item, if as you say "to make a computer a 7/10 (or better) transport in the grand scheme of transports in general .. [is] .. a surprisingly big ask, given the IT junk we're working with" why bother? Why not just go down the Linn DS/Naim NDX route, where the transport (ie streamer) and DAC are integrated and tailor made? Why try and fix junk?
Using a Mac Mini as a digital source
- Thread starter JTC
- Start date
Cheers - I use an HP something or other, a Sony VAIO and a Logitech Touch to stream to my HDR with indistinguishably superb results. I guess the iPads USB must be crap, or maybe the digits get buggered somewhere in the operating system. The Naim community say that an iPhone into a Naim DAC is pretty crap too, even though the Naim DAC gets the raw digital data, so there definitely seem to be some issues. Maybe Elias Gwinn could help cast some light on this.
Absolute garbage!
Unless, that is, you can provide detailed technical papers from recognised institutions ... oh why bother. You won't, because you can't, and it's against the interest of your business.
Extremely good question. The problem is that
1. You're then locked into something proprietary that
2. can't be upgraded and which usually
3. cuts corners with regard to parts specification because of mass production.
Also
4. Because of the way those products are sold and distributed, they're bad value for the customer: something like two thirds, or more, of the ticket price goes in the hands of wicked middle men like us.
So, yes, it's better to design something to do this job from the ground up. Problem is that low-end streamers are designed as lifestyle products without much respect for audio quality, and the bespoke jobs are overpriced and lock you in.
For those prepared to soil their eyes with the occasional sight of an operating system, there is an 'open-source' middle way: very similar performance to the high end devices (better parts vs board limitations), but completely open-ended: any DAC (easy to switch from AES/EBU to Firewire, for instance), any file format, any web radio station, any future upgrade, slick remote control with iPad and Android apps, and optimised for USB, which most bespoke transports aren't. Crucially, it's cheaper.
Equally (actually more profitably), my business is selling DACs. Bigging up converters as panaceas in web forums would be more lucrative, but untrue.
I spend time with every DAC customer trying to improve the performance of their transport: usually it doesn't cost them anything. Sometimes I recommend software I don't sell. Often, we lend converters and cables - without charge - to audition with the DAC so customers can make up their own minds, without pressure, about these 'controversial' questions.
In the majority of cases, they conclude that clocks and power supplies and cables and jitter all impact on performance in the digital realm. Welcome to the desert of the real.
We sell Benchmark DACs, for instance, but they appear to behave quite differently with different transports - in my experience, and everyone of my acquaintance who has made comparison. By even saying this, I risk opprobrium from the distributor, but - again - sometimes the truth is inconvenient.
sonddek
Trade: SUPATRAC
The endless shifting goal posts. What do your customers think about direct blind abx comparisons between HD and SSD?
I don't recall anyone saying jitter doesn't matter. Simply that adequate is perfect. How many of your customers are measuring jitter and identifying its audible effects in blind abx comparisons?
I put to you that the things you claim make a difference here on pfm, the things you can get your customers to hear by persuasion/suggestion, and the things you sell which produce an inadequate (bit-imperfect) solution are distinct, and the reason for your evangelical confusion.
I wish they had, but all I had to go on were the monthly mags, there was no internet back then...
sonddek
Trade: SUPATRAC
Amen
sq225917
Bit of this, bit of that
Item, It would be fair to state that I believe that bit identical transports can sound different. But I don't put the difference with the transport, i put it with the dac who is either able or unable to render the music accurately discarding the signal line noise/pollution/jitter or whatever else it is this week.
But there's no rabbit hole there. We can measure the signal noise, we can see the modulation of the power rail for the clock, for the dac chip, we cna see the pollution of the ground plane and measure the increased noise. There's no miracle here, just engineering, which you seek to obfuscate
But there's no rabbit hole there. We can measure the signal noise, we can see the modulation of the power rail for the clock, for the dac chip, we cna see the pollution of the ground plane and measure the increased noise. There's no miracle here, just engineering, which you seek to obfuscate
Just for the record (because I can't believe you believe it) are you saying that any form of jitter than doesn't entirely obfuscate or corrupt bits is irrelevant? That there are only 10 kinds of transport: bit-perfect and bit-imperfect?
