Author Topic: 3.8 reports, queries and bugs  (Read 125739 times)

Dotpitch

  • Posts: 2871
Re: 3.8 reports, queries and bugs
« Reply #450 on: 1 Feb '17 - 06:49 »
What output were you using in AIMP? DirectSound, ASIO or WASAPI? Does the Windows mixer also show problems? And WASAPI in exclusive mode?

Discarding the first 0.5 seconds is indeed an annoying habit. What kind of DAC is it?

moriez

  • Posts: 115
Re: 3.8 reports, queries and bugs
« Reply #451 on: 1 Feb '17 - 10:03 »
What output were you using in AIMP? DirectSound, ASIO or WASAPI? Does the Windows mixer also show problems? And WASAPI in exclusive mode?

Discarding the first 0.5 seconds is indeed an annoying habit. What kind of DAC is it?

Thanks for responding.

In AIMP I only tried WASAPI exclusive, event and/or push of which I don't know the difference.
I haven't tried Windows mixer yet but shall play with this tomorrow.
The DAC is a Chord Mojo that is connected over USB.

Theoretically, could it have to do with that AIMP does resampling by default? As far as I can see it can't be turned off.

Dotpitch

  • Posts: 2871
Re: 3.8 reports, queries and bugs
« Reply #452 on: 1 Feb '17 - 17:02 »
Theoretically, could it have to do with that AIMP does resampling by default?
Could be, but you can make XMPlay do that as well. First try the Windows mixer.

Ian @ un4seen

  • Administrator
  • Posts: 20437
Re: 3.8 reports, queries and bugs
« Reply #453 on: 1 Feb '17 - 17:14 »
I'm  trying a new DAC that has the annoying habit to silence the first 0.5 second of a new track. Even when I load the entire album. I think I tried every single option under Output but can't correct it. Both WASAPI output set to 24 bit/44.100 (also tried 16 bit) with pretty much my entire music library being 16bit/44.100. Setting buffer to as low as 0.100s doesn't correct it either. Then, as I've been curious try out AIMP, this player has no issues at all.

Does the problem only happen when using exclusive mode WASAPI output, not when using shared mode WASAPI or the default (waveout) or DirectSound outputs? And does it only happen when starting playback, or does it also happen when playback transitions from one file to another (with the same sample format), ie. is gapless playback affected?

moriez

  • Posts: 115
Re: 3.8 reports, queries and bugs
« Reply #454 on: 3 Feb '17 - 20:00 »
Had a little time to test but too fast the battery drained. What I can report though is that the Windows mixer cuts off the same way. Tested this on Bandcamp and YT.

moriez

  • Posts: 115
Re: 3.8 reports, queries and bugs
« Reply #455 on: 4 Feb '17 - 19:30 »
Apologies. Although the problem is with ALL output options I wasn't entirely correct about the issue. AIMP does cut off the first 0.5 second :\ .. with one difference to it:

XMPlay idle in tray and three tracks are queued. I press play and the cutoff is there. The one exception is gapless playback (so without pressing any of the controls ie. 'next track')
vs
AIMP idling in tray and the same three tracks are queued. I press play and the the cutoff is there but when I use the controls ie. 'next track' it does NOT cut off.

In this scenario AIMP isn't really benificial to my situation.


Edit: This gent on Head-Fi seems to be on to something. Play silence at startup for hardware synchronization
http://www.head-fi.org/t/784602/chord-mojo-dac-amp-faq-in-3rd-post/7245#post_12179712
« Last Edit: 4 Feb '17 - 20:31 by moriez »

saga

  • Posts: 2183
Re: 3.8 reports, queries and bugs
« Reply #456 on: 19 Mar '17 - 12:40 »
I think something is wrong with S3M "Fast Volume Slides". In the attached excerpt from Purple Motion's PANIC.S3M, the sound on row 6 should be about half as loud as the sound on row 8 (same later on row 14 / 16 and 22 / 24).

