Intro: As If you are going to do one thing right when reproducing tapes, set the "record levels" properly. Here are some tips, provided by Alex Grabinski (5/3/93):

• Meaning: Record levels affect the input from the original tape, and therefore determine what gets recorded on the copy. The levels can be turned up to boost the signals so that the sound isn't hollow and quiet and muffled, or they can be lowered if the signal on the original is so strong that it squelches, or shudders speakers.
• Check before dubbing: As John Stewart <johns@zebu.heurikon.com> posted (12/20/95), "Levels are not something that is static and set once by the person that originally dubbed the tapes. Each time you dub a tape, the levels should be set such that the peaks are just below a certain range for a specific type of tape.
• Only once per tape: Keep the levels consistent throughout both sides of the tape. Don't try to remaster your tapes by changing levels on the fly.
• Pick a loud reference point: When setting levels, find a 'loud' part of a 'loud' song and use that to determine peaks, a portion of the song where everyone is playing fully - heavy bass, low drums, power chords on guitar and keys. The finale to Bowie is good, or the chorus to Possum.
• Guidelines: Levels should "peak" (that is, reach their highest point on the tape, which would therefore be the loudest/strongest point of the recording) at the appropriate level: between 0 and +3 dB for Type II tapes, +4 to +6 db for metal tapes. The rest of the music should generally follow 4 db behind -- between -4 and 0 dB and peaks a little above 0 dB on a properly recorded Type II tape, 1 to +4 on a metal tape.
• Decks & tapes vary: Edwin (5/4/93) posted, "From my experience as an owner of several different decks and as an erstwhile studio engineer I've learned that for analog cassette decks there are no rules. Sony D5's have to be hit very hard so that the peak light is almost always on, while my portable Marantz PMD 250 is sensitive to the point that +2 saturated. I run the JVC decks at Wellspring Sound at between +8 and +10 with no distortion. Unless I record at +7 on my Nak it sounds quiet on other machines. Newer tapes can certainly handle a lot more level. The bottom line is that every responsible taper needs to take the tape of his/her choice and record at successively hotter levels until audible distortion occurs, then back off 2db and note the level. This level will be different for different tape types."
• Boosting isn't selective: Eric Launder <eric@livingston.com> posted (12/20/95) that "if a tape that is well below 0dB is copied to a high bias tape and you boost the levels to meet the dB capacity of the tape, you are boosting everything on the tape, including the hiss. Boosting tapes too much also introduces a certain amount of distortion to the tape. The dynamics of live audience recordings don't lend themselves to higher recording levels as CDs do. The mix of keyboards, distorted gutairs, cymbals, and crowd noise without the benefits of a clean mix make some tapes sound totally saturated at fairly low levels. So boosting a tape like this makes it sometimes unlistenable. The key is to listen to each tape carefully with your recording monitor on and adjust the recording levels in such a way that adds the least amount of distortion and hiss to the recording."
• Seek a balance: David A. Craig <dac@cosmic.physics.ucsb.edu> posted (12/21/95), "Set the levels high, but not too high. The bottom line is, *you maximize the sum total signal-to-noise ratio by recording each new copy at optimal levels, in particular by maximizing the s/n of each new generation. Just because some dimwit screwed up one generation doesn't mean each generation thereafter has to keep repeating the same mistake. It is true that boosting levels boosts both the music and the hiss. Recording at lower levels reduces the ratio of the total signal on the copy to the *new* layer of hiss. The best final s/n is achieved by maximizing the s/n ratio of each step of the recording process. Too high, and you saturate the tape, leading to unpleasant distortion. The low level thing gets the most emphasis because it is the more common error by far, in my experience. This is one reason tapers loath dubbing decks - the usually end up setting levels uncomfortably low."

