or Connect
Head-Fi.org › Forums › Equipment Forums › Sound Science › What makes piano sound so hard to reproduce?
New Posts  All Forums:Forum Nav:

# What makes piano sound so hard to reproduce? - Page 12

Quote:
Originally Posted by spaark

Hold on, why does the other frequency have to be in the audible band? Why couldn't it be 45 kHz, for example? In that case, the difference would be 5 kHz, and the harmonics of this frequency would also be affected. I also don't understand what you mean when you say noise is greater at higher frequencies.

Sorry, I was assuming that we were talking about audio intermodulating with high frequency noise. If not, then I entirely misunderstood the model.  And if not, what frequencies are we talking about and why?  Where do they come from?

### Head-Fi's Best Sellers

Quote:
Originally Posted by jaddie

Sorry, I was assuming that we were talking about audio intermodulating with high frequency noise. If not, then I entirely misunderstood the model.  And if not, what frequencies are we talking about and why?  Where do they come from?

I'm talking about how high-frequency energy (beyond audible band) can actually distort the audible band. Let's say the audio signal has energy at 45 kHz and 50 kHz. Due to IMD, that energy can, as I understand, be transferred to other frequencies, at 5 kHz and multiples of 5 kHz.

Quote:
Originally Posted by jaddie

Sorry, I was assuming that we were talking about audio intermodulating with high frequency noise. If not, then I entirely misunderstood the model.  And if not, what frequencies are we talking about and why?  Where do they come from?

Even if recorded signal on a LP is strictly between 20Hz and 20 kHz, there IS always at least "garbage and noise" below and above the audio band. Below to 0,55 Hz at 33 1/3 RPM, above up in the megahertz region - a common scratch/tick/pop on vinyl record can go that high, it represents a pulse response to the stylus/cantilever/suspension/generator system/cartridge body/headshell/tonearm tube/tonearm bearings/...../....../ - before finally reaching preamplifier stage that will amplify it all - and cause all kind of troubles downstream the audio chain, with the end transducer (headphone, speaker) as perhaps the slowest/weakest link.

CBS Columbia Technology Center http://en.wikipedia.org/wiki/CBS_Laboratories was aware of this and issued a CBS CTC 310 Test Record - with both CORRECTLY recorded 1 kHz square

wave,

without ringing at approx 35 kHz as in its predecessor, incomparably more known  and oft used and cited CBS STR 112 Test Record , AND IM bands going from 1 kHz to 50 kHz.

One extreme case might be this: undamped high Q resonance in cantilever at 35440 Hz, effective stylus tip mass yieading against the elasticiy of the vinyl groove wall producing resonance at 35000 Hz - IM difference being the most sensitive frequency it is used for instrument tuning - 440 Hz - certain to be audible. In real life, this rarely happens, but is on the other hand entirely possible and plausible. In practice, these two resonances might even intentionally be placed at the same frquency, hopefully to cancel each other out - or be placed so far apart that causing audible trouble is the lest likely. All such cases are possible to measure using CTC 310.

I have decided to post this due to the fact that Google search did not yield a single hit regarding CBS CTC 310 - not even a sleeve picture. No wonder - it appeared in early 80s, just prior Sony purchased CBS and effectively shut down CTC afterwards. All the rage then was CBC CTC CD test disc, the first CD test disc originating from any country outside Japan - and vinyl was suppressed ASAP.

Will ask a friend to scan the whole sleeve and instructions, as this test record  is a significant  contribution to understanding and measuring reproduction of analog vinyl records.

Quote:
Originally Posted by spaark

I'm talking about how high-frequency energy (beyond audible band) can actually distort the audible band. Let's say the audio signal has energy at 45 kHz and 50 kHz. Due to IMD, that energy can, as I understand, be transferred to other frequencies, at 5 kHz and multiples of 5 kHz.

Yes, that is correct.

Now that you have 5 kHz, now we can generate more IMD sub-products:

45-5 = 40

50+5 = 60

etc.

Anyway, now you have introduced some very un-musical "noise".

