Strongarm
Strongarm's picture
Offline
Last seen: 12 years 2 months ago
Joined: Jun 6 2006 - 12:30am
USB and high-res limitations
jackfish
jackfish's picture
Offline
Last seen: 10 years 5 months ago
Joined: Dec 19 2005 - 2:42pm

what limits it to 16-bit/48kHz. There are other USB DACs which will only do 24-bit/96kHz. I will have a Schiit Bifrost Asynchronous USB DAC which will do 24-bit/192kHz in USB.  Its all in the implementation.  Now whether you can modify the Isis to do 24-bit/192kHz is out of my realm of understanding and it would seem that it might not be easy to do. You might need a dedicated DAC with a full array of inputs and/or USB 2.0 implementation which handles 24-bit/192kHz.

Strongarm
Strongarm's picture
Offline
Last seen: 12 years 2 months ago
Joined: Jun 6 2006 - 12:30am

Thanks Jackfish.

I thought as much. I'll stick with CD res for the moment.

Incidentally, do you have any thoughts on the benefits of high res vs CD res? As Art Dudley mentions in one of his recent columns, the Nyquist Theorem would suggest that higher frequency would not provide any benefit. And yet many people feel that is does.

Regards,

Grant

Jagatai
Jagatai's picture
Offline
Last seen: 12 years 2 months ago
Joined: Jan 14 2012 - 9:35am

 

I have just recently started experimenting with hi-res files and my experience is far from conclusive.

 

A year and a half ago, I re-ripped my CDs from 128 Kbps AACs to 256 Kbps AACs.  While much did not seem distinctly different, there was a noticeable increase in the clarity of certain high frequencies.  (Note I am 46 and my high frequency hearing ain't so great.). After that I started re-ripping the CDs to apple lossless without much noticeable improvement, but it made me feel better and I had the hard disc space to do it.

 

Recently I upgraded my speakers and started exploring hi-res files.  I got a couple of blu-ray audio discs (Naxos' "Dvorak symphonies 6 and 9" - 24/88.2 and 2L "Grieg piano concerto" from rolls performed by Greig and Granger 24/196). I have also downloaded some FLACs at varying resolutions of some John Coltrane and some bluegrass interpretations of Bartok.

 

To my untrained and unexperienced ear, the higher resolution audio does sound better, but it is relatively subtle and I can't discount my own subjectivity in the assessment.  Because I know I am listening to a higher resolution version, I might be crediting the sound with finer detail and clarity that I dismiss in a known lower quality file.

 

I ended up making an accidental test when I made a 128 Kbps AAC file from a hi-res file and then unknowingly played the hi-res file back to back with the low re-file.  I had a vague sense that something was different, but I certainly didn't hear a huge reduction in quality.  It should be note, however that I was very tired when this occurred and I had been listening to music for some time already.

 

If you are able, you might try downloading a few samples (HD Tracks offers a free sample of five or so tracks if you register with them) and then convert them to a lower quality.  Run a blinded test, having a friend select the quality of the track played while you guess if it is hi-res or low res.  if you can accurately guess the quality, then it might be a good idea to further explore hi-res files.  If not, you can spend your money on greater variety of music rather than file quality.

 

If I had to guess, if there is something noticeably different for me between CDs and high res, it is the 24 bit depth rather than the sample rate.  I think I am able to hear the improved dynamic range better than I can hear high frequency fidelity that comes with higher sample rates.  Perhaps a good test of this would be to create a 24/44.1 file from a 24/198 file and a-b them.

Strongarm
Strongarm's picture
Offline
Last seen: 12 years 2 months ago
Joined: Jun 6 2006 - 12:30am

Hi Jagatai,

Your experience is very interesting. I think you make a good point in separating the frequency from the bit depth. Many of the articles I have read on the topic don't make this distinction. As I understand it, Nyquist's Theorem refers to frequency only, and 44.1kHz should be sufficient to cover the audible range; hence the adoption of this frequency in the Redbook standard. However, I have been unable to find any information on what motivated the choice of 16bit depth for CD (apart from the constraint of trying to fit 70-80mins of material on one disc). Do you know of any analogous theoretical support (i.e., a Nyquist-type theorem) for the choice of 16bit depth?

