|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
Going to build a DAC. Go for digital filters or filterless? Your opinions , please. THANK YOU .
Follow Ups:
Dear Make,I would think that the answer depends on your level of technical ability (as well as your sonic preferences) - there is no sense in biting off more than you can chew comfortably. Remember, a given design or design approach is only a tool - to a large degree the outcome will depend on how well you can make use of that tool.
If you have very good multilayer circuitboard layout capabilities as well as very good schematic design abilities, a digital filter will be fine, and there is a good chance that you can get it to sound better than most filter-less designs.
On the other hand, if you donÕt have good (perhaps I should amend that to ÒexcellentÓ) layout, schematic and construction abilities, doing without the digital filter will give you less implementation headaches and will probably sound better.
I should also point out that some DAC chips come with the digital filter built-in, so those who are philosophically against digital filters should best avoid these chips.
hth,
jonathan carr
Speaking of analog filters, when I added balanced to my Samsung I had at first added in a fiter circuit. Then a friend told me to try it with out the filter. At first I was shocked and wanted to turn the tweeters on my Maggies down. But once I was used to the new sound, I'm happy to leave the tweeters right where they are. The DAC has four DC blocking caps and my cables come right up to the output side of those caps. bobwire
Just like the 47 Lab Progression.It's really very good.
George
Hi there,If you are living, dancing on the edge, if your equipment doesn't mind a bit of supersonic stuff being chucked out by the DAC, if you want to connect intimatly to the music, go Footlose and Filterless. You will be dancing and listening through many a night.
If you want a boring square and make your DAC sound like all the others out there, rather than sounding like music - then put a filter in. If you listen through the Tube of a 'scope and the screen of Audio Precision One, use a filter.
Remember, you got the power - use it.
Ciao T
This is one of those things that are, out of context, meaningless.You should be more concerned about a good circuit design, and address it's needs rather than surcumb to "conventional wisdom". I wonder about these people who abhor transformers (for example) and go to great lengths to avoid their use with a quirky result.
If the rethinking produces an elegant -- I am very much into elegant -- solution, then fine. Otherwise, I'd say, do what's right.
As we say in Perl, TMTOWTDI, or, There's More Than One Way To Do It.
Do you really mean digital filter? I don't think you can build a half-decent DAC with no digital filter. An analog filter is a different story, and many designs don't have it.
--Mr.Khomenko to post your answer to Mr. Qvortrups polite request to your statement in your post.I think there are many of us here would like to know the answer.
Curious,
Des
The answer is contained in that sentence:"I don't think you can build a half-decent DAC with no digital filter."
"I don't think" expresses an opinion. "I don't think" having your DAC run at 44.1kHz is good idea given today's understanding, but that dosn't mean everyone has to agree with it.
If Peter stated that in his opinion the only way to build a good DAC was with no filter, I would simply take it as such, his opinion. I would not choose to confront him.
Arguing about such opinions here would be close to agruing about particular products and their merits - something the customers should do, not manufacturers.
Dear Victor,Thanks, we do not employ digital filtering at all, so your objections most certainly apply.
I really do wish you were right about the market place deciding, if this was so then we would have better products available overall and thus a better market place, unfortunately for those of us who are serious about sound quality, the marketing and advertising budget is more market relevant than the research and development budget, so real improvements are rare.
I was not asking for a public argument, but a debate about the fundamental reasoning behind the choices either of us make when designing products, I think many members of the forum would find such an exchange interesting and enlightening.
There is too little publicly accessible debate about all matters in audio design, people are only told what is considered useful to the "image" of the company i.e. what makes it look innovative and clever, measures which in most cases just support already existing notions of "advanced technology", rather than more wide ranging improvements in sound.
I, for one, is prepared to invest some time in discussing this.
Sincerely,
Peter Qvortrup
Hi Victor & Peter,Maybe the issue is not to get rid of the dig filter,
but to design a proper one. Even the common DF's
such as 1704 and in Wolfson DACs with slow rolloff
response show very good step response in that mode.
The 1704 certainly sounds somewhat more 'analog like'
in slow rolloff mode.I feel the oversampling function of DF is a benefit,
apart from its rolloff characteristics.