Interesting link here WRT IMD:

https://passlabs.com/articles/audio-distortion-and-feedback

Quote:
Originally Posted by jaddie

Let's take a look at what intermodulation distortion is and see if this can be the mechanism.  IMD happens when one frequency modulates another causing a third result.  There are three basic types.  One, where low and high frequencies interact.  An example would be bass notes modulating mid-band or high frequencies because of non-linearity in the system.  In testing for this, an IMD analyzer typically uses the SMPTE standard of 60Hz and 7KHz mixed 4:1, and looks for whatever other frequencies are produced.  The audible nature of this kind of distortion is audible as the effect of low frequencies modulating higher ones, which is hard to describe, but easy to hear.  Find the loudest FM station on the dial and listen. Their on-air processing produces quite a bit of this kind of distortion.

The second type is difference IMD, where two closely spaced frequencies mix non-linearly to produce new frequency as products of the two.  For example, two tones of equal level, one at 10KHz, the other at 11KHz, may produce a 1KHz tone, which would be the low frequency second order product.  The non-linearity that produces this kind of result would also produce a high frequency product, and possibly other order products.  If one tone is of a significantly different level than the other, the IMD products change in level according to a non-linear relationship.  For example, if the two tone were 10dB apart, the products would likely be reduced by much more than 10dB.

The third type of IMD is transient-induced (TIM) or slew-induced intermodulation distortion (SID) or dynamic intermodulation distortion (DIM).  Three names for the same thing.  When a device is pushed to the speed limit of its ability to change output voltage levels, it's said to be slew-limited.  While it's slewing to a different output voltage, it looses its ability to output anything else but the changing voltage.  This means that during the transition all other audio will be essentially modulated downward until the transition is complete.  The typical test was to mix a mid-band square wave with a high frequency sine wave.  The exact frequencies depend on the bandwidth of the system being measured.  The output spectrum is then compared with the input (undistorted) signal, and the level of new frequency products is noted.  The technique was presented in the paper "A Method for Measuring Transient Intermodulation Distortion (TIM)" by Leinonen, Otala and Curl, J-AES April 1977.  Subsequently, it was found that amplifiers with proper negative feedback design reduced or eliminated the TIM effect, and that coupled with the poor correlation between measurement and audibility caused this test to fall into disuse.

The points to note here are: for IMD of any kind to take place, the system has to have some degree of non-linearity, and at least one frequency component has to push the system well into that non-linear region.

So, what IMD mechanism would cause high frequency noise to cause distortion products to land in the audible range?  We can eliminate the first type, LF modulation HF, because what's suggested is in fact the reverse.  The spectrum and amplitude of the noise is unspecified, but it's a fair assumption that the noise is fairly low level and above the audio band.  We can thus eliminate the second type, as the noise is non-specific, low level, and above the audio band.  If the noise forced a system into non-linear operation to produce intermod products, they would relate in frequency to the audio and the noise, placing the products out of the audio band as well.  To put it another way, for IMD to be audible the resulting products must be in the audio band, and the result of IMD of low level high frequency noise and audio frequencies would be either out of the audio band or extremely low level.

It's not looking to me like high frequency noise would cause audible problems because if intermodulation.

The first two look like the same distortion mechanisms to me, just different ways of measuring it:

IMD products would be (forgive me for stating the obvious and running off topic):

7 kHz - 60 Hz = 6940 Hz

and

7 kHz + 60 Hz = 7060 Hz

11 - 10 kHz = 1 kHz

and

11 + 10 kHz = 21 kHz

they both generate a form of "noise" in the audio bandwidth, i.e. non-harmonically related distortion.

TIM and SID are more complex than that, Otala didn't really say this much fedback or that much feedback.

Real short version is:

TIM has a lot to do with input stage overload and open loop gain and open loop linearity.

SID has to do with having adequate slew rate to reproduce audio signals, which are really horrendously slow.

Not much of a concern in modern SS equipment.

The above posts by Chris J highlight the fact that it is theoretically possible to produce IMD products in the audio band from stimulus outside the audio band.  I've never disagreed with that.  The problem is, presenting just the math of the situation is like saying IMD is an on/off condition.  Far, far from it.  IMD is caused by nonlinear mixing.  The degree of that nonlinearity directly affects the intensity of the IM products.  The intensity, frequency and duration of those products dramatically affects their audibility.  You can, for example, introduce two high frequencies equal and very high level into a system such that they are well into the nonlinear operation point, and yes you'll have IM products all over the place.  But what if the level of those frequencies is very low, where the system is much more linear?  The IM products are even lower in level.  What if the stimulating frequencies are not of equal level?  Again, the products are lower in level.

