If the switch is not in the signal path then how does it work ? Any switch has to be in the signal path for switching to work. One of my switch is exactly like the dac switch (electronically operated instead of mechanical) - soft press a button and the input changes just like the dac switch.
The DAC Scam?
- Thread starter Analogous
- Start date
reignofchaos
Well-Known Member
The signal in the dac is still in digital domain at the switch not in the analog signal path where your switch is which. The latter is will lead to way more distortion.
If you say switching before the DAC chip is good and switching after the DAC chip is bad then I do A/B of that too, using a USB switcher and reclocking the output to check the difference between two dacs. In reality a passive swich is better than the electronic switch in the DAC. The electronic swich in the DAC generates noise as it is an active device.
This seems to be true according to a quick perusal of the question on the interweb.
But in this context (AB testing using switches) both types need to be in the signal path in order to make sound.
Whichever type of switch one uses to AB test, the distortion (larger or smaller) would apply to both the DACs under evaluation.
Indeed. The switch will not differentiate. In reality these switches do not colour the sound. Any moden dac has multiple of them even in the signal path. One great example is the speaker protection circuits and pop avoidance circuits in dacs. The good dacs use relays to mute the signal on sample rate change, when switching the DAC inputs and sample rate changes which causes issues with few dac chips (like AKM). Few dacs like Toppings use an electronic relay like SGM3710, SGM3711, etc which are cheaper than the relay based switches. You can't avoid switching in any device. And then there is the volume control. I won't even talk about it.
The DAC has an analog section. That is permanently connected either to the volume control or the XLR/RCA output sockets through an analog preamp. The switching happens on the DAC chips digital input section (USB, Coaxial SPDIF and an optical board that converts optical to SPDIF). There is no other way. However when switch is deployed on the digital section it will introduce noise in the digital section which theoretically will be corrected and whatever noise that remains will be below the threshold of human hearing. But all DACs now also deploy protection circuit in the analog section which mute the output during startup and during sample rate changes. Some DACs don't have them and you will hear POP for example with the media format changes for example from PCM to DSD. This too will be subject to noise. But just like a well designed digital section, the analog section too has measures to eliminate noise below the threshold of human hearing.
Regardless of where the switch is implmented there will be compromises, but these compromises has far less impact than the human-ear-brain interaction. Even the construction of individual ear has profound impact. The ear's pinna too has a direct impact on how the sound gets processed
Last edited:
square_wave
Well-Known Member
The A/B switch is a valid idea. However, what I look for in an audio gear is not the typical resolution/detail, frequency response difference etc that one typically looks for in "nit picky mode" while switching between two dacs using small music samples. That is the usual protocol that I have seen while doing this.
The first phase ( basic checks and balances phase) - I have a few tracks that are used to check some basic technical stuff like frequency extension, frequency balance, resolution retrieval, soundstage, tone etc. This is usually done within an hour. Simple stuff. If there is a problem here, there is no point of moving to the next phase.
Second phase. I would like to connect one dac for a day and listen and connect another the next day and listen with the same music mix. This represents what really happens in the life of a audio enthusiast. I may even repeat this a few times over the week. Playing full songs. The only thing I would do is a level match between the two dacs so that there is no volume variation between the two.
This is my usual protocol when I want to change something. Possible only if you get dealer demos or if a friend allows you to borrow something.
I look for overall feelings the dac elicits in me while listening to a wide and varied collection of my chosen mix of music. I also look for repeatability of those effects.
For the second phase, I am not sure how to use A/B switches. I guess you don't need one.
I am not sure about this but I have heard many people who are naysayers of A/B switches talking about a few problems.
1. The short music samples and "looking for problems" mode emotionally disconnects you from the actual music.
2. The Switch acts like an equalizer between better audio gear and slightly inferior ones because there is something in a signal path now. So the qualities of a higher quality gear gets slightly diminished or changes its nature. The inferior one also will change its sound but not necessarily in the same way. An impediment of a certain nature may not allow some sought after quality of the better device to shine through but the slightly lower quality device may not have that quality so it is not affected the same way. A better quality car and a lower quality car are not affected the same way when you reduce the quality of the road although the quality degradation of the road is the same for both the cars.
