Rick, not sure if I understand you correctly - are you saying that it makes a difference how and with which equipment you rip your CDs and copy them onto your SD cards? Or in other words: when everything after inserting and playing the SD card (with your SDTrans as I read) is the same, you experience an audible difference from how the content of the SD card was created?
Last edited:
I know it might sound kooky but when you think about it HOW could creating the music file NOT require as much attention as playing it back? Well, it makes sense to me.
I can assure you there is as much difference between a rip made in a calm quiet environment as the difference I hear from my SDTrans as compared to my music computer. I was using WTFPLAY on the music computer the LINUX OS player which was by far the best player I had heard. I had been able to use it before its introduction to the world.
I embarrass myself when I use the audio adjectives but I have no choice.
There is much greater sense of naturalness - the big dynamics are bigger and the smaller dynamics are revealed. Of course this is most apparent on non-electric instruments. There is more assurance that you are listening to vibrations of wood and string, etc - not that smoothed over quality that turns the sawtooth into the sine wave. Yes, another layer of haze has been removed from the window. When the SATA SD writer was used the bass is powerful yet under control. When I was using the USB writer I was concerned that I might have done something to the bass with my OS mods. It was wimpy and lightweight. NOT the case. Last night my room was rumbling with bass extension lower than I have experienced before.
For a taste - get the SD writer I mentioned and hear the difference between USB and SATA. Then you can extrapolate from there.
Digital noise is inevitable but it doesn't have to be pervasive. Slowing down the system and giving the CPU much less work to do WORKS. With a minimized registry there is less to poll and, I figure, more polling of what matters to us within any timeframe. Imagine a stopwatch with only 20 "ticks" per revolution versus 200. If those essential 20 were within the larger clock you can see there would be much larger intervals between the instructions we need for a good rip.
I am not a computer guy so my explanations are conceptual but I think they are basically correct. I am a tinkerer who has tinkered extensively with XP over the years. The newer versions of WINDOWS are so huge as to be basically impossible to do much with. TRUSTED INSTALLER and extensive security make it better for what it was intended but not good for audio.
I can assure you there is as much difference between a rip made in a calm quiet environment as the difference I hear from my SDTrans as compared to my music computer. I was using WTFPLAY on the music computer the LINUX OS player which was by far the best player I had heard. I had been able to use it before its introduction to the world.
I embarrass myself when I use the audio adjectives but I have no choice.
There is much greater sense of naturalness - the big dynamics are bigger and the smaller dynamics are revealed. Of course this is most apparent on non-electric instruments. There is more assurance that you are listening to vibrations of wood and string, etc - not that smoothed over quality that turns the sawtooth into the sine wave. Yes, another layer of haze has been removed from the window. When the SATA SD writer was used the bass is powerful yet under control. When I was using the USB writer I was concerned that I might have done something to the bass with my OS mods. It was wimpy and lightweight. NOT the case. Last night my room was rumbling with bass extension lower than I have experienced before.
For a taste - get the SD writer I mentioned and hear the difference between USB and SATA. Then you can extrapolate from there.
Digital noise is inevitable but it doesn't have to be pervasive. Slowing down the system and giving the CPU much less work to do WORKS. With a minimized registry there is less to poll and, I figure, more polling of what matters to us within any timeframe. Imagine a stopwatch with only 20 "ticks" per revolution versus 200. If those essential 20 were within the larger clock you can see there would be much larger intervals between the instructions we need for a good rip.
I am not a computer guy so my explanations are conceptual but I think they are basically correct. I am a tinkerer who has tinkered extensively with XP over the years. The newer versions of WINDOWS are so huge as to be basically impossible to do much with. TRUSTED INSTALLER and extensive security make it better for what it was intended but not good for audio.
Thanks Rick.
Have you compared the exact and complete contents of SD cards ripped with your different environments and tools? I mean file and data comparison on a bit level?
From my understanding, a rip is either correct, i.e. identical ('bit perfect') to the content on the CD (as verified by Accurate Rip, I use Exact Audio Copy for this), or it isn't, there's no in between.
If you hear a difference, there has to be different content on the SD cards. It needs to be determined where this comes from. If all rips are bit perfect, i.e. exactly resembling what's on the CD, there's no difference in content on the SD card by definition, right?