Ian @ un4seen

  • Administrator
  • Posts: 20437
Re: 3.8 reports, queries and bugs
« Reply #457 on: 24 Mar '17 - 16:43 »
That problem appears to be caused by XMPlay processing the effect column before the volume column in S3M files, eg. the 1st channel's volume after the 1st tick of row 6 is 10 instead of 0 and the 2nd channel's volume is 25 instead of 13. XMPlay did originally process the volume column first (and still does for XM/MOD/MTM files), but that was changed for S3M/IT files in Dec 2013 for some reason. Unfortunately, I didn't make a note of why. Anyway, here's an update that processes the volume column first again:

   www.un4seen.com/stuff/xmplay.exe

Let me know if you find any other S3M or IT files don't sound right any more :)

Dotpitch

  • Posts: 2871
Re: 3.8 reports, queries and bugs
« Reply #458 on: 24 Mar '17 - 18:08 »
I think you changed it here because of this test file. Perhaps Impulse Tracker and Scream Tracker are slightly different in this regard.

saga

  • Posts: 2183
Re: 3.8 reports, queries and bugs
« Reply #459 on: 24 Mar '17 - 19:43 »
Thanks for the fix!
The "set volume" command in the volume column is "special" and always supposed to be processed first (also in IT), and since Scream Tracker has no other commands to which the IT special case would apply for, this fix is definitely wrong for S3M and only valid for IT. :)
« Last Edit: 24 Mar '17 - 20:00 by saga »

Ian @ un4seen

  • Administrator
  • Posts: 20437
Re: 3.8 reports, queries and bugs
« Reply #460 on: 27 Mar '17 - 16:17 »
I think you changed it here because of this test file. Perhaps Impulse Tracker and Scream Tracker are slightly different in this regard.

I do see a little code modification that was made for that 25.IT file, but I'm not sure if the volume/effect processing order change was also for that file. Does anyone happen to still have the 25.IT file to try? The 25.IT file that I have seems to sound the same in XMPlay and Impulse Tracker even after switching back to processing the volume column first, but I'm not sure if it's the original version of the file (I might have modified it while testing).

saga

  • Posts: 2183
Re: 3.8 reports, queries and bugs
« Reply #461 on: 27 Mar '17 - 17:56 »
The original files can be found here: https://resources.openmpt.org/player_tests/it/player-abuse-tests/
Test case 25 does indeed test volume vs effect column priority.

Ian @ un4seen

  • Administrator
  • Posts: 20437
Re: 3.8 reports, queries and bugs
« Reply #462 on: 28 Mar '17 - 16:19 »
Thanks. I see that I had indeed modified the file, and there was a difference after switching back to the original. Here's an update that seems to fix that problem without changing the volume/effect column processing order:

   www.un4seen.com/stuff/xmplay.exe

As always, let me know if you find it breaks any other IT files :)

Knurek

  • Posts: 523
Re: 3.8 reports, queries and bugs
« Reply #463 on: 1 Apr '17 - 19:54 »
I am using Wave Mapper for my output device, and have standard speakers plugged in the sound card (with a 3.5mm jack). Recently I bought a pair of bluetooth headphones, and while they work fine with XMPlay, there's a bit of wonkery going on while switching devices.

Say, I have a paused playlist and am using the speakers. The second I switch to BT, XMPlay will start playing on its own, and what's weirder, the global hotkeys will stop working for a single press. Same thing will happen when switching from BT back to speakers.

Is this the intended behaviour? Windows being weird? Or is it something that can be corrected?

Dotpitch

  • Posts: 2871
Re: 3.8 reports, queries and bugs
« Reply #464 on: 2 Apr '17 - 10:02 »
The second I switch to BT, XMPlay will start playing on its own, and what's weirder, the global hotkeys will stop working for a single press. Same thing will happen when switching from BT back to speakers.
Is this the intended behaviour? Windows being weird? Or is it something that can be corrected?
I remember this from a few years ago, apparently it's just Windows being weird.
If I pause playback and then the audio output is changed, xmplay continues playing track. It happens when I listen to music in my bluetooth headphones, pause and then switch off headphones. I click "play/pause" one time - it is changed on "playing" and nothing else happened, and after second click - playback is paused again.
Is it a bug?
I recall this coming up before. It isn't particularly related to bluetooth headphones, but rather the default output device changing on Windows 7. When that happens, and XMPlay is using the default device ("Microsoft Sound Mapper"), Windows is automatically resuming the paused output on the new device. DirectSound output appears to be unaffected though, so an equivalent default device option was added to the DirectSound plugin.
Windows resumes playback, but XMPlay still thinks it's paused, so that might affect the global hotkeys as well. Could you try with DirectSound?

