XXHighEnd

[PeterSt]

although I'm not sure this is correct as I haven't received a response to my question regards pricing,

 

I am sorry. Completely missed the post (but found it now).

 

Since it is 72 euro your earlier suggestion of 65GBP will be about right I guess.

 

Peter

 

[The Computer Audiophile]

Hi djp - I have to disagree with you on this one. In my opinion having a closed mind that only considers what a certain group of people considers "relevant methods" does not lead to progress. I believe we should think outside the measurable box. When we think we hear something we should try to measure the phenomenon. If we can't measure it we should not discount it. To discount it totally would really be argumentum ad ignorantiam (http://en.wikipedia.org/wiki/Argument_from_ignorance). Or, in other terms absence of evidence is not evidence of absence.

 

I do think objectivity and subjectivity have their places in good discussions. To use the oldest one in the book -> the earth was considered flat for quite some time. If people are more drawn to Hydrogen Audio and that style of objective conversations only I have a feeling they will only be happy on that forum. I will not turn Computer Audiophile into a big spreadsheet that is based on formulas where the top products can be arranged simply by statistics. That's totally unenjoyable to me and is quite closed minded even though it appears to be the most accurate route to take by those who value current measuring methods over all else.

 

There is no right or wrong on this one. If you prefer HA please contribute to that site. If you like Computer Audiophile please contribute to this site. If you like both please contribute to both.

 

 

 

 

[DavidJPettifor]

PeterSt wrote:

I agree fully, though Hydrogen's are not my way.

 

Care to elaborate Peter? I'd be interested to know what you mean by "not my way".

 

PeterST wrote:

Besides that, and possibly more important, it could be so that we *all* want to know what is going on here, and coincidentally that includes me.

 

A fair admission. Same here.

 

PeterST wrote:

I am sorry. Completely missed the post (but found it now).

 

Since it is 72 euro your earlier suggestion of 65GBP will be about right I guess.

 

No problem Peter and thanks for clarifying the cost.

 

--

djp

 

[The Computer Audiophile]

"The fact is that all this is measurable and provable so measure it and prove it or drop it. Proof is needed or this is simply meaningless conjecture. It really is that simple.

 

Ash under instruction"

 

Ashley - You commenting by proxy is actually hurting your arguments. Demanding measurements and scientific proof is antithetical to your style of posting here. You frequently provide information and say it is true because an unnamed source told you so. I highly doubt you'd be taken seriously if you presented a paper at the next AES conference and after every point you said, "A top engineer told me this so it is true," This leaves no room for anyone to contradict you because it isn't you "speaking." It is a discussion by proxy and leaves you an easy way out.

 

I've received many emails of late regarding your method of discussion. It is pretty similar to slapping someone in the face and getting them to thank you for it through the use of creative words. Please take this all in the spirit in which it is intended. I know you supported the site through advertising and many positive emails to me personally. But, I also have to look after the readers as a whole and keep the integrity of the site intact.

 

 

 

 

 

[DavidJPettifor]

The Computer Audiophile wrote:

I do think objectivity and subjectivity have their places in good discussions. To use the oldest one in the book -> the earth was considered flat for quite some time. If people are more drawn to Hydrogen Audio and that style of objective conversations only I have a feeling they will only be happy on that forum. I will not turn Computer Audiophile into a big spreadsheet that is based on formulas where the top products can be arranged simply by statistics. That's totally unenjoyable to me and is quite closed minded even though it appears to be the most accurate route to take by those who value current measuring methods over all else.

 

There is no right or wrong on this one. If you prefer HA please contribute to that site. If you like Computer Audiophile please contribute to this site. If you like both please contribute to both.

 

Nicely put Chris. Yes, I do think both forums are at either end of the scale, which is why I used the HA link. I suppose I put forward where my hat lies most, the objective crowd, though I'd be lying if I didn't add that my choice of current components was governed by reviews which would probably be more suited to the subjective camp (and also the CA crowd let it be known! Choice of components that is). I'm also a lover of tubes, so with that omission, all objectivity goes out the window!

 

--

djp

 

[The Computer Audiophile]

Thanks for the response djp! I like the honesty and the comments from a clearly reasonable and smart reader.

 

[DavidJPettifor]

The Computer Audiophile wrote:

I've received many emails of late regarding your method of discussion. It is pretty similar to slapping someone in the face and getting them to thank you for it through the use of creative words. Please take this all in the spirit in which it is intended. I know you supported the site through advertising and many positive emails to me personally. But, I also have to look after the readers as a whole and keep the integrity of the site intact.

