Author Topic: 3.8 reports, queries and bugs  (Read 124618 times)

Ian @ un4seen

  • Administrator
  • Posts: 20389
Re: 3.8 reports, queries and bugs
« Reply #225 on: 17 Oct '14 - 17:40 »
The thing is the sample loading is performed separately in XMPlay, ie. there's an IT file loading function (and other functions for XM/MOD/etc) which retains the sample file offsets, and then there's a sample loading function that's called afterwards if the file is being opened for playback (not just info scanning). Anyway, although inelegant, I think the current solution should work fine :)

saga

  • Posts: 2179
Re: 3.8 reports, queries and bugs
« Reply #226 on: 17 Oct '14 - 17:59 »
Indeed, for now it should be sufficient. I suppose to reduce false positives even further, one could also search for the "STPM" and only start looking for ".FSM" once these have been located.

Ian @ un4seen

  • Administrator
  • Posts: 20389
Re: 3.8 reports, queries and bugs
« Reply #227 on: 20 Oct '14 - 13:58 »
Good idea. I'll make it do that in the next update.

rst

  • Posts: 309
Re: 3.8 reports, queries and bugs
« Reply #228 on: 21 Oct '14 - 15:38 »
my idea:

to show the playlist with adjustable transparency over the VIS window.
to show the track title during at least 10 seconds when we use the VIS window, i am not saying full screen !

xaos

  • Posts: 172
Re: 3.8 reports, queries and bugs
« Reply #229 on: 28 Oct '14 - 07:10 »
As much as I hate to say anything ..

Win7, WASAPI default, auto-amp off, EQ on, .. volume *(output level)* fluctuates continuously as if the auto-amp were on steroids
« Last Edit: 28 Oct '14 - 07:26 by xaos »

Ian @ un4seen

  • Administrator
  • Posts: 20389
Re: 3.8 reports, queries and bugs
« Reply #230 on: 28 Oct '14 - 15:40 »
A few questions to help narrow it down... Does the issue occur when exclusive mode is enabled and/or disabled? Does it only happen when playing particular files? If so, is it at very loud moments? Does the same thing happen when using XMPlay's default output system with the "Resolution" set to 32 bit?

xaos

  • Posts: 172
Re: 3.8 reports, queries and bugs
« Reply #231 on: 28 Oct '14 - 21:10 »
No particular file, or any particular point in a file, it happens with anything that's playing.

It's "continuous" fluctuation, from beginning to end, as in, slightly louder, slightly quieter, slightly louder, slightly quieter, slightly louder, slightly quieter, ...  ...  ... as if the auto-amp were constantly being tweaked up and down, but it's OFF.

WASAPI exclusive mode doesn't fluctuate, Direct Sound and ASIO also sound correct.

Microsoft sound mapper, Speakers (soundcard), WASAPI non-exclusive :::  fluctuate.

The fluctuating only happens with the EQ turned ON. All other DSP settings are off, including the auto-amp.
« Last Edit: 28 Oct '14 - 21:25 by xaos »

saga

  • Posts: 2179
Re: 3.8 reports, queries and bugs
« Reply #232 on: 28 Oct '14 - 22:27 »
It would be interesting to answer the last question about 32-bit, because the mixer in Windows Vista and later (when not using exclusive playback mode) has some kind of auto-compression happening on 32-bit floating-point audio channels if they're too loud (even if they are then scaled down in the mixer).
« Last Edit: 29 Oct '14 - 08:43 by saga »

Ian @ un4seen

  • Administrator
  • Posts: 20389
Re: 3.8 reports, queries and bugs
« Reply #233 on: 29 Oct '14 - 17:30 »
Yep, that does sound like it's Windows built-in limiter kicking in. When XMPlay's sends 32-bit output through Windows (not ASIO or exclusive mode WASAPI), the sample data is unclipped floating-point, so it can go beyond 0 dB. When that happens, Windows (Vista and newer) will apply a limiter rather than clipping. In this case, the EQ is probably taking the level too high at times, and Windows' limiter is bringing it back down. To confirm whether that's the case, does the fluctuating disappear (possibly replaced by distortion) if you use the default output system (eg. "Microsoft Sound Mapper") with "Resolution" set to 24 bit instead of 32 bit? Shared mode WASAPI output is always 32-bit floating-point regardless of the "Resolution" setting, so it isn't possible to avoid Windows' limiter then.

xaos

  • Posts: 172
Re: 3.8 reports, queries and bugs
« Reply #234 on: 30 Oct '14 - 10:40 »
well, learn something new, flipping ms and their attempting to do everything.

The 24bit doesn't seem to suffer the fluctuation problem, but it does seem to suffer a bit of distortion .. and generally doesn't seem to (sound) as good.

.. only recently finally switched to Windows 7, had no idea they were now messing with the output like that. Can't say I remember having issues of any sort like that with XP, .... just assumed the output was handled by the APP and the sound card drivers ..

Seems the option to avoid windows interference is to use ASIO or exclusive mode WASAPI ..

Thanks for the help.

piovrauz

  • Posts: 967
Re: 3.8 reports, queries and bugs
« Reply #235 on: 30 Oct '14 - 11:14 »
ASIO would be the best if your card has an actual ASIO "device", but I guess ASIO4all works too. ;)

saga

  • Posts: 2179
Re: 3.8 reports, queries and bugs
« Reply #236 on: 30 Oct '14 - 16:27 »
The 24bit doesn't seem to suffer the fluctuation problem, but it does seem to suffer a bit of distortion .. and generally doesn't seem to (sound) as good.
Well, you have three options:
1) live with the distortion because your EQ or similar settings cause the signal to go above 0dB
2) live with compression because your EQ or similar settings cause the signal to go above 0dB
3) turn down the volume in XMPlay so that it doesn't go above 0dB anymore. That's the most sensible solution.