The first phase ( basic checks and balances phase) - I have a few tracks that are used to check some basic technical stuff like frequency extension, frequency balance, resolution retrieval, soundstage, tone etc. This is usually done within an hour. Simple stuff. If there is a problem here, there is no point of moving to the next phase.
Second phase. I would like to connect one dac for a day and listen and connect another the next day and listen with the same music mix. This represents what really happens in the life of a audio enthusiast. I may even repeat this a few times over the week. Playing full songs. The only thing I would do is a level match between the two dacs so that there is no volume variation between the two.
This is my usual protocol when I want to change something. Possible only if you get dealer demos or if a friend allows you to borrow something.
I look for overall feelings the dac elicits in me while listening to a wide and varied collection of my chosen mix of music. I also look for repeatability of those effects.
For the second phase, I am not sure how to use A/B switches. I guess you don't need one.
I am not sure about this but I have heard many people who are naysayers of A/B switches talking about a few problems.
1. The short music samples and "looking for problems" mode emotionally disconnects you from the actual music.
2. The Switch acts like an equalizer between better audio gear and slightly inferior ones because there is something in a signal path now. So the qualities of a higher quality gear gets slightly diminished or changes its nature. The inferior one also will change its sound but not necessarily in the same way. An impediment of a certain nature may not allow some sought after quality of the better device to shine through but the slightly lower quality device may not have that quality so it is not affected the same way. A better quality car and a lower quality car are not affected the same way when you reduce the quality of the road although the quality degradation of the road is the same for both the cars.
Last edited:
square_wave
Well-Known Member
Will you restart the song and listen to at least 2 minutes of the song while someone else switches the A/B switch without you knowing which is which ? Also do you level match ?
Level matching is most important. I do level match but without any instrument. My wife or son usually do the switching. Before all that happens I use ASR (anathema to many here on HFV), Archimago's Musings to look at the verdict and the community vote for the device.
The latest article by Archimago speaks about dbt.
MUSINGS: Trust, empirical testing, and evidence in the Audiophile hobby. And the Double Blind Test (DBT) strawman argument.
A blog for audiophiles about more objective topics. Measurements of audio gear. Reasonable, realistic, no snakeoil assessment of sound, and equipment.
archimago.blogspot.com
square_wave
Well-Known Member
Level matching with a decibel meter is important.Level matching is most important. I do level match but without any instrument. My wife or son usually do the switching. Before all that happens I use ASR (anathema to many here on HFV), Archimago's Musings to look at the verdict and the community vote for the device.
The latest article by Archimago speaks about dbt.
MUSINGS: Trust, empirical testing, and evidence in the Audiophile hobby. And the Double Blind Test (DBT) strawman argument.
A blog for audiophiles about more objective topics. Measurements of audio gear. Reasonable, realistic, no snakeoil assessment of sound, and equipment.
Also you missed a question. Will you listen to a full song after restarting the song while doing this ? This helps to see how the dac performs in dynamic swings throughout the song and also how you emotionally connect with the performance. If you switch between a song and let the song continue, you just tend to look for things differently. Also do you do this with multiple songs that spans genres ? And repeatability ?
I have to admit that I like ASR for what they do to bring some amount of accountability with specifications. That is the bedrock of a good product. But they don't give any importance to the subjective evaluation of a device. From that perspective, I do not like their overall approach.
I know that there are shortcomings the way I test it. I don't have a db meter. I should try out the mobile apps. But there are complexities here. What sine wave do I use for db level testing? We have to use many because the amp's frequency response may not be flat. All that just to buy equipment for my personal use? I don't post listening reviews. I do listen to the same song multiple times. A full album or two at the most and mostly limited to Jazz, Western Classical and old bollywoold songs. For these old bollywood songs all equipments sound the same. Most likely the only good source for these old bollywood songs are the LPs. So I know my setup may not be the best for genre I don't listen to. Ultimately the purpose is to enjoy Music. Many are in this hobby for Music alone and even a walkman is good enough for them. Many are here for the illusion of perfection and there is nothing wrong with it because you have a goal to reach. I'm here for the science behind it and achieving perfection without falling for snake oil and blindly follow stereophile magazines and reviews on the net. e.g. I listened to Denafrips Terminator extensively for a week and I really couldn't understand why people like it. Maybe it has a sound that appeals to many and I belong to a minority that has a different frequency curve.
