USB Microphone problems

Jerome Santucci wrote:
I have used a Samson c01u usb condenser mic for recording voice-overs for the last 3 years on a 12" G4 PowerBook. I got my MacBook 13" unibody last month, and the mic is useless with it. Recording levels are super low and the sound in muffled or garbled sounding. Samson has a software pre-amp which is not Leopard compatible. Samson says it did not update the software because it is not needed in Leopard. I have followed Samson's directions for adjusting the recording levels in Leopard (in the sound pref pane, Audio Midi setup app, and Garage Band 09), but the levels are still way too low, and the sound is still muffled. Also, if anything is plugged into the second USB port, it creates static on the microphone.
I have written to Samson and they say this is a known issue with cheap MacBook usb ports and/or hardware controllers. My usb mouse will sometime stop working for no reason and I have to unplug it and plug it back in, so this explanation sounds like it might be accurate. I am very disappointed if this is the case, because I upgraded my computer specifically for recording (the G4 is just soooo sloooow). Because this may be a hardware issue or a software issue, I have also posted this as a reply to another thread in the Garage Band forum.

Any ideas?

Something seems awfully suspicious here. A USB microphone has to include its own analog-to-digital converter and therefore the amplification must occur in the microphone. I suspect that your adjustments to the microphone gain aren't actually working and thus there is either a compatibility issue with the Samson hardware or there is a software conflict on your MacBook or there is a bug in Apple's USB audio drivers.

Are you sure you haven't installed any third-party USB audio software (that would include software from Samson)? Something may be in conflict with Apple's built-in drivers. Also, it's unlikely that this problem is being caused by "cheap MacBook usb ports and/or hardware." Once the audio signal has been converted to a digital form (for transmission over USB) it can't suddenly be changed as it goes between the devices (if that were true USB drives wouldn't have any chance of working).

Another possibility would be that the USB port on the MacBook isn't supplying enough power to adequately drive the microphone. However, Samson says this about the power requirements:

USB low-power device - draws 26 mA => 130 mW. In suspend mode 0.3 mA => 1.5 mW.

However, since UBS allows up to 500mA per port I can't see this as being the problem. Also, I haven't had any problems with bus-powered devices on my MacBook so I think a power problem is highly unlikely.

You might want to reset you PRAM since some information about audio levels is stored there. Here is an Apple Support article on resetting the PRAM:

http://support.apple.com/kb/HT1379

Beyond that I'd check for a software conflict.

As for the USB mouse, have you installed third-party software for the mouse? I'd suggest that you do NOT install third-party mouse software as that too can cause USB problems (although it would be somewhat unlikely to interfere with the gain setting on your microphone). Lastly, I would avoid using any type of hub between your microphone and the port on the MacBook (you didn't say that you were, but just to be safe).

As I said at the beginning, this seems like a problem in setting the gain on the microphone itself.

Also, if anything is plugged into the second USB port, it creates static on the microphone.

-- and --

The reason Samson said they suspected the usb controller was the feedback I was getting from having a second device plugged into the second usb port.

Are you sure this "static" or "feedback" is actually being recorded and not just something you hear when monitoring from the microphone? There is a big difference between those two possibilities. If it is actually being recorded then that would indicated that the USB digital data is being corrupted which seems highly unlikely. However, if you just hear the static or interference when monitoring the recording (i.e. in realtime, not during playback of the recording) then you could just be getting an analog interference which is certainly possible. If it is analog interference then the problem could reside in a number of different places (not just from the USB ports and USB hardware itself).

nother thread on this discussion board ( http://discussions.apple.com/message.jspa?messageID=9156522&tstart=0) says that the 10.5.6 update is what messed up their mic input.

Are you running 10.5.6 or 10.5.7? If you are still using 10.5.6 you might consider trying the 10.5.7 update.

The only way I could see another device affecting the actual recording is if there was a power supply issue. Thus, if there was a change in the power output to the microphone then this transient could cause the amplifier or ADC in the microphone to produce a similar transient (i.e. static) in the audio output. If that is the case then you might want to use a powered USB hub for those other devices. Thus, I'd plug the microphone into the MacBook and then attach a powered USB hub to the MacBook's second USB port and then plug any other devices into that powered hub.

Okay, I think I may have a workaround for your problem. Try this:

1.) Go into Audio Midi Setup and select the Samson microphone (default input and properties).
2.) Under Audio Input -> Format select 2ch-16 bit.
3.) Set the volume levels for channels 1 and 2 to maximum (24 dB).
4.) Now return the Audio Input -> Format to 1ch-16bit.
5.) Make certain that the volume level is set to maximum (24 dB).

