• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). There are daily reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

CHORD M-Scaler Review (Upsampler)

Rate this product:

  • 1. Poor (headless panther)

    Votes: 318 90.3%
  • 2. Not terrible (postman panther)

    Votes: 8 2.3%
  • 3. Fine (happy panther

    Votes: 7 2.0%
  • 4. Great (golfing panther)

    Votes: 19 5.4%

  • Total voters
    352
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
39,111
Likes
174,558
Location
Seattle Area
But, with all that said, I clearly hear a difference with the M-Scaler running at 16x and therefore remain curious as to why this is.
And I have explained in that video why it is. It has nothing to do with M-scaler. Or the USB cables you tested. Neither makes this kind of audible difference. The reason you perceive such is because you don't know how your perception confuses you this way. You have not lived through embarrassing situations where reality was different than what you perceived. And your livelihood depended on being right. I have been so situated so let me repeat a story I have told before.

I managed the signal processing team at Microsoft which developed our lossy audio codec (WMA). I was sick home for a week and manager of the group asked me if I could evaluate their new psychoacoustic model for the encoder. I said sure. He sends me a couple of samples and I tell him that there is degradation. He made some changes and it was still bad. The process of going back and forth was slow so I asked him to just give me the encoder with some parameters that I could tweak. He does this and gives me a config file with half a dozen floating point values.

I start testing and tuning all the variables. I get them to sound great using whole numbers. Then I realized that the decimal places also made a difference. With little else to do, I spent a week narrowing down all the variables to a few decimal places. I then email my signal processing manager the results. He writes back all confused telling me that the encoder actually doesn't use any of those fractions! I told him that was wrong and that I could easily hear the differences.

Next thing I know, he sends me two sets of encoding and ask him which is better. I easily find one set sounding better. Frustrated that the team couldn't hear that obvious difference, I wrote to him which was better and my anger at why he couldn't here the difference. You know what he wrote back? That the two sets of files were identical! Yes, identical!

I didn't believe him. My ears told me there was a big difference. Then I do a binary comparison and find the files were all identical just as he said! I then listened again. Now they sounded the same. :( So I assume that there could be a difference and once again, I could hear (really imagine) the difference! My perception was that variable.

So no, it is not sufficient to think you are right. You are advocating that people spend $6000 on a box that may do absolutely nothing. You owe it to your viewership to give them reliable information. And that is only made possible with a controlled/blind test. You know, the kind where only your ears are involved. After all, you don't want consider measurements or engineering explanation of why this device doesn't make a difference.

The sooner you arrive at this truth as the rest of us have, the sooner you will be on the right path to spend money in audio where it matters, and where it does not. It might now make for popular videos but it will be truth.
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
39,111
Likes
174,558
Location
Seattle Area
Not sure how you reach that conclusion. @amirm suggested earlier that I am expecting to hear something and so I hear it.
I did not whatsoever. I didn't say you were going to hear what you expected. I said your brain took the knowledge that something had changed in the system and you then perceived a change. Usually this is for the better because you focus more and hear more detail, darker background, veil removal etc. Later when you are just sitting back and enjoying music, this aspect disappears which means the device contributes nothing to the experience.

Per above, you could expect a device to do nothing, or even imagine it doing something bad, yet walk away thinking it has positive effect on sound after listening test comparison. This is simple distortion caused by sighted knowledge of a change in the system. You have to eliminate this.
 

Doodski

Grand Contributor
Forum Donor
Joined
Dec 9, 2019
Messages
13,555
Likes
13,363
Location
Canada
Not sure how you reach that conclusion. @amirm suggested earlier that I am expecting to hear something and so I hear it. By that logic, the natural assumption (and indeed my expectations) when demoing more expensive, "better" gear is that it will be better. Thus, if our ability to conduct listening tests were so fallible, my results would always align with the more expensive "better" product sounding better. In reality, by applying the theory in the articles I linked above, I tested assumptions and challenged biases to arrive at the conclusions I did.
That's the whole point though. That one cannot self-blockade bad thoughts of bias and subjectivity while open eye testing stuff. The only way to accurately test is double blind. For somebody with a psyche degree or whatever it is that you studied isn't this a given?
 

