Parts One, Two and Three of this series appeared in Issue 147, Issue 148 and Issue 149.
In listening to the direct-metal-mastered Stockfisch Records DMM Dubplate, Vol. 1, I almost couldn’t believe that the source was “only” a CD-quality recording, being so used to hearing substandard CDs and CD-quality recordings over many decades now. As with all kinds of audio technologies, it is not that the CD cannot possibly sound good. If anything, Stockfisch Records has done the medium a huge favor, proving that it is actually capable of good sound, if properly done. However, the CD as a medium and digital recording technology is more commonly associated with lowering the entry level in what used to be a very exclusive industry of audio recording and mastering. It was and still remains much cheaper and much easier for less-skilled and less experienced individuals to produce a recording in the digital domain, especially nowadays that computer ownership is so widespread, allowing pretty much anyone free access to software that will somehow record some sound in CD quality. It is much simpler, easier, cheaper, and almost an entirely automated process to burn the result on a CD using that same computer, and play it back on a CD player.
Even doing a factory run of 500 CDs is way cheaper and easier than doing a run of 500 vinyl records, with its complicated mastering process that requires equipment that is beyond reach for anyone but the most determined and serious audio engineers. This already filters out those that do not intend to really devote themselves to learning the art of disk mastering. I mean, few people in their right mind would invest six-figure sums to purchase industrial equipment weighing in excess of 1,000 lbs. to just randomly press buttons and see what happens, not to mention incurring the risk of explosions, fire, electrocution and having to overcome an aversion to screwdrivers. Avoiding all this was one of the major appeals and selling points of digital recording technology when it was introduced.
Digital recording was cheap and only needed that computer that you already had anyway. No heavy equipment, no major investment, not even what used to be standard equipment, such as mixing consoles and rack-mounted outboard gear. Digital audio workstations of a rudimentary nature were offered as free perks with many operating systems, or could be freely downloaded, including virtual mixing consoles and plugins that would replace the outboard gear. For many enthusiasts, this became a hobby, clicking around just to see what happened. There would be nothing wrong with that, if a clear separation was maintained between the professional industry, and the hobbyists.
What happened, though, was that countless overly self-confident hobbyists realized that they could get some sort of sound coming out of their computers, so they started believing that they were in the same league as the professionals and some even began selling their services as such to an unsuspecting audience that knew no better. Even worse, several short-sighted individuals in managerial positions in the industry saw this as an opportunity to increase their profit margins by eliminating what used to be a major expense in the production of an album: professional recording and mastering facilities and equipment, and suitably qualified engineers to operate them.
On the other hand, what Stockfisch Records has proven is what CD-quality digital audio can sound like in the hands of suitably qualified professionals, working in professional audio facilities packed full of very expensive equipment, using extremely high-quality mastering, and playback media.
Remember this every time you hear a bad-sounding CD.
There are several questions raised by this, though. If a CD-quality source can really sound this good, do we still need the higher sampling rates and bit resolution?
I really do wonder what the DMM Dubplate, Vol. 1 or a disc made in a similar manner would sound like using high-resolution digital sources (perhaps DSD recordings), or all-analog recordings from magnetic tape (Günter Pauler told me that this would be technically possible to do at their facilities, but he is not a big fan of tape, so we are not very likely to find out soon), or, even more ambitiously, direct-to-DMM-dubplate recordings, eliminating all other steps in the process.
Needless to say, you could only produce as many direct-to-DMM dubplates as the number of lathes in the facility (two in the case of Stockfisch Records), of each performance. For more than that, the performance would need to be repeated until the musicians collapsed (and of course, each performance would be different).
But I guess this would be the ultimate in pushing the boundaries of analog recording technology.