This should result in a notable increase in the microphone gain, but you will also have an increase in the background noise. Given that, the signal to noise ratio may be no better than when leaving Audio Midi Setup to its startup condition (i.e. when you don't do the above workaround).

Unfortunately, it appears that you may have to do this every time you restart your MacBook or reconnect the microphone. If you find this to be the case then you can just leave the Format set to 2ch-16bit and that should "stick" between system restarts. However, one problem in leaving it set to 2ch is that the channels won't be balanced during record/playback (one channel -- the left) will be higher than the other (at least that's what I've seen). I also don't know what effect this will have on the quality of the recording (that is, if you convert the stereo back to mono with an editor). This left/right balance problem doesn't happen when you set the Format to 1ch-16bit.

Right now I don't know if this is a bug in Apple's OS X 10.5.7 USB audio drivers or whether it is some problem with the Samson that has just surfaced with changes that Apple made to 10.5.7. It may also be something that can be fixed (more permanently) with removal of a preference or cache file. I'll be looking into that possibility.

By the way, I've read one report that the optimum sample rate for recording with the C01U is 48KHz since the microphone's A/D converter also samples at that rate (i.e. avoid 44.1KHz if possible). However, this same report claimed that GarageBand only worked correctly at 44.1KHz as any other sample rate may introduce clicks while recording. Actually, this reported problem with GarageBand seems a little odd to me, I haven't confirm it, so I don't know right now whether this is true.

I just checked on a different Mac running 10.4.11 and the behavior there seems exactly the same. So, this doesn't appear to be a new behavior (i.e. the switch between 2c-16bit and 1c-16bit to increase the gain happens even under 10.4.11 on my Mac Pro). The fact that you get substantially more gain after doing this workaround may be somewhat meaningless anyway since the signal to noise may remain about the same (thus, it may be no better than doing a level adjust or normalization after the audio is recorded). The interesting thing is that I'm almost certain that on both the MacBook and the Mac Pro the first time I used the microphone I didn't have to do this workaround. Could there be an internal state within the microphone that is getting set or corrupted that then carries over to other Mac systems?

I may try the Samson utility under a Mac OS X 10.4.11 system to see if that makes any difference in the results. Thus far I haven't tried to install Samson's utility (it doesn't work under 10.5 so it's no use with my unibody MacBook).

Okay, I think I have found out what is going on. This two channel versus mono (one channel) difference in the gain on the C01U is because Samson is using a two-stage amplification and it does in a sense work in a stereo mode (they apparently are using a stereo preamp and route the microphones one channel through both the left and right preamp channels). The default mode (which produces less gain) is most likely what Samson thinks offers the best signal to noise. However, if you boost both channels to maximum gain (as I suggested in my "workaround") and then switch back to mono mode you will get the maximum gain that can be set on the C01U (but possibly with a lower overall signal to noise ratio).

Thus, this behavior doesn't appear to be a bug in the Mac OS X drivers, it's just a rather obscure artifact of the way the C01U is designed. In a sense it may almost be a "feature." There is some discussion on the internet that you can actually optimize the system's signal to noise by adjusting each channel's gain independently (with corresponding decreases in the output level). However, one user said that the differences are so slight that it probably isn't worth the effort.

One other note, it was reported that there was a phase difference between the two channels when you record in stereo. One suggestion was that you set the right channel to full gain and the left to 60% and that will produce a stereo recording with equal levels. In this case supposedly the right channel will have a little less noise.

Try this:

Open Utilities/Audio-MIDI Setup.

In the menu, under "AUDIO" open the Aggregate Device editor.

Create a new Aggregate Device. (Add)

Select all available inputs and outputs (use), or at least The Samson and Built-In-Audio Inputs and all of the Built-In-Audio Outputs.

Try Selecting "Built-in-Audio output" as the clock master do not select resample. (might have to experiment with the this setting by selecting different devices as clock master)

Name and save the aggregate device.

Select the new device in Audio MIDI Setup Properties, there should be Input level there although I don't think this will affect Amadeus as it has it's own audio settings.
In Amadeus, set the audio output to Built-In-Audio or Line output, however it is you're listening to the output.

See if that improves the situation.

Don't be afraid to experient in the Aggregate device editor, you can always delete it in a secon.

pancenter-

Oh, and the PBG4....

It has higher quality audio/USB/Firewire components, mainly Texas Instrument chips. Also has a more Robust power supply.

pancenter-