And, here's a thorough discussion by Lee Silverman (now at <lee@www.phish.net>) (8/4/93):

Dubbing decks are designed to make exact copies of tapes, hence the "record level" knob is usually disabled in dubbing mode. To test this, try dubbing a sample tape. Let it record for thirty seconds, then turn the record level knob to zero, wait 15 seconds, and then set the record levels to 10 (or 11, if you have it B-) ) for 15 seconds. Stop the tape, go back, and listen to the copy. If the levels change, then you need to worry about setting them correctly. If they don't, you should probably read this anyway, but remember that there's nothing that you can do about the levels on the copy.

As Edwin pointed out, many tape decks display tape levels differently. However, almost all single-well decks have dolby and hence should have a dolby "double-D" [)(] symbol somewere on the record levels meter (some decks, like the JVC, have some other indicator that takes the place of the double-D). The location of this point on the levels meter can be used as a reference zero db value. On my Sony deck, as with many other brands, the double-D coincides with zero on the record level meter. On my JVC, it's at +4. The reason for this is that the Db values shown on the meter actually indicate the signal strength *relative* to a certain reference value that can differ between tape decks. That's why Db on the meter are written as +5 or -10; they indicate that the signal is that much louder or quieter than the reference value. When the Dolby double-D coincides with zero on the meter, it means that the company that made the tape deck has adopted Dolby Labs' reference value.

Just to demonstrate to yourself what can happen if the levels are set incorrectly, turn your record level knob to 10 and record a loud song, so that the peaks are way above what they should be. (Peaks at +8 or +10 should suffice.) Then listen to the tape (at lower volume) and you'll probably cringe. The broken, distored, almost grinding sound that you hear is distinctive of saturation, and it is the result of putting too much "energy" on the tape (speaking loosely). Furthermore, the amount of "energy" that a tape can handle depends on the type of tape. This complicates matters for people with dubbing decks who want to record from a metal tape to a CrO2 tape. If the peaks on the metal tape are +6 relative to the dolby standard, for example, and you use a dubbing deck to record to a CrO2 tape, the deck will try to record those peaks at +6 on the CrO2 tape and it will probably saturate. The best advice for people with dubbing decks is to always dub from metal to metal, and CrO2 to CrO2 or Metal, or Normal to all three.

Now set the record level knob to about 1 or 2, and record the same piece again, and see what happens. When you play it back, you'll notice that the hiss on the tape is very loud compared to the music. The hiss has effectively become 100 times louder, because the signal is that much quieter. The ratio of the level of the signal to the level of the hiss is called the Signal to Noise ratio, and is usually abreviated S/N. When setting the levels on your tape deck, the idea is to get the signal as loud as possible to reduce the amount hiss relative to the signal, but not to saturate the tape.

First, find the dolby double-D on your deck's record level meter; I'll call that level zero, so we're all working with the same standard. For a Normal (type I) tape, never exceed zero; set the levels so that the peaks barely reach it, if at all. For a CrO2 tape (type II), it is safe to let the peaks go a little (+2 db) above zero and you won't get any saturation, but I usually set the peaks so that they hit at about zero, to get better high-end response. On a metal tape (type IV), you can let the peaks go substantially above zero. I set mine to go to +4; some will say +6 is OK too; this number often depends on the brand of metal tape you use. I prefer to keep them a little lower than the max to take advantage of metal tapes' improved high-end response, but other people simply hate hiss, and think that adding hiss degrades a tape more than losing a little high-end, which the hiss might mask anyway. Experiment a little to get a feel for what works with different brands of tape. Finally, note that on the JVC, the reference zero value changes for different types of tapes (metal, CrO2, etc). Hence, when I talk about a reference zero level for a deck where the reference changes, use the reference for CrO2 tapes.

The bottom line is this: if you screw up a tape for someone, you screw it up for everyone that that person might want to give the tape to, and everyone that they might want to give the tape to. We all share this music, and it's our responsibility to take a little extra time to do things right, rather than to be lazy and screw our friends. Set the record levels for the high gear of your soul; you've got to tape like an antelope - Out of control!

"If I were to wish for anything, I should not wish for wealth and power, but for the passionate sense of the potential, for the eye which, ever young and ardent, sees the possible.""
-- Soren Keirkegaard"