PassionforSound

Member
Reviewer
Joined
Jul 24, 2022
Messages
45
Likes
11
And I have explained in that video why it is. It has nothing to do with M-scaler. Or the USB cables you tested. Neither makes this kind of audible difference. The reason you perceive such is because you don't know how your perception confuses you this way. You have not lived through embarrassing situations where reality was different than what you perceived. And your livelihood depended on being right. I have been so situated so let me repeat a story I have told before.

I managed the signal processing team at Microsoft which developed our lossy audio codec (WMA). I was sick home for a week and manager of the group asked me if I could evaluate their new psychoacoustic model for the encoder. I said sure. He sends me a couple of samples and I tell him that there is degradation. He made some changes and it was still bad. The process of going back and forth was slow so I asked him to just give me the encoder with some parameters that I could tweak. He does this and gives me a config file with half a dozen floating point values.

I start testing and tuning all the variables. I get them to sound great using whole numbers. Then I realized that the decimal places also made a difference. With little else to do, I spent a week narrowing down all the variables to a few decimal places. I then email my signal processing manager the results. He writes back all confused telling me that the encoder actually doesn't use any of those fractions! I told him that was wrong and that I could easily hear the differences.

Next thing I know, he sends me two sets of encoding and ask him which is better. I easily find one set sounding better. Frustrated that the team couldn't hear that obvious difference, I wrote to him which was better and my anger at why he couldn't here the difference. You know what he wrote back? That the two sets of files were identical! Yes, identical!

I didn't believe him. My ears told me there was a big difference. Then I do a binary comparison and find the files were all identical just as he said! I then listened again. Now they sounded the same. :( So I assume that there could be a difference and once again, I could hear (really imagine) the difference! My perception was that variable.

So no, it is not sufficient to think you are right. You are advocating that people spend $6000 on a box that may do absolutely nothing. You owe it to your viewership to give them reliable information. And that is only made possible with a controlled/blind test. You know, the kind where only your ears are involved. After all, you don't want consider measurements or engineering explanation of why this device doesn't make a difference.

The sooner you arrive at this truth as the rest of us have, the sooner you will be on the right path to spend money in audio where it matters, and where it does not. It might now make for popular videos but it will be truth.

I appreciate you taking the time to write this and it is an informative example. The problem I see here still though is that you are taking AN example of being tricked by perception errors and applying it to everything that is beyond obvious measurement. I don't question for a second what you are saying about the impact of our minds on what we hear. The concept of auditory illusions is well documented and I get it and I get that it plays a part in audio. I just don't agree that it cannot be controlled through other means or that it should be applied as a blanket explanation for a default position that everything sounds the same if there's no objective data to the contrary.

I do plan to arrange a whole series of blind tests hopefully later this year with multiple volunteers in an environment where we can include identical setups (i.e. controls) and systems with one variable altered, but all variations of the testing completely blind to participants. Ultimately, I see that as the only way to really move this discussion forward. I do not have the time or motivation to setup a blind test on my own (i.e. for me as the only participant) because it creates way too many challenges to remove me from any of the setup. To do it with a group, however, is much easier as I can do all the setup and they can be the blind listeners. We'll see what we discover and I look forward to the results regardless of their implications.
 

PassionforSound

Member
Reviewer
Joined
Jul 24, 2022
Messages
45
Likes
11
That's the whole point though. That one cannot self-blockade bad thoughts of bias and subjectivity while open eye testing stuff. The only way to accurately test is double blind. For somebody with a psyche degree or whatever it is that you studied isn't this a given?

One cannot self-blockade bias, but you can test against it. This is one of the theories discussed in the articles I posted. If you are conscious that you expect X to sound different to Y, you can do a round of listening and see what you hear. You can then do a fresh round of listening with the deliberate question in mind of "what if they actually sound the same and I am tricking myself?" Your brain goes looking for what you are focussed on. As Amir's example showed, when he was seeking differences in the decimal points of the variables in the WMA encoder he heard them. When he then expected the files to sound the same, he heard them the same. The expectations work in both directions and there is evidence of this in the form of auditory illusions studies and the studies on bias, consensus and group think in the broader psychological field.
 