A huge part of my enjoyment listening to music comes from the equipment design. I have been doing this since as a kid building radio sets, pre-amps, robotic equpments, amps, projectors, oscillators, repairing electronic equipment. I have a materials science background and to defend my thesis I built my own equipment to measure how lead titanate behaves when heated to around 1200 C and measurement was by making linear variable differential transformer to measure changes in the dimensions of the sample. I'm used to the rigour of testing and meeting with failure and that applies to this Music hobby too. I'm a measurement person and I know art too plays a very important role. Listening to Music is a by product DIY passion of indulging with electronics using unconventional methods. e.g. I don't use standard XLR or RCA cables. All my audio is routed through ethernet cables. The 4 pairs of twisted wires carry 2 channels for RCA and 4 channels for XLR. and absolutely zero hum from the speakers. I doubt anyone does what I do. All done to achieve sound of cymbals perfectly which I find lot of amps fail to do so. All ethernet cables are self made. I have all the tools, LCR meter, soldering station, etc. So I do listen to music critically but that stems out of passion for how these damn thing work. Class A, Class D Delta Sigma or Ladder dacs, NOS or OS doesn't matter. Measurement matters, technology matters, science matters, art too matters. Definition of art is different for different people. It could be the brand name for few. For a DIY it could be the journey behind making it happen. If it sound right, sounds good that is good enough for me. If it doesn't sound right, go back to the drawing board.
I route Balanced Audio and single ended audio through ethernet cable and keystone jacks soldered to RCA connectors. This scheme reduces the number of cables by half. The twisted pairs allow me to have very low inductance and I keep the capacitance low enough by having short custom length cables and using good quality unshielded CAT5/CAT6 cables. Using ethernet cables also allow me to switch equipments easily and very fast. Even My two pair of speakers are switchable at the turn of a switch or pressing a button on the remote. Speakers are the easiest animal to identify in a blind test 100% of the time and also almost impossible to test properly because they have to be placed in the same position for apple to apple comparision.
When experimenting I can connect any equpment in the chain simply be inserting an ethernet cable. This is how every equipment of mine looks from the backside. e.g. My headphone amp with keystone jack connected to the back. For critical listening I just patch an ethernet cable from the headphone amp to the DAC and that I do mostly in the nights.
mbhangui, what I like is there is a certain method to your madness to figure out what works and what doesn’t for you.
For me it’s normally an instant like or dislike. I usually end up disliking
If I like, then I follow it up with a week or two of listening. And then go back to the old to figure which connected with me more.
On a completely different note, I too did my bachelors in Material Science.
For me it’s normally an instant like or dislike. I usually end up disliking
If I like, then I follow it up with a week or two of listening. And then go back to the old to figure which connected with me more.
On a completely different note, I too did my bachelors in Material Science.
square_wave
Well-Known Member
@mbhangui , Totally love what you are doing ! You are such a mad scientist 
Unfortunately, I am on the same boat as Prem. The only level of sanity that I manage to bring to the table is a total emotional dissociation with the gear being tested which somehow manages to remove confirmation bias. For me, after a few songs I know so well, I know if I like or dislike the device. If I am on the fence with couple of them, I spend a week with those and then I surely know what will remain in the shelf.
Btw, w.r.t level matching, you are just level matching 2 dacs. Some dacs have slightly higher output level. You can test and then compensate that with the volume level on your amp when you use the dac with lower output level. I just do it with some electronic genre song that has beats on repeat. No need for a sinewave. The album random access memories has many such songs.
Unfortunately, I am on the same boat as Prem. The only level of sanity that I manage to bring to the table is a total emotional dissociation with the gear being tested which somehow manages to remove confirmation bias. For me, after a few songs I know so well, I know if I like or dislike the device. If I am on the fence with couple of them, I spend a week with those and then I surely know what will remain in the shelf.