winner

  • Posts: 260
I am getting the following message when launching XMPlay version 3.8.2.22:

Version mismatch
The XMPLAY.LIBRARY was written by and requires a newer version of XMPlay. To avoid losing the library contents, the library will not be read or written.

Please help to resolve. Thanks!

EDIT: I renamed my xmplay.library file to xmplay.library.old, then closed and reopened XMPlay. It generated a new xmplay.library file and I have no more error.
« Last Edit: 10 Apr '17 - 02:30 by winner »

Ian @ un4seen

  • Administrator
  • Posts: 20437
Re: 3.8 reports, queries and bugs
« Reply #466 on: 10 Apr '17 - 17:57 »
That's strange. Perhaps the XMPLAY.LIBRARY file got corrupted. Did the problem just suddenly start happening for no apparent reason, eg. XMPlay closed normally the previous time?

If you still have the file, please upload it to have a look at here:

   ftp.un4seen.com/incoming/

winner

  • Posts: 260
Re: 3.8 reports, queries and bugs
« Reply #467 on: 11 Apr '17 - 23:27 »
That's strange. Perhaps the XMPLAY.LIBRARY file got corrupted. Did the problem just suddenly start happening for no apparent reason, eg. XMPlay closed normally the previous time?

If you still have the file, please upload it to have a look at here:

   ftp.un4seen.com/incoming/
I've uploaded the file.

Ian @ un4seen

  • Administrator
  • Posts: 20437
Re: 3.8 reports, queries and bugs
« Reply #468 on: 17 Apr '17 - 16:32 »
The uploaded library file seems to be fine. Are you sure it is the file that the problem was happening with, eg. if you rename it back to XMPLAY.LIBRARY, does the problem return?

winner

  • Posts: 260
Re: 3.8 reports, queries and bugs
« Reply #469 on: 18 Apr '17 - 01:43 »
The uploaded library file seems to be fine. Are you sure it is the file that the problem was happening with, eg. if you rename it back to XMPLAY.LIBRARY, does the problem return?
Yes, that was the file that caused the error. I renamed it back to XMPLAY.LIBRARY and the problem returned.

Ian @ un4seen

  • Administrator
  • Posts: 20437
Re: 3.8 reports, queries and bugs
« Reply #470 on: 18 Apr '17 - 13:33 »
In that case, are you sure that you're using XMPlay 3.8.2? The error message you saw would be shown if you tried to go back to an older XMPlay version because it doesn't know how to handle the newer library file format. Please check the XMPlay version number in the bottom-right corner of the options window.

saga

  • Posts: 2183
Re: 3.8 reports, queries and bugs
« Reply #471 on: 20 May '17 - 16:40 »
Attached is a very simple MOD... apparently too simple for the auto-normalization! ;D
Using pretty much normal MOD playback settings (with 50:50 pan sep), the playback is distorted on the left channel.

Ian @ un4seen

  • Administrator
  • Posts: 20437
Re: 3.8 reports, queries and bugs
« Reply #472 on: 25 May '17 - 17:10 »
Here's an update with slightly tweaked module playback levels:

   www.un4seen.com/stuff/xmplay.exe

Leras

  • Guest
Re: 3.8 reports, queries and bugs
« Reply #473 on: 5 Jun '17 - 04:53 »
Here is the issue. If I click on MP3 file the player starts in background.

XMPlay 3.8.2
On top options are turned off
Under Windows 7 SP1

Under Windows XP all works OK.

Ian @ un4seen

  • Administrator
  • Posts: 20437
Re: 3.8 reports, queries and bugs
« Reply #474 on: 8 Jun '17 - 17:00 »
Here's an update for you to try:

   www.un4seen.com/stuff/xmplay.exe