PassionforSound

Member
Reviewer
Joined
Jul 24, 2022
Messages
45
Likes
11
One cannot self-blockade bias, but you can test against it. This is one of the theories discussed in the articles I posted. If you are conscious that you expect X to sound different to Y, you can do a round of listening and see what you hear. You can then do a fresh round of listening with the deliberate question in mind of "what if they actually sound the same and I am tricking myself?" Your brain goes looking for what you are focussed on. As Amir's example showed, when he was seeking differences in the decimal points of the variables in the WMA encoder he heard them. When he then expected the files to sound the same, he heard them the same. The expectations work in both directions and there is evidence of this in the form of auditory illusions studies and the studies on bias, consensus and group think in the broader psychological field.

I should add, that by testing in both directions as I described here, you will either discover no difference (as I often do) or you will identify an actual difference if it exists and can then take the time to explore and better define that difference subjectively.
 

Dogcoop

Active Member
Forum Donor
Joined
Feb 12, 2021
Messages
136
Likes
264
Not sure how you reach that conclusion. @amirm suggested earlier that I am expecting to hear something and so I hear it. By that logic, the natural assumption (and indeed my expectations) when demoing more expensive, "better" gear is that it will be better. Thus, if our ability to conduct listening tests were so fallible, my results would always align with the more expensive "better" product sounding better. In reality, by applying the theory in the articles I linked above, I tested assumptions and challenged biases to arrive at the conclusions I did.
This is from your 2016 review:

This is the most interesting comparison of all the cables I’ve tested I think because there is once again a significant change in the sound between the Coffee and the Diamond cables, but at first I found myself preferring the sound of the Coffee because it has a slightly greater sense of warmth and a beautifully focussed mid-range. Switching back to the Coffee though left me longing for the space I heard with the Diamond. With the Diamond USB cable, the soundstage stretches right out to each side whereas the Coffee brings it all in a bit tighter resulting in greater focus on the mid-range and the centre of the soundstage.

In a way I prefer the slight warmth of the Coffee, but then I wish for the same sense of openness and full-range clarity that the Diamond presents so it’s a really hard one to split here. I think the conclusion for me is that the Diamond is probably the cable I’d choose in an unlimited budget situation,
I started this test expecting to see a dramatic display of the law of diminishing returns, but I can honestly say that every cable in their range demonstrated a significant improvement over the lower model so you can safely spend whatever you’re comfortable with and know that you are really getting the best value for money with the AudioQuest range.




After reading the above attributes you ‘hear’ in different cables, I have no doubt that you believe you ’hear’ something magical that the m-scaler does to the sound. However, until you perform proper blind/abx testing of your ability to ‘hear’ these differences, I remain extremely skeptical of your ‘hearing’ claims.
 
Last edited:

PassionforSound

Member
Reviewer
Joined
Jul 24, 2022
Messages
45
Likes
11
This is from your 2016 review:

This is the most interesting comparison of all the cables I’ve tested I think because there is once again a significant change in the sound between the Coffee and the Diamond cables, but at first I found myself preferring the sound of the Coffee because it has a slightly greater sense of warmth and a beautifully focussed mid-range. Switching back to the Coffee though left me longing for the space I heard with the Diamond. With the Diamond USB cable, the soundstage stretches right out to each side whereas the Coffee brings it all in a bit tighter resulting in greater focus on the mid-range and the centre of the soundstage.

In a way I prefer the slight warmth of the Coffee, but then I wish for the same sense of openness and full-range clarity that the Diamond presents so it’s a really hard one to split here. I think the conclusion for me is that the Diamond is probably the cable I’d choose in an unlimited budget situation,


After reading the above attributes you ‘hear’ in different cables, I have no doubt that you believe you ’hear’ something magical that the m-scaler does to the sound. However, until you perform proper blind/abx testing of your ability to ‘hear’ these differences, I remain extremely skeptical of your ‘hearing’ claims.

That's OK. You're welcome to ignore them (as is anyone). For the record, I actually chose the Coffee cable as my preferred option despite the Diamond being more expensive and marketed as better. I think this conclusion was addressed at the end of that review or perhaps I landed there after further use. It is definitely something I have discussed quite often in comments and later videos.
 

the_brunx

Active Member
Joined
Apr 7, 2020
Messages
268
Likes
653
And it’s not just USB cables, he believes and says all cables make a difference, power, interconnects. The whole shebang. And when he does blind test or other tests he uses two set ups and attributes any difference he hears only to the cables, And as you can tell he is intelligent enough to know this is wrong but he does it anyway and to me this is what makes him a conman. Sorry to say.
 

