|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
216.23.55.198
In Reply to: Even if the software were perfect .... posted by Jon Risch on March 1, 2007 at 20:16:54:
It looks like this topic has been heating up since I was last here. Sorry I've been away and missed much of it, we had a critical office computer self-destruct and have had to scurry to get something up and running to replace it. Anyway..
----"To put it as simply as possible, unless the soundcard has enough resolving power and fidelity that you would be able to listen to music through it and use it for all playback chores, without being in the least dissatisfied with the playback quality of the soundcard, then this method would not be able to reliably detect any differences for a system that was at or above (less signal aberrations) the level of performance of the soundcard."
That is just plain nonsense. We do not need record and playback capability that reproduces things with perfect fidelity. We need only for it to respond *in some audible way* to the changes we are trying to detect. You neglect the fact that we are not trying to make something sound the same (as we wish in traditional listening tests in a stereo system) -- we are trying to see whether there is *anything audible* left in the processed difference.
It only matters that some audible trace of the difference is left in the record/play process. This is particularly true for this test --- but NOTICE THAT:
!! THIS IS ALSO TRUE FOR DETECTING CHANGES BY EAR. !!
Just look at where your reasoning leads ---Does an audiophile need perfect recordings in a perfect room in an otherwise perfect system in order to tell differences between cables, tweaks, etc? If so, I'd have to conclude that he must not really hear any of these differences, because no system or room is even remotely perfect. Do we agree with that conclusion?
Is every recording (that he uses to evaluate gear with) superior to every piece in his audio system? Even recordings that were made before the invention of the device/cable/tweak that is making the difference (wouldn't that tweak have to have been used in the recording studio, wired with better cable than is being tested?) One of my favorite recordings for ambience is Belafonte at Carnegie Hall, has nothing improved since that was recorded about 40 years ago?
For that matter, can an audiophile hear differences in a component when it is not the weakest link in his listening system because some less revealing component is in the chain somewhere? Can he detect a change of cables using loudspeakers less transparent than than either cable? (If so, I WANT those loudspeakers!!).
"In order to make this clearer, I will provide an over-the-top-example:
lets say the soundcard used had an ACTUAL linearity such that it literally failed to respond to any data below the 15th bit, or put into a different framework, that pretty much everything below -90 dB full scale was obliterated."Yes, that is over the top (one might even suggest, a "straw man"). But, ok, lets say we have that hypothetical soundcard. When recording, the signal being recorded is not below -90dB, it is dynamic varying with time over a wide range -- it contains large waveforms. Say there are differences between the cases being tested that, if isolated, are below -90 relative to fullscale, and are audible. So do these differences have *no effect* on the recordings made? Of course not. They will change values determined by the converter at many sample points. At some point, perhaps one sample value that had been on the edge of being +12,345 instead of +12,346 will get tipped over to +12,236 due to the tiny "difference" being made. An effect will show up in the subtraction result. And (back 'under the top), the actual dither or noise with real hardware actually allows signals to be resolvable below the least significant bit, just as sounds can be discerned below noise from analog sources.
'Under the top', still, in a real test with real gear, the difference track won't be 90dB reduced. In a very good null (made from a particularly quiet audio system), when turned up higher than usual, it will leave a soft, roughly white noise floor. But people get to listen to the result themselves, they don't need me (or you) to tell them what they can or can't hear or what they think may be significant.
And why the concerns about -90dB detections (reduced from the -120dB argument in an earlier post)? (That rules out use of vinyl, you know. An exceptional vinyl recording and setup can maybe get to 73dB S/N.) Are all possible differences between cables, etc., less than 90dB down? Because otherwise the example above could, at best, apply only to such circumstances.
****Should anyone who can solder want to really hear what 90dB below their listening level sounds like, you can make an attenuator to get that that with a series 330K resistor in the "hot" path from input to output, shunted to ground by a 10 ohm resistor on the output end. Just insert that attenuator into a line-level signal path and compare playing music with and without, at equivalent volume levels. No need to just read discussions about how low that is, hear it in actuality***
I have made a DiffMaker run using a decent (but not SOTA) soundcard. It was two recordings, made through two different cables, neither of premium status), and the resulting difference was of course not 90dB down, I'm sure. But when I listen to the difference at the same gain as used to listen to either one, nothing is heard in that case. Is the soundcard cutting the info out? To test, I took one of the recordings and dubbed my voice over it at a very low level, so that I couldn't hear it in the mix. When I subtracted the two recordings, I still heard nothing at appropriate gain. Until I turned the gain up, and then, there was my voice (but still no trace of the original recorded track). So, the differences must have been weaker than my overdubbed voice, at least to my ear in my room.
