Difference in digital sources

[Anonymous]

I really do not mean to be a pain but I disagree.

• You are going to compare a hard drive with an ipod, you have already suggested that different usb sources can sound different so this introduces a new variable into the data being sent to the DAC.
• You are going to test against a cd transport, this is a mechanical object which introduces numerous variables (vibration, 'on the fly' data processing etc).
• How are you going to rip the CD that is used in the transport to the hard drive and Ipod? There are more variables here than I can begin to understand.

This is why I suggested a controlled system where the same file is played from the same hard drive into the same DAC and amp where the only variable is the music server.

 

[Andrew Everard]

Ah, I see - the confusion arose due your choice of terminology. The devices you mention are all clients, not servers - the server is the unit actually holding the music, the client is the interface between that music sever and the rest of the system.

And based on some listening I have done to various client devices, including comparisons between music streamed direct from a computer to a DAC and the same datastream run via a high-quality client device, I'd say that yes, the choice of client can definitely affect the sound quality.

 

[Anonymous]

Ah well, there is me feeling a bit sheepish,

Thank you for your patience.

 

[PJPro]

Andrew Everard:I was as surprised as some posters here seem to be, but it does make a difference. Best explanation seems to be that different sticks deliver data at different speeds.

Hmmmm. Is this difference detectable in the first few bars of the track or does the sound deteriorate over time?

Ordinarily, the data will be buffered in memory and "delivered" from there. This buffer can usually be adjusted in the media software. In F2K this can be set up to a maximum of 16 seconds. I'd be very surprised if the buffer was ever emptied when set to this size. My buffer is currently set to 2 seconds.

So, if you are experiencing differences try uping the buffer size and see if the difference remains. If it does, you can rule out the speed of the device.....particularly if the difference occurs in the first few bars of the track.

 

[PJPro]

Given the growing attention being placed on computers as a source for hifi purposes I can imagine a time when audiophile computers and components emerge on the market. We are already seeing this with audiophile USB cables like this one.

Some claims may be easier to swallow then others e.g. SSDs over HDDs. Other claims will doubtless be more controversial. I can predict a debate between the theorists and the golden ears similar to those we currently see regarding exotic cables.

Is it all cobblers? You decide. It's your money after all.

 

[Anonymous]

Hi PJ,

Agree with you here. If the buffer on the recieving node is set high enough, then surely this will account for any jitter issues from the incoming source. The new Naim Dac for example, states that to minimize jitter, the incoming data stream is completely isolated from the data entering the DAC. Which to me, means that its stored in a buffer on the DAC device and then sent through with more accurate timing to the actual DAC function.

Would be a very interesting test Andrew, looking forward to the results.

 

[Anonymous]

chanukav@gmail.com:
Agree with you here. If the buffer on the recieving node is set high enough, then surely this will account for any jitter issues from the incoming source. The new Naim Dac for example, states that to minimize jitter, the incoming data stream is completely isolated from the data entering the DAC. Which to me, means that its stored in a buffer on the DAC device and then sent through with more accurate timing to the actual DAC function.

Yes, that is exactly how to do it and how it's done in other forms of digital communication.

And I still stand by my offer to eat socks, USB sticks and heck why not my keyboard as well if someone can prove differences in sound played from 2 different brands of USB sticks (bar one being faulty and the data cutting out of course). The idea is absurd. It's a piece of memory forgodssake -- and speed has nothing to do with it as every USB stick ever produced is capable of reading data at rates multiple times those needed for digital audio.

Data is read from USB sticks very much like from a regular hard drive, in chunks, into local memory. There is no concept of 'streaming' and therefore 'jitter', although a popular term in these discussions ;-), has no bearing here. Unless someone can explain how they think this would work -- I'm all ears... (although perhaps not golden ones.)

Unfortunately it's exactly the same as saying 2 brands of hard drives will produce different sound -- and if someone is prepared to suggest that, I may just give up.

 

[Anonymous]

This is interesting because it upsets one of the 'rules' of hi fi that has been around for as long as I have been interested in it. That is the assumption that you split your budget equally between source / amplifier / speakers and then add on 10% or so for cables etc.

Historically, in the days before cd, I think the rule was that you threw your entire budget at the turntable and skimped on the amp and speakers.

If the source becomes irrelevant (it is a hard drive and one hard drive is as good as any other) then that changes the equation so that all of the budget goes on to amp and speakers. It also means that it is about time that amps came with dacs as standard but I think I have laboured that point elsewhere.

 

[Anonymous]

I guess the bottom line here is.. can you hear the difference?

Professionals can tell you all they can about the wee little things that improve the sound but to your ear, if it sounds exactly the same...does it matter that you use the cheap usb cable vs the expensive one? or that you have a WD harddrive rather than a Seagate...

Someone in another recent post mentioned how he backed off his hifi obsession because he found himself so concerned with the detail of the sound rather than actually enjoying the music. I think this comment rings true for a lot of us.

