Michael Fremer defends Hi-Res digital while chewing out Gizmodo

Discussion in 'Audio Hardware' started by violetvinyl, Jan 25, 2015.

Thread Status:
Not open for further replies.
  1. Scott Wheeler

    Scott Wheeler Forum Resident

    Location:
    ---------------
    there is also the classic case of this....
     
  2. 2xUeL

    2xUeL Forum Philosopher

    Location:
    Albany, NY
    Love Key & Peele and love this skit!
     
    Scott Wheeler likes this.
  3. Bill

    Bill Senior Member

    Location:
    Eastern Shore
    I'm referring to the ability of the owner of a CD to easily, quickly and cheaply burn an unlimited number of perfect copies on virtually any PC to "share" with friends, precluding record company profits from additional sales to them. How often does someone make such an offer for a new release to any of us? While someone can dub a vinyl record to a cassette for someone or do a needle-drop, it's not as convenient a process. That's why I suspect industry may be embracing vinyl- more sales!
     
    2xUeL likes this.
  4. Bill

    Bill Senior Member

    Location:
    Eastern Shore
    As to my comment about Fremer's combatitive style, he reminds me of a certain presidential candidate who has since emerged!
     
  5. motionoftheocean

    motionoftheocean Senior Member

    Location:
    Circus Maximus
    Fremer has stated unequivocally, on numerous occasions, that vinyl has greater frequency response and bit depth than CDs. by extension he has also said that analog tape has greater frequency response and bit depth than do CDs. This is obviously a hotly debated issue with many engineers who have a far more scientific background than Fremer claiming fairly reasonably (if you care to read what they have to say) that vinyl and analog tape aren't capable of delivering dynamic range and frequency anywhere near that of digital. Nothing to do with the aforementioned topics are aesthetic or subjective merit; that they are still largely unresolved means they're about as far removed from objective knowledge as is possible. My issue with Fremer is when he makes comments like saying vinyl has a frequency response up above 50 kHz, says he's seen the spectral analyses to prove it, and then when challenged falls back on, "I just think it sounds better." That's likely where the claims of pseudoscience originate. I don't think anyone could argue with Fremer for saying he prefers the sound of vinyl over CDs - and in point of fact, I happen to agree with him - it's the stance of presenting something as scientific and then resorting to emotion that weakens his position.
     
  6. motionoftheocean

    motionoftheocean Senior Member

    Location:
    Circus Maximus
    haha you'd piss Fremer off so much with that comparison
     
    Bill likes this.
  7. Chooke

    Chooke Forum Resident

    Location:
    Perth, Australia
    Yes, it was a well designed study taking over a year to complete. The hi res source was played either "pure" through channel A or down-sampled to 16/44 through channel B. The loop allowed the music to also be played through both paths thus providing the channel X. When using the channel X loop, the listeners were asked to identify the points where the music went from hi res to 16/44 and vice-versa.

    I think you are conflating subjective preferences with objective measures and missing the point. Just because A is superior objectively to B, it doesn't necessarily follow that any one individual cannot prefer B over A. And just because some individuals prefer B over A, it doesn't invalidate the objective measures or the science behind it. For example, my audio preference is for a linear frequency response, telling it as it is, and a good friend of mine prefers more of a "warm" sound. Two different preferences but the measures will show that my equipment has a more accurate reproduction whilst his has a mid-bass emphasis. The measures don't lie but it also doesn't mean he should not listen to music in a way the he prefers. I respect my friend's opinions on why he prefers the warmer sound as he couches it in subjective and emotional terms, and he makes no claims that the better sound (to his ears) is based on some technical or scientific fact regarding frequency responses.

    Another, non-audio, example is the story of two high performance cars having the same standing start acceleration and top speed but the metrics show that car A develops most of its torque low in the rev range while car B has a more even spread of torque. Which is better? Objectively overall it is car B, but some drivers may prefer car A for its bottom end oomph and burnouts ability, while others prefer the even spread of power in car B. That is subjective but objectively, without even having to drive the cars, we can tell that car A will have better acceleration at lower speeds while car B will be better at higher speeds and more consistent overall. What it can't say though, is which of the two characteristics any particular individual would subjectively prefer.

    I think I read the same claim in one of his articles. This is an example of pseudoscience. Firstly it is true that vinyl has a greater frequency response than CDs (but not hi res) but what he omits is that CDs cover the range which humans can hear, let alone whether any of these supersonic frequencies has musical content anyway or that it was recorded and survived the production process. More importantly is the linearity of the frequency response. Vinyl starts rolling off around 16khz (contributing to its "warm" sound) whereas a CD is still within 0.5db at 20khz. Even this is moot as our hearing's maximum sensitivity is around the 4-10khz vocal mid range (which makes sense from an evolutionary perspective), so any harmonics at say 18khz would likely be masked by other content.

