|
Spoondrift posted:I'm playing devil's advocate somewhat here, but I'm pretty sure the whole "digital = no chance for imperfection" idea is patently wrong. A DAC is like any other circuit in that it has no inherent understand of bits. Voltages representing bits go in and an analog signal voltage comes out. Imperfections in those voltages going in will change the output. Bits are zeros and ones. If there isn't voltage, that's a zero. If there is, that's a one. Either there is voltage or there isn't. Not to mention your point doesn't even stand if you're talking about optical. Either there's light or there isn't. As long as the light is getting to the other end unobstructed, it doesn't make a difference if it's through a $5 fiber cable or a $500 one. Plus, if some of the bits weren't making it down the tube, it wouldn't change the tone of the sound, it would create a lot of high pitch static and cutouts.
|
#
?
Nov 19, 2009 19:14
|
|
|
|
| # ? Apr 23, 2024 21:37 |
|
|
here's a gem from the squeezebox forums:quote:I said I do hear a difference between networked SBS or TinySC, at this point I don't know for sure why. It MAY be different data, but I don't think so. Phil's test will prove that one way or the other. edit: oh even better quote:
JITTER 
qirex fucked around with this message at 19:45 on Nov 19, 2009 |
|
#
?
Nov 19, 2009 19:42
|
|
|
timb posted:Bits are zeros and ones. If there isn't voltage, that's a zero. If there is, that's a one. Either there is voltage or there isn't. That's not how digital signals work. You're going to have a voltage that represents zero (let's say 0V) and a voltage that represents one (let's say 1V). Now, because bits are an abstract concept and voltages are only their physical representation, you might end up with some zeros that are 0.1V and some ones that are 0.95V, etc. When you pass 0.95V through a network of resistors (i.e., a simple DAC), you're going to get a different voltage coming out than if you pass exactly 1V through it. This principle applies to optical as well. The flaw in your reasoning is your assumption that the DAC sees 0.9V or 1.1V as "one". This is not the case. The DAC has no concept of bits. Voltages go in, voltages come out. Let me point out that I'm NOT suggesting that a $500 optical cable is a good investment or will necessarily have an impact on the listening experience at all. I'm just showing how digital audio is not magically perfect.
|
|
#
?
Nov 19, 2009 20:00
|
|
|
Spoondrift posted:The flaw in your reasoning is your assumption that the DAC sees 0.9V or 1.1V as "one". This is not the case. The DAC has no concept of bits. Voltages go in, voltages come out.
|
|
#
?
Nov 19, 2009 20:14
|
|
|
qirex posted:If digital signal transmission worked that way we wouldn't have hard drives or the internet. There is a wide range of voltages that the DAC will accept as "1", probably anything between 0.5v and enough to damage the equipment. The DAC has no notion of "1". The DAC has no notion of "0". The DAC is an analog circuit that turns voltages into other voltages. Computer data does't work the same way because the voltages are only ever interpreted as bits. The variations in voltage aren't being propagated to an analog signal (computer audio, etc., is an obvious exception, but this is the general case). As an aside, computer networks and storage use error correction methods because bits DO sometimes get flipped.
|
|
#
?
Nov 19, 2009 20:27
|
|
|
Spoondrift posted:The DAC has no notion of "1". The DAC has no notion of "0". The DAC is an analog circuit that turns voltages into other voltages. The DAC will output the same no matter if the input is .95 or 1.05, it only records at what range the signal came in. The input voltages do not matter as long as the sample comes in with the same bits, it will then get ouput as the same analog signal as any other digital signal with the same sample regardless of the individual voltages of the bits.
|
|
#
?
Nov 19, 2009 20:37
|
|
|
Bensa posted:The DAC will output the same no matter if the input is .95 or 1.05, it only records at what range the signal came in. The input voltages do not matter as long as the sample comes in with the same bits, it will then get ouput as the same analog signal as any other digital signal with the same sample regardless of the individual voltages of the bits. Can you cite this or show an example of a circuit with this property?
|
|
#
?
Nov 19, 2009 20:42
|
|
|
Spoondrift posted:The DAC has no notion of "1". The DAC has no notion of "0". The DAC is an analog circuit that turns voltages into other voltages. What? Of course the DAC has a notion of 1 and 0... It's a digital to analog converter. S/PDIF transmits 20-bit data via biphase mark code. When using coaxial, 0.5-1v represents a 1, anything below that is a 0.
|
|
#
?
