Last page...
There has been an awful lot of research and work done to combat hearing loss, not only at Cobham but at Racal Acoustics, Bose etc
http://www.northropgrumman.com/capabilities/anvic5/documents/vic5.pdf
Stick to physics my friend.You guys really need to understand the converter technology before making up these scenarios. As I joked why not scale the second of two 16bit converters to make a 32bit one? Dis don't woik. Neither does the other .
Don't be silly. We were using a 16 bit and an 8 bit 21 years ago to generate absolute accurate 24 bit output.. Quite trivial to do, in fact.
Now, some details:
1. We ramp 24 bits in about 3 minutes.
2. We used a 10 digit dvm as the feedback mechanism.
3. We used a lookup table generated using the DAC's and the dvm.
4. We filter the output big time, BW is under 10 hz. So essentially, we have about 23 days between level transits..
But back in the day, 16 bitters were not capable of steps at 24 bit accuracy, and long term drift was also a bit problem. Hence the 10 digit dvm.
What does interest me is the concept that dither is mirror image. For brick type bw limiting, the math may certainly support it. But the up front 16 bit conversion to digital involves truncation of information. Anything beyond 16 is gone forever. Using horsepower at D/A, calculating to produce the "lost" information, does not restore the signal integrity to the level of interaural timing humans are capable of. It's lost.
Doing lots of 32 or 64 bit DSP on a truncated 16 bit source signal cannot return that interaural information. So taking the lower residual 8 bits of a 24 bit stream and applying it to some dither function should only "fix" a band limited (and inversely, time limited) signal. ITD accuracy is not restored.
I don't even worry about 16 vs 24. I'm content with cd's as they are now. But that's just me.
32 bit...hmm, nanovolts, picovolts, I can't wait for femptovolts, those systems would really be revealing..
jn
Last edited:
I don't know why anybody (almost everybody) here comes so strongly to the defense of 16 bit CD quality. We have worked with it for decades, and many 'hi enders' find it somewhat wanting. With storage so cheap and easy, what is the problem with working with 24-96K as a minimum for 'best fi'? Are you trying to rationalize your CD investment, so that you don't have to upgrade your software? IF your CD software gets you to the store and back, (figuratively) who needs a 'performance car', as well?
You certainly do not speak for me in respect to categorization.
As I said, I am content. I did not say it is sufficient for the full task of soundstage accuracy. That said, I really love how some make up technical fiction to support reasons why we need 24/96.
I enjoy the lack of scratches and the fact that my turntable performs poorly when there are many potholes in the road.
jn
No, we're just trying rationaize rationality.
So far no one has demonstrated anything lacking with regard to 16 bits on the playback side, other than to throw up numbers game smokescreens.
And you're talking about 24/96 being a MINIMUM?
As for what "many 'hi enders' find," hi enders find all sorts of crazy things. Your point?
se
John/RNMarsh and everyone else,
would you folks be prepared to vote and take part in the small listening trial I linked to a half dozen posts or so above this one. A minority have already listened and voted.
What would be great is if you guys who have access to really top notch DAC and playback systems evaluate these and see what you make of them. They are all short files around 26 seconds in length. If you don't fancy voting then just try and describe what you hear.
Thanks
Not at all, just i don't agree it is a minimum. 16 44.1 is the minimum, on my point of view.
That's the same as me John. If I wanted to play these files through my main system it would be via the PC into the amp rather than some dedicated music server/DAC or whatever.
(That said, if you have decent headphones then they are still worth evaluating, even using your just your PC soundcard and so on)
Anyone else ? the more the better. Lets try and get some useful info on what you hear.
(That said, if you have decent headphones then they are still worth evaluating, even using your just your PC soundcard and so on)
Anyone else ? the more the better. Lets try and get some useful info on what you hear.
Mooly, not necessarily so. Most modern CD players have USB inputs for digital signals. I can select the source on my NAD C 565 BEE player, and that one was hardly a madly priced unit (I šaid €400, app. US$ 500 or so). My PC has a built-in audio section, amd it's no surprise that the usic suonds better via the NAD, which actually cost as much as my entire PC did (sans monitor).
All it takes is a USB-USB link cable. Inputs selectable from the player's remote too.
All it takes is a USB-USB link cable. Inputs selectable from the player's remote too.
But this is all 2 channel talk. Do you think 5.1 or 7.1 doesn't add to the party?
Come on, trivial? I was speaking in the context of off the shelf DAC's and the DIY community. Sorry, still friends?
Yep, trivial. But pretty much a dc application.
There was no offense taken..how could there be??Sorry, still friends?
jn
I agree , any and all system wide improvements are necessarily to discern them bits.
I don't think you need to go all out "anal" battery powered to split the digital and
analog "camps" for exceptional fidelity.
Running primitive analog speakers should be done with good old fashioned
analog amplification (below - real trafo and capacitors).
No class D , even for my bass reproduction - a nice 20th century 300W class AB amp will suffice
.. SMPS's for audiophile amplification - Blasphemy !!!
For the source - digital is cheaper , capable of far greater fidelity , and more fun
.OS
Attachments
No one of golden eared gurus will participate. There will be tons of excuses and few brute force arguments ("I don't have to prove anything; I trust my ears!")
- Status
- Not open for further replies.