I sometimes wish i could go back to the day when i knew nothing about all of this hifi stuff! .. i could just buy a system straight from the store that moved me and be happy in the knowledge that no other competition existed out there that could possibly sound any better... Ignorance is sometimes bliss! especially when my economic circumstance doesnt allow for further investigation.. hehe

 

[Andrew Everard]

storsvante:Unfortunately it's exactly the same as saying 2 brands of hard drives will produce different sound -- and if someone is prepared to suggest that, I may just give up.

Actually, now you come to mention it...

 

[Anonymous]

demartin_1:
This is interesting because it upsets one of the 'rules' of hi fi that has been around for as long as I have been interested in it. That is the assumption that you split your budget equally between source / amplifier / speakers and then add on 10% or so for cables etc.

Historically, in the days before cd, I think the rule was that you threw your entire budget at the turntable and skimped on the amp and speakers.

If the source becomes irrelevant (it is a hard drive and one hard drive is as good as any other) then that changes the equation so that all of the budget goes on to amp and speakers. It also means that it is about time that amps came with dacs as standard but I think I have laboured that point elsewhere.

I agree with all that. The move to digital sourcing/transport of audio *does* change the rules. To pick up on your example - turntables are of course highly mechanical and highly susceptible to vibrations and all those other things audiophiles are traditionally worried about. In contrast most digital sources nowadays are capable of reproducing the original material with 100% accuracy (including USB sticks) which of course is a great leap forward. This should enable everyone to put their money where it matters still. The logical conclusion is to invest in DAC, amp and speakers. And analogue cabling. And storage capacity.

And +1 on your point about integrated DAC+amps. Interconnects are evil as you need to spend money on ones which don't interfere with the analgue signal. So if the DAC+amp is integrated you eliminate that step and everyone is a winner (except for Nordost).

 

[The_Lhc]

demartin_1: If the source becomes irrelevant (it is a hard drive and one hard drive is as good as any other) then that changes the equation so that all of the budget goes on to amp and speakers. It also means that it is about time that amps came with dacs as standard but I think I have laboured that point elsewhere.

A hard drive is NOT a source, any more than a CD is a source. It is media.

A CD PLAYER is a source, I'd say the equivalent was a streamer that reads directly from the media, either by USB or network, as Sonos does. In that case the hard drive cannot make any difference any more than two different copies of a CD would (not counting those strange Japanese premium CDs that Andrew mentioned in his blog a while ago).

The problem with streamers is that some of them require server side software in order to read the data, in which case you can't compare them directly to something like Sonos because you can't be sure what the server is actually doing to the data, if it's just sending it as data then fine but if it's converting to some intermediate format and therefore possibly messing with sample rates and so forth then there's a whole new set of variables to consider.

Either way though, I'd still argue the underlying hard drive is NOT the source. It's whatever is plugged into the amp (or DAC) that is the source.

 

[idc]

Andrew Everard:

storsvante:Unfortunately it's exactly the same as saying 2 brands of hard drives will produce different sound -- and if someone is prepared to suggest that, I may just give up.

Actually, now you come to mention it...

My second or third post on this forum was asking whether the PC/hard drive had any influence on sound since all it did was hold a music file, in my case for itunes and any influence on sound would start with itunes, its EQ settings etc. The answer was, no the PC/hard drive will not have any effect. Then a while back I put on a similar post on another thread to answer a similar question, but this time Andrew responded and suggested that that was not really the case. Now here he is doing it again, but with a bit more detail. I presume that is because more listening and work has gone into the subject. It is after all likely to be very controversial, as storsvante is already showing.

Storsvante, when you say you eat stuff if you get proof, what will constitue proof to you? Expert review, peer testimony, you going and trying it out yourself or something else?

 

[Anonymous]

andrew has to prove it he said it.

 

[idc]

Short of us all getting to go to What Hifi Towers to actually listen to memory sticks, how do you expect Andrew to prove it?

If you go to the doctor and they give you some medicine, do you say I will not take it till you have proved it works? Or do you trust them as an expert and take the medicine?

 

[PJPro]

idc:
Short of us all getting to go to What Hifi Towers to actually listen to memory sticks.....

Hey, that sounds like a quality way to spend an afternoon!

