NSFPlay 2.2

Post by **lidnariq** » Wed Nov 28, 2012 11:34 am

Sorry, I stated that badly. What I meant was: The problem is not the CPU speed in an Android device (although it may be), but rather that a CPU-intensive process is going to suck the battery dry very quickly

TmEE · Post by **TmEE** » Sat Dec 01, 2012 5:45 pm

Loop count does not seem to work for me in winamp. I have set it to 8 and all tunes still seem to fade after 2 loops. This happens on 2.2 and the 2.3b
EDIT: 2.3b seems to have it work on Bubble Bobble 2 NSF, time is not right in the playlist but the song continues to play after end has reached (and timer goes into positive, when winamp has countdown display mode).
Power Blade 2 is completely messed up on 2.3b, I am häppy that my amp wos not turned up any higher... haha
.....aaaand i have been loading playlist files in winamp all this time.... PowerBlade2 is still completely messed up

rainwarrior · Post by **rainwarrior** » Sat Dec 01, 2012 9:08 pm

Can you paste a nonworking playlist entry for me? The way the loop logic works is kinda weird, I'm wondering if there's a bug or I just need to document it better. It should work like this:

filename::NSF,[song],[title],[time],[loop],[fade],[loopcount]

[time] = length of intro + one complete loop
[loop] = length of loop (or if it ends with a -, start time of loop. if using - the [time] parameter must be filled)
[loopcount] = number of times to loop

Note the if your [time] and [loop] aren't set to match the song the loopcount isn't going to help. (I personally find it easier to leave loop and loopcount as default and just set the time for the length I want.)

Can you attach or link to the Power Blade 2 NSF that isn't working for you? I can find 3 rips of it and they all work fine (though one of them has VRC6 and VRC7 enabled for some reason).

TmEE · Post by **TmEE** » Sat Dec 01, 2012 10:20 pm

Messed up sound : http://www.tmeeco.eu/BitShit/Power%20Bl ... Taito).nsf
Plays (but sounds like something is missing, maybe JP is like that...?) http://www.tmeeco.eu/BitShit/Power%20Bl ... Taito).nsf

The time setting seems to take priority when I play NSF directly, I got a nice 10 minute playtime and it is good enough for me.

rainwarrior · Post by **rainwarrior** » Sat Dec 01, 2012 10:28 pm

Hmm, both of those are playing fine for me in NSFPlay 2.3 beta 2. Can anyone else duplicate the problem?

TmEE · Post by **TmEE** » Sat Dec 01, 2012 10:36 pm

This is what I am getting in Winamp (~1MB file) http://www.tmeeco.eu/BitShit/Power%20Blade%202.wav

The standalone player has no problems though.

Could Winamp version make a difference ? I got 5.35 installed.

rainwarrior · Post by **rainwarrior** » Sat Dec 01, 2012 10:46 pm

Hmm, I've got 5.63... the version shouldn't make a difference (in theory), since the plugin is a Winamp 2 plugin.

That's pretty weird though. I'm not having any problem playing that file in Winamp with NSFPlug. I'm not sure what to suggest.

koitsu · Post by **koitsu** » Sat Dec 01, 2012 11:22 pm

TmEE wrote:Messed up sound : http://www.tmeeco.eu/BitShit/Power%20Bl ... Taito).nsf
Plays (but sounds like something is missing, maybe JP is like that...?) http://www.tmeeco.eu/BitShit/Power%20Bl ... Taito).nsf

Both NSFs play fine for me using Winamp 5.63 on Windows XP + NSFPlay 2.3b2. I don't experience what you do here. If you want me to make a video of what I experience just let me know.

Have you tried deleting your in_yansf.ini from your Plugins directory and using the one included with 2.3b2? This might explain your looping oddities/behaviour/etc. too, but I'm not sure.

If you roll back to a previous NSFPlay release (and if so which one), does the problem go away?

The sound difference you hear between the two versions is that the strong bass/drum line is missing in the "JP" version. To me, that would indicate whoever did the ripping probably didn't do the right thing (some APU registers not being initialised correctly, etc.); AFAIK the two versions have the exact same music.

Edit: With regards to the sound difference between the two NSFs -- it's related to the DMC channel (which is used for the drum/bass line). Someone either ripped the NSF wrong or is initialising some APU registers wrong.

koitsu · Post by **koitsu** » Sat Dec 01, 2012 11:28 pm

Found another bug pertaining to the visualiser/vu metres/etc., rainwarrior. Try using a sample rate that's lower than 44100Hz (try 8000Hz) under Playback / Sample Rate -- the metres are "ahead" of the actual audio playback by quite a lot. The buffer/delta gets smaller and smaller the closer to 44kHz you get.