Btw, w.r.t level matching, you are just level matching 2 dacs. Some dacs have slightly higher output level. You can test and then compensate that with the volume level on your amp when you use the dac with lower output level. I just do it with some electronic genre song that has beats on repeat. No need for a sinewave. The album random access memories has many such songs.
That thrills me. I never expected anyone here having dabbled with Materials Science.
Indeed points to what we are discussing now and that was exposed by Golden Sound who exposed the DCS Bartok DAC thing and the Denafrips false NOS claims about their dacs. What I wrote above about denafrips is exactly what Golden Sound says. The DAC is too soft.I like this thread
If you ever doubt how gullible we are as far as audio is concerned - remember your stand on TIDAL MQA
Last edited:
@mbhangui you have expressed many aspects what I am thinking of when trying g to decide whether a particular component of the audio chain is what I like or not. However I don’t have the technical ability that you possess. My lazy approach is to just connect and keep listening till I can make a decision. Sometimes I realise after buying that I don’t like it as much as I initially thought. But it has all been an interesting learning process.
I now know the extreme detailed sound across the entire hearing bandwidth is not my preference and I like slightly rolled of treble frequencies with good coherence and a fairly precise sound stage and a seamless presentation (coherence?)
Another interesting insight I had is sometimes even with fairly expensive equipment after the initial “wow” feeling I don’t hurry as much to turn on the set up for longish listening sessions.
When I achieve the sound that I like I am tapping my foot and moving to the music and forget to analyse the sound.
This may not make much sense to others, But then it’s my insight into my own preferences.
All of the above you have written points to satisfaction. Each of us is different. I don't believe there is a right kind of sound and there ever will be. There are too many imperfections that start right from the recording and finally the last chain ending in the speaker and the listening room. The natural tendency amongst us to disbelieve what one person likes and which is very different than our own taste. Most reviews fail to address what kind of sound they personally like when they review an equipment and instead use silly adjectives, flowery audio language to describe a dac, amplifier or speakers. When you listen to such reviews it looks like they are describing the best equipment made on this planet and that is a bit dishonest because folks like us get carried away by the language, expecting a Shangri-la straight from Tibet's Kunlun Mountains.
When you mentioned rolled of treble, I understand why dacs like Denafrips have an appeal. I have a theory. Maybe you hear higher frequencies better than lot of us. This is something I have not understood. All headphones have different frequency curves. So is it possible that each of us hear sound very differently? Hence things like DACs sound different to each person? I don't think any research like this has been carried out. There is this Harman Target Curve based on extensive listener preference tests. See this https://www.soundstagesolo.com/index.php/features/217-where-are-we-at-with-the-harman-curve
Also read this if you get the time. IT is a 3 part series
A Deep Dive Into Harman Curves - The Long Way Round (Part 1)
In Part 1 of the Harman Target Curve deep-dive series, we explore the fundamental of human hearing and how it is related to the Harman Curve.
www.headphonesty.com
Even the race of a person determines how you will hear the sound.
Like you said, it is the journey and the learning process that is very important. We should actually talk a lot about our experience with our journey in pursuit of this hobby. I love the new open baffle speakers that you have acquired. Most of us are accustomed to boxed speakers
Last edited:
square_wave
Well-Known Member
@mbhangui , From all that has been said so far we have realised that a listener's preference is going to be a subjective one due to all the varied and myriad variables at play. We have discussed a lot of it already. We also know that almost all dacs adhere to basic engineering specifications. The difference in sound comes from the implementation, dac type, parts choices etc. If that is the case, do we have any other choice than a subjective listening test to decide what we choose for ourselves ?
I accept that due to nature of the audio industry, there are wild claims and devices being sold at unobtanium price points etc. Let's leave that aside for a while for rich collectors.
I accept that due to nature of the audio industry, there are wild claims and devices being sold at unobtanium price points etc. Let's leave that aside for a while for rich collectors.
Check out our special offers on Stereo Package & Bundles for all budget types.
Similar threads
