zaboomafoo

The reason I said 0db digital = +12db analog is that for the digiatl world 0db is the clipping level (going out). On the other hand, in the analog world, 0db is an optimum level and each device has some headroom available, at least +12db (usually more for the pro devices). So to keep both sides out of clipping, digital would have to be in -12db range and analog in 0db range.

For example, Roland's manuals for their digital (VS*) recorders recommend peak levels to be in -12db range to avoid clipping.

Of course, both devices would have to be at the same standard (consumer or pro) for this to work.

On that note, line outputs from audigy cards are at the consumer levels, am I correct?

Regards,
Frank

Max M.

>On the other hand, in the analog world, 0db is an optimum level and each device has some headroom available, at least +12db (usually more for the pro devices)

Exactly.

> So to keep both sides out of clipping, digital would have to be in -12db range and analog in 0db range.

Nope, i believe this is misinterpretation. The proper device outputs nominal analog level (0db) for the maximum digital level (same 0dB) - and in analog you have additional +12dB headroom above this. (Not that you should provide this headroom manually below the nominal).

By keeping maximum digital level at -12dB you only lose dynamic range and SNR (by same 12dB).
Again, to prevent the destination device from clipping you should keep the source device levels at its nominal (= 0dB) and tweak the destination device levels to the maximum level that causes no clipping (and not vice-versa, the only exception is when you have no control over input level for the destination device - there you obviously have no choice other than reducing the output level of the source).

>Roland's manuals for their digital (VS*) recorders recommend peak levels to be in -12db range to avoid clipping.

Could you give a link, please? I suppose this info was just some specific to this particular Roland product and it should not be generalized to anything else.

p.s. (i've just went through VS-2480/VS-1680 manuals and supplemental documents - and can't see anything related)

__________________

zaboomafoo

Sorry I did not mean maximum at -12db, I meant average (or whatever is the right term for that). Same as on the analog mixer one would keep the level at 0db in average, to allow for spikes in dynamics to go above that.

I will have to look at the manual again, it was for my VS890.
I am not sure if you can see it on this image: http://namm.harmony-central.com/SNAM.../VS-890-lg.jpg (look in upper right corner, right side of the LCD where the peak levels are).
but -12db is printed in reverse on the display, to indicate this.

I can see that when I send out something at -12db from the audigy to the mixer and set mixer's input trimmer to 0, input level on the mixer is at 0db (it is a behringer mixer, I believe it is at the consumer level). If i put a digital device in the chanell's insert on the mixer (PEQ for example), it shows -12db on the input too, so it matches audigy's output.

Thanks for your answers. I recently started mixing digital with analogue gear and this helps me understand how to properly set the overall gain structure. Now I would like to use a dedicated computer with couple of audigy cards in it as a digital mixer (working on a kx plugin for that).

Regards,
Frank