 

Hmm, I have to disagree with you on this one Chris! Ash, tfarney et al are positive reasons why I enjoy CA. I believe they actually do give balance to CA, and, possibly contradicting my previous post here, do prevent CA from becoming the subjective equivalent of Hydrogenaudio (which is a good thing in my book).

 

I realise as the web site creator, you have to do a fine balancing act, to keep both camps happy. Possibly the compromise to my previous "think this forum could benefit from a similar principle as that held over at HA" remark. I think your opposition to this was correct and trying to embrace both objectivity along with subjectivity is the way to go.

 

Not easy though.

 

--

djp

 

[Ashley James]

Chris

I'm sorry you felt it necessary to say what you did, for you know that I rely on three primary sources for guidance.

 

These are: Dr Chris Smith A Research Head at Roke Manor, which is a UK division of Siemens. Martin Grindrod BSC Electronics my business partner and ex Smith Industries and the design team at www.LS-design.co.uk who make our stuff.

 

I don't believe this reduces my credibility, I'd prefer to think that it shows how careful I am to try to stick to simple provable fact.

 

Ever since I returned to audio in 1985 I've seen good enthusiasts and engineers desert audio out sheer frustration with the subjective side of it and the reverence shown to various Gurus who may not have formal qualifications, leave alone experience in a disciplined working environment amongst peers.

 

Ashley

 

 

 

[PeterSt]

Care to elaborate Peter? I'd be interested to know what you mean by "not my way".

 

At HA there is no room fo discussion or new ideas for that matter, just because one has to scientifically prove his "statements" first. Nothing wrong with that by itself, but nothing can come from that in my theories, and all is as cold as stone because of that. Please let me suggest that CA is the 180 degree opposite of that, and this is a good thing for feeling comfortable and cosey for many people.

 

This may be one of the best examples of what I just said. Not because it is indirectly about me, but about the way it works overthere :

 

http://www.hydrogenaudio.org/forums/index.php?showtopic=64392

 

This thread was closed within the hour. Just closed. I don't know what rules were disobeyed here in order to justify closing the thread, but if you'd take the time to read it, this is only about improving Foobar, of which long gone was defined it can't because nobody can come up with a proof of that. SO FAR.

Now let's say we're all hoping for the real proof at last, never mind we can hear it already. It's not enough. Not for many, and at least not for myself.

 

On this matter I am completely honest in saying that I can't underatand myself, no matter I *can* control it. But this is by means of much much experience (mostly from users), and the data coming from that experience may or lead to "science" itself, or may lead to a new means of measuring which latter will be science in itself.

 

The latter is quite another subject which I brought up a few times at CA, just telling (and not hard to prove) that the standards for measuring DACs are not correct in the first place. Why nobody comes up with the real way to do it ... I guess it is as hard for everybody as it is for us to define it should be done.

 

Peter

 

 

[PeterSt]

My telling about the measuring may look like a kind of self defense, but it really is not.

 

This is what I wrote on my own forum at last Janaury 14 :

 

All,

 

Since Januari 6 (which is a whole long week already) I have this still unopened box with 10K worth of audio measuring equipment;

It has been a dream for me to have such equipment, which will allow me to see what actually happens within XX at certain settings. Also, it will allow me to find theoretical best settings, or even other influencing means. So, be ready for that.

 

I post this in advance to ensure myself doing it in the next coming days or weeks, and to at last prove that what we hear is no placebo.

Right now I don't know what I will find or where to find it. Part of it should be in the digital domain hence jitter, part of it will be in the analogue domain only (right behind the DAC).

 

The reason to at last buy this equipment originates from the Phasure NOS1 DAC which should be top of the bill, and which should go along with some real THD etc. data. But of course first there will be the process of improving by means of measuring and applying changes.

 

Whether it be XX or the DAC, you can bet that I will be measuring different things opposed to what is commonly done and accepted;

The fun is, that by now truckloads of experience tell me/us what influences how, and what we like and do not like. For example, I could grab an old XX version of which is appreciated it doesn't sound good, and compare with another of which is known it sounds good. From that possibly the properties can be derived which make what happen.

 

But now comes the combination ...