Regards, Grant

Jagatai
Jagatai's picture
Offline
Last seen: 12 years 2 months ago
Joined: Jan 14 2012 - 9:35am

Hi Grant,

I don't know a whole lot on the subject, but the following Wikipedia entry seems to cover it well.

http://en.wikipedia.org/wiki/Audio_bit_depth

Apparently for each additional bit of bit depth, you gain 6db of dynamic range.  I suspect you are right that 16 bits was chosen as a trade off between acceptable dynamic range while having a decent running time on the disc.  They may also have been hitting the limit of how fast the data could be read off the cd.  Obviously readers are much faster these days, but at the time attempting to pull 24 bit data off the disc without dropouts might not have been feasible.

Lars

GlacierJeff
GlacierJeff's picture
Offline
Last seen: 11 years 9 months ago
Joined: Aug 20 2011 - 9:48am

Any thoughts on the bifrost?

jackfish
jackfish's picture
Offline
Last seen: 10 years 5 months ago
Joined: Dec 19 2005 - 2:42pm

16-bit/44.1kHz. The digital files through the Schiit Bifrost sound better in my estimation than the CD playing on an Emotiva ERC-1.

HybridCore
HybridCore's picture
Offline
Last seen: 11 years 8 months ago
Joined: Jun 27 2012 - 8:49am

USB isn't limited to 16bit/48kHz. There are plenty of DACs out there that are capable of 24bit/96kHz (some can do 24bit/192kHz). There was an article I read though about if we can actually hear the difference between 24bit/96kHz and 24bit/192kHz. We usually can't hear past 20kHz, give or take a bit. Maybe there are some exceptions but none that I'm aware of. 48kHz is already far beyond our hearing. 96kHz too. Some people claim they can, but it's the hearing bias everybody has. Is it possible to get rid if it? If you're capable of, you must be either a robot or a very advanced version of a human (or an alien). Truth is, people can't really hear the difference. It might be nice for e-peen but we can't hear much of a difference past that, so 24bit/96kHz is more than enough. I'm not against people going for 24bit/192kHz. I'm just saying it's probably not necessary.

And first post. New here. :)

Demondog
Demondog's picture
Offline
Last seen: 6 years 11 months ago
Joined: Feb 22 2009 - 5:01pm

Welcome to the forum.

I prefer my audio files in as high a sample rate as I can get. I can't really say if it makes a difference or not, but it doesn't seem to hurt. When I have a choice between two Hi-Rez formats, I usually choose the higher sample rate, but not always. It kind of depends on how good I think the recording quality was to begin with as to whether I'll pay, what is usually more, for higher rate.

btw- I can't justify buying really expensive cables either. But then in informal testing with test tones, I can't hear past 11kHz, so I don't know why I spend more than $100 on an audio system anyway. haha 

HybridCore
HybridCore's picture
Offline
Last seen: 11 years 8 months ago
Joined: Jun 27 2012 - 8:49am

Thanks for the welcome.

I do too, considering the fact that we have the space, and most receivers (AV and stereo) these days can make use of 24bit/192kHz (which implies that anything better than that should be able to too). There's nothing wrong with it. It'll only become a problem if you're trying to save space, but by the time we need some more space, I'm pretty sure we would have bought another hard drive by then, considering how inexpensive they cost (1TB hard drives cost around $99.99-$109.99 these days, so about $0.10/gigabyte, although technically it's gigabit/gibibit because hard drives are actually measured in bits even if the computer says bytes. I have proof, although I don't think I really have to explain this right now).

Monoprice cables are already great, so I don't see the need to spend for cables much more, although I haven't heard the difference between a Monoprice cable and a more expensive cable claimed to be better, so I probably shouldn't say anything. $100? That seems awefully low. I think you mean $1000.

24x48
24x48's picture
Offline
Last seen: 8 years 4 months ago
Joined: Jul 2 2012 - 2:17am
Strongarm wrote:

Could anyone explain the limitation of USB to 16bit/48kHz, and whether I might be able to modify the USB connection to accept High Res files. I would have thought USB 2.0 would have enough bandwidth to handle 24 bit/192kHz.