I have tried PMD100 at 4x and 8x and it
is smoother and more natural in 8x. This would not
affect it's transient response at all (still brick wall)Given enough DSP power a DF should be possible
with proper transient response. I know Julian
Dunn implemented a custom DF in Prisms DA2 which
despite many ugly chips in sig path is very good.
Not sure whether it was FIR or IIR.One thing I'm not clear on is this term "time
smear" of DF's. If they are linear phase what
is time smear? How is the DF distorting time
relationship of signals if group delay is
constant?Regards,
Terry
Based on discussions about this in the past here, the "time smear" that is being attacked is the fact that a single sample exihibits energy before and after its interval when reconstructed. They seem to feel that a sample should have an energy contribution with a duration less than 1/Fs secs.For some reason people don't realize the only way to reconstruct a sine wave with a freq = Fs/2 over the interval of the two samples representing it is to represent an individual sample with a sinc function scaled by the samples magnitude. This is tried true and correct.
It is a poor misunderstanding of sample theory.
(but that doesn't mean incorrect reconstruction via a steeper filter doesn't enhance the listeners enjoyment.)
-CAL
Very good points, Terry. There is no question one doesn't have to use some "canned" filter, unless of curse the function like the HDCD is desirable.However, we should not lose the sense of reality here. The digital filter is but one brick in the overal building. You can certainly keep developing your favorite algorithms - and I have designed many diferent filters during my years with the "serious" industry - but let's not forget that there are quite a few other areas in the product like a CD player that deserve at least equal attention. The search for the "best" filter design can easily be made endless to detriment of other functions.
So for instance using a VERY nice canned filter like the DF1704 does allow you to concentrate on those other areas. Product design is always the balance of many conflicting approaches. Which area seems to the designer as the one with the largest possible contribution and improvement? That is largely a personal desicion and reflects in the overall product feel and sound.
***Thanks, we do not employ digital filtering at all, so your objections most certainly apply.Yes, I gathered that much from your reply. I didn't know that. My opening comment was of course not directed at any particular brand/product, but rather simply a personal opinion answering the DIY'er question.
***I really do wish you were right about the market place deciding, if this was so then we would have better products available overall and thus a better market place, unfortunately for those of us who are serious about sound quality, the marketing and advertising budget is more market relevant than the research and development budget, so real improvements are rare.
I see this slightly differently. Like in art, cigars, cars, women, etc, there is never single "best", but rather the whole slew of great products. We all have preferences and who's to tell which ones are right?Your point about marketing vs. the design is of course true, and is way of life in all the industries, running shoes, whatever. We all see life sometimes as fair, sometimes as not.
***I was not asking for a public argument, but a debate about the fundamental reasoning behind the choices either of us make when designing products, I think many members of the forum would find such an exchange interesting and enlightening.That is probably true, but often such debates take life of their own, and often the truth is the first casualty.
My personal preference is for simply expressing our design objectives and preferences. I certainly see nothing wrong with you stating firmly that you believe the best sound comes from a system designed your way. But I do make rule for myself of not commenting on any particular product from another manufacturer, other than making simple factual statements (correcting a mistake about the number of inputs, that sort of silly stuff...) or a positive remarks (I have done some of that too). Other that this, I would rather err on the side of caution. I am not saying this is the only rule, simply that it is mine.
***There is too little publicly accessible debate about all matters in audio design, people are only told what is considered useful to the "image" of the company i.e. what makes it look innovative and clever, measures which in most cases just support already existing notions of "advanced technology", rather than more wide ranging improvements in sound.That of course is true too, with the companies representing the whole spectrum of possibilities. Some are, shall we say, agressive, others don't say much at all. Me, I tend to be reasonably open about what we do and how.
***I, for one, is prepared to invest some time in discussing this.
I am sure many folks here do appreciate that. That also can be part of good marketing.
Best of luck,
Victor
Dear Victor,Thanks, I thought it might be interesting to explore the motivation and background to your opinion, have you for example actually listened to a filterless DAC??
I think many of our fellow asylum inmates and participants would be interested to know how you arrived at this view, and since you express openness about your views, why not??
In my view audio is not art, because it does not take part in the performance, it is merely a medium through which others art can be explored, experienced and enjoyed and as such is incomparable to the interpretative art available in recorded music.