"If it is not, then it could be masking other components or systems aberrations and differences to whatever extent that the soundcard lacks ultimate sonic perfection."
Picked that sentence out because it illustrates a fundamental misconception. "Masking" is relevant to conventional listening tests. People may not be able to notice some things in the presence of other things. In DiffMaker tests, though, such masking "other things" are removed in the subtraction. Masking isn't a factor -- at all, unless possibly the noise is so high as to be noticeable. The test is between hearing something or hearing essentially nothing.
"...no sound card out there can pass this test, they all add something that is not 'musical' to the sound, when compared to the original 'source'..."
And there's another one. If I "add something" to A, and add the same something to B, then subtract A from B, that "something" vanishes, per the simplest algebra. Unless the added something can also make the difference we want to isolate VANISH (not change, mind you, but VANISH without a trace), the difference will be left.
"If you record one set of components, and then make a change, and record another set of components, the actual timing of the sampling will not be synchronized."
Everyone, please read the writeup about the test we are talking about before offering such objections. The synchronization is a main function that the software DOES, as is clearly described on the Audio DiffMaker page.
Besides, think it through: if the synchronization were incorrect, then the result wouldn't be silent, right? Synch (delay) errors, and gain errors leave a (false) difference. Silence only happens when all that stuff is right."Now, we would have the sample to sample timing error CHANGING, and the jitter differences between the ADC and the DAC would only add to the difference error, not cancel."
Ok, but all these things you object to will leave a difference error, not a silent null. And any lister listens to that error. If he hears nothing he considers significant, then that problem must not have been a factor, right?
If he hears something significant, then possibly the test has an error; or else a true difference has been found. It's easy to check for the error, by the way -- record the exact same situation twice, and subtract. If you hear something significant in that difference, then the setup or gear is inadequate for what you are testing."Note that the kinds of faults and problems I am mentioning almost all result in a false null, that is, little or no difference file content."
How so??? I see only ONE conceivable scenario in the post, which is that rather absurd case of the A/D converter that somehow knows what part of a signal is the -90dB part so it can ignore it in its conversions. EVERY OTHER problem you mention would result in a false DIFFERENCE. A NULL is very difficult to achieve, everything must go right for that to happen.
How could the other claimed faults result in a "false null"?
How would the faults mentioned above result in the two recordings being the same to within an inaudible degree, even if what they recorded WAS different to an audible degree?"Of course, far too many objectivists will take such a result as "proof" that there are no sonic differences, but the actual truth would be, we just wouldn't know for sure."
I really try to not offend, but: have you tried making a DiffMaker test? The statement above sounds EXTREMELY defensive to me. You are in effect implying that you fully expect the test to result in no differences you can hear, and are getting explanations all ready to defend against that. Why?
Have you tried it already and found no difference in things? Because I haven't tested many of the controversial things, myself. I don't have access to premimum cables or tweaks, don't have extra $$ laying around to buy any. (I'd also would prefer to just provide the tools and let others test themselves and make up their own minds). So, how do you KNOW that the tests will show "no difference" to most listeners?
There is only one set of DiffMaker recordings (that I know of) which has been posted anywhere, and it, in effect, reveals a difference (to my ears, at least) between different types of coupling capacitors (though the scenario I used is pretty much stacked in favor of that result).
May I also remind other inmates that the results of the DiffMaker test are not purely objective (though they could be modified to be, but then someone would have to decide what each number meant...). The result of DiffMaker is a WAV file that anyone interested listens to -- no graphs, no numbers, no one can say what someone else hears in that file.
Another correction, contrary to reports elsewhere, DiffMaker is not a commercial product, it is freeware. And I claim it to be only a tool, not a religion!
Follow Ups:
Bill wrote:
"We need only for it to respond *in some audible way* to the changes we are trying to detect.