    His claim that vinyl has a greater bit depth equivalent is wrong and easily proved wrong by basic mathematics. If it was higher, then vinyl would have a greater dynamic range, higher signal to noise ratio and a quieter noise floor than CD's 16bits. It doesn't, and using the maths backwards it supports the widely held view in audio engineering that vinyl is close to about 13bits, which is about the same as analogue master tapes. And before anyone shoots me, I may add that 13bits is enough resolution to faithfully reproduce most music, albeit at the expense of some noise (eg vinyl surface noise or tape hiss).
     
    Last edited: Oct 3, 2015
  8. missan

    missan Forum Resident

    Location:
    Stockholm
    I don´t think we can say that our hearing sensitivity is that high in freq. It is most sensitive from around 400Hz up to 4kHz. 16kHz is not contributing to a warm sound, very few, and I mean very few, can actually hear 16kHz at these low levels we are talking about in music. The warm sound comes from mid bass bumps normally.
    It is not that easy that we can replace vinyl with 13bits, it is true that the dynamic range is about the same, but at the same time if we reduce bits we also reduce resolution, so it´s difficult to really compare.
     
  9. Chooke

    Chooke Forum Resident

    Location:
    Perth, Australia
    You are correct on the hearing sensitivity range, thanks for that.
    The 16khz roll-off does contribute to the warmer sound effect. It is not the whole story but that, and a mid-bass emphasis it does play a part. I partly agree with your comment on the 13bits in that it is an equivalent based on certain measurements. The bits can tell you about the levels of quantisation error and hence noise in digital and the equivalent "number of levels" in vinyl is limited by the margin of error, i.e. the noise floor which can be given a bit equivalent. There's no practical difference between the noise caused by surface imperfections or the noise caused by dithered quantisation, but to say vinyl has a resolution that is equal or better than 16bits is quite a stretch.
     
  10. Dave S

    Dave S Forum Resident

    Well, I would be careful of the science behind it all. Very few people here understand how digital audio works (and I would include Fremer in that category - 16 bit cannot accurately produce a sine wave!).
    [​IMG]
    Here's a picture of a record groove btw - I would say it's about 50 microns wide. That's about 500,000 times the size of the smallest atom. The smallest movements in the needle would be several times the size of an atom. You are not going to get 10^24 bits resolution, however hard you try.
     
    Halloween_Jack likes this.
  11. missan

    missan Forum Resident

    Location:
    Stockholm
    Yes I agree, it´s very much a stretch I believe. We must take into account the different vibrations and different needle velocities that will effect resolution when playing vinyl. So one system will be very different than another system, system is TT/arm/cartridge/record. I mean one system might be much lower than 13bits while another system might be 14 or 15bits. It will vary quite a lot I think. But that it should be equal to 16bits I cannot see.
     
    Chooke likes this.
  12. Scott Wheeler

    Scott Wheeler Forum Resident

    Location:
    ---------------


    It does have greater frequency response. That is just a basic fact. Not that the extra frequency response is of any use but it does have it. Bit depth is kind of a strange claim since vinyl has no bits.




    What do you mean "by extension?" Either he sadi that or he did not say that.




    I have read plenty on the subject of dynamic range written by various audio engineers. Enough to know that you are speaking in far too broad terms here. Digital can be anything from very low res MP3 to 24/192. There is some debate as to what the real world dynamic range is of vinyl. The debate stems from the fact that the dynamic range of vinyl is frequency dependent. There is also some debate as for the real world dynamic range of digital in terms of how it is figured based on bit depth and how that really works in terms of distortion at low levels. I have yet to read any audio engineer dispute the fact that vinyl is capable of a wider frequency response than red book CD. That's audio 101. The debate is whether or not that extended frequency response is of any use. I am of the opinion that it is not. I would be interested in seeing what Fremer actually said on these issues but I know I can not trouble you to actually cite any references for your claims.



    If that is where they originate than yall got a big plate of crow to eat. Because he is right. Here is a suggestion. do a little research on quadraphonic LPs and check out the carrier frequencies and then get back to us on this subject.


    I think you got some real reassessing to do on this one. He is dead right on the capacity of vinyl when it comes to frequency response. It is just a matter of whether or not it matters.
     
