Industry information that you can actually use
January 2005
Jitter, Dither, And Blither-Blather: The 1x vs. 4x debate: Part IV of my
"Don't Get Duped By The CD Duper" series.
DON'T GET DUPED BY THE DUPER, PART IV:
Jitter, dither, and blither-blather: The 1x vs. 4x debate
(To catch up and read parts 1-3 go to Moses Supposes Archives.)
I have a headache. It began when Micah Solomon of Oasis Manufacturing made this radical comment, "Single speed [1x] glass [mastering] is no longer recommended by most top glass-mastering engineers so we have stopped encouraging people [to buy it]." 1x refers to the process of making a glass master CD in real time, where every one second of music takes one second to transfer from your source files to the master disc. 4x speeds process up to four times that speed, so every four seconds of music gets transferred in one second. Or thereabouts. The question is: Is data corruption likely to occur at the faster speed?
Seems simple enough. But have you ever tried to get a straight answer about the advantages of 1x over 4x from a replication plant? Don't bother unless you have a degree from MIT. Most houses use 4x and charge between $100-$300 extra if you want 1x. Some, such as Oasis, try to discourage you completely from going the 1x route.
So, I asked Micah if he could turn me on to any "top mastering people" to back up his statement. He referred me to Joe Bradley, regional sales manager for Sony Manufacturing, who said, "At higher speeds the jitter is actually REDUCED. The higher the speed, the less jitter." Joe says it's really an "emotional issue" and likens the debate to the one in which silly old hippies argue that analog, vinyl LPs sound "more natural" than CDs. (I'm one of those silly old hippies, by the way.) But Joe also admitted to me that he's not a technical person and has "never tried to hear the difference." Sigh.
He did try, however, to turn me on to a technical expert who COULD hear the difference. But none from Sony would confirm, on the record, that 1x would create more errors than 4x. I was starting to feel like I was investigating Watergate.
From the mastering engineer's side of the debate, we get a more concrete viewpoint. John Vestman, whose opinion is representative of most of the pre-mastering engineers with whom I've dealt, insists that all his jobs be transferred to glass at 1x. In fact, he gives his clients a booklet that underscores this point IN RED.
So who's telling the truth? Obviously it takes more time to transfer 50 minutes of music in real time than it does at high speeds. When you're doing about 200 glass masters a day, time literally is money. So it stands to reason that manufacturers are going to favor an opinion that says 1x is unnecessary. And conversely, those on the mastering side have a vested interest in hanging their work on an esoteric process that only someone with trained ears can appreciate.
Tony Van Veen, VP of sales and marketing for Disc Makers and always the moderate said, "While there is some audible advantage to 1x mastering, it is usually reserved for recordings with huge major-label budgets." Bradley states that 1x requests come from only 1% of his jobs at Sony. And even these are merely the label giving in to a specific request by a very particular act, usually a rock act, and not a Jazz or classical artist as one might assume.
For the tie-breaker, I spoke to Frank Farance, who has owned a top computer consulting firm in New York since 1981. After all, this is really a conversation about data, not music, right? Wrong, according to Frank. The nature of the data in music files - as opposed to data in an MS Word file - requires extra care in the transfer. He disagrees that, as a general rule, a 1x process could introduce more errors than at 4x. "Unless," he adds, "there is something unique about the machinery used at Sony. Since they invented the process, they might know better."
So, should this matter to you? I've mixed and mastered more records than I care to remember, and I believe that if producers and labels spent as much time obsessing over audio quality as they did about talent we wouldn't even have this debate. A 1x transfer is not going to help sell more records. Fact. Good music will. If you're really an audiophile with bucks to burn, spend it in the mastering lab tweaking your product, not the transfer itself. That's where I believe your money will translate into a perceptible difference.
Some comments by several CD manufacturers will be posted on my website in their entirety in the next month, including some answers to the bigger question: Is there something about Sony's method that would create this counter-intuitive logic that faster is better?
OBITUARIES
Freddie J. Perren: Legendary veteran pop/R&B record producer and songwriter died of a stroke after battling a long illness on December 16, 2004. He played a significant role in the creation of many Motown productions, such as the Jackson 5's, "I Want You Back," "ABC," and "(Stop) The Love You Save." He was 61 years old.
Darrell "Dimebag" Abbott, the guitarist of Grammy-nominated metal band Pantera was fatally shot during a performance in Columbus Ohio on 12/08/04. Three other people also were fatally shot before police shot the gunman to death. He was 38.
Terry Melcher, a record producer and songwriter who aided the careers of Ry Cooder, The Byrds and The Beach Boys. Melcher was the son of actress Doris Day and struggled a long battle with melanoma. He died on 11/20/04. He was 62.
Cy Coleman, composer of the Broadway musicals "Sweet Charity'' and "City of Angels" as well as such pop standards as "Witchcraft" and "The Best Is Yet to Come," died of heart failure on 11/18/04. He was 75.
Mo Out.
[ to top ]