Have you compared the exact and complete contents of SD cards ripped with your different environments and tools? I mean file and data comparison on a bit level?
From my understanding, a rip is either correct, i.e. identical ('bit perfect') to the content on the CD (as verified by Accurate Rip, I use Exact Audio Copy for this), or it isn't, there's no in between.
If you hear a difference, there has to be different content on the SD cards. It needs to be determined where this comes from. If all rips are bit perfect, i.e. exactly resembling what's on the CD, there's no difference in content on the SD card by definition, right?
Thanks for this info Rick,
An issue that needs serious looking into ...if reading SD is important...writing has to be just as important....one would guess
Well, my first question is why would anything make a difference when playing music from your hard drive but not make a difference when writing to your hard drive and subsequently to the SD card?
There is obviously something going on.
I have no way of reading the one and zeros on anything nor am I interested in that.
I tried to use logic in my approach and not conventional wisdom.
If you do not think any computer based players sound different from one another then I can understand being skeptical of the ripping process making a difference. You're likely not to hear the difference in ripping, either.
My comments were made because I, too, had assumed ripping was not important. I think it was more HOPING it was not important.
A little work can bring some big rewards.
Just as with a good phono cartridge - that which is lost at the beginning of the chain can never be regained.
I am an LP guy and I do think through digital the same way I would with analogue.
Try the SATA SD card writer and listen to the difference as a starter.
Well, when you play back music from your computer, you have all the crappy, noisy PC environment influence and corrupt your digital data output, with interferences, noise and artifacts etc. while streaming your music out. In contrast, when you write an SD card, you create a static set of files, an image on a storage media, the only thing that stays after that creation process is the digital data. This is asynchronuos, two cards by design and concept are identical if their content is identical, that's why I have asked for a bit-wise comparison. Or in other words, information theory does not allow for influence of 'noisy' or 'clean' data extraction and creation methods, once the data created and stored are identical, the resulting source is identical.
I have vast experience with computer audio, including cPlay and many others - that's why I have moved away from PC audio many years ago. I base my judgements on measurements first, then listening - some six or seven years ago I had built a number of PC based audio sysems with extremely stripped layout, special Linux OS, entirely powered with batteries, soundcards (Juli@ ;-) ) completely separated and also battery powered - and when I entered a professional lab (run by a guy who does calibration and tuning of professional measurement gear) with these truly extreme, best-possible PC audio machines, I was shocked that the i2s signal was still buried under heaps and loads of noise and crap and trash, and barely recoverable. That's when I moved away from PC audio and went completely SD transports. So absolutely, I'm a convert already ;-)
My point is: We need to know where the difference you hear comes from. There's no binary comparisons, there's no measurements, so there's no evidence to base further procedures and recommendations on. That's what I was asking for, and hoping we can collect these.
Last edited:
It is obvious you find no need to do this.
I cannot figure out why you would respond if you want graphs and numbers.
If you are the graphs and numbers guy I would say build something similar and find the graphs and numbers that show there is something here or that it is nothing more than placebo. I have no way of doing that. I have made an observation. If you think there is something here then pursue it. If you don't continue on as you have.
I hear what I hear. If you do not believe I am hearing what I say I am dismiss me and move along.
I am baffled at your questions and how they relate to what I said.
I did not make my post to get into arguments over the most dreary of all audio arguments, "can aural judgements be quantified". You must think YES and I am confident the answer is NO.
I cannot figure out why you would respond if you want graphs and numbers.
If you are the graphs and numbers guy I would say build something similar and find the graphs and numbers that show there is something here or that it is nothing more than placebo. I have no way of doing that. I have made an observation. If you think there is something here then pursue it. If you don't continue on as you have.
I hear what I hear. If you do not believe I am hearing what I say I am dismiss me and move along.
I am baffled at your questions and how they relate to what I said.
I did not make my post to get into arguments over the most dreary of all audio arguments, "can aural judgements be quantified". You must think YES and I am confident the answer is NO.
Not sure I understand "fixing" a rip?
I agree with your assessment of SDTrans. What a breath of fresh air after years of the triumph of hope over reality with computer based players.
I am using a large capacity LiFePo battery pack into the main connector for the moment. Will be going for four supplies - 3 - 3.3 volts supplies and leave the five volts into the main connector the FPGA supplies. Also have some trick clocks to install. BUT for the moment it sounds amazingly good.
