Cracking the Columbia Records Code

You can find a great deal of information online about matrix codes and their meaning. Unfortunately some of it is incorrect.

For example a post on the often useful Steve Hoffman Forums claims that Columbia “cuttings” are A-1st, B-2nd, C-3rd through L-11th and then AA is 12th “cutting” with AB being 13th. AA through AL would be cuttings 12-22, BA through BL would be 23-33, and so on.

However, that is incorrect according to Phil Brown, a Columbia Records cutting engineer who recently contacted analogPlanet. He worked for the label in the 1970s, so in addition to lacquers, Dolby tape copies were included with the lacquers for “file recuts and tape duping”.

According to him: “ An initial lacquer order was 6 sets plus a ref for QC and 4 Dolby copies (of course. 1A&B went to Pitman (N.J.), 1C&D to Terre Haute, (IN) 1E&F went to Santa Maria (CA).”

In other words the claim that “A” is the “first cutting” “B” the second and “C” the third, is incorrect. In fact 1A-F were all “first cuttings”.

This explains why one could find a record with “-1A” on one side and “-1B” on the other, since both were sent to the same Pittman pressing plant.

According to this individual, “ Any change in program triggers a -2. another change, -3. I saw a lot of -2, never saw a -3. In other, more common systems the -2 would be an RE-1.”

A “change in program” could be a revised mix of one or more tracks or a different take entirely. For instance, the first edition of Bob Dylan’s Highway 61 Revisited (CS-9189) had a different version of “From a Buick Six” than did later ones.

So the 1A-Fs had the original fast version, while the “-3” had the slower one, no doubt changed per Bob Dylan’s request. I have a -3 so I can be sure about that, I don’t have a -2 so I can only speculate about what was changed there. It could have been the “From a Buick Six” change and perhaps -3 has yet another change that I’m not aware of.

Interestingly, you can read detailed descriptions of this album and its recorded history, for instance this one on Wikipedia, yet it does not reference the substituted take on the later pressing.

I have a -3A of Blonde On Blonde that I bought shortly after the record was first released. It came with the cover sticker, but good luck figuring out what “program changes” were made between it and the original -1A-F, though if you want to see “obsessive-crazy” with no particular purpose other than to point out label and/or jacket variations check this out

It’s also possible that if the first lacquer cut was ruined the next cut was 2A and that was really the “first cut”. That is how it was done at EMI for instance.

According to my source: “Each plant got 2 sets because it's easy to lose a lacquer in plating. All the majors sent 2 sets to each plant. The reason is that it saved time if they lost one and we were always up against release schedules.”

“The only time I cut 1 set was for indies. We would sometimes at Columbia and later almost always at Warners have one plant do the plating for everybody. This is central plating. At Columbia it was almost always Santa Maria because they had the newest-and best-plating operation. The masters would still carry the dash number for the plant doing the pressing. Central plating generated the mothers that the plants would then pull stampers from. At Warners they bought Sheffield Plating which did plating for all the plants.”

The system may have been different for mono or different during the ‘60’s because I received this image from either Sundazed or Sterling Sound when the mono Dylan titles were mastered:

It shows but a single lacquer sent to each plant and the 1A going to Terre Haute. Perhaps by then mono was in decline and a single lacquer sufficed, but it doesn't explain why the 1A went to Indiana.

These symbols, often found in the inner groove area identify Columbia Records pressing plants: CT or CTH = Columbia, Terra Haute, IN
CP = Columbia (Pitman, NJ)
CS or CSM = Columbia (Santa Maria)

Incidentally the number found on the back jacket, right hand lower corner of all or most vintage Columbia jackets refers to the jacket manufacturer. It has nothing to do with the pressing inside (so #4 doesn't mean fourth pressing!)

jon9091's picture

Can you share who the mastering engineer was?

Michael Fremer's picture
I will have to ask him if I can use his name!
Michael Fremer's picture
Phil Brown (with one "l")
gMRfk6LMHn's picture

I would like to compliment you on the interview with Phill (with two l's). It was probably the best interview you have ever done, period! You 'teased' a lot of information from him in a very relaxed atmosphere.

James, Dublin, Ireland

gMRfk6LMHn's picture

I would like to compliment you on the interview with Phill (with two l's)Brown. It was probably the best interview you have ever done, period! You 'teased' a lot of information from him in a very relaxed atmosphere.

James, Dublin, Ireland

jon9091's picture

And Phil. Very valuable info.

azmoon's picture

I have a 6 eye with this in the deadwax: XSM47326-1AA on side one. Is it confirmed that AA means 2nd pressing? XMS must mean the pressing plant.