Back to the worst case: two high level, equal high frequencies.  The IM products produced will be lower in level than the stimulating frequencies, and that ratio determines the specific percent IMD.

Now, exactly how hot are these 45 and 50KHz frequencies we seem so worried about?  And how nonlinear is the system?  Are the stimuli stead-state or transient?  Are the stimuli single frequency tones or bands of noise? What's happening in the audio band at the same time that may or may not mask the IM products?

My point here is, just because you can dream up a worst case doesn't mean it actually happens all the time, nor does it mean it ever happens at all.  The biggest problem I have here is probably with the original posts that started all of this: "I read somewhere that the high-frequency noise can cause distortions in the audible frequency range, but this sounded more like a hypothesis than anything else. "

And the reply: "Because of intermodulation distortion, this actually does occur to some extent."

We then talked about the mechanism, but never the magnitude.  It's the magnitude that makes the difference.  There's all sorts of distortion in analog systems.  Does it matter?  Can you hear it?  The audibility of various kinds of distortion in various conditions has been the subject of several papers, but you can't just talk about the mechanism without including specific, and hopefully real world conditions.

Quote:
Originally Posted by analogsurviver

Unfortunately true. This is perhaps one of the reasons why phono cartridges with almost indistinguishable performance in the audio band can sound markedly different. Cartridges with high Q (undamped) resonance anywhere in their response, be it so ludicrously outside audio band, are more likely to cause trouble than the ones that show reasonably smooth response throughout their entire operating range of frequencies. In 1980 or so, Peter Moncrieff of International Audio Review published a comparative survey of then pretty much any cartridge then jostling for the place on audiophile's tonearm : frequency limit for the measurements were set by the equipment ( spectrum analyzer ), which at the time could not measure beyond 256 kHz ( in a word: two hundred  fifty six kilohertz ). No test record can go so high, a pulse response ( simply lowering the stylus to the glass support, creating vertical pulse ) was measured. The first in the series of Dynavector Karat MC carts, the 100R Ruby and 100D Diamond ( respective materials cantilevers of these carts were made of ) plus some others could easily exceed the measurement limit. 100D was the best behaved of the lot in this test. Today's Dynavector DV17D3 , the latest version of the venerable 100D which has by present standards become middle priced cartridge, should be a bit faster / better still. Mr.Moncrieff performed a lot of unconventional measurements, which he felt represent the actual behaviour of cartridges in real life better than conventional measurements of requency response and channel separation and provide reasonably good correlation between objective measurements and subjective evaluation by listening. Those seriously interested in analog vinyl might try inquiring if this survey is still available : http://www.iar-80.com/

I wasn't going to reply to this, but after reading it over a couple of times...well, there are some problems here.  I'll refrain from commenting on Moncrieff's writing or web site, it speaks for itself.

Now this needle-drop test he did in 1980.  The paper isn't available on his web site.  First problem - lowering the stylus onto a glass support creating a vertical pulse - you can force anything beyond normal limits.  There's no way that kind of test reflects any normal function of a cartridge on a record.   Next problem: all that output bandwidth.  Creating a step function at the stylus certainly would generate lots of high frequency spectrum.  But that's like strapping a rocket engine to your bicycle and saying "See! Bikes can to 200mph!" But more importantly, how was the output of the cartridge measured?  Was it connected to the spectrum analyzer directly, or through a preamp?  If through a preamp, preamps of that age were notorious  for being under-designed, and could be driven into distortion even with normal high level groove modulation.  Dropping the stylus on a glass support would certainly generate a far higher cartridge voltage, potentially creating distortion products in the preamp.  Now, the spectrum analyzer...there certainly were analyzers that went well beyond 250KHz in 1980, and even before that, he just didn't use one. Sorry, if you were really testing for ultrasonics would you choose equipment that limited the test results?  However, that's far less important than the supposition that dropping a cartridge on a piece of glass and measuring the ultrasonic spectrum it produces has any bearing whatsoever on the representation of the actual behavior of cartridges in real life.  You can take a hammer and whack the cone of your woofer, and measure the output voltage that produces and then try to correlate that with what happens when it's acting as a microphone picking up noises in your room.   You can do that, but it's not going to have any bearing on operation in the real world.