I have always found it very pretentious and overly self important when audio designers elevate themselves to stand on the same timeless rostrum occupied by real artists like Mozart, Furtwangler, Godowsky or Mahler, these are the true artists whose art has made a genuine contribution to the human spirit, us mere mortals will have to contend with being the servants that provide some of the mechanics, our art is not timeless, theirs is, but only if we treat and preserve it with care and respect.
Everything people like you and I do will invariably be scrutinised and viewed as "marketing", in the US there seems to be only cynicism in relation to statements by industry people, it has to be part of some ploy to sell more.
Some of us may actually be doing what we do for reasons that are slightly broader and more widely considered than just cold mammon, at least here in Europe there still exists value systems that involve and recognise aesthetics and ethics.
More than a little sad that we cannot be part of a discussion without being accused of always having a purely commercial agenda.
Sincerely,
Peter Qvortrup
***Thanks, I thought it might be interesting to explore the motivation and background to your opinion, have you for example actually listened to a filterless DAC??Not lately, and perhaps I should revisit this issue, although it is not too easy to compare oranges with oranges... I'll think about it.
***I think many of our fellow asylum inmates and participants would be interested to know how you arrived at this view, and since you express openness about your views, why not??***In my view audio is not art, because it does not take part in the performance, it is merely a medium through which others art can be explored, experienced and enjoyed and as such is incomparable to the interpretative art available in recorded music.
***I have always found it very pretentious and overly self important when audio designers elevate themselves to stand on the same timeless rostrum occupied by real artists like Mozart, Furtwangler, Godowsky or Mahler, these are the true artists whose art has made a genuine contribution to the human spirit, us mere mortals will have to contend with being the servants that provide some of the mechanics, our art is not timeless, theirs is, but only if we treat and preserve it with care and respect.
I don't see every reference to art as pretentious. Art covers wery wide range from very simple to the most delicate. But to say that the boy next door is not creating art when he is just studying oil painting is incorrect. One doesn't need to be Godowsky to be an artist. Art is more about the method, an approach, the internal desire and search for beauty. Lot of that IS present in audio design, so the word "art" doesn't scare me there.
***Everything people like you and I do will invariably be scrutinised and viewed as "marketing", in the US there seems to be only cynicism in relation to statements by industry people, it has to be part of some ploy to sell more.I think this may be too broad a statement. I hear many different reactions from people, and while some are definitely cynical, there ARE many people who do understand things and are capable of separating the fluff from truth. I have to agree, that in this industry this is getting harder every day. There was much less voodoo in the fifties...
***Some of us may actually be doing what we do for reasons that are slightly broader and more widely considered than just cold mammon, at least here in Europe there still exists value systems that involve and recognise aesthetics and ethics.
I am of course less in touch with things in Europe, but there you too have people fighting over the diametrally opposed philosophies (and buying such dissimilar products) - so perhaps things are not that different after all.
***More than a little sad that we cannot be part of a discussion without being accused of always having a purely commercial agenda.
I think we always should be mindful of that. But my personal "line in the sand" is drawn somewhere between the world of picture-perfect responses and the one of completely loose ones.
See the "Phenomenology of Perception" by Maurice Merleau- Ponty. I believe that our striving to get a clearer perception of a musical performance through our equipment is an artistic endeavor. Certainly not as significant as the work of a great artist, but "art" none the less.Dave
Dear Dave,And herein lies the problem, mass popular culture has elevated everything to a position of being "art" thus drowning and devaluing true art and imparing our judgement in a sea of mediocrity.
Says everything about our age, and nothing about real art!.
Sincerely,
Peter Qvortrup
Dear Peter,I think an important distinction has to be made between what is not art and bad art. When I took aesthetics philosophy, it struck me that most of the people I was reading had something in the art world that they didn't like and were trying to come up with a definition of art that ruled that thing out as art. Bathroom graffiti is not on the same level as Picasso, and us usually not even an attempt at art. But, I have seen some graffiti that was art (both good and bad). Even if we decide something is art, we still have to decide if it's good art.
