Slashdot Mirror
Cheap to Audiophile with Simple Hacks
petertrog writes "The IEEE has a story showing how you can turn a cheap DVD player into something that sounds a whole lot more exotic. All you need is a small budget, a soldering iron and a desire to void your warranty."
Wow, an audiophile article from the IEEE. Next thing you know, we'll have witch doctors contributing to the Journal of the American Medical Association.
Strange things are afoot at the Circle-K.
I dunno, my time is too valuable to bother doing the upgrade myself. Better just to buy the high-end at 10x the price and save 100x in the cost of my time.
Ruby Neural Evolution of Augmenting Topologies
That is the first real 'Audiophile' tweaks article I have ever seen. It actually detailed real changes you can make to improve the sound of your equipment.
The only reason people purchase expensive interconnects etc is because those components are very easy to change. NOT because they have a significant effect on the fidelity of reproduction.
To really improve the sound you have to improve power supply, decoupling caps etc, but even though the components are very cheap, it's a lot harder than buying a $500 interconnect cable.
I hope to see more articles like this in the future!
This article is basically an advert for LC Audio (whose own stuff doesn't look anything special - look at the ringing on the scope trace of their wunderkind clock oscillator), mixed in with the usual audiophile crap (where's the blind A/B comparison?) with a healthy dose of stupidity; anyone advocating replacing safety-rated components on the mains side with unrated "audiophile" grade parts deserves to be sued by the first idiot who burns his house to the ground. The mains is a hostile environment, those components are designed to fail open-circuit for a REASON!
Jon.
The real reasons for recording with greater resolution (bit depths and sample rate) than your target media are :
To allow more headroom while recording.
To prevent generation loss while editing and mixing.
To enable releases on newer media than CD. (Or just upsample the 44.1, no one complains anyway.)
Sure, 24/96 does sound a little better than 16/44.1 on a solo'd vocal, but once you have your final master, all compressed and eq'd up, it makes very little difference. There is only around 6db dynamic range on modern releases, and the majority of playback equipment does very little above 18k anyway, so even 16/44 is overkill for domestic systems.
You know... I really fear this isn't a joke, is it?
$1500 power cord (The JBS review) the guy actually implies it is justified!
Does he not know what crap is behind the walls you plug into? How can you think that just the last 10 foot of a power cord is going to make any sigificant difference given the other kilometers of wire involved?
My $35 DVD player has a digital (coaxial) output, and my PS2 has an digital (optical) output (but, the laser is blown and it can't play disks with even the smallest scratches). Why mess with the electronics inside when you can get the audio data right off the disk into your system?
autopr0n is like, down and stuff.
You underestimate the power of suggestion (and ignorance).
HAND.
Well, maybe if your dog is an audiophile, but as a human I'm perfectly satisfied with equipment that reproduces sound in the frequency range I can hear.
English is easier said than done.
... "vibration-induced clock jitter"? Get to fuck...
A signal which comes out cleaner on the scope, up to a certain point, will also sound better to the human hear, but past that point, it just comes down to preference. This is why studio engineers often add "color" to a song, and why some audiophiles still swear by vaccum tubes. The vaccum tubes don't produce anywhere near a flat frequency response through the 20kHz range, but instead color it in a way that people describe as "warm."
The point is, you can try and make changes to flatten the frequency response as much as possible, but it may NOT be the sound output you're looking for. The scope would, of course, be useful to track down problems with power supply noise, but when it comes down to swapping op-amps or other stuff, it's often times more subjective than not, which is what his article says. Here, seeing the scope output is useless, because the only important this is whether you like the resulting sound output.
I'd like to agree with you on the part about the clock though, but I have never looked clock outputs when they get shaken/etc, so can't really comment.
The author provides no objective evidence of improvement. Instead, we get: "The tone had been slightly light(?). Modification increased the body(?) of the tone--for example, a guitar sound that previously was all string now includes the wood of the instrument. The stock unit had a bit of congestion(?) on dynamic passages, especially evident on massed strings. Not anymore; the top and bottom ends are detailed(?), extended, and inviting(?). The soundstage, that is, the virtual placements of the instruments that you hear in front of you, was originally very good--definitely not an in-your-face kind of sound that you would normally expect from a cheap player. Nevertheless, modification added an ease(?) and presence(?) to the sound; a liquidity(?) that was not there before."
I added "(?)" after terms which, sorry for my ignorance, would have no idea how to measure objectively. (I was tempted to put one after "soundstage", but I guess it could mean stereo separation, so I'll give him the benefit of the doubt. And I suppose "wood" is bass but not sure.)