PassionforSound

Member
Reviewer
Joined
Jul 24, 2022
Messages
45
Likes
11
And it’s not just USB cables, he believes and says all cables make a difference, power, interconnects. The whole shebang. And when he does blind test or other tests he uses two set ups and attributes any difference he hears only to the cables, And as you can tell he is intelligent enough to know this is wrong but he does it anyway and to me this is what makes him a conman. Sorry to say.

When I use two setups, they are identical. If not, I am testing two setups to see if what I hear is consistent on multiple devices. Please get your facts straight
 
Joined
Jul 19, 2022
Messages
12
Likes
39
And I have explained in that video why it is. It has nothing to do with M-scaler. Or the USB cables you tested. Neither makes this kind of audible difference. The reason you perceive such is because you don't know how your perception confuses you this way. You have not lived through embarrassing situations where reality was different than what you perceived. And your livelihood depended on being right. I have been so situated so let me repeat a story I have told before.

I managed the signal processing team at Microsoft which developed our lossy audio codec (WMA). I was sick home for a week and manager of the group asked me if I could evaluate their new psychoacoustic model for the encoder. I said sure. He sends me a couple of samples and I tell him that there is degradation. He made some changes and it was still bad. The process of going back and forth was slow so I asked him to just give me the encoder with some parameters that I could tweak. He does this and gives me a config file with half a dozen floating point values.

I start testing and tuning all the variables. I get them to sound great using whole numbers. Then I realized that the decimal places also made a difference. With little else to do, I spent a week narrowing down all the variables to a few decimal places. I then email my signal processing manager the results. He writes back all confused telling me that the encoder actually doesn't use any of those fractions! I told him that was wrong and that I could easily hear the differences.

Next thing I know, he sends me two sets of encoding and ask him which is better. I easily find one set sounding better. Frustrated that the team couldn't hear that obvious difference, I wrote to him which was better and my anger at why he couldn't here the difference. You know what he wrote back? That the two sets of files were identical! Yes, identical!

I didn't believe him. My ears told me there was a big difference. Then I do a binary comparison and find the files were all identical just as he said! I then listened again. Now they sounded the same. :( So I assume that there could be a difference and once again, I could hear (really imagine) the difference! My perception was that variable.

So no, it is not sufficient to think you are right. You are advocating that people spend $6000 on a box that may do absolutely nothing. You owe it to your viewership to give them reliable information. And that is only made possible with a controlled/blind test. You know, the kind where only your ears are involved. After all, you don't want consider measurements or engineering explanation of why this device doesn't make a difference.

The sooner you arrive at this truth as the rest of us have, the sooner you will be on the right path to spend money in audio where it matters, and where it does not. It might now make for popular videos but it will be truth.
Perception is a big factor in our judgment in sound. I used to be a live sound engineer. When i was doing monitors for musicians, many times they asked me for more volume in their wedges , done nothing but signaled them it was done . And they were convinced i had raised their volume and happy ! I had to trick them because the raising path leads to feedback. But they were happy as i'm sure Chord clients are (and lighter of 5000 USD of course!)
 

raif71

Major Contributor
Joined
Sep 7, 2019
Messages
1,909
Likes
1,871
That's OK. You're welcome to ignore them (as is anyone). For the record, I actually chose the Coffee cable as my preferred option despite the Diamond being more expensive and marketed as better. I think this conclusion was addressed at the end of that review or perhaps I landed there after further use. It is definitely something I have discussed quite often in comments and later videos.
Maybe you can find a suitable coffee table to put your gears along with the coffee cable :)
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
39,111
Likes
174,558
Location
Seattle Area
One cannot self-blockade bias, but you can test against it. This is one of the theories discussed in the articles I posted. If you are conscious that you expect X to sound different to Y, you can do a round of listening and see what you hear. You can then do a fresh round of listening with the deliberate question in mind of "what if they actually sound the same and I am tricking myself?" Your brain goes looking for what you are focussed on. As Amir's example showed, when he was seeking differences in the decimal points of the variables in the WMA encoder he heard them. When he then expected the files to sound the same, he heard them the same. The expectations work in both directions and there is evidence of this in the form of auditory illusions studies and the studies on bias, consensus and group think in the broader psychological field.
Please don't pull me in as a witness because there is no validity to what you just described. Your references are not remotely on topic either. Group think? What on earth does that have to do with you singularly detecting a difference in an audio device or not?