I agree, jaddie. In practice, the effects are probably negligible in most cases. There are several factors like you said. I suppose I should've mentioned that, but that's why I said "to some extent".

Quote:
Originally Posted by jaddie

The above posts by Chris J highlight the fact that it is theoretically possible to produce IMD products in the audio band from stimulus outside the audio band.  I've never disagreed with that.  The problem is, presenting just the math of the situation is like saying IMD is an on/off condition.  Far, far from it.  IMD is caused by nonlinear mixing.  The degree of that nonlinearity directly affects the intensity of the IM products.  The intensity, frequency and duration of those products dramatically affects their audibility.  You can, for example, introduce two high frequencies equal and very high level into a system such that they are well into the nonlinear operation point, and yes you'll have IM products all over the place.  But what if the level of those frequencies is very low, where the system is much more linear?  The IM products are even lower in level.  What if the stimulating frequencies are not of equal level?  Again, the products are lower in level.

Back to the worst case: two high level, equal high frequencies.  The IM products produced will be lower in level than the stimulating frequencies, and that ratio determines the specific percent IMD.

Now, exactly how hot are these 45 and 50KHz frequencies we seem so worried about?  And how nonlinear is the system?  Are the stimuli stead-state or transient?  Are the stimuli single frequency tones or bands of noise? What's happening in the audio band at the same time that may or may not mask the IM products?

My point here is, just because you can dream up a worst case doesn't mean it actually happens all the time, nor does it mean it ever happens at all.  The biggest problem I have here is probably with the original posts that started all of this: "I read somewhere that the high-frequency noise can cause distortions in the audible frequency range, but this sounded more like a hypothesis than anything else. "

And the reply: "Because of intermodulation distortion, this actually does occur to some extent."

We then talked about the mechanism, but never the magnitude.  It's the magnitude that makes the difference.  There's all sorts of distortion in analog systems.  Does it matter?  Can you hear it?  The audibility of various kinds of distortion in various conditions has been the subject of several papers, but you can't just talk about the mechanism without including specific, and hopefully real world conditions.

Thanks for the props!

Read the Pass Labs article for Nelson Pass's thinking on this.  He calls IMD the elephant on the dance floor.

Keep in mind adding an unharmonically related 5 kHz IM distortion product into the signal will subjectively be much worse than harmonically related 2 or 3rd order harmonic distortion.

Interesting observation:

Manufacturer's often do not publish IMD specs.............hmmm what are those manufacturers hiding?

Quote:
Originally Posted by jaddie

I wasn't going to reply to this, but after reading it over a couple of times...well, there are some problems here.  I'll refrain from commenting on Moncrieff's writing or web site, it speaks for itself.

Now this needle-drop test he did in 1980.  The paper isn't available on his web site.  First problem - lowering the stylus onto a glass support creating a vertical pulse - you can force anything beyond normal limits.  There's no way that kind of test reflects any normal function of a cartridge on a record.   Next problem: all that output bandwidth.  Creating a step function at the stylus certainly would generate lots of high frequency spectrum.  But that's like strapping a rocket engine to your bicycle and saying "See! Bikes can to 200mph!" But more importantly, how was the output of the cartridge measured?  Was it connected to the spectrum analyzer directly, or through a preamp?  If through a preamp, preamps of that age were notorious  for being under-designed, and could be driven into distortion even with normal high level groove modulation.  Dropping the stylus on a glass support would certainly generate a far higher cartridge voltage, potentially creating distortion products in the preamp.  Now, the spectrum analyzer...there certainly were analyzers that went well beyond 250KHz in 1980, and even before that, he just didn't use one. Sorry, if you were really testing for ultrasonics would you choose equipment that limited the test results?  However, that's far less important than the supposition that dropping a cartridge on a piece of glass and measuring the ultrasonic spectrum it produces has any bearing whatsoever on the representation of the actual behavior of cartridges in real life.  You can take a hammer and whack the cone of your woofer, and measure the output voltage that produces and then try to correlate that with what happens when it's acting as a microphone picking up noises in your room.   You can do that, but it's not going to have any bearing on operation in the real world.

