the CPU time for software rendering alpha GFX considerably. This was really quite a complex optimisation exercise, and involved additional processing to actually save on raw pixel mix, layer and write operations. I'm pretty pleased with the results and so were the testers - It has made complex skins viable now.

I've just added left/right scrolling for playlists, as well as working listview slider gadgets. In a way I've written a basic GUI system with some pretty flexible means of positioning, moving and laying objects. (Future use... maybe..)

In a bid not to delay the release any further I'm trying to clean up the rough edges and release as frequent updates to the TuneNet site. I have 8 items left on my list to complete before I'm happy to release.

That's really all I can say at the moment, except for a big thanks to all involved in helping move this project along!

Cheers,
Bean.

@ChrisH

Well done Chris! You've clearly put a lot of time and effort into this.

I've never touched E before, but it looks intriguing. When it goes public I'll be sure to try it out.

Cheers,
Bean.

@thread,

Just a little update. I'm currently brushing up one or two of the features, trying to squeeze some additional speed out of the system, + I also have 1 or 2 bugs I'd like to track down.

As someone mentioned to me; there's no point releasing something I'm not entirely happy with. Therefore it's going to take a little longer I'm afraid - at least 2 more weeks before it's released as a preview.

It's very encouraging to see people's interest in the app!

Cheers,
Bean.

@ZeroG

It's listed but not supported - a little confusing I know. It's on my "to do" list! At the moment the streaming part of TuneNet specifically looks for mp3 streams, this is fairly straight forward to change, but I've also a bit of work to do with the plugin API.

Cheers,
Bean.

@ZeroG

No, not at the moment. The AAC plugin is from Curty, I need to extend the plugin API to allow different format types when streaming other than mp3.

Cheers,
Bean.

@FuZion

Quote:

I'm holding these guys back a little here, they are doing a great job creating and converting skins and I'm trying to cram in as much as possible into the TuneNet preview.

Soon you'll all be able to get creative with a fairly powerfull (if un-optimised at the moment) Skinnable system.

Cheers,
Bean.

P.S. I'm pretty pleased with it, can you tell!

P.P.S. It will be demo'ed again at ABW (former ANT group) this Sunday.

http://www.amiganorththames.co.uk

Excellent idea, and nicely implemented.

Thanks,
Bean.

@Raziel

Can you please try doing the same tests with both AmigaAmp and ANR and then PM me? Please test for some time as well.

The skinnable interface is my major concern at the moment, I'm already a day overdue!

Cheers,
Bean.

P.S. Whilst developing I do have TuneNet running (usually playing the Groove Salad station ), along with iBrowse and Simple Mail. But then I am using the latest compiles.. Ho humm...

@Mikey_C



mmmmmmmmmmmm, Chocolate Gataux!

@Snuffy

Hi,

There is no new version just yet, just information about the upcoming version! The version on the downloads is older than the one on OS4 Depot - and is a bit buggy.

I'd update it, but the new version is just round the corner now and I'll sort the site out then. Not long now.

Cheers,
Bean.

@Mikey_C

The biggest problem I've got with all of this is the apparent pain the developers have all gone through entering this risky business in the first place, and the potential issues if OS4 is not resolved correctly.

If one day it is all over, and there is no resolve, we'll all look back at the past 10 years and say, "if only...". ..and that'll be sad.

It's in the hands of the courts now, speculating and worrying won't achieve anything, but I guess there isn't much else to do! Hold tight and brace for impact, or something!

Cheers,
Bean.

@acefnq

Yeah he did! There was quite a lot of interest in the way skins are constructed, not so much in the actual application itself! (I suppose at the end of the day all it does is play music!). I'm gearing up to a last minute "preview" release at the end of the month which will be released on the TuneNet homepage, but not on OS4 depot as it is not ready yet for general use.

Those interested in creating skins (or wanting AHI Device Audio as apposed to the Music Unit) should download it when available. I'll give you all a shout when ready.

Cheers,
Bean.

@Raziel

It'll be out soon enough in a preview version! Not long now to wait.

Cheers,
Bean.

For what it's worth I got AWeb to crash here whilst using latest
version.

I'll do a thorough run through of my network code and let you
all know.

@Rudei

Quote:

Good luck!

@Rudei

..and possibly BSD Socket!

Are you comming to ATBW (ANT) tomorrow, perhaps you can test the latest TN for this problem on your machine?

Cheers,
Bean.

@Raziel

I think we need to know what is happening in AWeb at the point of the crash and what BSD socket function (Opensocket by the looks of it) and the parameters that Socket function is being called with before going any further. After all the crash is being reported on the AWeb task and not TuneNet's. Logically we need to know why the crash occurs in AWeb first.

I tend to think it may be a bug in the socket library.

The main body of TuneNet's streaming is pretty much made up of a non blocking:

ISocket->recv();

It always requests a value greater than 0 and smaller than the buffer allocated for it. In terms of using the socket library there's not much more to it than that!

As I said I can't spend much time on this at the moment, but I think TuneNet is the red herring here!

Cheers,
Bean.

@Mikey_C

I'll be there.

Once you've finished laughing at Mikey.. ahem I mean laughing with Mikey... come and also have a look at Skinnable TuneNet due for release later this month.

Cheers,
Bean.

@All,

Hi guys, yep I am listening - just a bit busy! I have no idea what it is at this point in time, as you say it could be a variety of things.

As I'm working full pelt at getting the next "preview" release in a usable state I'm not going to have time to look in great detail at this just yet. I did have a quick look at my streaming code and network class, but I couldn't see anything obvious.

Some questions for people who suffer with this:

1/ Does TuneNet still continue playing, after the AWeb crash?

2/ Does another browser, like IBrowse, operate after the crash has occured.

3/ Can you confirm you have sufficient stack on AWeb?

4/ Can someone try running TuneNet with the -nogui and -nodock switch from cli with a radio station - does that cause the same problem.

Each socket opened using the bsdsocket library can only be (or should be) accessed by the calling task, which seperates things nicely. Perhaps there is something in TuneNet that is trashing a memory in AWeb or the socket library? or perhaps the additional load of TuneNet streaming in addition to web browsing is causing problems with the bsdsocket library? Perhaps it's an issue with AWeb which only comes to light when TuneNet is running?

There are many factors involved here I'll try and help out the best I can.

Cheers,
Bean.

P.S. I can't recreate the problem at the moment.