I can say with confidence that the difference between SDTrans with my old rips and the new is akin to what I heard when going from computer/WTFPLAY to SDTrans.
I love this little device sincerely.
Using a highly modified SOEKRIS DAC - IIS lines less than three inches between them. SOEKRIS is A123 LiFePo powered throughout - no OEM regulation anywhere on the board.
My ripper OS can only do one thing and one thing only!
Now there might be something gained by using the SATA SD writer when copying to the SD card but what is missing will never be there.
If it isn't there from the rip ...
Forgot to mention that this thing (obviously) cannot connect to the internet.
If it could it would be corrupted in about 15 seconds. It has NO security - all of it has been removed Important for better sound.
So there is no ACCURATE RIP info (I did enter the offset into the registry since I use the same CD-rom in my other machines) and no filled int he blanks song titles. I figure ACCURATE RIP is only so useful - if it tells you the disk is not perfect would you really go out and buy a new one? I know I would not and never have. Plus with SDTrans - I cannot read the display very well so I am happy with Track 1, Track 2, etc. You could always type this stuff in for yourself but I am not interested in that!
Now there might be something gained by using the SATA SD writer when copying to the SD card but what is missing will never be there.
If it isn't there from the rip ...
Forgot to mention that this thing (obviously) cannot connect to the internet.
If it could it would be corrupted in about 15 seconds. It has NO security - all of it has been removed Important for better sound.
So there is no ACCURATE RIP info (I did enter the offset into the registry since I use the same CD-rom in my other machines) and no filled int he blanks song titles. I figure ACCURATE RIP is only so useful - if it tells you the disk is not perfect would you really go out and buy a new one? I know I would not and never have. Plus with SDTrans - I cannot read the display very well so I am happy with Track 1, Track 2, etc. You could always type this stuff in for yourself but I am not interested in that!
No. If the rip failed to copy the bits correctly, then there is no way to reconstruct the correct bits. That is why I recommend using a ripper program that compares the ripped result with an on-line database (accurip) to verify that the rip was done correctly.
It is important to understand that the only errors/distortions you can get when doing a rip is missing and/or changed bits. Either the bits got ripped correctly, or they didn't. Yes, the accuracy of the rip might be affected by various things (dirt and scratches on the CD, and various analog errors in the CD drive mechanism), but these days they are rare - I get rip errors on maybe 1 CD in 20, and usually it is because copy protection schemes or non-audio bonus tracks. In any case, a check with the accurip database would show the error.
So the point is that as long as the rip is error-free on a bit level, as verified with comparing to a database, the bits are, as the saying goes, just bits. Your DAC/player might reproduce them well or less well, but there is nothing in those bits that carries any trace of what computer, power supply, or magic crystals you used to rip the CDs with. It could be a dirt-cheap laptop or a hyper-expensive, audiophile tweaked computer with all-rhodium cables and titanium enclosure, and it would sound exactly the same.
So please explain why this same assurance does not go both ways?
If it is all JUST BITS why does it make a difference during the playback cycle?
Like I said, I made an observation.
I know what I am hearing.
Let's face it, most of the concern is about having to re-rip again. Much easier to convince oneself it cannot make any difference.
Back to the turntable - all cartridges trace the same grooves and there should not be a difference between top echelon cartridges, but there is.
Wouldn't we all like proof of anything before we try it but that is not how it works.
I could send someone an SD card with a few of these rips. I could send someone a SNAPSHOT image of the thing. They would have to make some BIOS changes and the latest one was made with an H81 board and a HASWELL 4130T - would probably be best to use a similar configuration.
They would need the SATA based writer with the image to get it right.
I am no more interested in re-ripping all of my disks that anyone else. If there was not a clear cut reason to do so I would not be doing it.
If it is all JUST BITS why does it make a difference during the playback cycle?
Like I said, I made an observation.
I know what I am hearing.
Let's face it, most of the concern is about having to re-rip again. Much easier to convince oneself it cannot make any difference.
Back to the turntable - all cartridges trace the same grooves and there should not be a difference between top echelon cartridges, but there is.
Wouldn't we all like proof of anything before we try it but that is not how it works.
