Columns Retired Columns & Blogs |
May 21, 2010 - 4:31am
#1
DAC's only allowing specific sampling rates
Loudspeakers Amplification | Digital Sources Analog Sources Featured | Accessories Music |
Columns Retired Columns & Blogs |
Loudspeakers Amplification Digital Sources | Analog Sources Accessories Featured | Music Columns Retired Columns | Show Reports | Features Latest News Community | Resources Subscriptions |
How many other music titles are limited to specific sampling rates which will likewise limit high resolution file playability?
There are a number that do not accept 88.2 but this is becoming much less common.
Just briefly looking at the high rez selections on HDTracks.com, it looks to me at least half of the current offerings are at an 88KHz sampling rate. So if other DAC's limit the sampling rates like the Cary 306, to me this is a significant problem with the state of high resolution downloads.
Yes, and perhaps the majority of those are SACD transfers. There is one prominent orchestra that acknowledges this is what they provide HDtracks, a hybrid SACD. How they transfer it to a high rez download they have no idea. The few machines capable of extracting the data can do so only at a 44.1 downsampled rate. Are they then upsampled to 88.2 and sold to you as an exact copy of the original. I believe it was Ed Meitner who said something to the effect that it would take a fairly arrogant attitude to suggest there is not some loss of quality _ every _ time you SRC.
Why can't we get those Chesky boys to fully divulge how the high res files in each and every case have been derived? As consumers we have a right to know. Meanwhile we jump through hoops to purchase equipment capable of playing, ...what???
That is not correct. The professional dCS 972 and 974 converters - I purchased a 972 some years back - can convert DSD data to any sample rate desired. However, as the conversion of DSD to 88.2kHz or 176.4kHz PCM is computationally more simple, that is what is usually done.
Regarding the inability of the updated Cary to accept these sample rates, it is possible to sample-rate convert 88.2kHz files to 96kHz or 192kHz. Depending on the program used - see http://src.infinitewave.ca/ - the loss of quality will be minimal. However, when done with the highest precision, the conversion can be time-consuming. Bias Peak Pro 6 at its highest-quality setting running on my Intel MacBook takes about 3 minutes to SRC each minute of an 88.2kHz file.
John Atkinson
Editor, Stereophile
Let me ask a really dumb question...do any of these numerous sample rates sound better than any other? If not, why in God's name do they makes so many? This is worse than Beta/VHS or CD v SACD v what all...
A DAC standard that covered the best sample rate would seem to be an answer.
Then let me ask another really dumb question...do any of these different car engines work better than any other? I mean boxer four v straight six v V8 v W10...
Why can't they just agree on a standard and stick to it?
Poor comparison...99.99% of all cars run on gas or deisel and get from point A to B...The DAC's in question only run on the fuel they like and there are apparently a lot of fuel choices...My question is simple, is one sample rate better than the others and if so, why not standardize on that??
Hello Jim,
You're right, it was a poor analogy. Perhaps gas would have been a better one? Even in a small market like Sweden there is a choice of 4 different types of gas at most stations (regular, premium, diesel and ethanol). If one is better than the others why not just standardize on that?
My point is that the premise implied by your question seems flawed on a number of levels:
1. Define 'better'/'best'. I suspect you mean sound quality wise (however please acknowledge that even that is subjective). Problem is there are other parameters to consider such as cost or negative externalities such as enviromental impact. These are typically in conflict with each other so what is 'best' on one paramter may be 'worst' on another and different consumers have different priorities. The gas analogy works well here.
2. Is it an attribute worth standardizing? Pretty much all common sampling rates in consumer audio are multiples of either 44.1kHz or 48kHz, rates chosen not for their sonic superiority but for compatibility with older analog broadcast/storage technologies. If crystals for both these sampling rates are present then supporting all the other common ones is quite straightforward with most DAC chips. Plenty of DACs out there support multiple sampling rates, the ones that don't don't because they use old DAC chipsets that can't or the vendor simply decided the marginal benefit was not worth the marginal cost.
3. Consider the way that standards actually get created in the real world. If we confine the discussion to standards that broadly speaking enable interoperability (which I think is your point here) then there are three basic mechanisms:
In none of these categories do notions of sound quality, 'better' or 'best' play any significant role in shaping the standard. Beyond the basic desire to promote interoperability (including backwards compatibility) factors such as cost and convenience and the compromises inherent in any negotiated settlement are the main drivers.
4. Even in the presence of a dominant standard, vendors continue to innovate outside the standard in order to differentiate their products. Even if the market 'standardized' on one sampling rate, somebody would still use another and claim it was better. Even as the market seemed to converge around PCM, someone came along and introduced DSD, which was fundamentally incompatible, claiming it was better.
Anyway, enough of me picking holes, please explain how you envisaged this working.
V12. No question. If only upsampling the boxer four in my Subaru made the difference they said it would!
The multitude of sampling rates and bit depths does seem a bit silly until bandwidth and storage mediums are considered.
You drive a Subaru? Elk, you just shattered all my illusions about you!
As an aside, with the exception of Porches almost all my favourite engines (diesel, of which I am a huge fan, as well as gas) have been 4.2 litre V8s. IMO V12s are just too heavy and long and cause all sorts of packaging problems. I spotted the pattern with the second one I owned, I doubt there will ever be a third. Interestingly, a couple of the cars I like that used to have V8s in have now gone over to V10s which I have never tried. Maybe this is the ideal compromise?
The cute little Subie is my practical car. The boxer engine is fun as it sounds like a little Porsche.
V12s are a pain, but they sound and feel wonderful.
Otherwise, big V8s are delicious - lots of torque, willing to spin. The Z06 is 7.0 liters of fun!
I also really enjoy rotary engines. The Mazda RX8 engine is a wonderful modern example - silky.
The only V10 I have driven is the Dodge Viper. Miserable car and the engine sounds like two slant 6s parked next to each other. Bleh.
Perhaps and I admit I do not understand why so many different and conflicting formats are necessary. If the highest sample rate produces the best sound, or any other sampling rate, why all the choice, with some media working and other not?
I am reminded of the old video tape format wars..Nothing was gained but market confusion. The CD was an example of doing it right...one format agreed on by everyone, and before one complains that the red book standard is flawed, remember, the same standard is used today in CD playback that would astound the folk who created it all those year ago.
Would you buy a DVD player that only played some DVD's? or a CD player that only played some bran names? I wouldn't.
I understand that, but the public also must be considered. The overwhelming majority expect their gear to work always and under almost all conditions. If one was to buy a cassette deck, having a score of different size cassettes would make most folk very reluctant to buy one.
That is true, there will always be niche hobbyists BUT, again, the majority of folk are not going to invest in technology that shows minor alleged improvement. They will wait for acceptance and universality.
Think the CD...for the overwhelming majority of adapters, the sound on their gear was both better and easier to use than the vinyl it replaced..after 30 years, the best vinyl is as much better than mass market CD as that CD was mass market vinyl.
People want to be knocked out by new technology, not baffled and befuddled by it. Conflicting standards never is good for a medias success.
I would like to see a red book type of decision, something that produces something that everyone can use and that all music providers can work toward so when one buys a modern DAC, it will play ALL commercial media, not just some. If such a standard is agreed upon, economy of scale will drop the cost very quickly. Look at how fast CD players went from $1000 to $100 or less...what, 2-3 years.
We pretty much had this with the introduction of DVD-A and SACD. Neither format stirred the emotions of the average consumer sufficiently to become accepted.
It's pretty uncommon for a modern DAC not to play all sampling-rates/bit-depths commonly available in downloadable format. All will play the digital out form a DVD, CD transport, etc.
There was a time when a number of DACs didn't play 88.2 but this is uncommon now. This is the only "commercial" format that posed any problem that I can think of. It is commercial only in a very limited sense.
Is there another specific example you see as an issue?
I would argue that comparisons with CD, DVD and SACD all fall down as the latter are physical media requiring player/software compatibility. In this case however the only requirement is that the ADC and DAC support the same rate, nothing else in the distribution or replay chain is affected.
IMO this is a non-issue which has been sparked by a handful of deficient products rather than a systemic problem. I have no doubt it will sort itself out one way or the other. Either 88.2 will fade away and the industry will converge on something else, e.g. 96, or the offending products will be fixed or superseded. Looks like a storm in a teacup to me.
Agreed. It's essentially a non-issue already.
But it could be frustrating if you already bought a DAC not realizing it couldn't play what you already have/plan to purchase.
Caveat emptor, right?