W B's picture

-1AA was the 12th lacquer cut over the life of this album's being on vinyl (we would appreciate where this lacquer landed). 'XSM' was the stereo code used by Columbia for 12" LP's ('XLP' - or as they put it on their labels in those days, 'x"Lp" ' - was for mono; on Epic the codes were XEM for mono and XSB for stereo.)

Michael Fremer's picture
Was Columbia's matrix code prefix. All pressings would have that. 1AA means after they got to 1L, they would go to 1AA, 1AB etc. So it's really an early pressing since 1A-1F was "first lacquers cut" (2 each to the 3 pressing plants) and then 1G-1L would be second set. 1AA would be third set but 1 means still same "program" or tape.
pessoist's picture

is useful information. thank you Sir.
Still, I wonder when you'll do the research and publish a book related to such kind of information filled with some of your stories. Would rip it from the Bookshelves instantly.
thank you.

alucas's picture

you should do a book, you have a lot of good stories and you tell them so well. very entertaining! thanks for the info. although i buy all nm to mint originals, i still have not gotten into the run-out grove info ...yet. i am just thankful i can get a really clean copy!

Michael Fremer's picture
Can be tricky! For instance you can buy two "originals" of Buffalo Springfield "Again". Both with the "original" label but one will sound awful and one spectacular!
Bigrasshopper's picture

I gave up trying to find a clean original copy of Buffalo Springfield Again a couple of years ago simply because the first well used copy I came across made any further pursuit seem pointless, it sounded bad. Are you suggesting that there is a matrix number that can guide me to a copy worth keeping ?

Michael Fremer's picture
Ironically, in this case steer clear of the one with the Columbia stamped matrix info. It's awful yet it has the "original" label! There's another one mastered at Atlantic that sounds great.
Bigrasshopper's picture

This doesn't look like something that's going to be casually available on Discogs, anyway. Looking over the listings there, there are several US variations all from 1967. There are mono and stereo variations and pressing plants all under the ATCO label. There is listed a mono Monarch pressing. Where do you find " Original " printed ? Discogs lists an "original" but doesn't say if that's printed or what. Just to make this simpler, what are the matrix numbers and pressing plant on the example that you find pleasing ?

Michael Fremer's picture
as you might know, Atlantic/Atco was one of the few labels to put the pressing plant info and matrix code on the label. So avoid (under "Bluebird") (ST-C671117CT) and that stamped in lead out groove area even if followed by -1A. The CT means it was pressed at Terre Haute plant. The one you want has handwritten ST-C-671117- (A rubbed out) C on side 1 and "D" on side two. Also has "AT and LW" hand written elsewhere on the lead out groove area (as do many Atlantic/Atco cuts done at Atlantic Studios. THAT is the one to have. I also have a yellow label version some would say is a "second pressing" but it has the same info as ST-C-671117- (a rubbed out) C and D except it has "PR" on the label, which means it was pressed at Presswell, Ancora, NJ (RIP)
parman's picture

I have been getting into researching matrix info for the last few years and have thought putting the matrix on the label was the way to go. So much easier to see and sometimes on the etched deadwax I can't tell a 3 from a 8 or a / from a 1 or an I

Bigrasshopper's picture

It would seem that the greater my record collection expands, the less I am able to recall anything specific about it. I am not a record collector who is compelled to pursue it out some intrinsic capacity of my character, but only because I enjoy it when I am rewarded with emersive music. So I'm willing to wade into details only when I have some sign or have nurtured and developed a hope, a real or imagined expectation of some reward. In fact when I consider the great sounding records that are out there to be heard, that I will surely never hear because it would involve to much work, or risk, going out on a limb in the dark, I try to overcome that sudden sense of depression by reminding myself that my soul will remain unaffected by this lack, even though I am surely not fully aware of it, that is, my soul. So I agree to for the present to remain unaware of that musical reward for the sake of peace, complacency or sleep while the soul rests in itself as its own reward. I take a sigh and let go, ease my grasp, release a spark, and close down that webpage. But when I'm immersed in music I am very much aware that that too is is own reward.
I suppose that's a way of saying I appreciate you for lighting the guideposts that you do and sharing your valuable experience. You are one those who has a unique set of qualities and are in a unique position to use them. As it happens I listened to the record I thought we were talking about last night and again today and was glad to discover that my copy sounds very much present and satisfying. It must have been a different Buffalow Springfield record that had disappointed. My record is similar to yours, has the same scribed matrix but has a scratched "A" followed by CC. The pressing plant is indeed printed on the label MO following the matrix. Apparent Atlantic / ATCO only started doing that, in Janurary of that same year, 67. Which is really handy. But not all of them are printed there, perhaps the first runs. I did find one with similar marks as your own, am curious to compare, but have to wait to hear back.
This album has great energy and to me seems like a easy candidate for an AAA reissue from Rhino mastered by CB if the tapes are still around. It has simpler more rootsie vibe but in many ways similar to to what came later. I think it's essential.

gMRfk6LMHn's picture

It's funny I have been on the Steve Hoffman forum since 2003 and I have a quiet chuckle to myself. I see members who didn't have a clue what the word 'deadwax' meant a couple of years ago have become overnight experts. I have been following the 'deadwax' since around 1984 when I became 'aware' of all the information that is on the deadwax and to this day I am still trying to figure it all out! I agree with the comments already made that you (Michael)should definitely consider putting all your knowledge down on paper!