Do you have any proof point that it works? That is, someone tested you multiple times to see if you detect valid differences when there, and not when there isn't? I trust not.

You said there is clear audible difference with this device. I am happy to throw out all my measurements in favor of that proof point. Can you turn on the camera and have someone change the upsampling method a dozen times and see how often you get it right? It should only take a few minutes.

Or are you telling us the only way you can tell this improvement is by using your eyes in addition to ears?
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
39,111
Likes
174,558
Location
Seattle Area
Perception is a big factor in our judgment in sound. I used to be a live sound engineer. When i was doing monitors for musicians, many times they asked me for more volume in their wedges , done nothing but signaled them it was done . And they were convinced i had raised their volume and happy ! I had to trick them because the raising path leads to feedback. But they were happy as i'm sure Chord clients are (and lighter of 5000 USD of course!)
Indeed. We had a demonstration of JBL M2 speakers for the local audiophile society. We had run the calibration and knew response was right in room but folks insisted the bass was too much. So we changed it just by -0.2 dB (but didn't tell them that) and all of sudden they declared it sounded correct!
 

PassionforSound

Member
Reviewer
Joined
Jul 24, 2022
Messages
45
Likes
11
You said there is clear audible difference with this device. I am happy to throw out all my measurements in favor of that proof point. Can you turn on the camera and have someone change the upsampling method a dozen times and see how often you get it right? It should only take a few minutes.

If it were that simple, I would, but the very act of changing the upsampling setting (and the time between switches) makes the test no longer truly blind hence my point earlier about having to set this up properly if we're going to do it at all

Edited to add clarity: to move from 16x to bypass takes around 0.5 seconds, but to cycle from bypass to 16x takes noticeably longer which will quickly indicate what the setting actually is

Also, while we're on that topic, Amir, I'm curious still why you didn't personally conduct listening tests (and record the tests) for the M-Scaler at the full 16x oversampling?
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
39,111
Likes
174,558
Location
Seattle Area
If it were that simple, I would, but the very act of changing the upsampling setting (and the time between switches) makes the test no longer truly blind hence my point earlier about having to set this up properly if we're going to do it at all

Edited to add clarity: to move from 16x to bypass takes around 0.5 seconds, but to cycle from bypass to 16x takes noticeably longer which will quickly indicate what the setting actually is

Also, while we're on that topic, Amir, I'm curious still why you didn't personally conduct listening tests (and record the tests) for the M-Scaler at the full 16x oversampling?
Answering the last part, I didn't know that a talking point would develop that 2X was no good and that only 16X had value. This came about post my testing. How is that a 2X upsample is no good when it invokes the same super long filter? How is that the magic only exists at 16X and all disappears at 2X?

As to the rest, didn't the two of you in the video ridicule me for not testing 16X against bypass saying it is seamless? Now you say that test is impossible blind because it is not? It seems that both you and Rob didn't know about this delay. I tested bypass to 2X because it is completely seamless.

Still, I don't know why it should stop you from running this blind test. Have you leave the room between switching and come back when it is changed (or not). Do this a dozen times and see if you can identify the difference.
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
39,111
Likes
174,558
Location
Seattle Area
I will summarize where we are:

1. I have both measured and provided listening tests and indicate there is no audible difference with M-scaler. I could have dismissed it out of hand but instead, provided extensive testing that actually brought out design deficiencies in increased noise and jitter.

2. You say that you can ask Rob Watts about his claims of transient response. Once he provides such measurements, references and or controlled blind listening tests, then we can discuss.

3. You propose a way to remove bias from sighted listening tests which has no proof point or valid references. You can remedy that by shooting a video that demonstrates efficacy by having someone randomly run trials on you. Until then, I hope you appreciate that your testing in this regard has no probative value.

4. You don't want to run a blind test even though you say the difference is obvious. You have not articulated why this is hard let alone impossible seeing how the difference is significant as far as you say.

5. I still have not heard a valid response as to why you believed everything Rob Watts said without due knowledge to know if he is right or wrong. You should have asked for research references, valid blind tests, etc. Instead, you just gave him a microphone and nodded to everything he said. In US we call this an informercial. Something to sell products by exaggeration and not any search for truth in advertising.
 
Top Bottom