USB2.0 interface can handle up to 480mbps. USB3.0 up tp 3400mbps. Current maxium for sound stream for PCM stereo is only 24bitsx192kbpsx2channel is only 9.216mbps! So you always have enough bandwidth to cover maximum sound streams. The problem will be that you cannot use other fast USB-drives such as flash memory, external hard disks, external optical drives, etc. If you use such devices at the same time with your USB DAC, you hear lots noise through transmission interruptions. At worse, Windows OS will delete your DAC device drivers! You need to reinstall your device drivers.

HybridCore
HybridCore's picture
Offline
Last seen: 11 years 8 months ago
Joined: Jun 27 2012 - 8:49am

I don't think bandwidth is going to be the issue here because we know there are DACs out there capable of handling 24bit/192kHz over USB 2.0 right now (okay, I lied. You guys might not know because headphones aren't really your interest, as speakers can perform better).

USB 3.0 can hit up to 5GB/s in bursts sometimes. It's no Thunderbolt though which can usually burst up to 10GB/s. I wish Thunderbolt would become mainstream. Then we'd have a single connector for video and data transfer (Thunderbolt is essentially a Mini DisplayPort cable used for data transfer instead).

24x48
24x48's picture
Offline
Last seen: 8 years 4 months ago
Joined: Jul 2 2012 - 2:17am
HybridCore wrote:

You guys might not know because headphones aren't really your interest, as speakers can perform better).

I don't think you follow the modern digital technology. World's cutting edge R&D sound engineers were predicting the demise of loud speakers many years ago, for the reasons of "laws of physics". It's science! I explained about this in detail in the section "Loud Speakers". Please read more from there.

HybridCore
HybridCore's picture
Offline
Last seen: 11 years 8 months ago
Joined: Jun 27 2012 - 8:49am
24x48 wrote:
HybridCore wrote:

You guys might not know because headphones aren't really your interest, as speakers can perform better).

I don't think you follow the modern digital technology. World's cutting edge R&D sound engineers were predicting the demise of loud speakers many years ago, for the reasons of "laws of physics". It's science! I explained about this in detail in the section "Loud Speakers". Please read more from there.

Hm...well, I've only really paid attention to Head-Fi and headphones. I'll give it a read.

Demondog
Demondog's picture
Offline
Last seen: 6 years 11 months ago
Joined: Feb 22 2009 - 5:01pm
HybridCore wrote:

I don't think bandwidth is going to be the issue here because we know there are DACs out there capable of handling 24bit/192kHz over USB 2.0 right now (okay, I lied. You guys might not know because headphones aren't really your interest, as speakers can perform better).

Some of us guys know headphones pretty well. Plus I personally wouldn't make the statement that speakers can "perform better". Headphones are a very different way to feed sound to our ears, and I don't think headphones and speakers can be compared with each other. Speakers are better at filling a room with sound, well sure.

And since when have DAC's been exclusive to headphone use?

btw- I've got a few posts over at Head-Fi under a slightly different user name, though still a dog.

struts
struts's picture
Offline
Last seen: 2 years 10 months ago
Joined: Feb 1 2007 - 12:02pm
Strongarm wrote:

Could anyone explain the limitation of USB to 16bit/48kHz, and whether I might be able to modify the USB connection to accept High Res files. I would have thought USB 2.0 would have enough bandwidth to handle 24 bit/192kHz.

Hi Grant,

I am guessing that the limitation is in the USB receiver chip, not the DAC circuit per se (couldn't find a teardown or internal pics of the Isis in a quick online search but this would be easy to confirm if you opened it up).  If my speculation is correct you won't be able to work around it.

donniewn
donniewn's picture
Offline
Last seen: 10 years 11 months ago
Joined: Apr 14 2011 - 6:40pm

upgrading the bit rate and sample rate gives you less jitter and more amplitude,so to say you cant "hear" the difference is absolutly untrue.

Log in or register to post comments
-->
  • X