I agree, I suspect in my case the cables worked fine on the HD's because the driver was better and probably handled the errors. In the case of an audio stream, who knows what difference the errors will make? It will be dependent on the software. So the perceived improvements discussed here could be as simple as removing errors, and the software not having to deal with them.
Note that the cables I was using were the asus provided ones, then I bought some $5 ones that made no difference, it wasn't until I tried the more expensive one with all the marketing BS that the problem went away. I thought digital is digital, it either works or it doesn't, that was sort of the case here, however it worked fine in one application but not in another. This I put down to software and error detection/correction.
Tony.
I just copied data over a standard SATA cable with ~80MB/s to my SSD 🙂
I have build many PCs and worked in a building with ~800 Computers that had at least 1 SATA HD. I can´t remember we had to change only "1" of these cables ever.
I doubt we will ever have to wonder about an errorate that can even remotely confuse audiofile playback until something is very broken.
I have build many PCs and worked in a building with ~800 Computers that had at least 1 SATA HD. I can´t remember we had to change only "1" of these cables ever.
I doubt we will ever have to wonder about an errorate that can even remotely confuse audiofile playback until something is very broken.
SATA communication is guarded by checksums in hardware. If the checksums fail, errors get thrown and logged (smart, OS logs, etc.). There is no way this hardware can distinguish between audio data or e.g. executable code being transfered over the SATA interface.
I have encountered broken unreliable SATA cables. But that comes to the point - they either work reliably or don't. Regardless of what type of data is being transferred.
The improvements that are possible to hear by audiophiles when as an example a SATA cable is replaced with a silver SATA cable have little to do with error levels or error correction.
The hearing are an extremely good measuring instrument that are not limited in function.
What you are able to hear as a difference is simply a difference in the total accumulated noise level and sometimes in the jitter level.
I myself do not only use improved cables, but I have galvanic isolation on SATA, USB etc. etc..
The galvanic isolation gives a greater improvement in audio performance than cables alone can do.
These improvements in audio performance are as said related to noise levels and jitter levels and have little to do with error correction and basic function.
If the ear or eyes (noise and jitter will change the visual aspects) are not involved then the noise and jitter levels only need to be good enough to avoid data loss or data errors...
The hearing are an extremely good measuring instrument that are not limited in function.
What you are able to hear as a difference is simply a difference in the total accumulated noise level and sometimes in the jitter level.
I myself do not only use improved cables, but I have galvanic isolation on SATA, USB etc. etc..
The galvanic isolation gives a greater improvement in audio performance than cables alone can do.
These improvements in audio performance are as said related to noise levels and jitter levels and have little to do with error correction and basic function.
If the ear or eyes (noise and jitter will change the visual aspects) are not involved then the noise and jitter levels only need to be good enough to avoid data loss or data errors...
Last edited:
Wow, what an entertaining thread!
Anyone who believes shielded silver or even shielded gold wires affects a digital transmission node knows nothing about digital transmission nodes.
Audio?.....that's the analog wavefront that stimulates your ear hairs!
From your speaker cones.....hmmmm...where's the weak link in this node scenario.
Hmmmmm.....sounds better to me!
Taiwan man needs to keep trying though.....
It sort of makes you wonder if these guys who claim to be engineers and beta testers from S. Africa really know anything about electronics or if they just think they do.
I could see using Taiwan shielded silver for runs from stage to the mobile truck to cut down on noise generated in the audio/video feeds....but then that would be too easy......just the slighest MADI drift will sound soooo much diffrent over the arrays.
Anyone who believes shielded silver or even shielded gold wires affects a digital transmission node knows nothing about digital transmission nodes.
Audio?.....that's the analog wavefront that stimulates your ear hairs!
From your speaker cones.....hmmmm...where's the weak link in this node scenario.
Hmmmmm.....sounds better to me!
Taiwan man needs to keep trying though.....
It sort of makes you wonder if these guys who claim to be engineers and beta testers from S. Africa really know anything about electronics or if they just think they do.
I could see using Taiwan shielded silver for runs from stage to the mobile truck to cut down on noise generated in the audio/video feeds....but then that would be too easy......just the slighest MADI drift will sound soooo much diffrent over the arrays.
Digital data dose not accumulate noise unlike analogue, and jitter in the Audiophile sense is not a problem with SATA communications, once the data is read from the disc (and checked via CRC, as Wombat pointed out) it is buffered, so the squarewave is recreated when transfered from the buffer to the next reciever in its path from hard drive to sound card, and jitter would only be of concern if you use the SPDIF interface, as the last link in the chain (and beleive such low levels of jitter are audible). As to noise, when you refer to accumulated noise, galvonic isolation at the final stage before the DAC will avoid the transfer of the general background digital noise. This wont get rid of the digital noise that is on your DAC board, but that will be minimal dependant on the layout and decoupling scheme.