Real world unfortunately does present similar if lesser in amplitude signals to the cartridge. Any imperfections in the vinyl, be it scratch, particle of dust, whatever, will present a pulse to the stylus, exciting any of its resonance(s). Want a proof of the audible differences that stem from these causes? Simply play any real LP with  a cartridge with peaky response anywhere in its operating range, not just within audible band, and one that is reasonably flat without high Q resonances. The flat one will sound appreciably quiter in the groove, not emphasizing any vinyl noise, ticks, pops etc that are real facts, not some imaginary hypothetical prospect. Yet the two might have very similar performance in the audible band. When an subjective audio reviewer will keep on telling you that any given cartridge is particularly quiet in the groove, he is effectively reporting that said cartridge has very well behaved response up to the limit of its operating range of frequencies.

I certainly agree that lowering the stylus to the glass surface is not the same thing as vinyl, for numerous of reasons. Lowering, not dropping to the glass surface, should not result in overload to a decent RIAA preamp - no more than it does with lowering the stylus to the rotating vinyl record whenever it is played. The difference is that hard glass surface will excite the stylus above the frequency of stylus mass yielding against the elasticity of vinyl groove, which acts like a low pass filter - but still occurs far above audio band in any decent cartridge.

It does not take to use the amplitude Moncrieff used in order for the test setup could distinguish from the noise with good signal to noise ratio that looks good on graphs. Phono cartridge is the most sensitive mechanical device you can possibly have in the home - it reveals information in the groove it is impossible to see even with scanning electron microscope - yet it will be displayed in the spectrum analizer plot. The dimensions and orders of magnitude this superficially crude mechanical setup is capable of accurately retrieving from the vinyl record go beyond what average or even not so average person can think of as being possible. Just calculate what is 100 micrometer ( about maximum amplitude for real records ) at - 80 or so dB. Electrical S/N can meet or exceed this 80 dB figure if proper parameters for cartridge impedance etc are chosen. Doubt it will ever be possible to directly measure mechanical deformations/resonances in cartridges this small - yet if you ever listened to a LP, you should at least know the general idea how it sounds. Vinyl records do not "noise" per se - what we perceive as vinyl noise is in fact sum of mechanical response of stylus/cantilever/etc from the stymulus of vinyl imperfections.

And the better/best preamps of that about 1980 age will still put most ( not all ) of the current available phono preamps to shame. Some performance in phono preamps from about that time is  still unsurpassed today. Whatever one might think of Moncrieff - fact remains, he was the only one capable of accurately measuring the deviation from ideal RIAA curve - down to the incredible mB ( in a word - mili Bel ) - any other test published simply printed a straight line and stating deviation was below measurement capabilities of measuring setup being used. He must have had state of the art equipment available to do that - I remember the name of the FFT Analyzer for the lowering-to-glass test - GenRad, Model # escapes my memory.

CBS CTC 310 test record allows for the most scientfic/repeatable etc way to check for the problems that might stem from above 20 kHz characteristics of cartridges. On the recording side, there are far fewer things that can go past linear 27 kHz using real time cutting - not so with catridges. One of the biggest offenders in MC artridges can be the method the wires leading from the coil attached to the cantilever leading to the outside world is executed - if not done right, these wires can and will exceed with their spurious output the output from the coil proper causing havoc with HF response. If not audible otherwise, it will be audible as vinyl noise being higher than necessary. Opening a few random MCs and examining them under microscope will  instantly give you an idea regarding this - and it is only one of the mechanisms that can affect a phono cartridge above audible band. Recent crop of MCs are far better in this regard than previous generations - do you think manufactures would go to all the additional trouble it brings for no apparent benefit ? In a way, it is a matter of survival in the marketplace - if your competition becomes generally known to produce cartridges that are quieter in the groove than yours, you might soon have to close the doors.