Nov 19, 2009 21:00
|
|
|
Spoondrift posted:Can you cite this or show an example of a circuit with this property? I'm with Spoondrift in this particular bit. No one here is seriously arguing that the data path from the media to the DAC can somehow alter the digital representation of the audio unless a piece of equipment is fundamentally broken. Different DACs will likely have some (almost entirely imperceptible) quirks on the analog side, for that is the nature of analog audio. The insanity from audiophiles comes from people trying to improve the digital source material going into the DAC (outright snake oil), and improving the analog output way past the point of diminishing returns. I have no doubt that a $100,000 amplifier will sound better than a discman, all else being equal. But I'd rather spend that money touring the world and going to live music venues where I can complain about the jitter coming from the trombone.
|
|
#
?
Nov 19, 2009 21:04
|
|
|
^^^ I'm actually questioning why improving the digital input to the DAC is outright snake oil, at least in principle. Re-edit: By "improving the digital input" I mean reducing variation in the signal, not correcting bad bits which I don't consider to be an issue.timb posted:What? Of course the DAC has a notion of 1 and 0... It's a digital to analog converter. A resistor ladder is a digital-to-analog converter that clearly has no notion of bits and clearly produces a different output voltage depending on variations in the input voltages. I'm totally open to opposing arguments here but all I'm getting is "...but it's DIGITAL!" Spoondrift fucked around with this message at 21:15 on Nov 19, 2009 |
|
#
?
Nov 19, 2009 21:08
|
|
|
Spoondrift posted:A resistor ladder is a digital-to-analog converter that clearly has no notion of bits and clearly produces a different output voltage depending on variations in the input voltages. But the DACs used in the amplifiers we're talking about aren't resistor ladders...
|
|
#
?
Nov 19, 2009 21:20
|
|
|
timb posted:But the DACs used in the amplifiers we're talking about aren't resistor ladders... Of course not, but your last post implied that DACs have an understanding of zeros and ones by virtue of being DACs. Can you give any explanation of how a more advanced DAC "understands" what bits are?
|
|
#
?
Nov 19, 2009 21:25
|
|
|
Spoondrift posted:Can you cite this or show an example of a circuit with this property? timb posted:S/PDIF transmits 20-bit data via biphase mark code. When using coaxial, 0.5-1v represents a 1, anything below that is a 0. Lets say you got a sample consisting of 01, you can have that be sent digitally as a 0.3V and 0.7V, or 0.2V and 0.8V, both will have the same output in the same DAC. The input is only used for reading the bit, it has no other effect on the output (on a properly designed DAC). Therefore as long as the sample is the same the input voltage is irrelevant to the output. Spoondrift posted:Can you give any explanation of how a more advanced DAC "understands" what bits are? They're inherently designed to either read the input as a 0 or 1 based on voltage ranges and then transmit to the ouput generator only as a guide, not as the actual signal to output. They have no need to "understand", just like how a certain voltage will punch through a dielectric a certain voltage will trigger a signal from the input stage to the output stage. Bensa fucked around with this message at 21:33 on Nov 19, 2009 |
|
#
?
Nov 19, 2009 21:30
|
|
|
Before we get too much like the thing we're trying to mock, a few questions: Can we all agree that there are varying ways to build a DAC, and that there are DACs of varying quality in the marketplace? Can we all agree that trying to scrub the digital input stream coming to a DAC, in the ways cited above, is idiotic? Can we all agree that placement of the DAC in a digital audio stream (as in, on-board with your sound card, or offboard at your amp) does make a difference? And finally, can we all agree that the method of transmitting digitally encoded audio to an offboard DAC, be it coax, hdmi, or optical, is unlikely to be affected by quality of the cable as long as the cable meets minimum acceptable specification?
|
|
#
?
Nov 19, 2009 21:33
|
|
|
fahrvergnugen posted:Can we all agree that placement of the DAC in a digital audio stream (as in, on-board with your sound card, or offboard at your amp) does make a difference? Its not really the placement of the DAC that matters since the signal to it should be unaffected, the portion that matters is the path that the analog signal from it takes since its immensely more susceptible to interference.
|
|
#
?