I certainly was not trying to trivialize the works of great artists. My point was that, in creating a link between us and the performance through our equipment and then trough our senses, that we are not just passively receiving sense data. We are actively working to perceive the performance. This is an artistic process. It is not nullified simply because people do art on a grander scale.
Dave
Dear Dave,Like so many other aspects of mass commercial culture, the lowest common denominator rules, if your equipment is not capable of presenting you with a performance that does not require you to strain your senses to hear what is going on, then it is doing a pityful job and cannot be good enough.
If there is "art in listening" then I fear for our common future, because under that scenario no timeless art can be created, it can only be consumed.
Sincerely,
Peter Qvortrup
Dear Peter,The equivalent of "lowest common denominator" in the audio world, would be rack systems, Bose, cheap direct drive turntables etc. If you’re not sure what this is you can go to Circuit City and they'll be happy to demonstrate it. In visual arts its velvet Elvises and kids with big eyes. There will always be, and always were lemmings lusting after garbage. It has yet to destroy societies ability to create art and it never will. I never tried to say that bad equipment wasn't bad or that we should empty the National Gallery and let kids spray paint on the walls.
What I am saying is that listening to music (or looking at art) is an active creative experience in which you are communicating (on-way communication) with the artist and is much more than passively receiving sound (or light) waves. This observation does not rationalize mediocrity nor does it negate expertise.
When doctors say that there is and art to practicing medicine, they don't mean that you should let any bozo do medicine. In the same vein saying that there is an art to listening does not mean that any equipment flies and all opinions have equal weight. There's still room for expertise.
As far as timeless are goes, time is the only test of that!
Sincerely,
Dave
.
--- very Diplomatic answer-you might have missed your calling in NY!---I like Peter's DACs and your Amps--guess that kinda makes me 'half decent'Listening to MUSE/le mystere des voix bulgares--involving.
Des
nt
Well, both really.I presume by not using a digital filter you mean having your DAC run on 44.1kHz, and that is something I don't like.
Or perhaps we are simply talking about differen things. So maybe you could explain quickly what you are suggesting. But without a filter - and we are using the DF1704 now - you also get aliasing.
But as a geberal rule, I am not completely married to measurements.... unless it looks too bad.
I`m not really suggesting anything specific, merely that the prime function of a DAC is surely to turn digital information into music? If DAC `A` (without filter) sounds better than DAC `B` (with filter), then surely we can assume that any contradictory measurements from the Laboratory must be put in perspective. These measurements are only an indication of sound quality, even the likes of Paul Miller concede this point; we`ve all read of amps which measured fantastically but sounded terrible - I`m now of the opinion that it`s near impossible to determine which measurements are more important than others, and that the subjective importance would probably vary from component to component anyway. The analogy I`d make is that if you had a room which was out of square, sloping with an uneven floor and crooked doorway, how would you then hang a large painting? Would you simply take your spirit level out, tape-measure and place it perfectly in the centre of a wall - even though it looked badly out of true, or would you hang it so that anyone looking wouldn`t notice that it wasn`t level? How can you measure the effectiveness of a digital filter as opposed to an analogue filter anyway, more effectively than with the human ear? How long has it been accepted that we can`t hear above a certain frequency, only to learn that super-tweeters can bring definate improvements? At the end of the day, the proof of the pudding is in the eating - not a scientific analysis of the pudding viscosity, temperature, or nutrients.
"If DAC `A` (without filter) sounds better than DAC `B` (with filter.."if this is the case you can assume 1 of 2 things, the pre-amp-amp-speakers used where not affected by the hf image components from the dac, or 2, they had a negative affect that is enjoyable, at least for the time being with the particular source material used. Seems like a very abitrary way to build a good system.:)
steve
...I`ve heard some systems which displayed all sorts of distortions, totally unlistenable to for more than a few minutes - now I realise that they were really true hi-end, merely revealing limitations which my own system was incapable of exposing. In fact, recently I`ve been listening to hour upon hour of music in the vain hope that some sort of distortion will make itself known to me but alas, all I ever get is clean, pure vocals and dynamic instrumentals.....really is a sickener! Of course, somebody suggested that there may be audible and inaudible distortion, but what would they know? Keep smiling Steve, `cause I`m grinning from ear to ear permanently nowadays :0)
Dear Steve,I have said this before, the "out of band" signals or as you call them hf image components, present on the measurements may be a function of the way we measure rather than a reality when music is played, since they are not generally audible.