Like I said - I like to gather all info possible and try to get the whole process from musician performing live or in studio to the moment one wants to reproduce this recording at home as close to the original as possible. Finding any possible option(s) for improvement(s) anywhere in the chain. I know the cutting side can always say they can put the signals of amplitude on disc reproducing side can never play back - only to be reversed with frequency response with real time cutting/reproducing. Ping.pong ad nuseaum. I simply suggest sitting together and finding best possible solution for the end consumer, nothing else. "Translated" to piano sound, that means less or  ( next to ) none glassines added to the original piano sound - something listeners of analog piano recordings must have heard - at one time or another - along their journey regarding reproduced piano sound.

FYI, bicycles DO go over 200 km/h (not 200 mph - YET :http://en.wikipedia.org/wiki/Cycling_records - that list needs updating - BADLY ) -  not normal, regular ones in any sense, but either downhill on snow surface or behind air cutting vehicle on level flat salt lakes. It takes a lot of experience, courage or , if you wish, craziness to pull off such a feat; yet there are descents where it is possible to exceed 100 km/h with a regular road racing bike. The road where I  used to be regularly possible to do it no longer allows it - after they built a motorway some 50 km away, there is constant wind blowing in the opposite uphill direction, minimum about 10 km/h, making anything above 87 or so km/h impossible. No rockets needed. Well assembled and properly taken care of and maintained quality bike - needed for certain-  or else this is gambling with your own life.

Sorry, I really don't see the point of continuing this discussion if we're going to only speak in generalities with the reference data being an out of print paper being recalled from memory.  I also don't see the point of talking about extremes as if they had bearing on the usual and normal.  Referencing a world bike speed record has nothing to do with what the average bike rider experiences.  Without access to the "Marcrieff Papers", I simply cannot evaluate any hypothesis based on them.  I'm not a disciple of his.

Let me take the burden off others and go ahead and discredit myself: I cannot cozy up to anyone who spends time doing such things as subjective reviews of the audible effects of wire.  I haven't sipped that Cool-Aid, and I'm not likely to any time soon.

I don't care if a distortion product is harmonically related or not unless the magnitude is also specified.  As to manufacturers not publishing IMD specs, I agree they should, but manufacturers have historically published meaningless specifications.  For example, "Frequency Response: 10Hz - 20KHz"...means nothing without another figure "+/- XdB".  Welcome to marketing 101.  But having actually performed all three types of IMD testing on many different devices, I can also say that IMD never exists in isolation of other forms of distortion resulting from non-linearity.  If the THD+N figure is vanishingly low, IMD will be there too.  And, conversely, you won't measure high IMD without other measurements indicating nonlinear behavior as well.  It's simply never an on/off situation.

So, with respect to everyone, I'll have to bow out of the discussion unless some quantitative data is entered into the stated claims.  Just stating that IMD exists without any reference to quantitative data or audibility is just alarmist thinking.

There is a tendency in audiophilia to swing to extremes... Either flowery vague poetry about sound, or specialized technical esoterica. Neither approach really helps people who just want a good sounding stereo in their living room. I've never seen the point in counting angels dancing on the heads of pins. I'd rather talk about straightforward nuts and bolts.

The MOST important thing in discussing specs is knowing what all those numbers sound like. You can chase fractions of fractions from now until the cows come home and it won't make a lick of difference.

I have a pretty good idea of what makes stereos sound good. My theories are sitting in my living room for all to hear. They aren't just floating around in my head like an armchair expert.
Edited by bigshot - 1/6/13 at 11:37am
Quote:
Originally Posted by jaddie

I don't care if a distortion product is harmonically related or not unless the magnitude is also specified.  As to manufacturers not publishing IMD specs, I agree they should, but manufacturers have historically published meaningless specifications.  For example, "Frequency Response: 10Hz - 20KHz"...means nothing without another figure "+/- XdB".  Welcome to marketing 101.  But having actually performed all three types of IMD testing on many different devices, I can also say that IMD never exists in isolation of other forms of distortion resulting from non-linearity.  If the THD+N figure is vanishingly low, IMD will be there too.  And, conversely, you won't measure high IMD without other measurements indicating nonlinear behavior as well.  It's simply never an on/off situation.

So, with respect to everyone, I'll have to bow out of the discussion unless some quantitative data is entered into the stated claims.  Just stating that IMD exists without any reference to quantitative data or audibility is just alarmist thinking.

