Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jun 9, 2012 23:51:44 GMT
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jun 10, 2012 9:49:57 GMT
I certainly agree with Steve Nugent on this comment (about IP): The only clock that is important is the one in the end-point device. Examples of end-point devices are: Squeezebox, Duet and Sonos. This would seem to be the ideal situation, which it certainly is.This means the solution to jitter problems in reading and transport (cables) are essentially solved/nullified by a DAC that converts USB/SPDIF/I2S to (Gb) ethernet (there are simple modules for this) and simply convert it back to I2S which is then directly fed into a DAC. Problems solved and none of all the other jitter before it matters anymore except the clock and PS of the DAC/clock circuit. Also the jitter that is introduced in the recording stage and cannot be removed as it is stored in the signal. So a device (DAC) that converts USB into packages (at a much higher speed than the data speed) and reclockes to it's own internal stable clock. hmm... isn't something very similar (storing data and reclocking, not IP) happening in almost all newer DAC's that are inherantly low on jitter ? Sabre and Wolfson come to mind as cheap solutions (when having a proper powersupply scheme). otherwise he 'believes' a lot to be the case (he actually uses the term 'I believe' a lot and offers no proof), depending on subjective found differences (not blind in most cases ?), and creates sales arguements for his devices. Just my uneducated opinion of course and not disputing Steve's technical knowledge.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jun 10, 2012 10:20:47 GMT
Hi Frans I posted it for info only. I certainly do NOT agree about earlier "Jitter"or whatever, as in the comparison .wav files that I UL being corrected later on. I am starting to not give a rat's anus whether others believe me or not (It's their loss.)when I say that the damage is IRREVERSIBLE with those previous wav files. The best you can hope for is to make everything sound a little better, but with the gap remaining, no matter how good the DAC is. To that end, this afternoon I fitted a "Cricket": as in the SC DAC, inline with the .3PPM TCXO in my PCs's X-DAC V3. Afterwards,I even felt the need to listen to Hotel California in 24/192 right through a couple of times. Very good stuff sounds even better, but the gap is still there with earlier rips, AND TO HELL WITH THE CHECKSUMS !!! Kind Regards Alex
P.S. I have been just listening again to "Norah Jones -Come Away With Me" in 24/192. WOW ! It seems that good high res material really comes into it's own with very close tolerance reference TCXOs fed by VERY low noise , highly stable power.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jun 10, 2012 17:19:07 GMT
If, in the days the holy books were written, jitter would have been known to exist in all its forms they surely would have called the devil 'jitter' instead of 'satan' . It corrupts all that is GOOD.. and is the root of all evil in digital audio, especially for believers ;D
|
|
|
Post by dalethorn on Jun 10, 2012 20:10:01 GMT
If, in the days the holy books were written, jitter would have been known to exist in all its forms they surely would have called the devil 'jitter' instead of 'satan' . It corrupts all that is GOOD.. and is the root of all evil in digital audio, especially for believers ;D There's an analogy in photography called Gaussian Blur, which takes the edge off of overly sharp images. If we can apply something like that to individual WAV files permanently then we would not need the low quality DAC for edgy music files. On a related note, I never use checksums. I use my own code (dalethorn com) which is 100 percent reliable. EDIT: The very reason I started writing utility code (1988 to date) is because of bugs in the most common checksum etc. utilities.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jun 10, 2012 21:11:19 GMT
The root of all evil is closed minded EEs with ageing/poor hearing. ;D
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jun 10, 2012 21:58:37 GMT
That must be my evil twin you are referring to ! ;D
|
|
|
Post by dalethorn on Jun 10, 2012 22:18:32 GMT
The root of all evil is closed minded EEs with ageing/poor hearing. ;D The only closed minded person I know is one who argues against pure logic itself - i.e. "if something is identical, it's not identical because I say so."