 

[Anonymous]

Looking at the problem somewhat differently: For some hifi parts (in particular loudspeakers, amplifiers) it is clear that there are differences. And there is a lot of equipment that is seriously good, many of it expensive. Of course a setup appreciated by a heavy metal fan will differ from a classical music fan. All very clear. But for other parts I think there should not be a difference, unless something is really wrong (like bad contacts, bad ground). Usb cables and sticks, harddrives for instance. But also (loudspeaker) cables. The perfect cable to me is still one that does not get in the way between amp and speaker. If a good setup needs tweaking using a certain cable (or USB, or harddisk) that alters the sound there is something wrong with the combination I'd say. This does not mean that I might even prefer that sound (that is indeed possible) but it does not change my conclusion that there should not be a difference between two good cables unless at least one seriously alters the sound in a way not intended by amp and source designer. This also means, for instance, that I will never buy those ebay cables that get such good reviews on this forum: I just do not trust them and suspect that if many people are hearing a difference that the cable does something with the sound, and even if they prefer it I tend to stay suspicious. There is a distinction between discussions about differences in types of sound, sound quality and suitability for certain types of music on the one hand, and faulty and non-faulty functioning parts on the other. So I hope we will not get discussions on audiophile grade usb sticks or hard disks.

 

[8009514]

PJPro:idc:

Short of us all getting to go to What Hifi Towers to actually listen to memory sticks.....

Hey, that sounds like a quality way to spend an afternoon!

Not likely to happen though. Andrew has already said no when I suggested it yesterday.

 

[Anonymous]

idc:
Storsvante, when you say you eat stuff if you get proof, what will constitue proof to you? Expert review, peer testimony, you going and trying it out yourself or something else?

Apologies for slightly going off on a tangent but I'll answer the question.

Proof to me is a well conducted trial. There are many ways you can do it. But how about, for example:

a) take 2 identical system setups which are each capable of reading audio off an USB stick. Let's say 2xNaimUniti and let's call them A and B.

b) feed speaker output from both Naims into a switch which controls output to the speakers. Give the switch to the control person (Mr E?) and put a blindfold on him.

c) copy the same audio files onto a couple of USB sticks. Let's say we want to compare 2 brands, Foo and Wiz, so let's use 2 sticks from brand Foo and 2 sticks from brand Wiz.

d) the experiment begins... Pick 2 sticks from the pile at random. You'll either have 2xFoo, 2xWiz, or 1 Foo and 1 Wiz. Load them into the Naims, press play, and let the test person listen and toggle back and forth between A and B to compare. Ask him whether he thinks they are from the same brand or not. Make note. Repeat 10 or 15 times.

e) if he's right more often than he's wrong, and this is statistically significant (let's not go into that), it implies that there are audible differences between Foo and Wiz.

Or, if this setup is too expensive/awkward, how about this (slightly less good):

a) single system, 1 stick of each brand loaded with identical data. Let the test person listen to both brands for as long as he wants, and switch back and forth, always knowing which one he listens to, until he has identified the differences (or given up!).

b) blindfold on

c) pick a stick at random. Play chosen parts of music (let the test person decide). Ask whether it's a Foo or Wiz stick. Repeat 10 or 15 times.

d) is he more right than he's wrong and is it statistically significant?

Just 2 ideas.

 

[idc]

Pete10:

........ So I hope we will not get discussions on audiophile grade usb sticks or hard disks.

Whereas I hope we do! My theory is that those who have heard loads of kit, especially high end and very revealing kit become so used to listening for differences that they find them easy to spot. Hence the What Hifi team report differences that others struggle to hear. I have read on another forum a thread where there was an agreement that you can 'train' yourself to hear differences between bit rates with practice. There are also those with golden ears, such as my wife, who can pick out differences between bit rates and kit better than I can.

If I cant hear a difference, but another can and I then ask for proof that there is a difference beyond their own testimony, what is left?

 

[Anonymous]

idc:

If I cant hear a difference, but another can and I then ask for proof that there is a difference beyond their own testimony, what is left?

Put them to the test. See above.

 

[idc]

storsvante:

idc:

If I cant hear a difference, but another can and I then ask for proof that there is a difference beyond their own testimony, what is left?

Put them to the test. See above.

But who is going to do the test and provide testimony that you trust? Will What Hifi do since you have already appear to have rejected Andrews listening abilities?

 

[Anonymous]

this is silly if you think the hdd/usb stick can make a difference you think pigs have wings and can fly sorry its true

 

[Anonymous]

idc:Pete10:
........ So I hope we will not get discussions on audiophile grade usb sticks or hard disks.

Whereas I hope we do! My theory is that those who have heard loads of kit, especially high end and very revealing kit become so used to listening for differences that they find them easy to spot.

It don't mean to say that they all sound the same - just that they should if they are working properly. I also do not want to insult Andrew and definitely not your wife by stating that they are liars and imagining things - that would be unwise . But no 'This USB stick has a warm and natural sound but lacks some clarity in the lower range' kind of discussions, and people that start selling USB sticks for hundreds of pounds/euros/dollars, please.

 

[Anonymous]

idc:

But who is going to do the test and provide testimony that you trust? Will What Hifi do since you have already appear to have rejected Andrews listening abilities?

I think the point is that it is easy to claim to hear differences, and particularly when someone throws the 'golden ear' trump card, it's difficult to put rhyme and reason against the claims however absurd they might be. But it need not be that way.