Please don't leave us, your contributions are really valuable, there is a lot of mumbo jumbo pollution creeping into this once more or less rational subforum. Touching on your IMD comments there is an interesting parallel with Bob Adams discussion of jitter (Audio Critic back issue no 21) where he concludes with saying that anything that has a poor jitter performance will show this in other more easily measured forms of distortion

Quote:
Originally Posted by nick_charles

Please don't leave us, your contributions are really valuable, there is a lot of mumbo jumbo pollution creeping into this once more or less rational subforum. Touching on your IMD comments there is an interesting parallel with Bob Adams discussion of jitter (Audio Critic back issue no 21) where he concludes with saying that anything that has a poor jitter performance will show this in other more easily measured forms of distortion

Not leaving, but I don't have anything else to contribute until we get back down to earth and start talking specifics.  And, I'll take at least part of the responsibility for hijacking the thread...badly.  I'll monitor, though.

Perhaps I'll see if bigshot will demo his stereo for me...(really like to hear it, if only we were near by!)

Quote:
Originally Posted by jaddie

Sorry, I really don't see the point of continuing this discussion if we're going to only speak in generalities with the reference data being an out of print paper being recalled from memory.  I also don't see the point of talking about extremes as if they had bearing on the usual and normal.  Referencing a world bike speed record has nothing to do with what the average bike rider experiences.  Without access to the "Marcrieff Papers", I simply cannot evaluate any hypothesis based on them.  I'm not a disciple of his.

Let me take the burden off others and go ahead and discredit myself: I cannot cozy up to anyone who spends time doing such things as subjective reviews of the audible effects of wire.  I haven't sipped that Cool-Aid, and I'm not likely to any time soon.

I don't care if a distortion product is harmonically related or not unless the magnitude is also specified.  As to manufacturers not publishing IMD specs, I agree they should, but manufacturers have historically published meaningless specifications.  For example, "Frequency Response: 10Hz - 20KHz"...means nothing without another figure "+/- XdB".  Welcome to marketing 101.  But having actually performed all three types of IMD testing on many different devices, I can also say that IMD never exists in isolation of other forms of distortion resulting from non-linearity.  If the THD+N figure is vanishingly low, IMD will be there too.  And, conversely, you won't measure high IMD without other measurements indicating nonlinear behavior as well.  It's simply never an on/off situation.

So, with respect to everyone, I'll have to bow out of the discussion unless some quantitative data is entered into the stated claims.  Just stating that IMD exists without any reference to quantitative data or audibility is just alarmist thinking.

OK - will dig out Moncrieff paper, have friend scan it, and send it to you. It is something between 20 to 40 pages IIRC. I only wish THD+N figures (or any other form of non linarities ) for phono cartridges were anything like those in electronics that can be ignored for most part, because they are so low. Even with best cartridges, it is in percent, not tenths or hundrieths of a percent distortion(s) - and although a phono cartridge that has 20 Hz to 70 kHz +/-3 dB or about that tolerance response does exist in real world, it is unfortunately more exception than the rule. I find it surprising for a mastering engineer to even consider equating non linearities of electronics and that of transducers - they are order(s) of magnitude apart.

I never wrote/said anything regarding audio cables. It just happened that the link for IAR had one "cable" in there. BTW, cable IS important in phono equipment, particularly regarding capacitance when MM cartridge is used. And replacing perfectly good "cheap stock" cable with correct electrical parameters for the job with audiophool cable with inapropriate electrical parameters is detrimental and not beneficial to the sound. Replacing with better quality cable WITH APPROPRIATE ELECTRICAL PARAMETERS is OK - and yes, can be heard. But only if and when measurable parameters are the same or very similar to the correct stock cable. One can have the cable out of nonexistium and if it has too high or too low capacitance when used with a MM cartridge, it will sound poor no matter what.

Here the first result for Google picture search " Shure Hyperelliptical vs Micro Ridge" - along Shure's, some other current carts and their frequency responses and THD measurements with pictures of the actual styli:

http://www.audiomisc.co.uk/HFN/LP4/NewLampsForOld.html

Please note that all cartridges featured are pretty high quality - all stylus tip profiles are an equivalent of Shibata or better; an elliptical or even conical stylus would have given much worse results still.