Nov 19, 2009 21:37
|
|
|
Bensa posted:Lets say you got a sample consisting of 01, you can have that be sent digitally as a 0.3V and 0.7V, or 0.2V and 0.8V, both will have the same output in the same DAC. The input is only used for reading the bit, it has no other effect on the output (on a properly designed DAC). Therefore as long as the sample is the same the input voltage is irrelevant to the output. I'm satisfied with this explanation (I suspect that the details surpass my knowledge of electronics). Thanks.
|
|
#
?
Nov 19, 2009 21:40
|
|
|
Bensa posted:Its not really the placement of the DAC that matters since the signal to it should be unaffected, the portion that matters is the path that the analog signal from it takes since its immensely more susceptible to interference. Yes, that's pretty much what I meant. Thanks.
|
|
#
?
Nov 19, 2009 21:50
|
|
|
Spoondrift posted:I'm satisfied with this explanation (I suspect that the details surpass my knowledge of electronics). Thanks. No worries - at least you're willing to listen. People with knowledge in electrical engineering tend to bang our heads against the walls when we visit an audiophile forum.
|
|
#
?
Nov 20, 2009 09:39
|
|
|
Just so we're clear, this is representative of an SPDIF signal. Oh no, it's not a perfect square, will I hear distortion?? All electrical signals, including digital data, are carried by and analog waveform. It will never be a perfect square wave, and often doesn't even need to be close to square, or close to 0V or 5V. A data sheet for a DAC will have an input specification that reads somethng like this code:(there are a few other considerations, but that should drive the point home that small fluctuations of the "digital signal" voltage will have no affect on the output of a DAC.)
|
|
#
?
Nov 20, 2009 16:52
|
|
|
proudfoot posted:No worries - at least you're willing to listen. People with knowledge in electrical engineering tend to bang our heads against the walls when we visit an audiophile forum. Audiophiles are really great at completely missing the point of any argument and hanging onto any scrap of the argument that seems to contradict what you said, especially if it is the most superficial contradiction possible. For example, I was discussing how the data written onto a CD couldn't possibly affect the bass and the treble, since it would either be the signal from the digital master or it would be some form of nonsense if there was an error. The audiophile then talked about jitter and read errors and other poo poo. I respond by asking "how do we use CDs for data storage if they are so prone to errors" and then going into a deep explanation of Reed-Solomon error correction and bit-parity and how it would mask errors, and since unless it was a particularly bad error, you wouldn't even notice it, and if you did, it would be a total drop out or reduction to mono, or at the catastrophic level, the player stopping and stopping, and skipping ahead randomly. His reply, "Well, Red Book Audio doesn't require a perfect audio stream." I decided to give up there. He took my initial question, ignored everything I had to say about how accurate CD lasers really are, and just went ahead and repeat some fallacy that ignores all the facts. Arguing with an audiophile is like trying to explain to an atom what the color green looks like through a red lens.
|
|
#
?
Nov 21, 2009 03:49
|
|
|
Spoondrift posted:I'm satisfied with this explanation...except that one time I put different things into my DAC and different sounds came out. I shoved a cocount into the input because it's round like a zero, I should hear low frequency like bass, but instead I heard this rather purple bit of high frequency audio. Then I used the banana and got a different, though still purple sound. SEE DACS DON"T UNDERSTAND BITS1101 RAWR11001011 fixed, although I would like to point out that DACs are in fact designed to decode bitstreams, which have defined tolerances.
|
|
#
?
Nov 21, 2009 19:37
|
|
|
davepsilon posted:fixed, although I would like to point out that DACs are in fact designed to decode bitstreams, which have defined tolerances. Huh, I guess because I don't have an electrical engineer's knowledge of how digital audio works but am not intellectually incurious enough to be satisfied with an explanation of "IT'S DIGITAL!!!" I'm a crazy audiophile.
|
|
#
?
Nov 22, 2009 21:48
|
|
|
Spoondrift posted:Huh, I guess because I don't have an electrical engineer's knowledge of how digital audio works but am not intellectually incurious enough to be satisfied with an explanation of "IT'S DIGITAL!!!" I'm a crazy audiophile. You pretty much are, being you were claiming that a digital signal would be substantially changed by minor interference to the analog signal conveying it.
|
|
#
?
