This review page is supported in part by the sponsor whose ad is displayed above

Reviewer: Srajan Ebaen

Of masters and slaves
When it comes to music software, we're all slaves. Not just slaves but descendents of slaves. The down-trodden.
Think crapillions of CD slaves stamped out in the robotic pressing plants from the original glass masters where duplication speed is king. We're slaves to these slaves' inherent physical imperfections. One phenomenon that bona fide digital researchers agree upon as existing outside of sheer imagination is pit jitter. That ain't wobbly female knees over Brad Pitt. It's imperfectly formed digital pits. It's somewhat analogous to a worn record groove whose walls are no longer perfectly perpendicular; whose floor is warped; whose floor/wall corners are out of true. Except that with CDs, it's not physical wear on the software we're considering now. After all, data pickup via laser doesn't cause physical contact. (And yes, you can certainly scratch or otherwise abuse a CD to render it unplayable or at best, with serious drop-outs). No, we're talking wear on the software creation stampers. Or inferior stamper execution in the first place. High-speed duplication errors. Other loose tolerances during the manufacturing process. Think rubber stamp that wasn't fully saturated with ink. Its stamped lines and edges aren't as crisp on the paper as they should or could be. xrCD demonstrated in the past how ultra-strict manufacturing protocols in the commercial CD pressing process can pay huge dividends. No voodoo, just precision.


You see, many high-speed processing plants use stampers with deliberately sloping edges on the land/pits so the polycarbonate disc will easily detach from the press. Further, submerging a CD under water demonstrates the permeability of its layers. This lends credence to the theory that certain substances like mold releases and cleaning solutions can get trapped between the CD substrate and reflective shiny top layer during the manufacturing process. It is those don't-belong inclusions that may get removed/evaporated by brief flash exposure to very bright light (the Nespa treatment) afterwards.


The upshot of pit jitter? Read errors. Not laser missiles but laser misses. Better formed pits are said to result in less errors. Read without squinting. Better sound. Or so the theory has it. Puzzling it's been to the slaves who would become masters again - audiophiles who discovered by various means that CD duplication can often spawn rather better-sounding offspring than the commercial originals. Genetic leapfrogging. Think of this as underground home-based remastering. It's not remastering in the sense of added reverb, altered dither settings, increased word lengths or other parameters - though the Alesis MasterLink for one is perfectly capable of that. It's simply copying commercial CDs such that the copies sound better than the originals. Whatever you want to call that.


If you've never EAC'd, Plextor'd or flashed your CDs; never circumcised 'em, green-edged 'em, spit-polished them with squalene; never cryo'd em; never George Louis'd 'em; never applied any other wacky or not-so-wacked treatment on them... it does quite suggest audiophile weirdness and wishful beliefs in magick. A few measurements below show that something's going down with data changes. What it all means, what mechanisms caused it - that's best left to the serious scientists and lunatic fringe theorists. The purpose of today's brief is to merely corroborate what a lot of music lovers already know and alert those who aren't in on it yet: there's more gold hiding in them thar poly-carbonacious spirals than you might have mined yet.


For our little adventure, Marja & Henk provided the original commercial pressing; an EAC'd copy; and an EAC'd copy subsequently Nespa'd. Go to their Exact Audio Copy and Nespa #1 articles for the skinny on those processes. To understand what you're looking at with the graphs, Marja & Henk submitted the following:
"CoolEdit's Spectral View is most useful for analyzing your audio data, to see which frequencies are most prevalent throughout the sound file. The more abundant a frequency (the greater a signal's amplitude component within a specific frequency range), the brighter the display color. Colors range from dark blue (next to no frequencies) to bright yellow (frequencies in this range are very strong). Lower frequencies are displayed near the bottom of the display and higher frequencies near the middle or the top. The display is linear. White lines on the left and right divide the display into 1/2, 1/4, 1/8 and so on. The very top of the display represents frequencies at just below the Nyquist frequency. If a bright spot appears near the top of the display for a signal sampled at 22KHz, the frequency being represented is near 11KHz (from the CoolEdit Help menu)."