It is possible there was an improvement. But from this kind of babble I can't tell. It is very possible he wants it to be better and thus perceives it to be so.
An improvement in audio or video, if any, can be measured objectively with appropriate instrumentation. If the author had done this, he would have determined which, if any, of the capacitors he's boosting have a measurable effect, saving a lot of work. Instead he seems to be on this mantra of blindly replacing them all in hopes of an improvement.
At a very minimum, without instrumentation, there should be a blind test comparing a modified and unmodified unit by a third party.
And just what are the author's qualifications? "Robert McNeice is a business and information-technology consultant for the financial services industry. He is an audiophile and occasional tweaker." I suppose he could be an EE, but if so he needs to go back to school to learn how science works.
This reeks of the kind of subjective nonsense you see with the high-end audio bs with its $600 cables. Shame on the IEEE (an otherwise respectible organization for electrical engineers) for publishing this crap.
Wrong. The clock is in the player, transmitted over S/PDIF, and recovered by the DAC. That transmission and recovery step is fraught with peril, and it is guaranteed by the design of the S/PDIF signalling to have data-correlated time domain distortion. The only way to avoid this problem is to put the master clock in the DAC (as you say) and slave the player's clock to it, so that the two run in synchronized clock domains. But only DIYers and professionals have such equipment. To the best of my knowledge no manufacturer has marketed a consumer DVD player with a clock-sync input.
Audiophiles believe in paranormal phenomena that cannot be verified by any scientific process.
So this guy's player sounded better after he replaced the caps, resistors, power supply? He could tell by listening to it? With how much - a day - interval between the two auditions?
One of two things happened - he made no discernible difference and only imagines he improved the equipment. Or he made it much, much worse, and likes the distortions he introduced.
Correctly functioning players - even the cheapest - have such low noise, low distortion, and flat frequency response within the human audio spectrum that nobody has yet been able to demonstrate - in double-blind level-matched synchronized A/B comparisons - that they can tell the difference.
If you want to improve your stereo performance, concentrate on the "I/O" devices - speakers, monitors, and microphones. They introduce many orders of magnitude more color than the electronic components.
That's only true if you're using the analogue output, think the original poster is talking about the digital output.
At any rate why would you spend so much time and effort improving your DVD player if you have it hooked up to a crap amp? It's the equivalent of trying to turn your economy car into a hot rod by simply installing a sports exhaust and air filter (and maybe a huge spoiler and some stripes).
1) For something as subjective as viewing paintings and sculpture, I find it interesting that everyone here is trying to bash people for trying to increase the quality of sound in their stereo. One reason may just be a generational problem where modifications were much more apparent in equipment 30-50 years ago. 2) I also find it funny that everyone is bashing the audiophiles when only a few stories below is about a guy who memorized >80k digits of PI (far less useful than modifiying stereo gear). Just so you know how relevant 80k digits of pi is: "If one were to find the circumference of a circle the size of the known universe, requiring that the circumference be accurate to within the radius of one proton only 39 decimal places of Pi would be necessary."
http://www.brentcastle.com
Higher frequency response isn't the reason to go to 192kHz.
If you think you can really hear past ~22kHz, and if your speakers went there, and if your cable could pass that frequency, well, so be it, but you're hearing what isn't intended to be heard, dog.
The reason to use higher sampling rates is to obtain more accuracy in that critical 20Hz to 20kHz range.
Consider how many samples a 10kHz sine wave receives, if the sample rate is 40kHz. Yup, 4 samples. So you have four digital 'dots' which get connected together to form the same waveform on the analogue side. Not going to happen.
PCM audio is inherently flawed in that as frequency increases, resolution decreases.
In general, sounds in that higher-frequency register require more accuracy anyway.
You're right of course, it all gets thrown on a 44.1/16 CD, and then the kids put lossy compression on it, and the radio station compresses the mix to the last 4dB.
Consider this though, if we ditched PCM for something better, would anyone really sell any more CD's?
when the warranty expires for a gadget, an average joe would be like, "shoot, i need to be more careful now". when the warranty expires for a gadget, a geek would be like, "sweet! i get to take it apart now!"
HD Trailers
audiophile, n: Someone who listens to the equipment instead of the music.
while (!asleep()) sheep++
Reminds me of a different story about those fancy expensive wine glasses that supposedly improve wine tasting experience.
They are still beeing hyped by some most prominent wine critics. In all professional reviews there was a clear improvement of wine score when tasted from those glasses.
The problem was - after those experiments were properly repeated as a double blind study, any difference completely dissappeared.
