Anyone else find this misleading and wrong? Page 338
No
“as loud as possible without exceeding 0dB” That would assume the user knows that -18dBFS is the general equivalent of 0dB. I had a user question this on another forum, and I feel it is misguiding for someone new to recording.
That would assume the user knows that -18dBFS is the general equivalent of 0dB
I’d say you’re wrongly assuming they are assuming that. I would read it exactly as written. They are suggesting you record as loud as possible without clipping (at 0dbfs)
Most would say you’re better off recording lower level than this nowadays so in that case it could be seen as misleading, but not specifically wrong…if you’re not clipping then you have a good audio file after all.
But it states ‘0 dB’. It should be stated as ‘0 dBFS’ to make it clear. That is what the forum member I replied to was unclear about. He wondered if the equivalent was correct or whether he should make changes in his input chain.
In any case I understand completely how this works and what is best for any instrument input level. Hell, I have pushed live drum tracks damn near the point of clipping when recording. But the wording in the manual kind of assumes the user knows the boundaries from experience.
Well arguably because they are referring to a specific meter there is no need to clarify the calibration.
As far as I see in the other thread the user is just wondering how the UR22 signal/clip light is calibrated rather than the metering in Cubase. (I have no idea…I think your advise to them covers it perfectly though!)
The manual is correct, assuming you are not using 3rd party plugins that are modeled after analog gear, including gain structure.
But that should be pointed out in the manual of the 3rd party plugins gain structure.
You are both right and you are both wrong. To get technical:
The dB scale is a referenced scale. That means you must state the reference level (dBFS, dBu, dBv, etc) for it to mean anything.
The exception is when the reference level is implied by the context. The manual refers to the scale on the MixConsole, and the reference level is therefore implied (dBFS in this case).
This, however, is wrong. 0dB"what"? This statement means absolutely nothing. Since the reference level used on most analogue equipment is dBVU, the correct way of describing this would be “18dBFS is the general equivalent of 0dBVU”. Those two letters means everything!
Or it could be interpreted as -18dBFS = 0dBFS, if we go by the rule of implied reference scale. Which, obviously, is impossible.
You could also use the dB scale to compare two values. It’s correct to state that “The difference between Y volts and Z volts is XdB.”
what i rather find missleading in this chapter is that is sound like you can acutally see overs produce at the analoge stage in the meters of you input channels. In fact you can not. Every over you will get indicated can only be a result of an altered fader position, Gain position or any inserts that will cause a boost in levels in the digital domain inside Cubase after A/D Conversion.
by definition there is nothing higher than 0dbfs in the digital realm. Full Scales has its highest possible value at 0dBfs.
thats why its actually useless for judging clipping on the internal metering of Cubase for the analoge stage. Also all converters on the market work with interger bit depth
Yes i know with 32bit float you can have higher peaks than with 24bit interger thats why cubase internally can have peaks above 0dbfs
Formally you would have to rescale the metering making 0dbfs the maximum possible value for 32bit float. The problem is that this wouldn’t be any usefull because the levels that are resonable for audio would be so small that they wouldn’t even be indicated making the metering completely useless.
Input/output metering on every insert would be a good start 
Thanks to Svenne, I was to lazy to look up the right way to describe the dB relationship.