Nov 22, 2009 22:07
|
|
|
fishmech posted:You pretty much are, being you were claiming that a digital signal would be substantially changed by minor interference to the analog signal conveying it. No, I didn't. I claimed that variations in the digital signal could be propagated to the analog signal after coversion, and this is indeed the case for something like a http://en.wikipedia.org/wiki/Resistor_ladder DAC, as we've already seen. Now, since you didn't seem to comprehend my posts at all, I'll lay this out very carefully for you: based on my knowledge of electronics, I can see that it's possible to design and implement a DAC that is not tolerant to variations in the digital input signal. However, since my knowledge only goes so far (i.e., I'm not an electrical engineer), I'm not able to pull a circuit out of the ether that takes an input voltage from a certain range and outputs the "correct" voltage that it's supposed to be. What I wanted was for someone to explain how such a circuit works, not because I doubt the existence of such a circuit, but because I want more than a superficial understanding. What I don't want is some poo poo about purple sound that I never said nor came close to saying. Are you getting this now, buddy? The great irony of this thread is that many of the posters seem to base their criticism of audiophiles on as little hard knowledge as on which audiophiles base their purchases of snake oil products.
|
|
#
?
Nov 22, 2009 22:25
|
|
|
Spoondrift posted:No, I didn't. I claimed that variations in the digital signal could be propagated to the analog signal after coversion, and this is indeed the case for something like a http://en.wikipedia.org/wiki/Resistor_ladder DAC, as we've already seen. Then you don't know what you were talking about because you brought it up for when digital signals are transferred between two devices over an analog carrier. And by the way if a digital signal doesn't vary it's carrying either all 1s or all 0s.
|
|
#
?
Nov 22, 2009 22:42
|
|
|
fishmech posted:Then you don't know what you were talking about because you brought it up for when digital signals are transferred between two devices over an analog carrier. And by the way if a digital signal doesn't vary it's carrying either all 1s or all 0s. Dude, really? Do I have to spell out what I mean by "variations in a digital signal" so you don't egregiously misinterpret me? Protip: Electronic circuits are no longer abstract concepts once you actually build them. They don't work perfectly as they do on paper. That's pretty much the root of my entire curiosity.
|
|
#
?
Nov 22, 2009 23:18
|
|
|
Spoondrift posted:Dude, really? Do I have to spell out what I mean by "variations in a digital signal" so you don't egregiously misinterpret me? And any device you buy that's legitimate is designed to handle minor variations, and (almost) all digital signaling systems involve some form of error-correction. It is very difficult for a digital signal to spontaneously become a valid, different, signal, instead you just have drop outs in video/audio or somewhat degraded quality for a few milliseconds. Furthermore it'd be extremely difficult to build a DAC that only accepted a precise voltage for 0 and a precise one for 1 - most of the reason we have tolerances in DACs is because adapting them to only accept say 0 v and 1 v ONLY would require a shitload more components.
|
|
#
?
Nov 22, 2009 23:41
|
|
|
fishmech posted:And any device you buy that's legitimate is designed to handle minor variations, and (almost) all digital signaling systems involve some form of error-correction. It is very difficult for a digital signal to spontaneously become a valid, different, signal, instead you just have drop outs in video/audio or somewhat degraded quality for a few milliseconds. None of my posts contradict any of this. fishmech posted:Furthermore it'd be extremely difficult to build a DAC that only accepted a precise voltage for 0 and a precise one for 1 - most of the reason we have tolerances in DACs is because adapting them to only accept say 0 v and 1 v ONLY would require a shitload more components. I never even came close to suggesting that a DAC would only accept a perfect voltage for a 0 or 1. Seriously, did you even read my posts? The only thing I was trying to learn was how DACs are able to read the digital signal perfectly despite the imperfections that are inevitable in, you know, physical reality. I never disputed that DACs were able to do this; I only wanted to know how they did it instead of blindly accepting it. How are you not getting this?
|
|
#
?
Nov 23, 2009 00:30
|
|
|
Hey, you know what's awesome? Whiny bitch-fests about minutiae! Wait, that's a lie, those are awful. Go back to laughing at stupid people.
|
|
#
?
Nov 23, 2009 01:02
|
|
|
Hey he's not being hard-headed, he's just asking questions The Fox News troll.
|
|
#
?
Nov 23, 2009 02:03
|
|
|
Digital signals are fair and balanced.
|
|
#
?