I am curious as to why in the proffesional world of digital design (audio/communications related) they dont rely on their ears alone to design but use high speed scopes, network analysers and simulation software for the digital side of the design. And use ears to listen to the final output only
I am curious as to why in the proffesional world of digital design (audio/communications related) they dont rely on their ears alone to design but use high speed scopes, network analysers and simulation software for the digital side of the design. And use ears to listen to the final output only
IMO there is no point in discussing causes of an effect which has not been confirmed to exist objectively. Ears are definitely an extremely good measuring instrument, but the brain down the stream is just as good at giving subjective impressions.
Once someone does an objective undisputable listening test confirming the effect, we can start analyzing the causes.
Digital data dose not accumulate noise unlike analogue, and jitter in the Audiophile sense is not a problem with SATA communications, once the data is read from the disc (and checked via CRC, as Wombat pointed out) it is buffered, so the squarewave is recreated when transfered from the buffer to the next reciever in its path from hard drive to sound card, and jitter would only be of concern if you use the SPDIF interface, as the last link in the chain (and beleive such low levels of jitter are audible). As to noise, when you refer to accumulated noise, galvonic isolation at the final stage before the DAC will avoid the transfer of the general background digital noise. This wont get rid of the digital noise that is on your DAC board, but that will be minimal dependant on the layout and decoupling scheme.
I am curious as to why in the proffesional world of digital design (audio/communications related) they dont rely on their ears alone to design but use high speed scopes, network analysers and simulation software for the digital side of the design. And use ears to listen to the final output only
System noise accumulate.
Even if data is read from a SATA drive and into RAM and played then the noise generated from the SATA drive / connection when physically connected will accumulate and alter the audio performance.
If the SATA drive was physically disconnected / powered off during play from RAM the accumulated noise would be altered and the audio altered (most possibly improved).
Yeah as I've said its general digital noise that all digital systems create, so galvonic isolation at the DAC stage will stop the noise getting onto tha analogue wave. You cant engineer that out without a good scope, and even then all you can do is cut it down as a lot of the noise is created by multiple gates switching at the same time ( call simultaneous switching noise). The way you were describing the jitter and noise sounded like it accumulated on the digital data stream, ie that the noise was supperimposed on the wave and transferred through the various buffering as it travels from disk to DAC.
If you isolate between the PC and the DAC then you will limit the noise to the minimum. As to digital systems, that it the whole point of using it, it is pretty tollerant of noise, but to examine this noise you need a good scope, 13GHz + will suffice.
Henry Otts book Electromagnetic Compatability Engineering is an essential read if you are doing digital (and analogue) design, and at the end of the day it is all engineering and basic physics.
Still waiting for the revalation that will open up a whole new world of design, or is noise it, if so its already known and a hell of a lot of design effort goes in to reducing it and keeping it to acceptable levels for systems to operate...
So isolate the DAC from the PC, a good qaulity fibre link will do the trick, no ground loops or PC noise to worry about, or use an ethernet interface, they are pretty well isolated.
If you isolate between the PC and the DAC then you will limit the noise to the minimum. As to digital systems, that it the whole point of using it, it is pretty tollerant of noise, but to examine this noise you need a good scope, 13GHz + will suffice.
Henry Otts book Electromagnetic Compatability Engineering is an essential read if you are doing digital (and analogue) design, and at the end of the day it is all engineering and basic physics.
Still waiting for the revalation that will open up a whole new world of design, or is noise it, if so its already known and a hell of a lot of design effort goes in to reducing it and keeping it to acceptable levels for systems to operate...
So isolate the DAC from the PC, a good qaulity fibre link will do the trick, no ground loops or PC noise to worry about, or use an ethernet interface, they are pretty well isolated.
I think this kind of claim is like junk mail or phishing. Say an incredible thing to a big enough audience and there will be a just few who find it credible. These are your income generators.
The more debate the subject gets, the greater the audience. And so it goes on....
The more debate the subject gets, the greater the audience. And so it goes on....
That was my reasoning before buying the new cable. Note that the same (broken) cable works fine on HDD's just not on my pioneer DVD RW drive running on linux. Many people have reported the same issue (linux broken, windows works fine). It can't be purely a hardware level issue it has to be the driver (otherwise it would be broken when dual booting windows).
Anyway It's a problem I had in the past, and it's anecdotal in nature so I'll leave it at that, believe me or not as you see fit 🙂
Tony.
I am beginning to tire of this Thread.
Some clearly show competence in the subjects being discussed.
Others show they are almost completely ignorant of the science.
Yet one does not seem willing to learn from the other.