The other question is whether we need the DMM Dubplate, Vol. 1 at all, if the digital source is already that good. Why not just listen to it in its digital form? In the digital domain, the 0s and 1s that comprise the digital audio signals will (in theory at least) remain identical (sample rate conversions, sometimes occurring under the hood, will certainly change that). However, the process of reconstructing an analog electrical signal from these 0s and 1s relies on the digital to analog converter (DAC) at the consumer end. Likewise, the extraction of the information from the grooves of a disk relies on your turntable, tonearm, cartridge and phono stage.
If you have an outstanding disk reproduction setup, but only a mediocre DAC, then you are better off listening to the DMM Dubplate, Volume 1 (or other high-quality LPs cut from digital sources) with your turntable setup. If you have a top-of-the-line DAC but only a budget turntable, stay with the digital files.
If you have the finest examples of both…well, just listen to both and compare!
Just to make things even more interesting, one of the tracks on the DMM Dubplate, Vol. 1, “No Sanctuary Here” by Chris Jones, was also featured on Stockfisch Records’ DMM-CD/SACD Vol.1, released in 2013. The same digital source was used to transfer to a DMM disk, as with the DMM Dubplate, Vol. 1, but then the DMM disk was played back right there on the VMS-82 lathe vacuum platter, using an EMT tonearm, cartridge and phono stage. The signal was then fed to an analog to digital converter and converted back to a digital 44,1 kHz/16-bit file (to create the CD layer) and a 2.8 MHz DSD file (for recording onto the SACD layer). You could then listen to a digitized version of the sound that came off the DMM disc. This was meant to somehow do something good to the sound, and “improve” upon the plain digital source. The end result (the DMM-CD/SACD Vol. 1) did also sound good.
I have yet to encounter anything that Günter Pauler has done that did not sound good. But I am not very convinced that the sonic “goodness” of the DMM-CD/SACD Vol. 1 came from the DMM transfer. Compared to the DMM Dubplate, Vol. 1, DMM-CD/SACD Vol. 1 sounds somewhat less exciting, which was not much of a surprise; the additional and unnecessary analog to digital conversion and subsequent digital to analog conversion is a process that’s never done anyone any favors. Any conversion or transfer for that matter that can be avoided, is best avoided. In this case, given a top-of-the-line DAC and disk playback setup, I would expect the digital source to sound slightly better than the DMM Dubplate, Vol. 1. (You can compare for yourself, as both the DMM dubplate and the digital source file are available from Stockfisch Records.)
The Stockfisch Records cutting room.
A direct-to-disk DMM dubplate (as opposed to one sourced from a digital or analog magnetic tape master) would most probably be the ultimate recording medium of all time. The commercial viability of creating such disks would be questionable at best, but in terms of sound quality alone, I would think that a direct-to-DMM dubplate would be about as good as any form of sound recording can get, all technical points considered.
But what if we were to do a direct-to-disk recording on lacquer and on DMM at the same time? Which one would win?
Unlike the nickel mothers grown from processed lacquer master disks, which are ferromagnetic and as such, impose restrictions on the type of playback cartridge used to reproduce them (the stray magnetic field on the underside of a typical moving coil cartridge would create strong forces of attraction between the cartridge and the nickel mother, making it impossible to maintain proper vertical tracking force), the lacquer disk itself does not have such issues (it does have an aluminum substrate, but this is not ferromagnetic). On the other hand, lacquer disks are soft, they tend to deform under the playback stylus (low vertical tracking force (VTF) cartridges offer an advantage here, and a line contact stylus is an additional bonus, maximizing the area of contact between stylus and groove), and they have a very limited lifespan (about 10 plays before audible deterioration).
The DMM disk, on the other hand, is hard enough to have a reasonable lifespan and lower deformation under the stylus, while both the copper recording layer and the stainless-steel substrate are nonferromagnetic, allowing various types of cartridges to be used. It almost sounds perfect, but there is a little-known fact that unfortunately stops us from reaching the highest states of enlightenment and sonic bliss.
