apolkosnik[m] | ACTION uploaded an image: (440KiB) < https://libera.ems.host/_matrix/media/r0/download/matrix.org/cJuOGtVoMwARRxoqOAENMxmF/PXL_20220420_223909689.jpg > | 00:51 |
---|---|---|
apolkosnik[m] | So, it looks like the AHI-Noteplayer is looking for a realtime mode, when Use best audio mode is checked. | 00:52 |
minute | should be a realtime mode though, weird | 00:53 |
minute | apolkosnik[m]: do you use the -tw driver version? | 00:54 |
apolkosnik[m] | yes | 00:54 |
apolkosnik[m] | it plays much better | 00:54 |
apolkosnik[m] | I was trying to set up that octamed sound studio earlier, just couldn't find any octamed modules | 00:55 |
- xet7 (QUIT: Remote host closed the connection) (~xet7@user/xet7) | 03:36 | |
+ xet7 (~xet7@user/xet7) | 03:37 | |
- Jadek (QUIT: Quit: Connection closed) (~Jadek@87-207-204-28.dynamic.chello.pl) | 08:33 | |
neoman | apolkosnik[m], I got it running with AHI Noteplayer but you have to choose the driver manually | 08:44 |
+ mnemotron (mnemotron@194.135.47.239) | 09:45 | |
- xet7 (QUIT: Remote host closed the connection) (~xet7@user/xet7) | 11:05 | |
+ xet7 (~xet7@user/xet7) | 11:06 | |
minute | here is a testcase for ZZ9000AX, if you have problems with distortion or stuttering (?), you can help me https://www.a1k.org/forum/index.php?threads/83561/page-7#post-1587500 | 12:58 |
minute | it's in german but i think the list is readable as english | 12:58 |
minute | i notice that at the beginning in the very high freqs, there's some harmonics that might be clipping hmm. when i turn the volume in songplayer down to 8 they're gone | 13:01 |
minute | ah yeah, i think it happens in the 24->48 upsampling in zz firmware | 13:05 |
minute | http://dump.mntmn.com/sweep48.wav this sounds fine | 13:05 |
minute | _Bnu: do you happen to have something better than this? https://source.mnt.re/amiga/zz9000-firmware/-/blob/master/ZZ9000_proto.sdk/ZZ9000OS/src/ax.c#L513 | 13:13 |
+ gdonner (gdonner@c-98-253-93-116.hsd1.in.comcast.net) | 16:16 | |
gdonner | minute: Both files played ok here, on both 24 and 48 Hz AHI settings | 16:17 |
gdonner | My setting are as follows: | 16:17 |
gdonner | firmware and drivers v1.10 | 16:17 |
gdonner | Kickstart 47.102, Workbench 47.3 (OS 3.2.1) | 16:18 |
gdonner | 68040 (A3640, rev. 3.2, recapped/fixed polarities) | 16:18 |
gdonner | Only MuFastROM running | 16:18 |
gdonner | 68040.library v46.4, mmu.library 47.2 | 16:18 |
gdonner | This is on my A3000 | 16:19 |
gdonner | Oddly enough, Play16 (from Thomas Wenzel) seems to play the sweep48.wav more accurately than SongPlayer does | 16:19 |
gdonner | I'll try all the above on my A4000 and report back | 16:20 |
gdonner | (this is using the -ahi-tw.lha ZZ9000AX.audio file with a reboot) | 16:20 |
gdonner | there were a few clicks on first play of the sweep24.wav file | 16:21 |
gdonner | but I believe that's normal | 16:21 |
apolkosnik[m] | is there a way to quiet the spikes on the startup (when the machine is powered up, then couple of spikes while it's booting up)? | 16:21 |
gdonner | no distortion, no stuttering | 16:22 |
gdonner | minute: ok, both files played fine on my A4000/060 as well | 16:42 |
gdonner | specs: same as the A3000, except A3660 (50 MHz) | 16:43 |
gdonner | 68060.library v47.1, mmu.library v47.4 | 16:43 |
gdonner | MuFastROM also running | 16:43 |
gdonner | hope this helps! and confirmed: Thomas Wenzel's Play16 plays audio better than SongPlayer :) | 16:44 |
gdonner | (more complete, more detail) | 16:45 |
minute | oh, play16 | 16:51 |
minute | apolkosnik[m]: will be in next fw rev | 16:51 |
minute | gdonner: thank you very much for testing! so it seems to work as designed then on your machine | 16:52 |
minute | i already know that the resampling (upsampling to 48khz) in the current firmware is not ideal, as it is missing a low pass filter stage | 16:52 |
minute | so if you play 44.1khz stuff and it is resampled in the card, this will sound sub optimal | 16:53 |
gdonner | sure--happy to help! | 16:55 |
gdonner | BTW, thanks for all your work on this (and Thomas W. too) | 16:56 |
apolkosnik[m] | awesome, thank you! | 16:56 |
apolkosnik[m] | Hippo is doing much better once I bumped Priority in PS3M settings to +1 | 16:58 |
minute | aha? | 16:59 |
minute | what exactly is much better? no more stuttering? (did you have stuttering before)? | 16:59 |
minute | btw i had to set the AHI volume in hippo to 500 | 16:59 |
minute | instead of 1000 | 16:59 |
minute | or it would be distorted | 16:59 |
- mnemotron (QUIT: Quit: Quit) (mnemotron@194.135.47.239) | 17:23 | |
- gdonner (QUIT: ) (gdonner@c-98-253-93-116.hsd1.in.comcast.net) | 18:17 | |
_Bnu | minute: My resampling is probably worse, quality-wise. | 19:13 |
minute | _Bnu: oh, ok? | 19:13 |
_Bnu | https://github.com/captain-amygdala/pistorm/blob/main/platforms/amiga/ahi/pi_ahi.c#L282-L305 | 19:14 |
_Bnu | I thought it sounded fine, but others told me it sounded bad, so I ignored it forever...! | 19:15 |
_Bnu | I was just happy it worked and didn't desynch to hell and beyond, lolb. | 19:15 |
minute | _Bnu: looks very similar to mine | 19:19 |
minute | _Bnu: a physicist told me that it might just be missing a low pass filter | 19:19 |
_Bnu | My idea on how to make it "better" was to render it as a texture, using bilinear filtering or something. | 19:19 |
_Bnu | Abuse the Pi GPU to smooth out the waves, 16-bit luma. | 19:19 |
minute | funky | 19:20 |
_Bnu | I figure you could probably even do some sort of mixing that way, with some alpha channel blender setting. | 19:23 |
minute | yep | 19:25 |
Jope | phew, ups didn't tell me that the ax arrived on the 12th :-P | 20:23 |
Jope | by chance I checked the tracking today and it would have gone back tomorrow | 20:24 |
Jope | delivering parcels: difficult | 20:24 |
minute | oh :D | 20:27 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!