I could send someone an SD card with a few of these rips. I could send someone a SNAPSHOT image of the thing. They would have to make some BIOS changes and the latest one was made with an H81 board and a HASWELL 4130T - would probably be best to use a similar configuration.
They would need the SATA based writer with the image to get it right.
I am no more interested in re-ripping all of my disks that anyone else. If there was not a clear cut reason to do so I would not be doing it.
Because the playback involves a digital-to-analog conversion that is susceptible to noise, distortion and jitter.
And if "I know what I am hearing" is your viewpoint, then there is not much I can tell you - my views are just based on science - the science that your computer, your audio systems and all the stuff in the studio are based on - and I know science will never win over faith.
I assume you don't believe in the existence of the placebo effect or confirmation bias? Your senses (or mind) has never deceived you? Do you believe in optical illusions?
Do you understand the difference between analog and digital? A turntable and cartridge are both analog devices. They are subject to a heck of a lot of distortion, non-linearity, noise, rumble, wow and flutter. A computer is a digital device. When was the last time a cell in your spreadsheet was almost, but not quite what it should have been? A few characters distorted in your email?
That is how science and engineering works. Who do you think designed the chips and circuits you use?
Or you could simply run a bitwise compare program on your computer. Or send me a CD card with a rip of a CD that I already have, and then I could rip it on the cheapest, noisiest computer I have, and show the result is exactly the same.
If you feel the need to re-rip your CD's, go ahead - it is of course your time, and you are free to use it as you wish. My comments are purely for other people who might have started worrying because of your posting.
Well, I bet they are completely reassured now.
I expected you to LISTEN to the SD card, not to test it.
I should not have expected anything different. Do you understand how analog the digital stream is? What are those numbers signifying?
Do you even listen to music or just look at the numbers?
Your experiment, using a noisy computer defies logic and belief. What would that tell you?
I expected you to LISTEN to the SD card, not to test it.
I should not have expected anything different. Do you understand how analog the digital stream is? What are those numbers signifying?
Do you even listen to music or just look at the numbers?
Your experiment, using a noisy computer defies logic and belief. What would that tell you?
Well, I have only worked with digital sound processing for the last 35 years or so, so what do I know?
Do you understand how using analog media to store and transport digital data works? Do you understand how we can transmit terabytes of data over thousands of miles of undersea fibre optic cables using analog laser light - without any errors?
I would love to hear *your* view on what those numbers are signifying - and what could cause audible differences between bit-identical data, if reproduced with the same system, independent of *how* those bits got onto your SD card.
Understanding the technology doesn't rule out enjoying the music...
I am sure it defies belief (as in "faith"), logic is another matter. It would tell us that the bits are identical, no matter how they are ripped, as long as the rip is bit-perfect.
If you are saying that two bit-identical copies of the same track, on your SD card, sound different depending on what the digital path from the CD was, I think the onus is on you to prove that 90 years of signal and information theory is wrong. If you succeed, there is a Nobel Price waiting for you.
I would also recommend studying the effects of pareidolia, confirmation bias and cognitive bias.
I am sure this discussion won't change anybody's views. Believers will continue to believe, people who rely on knowledge, facts and evidence will continue to rely on knowledge, facts and evidence.
I am glad there are people such as yourself who know what you know. I am not trying to discount this knowledge.
What I and other kooky folks like myself are trying to find/discover is what works and what doesn't. We welcome the brilliant mistake. We try things based on our tinkerer's logic that symmetry in the recording chain makes sense.
If we find that doing all sorts of stuff that fellows such as yourself have dismissed as needless and NO WAY will work and then they do make an audible difference to the ear and not necessarily to the microphone we wonder why not try it on the other side of the equation.
Just think of the, now, accepted wisdoms that were dismissed as foolishness. You may continue to dismiss them and as I say to anyone who is happy with the way their systems sound, or measure, then leave it be. Be happy you are satisfied. You are a lucky fellow.
But for those who hear what is wrong and try to find ways to allay the errors what is the point of this typical tired criticism? There were people saying the same thing about wire decades ago.
Measurements are wonderful but what about things fellows like yourself have not figured out how to measure? And if you assume at the beginning that there is nothing there then you stymie yourself in your pursuit of proof.
The end of discovery has not yet occurred, I assure you.
