What Makes a Good RIAA or Line Stage?

Hi Raul,

Indeed, holistic thinking is required - both in life and in hi-fi.

As you well know from how long it took you to get the Essential to this high level of performance, thinking holistically can drive you crazy as the circuit gets more complicated.

Not that the goal cannot be achieved, as you've demonstrated with the Essential, but the problem certainly becomes exponentially more challenging with each feature (or circuit block) you add.

I'd love to hear other designers' comments ...

To Doug ... a bit of embarrassment brings a rosy glow to your cheeks (grin).

Cheers,
Thom @ Galibier

Hi all,

It's great to get some brilliant designers involved in this thread (ain't sucking up ... I call 'em as I see 'em).

Whoah, Thom! Look what you started here.

I'm loving this discussion and hope that it will bring light to the general philosophy of how we achieve the goal of a realistic and SATISFYING musical illusion in our homes. I hope this is everyone's goal. Yes, I'm 5 small steps ahead of Doug in this area and miles behind you guys, and that's part of the fun in watching this thread unfold.

Indeed, the automobile metaphor was incomplete at best, but you need to kick off a conversation with a provocative statement. Like all designs, it is the entire system context that is important, as Nick, Jim, Ralph, Raul, and Jose all pointed out.

Jose brought up some great points, but I'm still puzzling through the difference between what Arthur Salvatore would call "noise floor" and "sound floor". My Artemis is not the quietest piece of gear out there (noise floor - it is quiet but not the quietest) and yet it has the uncanny ability to extract musically significant nuance from recordings (sound floor). Other great designs accomplish this differently

I too have read from more than one individual that a phono cartridge is not balanced, but rather floating single ended. Victor Khomenko's (BAT) writings over on Asylum are one source, and I've heard Jim comment about this on several occasions.

If I recall correctly, when you "balance" a phono cartridge (using a center tapped transformer, for example), you don't get the normal 6dB common mode rejection, because you are "halving" the signal in order to balance it. My explanation may not be perfectly clear on this.

I too have to wonder about the necessity of obsessing over small fractions of a dB in RIAA accuracy when in room speaker response varies by such huge amounts. I would agree with Jim however about left vs. right channel balance as far as treatment of the delicate stereo signal and its implications for imaging and such.

Now, I certainly understand that distortions build through the signal chain or through the gain stages in a single component for that matter - that if you were forced to choose in reducing distortions within an active amplification stage, that larger gains would be had by "improving" the earliest gain stage.

Surely you want to address all issues if you can, but the point here relates to the impact cleaning up the signal as early as possible. How much is too much, and when should we shift our focus away from trying to get the RIAA eq. to vanishingly low levels? I suppose the only way to validate this would be to construct an experiment and introduce larger and larger RIAA errors in several different systems.

Cheers,
Thom @ Galibier

Thanks Ralph (about balanced) ...

Yours is a different perspective from the limited commentary I've read on the subject. I'm not trying to be contentious, as I'm clawing to grasp some of the subtleties here. A while ago, I dug up and filed away a few links to threads on the topic of balanced. Some on the list may be interested in reading more, so I'll paste these in as hyperlinks.

Asylum Thread - on Balanced Phono
Asylum Thread - another Balanced Phono thread
Asylum Thread - on the noise disadvantage of running phono as a balanced device

Hi Raul,

I'm confident that everyone who is active on this thread has exactly the same sonic goals as you do - goals which you share with your partner Jose but which he stated in a way that I think more of us can relate to. I think we're working our way through a language issue here.

I think that your comments about searching for perfection is another way of your saying that every little bit helps (as Ralph agrees).

Now, in the case of a passive RIAA circuit, achieving low variance is more a matter of painstaking attention to parts matching, along with possibly cost, as you end up having to reject R-C pairs which don't result in a correct time constant (turnover frequency) within your specifications.

This is a time and expense sort of thing and not a design challenge. From a design perspective, it's a "freebie". Correct me if I'm wrong.

OTOH, design approaches have the potential to result in sonic penalties elsewhere in the design.

Perfection is a noble goal. It is however like saying you want to stop world hunger. Everyone interested in stopping world hunger raise your hands now. It becomes meaningless to the point of sounding like marketing.