AND
It only matters that some audible trace of the difference is left in the record/play process."I tried to explain this in my original post, but I didn't think it would have to be spelled out word for word.
OK, let's look at that example I made, of the soundcard with the last bit missing, everything below -90 dB is GONE.
Now, we record the output of DUT A, and of DUT B using that same soundcard. We play back the DIFF file, and low and behold, we hear nothing but a slight amount of noise. This means that there is no difference between the DUT A and DUT B, right?
Nope, in fact, they could sound quite different, and we would never even know it if all we used was the DIFFmaker test.
As an extreme example, let's say DUT A _ALSO_ lost the last bit of data, everything below -90 dB was GONE. Unit B does not do this, and is quite accurate all the way down to the 16th bit. Obviously, if we listened to DUT A vs. DUT B, we would hear this complete loss of low level information, the truncation of the audio signals. But the soundcard would be inherently unable to record and thus, show a difference, because IT TOO was limited to this level of resolution.
So how good the soundcard is, makes a difference in how well we can detect a difference between two other DUT’s being recorded.
I know that some have been proposing that we examine the difference between the input of DUT A and the output of DUT A, thus eliminating any loss of resolving power between the two units A and B, but the same problem rears it's ugly head, if we tested DUT A using that hypothetical soundcard, and adjusted the gain for the two different output levels, and listened to the resulting difference file, we would again hear virtually nothing but noise, thus we would erroneously conclude that DUT a was "perfect", that is, it had no difference between it's input and output. DUT could also be tested and also found to be "perfect", but in reality, the two would have quite different performance capabilities, yet the DIFFmaker test would not reveal ANY differences between the two DUT's.
This is why I am saying that whatever the limitations are of the soundcard used to make the recordings, this is the limit of the resolving power of the DIFFmaker test.
Now, I know that we are not dealing with soundcards limited to 15 bits of resolution, that was a gross example to make my initial point that the soundcard can and WILL adversely affect the results if it is less than perfect itself.
I once again point to widely accepted opinion and professional practice, that NO soundcard is considered sonically perfect or even sonically SOTA, much less a $150-200 soundcard, therefore, the DIFFmaker test is going to be limited in how much resolving power is availble to record differences.
If we postulate that no soundcard even approaches the SOTA sonic quality as exemplified by such components as Mark Levinson, Krell, Audio Research, etc, then we must also acknowledge that the DIFFmaker test will NOT be able to provide a valid difference file for one of these components, nor will it be able to provide all of the subtle differences present and of interest in audio cable, and many other audio components.
None of this addresses the obvious issue of the gain adjustment when trying to do a DIFF maker test for an input/output test of the same unit, obviously, if the soundcard used does not adjust ONLY gain, but introduces any other signal aberrations when a gain adjustment is made, then this distortion of the soundcard will be added to the resulting difference file.
In the hypothetical case we have been examining, instead of hearing just noise, we might hear some very small residual of distorted music, just enough to look as if the DUT was slightly less than perfect, BUT, with a difference that was very low. This might induce us to believe t6hat the DIFFmaker test had actually been wholly unsuccessful and had provided valid and worthwhile data, even though the entirety of the 16th bit and below was missing.Bill wrote:
"Just look at where your reasoning leads --- Z"Indeed, it leads to the inescapable conclusion that the DIFFmaker test is not magic or supernatural, but is limited in the same manner as all other audio systems.
Can we hear 'past' a weak link component? Yes, to a certain degree, but I think that it is wise to point out, that with a RECORDING process taking place, anything lost is gone forever, that once it has been 'left behind', it is never coming back, while with real world audio systems, we can can kinda sorta listen past a weak link component, and when that weak link component is upgraded, we can now hear what the other components are doing that we were not fully able to do belfore. However, if the soundcard is upgraded or replced, all the old recordings are still limited in what they can show or reveal, no new difference information will become known until we make that recording comparison again, using the new soundcard.
Bill wrote:
" Say there are differences between the cases being tested that, if isolated, are below -90 relative to fullscale, and are audible. So do these differences have *no effect* on the recordings made? Of course not. They will change values determined by the converter at many sample points. At some point, perhaps one sample value that had been on the edge of being +12,345 instead of +12,346 will get tipped over to +12,236 due to the tiny "difference" being made. An effect will show up in the subtraction result."Of course, some effect of a difference down at -93 dBFS will still alter bits higher up the ladder, BUT, these changes are greatly reduced, and would tend to fall, by definition, into that realm where they would register as if they were less than a true difference at -93 dB.