|
|
|
Post by dalethorn on Jun 10, 2012 22:21:50 GMT
BTW, with the ATH M50 I can clearly hear 16 khz, just like the Beyer DT 48. I don't hear that tone clearly on the Shure 1840 or Senn HD800, even with the best gear. How about that! HF rolloff on the $1500 headphone.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jun 10, 2012 22:45:20 GMT
Dale Then clearly, your "pure" logic is not revealing the whole picture. You are so fixated about pure binary data that you can't accept the possibility of timing errors or noise along for the ride with the data. You need to start looking OUTSIDE the square. There are far more members here that have identified clear differences between .wav files that you insist are identical, than those that haven't , AND selected the same ones as sounding better,than those that have been unable to hear the differences. This was also verified recently by Martin Colloms in HFC, and an additional 6 blind A/B/A/ 3 minute listening sessions with different people.There is only one person who has been unable , out of around a dozen people who have listened to these files directly from my P.C. who was unable to hear the differences, and select the same file as sounding better, DESPITE the check sums being identical.You could also ask the guys who have listened directly to the Corsair Voyager now being passed around Europe whether those files sound the same or better than those they had previously downloaded, which also had identical checksums. Alex
P.S. As Ian suggested earlier, I will soon tidy up this thread, as most of the recent replies, including some from Chong , have no relevance to the title of this thread.
|
|
|
Post by dalethorn on Jun 10, 2012 23:23:14 GMT
As I told you so many times Alex, I don't dispute a difference where you hear a difference. Now as to whether that difference is in the file *contents* on your system, it would not be in the measured file contents. But if the file contents were distributed in a funky way on your system or played from different drives, they could sound different. Even some innocuous difference such as a filename could *possibly* make a difference depending on the secrets behind the music player code.
Where the dispute lies is simple. When I copy *measurably* identical files to my system, *all* of those variables are eliminated, because: 1) They are on the same drive in the same folder. 2) The filenames are inconsequential, i.e. x1.wav and x2.wav. 3) My computer did not *receive* a physical thing when copying from the Web, i.e. it received an electric signal and *created* the bytes indicated in that signal one at a time, *exactly* the same as if I wrote those bytes myself, *exactly* the same as if I had written "abcde" twice on a single sheet of paper. 4) Until you are willing to look for the actual differences instead of where they don't exist, on my system at least, we'll never get anywhere. 5) It has nothing at all to do with the resolution of my system, any more than it has to do with the difference between "abcde" and "abcde". 6) I don't know how other people could copy measurably identical files across the Web and hear a difference, but if they hear a difference, something is happening different than happens for me - writing the files to different locations maybe? I don't have any idea, but that needs to be explored. 7) Measurements of electronic components are extremely complex if done 100 percent, for reasons you already know. Measurements of file contents are not complex, in fact, they follow the simplest of mathematical truth tables, and will tell you exactly with zero error 100 percent of the time what the file contents are. Assuming you have a 100 percent reliable utility program.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jun 10, 2012 23:44:07 GMT
It would appear then, that your computer works differently to the computers of the many members, and others who have heard these differences. It is a waste of time, and boring to the vast majority of members, for me to continue discussing this with you. Have a nice day. Alex
|
|
|
Post by dalethorn on Jun 11, 2012 1:08:16 GMT
A waste of time to investigate? That's pitiful.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jun 11, 2012 1:12:49 GMT
A waste of time to investigate? That's pitiful. No. A waste of time arguing with someone who claims to be a world's leading Software Expert. Investigations are still ongoing elsewhere.
|
|
|
Post by dalethorn on Jun 11, 2012 1:51:49 GMT
A waste of time to investigate? That's pitiful. No. A waste of time arguing with someone who claims to be a world's leading Software Expert. Investigations are still ongoing elsewhere. So are you claiming I'm not a world-class database and software expert? That's pretty brazen of you, and wrong of course. I've tried to educate you on what computer file contents mean, but it's like talking to my parrot. Stubborn is one thing, but you're raising it to an art form. And no, your stubbornness does not equal my expertise. My expertise comes from tens of thousands of hours of work, not from stubbornness and blinders. Do the work. Investigate. Find the differences.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jun 11, 2012 2:39:22 GMT
No. A waste of time arguing with someone who claims to be a world's leading Software Expert. Investigations are still ongoing elsewhere. So are you claiming I'm not a world-class database and software expert? That's pretty brazen of you, and wrong of course. I've tried to educate you on what computer file contents mean, but it's like talking to my parrot. Stubborn is one thing, but you're raising it to an art form. And no, your stubbornness does not equal my expertise. My expertise comes from tens of thousands of hours of work, not from stubbornness and blinders. Do the work. Investigate. Find the differences. Modesty is clearly not one of your strong points.