Added Henk: "When you study the three spectral views, focus on the rightmost part of the graphs. Here you see EAC is cleaning up the dark blue background (noise?). After Nespa'ing the EAC'd CD, some noise is back as the graphs show. Does this correlate with what you'll hear?" Being a frugal Dutchie, Henk didn't tell me. I'd have to work it out myself. Thankfully, the CD in question was Renaud Garcia-Fons' newest Arcoluz [Enja-9478 2]. It's an epochal live recording at the very peak of what's musically possible from 5-string contrabass, Flamenco guitar and percussion in a super-virtuoso trio context. The blank CDs are MAM-E Prostudio branded, the original data was imported with Exact Audio Copy, written with Plextor's PlexTools software and one of two discrete burns off the same imported files was then finished off with the Nespa lumen treatment.

Spectra from the original, track 1, album Arcoluz
Spectra from the EAC'd copy
Spectra from the Nespa'd EAC copy
Original wave
Mysteriously, all RMS power figures are very slightly lower with the EAC'd copy
The Nespa treatment makes zero alterations to the EAC'd copy
The dinosaur signal hump with the right-channel red spikes from the original
The dinosaur has clearly submerged/attenuated some after being EAC'd. Bye bye Godzilla signal?
Virtually identical with the EAC'd version after Nespa flash treatment. Godzilla sticks around.

The measurement-über-alles crowd might be prepared to grant us something now. There could be audible differences between original and EAC'd version. After all, the graphs and figures aren't identical. Yet they'd absolutely refute any possible audible differences between EAC and Nespa versions. Those graphs are identical. Predicting what the measurement offsets should mean (since EAC data acquisition and subsequent PlexTools software writing incurs certain signal changes which you'd be inclined to call the artifact of a lossy transfer; signal loss couldn't possibly be a good thing - if that's what it is)? Knock yourself out predicting what it should mean and will sound like. It's all dice.


If I can't hear it, the graph means nothing. If I can hear something and the graph says I shouldn't; or suggests I should hear something negative when I'm hearing a clear improvement? It's a meaningless graph all the same. If graph and listening do correlate? It's ego gratification and a faint notch in the ol' audiophile belt. But in the end, the graph really remains meaningless. After all, I can't hear it. I can't even eat it. Graphs can be fun to conceptualize on these matters. But without a true engineer's grasp, that's entertainment at best. Window dressing. This by way of stepping back again from the measurements to rely on purely subjective listening. The field is certainly wide open for all manner of conjecture. Brighter digiphiles might have ideas on how to correctly interpret such graphs.


I'm here merely to tell you that home-based 'remastering' -- to improve the audio experience within your own four walls, not to get into wholesale pirating and on the FBI's wanted list -- works. It only takes desire and time and blank CDRs (and not all of those sound the same either, perhaps a subject for a future brief). The EAC data acquisition process combined with slow-speed writing/burning makes for better-formed land/pits which aren't imprinted with CDRs like on original pressings but written in ink. It's a post-facto fix for a product you paid good money for, under the normal assumption that you were buying something well made. Anyone with ears will have to conclude that because home-based rewrites of your favorite CDs can improve their sonics -- and you're not doing anything other than rewriting them by copying -- something in the commercial high-speed duplication process causes sub par results. Imagine having to redo the seams on a brand-new pair of Jeans. You just bought 'em at full designer retail. Wouldn't you scream bloody murder?


Though I merely had one CD for this brief -- interestingly enough, a far-above-average commercial pressing to begin with and packaged with a live-concert DVD -- Marja & Henk have been at the personal compilation burning gig for many moons. They've built up quite a cache of subjective evidence. Most recordings benefit from "remastering" - some rather drastically, others less so. It's the rare one that doesn't improve to some extent, somewhere, because it was so flawlessly pressed to begin with. Right. What you should take home with you is a perhaps radical seeming notion. Most commercially available CDs can be improved upon by simply using EAC for data capture, PlexTools for file writing and Nespa for optical finalizing after you've burned your CDR. Try it out. It's cheap. EAC is freeware in fact. No need to believe any of this outré stuff. Your own experience is so much better - and the only one valid for you.