I don't know whether they're performing ABX comparisons: I leave it entirely up to them, in the comfort of their own home, hundreds of miles away, to judge for themselves. That's some mind trick.
Absolutely - you find yourself reluctantly on the same page again: signal noise, power rail and ground plane pollution, jitter - many would add EM/RF contaminants - all straightforward, known, measurable stuff that makes transports behave differently. No obfuscation. Just engineering problems and engineered solutions. Without recourse to sonddek's charming voodoo elves.
And - duh - obviously DACs handle crud differently. But there's no escape from the garbage in, garbage out maxim: the problems created by a transport are too large for any DAC to just make go away: this from wide experience. Anything else is manufacturer hype. I'm talking about tackling the problem at source, not allowing it to roam the system and trust in half-assed damage limitation.
Once more the usual garbage and unsubstantiated crap. In your case it seems to be poorly understood concepts in and garbage out
I don't know what you're trying to gain by continuing with this - clearly not much in the way of credibility, anyhow.
If you want to convince people then forget the homespun comfy speeches and do what has been asked so many times, give references, and I mean proper references.
Darth Vader
From the Dark Side
Are you aware of accuraterip?
http://www.accuraterip.com/
Transports ARE different.
Drive accuracy
http://forum.dbpoweramp.com/showthread.php?23074-CD-DVD-Drive-Accuracy-List-2011
Cheers,
DV
It's always interesting to realise how computers are fallible, and how readily their complexity is underestimated by many. But, the kind of generic, characteristic differences clearly audible between computers can't be explained by occasional bit errors: only by electrical factors (including jitter) during real-time playback. In the time-suspended storage domain, bits really are bits.
Once more the usual garbage and unsubstantiated crap. In your case it seems to be poorly understood concepts in and garbage out
I don't know what you're trying to gain by continuing with this - clearly not much in the way of credibility, anyhow.
If you want to convince people then forget the homespun comfy speeches and do what has been asked so many times, give references, and I mean proper references.
You keep asking: the net keeps supplying. Noise impacts on clock performance: check. Computer modifications reduce jitter: check. DACs sensitive to input jitter: check. You're asking for proof that the sky is blue and people keep handing you photographs.
The trouble is that noise/jitter/contaminants are different in every system, and null testing of subtle, noise-related phenomena (and cables) is difficult. These are fairly subtle effects, easily masked by off-the-shelf inversion algorithms.
However, the difference between a standard computer and, say, the top-end Auraliti (choosing one we don't sell), is greater than the difference between 192kbit MP3 and uncompressed. When the difference is so gross, and so well attested by so many, you have to be perversely bloody-minded not to get it.
Poor rips aren't exactly hard to spot.
It's basically a problem of reading red book discs on drives designed for CD rom, which has a far more robust form of error correction built in.
avole: tell me what you want. What would it take?
Here's the problem:
When you see data, you say: “that's not audible”.
When people talk about audibility, you switch to: “show me the data”.
Both are forms of evidence; both need contextualising. But both are relevant. You can't discard one or the other to suit your prejudice. In this instance, they generally corroborate, as well as you can expect.
For anyone with an iota of audio nous, or hearing - but, mainly, without an axe to grind - all it takes is: “Reducing noise and interference in a digital audio transport improves the system.” Bing! If you don't get it from that, there's no helping you.
Here's the problem:
When you see data, you say: “that's not audible”.
When people talk about audibility, you switch to: “show me the data”.
Both are forms of evidence; both need contextualising. But both are relevant. You can't discard one or the other to suit your prejudice. In this instance, they generally corroborate, as well as you can expect.
For anyone with an iota of audio nous, or hearing - but, mainly, without an axe to grind - all it takes is: “Reducing noise and interference in a digital audio transport improves the system.” Bing! If you don't get it from that, there's no helping you.
Define 'sound better'...
Also, if this 'noise' can somehow corrupt the resultant analogue signal, why does it not make the computer crash when this 'noise' corrupts the workings of the O/S?
Or are we talking some kind of 'magic', discriminating noise?
sonddek
Trade: SUPATRAC
I'm amazed by these accuracy figures. I can't believe that there are so few imperfect rips. My sister's CD collection alone would account for all the imperfect rips that dbpoweramp has received. Dbpoweramp users must be a fastidious bunch.
The fact that there is so little deviation in the error rates across different drives suggests that all transports are the same. Film at eleven.