Asto the question of whether they add "enjoyment", my contention is that it is the removal of the massive (and ever changing) time domain interference (see the much improved pulse behaviour) that make the 1xoversampled DACs sound the way they do, not some weird intermodulation of the out of band signals with the signal.
A negative "enjoyable" effect sounds like a very contentious proposition to me.
Is there really no limit to the arguments the proponents of the status quo will use to defend their position??
Sad in my view.
Sincerely,
Peter Qvortrup
Are you guessing the image is not creating a pleasent distortion(or have you seen it in measurement graphs)Peter, also what of the inclusion of differing DF modes on the newer hardware that does the job of creating a narrower impulse response as you get with the 1x/analog?
Sure the advantage is that you have a filter that takes an incredible number of poles (and extreme precision) to recreate digitally but you also have the noise and phase irregularities it brings in with it. (I also understand it is also easier of the DIYer to adjust the sound via a mod of the analog filter and that is part of the appeal of your DACs)
Do you not see the 1x/analog approach as a short lived approach that will ultimately be surpassed by the silicon solution?
-CAL
Dear Cal,What is a pleasant distortion?? That sounds like a contradiction in terms.
When Michelangeli runs his fingers across the keyboard in several fast glissandi in his 1948 recording of Brahms' Paganini variations then I hear no distortion that smears the individual notes when the oversampling and filters are removed and replaced by a good and simple analog filter, when the digital filters are in place then I hear all kinds of smears, regardless of type of device and measured "precision".
The same is the case when at the start of Sergiu Celibidache's recording of Bartok's Concerto for Orchestra you can feel the audience "breathe" in the hall just as he walks onto the podium before the music starts.
> > Peter, also what of the inclusion of differing DF modes on the newer hardware that does the job of creating a narrower impulse response as you get with the 1x/analog? < <
"Creating" is the key word here, how do we know what is supposed to be there??
Methodology is irrelevant, every experiment I have made over the past 25 years have reinforced my view that ANY attempt to "correct" or "recreate" anything in an audio signal is doomed to complete failure, we are not dealing with properties we can fully or even partly describe even by complex algorithms, the best we can do is to observe and then only correct the faults where they occur, not "recreate" what we in our limited wisdom think might have been there to start with.
I can become just as fascinated with the latest greatest technology as anyone else, but anyone who has even a shortish memory will recall numerous claims made by numerous companies and individuals about having the solution to this or that problem and each time time has proven them wrong!
Silicon solutions to these problems are about as effective as they are in analogue audio, which is why a growing number of music lovers are listening to single ended triodes with no correction, these also measure poorly, but sound far far superior under real life conditions.
Why are none of you guys willing to consider the possibility (I would call it fact, but for the same of argument "possibility" seems more appropriate) that our measurement methods and technologies are woefully inadequate and it is they that need to be revised??
By refusing to face up to the inevitable and continuing endlessly throwing incomplete and inadequate theory and its resultant technology at the problem you are only delaying the real revision of our measurement technology that should have taken place several decades ago.
Whatever happened to 19th Century observational science??
Sincerely,
Peter Qvortrup
This is getting interesting, dig filters, no DF's,
what is really accurate to the original event
or what gives the "feeling" of the original event,
what gives the emotive effect of the original event.Are we confusing them?
Sadly an area which is totally missed is the record chain
of the original event.The other day I was looking at the channel strip schem of an SSL mixer and there must have been 10 5534's just in the channel. Now there's the group VCA's and summing amps etc etc. Most pro gear just throw chips at everything, and lots of 'em and I don't think I've ever seen something like a 627 in a bit of pro gear, not even hi end pro gear.
Even outside the much maligned studio and into the world
of minimalist recording. *the* most widely regarded hi end
pro mic pre used for minimalist stereo recording is the
Millenia Media HV3. Look inside and there are plenty of chips
mixed with discrete, all SS and lots of global FB.