  13. Scott Wheeler

    Scott Wheeler Forum Resident

    Location:
    ---------------
    I think it might be you who is missing my point. I did quite the opposite of conflating objective measures and subjective preferences. By pointing out that there is no such thing as an objectively "superior" or "inferior" measure of anything without someone subjectively setting a standard by which better or worse are gauged. What I did was place the proper separation between subjective goals and objective measurements.

    You already made a false premise in this argument. No such thing as an objectively "superior: measurement. "superior" is a purely subjective abstract idea. You are conflating subjective goals and standards with objective measurements.

    Let me give you some simple examples.

    What is a "superior" weight? 100 lbs or 200 lbs?
    What is a superior length? 1 foot or 1 mile?
    What is a superior frequency 100 hz or 100 khz?

    There is no objective superiority inherent in any of these measurements. Superiority can only be determined once someone has made a subjective decision on what is an ideal in each and every case above. Objectively things measure the same or differently. they don't measure better or worse.





    That is true. But those measurements are not objectively "better" or "worse." They are just facts with no intrinsic values.



    Yes I would agree that your preference is for a *more accurate* frequency response. Not a "superior" frequency response.


    But it is based on a technical fact regarding frequency responses. The different frequency response that he prefers is *in fact* less accurate but to his ears is "superior." That is the standard by which he measures better or worse and it can be correlated to objective measurements. Again, once you start labelling any objective measurements as "better" or "worse" you are subjectively setting a goal or standard by which objective measurements will be judged **subjectively** In this case you and your friend have different subjective goals. What constitutes a "superior" measurement for you and him are simply different.
     
  14. Chooke

    Chooke Forum Resident

    Location:
    Perth, Australia
    Quite amusing... putting aside the embellished claims, it is amazing though how the engineers have refined the old technology to produce the sound quality it has.
     
  15. Chooke

    Chooke Forum Resident

    Location:
    Perth, Australia
    Mate, lets just agree to disagree. I still don't think you get it. What you say is meaningless without a context. If you define the power of spotlight in distance of light, then obviously 1 mile is objectively superior to 1 foot. If you define convenience as the distance it takes you to get to the water closet then obviously 1 foot is objectively superior to 1 mile. Someone may prefer planning a trip to the toilet with a leisurely walk so subjectively prefer 1 mile to a foot, that doesn't mean that the definition of convenience is merely a subjective goal.
    If you define pure hi fidelity as music reproduction accurate to the source, then a linear frequency response is objectively superior to a non-linear response - though subjectively you may prefer some colourisation, it doesn't change the meaning of high fidelity (given it has relative meaning, as no system will be 100% accurate to source). Some young people I've come across prefer, or are indifferent, between 128kbps and CD, doesn't mean that a statement of CD being objectively a higher quality format is based on a subjective notion.
     
  16. 2xUeL

    2xUeL Forum Philosopher

    Location:
    Albany, NY
    Don't forget that at the end of the day, the speaker has to move continuously, be the source a record or a CD.

    On a side note, how is the smallest movement of any object in reality incremental as you have described (the needle's 'smallest movement' being several times the size of an atom)? Space is continuous, right? So for an object to change position it can't magically warp to a different position as you seem to be describing...? Sorry for the confusion.
     
  17. head_unit

    head_unit Senior Member

    Location:
    Los Angeles CA USA
    Hey! It's digital! It MUST sound great!
    :D

    I think we've already established that it depends on the sampling rate and bit depth.
    :laugh:



    Actually there is another dog in the hunt :)hide:) so to speak, and that is the QUALITY of the bits and perhaps the sampling. But that is another thread
    http://forums.stevehoffman.tv/threads/doesnt-anybody-test-audio-quality-in-depth-any-more.468564/
     
  18. head_unit

    head_unit Senior Member

    Location:
    Los Angeles CA USA
    Well, professionals can be short-sighted morons. Witness as I've ranted before the lack of artist/track names embedded into the CD, and the idiocity that PLAY doesn't actually play a DVD.
     
    Scott Wheeler likes this.
  19. head_unit

    head_unit Senior Member

    Location:
    Los Angeles CA USA
    Well, when I said "stereo enthusiasts" I meant (and *think* I said) assembling components for sitting and listening-JUST listening-at home. I know a LOT of young people and NONE of them are in this category.*

    I doubt they would call themselves "stereo enthusiasts"-if you asked them that they would look blank, it is a meaningless term. "Music lover" they could respond to, and some are putting boomy systems in their cars.