In other words, loss of the entire 16th bit will render the difference virtually deaf to signal aberrations below -90 dBFS.Bill wrote:
"And (back 'under the top), the actual dither or noise with real hardware actually allows signals to be resolvable below the least significant bit, just as sounds can be discerned below noise from analog sources."Sure, but now we are limited to 15 bits and a little more with dither, instead of 16 bits and a little bit ore with dither. BTW, dither designed to augment and extend a 16 bit system WILL NOT adequate exercise the 15th bit to "fix" it, the dither level would have to be increased correspondingly, and so, you argument falls apart in a very real sense.
Bill wrote:
"But people get to listen to the result themselves, they don't need me (or you) to tell them what they can or can't hear or what they think may be significant."Of course, but what they may not know, probably DON'T know, is that the difference file is still limited in a very real sense and manner, to the sonic quality of the soundcard.
It still boils down to whether or not you think that the specs are enough, or the sound as it listens is what matters, if the soundcard is not SOTA sonically perfect, then you can not test SOTA audio components with it and expect to see/hear all that might be there.
Bill wrote:
"And why the concerns about -90dB detections (reduced from the -120dB argument in an earlier post)? "I think that you must be referring to someone else's post here. My argument is that the known and obvious changes in the sound when different dither algorithms are used, is a form of proof that signal aberrations that are at or below -90 dBFS DO impact the sound. Each dither algorithm does sound different, and not just at low levels, but all across the range of the music signal. Music doesn't 'live' at 0 dBFS or even -10 dBFS, but rather, across a wide gamut of mini ranges as it swells and flows.
Bill wrote:
"(That rules out use of vinyl, you know. An exceptional vinyl recording and setup can maybe get to 73dB S/N.)"No, no, and no.
First, I have worked extensively with measuring and characterizing the vinyl playback system, I worked for Discwasher for years, and have actually done the research (in some cases, original and unique work). A well pressed virgin vinyl recording can reach a dynamic range in excess of 80 dB, from approx. a clean noise floor of -70 dB (where signals can be heard below for approx. 6-10 dB) in the midrange, up to about +10 to +12 dB above 0 dB (where 0 dB is defined as 3.54 cm/sec modulation velocity per channel, or 5 cm/sec lateral, the Shure tracking test record had levels of 25 and 30 cm/sec, which approximate to +18 dB).
If the record was from a direct to disc session, it could even exceed that range, up to about 85 to 88 dB. Some of the Sheffield direct to disc records regularly hit +18 dB, and had a noise floor of -70 dB or better, all in the midrange.
Second, it does not necessarily rule out the use of vinyl as a test medium, because as we are all aware, you can hear things on a vinyl record that allow you to discern differences between SOTA power amps, preamps, and audio cables. It is not JUST the dynamic range, or the low level resolution, but rather, the entire package of performance during the playback event. Where vinyl excels, is that when it is reproducing the music where it 'lives' at moderate to low levels, it is fairly linear and resolving, without some of the issues that plague digital audio.
Bill wrote:
"Should anyone who can solder want to really hear what 90dB below their listening level sounds like .... "Wrong. I did not say -90 dB below the music level. Music level implies the average level, not the peaks, and not 0 dBFS. Relative to the average music level, a -90 dBFS signal is really at somewhere around -50 to 55 dB from the average level of the music, which is quite a lot off from the -90 dB straw man you set up.
Bill wrote:
"In DiffMaker tests, though, such masking "other things" are removed in the subtraction. Masking isn't a factor -- at all, unless possibly the noise is so high as to be noticeable. The test is between hearing something or hearing essentially nothing."Perhaps masking was a poor choice of words on my part, I was not referring to the classic 'masking' as applied to psychoacoustics and MP3's, but rather more in the nature of the poorly performing soundcard occluding or losing the low level signal information to one extent or another, and thus, 'masking' what was going on with regard to the difference file.