The lesson was - hype does affect your taste.
Actually, I am not saying it is bad. If they enjoy the illusion - that's just fine.
<^>_<(ô ô)>_<^>
Well then, in the interest of forcing audiophiles to make up some new words, let's come up with some concrete definitions for these. To start us off, here's how I would read some of these terms. I doubt any self-respecting audiophile would agree with me :-)
:-)
soundstage is, as you suggested, stereo separation. I might also include the presence of very high-frequency response in it, since there's some evidence frequencies we don't perceive conciously still affect our ability to place the source.
extended is easy, lots of frequency range on both ends.
detailed, I would read as the ability to produce quick, quiet broadband effects well, even when playing a much louder main harmonic (think guitar frets). This basically boils down to the ability to put a small step/square-wave into a larger-magnitude waveform.
congestion makes me think of the sort of pre-echo sound of a piece that's been transformed into freq domain using an FFT, then resynthesized. This causes frequencies found in a short burst to get spread out in time, over the full length of the FFT window that detected them. This is a very characteristic artifact of digital compression, particularly MP3, but not something a player should normally struggle with.
liquid... I can't think of something for this w.r.t layback, but in a piece of music I might use it to describe a section played with a slightly irregular tempo (just a little bit rubato). If my playback gear is producing this effect, I doubt I will like it
The Matrix is going down for reboot now! Stopping reality: OK. The system is halted.
Here, seeing the scope output is useless, because the only important this is whether you like the resulting sound output.
The scope isn't just to ensure a flat response. It can also tell you if changing one capacitor, op-amp, etc. for another one had any effect at all, or if the difference is psychological.
"...always new atoms but always doing the same dance, remembering what the dance was yesterday." -Richard Feynman
- Making the power supply filter capacitors "20% bigger" is a silly idea, for MANY reasons:
- Most electrolytics come with a -20% to +100% tolerance. Because of their construction, it's hard for the manufacturers to get them much closer than that.
- Plus in any well-designed power supply the capacitors are intentionally chosen a bit oversize to handle 50Hz or low line voltage situations.
- Electrolytics have a steep cap versus temperature curve. The engineers know this and specify 40% bigger caps to handle the times you use your CD player in Alaska.
- The filter caps are isolated from the audio circuits by a voltahge regulator chip, which provides about 60 to 90 db of isolation. There's just NO WAY one can notice the effects of a 20% change in capacitance, when the effects are mulffled by a factor of a million to a billion.
- The original filter caps have to be very specially chosen for compatibility with the high frequencies and ripple currents. Is it likely the average joe tweaker is going to choose something that approaches what the actual power supply designer chose? Not likely.
- Replacing the power supply diodes with "faster" ones is a waste of time and money. Any noise the old diodes generate (if any) is many decades above thre audio range. Plus the CD player has to pass FCC emission limits, so they can't be too noisy to begin with. Skip this mod.
- Changing op-amps is really ridiculous. Op amps are always used with huge amounts of negative feedback, which reduces their individual quirks and distortion by a huge factor. I've worked with dozens of op-amps, and have never found one that's not capable of handling your typical audio. A typical 30 cent op-amp already has about 0.001% distortion, thousands of times lower than a golden-eared indivuidual can discern. Skip this step too.
- Tapping into the DAC outputs is a REALLY bad idea. Apparently this guy hasnt a clue about Nyquist limits and sampling rates. You HAVE to filter the output of the DAC's, as they're intrinsically rife with sampling-rate related harmonics and aliasing. Those op-amps are there for a reason!. Don't even think of doing this.
- Putting caulking on the crystal is wet-your-pants funny! There's absolutely no need for this. Crystals are designed to resonate at one frequency. They're totally insensitive, by factos of a billion or more, to any other vibrationary frequency. As an example, there are very precise aerospace radios, with dozens of crystals, none of them caulk-damped, used for life-critical navigation and landing systems, and they work just fine for decades of constant use in vibraty, shaky old prop planes. Put the rope caulk around your windows, not on your crystals.
- If you like the look of gold-plated jacks, install them. There will be absolutely no discernible difference in the sound, but they look neater.
Sorry to rain onthis guys parade, but IMHO there should be at least a token nod towards reality."I'll fill in from a buffer while you're looking" is correct. "Usually this is done by repeating briefly the last good data" is complete crap. The player either plays the data from the buffer or it plays nothing. Shock buffering the servo does nothing in this case, where we have a CD player sitting on a shelf.
Not everybody knows how to recognize this in what they hear, as it is very short. Typical crazy audiophile bullshit.
-ZOD-