-
Here’s an audio puzzle for you folks. Solution included. Check this out.
So the solution to this has been found, but I thought I’d share it with you, might save you a ton of time, because neither I nor my buddy who is a professional sound engineer had ever encountered it before. Rather than spend a bunch of your time pondering what was wrong, on this, you can read this and then read the solution on the bottom and go, “sure”. Or you can go look at the files while they are still up (I’ll rename them to identify them when I push the fixed files up). There are clues to the answer in the story below. There are also questions based on those clues you might ask yourself.
I recorded a lecture in a hall with a decent podium mic at 48khz in WAV on my Marantz PMD 661 (shooting video to marry up to it, but also wanting to output an audio podcast). I was taking line in off the floor jack, which the guy who did AV for the relatively new building had never used. Cans said it was working fine when I started, and when I ended. So it wasn’t anything that happened during the recording.
Playing it back on both my Windows Lenovo 520 laptop speakers which usually are only so so, and through my powered studio monitors, it sounds “fine”.
I output it to anything, 44k or 48k MP3 and while it sounds OK on the same speaker sets as above, on Windows, when I stick it into my Macbook Pro, the same MP3 file sounds softer, tinny and hollow, at even the highest volume the mac is capable of. (this is a newer Macbook Pro) with the latest iTunes.
All other music plays fine with the Mac and has perfect balance. So it’s not that the speakers on the MacBook Pro are lousy. I usually watch The Daily Show on this machine because the speakers usually sound better and louder than my Windows laptop. It’s something with these files.
I recorded an intro to the talk, with a local mic on the Windows Laptop. Sounds balanced and full, as I would expect. No sweetening needed if I just want it out quick and dirty. I put that same intro onto the Mac alone, and it sounds fine. It’s only the stuff from the talk that is screwed up.
No matter what I do, the talk will not sound good on the Mac internal speakers, though everything else comes out fine with those same speakers.
I’ve tried EQ, boosting volume everything, but the sound is not fixed. I’ve tried resampling to 44 kHz. WAV, MP3 at 256kps. everything. Even the WAV file itself sounds bad when not on powered monitors.
Another oddity is that when I plug in the same powered speakers that are usually plugged into the Windows laptop and sound so good. The MacBook then sounds fine!
For you to hear. Here’s a few samples I’ve put out last night on Soundcloud. You can try it on your own machines and give me your thoughts. Some of these samples show a problem, some don’t but do have the problem!
I’ve not had this problem before and there seems to be something in the range of the Mac speakers with this specific set of audio that is acting odd. My customer who has a Windows computer with lousy speakers, said it sounded fine to her.
A non intro’ed output of a talk.
https://soundcloud.com/mountainstone/2012-nwstraits-annual-meeting?goback=%2Egde_1825429_member_182176220My studio engineering buddy says that they also sound terrible playing on his Mac!
SOLUTION: After spending a half hour puzzling this out with my buddy, I went back to the raw file. And noticed something odd. While very close, the two channels, L and R, did not match. That’s odd, I thought, as this should have been a mono recording. Then it dawned on me. There was a second mic, that was used to walk around and hand to folks in the audience. It of course was left on. The Line Out on the floor was giving me a stereo feed with the podium on Right and the floor mic on Left. When I turned off Left, the problem immediately went away. A quick stereo to mono conversion and my problems were solved.
My buddy wonders whether there was a phase shift going on or not, and why only the Mac internal speakers picked up the problem? Why did the good studio monitors not amplify the problem, rather than mask it? Odd…
I bet they don’t teach you this one at Berkeley!
Al
-
5 Replies
-
I love a puzzle.
You should have kept the qu and A separate.As soon as i listened to the audio sample here at home, my ears said there was a phantom image. Couldnt pin it to the screen with my ears. drifting as my head turns. Phase yells out! but not 180.
I only have PPMulator here at home but ill listen in the studio at work. My guess is that the l – r relationship is not 100% mono compatible.
radios always, by definition have to have a delay, but only a millisecond or 2 (40ms in 1 frame, 25 frames in 1 sec here in the uk)
PeterPost Production Dubbing Mixer
-
Thanks. What kind of speakers were you listening to it on? (that seemed to mask the issue to me). I thought about keeping the answer separate, and maybe next time will, but I value people’s time, and was worried someone would get upset at me wasting theirs over what might have seemed a stupid issue to them.
And yes, I didn’t mention that the mics were both radio, not wired. I guess I assumed you folks would assume that.
Anyway, thanks for your feedback!
Al
-
There’s a giant hole in the middle between the speakers. That says something is out of polarity. Normally the dialog should be rock solid in between the speakers with no L or R standing out.
Eric
“I push the RECORD button and hope for the best”
-
By means of an explanation of this territory for those on the cow who perhaps don’t know.
Were talking about how sound is heard by a listener sat between 2 speakers. If you play a voice track and pan it hard left and hard right (providing your speakers are equally set for power, then if you look straight ahead, you should be able to pin POINT at a place between the speakers, where it appears you have a third speaker. It is pinned centre and it rock solid.
This audio is called in phase (ie the coils of the speakers are pushing and pulling synchroniously)
If the audio or the speakers are ” out of phase” then the coils don’t work together and the centre Image of the audio is not pinnable like before.
This is something your ears will quickly learn to spot. The sound is diffuse, and not really pinnable as to exactly where it comes from.If 1 leg of audio is not phase coherent with the other (which can be caused by delays, cable lengths etc etc) then if you mono the 2 tracks you will find a lessening of the signal, bass frequencies are 1st to go) and worst case, if the audio is 180 degrees anti-phase to each other, the audio will disappear. Obviously this only affects audio in the middle area of a mix.
Hope this helps
PeterPost Production Dubbing Mixer
-
Yep. Listened in the studio
180 degreses out of phase.
Mono it and you get nothing.
PeterPost Production Dubbing Mixer
Reply to this Discussion! Login or Sign Up