The problem is, TL, you're prepared to contribute plenty in terms of talk or graphics ripped from the stereophile site, but absolutely nothing in terms of practical help.
w
w
So, build a USB bitstream splitter. Can probably do this with a cable, maybe you need a bit of hardware. Make one HiFace passive in terms of not talking to the PC, but leave the other active so that the computer recognises it. Trick the second (passive, parasitic) device into receiving the info from the computer if necessary. Connect both to DACs, one with attenuators in circuit, one without. Connect the DAC outputs to a differential amplifier. Tweak the amplitudes if necessary. See if anything comes out. Not exactly straightforward to arrange, but I don't see anything wrong with it in principle.
Or you could do your spectrum analysis with an analog (scanning) type analyser rather than an FFT. Might show up something.
Or you could just do the double-blind test and forget the instrumentation.
I wish I'd never got into this. I don't care about hi-fi that much, I think it's just a pacifier for people who can't play themselves.
w
Or you could do your spectrum analysis with an analog (scanning) type analyser rather than an FFT. Might show up something.
Or you could just do the double-blind test and forget the instrumentation.
I wish I'd never got into this. I don't care about hi-fi that much, I think it's just a pacifier for people who can't play themselves.
w
I think here is a rather subtle point. I think everyone would tend to agree that 'no analogue output difference' means 'no audible difference'. However SY is interpreting 'no analogue output difference' as 'no difference in FFT'. The map is not the territory though - the analogue output clearly (its trivial to show this) can change with no discernible change in the FFT. Perhaps this is something that SY is unaware of as he does put great faith in his FFTs.
Nothing wrong with long and woolly, like Dr. Who's scarf you mean?
OK, I can follow this but only up to a point. Presumably you do not spend time listening to what amps and speakers do to pure tones (or do you?). So then when listening to music how do you know its the 'harmonic structure' that you are listening to and not something else? In my understanding, its not harmonic distortion but intermodulation distortion that dominates when real music is playing, as there are so many frequency components present. So its this jump, from 'I know what my amp does to pure tones (at various frequencies and levels)' to 'I know that's what's responsible for my amp's sound' that I can't yet make. There's some missing step in the chain of logic for me.
Something you didn't mention in your answer is how Gary goes about making his system sound cleaner. Has he correlated 'clean sound' with 'absence of higher harmonics in FFT' for example? If he has, that would be very interesting because I've made my system sound cleaner without a single FFT being necessary. The reason I warmed to your post about Gary's system is it coincided with a set of mods I included in my system which resulted in solo violin sounding so much sweeter, just like you said in your impression of Gary's system. Yet I really have no idea what the 'harmonic structure' of my amps is.
And here lies a number of rubs.
Does your system faithfully reproduce? By that I mean does the violin sound like a live violin?
Or does it more correspond to your idealised sound of a violin?
Sweeter to me indicates layers of harmonics - are they real, or the result of the many parts of the reproduction chain? How much sweeter would it have to become to be saccharine?
Sweeter is one of the terms that I would refer to as "preference" language.
Clear - thats not preference, its perception
Veiled - again, perception
Warm - preference.
Maybe I'm wrong but again its a matter of having a language to describe universally what we hear and perceive.
I've never heard Gary's system and are never likely to. Most of the music I listen to is electronically derived - I'm a child who grew up in the 70's and I never got into the acoustic and classical music of my parents. So ALL my music has been harmonically manipulated!
I'm still concerned about clarity of separation of instruments and the like, but its increasingly difficult in the context of modern (largely electronic) music to get hung up on the minutiae of FFTs ...
Rambling diatribe over - back to the battle gentlemen.
Last edited:
Although I've never heard a live violin playing in my (current) room, clearly no, what comes out of my system does not sound like a live violin in my room. That's because what comes out of my system also includes the acoustic of the recording venue. If I had an anechoic recording of a violin, then perhaps it would but I'm not really interested in such levels of 'fidelity'.😀
As far as I'm aware, I have no idealised sound of a violin. I'm no Platonist in terms of violin 'templates'.
In this case, I wasn't particularly clear in describing the sound as it was an aside, not really germane to my argument. Better to say it had fewer artifacts of the reproduction system associated with it.
It became 'sweeter' by the removal of sourness, harshness, so no, it could never become saccharine by the same mechanism.
Sure, so whilst clearly preferring the 'new' violin sound, its not preference language in this case. I agree with your characterisations between 'preference' and 'perception' btw.
Yep, we lack this. It would hasten the science of audio to have such a universally agreed upon language for talking about such things.
BTW, here's a quote from the latest person to try attenuators which is pertinent to the discussion above re the sound of reproduced music post here M2Tech HiFace 24/192 USB->S/PDIF penstyle interface
And my reply
And my reply
Let me re-phrase - you have shown two FFT plots from different devices & as a result declared them to be sonically the same, no? If you could stop playing cat & mouse this could be less tedious!
I know, I know! 😀
And not only that, you have really missed the point. Are my posts that opaque? (it's possible) I did not ask for a simply a lower noise floor. I'm asking for a test that SHOWS jitter between -140 and -120dB, but not above -120. That is the type of test that could invalidate SY's. Where is that test? If you've shown it to me and I haven't seen it, I apologize. Please point it out to me again. If it's marked up with arrows and notes, it would really help. Of course it would have to be audible to make a difference.
I take your point about the noise floor and it would be nice to do better, but that's probably beyond the capability of most of us without real lab equipment. (I have seen it done, tho). Below is a 16bit measurement I did last January of the stock DCX. Noise floor looks like it's down around -137dB. But the spurs still rise above -120.
|
Attachments
I wasn't asking the questions.
You're trying to get me to say things that I do not believe or agree to statements worded so badly as to be meaningless. Sorry, but I'll have to disappoint you.
You're trying to get me to say things that I do not believe or agree to statements worded so badly as to be meaningless. Sorry, but I'll have to disappoint you.
Good old toothy Tom Baker, one of the greatest Doctors!
No. I dig the sweeps. 😉
Experience. I know people who are much better at it than I.
Don't get carried away with the FFT thing. It's just a tool. But it is a very powerful tool and so much more useful than the old standard %THD@1Khz.
There seems to be the idea here that folks how trust measurements think that "If it measure the same, it is the same." That unfair. I sure don't think that. That's like saying "this red brick weighs the same as this grey stone, therefore they are the same." I don't think anyone here would go that far! They may be the same weight, true, but they aren't identical in other aspects.
Taking that into the realm of FFT and jitter. If there is jitter, you will see it as side bands on the FFT. It's that inharmonic distortion that we don't want. If it changes, so will the sidebands. That's not likely to tell you how it sounds, but will tell you if the jitter is different, absent, present, etc. That's all the test is looking for. Is the jitter different.
OK, I'm not going to quote the post again, and with no insult intended to the OP, it is representative of the worst kind of anecdote.
As the developer, jk, it is incumbent on you to make some effort to sort the wheat from the chaff.
Just because somebody writes you an endorsement you are not compelled to believe it or pass it on.
This guy has a card with a standard level output. There is every chance that if there is a difference in sound, it is actually a degradation, which is exactly why I took issue with you over this subject originally.
Didn't it ever occur to you that this was a possibility? There are a lot of impressionable people out there. It may be possible to take advantage of them, but it's not right.
w
BTW, I have a quantity of guaranteed lucky numbers for the National Lottery available for sale. Prices are negotiable. PM me if you are interested.
- Status
- Not open for further replies.