[clepsydrae]

Originally Posted by Reno.thestraws

So, You didn't test the input, but the output. That means taht you REDUCE the volume of the output (and so the SNR) before reinjecting.

I see your point: yes, the SNR goes down when reducing the output level due to the noise floor of the output, and I should have isolated for the SNR reduction on the input only. Let's do your version:

-Play a 8k tone on your speaker (not too loud)
-Place a SM57 or a similar dynamic microphone if front of your speaker (DO NOT MONITOR THE INPUT or you'll get larsen)
-Plug the SM57 on a preamp
- create a new track in reaper with the input set to your SM57 (Do not monitor input!)
- Set the preamp to achieve a level of -0.5dbfs in REAPER on the track input and record
- now ,mute the track you just recorded and create a new one
- set the preamp to achieve a level of -18dbrms in REAPER on the track input and record
- Normalize the two recorded track and make your analysis again

I have done this now; with an AKG C-535-EB into a pre-amp on the FireStudio. (The 535 is condenser -- is there a reason to favor a dynamic? Unfortunately I only have condensers...) Same basic result: the hotter signal has less noise. There is not as big a difference as before, but it's still worse at the lower level. (Of course all this noise is effectively silent.)



I presume that besides the pre-amp stage there is another analog component in the ADC that adds a constant amount of noise, regardless of the gain (the Lavry tech mentioned "intermediate stages" between the input and the actual sampling AD converter). So you turn down the preamp, and along with it the signal, but there is other noise that does not change, and the SNR is reduced.

I'm glad you had me do this second test to isolate to the input: the noise issue (which was never a significant problem) is less than we saw before, but it is still there.

All I want to show is that the signal doesn't get worse at -0.1, and that seems pretty clear both from these tests and the listening tests I've done, not to mention the four manufacturers I've been able to track down and the other authorities quoted.

Originally Posted by Reno.thestraws

So, if you want to be sure to don't overload the analog part of your signal chain, you have to rely on your digital level

I understand what you mean: it's a point also made by others here, that getting levels at -18 dBFS RMS (or whatever particular calibration) may be a good idea for some or most analog gear coming before the ADC.

But a lot of people advocating for -18 dBFS RMS present it the opposite way: they say that you need to do it because "hitting your converters too hard makes it sound bad, distorts their input, etc". That's the point I'm trying to get at.

And it's looking increasingly like that part is in fact wrong. We just need to stop saying it backwards.

I propose this rephrasing:

Originally Posted by rephrasing

Recording to levels at -18 dBFS RMS is generally a great idea; it doesn't change anything as far as your ADC is concerned (except adding an insignificant amount of noise, and possibly an even less significant amout of THD), but it leaves you a safe amount of headroom and tends to result in you setting appropriate output levels from other analog gear going into your ADC, and depending on your plugins it may result in more convenient levels once in the DAW; however, some analog gear may sound better outputting hotter or quieter, so use your ears, and don't worry about 'driving your ADC too hard' -- as far as any contemporary and reasonable ADC is concerned, the harder the better; just prioritize the optimal levels for the analog gear before the ADC, and make sure you have enough headroom to guarantee that clipping won't happen.