|
|
|
Post by dalethorn on Jun 11, 2012 2:59:21 GMT
So are you claiming I'm not a world-class database and software expert? That's pretty brazen of you, and wrong of course. I've tried to educate you on what computer file contents mean, but it's like talking to my parrot. Stubborn is one thing, but you're raising it to an art form. And no, your stubbornness does not equal my expertise. My expertise comes from tens of thousands of hours of work, not from stubbornness and blinders. Do the work. Investigate. Find the differences. Modesty is clearly not one of your strong points. False modesty is lying. I wouldn't want you to call me a liar, although given all that's happened....
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jun 11, 2012 3:30:52 GMT
Dale You seem to think that I am the only one making similar claims. I am not. I am in the unusual position here in R.G. of being more open than most, whereas others in other forums often get silenced by never ending sarcastic replies from resident EEs and software writers, and learn to keep quiet. Many RG members are in the same situation where they no longer wish to get involved for the same reason.Many HiFi writers have been saying similar for quite a few years. Recently there were the T.A.S. reports about .wav vs.flac, where they came under heavy attack from elsewhere. In the jPlay forum "screenmusicargentina"also came up with similar findings to mine independently, and even supplied some of his own original music for comparison purposes..There have been quite a few tests done also involving myself and Marcin,who is a co-developer of jPlay, and John Kenny.Erin from DIYAudio and now an R.G. member, was effectively silenced in DIYAudio after he reported hearing improvements with .wav files that were ripped in Windows using Safe Mode.Erin's claims were put to the test and verified by a member employed in the I.T. area. Later, after much "peer pressure" (more like bullying) he retracted his original report. Due to a lot of sarcastic replies , and some personal attacks on Erin, the DIYAudio moderators ended up closing the thread.Martin Colloms, who is a chartered Engineer,Acoustic Consultant, and HiFi Reviewer and magazine editor with experience dating back to the 80s has also verified my claims in print. I have attached a copy of an article from another HiFi writer who stated that PC rips were far below the SQ of those from a Naim Unitiserve. ( I will shortly delete that page, as it is copyrighted material.) The writer of that article also came under heavy and sarcastic attacks in their forum for daring to make such comments. It appears that bullying tactics are the norm for many Objectivists. Alex Attachments:
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jun 11, 2012 5:10:51 GMT
BTW, with the ATH M50 I can clearly hear 16 khz, just like the Beyer DT 48. I don't hear that tone clearly on the Shure 1840 or Senn HD800, even with the best gear. How about that! HF rolloff on the $1500 headphone. Attachments:
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jun 11, 2012 5:26:26 GMT
Dale,
It is quite possible you do not hear 16kHz with a HD800/SRH840, but can with an M50/DT48 or even other HP's for several reasons.
The DT48 has (opposite the average SPL in the 300Hz to 2kHz region) an average level of highs (in the mentioned region) of +10dB while the HD800 and SRH1840 are (average) -5dB around the same point so on average that region is 15dB louder (and is so for a reason) which may also make a difference when nearing the hearing treshold. The M50 has about the same average level as the other 2 (-5dB) b.t.w. but there is another reason as well.
The difference between the M50 and the other 2 headphones might very well be due to your test method. When you look at the raw graphs from Tyll it can be easily seen that at those frequencies the highs are behaving eratically having dips and peaks close to each other in that specific frequency band, meaning a 15 kHz sinewave can be in a dip (-30dB or more) in one headphone and on a peak in the other one. This difference can let you hear or not hear a sinewave while reproducing a slightly different tone better.
