Testers wanted for NSF player

[Adam]

There is now an x5 build up here: http://hcs64.com/rockbox .  I am very interested to hear how/if it works for you.
Also, for any interested, it takes 16 minutes on my machine to build all 6 versions I am now providing.

[Sherv]

Thanks Adam, it works perfectly so far!! This is amazing, thanks so much for coding this...I'll be using it heavily over the next few days, I'll let you know if anything comes up!

[Sherv]

Hey guys. So after a day or two of heavy usage I run into one problem on my X5 which I can't reproduce faithfully but which occurs rather frequently. I ROLO into the custom build Adam has compiled, navigate to my NSF directory, chooes a file, and play it. It loads fine, and switching tracks usually proceeds as normal until randomly one track will just be silence. It remains silent if I try to advance past it, but if I move backwards through the tracks in the NSF then it will play. I'm not sure if this makes sense, I'm still trying to pinpoint the exact problem, but it doesn't affect any specific NSFs from what I can tell.

[massiveH]

Any chance of an H3xx build?

[Adam]

H3xx build now built (and others updated), please let me know how it works.

[massiveH]

I have an H360, and it appears to be working fine, I'll test it throughly over the weekend, and let you know my findings, THANKS AGAIN!  8-)

[Sherv]

Kudos on getting this feature committed to the main source dude!

I have two questions for you...firstly, how feasible would it be to change/customize the navigation scheme? That is, instead of the next/prev button presses skipping to the next NSF file how about having the track change instead?

Secondly, suppose I wanted to set up a playlist of specific tracks in different NSFs...is there a way to properly address these tracks or is that impossible?

Thanks bro!

[Adam]

how feasible would it be to change/customize the navigation scheme? That is, instead of the next/prev button presses skipping to the next NSF file how about having the track change instead?

This is already somewhat supported, when you set repeat mode to "repeat one" and press next it will go on to the next subtune in the NSF.  Other than that (which is sort of hackish) there isn't much that can be done from the codec side.

suppose I wanted to set up a playlist of specific tracks in different NSFs...is there a way to properly address these tracks or is that impossible?

The NSFe format supports playlists, so theoretically you could make a bunch of copies of one file and set each to only play a single track.  I am considering working on a tool to do this automatically.

[Sherv]

Ahh, I didn't realize the change in function when in repeat-one mode. Sweet, that's good enough for me! I dabbled a bit with NSFes but that was a while back, I'll look into em again!

[dma]

Great work on this codec as well. Marvelous that it supports some additional chips.

[oblib__]

Thanks a bunch for this codec. One request -- winamp is able to create a playlist indexing the individual tracks within the NSFE file. Would it be possible to make these playlists compatible with your codec? I tried and it froze up. An entry looks like this:

nsf://6:Zelda.nsfe

That will play track 6 of Zelda.nsfe. I guess for now I can make duplicates of the files as you indicated above. Thanks again!

[oblib__]

Well, I went ahead and did it myself, though I'm sure it's not to the point where it could be submitted. For one thing, it changes the codec API without updating the version  

In any case, I'm not a programmer and I've never actually modified open source code before -- you are welcome to take what I've done here and fix it, or throw it out completely!

Theory behind my changes:
I changed the playlist parser so that if a line starts with nsf:// it strips that prefix off, reads and stores the following number, and then passes that number to the nsf codec via the codec_api class.
The nsf codec requests the track number everytime it skips to another track, and if it's not there (-1) then plays as usual. If parsed track number is there, the nsf codec forces that track to play, and then forces the end of that .nsf file when the track is done.
I tested it with a winamp generated playlist and it seems to be working perfectly on my nano.

BTW, I found that the main music of Solstice.nsfe (track 2) from Slick's sounds way off versus what it sounds like in the game and in Winamp. I'm not sure what would cause that...

[attachment deleted by admin for age]

[Sherv]

Hey guys, I've been noticing a strange phenomenon recently while switching to playing NSFs from MP3s: basically, my hard drive on my iAudio X5 will be constantly accessed and navigation through the currently playing NSF or to a new NSF or even to an MP3 will basically freeze. The music will continue to play, the WPS will load all the info for the loaded file/track, but the hard drive will continue to spin like crazy and the player would basically be stuck until suddenly the new file/track will play.

I am running the latest daily build...and I have things such as directory cache on. Any ideas? Thanks!

[Mr. Brownstone]

I just upgraded my Rockbox build from some early/mid-2006 custom patched-version to the latest in the Current Build section. What a difference! The new menu layout is great, and easy to get accustomed to thanks to the icons.

The main reason I upgraded was to hear some classic NES music on my H300. The only track I've listened to so far is the NSF from Solstice: Quest for the Staff of Demnos which I have rather a soft spot for, But there are a couple of issues with playback:

1. The "bell" noise that plays throughout track 0:01 (in-game music) is not true to the original sound, and jitters occasionally, so there's a bug in the instrument playback somewhere.

2. Tracks 0:05 and 0:06 (title-screen music) are quite long, and fade-out before they're finished. Could there be an option to change this limit in the Sound settings possibly? The title-screen music is the best bit!

Other than these issues NSF playback of the Solstice music is flawless, which is excellent considering how inaccurate I've heard it through some emulators.