TmEE · Post by **TmEE** » Sat Dec 01, 2012 11:57 pm

I just tried 2.2 and it plays nicely, no problems.
And now I copied over 2.3 in its entirety and seems there is no problem anymore.
I cannot remember if I overwrote the settings file last time or not... maybe not then ?

Bottomline is things work now ^^

rainwarrior · Post by **rainwarrior** » Sun Dec 02, 2012 12:21 am

koitsu wrote:Found another bug pertaining to the visualiser/vu metres/etc., rainwarrior. Try using a sample rate that's lower than 44100Hz (try 8000Hz) under Playback / Sample Rate -- the metres are "ahead" of the actual audio playback by quite a lot. The buffer/delta gets smaller and smaller the closer to 44kHz you get.

Hmm, well the delay for the keyboard view is configurable (click the "Settings..." button). I've tried to tune the default for 44.1kHz on my own setup, but I am not confident that audio latency will be consistent from user to user even if they use the same samplerate as me. I could try to make the delay setting proportional to samplerate, though that might make it confusing (what unit would I use? right now it says ms, which I think is clear), so I'm not sure it would be worthwhile to do that.

koitsu · Post by **koitsu** » Sun Dec 02, 2012 8:50 am

rainwarrior wrote:
koitsu wrote:Found another bug pertaining to the visualiser/vu metres/etc., rainwarrior. Try using a sample rate that's lower than 44100Hz (try 8000Hz) under Playback / Sample Rate -- the metres are "ahead" of the actual audio playback by quite a lot. The buffer/delta gets smaller and smaller the closer to 44kHz you get.
Hmm, well the delay for the keyboard view is configurable (click the "Settings..." button). I've tried to tune the default for 44.1kHz on my own setup, but I am not confident that audio latency will be consistent from user to user even if they use the same samplerate as me. I could try to make the delay setting proportional to samplerate, though that might make it confusing (what unit would I use? right now it says ms, which I think is clear), so I'm not sure it would be worthwhile to do that.

Hm, maybe I wasn't clear in my explanation -- by visualiser/vu metres I'm referring to Winamp's. It's labelled "spectrum analyser". We had a conversation about this problem before, where the issue turned out to be due to the sample size being fed to Winamp was too big so it resulted in a delay between the spectrum analyser results and what you actually hear. What I'm saying is that this problem (the effect itself) happens if you use a sample rate in NSFPlug that's lower than 44kHz.

Question: why is there an intentional delay (referring to the Delay setting under Keyboard) of any sort to begin with? What purpose does this have?

Jarhmander · Post by **Jarhmander** » Sun Dec 02, 2012 10:42 am

I guess it's sort of a cheap audio-video manual syncing facility.

rainwarrior · Post by **rainwarrior** » Sun Dec 02, 2012 12:52 pm

Yeah, it compensates for the audio latency. Ideally I would use something like waveOutGetPosition and synchronize, but since it's a winamp plugin, the plugin doesn't have any direct access to the audio system (it just fills a buffer on demand).

As for Winamp's visualizer being out of synch at other samplerates, I have no idea what to do about that. The last time the fix was to make sure the buffer length was 576 samples, which Winamp likes best for historical reasons or something. I dunno what to do if it isn't synched at other samplerates...

B00daW · Post by **B00daW** » Sat Dec 08, 2012 5:39 am

Found another bug in NSFPlay 2.3 I believe... If you download the newest Famicompo archive, there is a song that is all DPCM in the originals section. NSFPlay doesn't detect it playing anything and fades out after only a few seconds. Since I'm at work, I remember it being about a man saying "I'm hungry..." and him eating various things.

nesdev.org

NSFPlay 2.2

Re: NSFPlay 2.2

Re: NSFPlay 2.2

Re: NSFPlay 2.2

Re: NSFPlay 2.2

Re: NSFPlay 2.2

Re: NSFPlay 2.2

Re: NSFPlay 2.2

Re: NSFPlay 2.2

Re: NSFPlay 2.2

Re: NSFPlay 2.2

Re: NSFPlay 2.2

Re: NSFPlay 2.2

Re: NSFPlay 2.2

Re: NSFPlay 2.2

Re: NSFPlay 2.2