Again, in my hypothetical example, the soundcard will literally not record anything below -90 dB, it is MISSING, gone, poof! ALL differences below that level will be utterly and completely 'masked', or as a better choice of words, obscured. The difference file will contain virtually nothing but noise, or perhaps just enough residual of the original signal that it seems as though the test worked, but in reality, YOU COULD NEVER KNOW FOR SURE whether or not the soundcard was adequate to the task, or not, unless you had conducted a huge series of well done and executed controlled listening tests that incorporated numerous controls and verifications, as well as open testing, to try and verify or confirm that the soundcard was sonically transparent.
Without that kind of extremely costly and time consuming verification, you could not automatically assume that the soundcard was god enough, or did not have it's own flaws literally covering up those of the DUT's.Bill wrote:
"
And there's another one. If I "add something" to A, and add the same something to B, then subtract A from B, that "something" vanishes, per the simplest algebra. Unless the added something can also make the difference we want to isolate VANISH (not change, mind you, but VANISH without a trace), the difference will be left."Sorry, but the "added something" may not be a simple addition, but rather, an alteration of the signal, a signal aberration would again be a better term. In the case of the hypothetical soundcard which loses the 16th bit, nothing at all is added to the signal, just removed. However, in my extreme example, (in order to clearly make my point), the loss of signal content is a well defined and obvious thing, let's call it truncation to 15 bits. But what about a soundcard that is not so obviously distorting the signal? What if it is 15 1/2 bit s accurate (a not uncommon situation with a 16 bit based card), and that it has a frequency and level dependent loss of signal, these things WILL NOT necessarily cancel out with a set of recordings of DUT A and DUT B, or with an input/output comparison, but will still tend to obscure any low level differences that the DUT's do have that are below the level of soundcard corruption.
Bill wrote:
"Have you tried it already and found no difference in things? "Actually, I have tried to use a soundcard in a similar manner, by nulling the test signal with the output of a DUT, as well as to compare two DUT's with each other via a difference file.
I manually edited the timing to match the nearest sample, using both 'ways' {creating two diff files}when it was not coincedent with each other (almost all the time).
At first, I too thought that the soundcard should not matter, but found that using a SB16 just would not do the job; even after I tried a studio grade soundcard, the Echo/Event Darla, there was still too much contamination from the soundcard to do the job properly.
I sincerely hope that you do finally understand that the soundcard quality DOES matter, and that it can not be ignored, that it DOES NOT fully cancel out, only a portion of the soundcard distortions are cancelled out. Losses of the signal, non-linear or otherwise, simply do not cancel out. Non time coincident additions do not cancel out (computer EMI/RFI contaminants) either, and their resultant effect on the audio circuitry can be erratic or misleading with regard to how the DUT's AND the soundcard respond to them.
If you can not see this is the case, then I am sorry for you, it is not THAT hard to figure out, especially when it has been spelled out for you.
Jon Risch
It is pretty obvious from your reply, that you really didn't get most of the points I was trying to make, unless it is deliberate obsfucation on your part.I will reply point by point when time permits, as it is late here now.
Jon Risch
Bwaslo, you should have more experience before commenting.
First, sound cards DO effect the sound. IF they didn't, we would pronounce them perfect and not go any further. It would mean that if I added the AD-DA converter that I purchased for less than $200 on sale, from Leo's Music, a pro audio outlet, in series with my preamp, that it would be perfectly inaudible. We have found that this sort of thing IS audible.
Next, you act as if we think that all the subtle changes that we make in audio can be made obvious to everyone, everywhere, with cost effective audio equipment. No it can't, BUT we can detect differences in our own equipment that we have become used to, with audio sources that we know are recorded to a very high standard of quality.Technically, this difference test, while potentially being an important tool, will suffer from the limitations of the sound card. Perhaps, not as much as normally, because some of the more obvious harmonic and IM distortion will be cancelled out, that was generated by the op amps before and after the A-D processor, but the noise will still be added and will be at least 3 dB higher than the wire as it would normally be evaluated. This is because the subtraction of the coherent parts of the signal will not cancel the random effects of the noise from both samples taken. In effect, I would think that I would see mostly noise in the error signal. Another big problem is phase differences at both low and high frequencies. This might not be too much of a problem with a wire, but a component (such as a cap) would seriously suffer differences with another component.