To achieve the required surface finish of the copper groove, the machining parameters at that surface speed range (linear cutting velocity) call for a significant positive rake angle (the angle formed between the cutting face of the tool and the surface of the blank disk) of 15 degrees. Lacquer has traditionally been cut with a stylus rake angle of 0 to 1 degrees. Playback cartridges are usually designed to have a stylus rake angle of 0 to 1 degrees, to match the angle of the cutting stylus, for accurate tracing of the groove wiggles. Similar to the concept of vertical tracking angle (VTA), which has troubled many a self-respecting audiophile and, at times, every major and minor publication devoted to high-fidelity audio, the geometry of the playback stylus must match that of the cutting stylus, or distortion will start creeping in. Just as the vertical plane of modulation (VPM) at the time of cutting a record has to be in close agreement with the vertical tracking angle of the reproducing system, the stylus rake angle (SRA) at the time of cutting should match the SRA of the playback stylus.
Neumann and (record label) Teldec were of course fully aware of this, and addressed the issue in German overcomplicated engineering fashion. The electronics package of the Neumann DMM system came with a pre-distortion module, adding the inverse form of distortion to the signal, of what would be expected to be generated upon playback, so that it would cancel out…if everything worked exactly as calculated!
J.I. Agnew with a Neumann cutting lathe.
Since the distortion generated is dependent upon the exact shape of the playback stylus, marketing and statistics came to the rescue. Most listeners at the time of development of DMM technology were using spherical styli, so the pre-distortion system was designed to compensate for spherical styli of a specific tip radius.
If yours deviates from that, then the mathematics won’t quite work out as intended! Curious, as I always tend to be, I played the DMM Dubplate, Vol. 1 using a spherical stylus as close as possible to the original Neumann specification, and I also played it with my favorite line contact stylus, which is as far as a stylus can get from what the engineers at Neumann were wearing out their slide rules on. I think it sounded better with the line contact stylus, but I cannot explain why. It would take a few DMM dubplates containing test signals to conduct proper laboratory measurements with, and make conclusions regarding the magnitude of error and the exact form of distortion generated in practice, using different calibrated playback systems. There is quite a bit of intense science behind record cutting and accurate playback.
Listening to direct-to-disk recordings on lacquer disks (unprocessed disks, not the ones that have gone through the plating process used for mass manufacturing of vinyl records) has been a fascinating experience for me, as has been listening to the metal mothers made from my lacquer masters. But I know of no examples where a direct-to-disk recording has been simultaneously done on lacquer and DMM.
While vinyl records made from direct-to-disk recordings are perhaps the most sonically rewarding mass-manufactured audio media available, I expect that direct-to-DMM dubplate recording would possibly be beyond anything that has ever been attempted in the world of audio.
Has it already been done?
To the best of my knowledge, not yet. Even the Church of Scientology is, I believe, using recorded versions of L. Ron Hubbard’s speeches to transfer to DMM.
One major obstacle to attempting direct-to-DMM recordings is the lack of enough Neumann VMS-82 cutting lathes, as the VMS-82 is the only machine ever designed to cut DMM. However, modifications are always possible, and the author knows of at least one Neumann VMS-70 lathe that has been successfully converted to cut DMM. This is a recent development, prompted by the shortage of lacquer disks following the 2020 fire at the Apollo-Transco plant in Banning, California. I expect we will be seeing plenty of innovation and a lot more pushing of the boundaries in the years to come. In many ways we live in very weird times, but these are also exciting times indeed when it comes to audio. Perhaps spending more time at home due to the pandemic will rejuvenate the appreciation of high-fidelity sound and the importance of recorded music in general.
If you have a good record playback setup and want to discover how far the boundaries of grooved media can be pushed, the Stockfisch Records DMM Dubplate, Vol. 1 comes very highly recommended. It is a great recording of great music, using top-shelf equipment, transferred to DMM by one of the world’s few remaining extremely knowledgeable and experienced audio engineers and his very capable team.
My copy even has my name on it!
Handwritten, to drive the point home even further!
0 comments