Plug that into a nice AD and the choice is usually Prism,
dB Tech, dCs, Benchmark and a few others. Yep, more chips,
still haven't seen any of those nice 6172 or 627's or
even any nice class A circuitry in there!The point I'm eventually getting to is this, are we
exponents of 0 FB triodes and iron making up for inadequacies
in the recording chain by adding something that makes
the whole thing more digestible?People here often talk about 'real sounding'
but the only way to get real sounding is to
AB the source in other words be at the recording end
when it was done. To witness the original event.I'm currently working closely with a minimalist recording
engineer recording in chuches and recital halls. I'm
building him a series of mic pre's and after that AD converters.
and DA's.I'm sure by the time we get to the other end of this
I'll have a more accurate perspective on what perceived
accuracy really is. And we are not scared at all to get
euphonic, ultimately it's gotta be engaging and emotive,
transfer the message, whatever it takes.Maybe the message is damaged at the microphone membrane and so forever
thereafter?Food for thought.
Terry
A while back I was just screwing around with a Shure SM57 and recorded at 24/96 my bedroom's enviromental sounds with the TV on tuned to the news. You know what, the life was already gone from from the live feed and the 24/96 was undiscernable. Even decimated to redbook, there was little difference. It wasn't the format, it was the recording.Granted that isn't a great mic but I was just recording a very limited source(builting tv speakers).
I personally thing the majority of improvement will be in newer extremely low mass mics and less so in the playback chain. Fortunalty some of this can be recreated by a minor digital prep. Even the new Panasonic DVD RK91 I picked up that has 3 upsampled "remastering" modes do a decent job of opening the presence again. It works by adding over nyquist components via upsampling and minimal filtering.
Sound familiar?
-CAL
Dear Terry,You are mostly right, the message is now so much the hostage of the medium that it is almost impossible to transfer the musical event untarnished from the recording equipment to the software.
Microphones with superwide bandwidth, damage the signal, they are very inefficient and therefore super insensitive to low level information, (it takes a fair amount of overall energy to make them start moving).
One of the major secrets behind why the great recordings of the 1950's and 1960's were so good, were the "old fashioned" large diaphragm microphones, what would Ella Fitzgerald or Louis Armstrong or Bing Crossby sound like with a modern mike? I feel sad for the artists of today who have to leave their art in the hands of the kind of studio equipment we have these days, I really do.
Like you say, to compensate for the low sensitivety the modern microphone has to have lots of gain, hence the large number of ICs and so on.
The mixers are stuffed to the gills with cheap IC's, cheap pots, cheap power supplies and poor interfaces.
How can we expect to get any decent and worthwhile music from this hotch potch of junk?
I wish you the best of luck with your project, if you need any microphone transformers designing let me know.
Sincerely,
Peter Qvortrup
nt
Thanks Peter,> > You are mostly right, the message is now so much the hostage of the medium that it is almost impossible to transfer the musical event untarnished from the recording equipment to the software.
But it is a hugely rewarding objective to strive for. I feel
lucky to know and work with a recording engineer who actually has hi end values.> > Microphones with superwide bandwidth, damage the signal, they are very inefficient and therefore super insensitive to low level information, (it takes a fair amount of overall energy to make them start moving).
I don't think many of the modern or old mics had BW much past 20K.
The only ones I know now are some B&K and earthworks, however
these are used infrequently.> > one of the major secrets behind why the great recordings of the 1950's and 1960's were so good, were the "old fashioned" large diaphragm microphones, what would Ella Fitzgerald or Louis Armstrong or Bing Crossby sound like with a modern mike?
A lot of old recordings were actually done on ribbon mics which had very low sensitivity and so SN ratio. However these ribbon mics had a beautiful sound due to their simplicity which is just ribbon element and OP step up transformer, no active electronics at all.
The mic which my friend uses is a modern ribbon mic and uses neodymium magnets to get better sensitivity but still only puts out MC levels. Also note that is with a 1:40 step up transformer after ribbon element. So the element is putting out micro volts. But it sounds absolutely beautiful and natural with incredible imaging.> > Like you say, to compensate for the low sensitivety the modern microphone has to have lots of gain, hence the large number of ICs and so on.
This is not so. Modern mics have better sensitivity and dynamic range than old ones. This includes large diagphram tube mics. There are still many modern mics which emulate old tube designs such as Brauner etc but with lowe noise electronics.
