Closed GoogleCodeExporter closed 9 years ago
GoogleCodeExporter
commented
9 years ago Whats allowed in WAVEFORMATEX is defined by the MSDN. Anything above 44100 Hz
will get WAVEFORMATEXTENSIBLE.
If your driver does not cope with 100% valid values, well, too bad.Original comment by h.lepp...@gmail.com on 15 Sep 2011 at 5:13
GoogleCodeExporter
commented
9 years ago Where does it say on MSDN that 44.1 kHz is the limit for WAVE_FORMAT_IEEE_FLOAT?
"If wFormatTag is WAVE_FORMAT_PCM, then common values for nSamplesPerSec are
8.0 kHz, 11.025 kHz, 22.05 kHz, and 44.1 kHz. For non-PCM formats, this member
must be computed according to the manufacturer's specification of the format
tag."
There is no limit specified even for WAVE_FORMAT_PCM, only a few common values,
but for WAVE_FORMAT_IEEE_FLOAT there's not even a hint of a limit. Here's
mpc-hc playing 48000 mp3 with WAVE_FORMAT_PCM with no problems:
Filter : Default DirectSound Device - CLSID :
{79376820-07D0-11CF-A24D-0020AFD79767}
- Connected to:
CLSID: {18C16B08-6497-420E-AD14-22D21C2CEAB7}
Filter: Audio Switcher
Pin: Out
- Connection media type:
Audio: PCM 48000Hz mono 768kbps
AM_MEDIA_TYPE:
majortype: MEDIATYPE_Audio {73647561-0000-0010-8000-00AA00389B71}
subtype: MEDIASUBTYPE_PCM {00000001-0000-0010-8000-00AA00389B71}
formattype: FORMAT_WaveFormatEx {05589F81-C356-11CE-BF01-00AA0055595A}
bFixedSizeSamples: 1
bTemporalCompression: 0
lSampleSize: 2
cbFormat: 18
WAVEFORMATEX:
wFormatTag: 0x0001
nChannels: 1
nSamplesPerSec: 48000
nAvgBytesPerSec: 96000
nBlockAlign: 2
wBitsPerSample: 16
cbSize: 0 (extra bytes)
__________________________________________
I can name three filters that do higher than 44.1 with WAVE_FORMAT_IEEE_FLOAT:
ffdshow, mpc-hc internal, vlc internal
with no problems. Are they all wrong?
Since you fixed Issue 96, it doesn't make sense to WONTFIX this, 48kHz is very
common.Original comment by laxera...@gmail.com on 15 Sep 2011 at 6:08
GoogleCodeExporter
commented
9 years ago Can you please review your position that MS docs forbid sample size > 44100 in
WAVEFORMATEX. If you were wrong, wouldn't you want to know?
And if that was the sole objection, will you reconsider the wontfix?Original comment by laxera...@gmail.com on 15 Sep 2011 at 7:04
GoogleCodeExporter
commented
9 years ago I really don't care much about this. I write modern filters for modern systems.
Windows XP is not a modern system, its legacy. Any newer Windows OS has a
proper DirectSound mixer which will gladly take any format, independent off how
broken your driver is.
Anyway, it looks like i found some reference that the 44100 Hz quoted in the
Windows SDK is not correct, but instead it will accept up to 48000 Hz without
using WAVEFORMATEXTENSIBLE.Original comment by h.lepp...@gmail.com on 16 Sep 2011 at 5:53
GoogleCodeExporter
commented
9 years ago Can you post you reference?
48000 is much better it covers the vast majority of current content, but to be
clear, its not the limit either. I've played 96 kHz files as WAVE_FORMAT_PCM
and WAVE_FORMAT_IEEE_FLOAT with no problems. I think you can safely set it to
at least 96000 as mpc-hc, ffdshow, and vlc do (perhaps they don't enforce any
limit)
Original comment by laxera...@gmail.com on 16 Sep 2011 at 9:19
GoogleCodeExporter
commented
9 years ago I will not change it anymore. This ticket is closed.
The documented limit is 44100 Hz, but some explanations from a Microsoft MVP
cleared it up that DirectSound and WaveOut can deal with 48000. Anything above
will need WAVEFORMATEXTENSIBLE.
This is my last comment on the matter. Get better drivers, a newer OS, or just
use a audio renderer that does it properly. I'll not risk breaking playback if
some component actually enforces the spec.Original comment by h.lepp...@gmail.com on 16 Sep 2011 at 2:50
GoogleCodeExporter
commented
9 years ago You must be referring to this thread:
https://groups.google.com/group/microsoft.public.win32.programmer.directx.video/
browse_thread/thread/c24f0fcf7b3f0d5d
Though the current MSDN WAVEFORMATEXTENSIBLE page:
http://msdn.microsoft.com/en-us/library/dd390971(VS.85).aspx
does not mention the limit even for PCM, the version still in Google cache:
http://google.com/search?q=cache:http://msdn.microsoft.com/en-us/library/dd75772
1(v=vs.85).aspx
does, but specifically and only for PCM (i.e. WAVE_FORMAT_PCM)
so for PCM, sticking to 48000 makes sense, to be on the safe side
But you have posted no evidence for a 44.1 or a 48 kHz sample rate limit for
WAVE_FORMAT_IEEE_FLOAT, and empirically it's not there, as you can confirm
yourself. mpc-hc, ffdshow, vlc, when configured for 2 channel float output,
all successfully output WAVE_FORMAT_IEEE_FLOAT, even for 96000. If that were a
problem, they would have changed it by now, but probably the opposite is true,
they do it because it's less problematic, as my experience with Audigi SE, a
fairly popular entry level card, shows.
You can of course do whatever you want, but I think you're making this decision
based on a misconception.Original comment by laxera...@gmail.com on 17 Sep 2011 at 7:04
Original issue reported on code.google.com by
laxera...@gmail.comon 15 Sep 2011 at 5:03