So - how then did the two copies (whose graphs tell us that they're identical) compare to the original (whose graphs tell us that its copies aren't 100% perfect clones)? What delta of difference did I hear ? Was it just different? Did portions of the difference go retrograde to cause complaints? Or was it all a free ride to full glory? 'twas John Wayne at his best.


Under the right circumstances, the EAC/Nespa version makes the original sound downright broken while the EAC-only version gets you about 75% of the way. Right circumstances simply means a system of sufficient resolving power to magnify these changes. One evaluation of the three CDs occurred in the presence of an experienced audiophile. He had no clue what I was up to. His setup ends with a 5-way speaker of four horn-loaded ALE compression drivers (including 30' long straight bass horns) and Acapella ion tweeters. The system is about 110dB efficient. It's ultra resolved, very fast and ferociously dynamic. It's in a good-sized room to develop appropriate scale.


I played its owner the fully tricked-out CD first. This netted an unsolicited comment. "Unbelievable recording". It bloody well is. Then I offered that we had listened to a copy. Would he like to hear the original? He looked dubious. "Copy?" A few seconds into the original, he began to accuse me. Of trickery. Surely this was the copy and the other the original. The difference was that dramatic. He only believed that I wasn't setting him up when the intermediate EAC-only copy got spun. It was clearly better than the original but not as good as the Nespa'd one. These graduated improvements made the thing real for him.


The best copy had far more energy than the commercial pressing. This was apparent in transients with their higher shock value, more accentuated edges and better focus. It was apparent in more pronounced dynamics. Lastly, it was apparent in more overtly audible space as heightened ambience or 3D context. All these apparencies added up to more - more juice, more color, more life, more realism. Reverting to the original became an exercise in tamer, bleached, color deprived and spunk deficient. The lesser copy repaired most of the ambience but didn't equal the transient incisiveness and dynamics of the fully optimized version.


In my own system, the same differences applied. The magnitude of stretch between the three versions simply shrank. My rig isn't hornloaded. It's not as resolved and, though very dynamic, not as charged. The advances in percussive attacks and sheer dynamic scaling of the fully 'remastered' copy were maximally heightened by the Xtreme rig's specific strengths. Those acted like a dynamic expander. Its results rendered the original plain unlistenable once you'd heard the Nespa'd EAC version to believe it was the original.


In my less extreme setup, the trick copy was clearly superior. Just not as superior. Less overall resolution and dynamic envelope, less magnification of these qualities. But it's fair to say that any Joe off the street could tell the difference. It's that obvious. It's a pisser alright. That's how a commercial CD should sound, right off the shelf. But for a change, it's a pisser you can do something about - without asking for a refund or throwing a tantrum with the store manager. Rather, acknowledge the reality of CD manufacturing. It's got demands on quantity and speed. It is what it is. Without it, we'd have no music. Simply plan on earmarking your favorite CD stack for finessing afterwards on your PC with its dedicated hard-drive, PlexTools and EAC freeware. It's not news, really. We've been talking about this for nearly two years. As have many others. It's simply a good subject to talk about again.


The fix is essentially free (the Nespa box for the full monty is a one-time expense but stretched out over each treated CD, it'll end up being pennies). The fix is brilliantly effective. The fix empowers audiophiles to get hands-on involved rather than crying foul, dreaming on or otherwise avoiding action (but the Nespa treatment won't yet give you the backup of measurement changes, at least not the kind of measurements we took). For once, we get to rectify a situation on our own. No special skills, advanced knowledge or secret handshakes required. The man who would be king. The slave who would be Cesar. Audiophile self help for legit aggrandizement. Pass it around then on the forums and to friends who aren't hip yet or believe it's more imaginary stuff not worth the bother. This fix is real and thus very much worth the bother.


Remember you read it here first. Not. I'd rather hope you've read about this topic before - multiple times. It might just take repeat exposure to start taking it serious. At first blush, it does seem rather preposterous that a copy -- a supposedly pit-perfect clone -- could sound better than the original. Preposterous it may be. But there's plenty of folks who agree that it's audible. Which side of the fence you choose to sit on is up to you. But shouldn't you at least give it a try to have your own opinion based on personal experience?