This is why testing with a single tone often is not done with mechanical transducers but is with electrical devices.
Use a stepless tone generator for this 9and manually sweep through that area), a sweeping tone or the use of warble tones (a sinewave that sweeps up in down closely around the target frequency in a a certain pace (frequency) or terts noises (very narrow band parts of a white noise signal) as one would while measuring speakers would give more reliable test results.
It's experiments with transducers, single sine waves and hearing abilities that give proper measurements and science their bad name in the world of subjectivity.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jun 11, 2012 6:47:15 GMT
Where the dispute lies is simple. When I copy *measurably* identical files to my system, *all* of those variables are eliminated, because: 3) My computer did not *receive* a physical thing when copying from the Web, i.e. it received an electric signal and *created* the bytes indicated in that signal one at a time, *exactly* the same as if I wrote those bytes myself, *exactly* the same as if I had written "abcde" twice on a single sheet of paper. 7) Measurements of electronic components are extremely complex if done 100 percent, for reasons you already know. Measurements of file contents are not complex, in fact, they follow the simplest of mathematical truth tables, and will tell you exactly with zero error 100 percent of the time what the file contents are. Assuming you have a 100 percent reliable utility program. Dale, In Alex his defense and to make sure the claims are understood properly. 3: In theory you are correct and as a computer programmer you are correct and here is the thing Alex probably agrees with you. Funny you mention the writing of ABCD and even when you would have said printing of ABCD Alex could make his point. Here is the point: When you write ABCD twice it still reads as ABCD (if you handwriting is readable) in both cases. When you look at the writing you will see they are both different in letter form, pen pressure, line thickness e.t.c. Even when printing twice even the same file and looked at miniscule detailes under a microscope or on 2 different brand printers or even 1 printer and differnt types of paper the letters differ. When you write it down twice, once in a sloppy handwriting and once in very beautifull letters the message is the same but it reads different and you may get a totally different impression/feel with it. Even reading the same page/print in different lighting conditions may evoke different emotions. That's where Alex (and others) feel the differences lie, not in the bit contents they are not disputed. 7: is about data integrity. It is also not disputed eventhough it appears it is. Some feel the timing of the data is important in the whole process where I (with no software but analog and digital background) am convinced jitter is NOT the root of all evil in every stage and have technical reasons for that + my own experience in audio and perception. This line I wrote above: Even reading the same page/print in different lighting conditions may evoke different emotions.Is IMO the root of all perceived differences, but is just MY opinion untill proven otherwise. I have not seen proof otherwise (other than subjective findings) so stick with it untill it is shown to exist in certain tests which in turn will never take place. As I already stated grains of truth mixed with more or less educated assumptions or theories based on subjective and sighted findings can create compelling evidence to some and raise questions for others. Much in the same way as this. www.amazon.com/Diamond-Digital-Audio-Ethernet-Connection/dp/B003CT2A6I/ref=cm_cr_pr_product_topScroll down and read the reviews of this cable and the differences in emotions that are invoked using expensive video cables so it's not only audio. IMO only bandfwidth limiting, which can be expected to be happening, should not envoke these reactions. It's really exactly the same as in the audio world (perception thingy). As a photographer I can appreciate good bokeh or the way a good lens makes an image more pleasant to look at, tonal scales in B&W, composition, shapes of diaphragms and lens combinations do create. Thing is not everyone sees those details as they don't know WHAT to look for an appreciate. Most will just notice the content and can only tell if they like it. Some like technical horrible pictures but because there is a child they know well in there makes the picture great for them. Looking at a TV screen I do not see what's been reported... others might very well have this experience. I hope this clears the air between you guys as I simply think it's a misunderstanding rather than disagreement. Perhaps my explanation in a different way that may or may not appeal to you sheds some light.
|
|
|
Post by dalethorn on Jun 11, 2012 14:10:09 GMT
To Alex and Frans. Firstly, I use many tones and frequency sweeps, and BTW the M50 I have is not that choppy in the highs above 10 khz. A little choppy, but not that much. The DT-48 is much worse. But if you test enough you can figure it out. This isn't some random thing I tried once - many, many tests and I continue to test.