However most the IC's you mention are in the mic pre amps, the worst being resident in multichannel mixing desks. Mixing desks IMO are responsible for a lot of the bad sound we hear. Trying to cram 48> 72 channels of stellar electronics that has totally flexible routing with
totally parametric 4 band EQ (on each channel) and dynamics processing (VCA's) aswell is pretty scary!> > I wish you the best of luck with your project, if you need any microphone transformers
I may email you re this down the track.
Currently best commercial avail is Jensen JT16A/B 1:2
Full performance specs are avail on their website
www.comtran.com/mic_in.html
Regards,
Terry Demol
What is a pleasant distortion?? That sounds like a contradiction in terms.Distortion is any signal or alteration that is present at the output that wasn't present at the input. It is not a subjective entity and as such the effect on the listener isn't accounted for.
Tone controls and 2nd order harmonics exemplifiy the term pleasant distortion Just because the signal is mangled doesn't mean it doesn't sound pleasant to the listener.
Why are none of you guys willing to consider the possibility (I would call it fact, but for the same of argument "possibility" seems more appropriate) that our measurement methods and technologies are woefully inadequate and it is they that need to be revised??I've preach the same thing. There are too many variable and permutation to be able to account for them all and declare superiority alone via measurements. Measurements are good only at giving a gauge of performance with the conditions of the test, not a dynamic situation
Whatever happened to 19th Century observational science??
The problem comes from your spanning the bounds of the two branchs:
One hand you try to attack very sound science that is already correctly applied and as such trying to bring measurements into the game.
On the otherhand you say let the ears be the judge and toss conventional thought aside.
This is an open invitation for strife. By sticking to the subjective argument you at least have an unattackable position.
Scientifically though all I have to do to disprove you is to say: lets see a plot of a resonstructed 22050khz sine wave. If the output from the DAC without energy before and after the sample is correct, you'll have just put your name in the books of science forever. Otherwise the technic is flawed technically. You do feel the objective is recreating the input correctly don't you(and not synthesizing false reality)?
See what I mean.
The first allows for a consensus. The second only allows for an opinion.
Later-CAL
I think this is why Victor bowed out of the conversation. One can't say, (I'm paraphrasing) 'Measurements don't mean very much, nobody's sure what should be measured, it's the sound that matters' and then say 'Victor, tell me why your way is better.' If it's all subjective, what's to discuss? BTW - I've never heard a VK-D5, although I've read many glowing reviews. I do have an AN DAC 1.1 sitting in my living room, connected to absolutely nothing, although I've read many glowing reviews.
No text.
Well, we ARE talking about different things. When saying "digital filter" I mean what most designers would mean - the one before the DAC, the one that does say, 8x upsampling and anti-aliasing. It allows the DAC to run at much higher frequency than the 44.1kHz. Most commonly this is performed by either the PMD100 HDCD decoder/filter, or something like the DF1704 part. So I think you are simply confused here.
.....as it`s very easy for me to become confused in the paralysis of analysis which is audio. The way I (probably mistakenly) understood it was that the Audio Note DACs do not have the filter before the DAC - yes, I knew it was before - but that a very gentle third-order analogue filter (silver-wired inductors) was employed after the DAC. In effect, the raw data is fed to the DAC, then the output is current-to-voltage converted via a transformer. Please feel free to correct me if I`m wrong, as it appears that for all your posturing you are not prepared to engage in a debate with someone like Peter which would benefit the technically illiterate such as myself. Stating that such debates take on a life of their own reminds me of the defensive measures taken by Members of Parliament who wish to appear above such `confrontation`, but who really don`t wish to answer any questions which may challenge, or even disprove the validity of their own conclusions. At the end of your initial reply to Peter, you also stated that such debates could be viewed as part of the `marketing` for the participants - not exactly subtle was it Victor? Surely though, in audio the maxim that "there`s no such thing as bad publicity" is far from true, and marketing is not simply the publicising of a product, but the publicising of the positive aspects of a product? If you could have demonstrated that a product is fatally flawed, then the `marketing` accusation falls flat. You don`t wish to comment on another manufacturers product? Fair point, but when actually invited to, where`s the problem? Sorry to be so blunt, but I`ve had two e-mails about this issue since I got in from work, both of them less than complementary about your evasiveness, one which mirrors my own incredulity that you haven`t even listened to a `filterless` DAC "lately". For any opinion to be worth the attention of two brain-cells, it should be capable of standing up to scrutiny - a condescending attitude fools no-one with aforementioned two brain-cells.