 

What is currently done via software so preciously and which is very fragile and far from absolute science, most propably and hopefully can be mimiced by hardware. If all is just about less jitter, we're done very fast. But I don't think it is and other influences may play a larger role than we expect. This is why I will be measuring quite different things than commonly done.

If all works out as I expect, think of a Q1 button on the DAC that changes sound ...

 

So far for now.

Peter

 

(Measuring XXHighEnd ...)

 

[BEEMB]

 

Peter,

 

I had no idea that you were infact using 1,000's of dollars worth of equipment to measure the results of your programming.

I too saw the thread over at Hydrogen Audio and I believe that sometimes, your English is a little difficult to understand. That's not an insult to you by the way - for what it's worth I read and re-read your posts to try to understand them as best I can and thankyou for your contribution to the site.

 

Ash, why the instruction to "drop it" ... this is an open discussion which is certainly producing more than which I intended with my original post simply encouraging people to try XX and see if they can hear the difference. I'd like to see the results of some of Peter's tests before we finish the discussion - I'm simply eager to learn more.

 

I'd still urge anyone who hasn't tried it, to do so. Maybe those that have could post their comments here.

 

 

 

 

[tino]

"Or, in other terms absence of evidence is not evidence of absence."

 

This is perfectly put Chris. Facts are all very well in an argument of this kind (and actually there look like a lot of facts being presented from both sides), but some people on here appear to believe that unless we know the scientific reasons behind something NOW, then that something can not be true or exist. If everyone had always thought like that, we wouldn't even have sound systems full stop..........

 

 

 

 

 

[DavidJPettifor]

At HA there is no room fo discussion or new ideas for that matter, just because one has to scientifically prove his "statements" first. Nothing wrong with that by itself, but nothing can come from that in my theories, and all is as cold as stone because of that. Please let me suggest that CA is the 180 degree opposite of that, and this is a good thing for feeling comfortable and cosey for many people.

 

Thanks Peter. People have commented that english does not appear to be your first language and a few takes are needed to understand exactly where you are coming from. I got that in one.

 

Again, thank you.

 

--

djp

 

[glt]

Inside the computer, when dealing with digital files, jitter and noise does not matter because the software and the hardware are designed to tolerate those things (error checking, design tolerances, etc). Besides, data buses run at 1GHZ which is several orders of magnitude higher that audio data frequencies so designs have to be more robust. The computer will never read a 1 as a zero because imagine what would happen if your spreadsheet pulls a file and misses the MSB of $1,000,000. As one poster said above, jitter only matters in the DA conversion. So if you trace back, the digital file is the same everywhere except when it is converted to SPDIF (and equivalent).

 

Additionally, what is the noise spectrum of the PC PS? and how does that spectrum changes with load? How was the PS designed? What is the optimum operating condition of the PS?. The Power System in a Motherboard is fairly complex. If you look closely, there are a bunch of local regulators and depending on the noise rejection and noise of the local regulators, it may not matter what happens in the PS. I've seen many posts where people change the PS in their audio equipment with mV noise figures to nV noise figures but the local regulator still generates mV noise and they claim better sound. But how can this be? the noise is dominated by the local regulator. The sound card may have its own local regulator too. For a meaningful measurement you will need to put a scope at the output pin of the local regulator of the sound card and then measure the noise (ripple) depending on processor load. But does this noise matter? The chip that is feeding on this power may be bypassed with multiple capacitors and the input pins may have noise rejection rations of tens or even 100 db. So does it still matter?. OK we can try to access multi-mega buck jitter measurement outfits and measure the output of the soundcard, but does it matter? Well, we are back to how good is the receiver.

 

 

 

[BobH]

It is difficult to make comparisons between 'static' data and data 'streams'. When static data is received it can be checked and, if found incomplete, can be re-requested. This process is very difficult, if not impossible, to achieve with a data stream.

 

The most extreme example of this can be found with internet transmissions. If a packet of static data fails to reach its destination intact then it is simply re-requested - you see nothing. If a packet of streaming data is not received correctly then the video stutters, or the song skips - you will notice.

 

In other words once the audio data blocks have been pulled from the hard drive and re-assembled into a stream of audio data, it becomes an 'on-demand' service - it stops for nothing. If something subsequently happens to it, there is no gong back and starting again - I guarantee you wouldn't need measurements to tell you about that, if it was happening

 

[glt]

I think you are talking about "packet" based protocols. If the packet is not received, it is resent. But packets have a lot of data and the packet protocol knows nothing about the data inside the packet. Even a stream is made of packets. So for example, you use an Airport Express, the data is send in packets. If your network is slow, you will miss a packet (and you are late in resending) and you will hear a BIG gap. But having used AE for many years, I've yet to see (hear) random droppings of packages. If you loose data in your SPDIF interface, then there is something wrong with the hardware.

