Whoa, that was fast! I assume you mean polling the IRQ flag, AKA the Deflemask trick?Karmic wrote:It's the typical trick where APU frame IRQs are used to simulate the 60Hz PPU frame NMI. Both use non-returning play.
More NSF Requests
Moderator: Moderators
Re: More NSF Requests
Re: More NSF Requests
Yep. It's unfortunately the only real way to do this kind of timing with the limitations of NSF.Rahsennor wrote:Whoa, that was fast! I assume you mean polling the IRQ flag, AKA the Deflemask trick?Karmic wrote:It's the typical trick where APU frame IRQs are used to simulate the 60Hz PPU frame NMI. Both use non-returning play.
Re: More NSF Requests
Well drat. I pass all the test ROMs for the IRQ flag so it looks like this one's going to be a headache.Karmic wrote:Yep. It's unfortunately the only real way to do this kind of timing with the limitations of NSF.
Thanks for the ultra-speedy answers.
- rainwarrior
- Posts: 8734
- Joined: Sun Jan 22, 2012 12:03 pm
- Location: Canada
- Contact:
Re: More NSF Requests
It uses the "deflemask" technique of disabling IRQ and polling $4015 for the frame counter's IRQ flag.Rahsennor wrote:Thank you, good sir! I've been waiting for a proper rip of this game.Karmic wrote:The 100% complete rip of Battletoads, with the PCM title and victory tunes, pause tune, sfx, and PCM sfx.
Now if I could just figure out why it plays way too fast in my NSF player... and only my NSF player. It does the same thing on your Skate Or Die 2 rip. What are you using for timing on the PCM tracks?
Anyhow, yeah I'm also glad to see this! I wanted to do this very thing for a long time, but it was never high enough on my priorities to get it done. Thank you!
Edit: whoops, missed the last page, you've already discussed it.
Re: More NSF Requests
Great job, especially because now we can listen to the samples without any NMI or OAM DMA cutting into the sound like in-game.
Re: More NSF Requests
Definitely seems buggy that these two NSFs rips — Skate or Die 2 and Battletoads — don't work as intended in a more complete implementation of the APU, but work when turning off a few lines of code for 4017 and the frame counter, lines needed for the test ROMs to pass. (I added an option in my emulator to get them to work by doing just that— turning off those lines.) They work in FCEUX because its APU has never been updated for authentic frame timing.Rahsennor wrote:Well drat. I pass all the test ROMs for the IRQ flag so it looks like this one's going to be a headache.
However, the battletoadsdoubledragon_pcm NSF works with the more complete APU.
Last edited by ap9 on Sat Jun 24, 2017 10:55 pm, edited 1 time in total.
Re: More NSF Requests
You're right. I didn't know the APU frame-flag was set thrice in a row :pap9 wrote:Definitely seems buggy that these two NSFs rips — Skate or Die 2 and Battletoads — don't work as intended in a more complete implementation of the APU, but work when turning off a few lines of code for 4017 and the frame counter, lines needed for the test ROMs to pass. (I added an option in my emulator to get them to work by doing just that— turning off those lines.) They work in FCEUX because its APU has never been updated for authentic frame timing.Rahsennor wrote:Well drat. I pass all the test ROMs for the IRQ flag so it looks like this one's going to be a headache.
However, the battletoadsdoubledragon_pcm NSF works with the more complete APU.
Fixed to acknowledge this, it works fine in Nintendulator now at least.
- Attachments
-
- Skate or Die 2.nsf
- (113.86 KiB) Downloaded 379 times
-
- Battletoads.nsf
- (31.56 KiB) Downloaded 396 times
Re: More NSF Requests
Much better in repeat, but still not quite fixed. If you restart track 1 of Battletoads repeatedly you'll find the timing wobbles a bit with each playback, sometimes by quite a lot.
Could still be a bug in my player, of course.
(Also ap9 that's me you quoted not Karmic.)
Could still be a bug in my player, of course.
(Also ap9 that's me you quoted not Karmic.)
Re: More NSF Requests
Yeah, the timing/tempo is slightly inconsistent (in Nintendulator, but fine in others). An improvement nonetheless.Rahsennor wrote:Much better in repeat, but still not quite fixed. If you restart track 1 of Battletoads repeatedly you'll find the timing wobbles a bit with each playback, sometimes by quite a lot.
My reply post began one quote in another, and I deleted the wrong element. My bad.Rahsennor wrote:(Also ap9 that's me you quoted not Karmic.)
-
- Posts: 63
- Joined: Tue Jul 05, 2016 10:53 am
Re: More NSF Requests
i think the timing is off somehow in that blood of jurassic .nsf...if you listen to track 5, the dpcm channel is, well i dunno how to explain it really, but theres some sort of drift going on, even after one rep through...maybe its a pal/ntsc issue or something? im not sure, but you can hear that it doesnt sync up with the tridrops in the song..i havent listened to enough of it to be able to say if it affects other tracks or not, but maybe someone can confirm this, or that its supposed to do this? ahaha, i donno.
- MrNorbert1994
- Posts: 988
- Joined: Tue May 06, 2014 1:50 pm
- Location: Szeged, Hungary
- Contact:
Re: More NSF Requests
Ripped the PCM samples for this.
- Attachments
-
- Mahjong Club - Nagatacho Sousaisen (SFX).nsf
- (288.13 KiB) Downloaded 369 times
- MrNorbert1994
- Posts: 988
- Joined: Tue May 06, 2014 1:50 pm
- Location: Szeged, Hungary
- Contact:
Re: More NSF Requests
A generic SFX rerip.
- Attachments
-
- Star Trek - The Next Generation (SFX).nsf
- (16.14 KiB) Downloaded 366 times
- MrNorbert1994
- Posts: 988
- Joined: Tue May 06, 2014 1:50 pm
- Location: Szeged, Hungary
- Contact:
Re: More NSF Requests
Added SFX and jingles to this one.
- Attachments
-
- Chuck Yeager's Fighter Combat (SFX).nsf
- (32.08 KiB) Downloaded 365 times
Re: More NSF Requests
Mahjong Club - Nagatacho Sousaisen (SFX) crashes or winds up at a bad address when a PCM sound effect completes. (Some players handle it better.)
Re: More NSF Requests
Regarding Mahjong Club - Nagatacho Sousaisen, at the end of playing PCM the RTS goes to a routine that ends in an RTI and then pushes the PC to $2022; which is constantly updating RAM; and eventually reads illegal opcodes and crashes some players. I'm looking into fixing it.