It would seem that only the simplest things, like a wire, could be tested easily with this device without linear distortion adding and confusing the error output.
Another really important factor is the OMISSION of information caused the the sound card. If the sound card was so good in the first place, would it omit infomation? Listening tells me that it does do just that. If you omit information, how do you get it back, to see how it has been changed?
It should be noted that we have been trying to do this sort of thing for decades. We did it with 10 bits and a 50KHz clock rate, 1/3 of a century ago. We could not see any significant differences in good (Levinson JC-2 level) op amps, and I have always wanted even more resolution like 16 or even 24bits, but like Jon Risch, I doubt that it will be very effective. However, it would be a Godsend for us, IF it really worked.
John,I am quite well informed on this, I think, thanks. I bet I've tested and listened to many more sound cards than anyone reading this forum. I fight problems in (some of) them way too often, if fact. I read the disagreements, here, but they don't seem to hold water to me. Let me try to spell out why.
"First, sound cards DO effect the sound. IF they didn't, we would pronounce them perfect and not go any further"
Guys, just where did I say that sound cards don't affect the sound? I admit I'm not the best writer, but I know didn't write that, not ever. In fact I test soundcards regularly and have a pretty good idea what most can and can't do. OK, soundcards can and do affect the sound. There!
But, SO WHAT? So does every loudspeaker ever made. Yet, somehow, amazing as it is, people hear differences in equipment while listening through loudspeakers! Same with some amplifiers that can affect the sound. And the sound being changed was recorded with mics that affected the sound, too. Clearly these don't automatically invalidate testing, then, so why do they invalidate a differencing test?
Yet again: the soundcard CAN vary the sound, and still be quite usable, just as can the the incredibly imperfect loudspeakers we all use in domestic listening rooms. For the differencing test, the soundcard only has to respond to the difference, not reproduce it verbatim.
Yours and Jon's argument seems to be (correct me if my understanding is wrong) that:
(1) one can never detect a difference, even with mathematical processing, between two signals recorded with soundcards. Because the soundcards can affect the sound. Yet,
(2) a listener *can* regularly detect such differences, by ear. Even if a number of the components he listens through affect the sound, too. Even if they affect the sound more obviously than a soundcard does. Even if that listener is a constantly changing organism every minute, with changing moods, states of hunger, thirst, tiredness, age, etc. , never exactly the same for any two tests.
Can't you see why I disagree? Can the argument of "not enough resolution" be valid in the one case, yet not in the other?I can rather easily make two recordings with an intentional inaudible difference, (inaudible to me, of course, who knows about others!) which difference can be exposed and made audible by the software. But I'd like to find situations where there are recordings that have differences that ARE said to be audible and can ALSO be exposed -- revealing the reported difference for more listeners -- by differencing. Now, surely, having that difference be audible can't automatically make it even MORE difficult to detect via differencing, could it??? The program can make something that is inaudible audible, it is not hard to do that. It doesn't seem to me to be much of a stretch to think that it could also render something that is audible to some (but not to others) more audible, and reduce disagreement about the situation.
Note that I have also not said that every difference can be resolved by this using any equipment. I sense people think I'm trying to prove nothing is audible unless it can be measured. I'm not. I don't care much about proving something that can't be proven like that. I'm only trying to make audible differences, if they exist, more obviously audible by removing the other program material that can mask them. If someone as a result finds that a differencing convinces them there is no relevant difference between two things they test, based on what they hear (or don't hear) themselves with their own ears, that will be their conclusion (which they are free to make).
Obviously, if a differencing test were made through a telephone, the difference test would leave (at least) obvious noise, and no conclusion could be made. But it wouldn't leave silence. There might be signal behind that noise, but not apparent SILENCE altogether. So that just becomes a simple case of inadequate equipment or setup for the test. That can easily be checked with a dummy test. We can all agree that the test can't be useful in all cases. But I think it can be useful in very many cases. Particularly useful, I think, is the characteristic that lets people evaluate audibility more readily using their ears to hear the difference recording, without feeling challenged or being unduly influenced by expectations in either direction. But I see that some here do not agree. Que Sera.