Do you really think music sounds like music coming from your system? Well, I do not and I am trying to find things that improve the experience.
When I see the disconnect between all kinds of tweaks on the playback side but nothing more than an average computer with the CPU running at default speed in a vibrating box sharing a bloated operating system that might be running twenty other programs/applications/services while the rip is being made I asked myself how could we delude ourselves into thinking this makes no difference?
So, I have to assume, the folks with single purpose music playback computers are kidding themselves, too? I would then have to agree that music playback computers are not a good idea at all but they are assuredly much better than general purpose machines. But with that, I suspect, you would take issue and say the general purpose machine offers THE SAME sound quality as the tweaked single purpose machine? I am giving you the benefit of the doubt when it comes to consistency. Those who are reading this exchange will need to hear your opinion on that.
I went for the ultimate in single purpose and find that it makes a difference.
One should not do anything to purposely upset their world view when they have it all sorted out.
What I and other kooky folks like myself are trying to find/discover is what works and what doesn't. We welcome the brilliant mistake. We try things based on our tinkerer's logic that symmetry in the recording chain makes sense.
If we find that doing all sorts of stuff that fellows such as yourself have dismissed as needless and NO WAY will work and then they do make an audible difference to the ear and not necessarily to the microphone we wonder why not try it on the other side of the equation.
Just think of the, now, accepted wisdoms that were dismissed as foolishness. You may continue to dismiss them and as I say to anyone who is happy with the way their systems sound, or measure, then leave it be. Be happy you are satisfied. You are a lucky fellow.
But for those who hear what is wrong and try to find ways to allay the errors what is the point of this typical tired criticism? There were people saying the same thing about wire decades ago.
Measurements are wonderful but what about things fellows like yourself have not figured out how to measure? And if you assume at the beginning that there is nothing there then you stymie yourself in your pursuit of proof.
The end of discovery has not yet occurred, I assure you.
Do you really think music sounds like music coming from your system? Well, I do not and I am trying to find things that improve the experience.
When I see the disconnect between all kinds of tweaks on the playback side but nothing more than an average computer with the CPU running at default speed in a vibrating box sharing a bloated operating system that might be running twenty other programs/applications/services while the rip is being made I asked myself how could we delude ourselves into thinking this makes no difference?
So, I have to assume, the folks with single purpose music playback computers are kidding themselves, too? I would then have to agree that music playback computers are not a good idea at all but they are assuredly much better than general purpose machines. But with that, I suspect, you would take issue and say the general purpose machine offers THE SAME sound quality as the tweaked single purpose machine? I am giving you the benefit of the doubt when it comes to consistency. Those who are reading this exchange will need to hear your opinion on that.
I went for the ultimate in single purpose and find that it makes a difference.
One should not do anything to purposely upset their world view when they have it all sorted out.
This is true. But which side are the believers? Seems to me there is plenty of belief to go around. If you think measurements say it all you most assuredly are using a belief system, too.
And we kooky engineers support and agree with that view.
This is where the difference is - tinkerer's logic is often based in what intuitively feels right, and human intuition was never designed to deal with electronics, digital systems (or nuclear physics). We realize that some processes are one-way (you can destroy information but you can't recreate it from nothing), so a concept of symmetry doesn't always make sense.
In most cases we haven't dismissed them just like that - we have dismissed them as a result of extensive research and experimentation - and when we try to find audible differences, we don't just rely on subjective and fallible evidence from sighted listening by one person, we actually deploy sophisticated listening test protocols such as ITU-R BS.1534-3 and BS.1116-2.
Contrary to what you might think, science works by actually challenging and questioning accepted wisdom all the time. It is just that some things have stood the test of time - and in this case, we are talking about 90 years worth of extensive challenging.
Then we use listening tests. Not just casual, informal sighted listening tests, but controlled, rigorous listening tests like the ones I referred to above.
And as I said, there is a Nobel Price waiting if you prove you are right. That is not a joke. It would totally upset modern information theory (amongst other things).
No. Did I claim it did?
I think it shows you don't understand the difference between digital and analog.
I already answered that. Playback involves a DAC susceptible to noise, jitter and distortion. Transferring bits from a CD to a SD card does not - just like copying a spreadsheet doesn't.
In that case, accept my apologies for (probably in vain) trying to do just that.