     
  20. Scott Wheeler

    Scott Wheeler Forum Resident

    Location:
    ---------------

    If you want to agree to disagree why say you still don't think I get it. I still don't think you get it. apparently you want to continue to argue certain points. And so here we go.


    That does seem to be the point you continue to fail to get. Objective measurements are not "better" or "worse" than one another. the "context" is always subjective because it always involves a person making a subjective decision about what is **wanted**

    No. It is objectively more powerful. If one is trying to simulate candle light for a close up for a photo shoot I can assure that in that context (meaning what is subjectively desired) 1 mile is definitely NOT superior to 1 foot.


    Exactly. Thank you for proving my point. There is no intrinsic "superiority" between the measurement of 1 mile or 1 foot. Some human being has to make a subjective choice about what they want as an ideal measurement. ***That is subjective.*** It **always** is.



    Not what I am saying. What I am saying is the idea of "superior" or "inferior" are purely subjective. As you have pointed out, whether or not "convenience' is superior or inferior depends on one's *subjective choice.*


    1. That depends on what you define as "the source."
    2. In audio "High fidelity" is nothing more than an old audio marketing term. Stereo recording and playback have nothing to do with literal "fidelity" to the original source. That being an original acoustic event in a real space. That simply isn't how audio works. If you want to let an old marketing term define your goals in audio that is a choice you get to make. It's not for me. I find it to be pretty arbitrary in practice.




    I suspect we don't even agree on the meaning of "high fidelity" as it pertains to audio. And to say no system is 100% accurate to it's source is the understatement of the thread IF one considers original acoustic event to be "the source." The original acoustic waveform will always be so far removed from the final acoustic waveform in the listening environment that any discussion of things like linear frequency response in the electrical signal of the playback chain will be beyond trivial. Audio recording and playback has nothing to do with literal fidelity to the original acoustic waveform in the original acoustic environment.


    It just means they have *different* measured levels of data and sampling rates. What is "superior" depends as always on what is subjectively desired.

    So do you want to agree to disagree at this point or do you want to continue to discuss the issue?
     
  21. motionoftheocean

    motionoftheocean Senior Member

    Location:
    Circus Maximus
    I've already done that. In fact, literally everything I've said about him has purposely been confined to the source I cited to make it as simple as possible. Instead you're blindly defending him and trying to pass off supposition as fact. At other times you're outright inventing facts (like claiming vinyl has a greater dynamic range than does CDs and saying that's "audio 101"). Every point introduced to you is met with another tangential question. You're not interested in science, just asserting that Fremer isn't wrong by way of conjecture.
     
    Last edited: Oct 4, 2015
  22. drbryant

    drbryant Senior Member

    Location:
    Los Angeles, CA
    That's interesting. I know quite a few young couples who have decent music systems for their homes. I have no idea if they sit and listen to music without doing anything else if that's your definition of a "stereo enthusiast" as opposed to a "music lover," but they have decent equipment. Perhaps we are talking about different ages - I am talking about younger professionals (30's usually) who are typically in their first home. As far as car stereos are concerned, I know that Honda considers it an important part of their brand imaging that the systems in their better lines be known for superior sound quality. Elliot Scheiner designed the systems in the Acuras. I have no idea if they sound good, but Honda thinks that people still care.

    Of course, some have no idea what they are doing - one guy I know has a beautiful apartment, and his speakers are in the ceiling right above the sofa . . . . .
     
  23. LeeS

    LeeS Music Fan

    Location:
    Atlanta
    This is not a good analysis. A vinyl LP can hold a 50khz signal. Using Nyquist that equates to a 100khz sampling rate. The LP is a hirez format.

    Now if you have doubts, listen extensively to a good analog rig and compare it to DVD-Audio. It's very clear that LP is at or exceeds 24/96 in sound quality.
     
  24. Dave S

    Dave S Forum Resident

    I was talking about the dynamic range, i.e., the bit depth.

    I'm not sure you can even do a fair comparison between the two formats, since they are totally different. Yes,
    you can compare DVD-Audio to CD if they have the same mastering and you use exactly the same equipment.
     
    Brother_Rael likes this.
  25. JBStephens

    JBStephens I don't "like", "share", "tweet", or CARE. In Memoriam

    Location:
    South Mountain, NC
    A loudspeaker cannot stop at one point in space, then magically re-appear at the next. It has no choice but to move through every possible iteration as it reproduces sound. Therefore, the speaker creates its own infinite level of quantization regardless of bit depth.
     
    crispi likes this.
Thread Status:
Not open for further replies.

Share This Page

molar-endocrine