I have absolutely no intention discrediting you or your work you obviously did with great devotion , attention to detail and competence. On the contrary,  I would simply like you to broaden your understanding of what is going on once past your work with mastering a lacquer disc / stamper process. Here the link for most test records still likely to be "procured" one way or another, with sleeves with description of recorded signals ( including record used from above link):

http://www.gennlab.com/test_records.html

I can see the results on an oscilloscope and/or make DSD recordings of phono measurements ( DSD at 5,6 MHz will capture most of the results quite correctly, it is essentially flat to past 50 kHz and rolls off gently 6 dB/0ctave above that, much like analog, capable of recording past 100 kHz, has 120 dB dynamic range, about similar channel separation, THD+N & IMD like best electronics, unmeasurable wow & flutter ... in short, DSD recording should be good enough to allow phono measurements to almost the same degree of precision as if measuring the real thing ). If you own or have acces to measuring equipment, needed to do IMD measurements, I can record some samples and send them to you on SD card or similar - DSD at 5,6 MHZ is 11 min audio = 1GB. You would very quickly discover the distortion figures in cartridges are unfortunately not small enough to be neglected - specially if frequency response above 20 kHz is considered. The same goes for scanning loss at high audo frequencies at inner grooves - there is a reason why only CBS STR 100 test record has such test and results can be quite embarassing - try anything from +3 to - 10 dB at 20 kHz as compared with the same cartridge at the outer grooves where usually all the high frequency tests are done. Similarly, only the best cartridges with the best stylus tip geometry will produce anything like same shape of square wave recorded at outer AND inner grooves - CBS STR 112. I have yet to see published result or photo for both outer ( better, usually published ) and inner ( not seen published so far , usually far worse ) groove 1 kHz square wave.

Here the link to my "nearest equivalent" - except he has measuring equipment necessary to provide hard copy, graphs etc :

http://www.resfreq.com/phonomusings/phonomeasurements.html

% of distortion for phono cartridges are hard to quantify,as it is very dependant on the exact test record used, due to geometrical reasons like Vertical/Stylus Rake Angle, etc, etc - but low enough to ignore them are certainly not. Here one link with IMD measurements :

http://www.edsaunders.com/shurereview.htm

I do own Shure C/PEK 3 Cartridge analyzer - NOS, but it is still at my friend's in the USA and I will have trouble importing it to Europe due to lack of CE certificate - it predates EU as we know it now for all practical purposes, but for customer officer all that is needed to see is US power plug for the game to be over. Will have somehow to get it here ASAP - one way is to declare it as vintage museum piece - not far from the fact, given the world today is predominantely digital...

I hope it is evident to everybody from this post my intentions are good - in no way I intend to  discredit anybody who has honestly contributed to the cause of better recording or reproduction of the sound - and specially not you, who have kindly shared your knowledge regarding analog mastering and azimuth in particular. It is just the fact I can not pull the magnitude or % of distortion from thin air due to (temporary) lack of measuring equipment - but if it is visible on the scope with naked eye, it can not be low enough to ignore.

The links giving at least sneak preview to the magnitude of distortion in phono cartridges posted were obtained using Google search (pictures) and although I am familiar with some of them for some time now,  I am in no way affiliated with them. The IMD distortion from 1 to 50 kHz result is nowhere to be found - as indicated in one of my previous posts, CBS CTC 310 test record LP with which it is possible to measure it did not even "exist" on Google prior to my postings. Same on Yahoo, etc. There are posts regarding the CBS CTC 300 test record LP from the same series used for trackability testing in Steve Hoffman's http://forums.stevehoffman.tv/threads/ortofon-540-phono-cartridge-review.91096/ - other than that, the last series of test records issued by CBS CTC  "never happened". Yet the creators of this last series did know exactly what they were doing -  the time this series of test LP records appeared coincides with the greatest rage for CD - the plug was pulled for analog and the whole of  CBS CTC not long after.

New Posts  All Forums:Forum Nav:
Return Home
Back to Forum: Sound Science
• What makes piano sound so hard to reproduce?

### Head-Fi's Best Sellers

Head-Fi.org › Forums › Equipment Forums › Sound Science › What makes piano sound so hard to reproduce?