"the noise will still be added and will be at least 3 dB higher"
Yes, noise will increase 3dB, but if the 3dB increased noise is not heard, I submit that it doesn't matter. The noise is always there, and it is always 3dB higher than some other level. (Anyway, even if it did matter, it wouldn't be hard to eliminate the 3dB increase: the tests could be recorded twice for each case and each recording set for a case averaged coherently. That would reduce noise by 3dB again, back to even. We could even average more than two for each situation and *reduce* the overall noise if we wanted to!).
"Another really important factor is the OMISSION of information caused the the sound card. If the sound card was so good in the first place, would it omit infomation? Listening tells me that it does do just that."
John, you are aware that such omission is in terms of your mind's rediuced ability to extract meaning and impressions from a signal that now less resembles the original signal you can relate to. We don't hear with spectrum analyzers. We can't hear degraded hash as well as we can notice degraded music. Information is a matter of context. But subtraction is not a matter of context or affected by masking (it has that very significant advantage over traditional listening tests for difference). The missing information is not from some buried signal completely vanishing from the larger signal with no effect, how could it *possibly* do that??
"Another big problem is phase differences at both low and high frequencies. This might not be too much of a problem with a wire, but a component (such as a cap) would seriously suffer differences with another component."
That is true, agreed (and experienced). Very small mismatches leave rather large difference results, the amount is easy to calculate, infact (see the Help file of the program). But that only means that care must be taken to handle these KNOWN effects. I've done such a test with capacitors, which of course make a first order filter (hp or lp, depending) in combination with the source and load resistances. It is a simple matter of adding resistances (or extra capacitance) to match the rolloff characteristics; at line levels, a 10 turn trim pot in shunt somewhere has done well. There is even a simple response analyzer system built into DiffMaker to help in adjusting these. What if the caps aren't matched well enough? Then, it will show up as an existing difference (which it is). Might not some characteristic of that trimpot cause an audible difference? Maybe, but if so, that will show as a difference, too. But, clearly (to me, anyway) neither has ANY way to show a false null. Mistakes make differences, not nulls. Presumably interesting difference results will result in further exploration to detect whether we are getting fooled by such things.
--
There are some comments in this forum that the DiffMaker test suffers from the same objections as do ABX tests: that the hardware used for the test can invalidate the result due to lack of transparency. I submit that is just not so. Hardware might degrade signals, so that they become less easy for a human to relate to, because they sound less like something humans care about (music). That might be a factor with human listeners involved with music during ABX. But after differencing, the listening is only for absense or presence of remaining signal, the main common signal is gone. The hardware simply doesn't know what part of the signal it sees is the original music and what is from the change made by the cable/tweak/etc. It's all just "signal" to it, one thing. I'm surprised that that, at least doesn't seem obvious to others here. Hardware must treat all parts of the signal about the same, it has no way to selectively strip one part off, based on how it got there, so that effects of the change completely go away.
....and sheesh, I'm tired of writing about this. Good night, all.
I think that what Bwalso is saying that regardless of the measuring ABILITY of the soundcard, it is still a valid CONTROL, since we're using the same soundcard for both tests.If you have mediocre speakers but can hear an improvement when you change sources - agreed - it must be QUITE a significant improvement. The trick was that you used the same speakers to audition BOTH sources.
It's a no brainer that one would want to have the lowest noise floor possible to see even the most minute differences.
There is also a difference bewteen the dynamic range and SNR ratings of a codec ON a soundcard and what the actual SNR will be when a loop-back noise-floor test is done.
I also imagine one can't do bit-for-bit comparisons unless the stimulus is synced with the recording equipment for each iteration. Is this correct?
Once again, I applaud your effort in making this program. It is possible that it will supplant linear subtraction in many cases, such as the setup that Walt Jung and I used to measure caps more than 20 years ago. However, I think that comparing wires will be difficult with this test, but that does not show that wires have no sonic difference between them. That has not been my experience, and since I don't make wires, I would prefer that all 'well engineered' wires sounded the same. It would save me a good deal of time and effort, in making my products.
The bottom line is the resolvability of the AD-DA converters and their attendant circuitry, which seems to remove too much low level information, and this would not necessarily be made audible by cancelling out the main signal. It would seem to me that you are looking at differences residing at the 1 or 2 bit level in many cases.
Perhaps I am wrong. What differences have you found between common components with this test?
This post is made possible by the generous support of people like you and our sponsors: