[Harry Manback]

Are these generally just snake oil? I have done some reading on them, but it wasn't from a critically neutral position.

Thanks folks.

P.S. I already know that if you sit it on one of those boxes of magic dirt that it improves plankton, speed, and PRAT and conditions the copper traces on the board.

 

[castleofargh]

http://www.head-fi.org/t/779572/r2r-multibit-vs-delta-sigma-is-there-a-measurable-scientific-difference-thats-audible
a lot of waste, a lot of great information, but if you expect something like X is better, prepare to be disappointed.
in the end a DAC that measures well is a good DAC and implementations tend to be more complex than just R2R vs delta sigma.

 

[mindbomb]

afaik, the only benefit to r2r is better noise modulation. That is to say the noise floor doesn't change as much as the signal level changes. Otherwise, low level linearity (how accurate the volume steps are at low levels), distortion, and resolution actually is typically better with the more affordable and more ubiquitous delta sigma dacs. 

 

[Harry Manback]

I was thinking in terms of DACS from reputable manufacturers using quality components and proper design and components.  If two companies did this, one with delta sigma, and the other with  multibit, would one have an inherent advantage over the other solely because of it's multibit or delta sigma architecture.
 
 

  afaik, the only benefit to r2r is better noise modulation. That is to say the noise floor doesn't change as much as the signal level changes. Otherwise, low level linearity (how accurate the volume steps are at low levels), distortion, and resolution actually is typically better with the more affordable and more ubiquitous delta sigma dacs. 

 
Is this because delta sigma gets more R&D?

 

[spruce music]

  I was thinking in terms of DACS from reputable manufacturers using quality components and proper design and components.  If two companies did this, one with delta sigma, and the other with  multibit, would one have an inherent advantage over the other solely because of it's multibit or delta sigma architecture.
 
 
 
Is this because delta sigma gets more R&D?

It is because with R2R DACs getting lower bit resistor values accurately enough becomes problematic.  If the values are exactly right that effects low level linearity.  Plus at those levels even a few degrees temperature change effect low level linearity.  I don't think anyone made a multi-bit DAC with more than 20 bits.  Even that increases expense.
 
So instead they use oversampling with sigma delta design so just a few bits are needed which are easily done and cheaply done.  The result is an equivalent of pretty much perfect low level linearity and most other aspects of DAC operation.  So then lower priced better performing DACs drive out of the market more expensive lesser DAC chips.  As a result it is also true there is more R&D for that.

 

[theearbone]

Sigma-delta dacs are often implemented with complex output stages.  For example the dac in my cd player takes output from two Wolfson dac chips, combines them with an opamp, puts it through both sides of another dual opamp and then through a buffer chip, with various negative feedback paths and filtering along the way - which is more active circuitry in the audio path than the vast majority of amps*.
 
DACs made with alternate techs like r2r or fpga are currently premium products that are often implemented with much simpler output stages, they will sound different from mainstream delta-sigma dacs that are generally built around a lower price point. 
 
So in a fancy dac vs sigma-delta you get three difference:
Different filtering and conversion process
Different output stages
Premium price point that can allow for fancier PSUs that can lower the noise floor of the device and improve the sound.
 
I know my Chord Mojo sounds better than my quad dac Arcam cd player, but I don't know how to attribute that difference to the various differences between totally different designs.
 
 
* I'd guess that the only more complex amp a consumer might have would be a parellel chip amp like the Levinson ones. 

 

[Mr Rick]

Are these generally just snake oil? I have done some reading on them, but it wasn't from a critically neutral position.

Thanks folks.

P.S. I already know that if you sit it on one of those boxes of magic dirt that it improves plankton, speed, and PRAT and conditions the copper traces on the board.

 
DACs are like amps. There are a lot of ways to get the job done. Each engineer has his own way of doing the job. Being engineers, they feel their way is the best way.
 
It's up to you to decide which DAC or amp is right for you.

 

[Harry Manback]

I currently use an ODAC feeding a custom OPA1622 based amplifier. I have been considering trying out the Modi MB. I don't want to shell out the cash for 0% improvement.

I have seen Purrin's dac rankings, but I do not have enough experience with the products he compares to know how audible the differences are.

 

[sonitus mirus]

I currently use an ODAC feeding a custom OPA1622 based amplifier. I have been considering trying out the Modi MB. I don't want to shell out the cash for 0% improvement.

I have seen Purrin's dac rankings, but I do not have enough experience with the products he compares to know how audible the differences are.

 
Unless there is an obvious problem with your current DAC, it is not worth it, and you would be wasting your money.  It seems like your are dedicated to finding supporting evidence for what you want to believe, but you would be making a fruitless purchase.  The ODAC measurements that have been posted show that it is practically audibly transparent and does a perfectly acceptable job.  What could be better, and how would it be better, from an audible perspective?

 

[Harry Manback]

Unless there is an obvious problem with your current DAC, it is not worth it, and you would be wasting your money.  It seems like your are dedicated to finding supporting evidence for what you want to believe, but you would be making a fruitless purchase.  The ODAC measurements that have been posted show that it is practically audibly transparent and does a perfectly acceptable job.  What could be better, and how would it be better, from an audible perspective?

That is the problem. Others who have heard other dacs claim they do sound better. I do trust the measurements. What I am not sure of, is are we measuring everything, and which measurements really matter in reality.

 

[castleofargh]

this is up to you, obviously measurements are strictly limited to what they measure and we may be missing something, or your device in your home might just not measure like the other odac measured under whatever conditions the other guy who published the results tested it. an objective approach isn't perfect and might require to do the measurements yourself for whatever you want to know.  also an objective approach will tell you about objective fidelity. but how do you know that the highest fidelity is really what you prefer most of the time as a sound? I believe too many people entertain the illusion that they will always enjoy high fidelity more because it's mixed with the concept of "better". but fidelity also means garbage in garbage out.
 
now about people who listened to stuff and have an opinion(AKA everybody). how reliable is that?
not at all! the reasons are pretty much why this sub section exists. preconceptions, biases, the utter inability to do a proper test and the obvious inability to realize it, and sometimes basic dishonesty.

 

[gregorio]

What I am not sure of, is [1] are we measuring everything, and [2] which measurements really matter in reality.

 
1. Are we measuring everything which affects the perception of hearing? No, we're not. ... Are we measuring everything (all the physical properties) of sound waves themselves? Yes, we are.
2. If you're talking about the latter question, physical properties of sound waves, then: frequency and amplitude. If it's the former, then the question is meaningless because it doesn't necessarily have anything to do with reality.
 
To be honest though, your question is irrelevant. The fundamental bottom line is: We can only record what we can measure! Or conversely; if we can't measure it, then we cannot record it! Obviously then, even if there were something that we cannot measure, your question is irrelevant because that "something" cannot be recorded and therefore does not exist in any audio recording you are replaying/reproducing!!
 
The problem which afflicts so much of the audiophile world is that they cannot (or will not) accept that there is difference between their perception of hearing and the sound waves themselves. This leads them to the fundamentally flawed belief that their perceptions MUST effectively be properties of sound waves, and anything which challenges this belief MUST be wrong. Pretty much any fact, "fundamental bottom line", science and even simple common sense will be sacrificed in order to maintain this belief. For example; digital audio isn't digital, audiophiles are not human, the proven mathematics (upon which all digital computer technology is based) is wrong and; if there is something they experience when listening to recordings which can't be measured, it MUST be because microphones, speakers and all past and present audio recording technology is capturing and reproducing something it was not designed to capture or reproduce and which science doesn't know exists and/or how to measure it! Unfortunately, pretty much all audiophile equipment is based in part or entirely upon this confusion (between sound waves and perception) and their marketing actively promotes, reinforces and exploits it.
 
G

 

[theveterans]

Are these generally just snake oil? I have done some reading on them, but it wasn't from a critically neutral position.

Thanks folks.

P.S. I already know that if you sit it on one of those boxes of magic dirt that it improves plankton, speed, and PRAT and conditions the copper traces on the board.

 
 
In my experience, delta sigma DACs measure much better than multibit. In terms of sound R&D, multibit is easier for manufacturers to get better sound with the multibit since multibit designs are more flexible than delta sigma. For example, multibit can be discrete (e.g. Soekris or Holo Audio) or system on a chip design. Manufacturers can tailor their own custom digital filter or run NOS. Multibit system-on-a-chip design can also be custom programmed by FPGA (e.g. Metrum Acoustics) or buy off-the-shelf chip (e.g. Analog Devices AD5547 that Schiit uses for example). In delta sigma, you are limited on a system-on-a-chip design that oversamples the data (No NOS here) into very high sample rate then reduce the bitrate then apply noise shaping techniques in order to create the analog sound. Inherently, they are limited to how good their noise shaping, digital filtering techniques can be. Chord uses a custom FPGA chip to do all the signal conversion (from beginning to end), but most manufacturers use a out-of-the shelf system on a chip design so they don't need to program it.
 
That's just the DAC part, then output stage implementation between the two comes into play as well.

 

[spruce music]

   
1. Are we measuring everything which affects the perception of hearing? No, we're not. ... Are we measuring everything (all the physical properties) of sound waves themselves? Yes, we are.
2. If you're talking about the latter question, physical properties of sound waves, then: frequency and amplitude. If it's the former, then the question is meaningless because it doesn't necessarily have anything to do with reality.
 
To be honest though, your question is irrelevant. The fundamental bottom line is: We can only record what we can measure! Or conversely; if we can't measure it, then we cannot record it! Obviously then, even if there were something that we cannot measure, your question is irrelevant because that "something" cannot be recorded and therefore does not exist in any audio recording you are replaying/reproducing!!
 
The problem which afflicts so much of the audiophile world is that they cannot (or will not) accept that there is difference between their perception of hearing and the sound waves themselves. This leads them to the fundamentally flawed belief that their perceptions MUST effectively be properties of sound waves, and anything which challenges this belief MUST be wrong. Pretty much any fact, "fundamental bottom line", science and even simple common sense will be sacrificed in order to maintain this belief. For example; digital audio isn't digital, audiophiles are not human, the proven mathematics (upon which all digital computer technology is based) is wrong and; if there is something they experience when listening to recordings which can't be measured, it MUST be because microphones, speakers and all past and present audio recording technology is capturing and reproducing something it was not designed to capture or reproduce and which science doesn't know exists and/or how to measure it! Unfortunately, pretty much all audiophile equipment is based in part or entirely upon this confusion (between sound waves and perception) and their marketing actively promotes, reinforces and exploits it.
 
G

+100 to this post.

 

[Harry Manback]

I appreciate all of the posts. What I am hearing is that, DS can give great results, if you have a good recording, and the output stage does its job. MB offers more customization in almost all aspects.

Thank you all.