Back to design traps, I think you'll agree with me when I state that the designer can easily paint himself into a corner by trying to track down microscopic distortion levels. Here's what I mean. The goal of lowest distortion might involve the implementation of feedback somewhere in the circuit (still using passive RIAA equalization, which seems almost universally accepted). One might be tempted to dial up the feedback until measured distortion is minimized. Overall "performance" might drop at this reading however. I'm reminded of a Tom Robins rant on the word "Performance" ... don't get me started.

Chasing down noise is another one of those demons which can get you into loads of trouble. In one of the above Asylum threads, Victor Khomenko (BAT) comments:

"Contrary to some beliefs, the signature of a good circuit is NOT no noise, but it is GOOD noise."

You can read his analysis in that thread, but I find this to be a provocative statement worth quoting, as it has great implications to how someone might approach a design.

I'm sure you will agree that the key is to think of the entire design holistically. The results in your Essential bear out that you embrace this philosophy. You don't get lucky with such a complex design. Please however let's not turn this into another thread on the Essential. We have an active thread for that.

BTW, you throw specs around. I'd be curious about how do you measure distortion - with sine wave input? with a wave form of an orchestra going full out?

Cheers,
Thom @ Galibier

Thanks ever so much for visiting this thread, Jonathan.

You make a very provocative point about varying sensitivity to different distortions (mechanical / speaker vs. electronic) which most definitely fits into the overall design process.

I have great respect for someone who has the courage to take on a complex design. It is a daunting task and one can easily get lost.

We need the electronic equivalent of the "Alpha Male" to push the collective envelope. Yes, some of them will suffer the electronic equivalent of the ill-fated Donner Party, but it's a choice they knowingly make, and we all benefit from the few successes along the way.

Cheers,
Thom @ Galibier

Thanks, Jonathan.

You've filled in the blanks better than I could hope to do, so I won't elaborate (too much). I completely agree with you that there is a difference between hearing flaws in a recording and being annoyed by them. A very good friend of mine runs a Lyra Olympos cartridge which exhibits exactly these positive attributes.

Perhaps a good mechanical analogy to what we're discussing lies in tonearms. A world-class tonearm can allow the turntable and cartridge to better do its thing by more effectively dissipating resonances. In being less resonant (perhaps a better term is "appropriately or artfully resonant"?), you can hear more of the music in the groves while at the same time ticks and pops fade to black much more quickly and become less annoying ... more of the good stuff and less of the bad.

The conversation about RIAA tolerances as well as the frequency width that you and Jose have been exploring is a provocative one. I think that all of us agree (at least no one has corrected me on this point yet), that getting low RIAA deviations is not so much a design constraint as it is one of labor and parts cost (component matching) to achieve the correct turnover points.

The meaningful threshold beyond which RIAA deviation becomes "specs-manship" and nothing more is what Jim and I were calling into question. I can't say for certain where this response threshold is, but your comments about the ear's higher sensitivity to electronic colorations over mechanical ones (speaker, room interaction, etc.) makes much sense, but still doesn't tell me whether .05% is just enough or if it is excessive overkill.

On a related issue, I'm curious about any psychoacoustic effects as far as the width of a frequency response deviation is concerned. I suspect we have several different thresholds along a continuum.

What I mean is that a very small width deviation might go unnoticed, whereas a slightly wider deviation (but still a spike) might be perceived as an anomaly, and a slightly wider band deviation may well be masked, or blended in. As Jose commented, these wider band colorations may well take some long term listening in order to be recognized and become potentially bothersome.

Some of the Lamm electronics exhibit such a wide band coloration which is quickly recognizable in completely different system contexts. While the gear is highly resolving, the colorations are very noticeable. I would never criticize someone for loving a Lamm, BTW, otherwise I'd have to own up to being an Audio Puritan (grin).

I have no insight into psychoacoustic experiments on this subject, and if someone does, I'd be interested in learning about it.

As Jim Hagerman (I think it was Jim) commented, we have to start from a technically correct baseline if we have any hope of coming up with a design that inspires us. I think everyone is in agreement that excellent technical performance is a necessary but NOT SUFFICIENT condition for greatness.

The good news (for me) is that when I get lost in a design change, my astute, musically trained wife drops by and immediately tells me whether we have music or merely hi-fi. Many of us are lucky to have perhaps the finest measuring tool known to man ... a smart, sensible wife who understands the goal of hi-fi.