Answer Peter's question.I've busted his balls before concerning the technical merits of a filterless design but I have also has good results simulating a nonoversampled designed through higher rate devices. Technically its not needed since the image is OOB and the effects of the image can be pleasing.
I'll still argue with him about the how a perfect impulse response should look but that doesn't mean the result of a technically inferior design can't actually sound better. Anyone with valve gear knows that.
-CAL
***I'll still argue with him about the how a perfect impulse response should look but that doesn't mean the result of a technically inferior design can't actually sound better. Anyone with valve gear knows that.I simply don't get that last comment. Tubes have very few fundamental limitations, but their alledged inability to properly handle some undefined "perfect impulse response" is not one of them.
Perhaps that statement was simply there as an attention getter.
let me clip out a bit...
that doesn't mean the result of a technically inferior design can't actually sound better. Anyone with valve gear knows thatTubed amp often have measurements that look like they were taken from a dish washing machine but the result of using them can be a brilliant realism acoustically.
-CAL
Dear Cal,Well if that does not prove what I just said in another response that what does.
Why has there been no whole hearted and concerted attempt by the audio industry to find out why??
Surely herein lies the key to longterm satisfaction and thus success??
Sincerely,
Peter Qvortrup
Peter, even though I attack you technically I have no subjective footing to stand on. (I wish I had one of your DACs to try out)It is the purist belief that is the problem. Once we forget that we are trying to satisfy our ears instead of magazine editors, 'net peers and audio salesman, all will be well. (Not to mention the problem of 'net hype but that is something else)
Unfortunately all too much is be bought to impress others rather than ourselves.
-CAL
.....but at the end of the day, you sit down in front of one of their `revolutionary` designs, press `play` and are totally underwhelmed - that`s what it all boils down to. How far have DAC chips `evolved` over the last ten years or so, from 16 bit, then 18, then 20, now 24........every DAC should be a world beater surely? Yet arguably the best `conventional` DACs have reverted to older 16 bit chips, and AN`s use old 18 bit. That`s REAL marketing - two identical DACs on a shelf, but one contains a 20 bit chip, the other a 24 bit chip - which gets sold? We`ve all been subjected to bull**it from an industry which lost it`s way badly, aiming to `refine` the presentation of music until the life was sucked out - too concerned with technical perameters than musical merit. Yes of course measurements ARE important, but they`re the servant, not the master - bearing in mind that one inch of the purest silver-wire will distort an analogue signal and we must accept that distortions are an innescapable fact of any component, but some are more noticable to the human ear than others, and lab equipment can`t really discern which I`d suggest. You appear to be open to this unlike Victor whatsisname; knowledge without wisdom is surely a heinous crime?
I learned long ago that even though I might have an excellent understanding of a particular subject, don't question another mans ingenuity until you get a chance to try what he has.You might be limiting the scope of thought just a little too much to the point that a relatively simplistic solution was there all along.
We all have bad hair and brain days!
-CAL
Dear Victor,So are you implying that a DAC without digital filter cannot be "decent"??
In what way?
Sincerely,
Peter Qvortrup
Dear Peter,As I stated in another post, I would not be proper for me to imply that your products are not decent, and I have no idea whether they do or don't employ digital filters.
Marketplace decides such issues, not arguments in public forums. As far as my objections - I have stated some already.
Best of luck,
Victor.
I would say this depends upon your pre and power amp bcause if you have no filter alot of high frequency noise is going to end up in your pre power if it is not a fast design (or a valve design) this is going to cause intermodulation distortion.
Also bare in mind if you do not use a filter your frequency response will not be flat as it will have a sin x / x function which makes it roll off toward 20K. Any half decent text on sampling will give you the math on this and how to correct for it. However to correct it will require a significant no. of components and if you don't correct it then most of the smoothness of you no oversampling DAC will be comming from the fact that it has the top rolled off.Andy
This post is made possible by the generous support of people like you and our sponsors: