Would someone explain what "up sampling" means and how it can affect the sound?


Hello.
I am looking at buying a DAC and after reading a few articles, there seems to be a question that comes to mind: what is up sampling, and does it affect the sound ? Some audio companies have up sampling in their DAC and some do not.
I just want to make an educated decision about what DAC I purchase that, hopefully, will not be obsolete in a few years.
Thank you for all your comments and answers.

rockanroller
If you want to make an educated decision, buy a dac based on sound and not specs.  For your purposes upsampling is when the sample rates are multiplied in hopes of improving the sound.  There are excellent products that use these techniques and others that believe that the less manipulation of the signal done, the better.  IMO, forget about  it.  
rockanroller,
upsampling is a digital domain process where the bit rate of the digital music stream is changed to something higher by an integer number. In redbook CD the bit rate of the music stream is 44.1KHz.
It appears to me that you are not versed in digital signal processing hence going into too many technical details is not advisable.
But, due to the 44.1KHz sampling rate of redbook CD, the analog filter at the output of the DAC needs to have a very steep filter skirt to attenuate the DAC clock energy. Well, making such a steep skirt analog filter has the major disadvantage that the in-band filter response (i.e. the response of the filter in 20Hz-20KHz audio region) attenuates the high frequency response of the music/program material + it creates a lot phase shift which destroys the integrity of the music signal.
One way around this is to upsample the music signal by an integer number, say, 2 to 88.2KHz. Now, the filter skirt of the analog filter following the DAC can be much less steep. This makes the analog filter easier to design, it adds less phase shift inside the 20Hz-20KHz audio band. So, from an analog filter design perspective this is a win.
What about the digital music bit-stream perspective? The way upsampling is done they zero-stuff. So, imagine your bit-stream was a 11111 at 44.1KHz & we want to upsample to 88.2KHz. then the new zero-stuffed bit stream would look like 1010101010. Now we can clock this new bit stream at 88.2KHz & the logic1s would be encountered at the same time as they were when we clocked the data at 44.1KHz. So, we have increased the bit rate. but we've destroyed the original data - our orig data at 44.1KHz had a string of 5 logic1s; the new 88.2KHz bit stream has alternating logic1 & logic0. what's up with that? You are correct if you asked this question. So, here comes the important part of upsampling, smooth the new higher speed bit-stream, you have to run the 88.2KHz bit stream thru a digital filter that will try to estimate whether the zero that was stuffed should remain a logic0 or be flipped to a logic1 to make the playback music sound right. There are n number of algorithms to do this many of them are quite sophisticated & almost all of them are proprietary. it is this digital filter algorithm (often called an interpolation filter) that imparts its sonic signature onto the final up sampled music signal. You might like the way the manuf does this interpolation or you may not. And, herein lies the crux of the matter i.e. the effect of upsampling onto the higher bit rate music signal. It is also why so many manuf exist & different people like diff manuf & you rarely get a consensus on who upsampling technique is the best. It's a matter of taste. Often you even see people who have upsampling CD players & DACs where they simply turn this feature off with the press of a front panel button because they say it sounds better without upsampling.
Anyway, that's the skinny on upsampling & its effect on the music signal. sorry it had to be so long-winded but it was important to explain a bit of the background.       
Just listen to the DAC and don't worry about upsampling.  Resulting sound quality depends on much more than that alone.  Can never hurt to understand what it is though.   

I have a non oversampling DAC and others that do.   They all sound lovely.  DACs are like phono carts.  No two sound exactly alike but most these days sound very good.  They need not cost a fortune to sound good though certainly some that do cost a lot sound very good as well, maybe even better.  You really have to do very careful compares to know what you like best.
Bombaywalla, I think that's the first time I understood someone's explanation of upsampling and remembered it minutes after reading it.
:-)
Thanks!

All the best,
Nonoise

https://en.wikipedia.org/wiki/Sampling_(signal_processing)#Sampling_rate

Also, I would have to disagree with chayro’s statement “If you want to make an educated decision, buy a dac based on sound and not specs.” Seems to me it would be best to make an educated decision based on all possible information available. That would include specifications, your own subjective listening, opinions of reviewers, value of the component, etc.

I used to design and implement up and resampling algorithms (and others) for image processing of satellite and aerial photography imagery.

Never for digital audio. The problem is different there but the principles are similar.

DCS Ring DAC technology (expensive) is the one I know of from a few years back that struck me as the best in regards to its ability to randomize noise and produce best results on paper. I’ve heard DCS gear and what I heard was consistent with what I expected.

I wonder to what extent other vendors have done anything similar in recent years to close the gap and maybe even help bring the cost down?

Theoretically, I would rather have a non oversampling DAC done well than an oversampling one done just so-so.

In practice these days, as I mentioned most good quality DACs I hear even at modest cost sound very very good and I care less about the technical details than what I hear, although if buying without opportunity to hear, it never hurts to know how things work in comparison under the covers. Assuming the information provided is accurate and reliable of course. In many cases, its just a black box with little information on how things work. You always only know for sure by listening.

The DAC in my main system is a several years old non-oversampling  mhdt Constantine I picked up used a few years back for less than $400  mainly as an experiment. I’ve never a/b compared to DCS, but the sound reminds me of DCS in many ways and its been a keeper.

Recently I’ve experimented with analog out from newer Iphone and Ipad. The sound is a lot different than the mhdt but surprisingly good and perfectly listenable as well in its own terms.

I also have a Bel Canto c5i digital integrated amp. Its also most listenable and a big hitter in terms of sound quality and features, but sounds much different from the others still.

So lots of very good flavors of digital out there these days at all price points. You just gotta pick a few and try.


I'm a "computer geek", so when I bought my current DAC, I was concerned as well when I read that it upsampled to high resolution DSD as part of its way of operating.

I concer with mapman: " Just listen to the DAC and don't worry about upsampling."
It’s just a change of the sample rate, any change of the sample rate is a lossy process and is audible especially on accurate setups.

Google the Grammy 5.1 guide, they mentioned it there as well.

I prefer NOS (sample rate unchanged), but modern DAC with 8x upsample sounds pretty damn good so there is no need to pay insane premium for NOS DACs unless you are made of money.

Well, I agree with chayro’s statement and disagree with gdhal's statements, "Seems to me it would be best to make an educated decision based on all possible information available. That would include specifications, your own subjective listening, opinions of reviewers, value of the component, etc." You don't listen to specifications and just because a reviewer raves over a unit doesn't mean you will like it. Bottom line is you buy what your ears like. Too many people buy for the wrong reason, and all you need to do is trust your ears.


I cannot tell you how many posts I've seen in which the person is looking for a DAC with a certain chip or a speaker with a first-order crossover, or a diamond tweeter or whatever.  IMO, it all doesn't mean a damn thing. It's how the product works as a whole that counts.  But then again, I truly believe that many audiophiles have no idea of what they're listening to anyway and just make purchase decisions based on reviews or pictures or whatever.  Of course, it's their money and their choices, so everyone is entitled to enjoy this hobby in their own way.  I just think that many audiophiles would be happier with their systems if they just listened more and read less.  
Oversampling - integer multiple of original rate
Upsampling - non-integer multiple of original rate

For instance, my Benchmark DAC1 is upsampling (and not the oversampling) DAC,
Specifications relating to how devices interface with other devices are always important. Many examples some of more consequence than others, but you ALWAYS want to know if or how well one device will work with another before buying.

Other specs or aspects of design are of interest also of course but harder to "bank on" in that there are so many factors that go into a quality design and product and what people judge as "good sound" that you never really know for sure what's good for you until you hear for yourself
tls49 and chayro - ordinarily I would ignore such a foolish and technically inaccurate response (from the both of you), however, I have ample time and desire to stir things up a bit to reiterate how wrong you are. Nowhere have I stated that someone should not listen with their own ears. What I have stated is that doing so is simply one more variable - albeit an important one - to the overall decision making process. The both of you, on the other hand, have gone as far as to completely discount everything except listening. And that is ridiculous, among many other other adjectives. 
I completely agree with Gdhal that all available information should be considered. For example, as Mapman alluded to and as I have stated in a number of past threads, specifications can often be particularly useful in **ruling out** components from consideration that would be poor matches to the components they would be used with. Thereby reducing the randomness of the component selection process, and the likelihood of expensive mistakes.

Also, "trust your ears" only goes so far. For example, obviously it is often not possible to audition a component in one’s own room with one’s own equipment. And just as obviously listening to a component in a less familiar environment with less than familiar associated equipment, probably during a relatively brief audition, will not necessarily say very much about how that equipment will sound to that listener in his or her own room with his or her own associated equipment, over the long-term.

Of course, this assumes that one has sufficient knowledge about specs to not misapply them. For example, if everything else is equal I would expect an amplifier having 0.0001% total harmonic distortion to be very likely to sound worse than one having 1% total harmonic distortion, when used in conjunction with many or most speakers. Why? Because the 0.0001% amplifier was probably designed with specmanship rather than sonics as the leading priority, and the 0.0001% THD figure was probably accomplished via heavy-handed application of feedback, with the adverse side effects that often go with it.

On another note, kudos to Bombaywalla for his excellent post.

Regards,
-- Al

"I just think that many audiophiles would be happier with their systems if they just listened more and read less."


It is possible to do both...
I have to concede that trusting your ears is not enough.  Unless you have ears you can trust.  
Ears don’t lie, but do have limitations:

1) you can’t hear something unless you buy it or go hear it somewhere else first

2) If you hear it somewhere else first, you are hearing an entire setup, not just the thing you are interested in hearing. So your ears won’t lie and you know what the thing of interest is capable of sounding like but not what it will sound like back home.

3) Once you buy if you don’t like what you hear then you are back where you started in the first place.

So best to use all information available to help deal with the limitations your ears alone are handed. That helps assure a happy ending sooner and probably avoids wasted funds.

Or just keep trying until....
I think we're getting a bit off topic here.  I agree with what most of you are saying in principle.  I don't think there is anyone who would disagree that the best way to audition any component is in your home system.  But that's not always possible.  When I was purchasing my system, I would bring my CD player to the dealer in order to have some point of reference as to the qualities of the new player compared to what I had.  Speakers are somewhat different in that you never know how they will work in your room, which is why I had a home audition before investing in expensive (for me) speakers.  So I think we all pretty much agree on this.  But what I am saying, and I'm sticking to it, at least for now, is that knowing about the mechanics of upsampling or what DAC chip is not going to matter one iota to whether you will like the player.  I know some people believe that each DAC chip has a sound, or that a ribbon tweeter is automatically brighter than a silk dome and so on.  In my experience, I have found those generalizations not to be true.  But that's me.  

Well, after reading all these comments and opinions, looking at reviews on TAS and Stereophile, and considering my budget, I have decided to buy a Schiit  Bitfrost with the upgrades, sinvce they have a 30 day try-out.

Thank you kindly .

But what I am saying, and I'm sticking to it, at least for now, is that knowing about the mechanics of upsampling or what DAC chip is not going to matter one iota to whether you will like the player.  I know some people believe that each DAC chip has a sound, or that a ribbon tweeter is automatically brighter than a silk dome and so on.  In my experience, I have found those generalizations not to be true.
I agree with this, Chayro, at least as a rule of thumb that is true most of the time.  It follows from my belief (shared by many others, of course) that what usually matters most in the design of an audio component is not the approach that is chosen, but how well the chosen approach is implemented.

Regards,
-- Al
 
Nice choice!  Should sound really good.  Little to loose if you don't like it for any reason.
I've heard very good things about that company.  The home audition is very prudent.  Leave it on for at least 24 hours before doing any serious listening.