On the computer files, I am well aware as you say about handwriting differences. But computer bytes do not vary, because they are numbers. And not floating point either - they are whole numbers from 0 to 255 (single-byte systems). You only get differences when you aggregate a lot of bytes and run them on a player. So how do we get those differences? A million or a billion ways - too many to describe. But I would suggest narrowing the possibilities when suggesting that people download and test measurably identical files. First, have them download them one file at a time, saving them in the same folder, and saving with sequential filenames such as 1.wav, 2.wav, 3.wav etc. Save the files on your fastest harddisk media. The reason I suggest harddisk is because you will get the least interruption playing from an internal harddisk, assuming of course you kill any background processes, Microsoft/Apple polling for updates, etc. etc. Also when using flash media, be aware of how flash controllers work, moving data around to account for the "wear" factor. There are articles on that and other flash media secrets. Harddisks don't account for wear factor or move things around, but your computer could be moving things around with those background processes you should stop.
Now if you can really get the process streamlined so everyone is doing the same thing, you may discover where some of those differences are coming from.
|
|
|
Post by freddypipsqueek on Jun 11, 2012 14:59:32 GMT
To me there are 2 points here.
Computers do not store the 'handwriting' contained in the data. As such it does not change the contents of the file. As such the playback of 2 identical files cannot be down to the way in which they they were created, emailed, zipped, or whatever. That is the definition of digital. It must be down to the way they are played back.
Having said that there are numerous other reasons why (I think) computer playback sounds different. When I was fussed about computer gaming performance a prescribed benchmark (say Quake 2) would get a number of results despite the PC being the same each run. In fact such benchmarks were often run in batches of (say) 5 and averaged. I could not tell you why the figures varied but I expect that's also true with music playback. It only needs a small change in file fragmentation to affect hard drive performance and file placement in memory etc.
Finally. Nobody knows if the files sound better. If you create a folder for one type of 'wav' on one drive and another for the others on a different drive it maybe down to the drive. Or it maybe that the different drives make one set of the files seem better to that listener but not to others.
There is no doubt that if I had the time I would look at improving the kit I use to rip files. I don't however and that is, I think, the weakest part of my argument.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jun 11, 2012 16:12:17 GMT
Computers do not store the 'handwriting' contained in the data. As such it does not change the contents of the file. As such the playback of 2 identical files cannot be down to the way in which they they were created, emailed, zipped, or whatever. That is the definition of digital. Exactly the way I see it Having said that there are numerous other reasons why (I think) computer playback sounds different. When I was fussed about computer gaming performance a prescribed benchmark (say Quake 2) would get a number of results despite the PC being the same each run. In fact such benchmarks were often run in batches of (say) 5 and averaged. I could not tell you why the figures varied but I expect that's also true with music playback. It only needs a small change in file fragmentation to affect hard drive performance and file placement in memory etc. IMO this would only be a problem if: a: the speed of the whole processing right up to the data being written to the DAC is not high enough for one or more reasons. Mind you digital audio storage/reading/processing is not a 'real time' event. Only the writing of data to the DAC itself is nearing a realtime event. Even streaming audio/playback is buffered by definition. b: An internal/external crappy sound card is used (the ones with little to no jitter rejection) or internal DAC's that are sensitive to powersupply variations/external DAC's with poor powersupply management. There is no doubt that if I had the time I would look at improving the kit I use to rip files. I don't however and that is, I think, the weakest part of my argument. IF there was only a slight hint in the direction of the reasons of said differences other than reported subjective findings I would bite myself into this.
|
|
|
Post by dalethorn on Jun 11, 2012 18:56:54 GMT
IF there was only a slight hint in the direction of the reasons of said differences other than reported subjective findings I would bite myself into this. Aw, c'mon - you're missing all the fun in life. Ask yourself - how long's it been since you've been in the mosh pit, laddie?
|
|