Generally you will not hear any difference between 24bit and 32bit because both have the same SQNR, so the only quality difference you will be able to notice is because of the clipping artefacts.

piovrauz

  • Posts: 967
Re: 3.8 reports, queries and bugs
« Reply #237 on: 30 Oct '14 - 16:33 »
Are the audio cards actually full 32bit? I mean, internally they mostly are 24 or les... ;)

saga

  • Posts: 2179
Re: 3.8 reports, queries and bugs
« Reply #238 on: 30 Oct '14 - 16:38 »
When talking about "32bit", this usually refers to 32-bit floating point, which has a 24-bit mantissa and thus the same SQNR. However, the extra 8 bits are used for the exponent, which (in basic terms) allows for the signal to higher than 0dB. No soundcard in this world will give you a DAC that can give you true 24-bit precision anyway, and it's not needed since your ears are simply not that sensitive. And in fact, 16-bit audio is just enough for listening.

winner

  • Posts: 260
Re: 3.8 reports, queries and bugs
« Reply #239 on: 30 Oct '14 - 17:00 »
well, learn something new, flipping ms and their attempting to do everything.

The 24bit doesn't seem to suffer the fluctuation problem, but it does seem to suffer a bit of distortion .. and generally doesn't seem to (sound) as good.

.. only recently finally switched to Windows 7, had no idea they were now messing with the output like that. Can't say I remember having issues of any sort like that with XP, .... just assumed the output was handled by the APP and the sound card drivers ..

Seems the option to avoid windows interference is to use ASIO or exclusive mode WASAPI ..

Thanks for the help.
Xaos, you've probably checked this, but have you gone to Start > Control Panel > Sound > Playback tab > click Speakers (soundcard} > Properties > Enhancements tab, and made sure that "Disable all sound effects" is checked, or that "Loudness Equalization" (last item) is unchecked?

saga

  • Posts: 2179
Re: 3.8 reports, queries and bugs
« Reply #240 on: 30 Oct '14 - 17:18 »
That should apply to 24-bit format as well, though.

xaos

  • Posts: 172
Re: 3.8 reports, queries and bugs
« Reply #241 on: 30 Oct '14 - 21:51 »
Yeah, I'd checked all the various settings I could find ... the more odd thing about it is that I'd never noticed any clipping or distortion in XP with the same EQ and volume settings that caused this is Win7. It's rather kind of annoying when the operating system takes it upon itself to mess with something that should be the domain of a specialized APP ... but anyway, thanks for the help and information, hopefully now it should be easier to sort out the details.

saga

  • Posts: 2179
Re: 3.8 reports, queries and bugs
« Reply #242 on: 31 Oct '14 - 10:12 »
There might have been other settings that have been different.  Vista's mixer is considerably better than XP's and it's not "taking anything upon itself", however the implementation of the limiter is very unfortunate IMHO, since it even applies if your floating point output clips but is then reduced to e.g. 50% in the mixer. It should rather be applied to the stream after attenuation. As said, since your audio clips, the only sane thing for you to do is the reduce the volume anyway, not revert to 24-bit. That's identical across all operating systems.

raina

  • Posts: 1163
Re: 3.8 reports, queries and bugs
« Reply #243 on: 21 Nov '14 - 06:57 »
There's this thing I noticed in 3.8.0.26 where tags from xmp-gme show just fine in Track info but they aren't listed in the Message info anymore and so xmp-scrobbler doesn't see them.
« Last Edit: 27 Nov '14 - 08:08 by raina »

Ian @ un4seen

  • Administrator
  • Posts: 20389
Re: 3.8 reports, queries and bugs
« Reply #244 on: 26 Nov '14 - 15:31 »
The current XMP-GME plugin version (rev.4) doesn't appear to implement the "Message" info retrieval function. Are you sure there were tags displayed in the "Message" info window previously? Anyway, to get around the problem, here's an XMPlay update that will display the tags it has when the plugin doesn't implement the "Message" info retrieval function...

   www.un4seen.com/stuff/xmplay.exe

raina

  • Posts: 1163
Re: 3.8 reports, queries and bugs
« Reply #245 on: 27 Nov '14 - 08:20 »
I thought I was sure but it does seem more likely now, that I previously had the tags overridden with Library stuff. ...And it looks like the soundtrack I thought I had tagged in the library, actually wasn't. I don't post a lot of posts anymore but when I do, I sure make 'em hasty and poorly researched!

saga

  • Posts: 2179
Re: 3.8 reports, queries and bugs
« Reply #246 on: 30 Nov '14 - 23:50 »
XMPlay's pattern vis seems to fail to display the lowest possible note in XM files, C-0:

Ian @ un4seen

  • Administrator
  • Posts: 20389
Re: 3.8 reports, queries and bugs
« Reply #247 on: 1 Dec '14 - 16:44 »
Oops, here's an update to correct that...

   www.un4seen.com/stuff/xmplay.exe

xaos

  • Posts: 172
Re: 3.8 reports, queries and bugs
« Reply #248 on: 16 Dec '14 - 05:10 »
I may have missed something, and I'm only saying this now because I haven't figured it out, .. it seems like title scrolling has been broken for several versions.

edit: if at all possible, it would also be nice to be able to 'grab' the 'title' in the main window and manually scroll it back and forth regardless of whether scrolling is turned on or not (flashback to winamp)
« Last Edit: 16 Dec '14 - 05:34 by xaos »

winner

  • Posts: 260
Re: 3.8 reports, queries and bugs
« Reply #249 on: 16 Dec '14 - 05:18 »
I may have missed something, and I'm only saying this now because I haven't figured it out, .. it seems like title scrolling has been broken for several versions.
I've noticed it too. Not working in version 3.8.0.28.