Nov 23, 2009 02:10
|
|
|
blugu64 posted:Digital signals are fair and balanced. Let's not get into balanced signals, please.
|
|
#
?
Nov 23, 2009 02:17
|
|
|
eddiewalker posted:Let's not get into balanced signals, please. I actually have a solution. So long as your receiver is perfectly balanced (you can buy one of my triblesonic vibration balancing levels) then the audio is balanced. If you still have issues you shouldn't worry as I'll throw in a free bag of rocks to tie to each speaker cable.
|
|
#
?
Nov 23, 2009 02:47
|
|
|
Laughing at stupid people: People reviewing the $195 CD Mat From the site posted:
quote:This thing actually does seem to make a difference - just did some blind tests with Lori as the listener - we agree that it makes a difference, and we're both consistent in our preferences. Opposite, though - she prefers the sound without the mat, and I prefer it with. Will try it for a few more days before sending it on. I really, really did not want to believe this could matter ... quote:I would describe the sound as a bit darker, with deep bass a bit more obvious, and more depth/spatiality to the sound. Not a huge difference, but surprising and worthwhile. quote:Thanks again Ken. quote:I liked it well enough that I bought the new "Signature 3D" version. I think I'm in love ... seriously, a very worthwhile improvement, though I can't say I'm too impressed with Marigo's own explanation for how and why it works. I wish I understood how it can make a difference at all, but I've gotten over feeling silly about buying the thing and am just enjoying what it does for the sound in my system. I will say, though, that the degree of improvement is highly disc-dependent. More like you realize in your gut you wasted $200 and are trying to justify it to yourself. quote:On my main system (Adcom GFA 545, GFP 555 II, Vandersteen 2C's, and Marantz CD 5001), I was only able to try the soft rubber mat. The rigid Marigo mat would not allow the drawer to close. There was a perceptible difference in sound quality with the rubber mat on all CD's, most significantly with the bass. It seemed more defined and a bit tighter. Where I found the most improvement was with burned CD's, the improvement was tremendous . To be fair to AK, most posts in the thread are skeptical, but you can see the stupid audiophile gears in their head working overtime to justify getting one.
|
|
#
?
Nov 23, 2009 03:22
|
|
|
I wonder how many audiophiles are going to be chomping at the bit after these statements? http://news.bbc.co.uk/1/hi/scotland/8368895.stm quote:The firm, which makes systems costing from £2,500 to more than £100,000, said discerning customers recognised the superior quality of digital streaming.
|
|
#
?
Nov 23, 2009 15:26
|
|
|
Neurophonic posted:I wonder how many audiophiles are going to be chomping at the bit after these statements? logictech is already cornering that market with it's $2000 audiophile Squeezebox quote:High-accuracy digital sound Jitter. For streaming audio.
|
|
#
?
Nov 23, 2009 16:05
|
|
|
HKR posted:logictech is already cornering that market with it's $2000 audiophile Squeezebox The Transporter is hilarious because you can tell they don't really "buy it" but when they figured out they could slap an extra screen and $200 more in "audiophile grade" components on top of a $300 squeezebox and sell it for $2k they figured why the hell not? There were people practically demanding it on their forums and they could get a jump on the rest of the audiophile market.
|
|
#
?
Nov 23, 2009 17:21
|
|
|
I also find it funny how audiophiles are more then willing to take apart poo poo to fix any perceived problems but seem to be incapable of hooking a good computer+soundcard to their stereo.
|
|
#
?
Nov 24, 2009 01:26
|
|
|
quote:Marigo Audio has further improved the Signature 3-D v2 CD mat by making the gold layer rough-textured instead of slick! That's it? Where's the explanation of why rough textured gold makes this coaster improve the audio from a CD?
|
|
#
?
Nov 24, 2009 02:14
|
|
|
|
| # ? Apr 23, 2024 21:37 |
|
|
qirex posted:The Transporter is hilarious because you can tell they don't really "buy it" but when they figured out they could slap an extra screen and $200 more in "audiophile grade" components on top of a $300 squeezebox and sell it for $2k they figured why the hell not? There were people practically demanding it on their forums and they could get a jump on the rest of the audiophile market. What will the people who sell audiophile grade cables say to wireless streaming! (audiophile grade wireless cards and antennas)
|
|
#
?
Nov 24, 2009 02:29
|
|