And yes Raul, Jose's brilliant effort is one of those very fine, top-tier components which achieves greatness. I have difficulty simplifying my English to help you with many of the subtleties in my writing.

I try to be clear ...

It made perfect sense to compare component colorations to those of concert halls. The language has been in our hi-fi vocabulary for some 30 years. I look at the term modern in a component to be analogous to the frequency bias of many modern concert halls like Avery Fischer. As an aside, I've heard that Portland has a fairly new concert hall that resembles many of the fine halls of the past with more of a mid-bass and lower midrange center of gravity.

Unquestionably, the Essential is Avery Fischer Hall, and I've received quite a few private e-mails to back me up on this. Is Avery Fischer a bad hall? Absolutely not. Do some people prefer the hall in Rochester, NY (can't remember its name). Certainly.

Back to our regularly scheduled programming ...

Cheers,
Thom @ Galibier

Yes, the beauty of the marketplace is that we all have choices. I buck strongly when someone tells me that they have achieved sonic perfection and that I am a member of the great unwashed peasantry if I fail to appreciate their brilliance.

Sure, as manufacturers, we're all proud of our achievements. Why else would we make so many financial sacrifices to achieve what we consider to be lofty goals? Trust me - none of the small guys are buying 40 foot sailboats off of their audio income. This should be taken as a given (personal pride), and we as manufacturers shouldn't beat someone over the head with it.

I try to deal with my customers from a point of respect, realizing that there are many reasons for them to prefer another product - from personal taste as far as what a hi-fi system should do in order to bring them enjoyment, to the fact that they may want a piece of gear that is a bit more plug 'n play, to ... well, you get the idea.

So in that sense, I am more in the camp of Thom. I can live with a +/-0.75dB RIAA error if the compromise gets me better connected with the musicians and their message. For me, this emotional connection is more important than absolute accuracy.

I am reminded of an article that JC Morrison wrote in an old issue of Sound Practices. I wish I could find a copy of it on the web to link to. In that essay, Morrison broke the audio world into two camps:

1. The Audio Puritans - folks who would accept anything as long as they were told it was correct. Correctness frequently involved great suffering in the listening experience. No matter ... they are Puritans after all, and have a tradition to uphold.

2. Everyone else. Folks who actually want to have fun with their tunes.

Guess which side I gravitate to?

Cheers,
Thom @ Galibier

Hi Raul,

Yes, layout is indeed important. Jim Hagerman is too modest to admit it in this thread, but he put enormous amounts of work into the layout of the Trumpet. I recall him mentioning that he went through some 14 major architectural layout changes.

In order to get the best performance out of a single chassis design, he kept returning to the two tiered layout. Yes, grounding schemes and general layout plays a big part the final product - especially one so sensitive to noise as an RIAA stage.

We're all coming from a perspective that all of the components under consideration are at a very high level of resolution - that we're all after a design that will give us as much resolution we can get. I'm debating those special components which make it past the final cut if you will, and "fun factor" has to be a major acceptance criterion. Anyone who tells you otherwise is either lying or is a tortured soul.

Surely, once you've peeled away layers of distortion, input overload and such, you can never tolerate these flaws in a component. As I've written on several occasions, a poorly designed RIAA stage can overload and sound uncannily like tracing distortion. Nobody wants this, and yet there is a surprisingly large number of highly regarded components which exhibit this and other design flaws.

-----

This thread is taking a philosophical turn, but I think that's o.k. because it gives us insight into the whole person - be it the designer or the end user.

I always look at any design as a manifestation of the designer's personality. I've known a couple of audio designers who could be characterized as having obsessive compulsive disorder. Their designs were equally dysfunctional. In spite of their painstaking attention to detail, their end product was as flawed as their personalities were.

I am not arguing for being sloppy and careless in stating the above. Please do not misinterpret this as being the case.

I have to "out" you on this whole "false colorations" thing however. Your "it has to be perfect" mantra is really tiring me out. You are beginning to sound like one of those Audio Puritans.

Given a choice of a "correct" design (whatever the heck that is) which doesn't allow me to enjoy 30% of my record collection and a "flawed", colored design which allows me to pull out ANY record in my shelf without having to ask if its sonics are "worth" putting on my record player, I'll take the latter in a heartbeat.

I can't begin to count the number of components I've owned which took all of the fun out of hi-fi because they were "accurate".

I'm sorry, Raul but as good as the Essential is, it is as colored as many of the other fine RIAA/line stages I rank in the top tier (and the Essential is a fine piece). Please get over it and realize that no one can be all things to all people.

Is the very fine Essential is more harmonically correct than many of the other fabulous components out there? Absolutely not. The Essential to my ears is very much an Avery Fischer Hall type of component - a very lean and modern sound. Other fine units we've discussed in various threads cover other parts of the sonic spectrum - ranging towards pre-renovation Carnegie hall, for example.

Which one is correct? Both are correct and neither is. At some point, one is forced to choose, and here's where the "fun factor" helps to break the tie.

I'm coming down on you hard, because in your writing, you are portraying yourself as being 180 degrees apart from my sense of you and your goals after our fun day together two weeks ago. I think you are more in the "audio fun" camp than you prefer to admit in public. This may be a language thing, but consider yourself "outed". Please don't redouble your efforts to prove me wrong. Resistance is futile.

Close your eyes, take a deep breath, and let your hair down. You'll feel better. I promise you.

Regarding Audio Puritans, I will go so far as to discourage perspective Galibier customers from buying one of my turntables if they give me so much as a hint that they are Audio Puritans. Life is too short to work with someone whose sonic goals are that different from mine. I would prefer that they purchase a mainstream turntable and let me spend a weekend climbing a cliff or going skiing.

-----

About marketing ...

"The lady doth protest too much, methinks"

You don't need to tell me that your work came out of passion, because as I mentioned earlier, none of us are buying 40 foot sailboats from our income in audio. All of us are about passion, and I applaud yours and Jose's efforts, because you have achieved something very special. Is it better than everything else out there? Absolutely not. Let's not create a mythology here. This is what I object to.

We need to return to the subject of this thread. I'm getting tired of this.

Cheers,
Thom @ Galibier

Hi Raul,

If I may offer you a bit of friendly advice, and remember ... free advice is usually worth a bit less than what you pay for it.

I know that we have " to fight " not only against limitations in electronic parts, technology limitations but more important than that limitations in the way people think: this is our challenge, 90% or more of the Essential 3150 ( presentations ) were on tube lover audio systems, not an easy task I can tell you.

This is where you need to be patient. You will not convert everyone, and most of those whom you do convert you will not do so overnight.

I think I can state with confidence that each and every one of the designers participating in this thread have the same amount of pride in their product that you do, as well as the vision that they have a unique window into musical reality. I would expect no less.

Fighting the limitations in peoples' thought is one of those Zen paradoxes however. The more you try, the further behind you get.

It's important to take a historical perspective on this - to realize that many great concepts did not benefit the innovator ... until years after their death. Now, none of us are arguing that we like this, and many of us have achieved some degree of notoriety in our lifetimes (still waiting on that 40 foot sailboat), but one still needs to accept the possibility that success (no matter how you define it) may not be in the cards for you.

There are all sorts of reasons why consciousness moves slowly. Certainly, people are slow to move out of comfort zones. Have you ever heard the expression: "whom are you going to believe? Me, or your lying eyes?". You have to accept that people change at their own pace, and you can't force your reality on others. If you push, they will push back.

Oh yes ... the last thing I want to do is to be the "boss" of any thread. I am humbled by the great minds who are participating here.

Cheers,
Thom @ Galibier

Hi Ralph, Jonathan, Raul ...

Back to our regularly scheduled programming. This is good stuff ! I need to take it in small chunks.

This concept of immunity from input overload is a critical one. Listen up to these wise folks folks.

If you're scratching your head and wondering if you screwed up your tonearm setup (or if your cartridge is unable to track a passage), try borrowing another RIAA stage ... hopefully one which is known to be both immune to input overload as well has having a lightning fast slew rate (the ability to respond quickly to transients without overshoot).

If you're not practiced at listening for this, you may well assume you have a mechanical issue to resolve (cartridge or tonearm setup) when the real problem may be with your electronics. Don't beat yourselves up about your setup skills until you validate what you're really hearing.

Cheers,
Thom @ Galibier

Groovey one ...

