He is right about suggesting a recorder that uses .wav format, but that is about it.
A couple points:
1) capturing in 24bit vs. 16bit provides at lot more accuracy in the analog->digital conversion, and considerably lowers the noise floor, which helps with later editing/fixing of issues with a raw recording (EQ/compression/limiting/adding gain or normalizing) and simply preserves a much larger amount of data and acoustical detail.
While 24 bit is of arguable utility for large venue/live PA recording if you intend to do nothing more transfer raw to a CD, once you do anything in the digital realm in post to a recording, even just to normalize or add a couple db of gain, that extra data at 24 vs. 16bit becomes much more important in minimizing any artifacts/noise/distortion that may be introduced by the post-editing. (many folks here could elaborate on these points in much more detail than myself).
2) typically, it misses the point of capturing the dynamics of the experience to apply a hard limiter or compression to a live 2 track recording at the show. While compression and limiting have their purpose, and are very useful editing tools for multitrack and matrix (SBD + AUD) recordings, there is a greater ability in the realm of digital recording today to do this
much more accurately in post-production vs. trying to do it live at a show.
With respect to 'hard' limiters, imagine the .wav to be like a person jumping up and down in a closed room where the ceiling of the room (or subflooring since .wavs go up AND down
) respresents a threshold of Odb - above that equals clipping. A limiter, whether 'hard' or 'soft-knee,' would set an artificial ceiling (threshold) beneath the real ceiling to stop the person jumping from hitting the actual ceiling (
limit the signal level/amplitude before it reaches the clipping level). To extend the analogy, a 'hard' limiter would be like making that artificial sub-ceiling out of concrete or brick...above which nothing can pass. A 'soft-knee' limiter, on the other hand, would be like making that artificial sub-ceiling out of thatched wood trellis work, where the person jumping could quickly push his hand through a hole before the rest of the trellis arrests his ascent. So, a rapid transient like a snare hit (hand pushing through a hole in the trellis) will be less affected by a soft-knee limiter than a loud bass thump (person jumping's head being too big to fit through a hole). A hard limiter will smack down both the hand and the head equally.
Graphically, a hard limiter will look like an ascending line that abruptly goes flat (horizontal) at the threshold, while a soft-knee limiter will look like the ascending line curves to horizontal around the threshold level.
Why is hard-limiting less than ideal for live PA recording? primarily because it sets an artificial maximum to the dynamic levels captured on the recorder, but also secondarily because many 'hard' limiters do not sound very musical and may degrade the sound.
Hard limiters have a tendency to invite the person recording to set their levels too high b/c of the known 'safety' of having the limiter engaged, which in turn results in 'squashed' sounding recordings where there is little audible difference in volume between what should be 'quiet' and what should be the 'louder' parts of the music/performance.
---------------------------------------------
Where the two points fit together: recording in 24 bit considerably lowers the noise floor when recording, thus enabling one to set levels more conservatively and then add gain later in post-processing. The use of hard limiting is overkill in this context unless the signal is too hot for the recording device, and you cannot turn it down on the front end. It is more of a multitracking tool when setting a final output level on a stereo mix, or to be used for EFP(electronic field production)/ENG(electrnoic news gathering) work where you only have once chance to get it right for a live broadcast. If you actually need some kind of limiting or compression to control a sudden dynamic spike that is out of proportion with the rest of the music, you may make it sound a lot more pleasing when you do this processing after the fact.
--------------------------------------
A couple other responses:
-many decks, some pre-amps, and most portable mixers will have a built in limiter, although it is not a function upon which you should base a choice in buying a recording device for live PA recording with a microphone pair. As long as you can leave the limiter off, you'll be fine with whatever you chose.
-I have no idea what he means by buffer, because it could mean two totally different things: (1) a 'true' threshold before clipping on a recorder that is __db above the clipping level according to the deck's metering, or (2) a cache to preserve a rolling memory of a data stream from the A->D converter of the recorder to reduce or eliminate the possibility of write-errors in the event that vibrations or other issues cause an interruption in the normal writing of data to a medium like a CF card or a hard drive.
-I believe the attorney's knowledge and/or experience with recording is from the studio/multitracking perspective, or audio production for video, and not with respect to throwing up a stereo pair of mics to capture 'what goes down.'
*edited for spelling/grammar