It has become a pointless (not point scoring where we could have a winner) argument where the two sides (or is it only one side) are not willing to ever listen to the other side.
Some clearly show competence in the subjects being discussed.
Others show they are almost completely ignorant of the science.
Yet one does not seem willing to learn from the other.
It has become a pointless (not point scoring where we could have a winner) argument where the two sides (or is it only one side) are not willing to ever listen to the other side.
Andrew either dont subscribe to the thread or simply stop reading it. And why oh why do you post on it. All that helps with is moving it back up the list..Stop posting! it will eventually go away all of its own accord..
Regards Ian
Regards Ian
I want to see a provable conclusion.
I have a feeling ( with me being a digital ignoramus) that I have a conclusion but I want to know if my tentative conclusion is flawed and if so why.
I expect to eventually learn something.
But this tennis match rally has no end in sight.
I have a feeling ( with me being a digital ignoramus) that I have a conclusion but I want to know if my tentative conclusion is flawed and if so why.
I expect to eventually learn something.
But this tennis match rally has no end in sight.
I have no problem believing you. The DVD drive or the SATA controller could have been configured differently in linux and the cable poor quality stood out. But it would have happened no matter whether you had a CD-Audio or a DVD-R with pictures medium in the drive. And that is the point of our discussion here - how come audio data would get corrupted while executable code was transferred OK, not causing any program/kernel crashes.
Im not a audiophool but I have encountered situations like wintermute which defies common engineering knowledge and practice. This subject with the sata cables may have some truth in it, some years ago there was a performance shootout between various harddrives published in a PC magazine (maybe I can still find a copy between all my junk in the garage) where the tester got a problem halfway through the test with the sata cable and made a change to the cheap cable provided by the hard drive manufacturer. The tests were started from the beginning but to his surprise the already tested drives were displaying slower speed in read and write. He couldnt understand it and neither could I but can one say he was lying. He mentions at the end of the tests that he purchased a new cable , the same as the first one used and speeds were again faster than the cheap cable, he left it at that. Can someone explain this to me......???
Second case, take a mp3 file, copy and paste it from one hardrive to another, even using a memory stick to change the said file to a different pc altogether. The more you move the file around the more youll notice the effect. When you make a copy of that file after the various transfers move the file back to its original drive and listen and compare to the original. I have found they sound very different and in specific the high frequencies have deteriorated. File size and number of bits dont explain the difference as they are always the same. The DJs I purchase these mp3 files for use in night clubs tell me they find the same effect and insist that I purchase the files direct from internet but download them direct to their own hardrives because of this. Ive found the same effect happens with the so called Flac lossless files. Anyone care to explain this one.......??
Isnt a 0 a 0 and a 1 a 1 ??? Ive had headaches trying to figure this out.
Second case, take a mp3 file, copy and paste it from one hardrive to another, even using a memory stick to change the said file to a different pc altogether. The more you move the file around the more youll notice the effect. When you make a copy of that file after the various transfers move the file back to its original drive and listen and compare to the original. I have found they sound very different and in specific the high frequencies have deteriorated. File size and number of bits dont explain the difference as they are always the same. The DJs I purchase these mp3 files for use in night clubs tell me they find the same effect and insist that I purchase the files direct from internet but download them direct to their own hardrives because of this. Ive found the same effect happens with the so called Flac lossless files. Anyone care to explain this one.......??
Isnt a 0 a 0 and a 1 a 1 ??? Ive had headaches trying to figure this out.
If you care to present an objective test results proving you really do hear what you think you do (10/10 correct answers on properly conducted DBT falls to 95% confidence interval that you could really hear the difference), perhaps someone will care to try to explain causes of the effect confirmed thus to really exist.
damm i came to Diy audio to get away from computer geekness. the only explanation i have at the moment is error correction getting things wrong . Only 1s and 0s not hard to understand . When we play a digital music file we are not able to use a checksum to detect errors so the error correction must make a guess to the missing bit being a 1 or a 0. Now that is easy to get wrong . the more error correction the more errors in the music file .
Just my thoughts at the moment .Im now off to search the web so take my rammblings with a pinch of salt for now . However i will be back . I agree with Andrew that this thread is becomming very frustrating and i guess others are feeling this way too.
Regards Ian
Just my thoughts at the moment .Im now off to search the web so take my rammblings with a pinch of salt for now . However i will be back . I agree with Andrew that this thread is becomming very frustrating and i guess others are feeling this way too.
Regards Ian
On the issue of linux and windows . I made an obsevation some time ago . Windows XP seems to make the hard drive more mechanicaly noisey than in linux . Anyone else noticed this. Surley as said it must have somthing to do with Hardware drivers as the hardware was the same used for linux and windows.
Regards Ian
Regards Ian
- Status
- Not open for further replies.