Yes, it goes without saying that halls, mikes (and miking techniques) can drastically alter the recorded sound for good or for ill.

I was on the phone with a recording engineer today and we were waxing philosophically about how under-emphasized pro-sound (recording) practices are in hi-end audio.

I have greatest respect for the opinions of Ralph Karsten, Jim Hagerman, Victor Khomenko, and Nick Doshi.

Having said that, Jim's and Victors' and Nick's arguments about balanced operation in a home audio context make more sense to me than Ralph's do.

Nick Doshi, for example works in a broadcast environment but in this case advocates a deviation from pro-sound practice by advocating single-ended phono stage operation.

What's to be concluded from this? If it sounds good, it IS good.

Sorry for the relativism here, but at the end of the day the only meter that you should be concerned with is the one that measures the width of the smile on your face.

It's we designers who have to sweat the details and numbers.

Now, as far as balanced is concerned, you can only try it and report back. Theory is just that and nothing more.

I have no doubt Ralph's experiences with balanced hookups are real and I would advise someone with an Atmasphere RIAA stage to experiment with a balanced hook-up.

It's Ralph's explanation that makes no sense to me however, as well how relevant it is to other balanced gear like Ayre, Hagerman, and BAT.

Please bear in mind that a correlation (Balanced hook-up in product X sounds good) does not imply causality (all balanced hookups are good), but at least you'll know what works for you in your situation.

Cheers,
Thom @ Galibier.

Hi Ralph,

Likely this best discussed (or put to rest) in the already heavily beaten to death thread on balanced inputs to phono stages - where you and Jim H. had an "enthusiastic" discussion - ultimately agreeing to disagree.

Two of the main points I can't get around are:

(1) Whether or not a cartridge is a balanced or a floating single ended device. Floating single ended makes more sense to me.

and

(2) Since everyone more or less agrees that there isn't a 6dB noise advantage to running a cartridge in "balanced" mode, what might the advantages of running a cartridge in balanced mode be.

Given the fact minds far greater than mine disagree on the subject, I'm content to ultimately let my (and your) ears be the final arbiter of goodness.

Cheers,
Thom @ Galibier

Doctor C ...

Give us some system context on this "perfect RIAA forever".

Cheers,
Thom @ Galibier

Hi all,

Thanks Jonathan, Mothra, Piedpiper and others for adding some very needed focus and perspective on the variability of the recording process, microphone selection, monitoring, etc. etc.

Ultimately, as Piedpiper stated, we need a small cache of recordings we can rely on to help us map out reality and point the way to the musical truth.

I've found that this doesn't necessarily demand that we depend on unobtanium records from the golden age of recording, although this certainly helps.

You can approximate this by listening to your RECORD COLLECTION ... no small task, I realize.

Now, I don't mean listening to your entire collection, but at the same time, realize that you can effectively interpolate your results by intelligently sampling your collection ... filling in the dots if you will.

You should be reasonably familiar with a collection of recordings - ones which duplicate some well known sounds - be it a tenor, a soprano, a Strad, some drums, a Strat (played through a Tweed Deluxe amp for example).

Now, that SAME tenor will sound completely different depending on the engineer, the mastering, the mike selected, etc., but if you listen to a dozen different records, you'll be able to find a commonality amongst them that will serve you well.

Add to this, your soprano, some strings, horns, drums, and a gee-tahr, and this process of triangulating (a dirty word in politics these days, but helpful in navigating this maze) on reality will take you home.

Is this process perfect? Nope. Will it help you get the most out of your record collection? Yup.

One problem with this method is that any single change in your system will upset the apple cart. Still, if you follow this process with an open and attentive mind, it can serve you well.

Broadening your "reference" record collection will serve another purpose.

It will shift your focus away from having a "shootout mentality" and will force you to be patient. Likely (hopefully) your attention will return to what's really important - MUSICAL COMMUNICATION.

All too frequently, when we use a few "pet sounds" (sorry ... couldn't resist), we focus on those sounds instead of on music.

The digital crowd stops at this point (sound) wondering what all of this analog mania is all about. How sad for them, and indirectly for us, because being the majority, they dictate their sensibilities to the recording process - but that's a whole 'nuther topic.

Cheers,
Thom @ Galibier