James, Dublin, Ireland

Corsair's picture

Thanks for the info!

OldschoolE's picture

Michael, thank you so much for posting this information!! Of course I am still trying to figure out the deadwax (I prefer the term "runout area") codes. Now I can go back and look at my Columbia records and try again when I get time. (I have a couple or more that look like the picture). I'm also delighted to know that the Hoffman forum is a key place for the matrix info because I was never sure what source was best. It's even sweeter that you made this correction.
Thanks again Michael.

alucas's picture

fine, i will start learning about the dead wax info and than start checking my collection...maybe. you should have a section on this site on dead wax info with pictures of different labels. it could be one big comprehensive learning area. in the mean time i am off to hoffman's site.

Ortofan's picture

... Based on the handwritten notation indicating "50 cy[cle] cut-off", it would suggest that this was not a standard practice.

Michael Fremer's picture
The Dylan record has massive amounts of bass, popped "p"s etc. The new Mo-Fi reissue has no rolloff and it sounds spectacular but I guess they worried that back then it would create's not standard to do that!
Martin's picture

Is that the mono or the stereo???
The stereo has "Mobile Fidelity Sound Lab" on it
The mono has "Original Master Recording"

Your recommendation will define which of these I will go pick up.

Michael Fremer's picture
All of the Mo-Fi Dylans are really good. The four monos are outstanding and the versions Dylan cared about.
Martin's picture

I'll tell the very friendly and helpful lady at Elusive Disc to put the mono in with my next order :-)

Your recommendations get acted on.

I also ordered "The Culprits Blues" from Nick Culp. Both vinyl and hi-rez. download :-)

Ortofan's picture

... The handwritten notation indicating "50 cy[cle] cut-off" would seem to suggest that this was not a standard practice. Is that correct, or not? Was it applied only to certain recordings? Did the cut-off frequency ever vary? What was the slope of the filter? Did the master tape have the full frequency range and was the high-pass filter inserted when the disc was mastered?

Michael Fremer's picture
That tape was bass-loaded I'm sure on purpose with Dylan popping many "P"s. It's possible that cut off was done just for that record. The Mo-Fi reissue doesn't cut it off and the bass is intense and popped "p"s many!
MarkLeviton's picture

I worked with Phil (definitely one "L") Brown at Warner Music Group. He knows stuff. The "other" Phill Brown is also a fine fellow but don't confuse them.

IR Shane's picture

Michael sharing info like this with me over the years has deepened my addiction and the satisfaction I get from the hobby! Michael, is it time to revisit cracking the Beatles matrix code? That's the one that sent me waaaaay off the deep end a few years back. Sincerest THANKS!

W B's picture

. . . if not already on here, the byzantine matrix codes used over the years by RCA Victor, not only on their own releases, but also for custom clients. What each letter or number signifies would make your head spin.

pjc's picture

If you're not familiar with LJC, it's a great place to learn details of labels, matrix codes, etc. An extensive discussion of Columbia matrix codes here:

Michael Fremer's picture
Yes fascinating details there but misinformation about Columbia too as they pass on what's found on Steve Hoffman Forums, which is often incorrect info.
pjc's picture

LJC refers to Hoffman for historical context, a starting place for discussion, not as useful guidance for reading matrix codes.

Martin's picture

But I checked a couple of my copies of Highway 61 last night.
The stereo is a 1B / 1D copy. With the alternate version of from a buick six.
The mono is a 1A /1D copy.


Michael Fremer's picture
There was some stamper trading going on....that's all I can think of.
W B's picture

The "claim" about the dash numbers signifying how many lacquers were cut to that point (#A-#L being the first 11, #AA-#AL signifying 12th through 22nd et al.) was not "incorrect," but rather cumulative, amassed over a period of years over a record's life in print, and sent to different plants. It should also be noted that this formula was only applicable to lacquers cut by Columbia's New York studios (with their trademark stamped type in the deadwax, spelling out the matrix and dash numbers). It does not account for other studios that cut for Columbia or their clients (whether Sterling Sound or other Columbia studios such as in San Francisco or Hollywood), where I've seen such dash numbers as -1M or -1P or -1Q. I figured this out by years of "reading" dash numbers of LP lacquers whose titles were in print for decades.

