I might be misremembering but I understand that the XHA1 is a little different from most cameras in this regard, in that switching the inputs to line level actually does not bypass its preamps. Instead, it switches in an inline pad before the preamp to take the line level it's being fed and knock it down to mic level, then it goes on to the preamps for reamplification. Seems a very weird way to do it but I understand that's how Canon chose to go with this camera. Possibly the rational was it's cheaper than putting in two separate recording amps, one optimized for mic level input and the other for line level input.

Whatever, using it the way I described it (line level from SD302 and pots fully open) I got over 90 dB S/N ratio, using mic level from SD302 and reamplification I got 80 dB S/N ratio.

A no-brainer to me.

Steve,

I'm pretty sure a lot of cameras do that now. Engineering in a pad is easier than a separate signal path around the preamp.

Regards,

Ty Ford

The 302 mixer does the same thing on its inputs as well. Also, Ty I think the 302 comes out of the box set for line level not mic. Of course as you've described in other threads the level can be lowered to mic via the software.
Thing to remember about the Canon XH1. The pots are attenuators so as you turn them to the left you are increasing the amount of gain going into the camera. This is counter intuitive to most audio people as we all know that Up is usually Louder. The sweet spot marking on the meter on the side of the camera is at -12. Don't set your tone to that mark. There is a -20 mark that is smaller. You want to set your tone to that. If you go to the -12 mark you will have 8 db less headroom and could easily distort on peaks. Also use the 302's vu plus peak reading setting so you can accurately see what the camera is getting. Petri is dead on in his assessment of how to work with this camera.
I know you are fully aware of all this Ty, just wanted to add some clarity.
Cheers,
Bernie

Hey Bernie,

Thanks for that. I don't know that about the Canon input controls. That's whacked!

Regards,

Ty

Yes. And I had a heck of a time trying to tell a HV30 shooter recently that the reason he was not happy with his "HD" footage was due to the fact that it was really HDV, and very compressed.

Regards,

Ty Ford

The way I understand the difference between DV and HDV audio is like this
The HDV audio is inferior to the DV audio on the same camera (like the XHA1) because the audio signal is compressed into and with the Video stream into the GOP while the DV audio is recorded seperately from the video stream onto the tape. Not sure if they could have chosen less compression for the audio and still fit it in the stream although I would have thought so. There apparently was space on the tape for simultaneous DV quality audio recording with the HDV Video but it was always out of sync by something close to the length of the GOP so no manufacturer seemed to want to deal with it.

Dan,

Yes, I had surmised that the mpeg audio was probably due to the latency problems; video encoding taking longer than PCM audio. So they needed to slow down the audio so it would sync with the video.

Regards,

Ty Ford

Ty,
Yes that was what I was told about the Mpeg audio as well but they still had room on the tape for the out of sync PCM signal to be recorded which could have been dealt with in post. A little tricky perhaps but might have been worth it. For example I am not sure if anyone would be able to access the PCM audio on firewire at the same time as the Mpeg. Might have taken more than a little engineering. HDV was an unusual advancement over DV in that a higher quality video image over the previous spec was married to a lower quality audio signal from the previous spec when it probably didn't have to be the only option due to physical constraints of data rate recording limitations.
Certainly not much we can do about it now but it is good to know that newer is not always better quality in all respects or better designed.