 

[Roseval]

This process is very difficult, if not impossible, to achieve with a data stream.

Streaming AV over internet is not different from sending any other kind of data. It are all packages and none of them are a real time stream.

There are two major differences when sending ‘static’ and ‘dynamic’ data over the internet.

When you download a file, you have to wait until the entire file is received.

In case of streaming AV it simply starts when the buffer is sufficiently filled. Due to the buffer, it is a quasi real time stream.

Another difference is the protocol. Normally TCP is used. It is very strict, has all the error correction options mentioned. AV is often done with UPD. It is more efficient at the expense of the error correction. If a package is dropped, well it is dropped.

 

 

[BobH]

Nope

 

What I'm saying is that error checking streaming audio is difficult, if not impossible to do.

 

If a piece of data from your spreadsheet example is received incorrectly, it will be noticed because of the error checking intrinsic to the data transmission. Let's say a number representing the volume at a given point in an audio stream is 25000. Because the audio stream is 'shit or bust', ie no error checking, if one of the bits in that number gets mangled it could arrive as, say, 24500. There is no way to check it, therefore that is what you get. This is extreme and may be unlikely to happen but it demonstrates my point. Everything will be in tact, bit wise, but errors will have been corrected by 'interpolation' - for want of a better word - rather than going back a getting the value again.

 

[Roseval]

I've seen many posts where people change the PS in their audio equipment with mV noise figures to nV noise figures but the local regulator still generates mV noise and they claim better sound. But how can this be?

I like your analysis of the PSU thing. It indeed far more complex than changing a switching power supply ( suspect by default in audiophile land) by a linear one. I’m afraid your analyses and your implicit conclusion are true.

 

On Hydrogen you will be moved to the trash bin if you claim a better sound due to another PSU without a proper ABX.

On AudioAsylum they will ask if you can supply the schema and the components used.

 

In case of a laptop you can test it a bit.

Play a song with the PSU in place and play a song when running on battery power only.

 

[glt]

Fair enough, TCP has error correction, UDP does not. But how different playback software sound different if delivering "bit perfect" data? The movement of data inside a computer is handled by the OS. The application accesses these facilities through and API to the device drivers. Error checking or lack of cannot be controlled by the application. Suppose we use a sound card to output audio through spdif. Application A and Application B will use the same device driver to write the file to output buffer. Any error correction or lack of (or any additional processing of the bits) is at the device driver level.

 

[PeterSt]

Any error correction or lack of (or any additional processing of the bits) is at the device driver level.

 

This is why MS for the PC "invented" Vista/WASAPI. Drivers supporting WASAPI should

 

a. Allow Exclusive Use to applications requiering it, meaning if A comes first, B is rejected throughout the life time of the A object;

 

b. Not mangle the data by any means (the OS ahead by passing through the bits around any "KMixer" like process inherently needed when A and B had to be serviced at the same time).

 

This by itself is just a proven concept, with the kind of downside that it does not allow resampling, or IOW an output stream of 16/44.1 cannot be played by a DAC supporting 24/48 only (the "High Definition Audio" of 95% of laptops and mobo chips as the first example).

 

[Roseval]

Application A and Application B will use the same device driver to write the file to output buffer

If this is the case and no other intervening activity like SRC is done by A and B, you might indeed wonder if there is any difference in sound quality other than in the ear of the beholder.

 

 

 

[BEEMB]

 

Roseval,

 

Just been looking at your website which I've stumbled across before.

 

Quote "Quality is hard to define but we recognize it when we see it. Likewise if we hear it. It is very subjective so it is up to you to decide what suites you best."

 

Says it all really.

 

Matt.

 

[Roseval]

b. Not mangle the data by any means (the OS ahead by passing through the bits around any "KMixer" like process inherently needed when A and B had to be serviced at the same time).

No, WASAPI plays what you feed into it. You might mangle the data to your hearts content, implement your own SRC, etc. before you call WASAPI

 

 

[Roseval]

Says it all really.

Thanks

Indeed, (sound) quality is about what I like