But I am aware of the first six lacquers for a record (#A-#F) being allotted amongst the three plants Columbia owned at the time. (It should also be noted that that lacquer log sheet for Dylan's John Wesley Harding mono issue came from Columbia's Nashville studios, which cut the lacquers in question on a Scully 501 lathe [as opposed to the Scully 601 they used to cut stereo records]; what eludes me is who exactly was "B Mc" who mastered such sides.)

As for that #4 on the back cover: You might want to check Part 2 (dealing with albums) of Bruce Spizer's excellent tome The Beatles' Story On Capitol Records to figure out which jacket fabricator turned out that one. Chances are (not to be confused with Johnny Mathis' 1957 hit of the same name) that the same numbers on Beatles' album covers would also be applicable to Columbia releases.

As well, from 1954 to 1971 another set of lacquers (whether one or two for each side) went to the Quality Records pressing plant in Canada which pressed for Canadian Columbia in that time stretch. (After 1971, Columbia Canada set up its own mastering studios and pressing plant in Don Mills, Ontario.)

Maury's picture

for the additional info and clear explanation. It is unlikely to the max that large record companies used any totally consistent methodology devised by the central office for their deadwax inscriptions. Stuff happens and when the deadline is approaching people improvise. Unfortunately each record is like a person, similar but different to all others.

Michael Fremer's picture
the older the record the more consistent were these codes. That's true for Columbia for sure. Once the indie mastering facilities began cutting, all bets are off... but those did not use the Columbia stamp or matrix system anyway.
W B's picture

The in-house system with Columbia was the focal point of my decoding. I agree, with indie studios, anything goes.

This system would have dated to February 1952, when they first got the stamping machine they used for over 31 years (up to the end of 1983) to embed the numbers in the runouts. (Would by any chance Phil know what make of machine that was?) This machine they used did not have any letter 'I', and 'O' was used for both the letter 'O' and number '0'. That is why the first 11 lacquers cut and spread out over many pressing plants would have been -#A, -#B, -#C, -#D, -#E, -#F, -#G, -#H, -#J, -#K, and -#L. For some reason, when this was set up, their system never went to M or higher letters, hence after -#L came -#AA and then on down.

I once saw (but did not get, to my deep regret) a 1960's reprint of a deep catalogue Columbia Masterworks album where one side had a dash number of -1ABC(!). By this code dating to 1952 or so, this would have been the 146th lacquer cut of that side. If not of a zillion-selling Broadway cast album like My Fair Lady, then it was a case of a classical LP where huge numbers of lacquers were rejected and a few rendered suitable for sending to the plants. I think -1LJ is the highest I've ever had in my collection.

jokerman's picture

Someone I know disagrees with the information in article and I'm copying his words here. I'm not saying I agree, just wanted to see the response to his assertions:

"the idea of lacquer letters being identifiable to pressing plant is wrong. Plus the guy claims to have never seen a -3 tape cut which is ludicrous."

Any thoughts?

W B's picture

Within Columbia, it varied over the years where -1A & -1B cuts from the New York studio went. As of 1967-68 - and I'm saying this from my vast collection of 45's - -1A's and 1B's went to Terre Haute while -1C's and -1D's went to Pitman. By 1969 Pitman got -1A's and -1B's from said studios. Generally, from Columbia's Hollywood studios, -1A and -1B sets went to Santa Maria first. There is not necessarily any "hard and fast" rule as to which plants got what -1A/-1B sets.

As indie studios got into the picture, Columbia's New York studios would cut later lacquers (i.e. -1F, -1G, -1H) after the indies cut theirs. And it's not necessarily true that Columbia didn't "stamp" matrix numbers on "other" lacquers: in 1972-73 they stamped matrix numbers onto lacquers from Frankford/Wayne (for Ronnie Dyson's "One Man Band (Plays All Alone)" 45 on Columbia) and Atlantic Studios (for Looking Glass' "Jimmy Loves Mary-Anne" on Epic). In 1959-60 Columbia occasionally stamped matrix numbers onto lacquers from studios that cut for labels such as United Artists and Cadence.

old_school's picture

Find a good original- don't be fooled by the 180 gram reissue that looks exactly like the picture above. Its not what you want. I just picked one up and the sound quality renders it worthless. Unplayable. I cant even hear the horns thru all the hiss. Even after a RCM, it sounds terrible. Maybe mine was just a bad pressing but I've seen others post the same. Thanks to Mike for helping us find good original pressings.

crispi's picture

Another aspect worth mentioning is that not all early cuttings went to one of the 3 Columbia pressing plants – some of those first cuttings travelled over the Atlantic to be pressed by the European branches of CBS. From my record digging experience, more often than not, European pressings from the '60s contain very early Columbia NY cuttings. I have quite a few UK, Netherlands or German-pressed records that contain -1A or -1B cuttings.