Article: SOtM Launches sMB-Q370 Motherboard

[extracampine]

Interesting to see a motherboard coming from SOtM. Their focus had previously been on separating the source component/PC (e.g. a Roon Rock) and the "endpoint" - e.g. using an SOtM SMS-200. As these two components have quite different requirements, there seems to be some sense in this - and this appears to have been the prevailing approach in the audiophile community. Having the source and endpoint separated by ethernet cable seems to negate the need for an "audiophile" motherboard in the source PC. Or am I missing something? If you just had one machine which acted as both source and endpoint (e.g. a CAPS), then maybe there might be some use in it....

[extracampine]

On 1/7/2022 at 5:24 PM, bbosler said:

BTW the joke is actually there are 10 types of people in this world - those who understand binary and those who don't

This made me lol....I know that one of course, and it's what I based my comment on....the "joke" in my case is just as stated! 

 

Regarding the debate between measurements and what people can hear, I agree with the above poster that the double-blind controlled trial is the only way to answer the question truthfully. It's how it's done in other fields where the results really matter. I think saying that it is difficult to do is a cop-out. It would be easy, for example, to have a hifi setup with 2 source components, one with a standard mobo and one with this one. You could then easily switch source and listen for any differences. You would need a certain number of listeners ("n") to make the results statistically significant. 

 

If someone would set up a website, start doing this sort of thing and publishing the results, I think it would be very popular indeed!

[extracampine]

On 1/13/2022 at 1:43 AM, fas42 said:

 

What would happen if this were done, is as follows:

 

* A system, to get rid of the switching mechanism, would be duplicated: one with standard mobo; the other with the one discussed here. And it would turn out that the mobo variation does make an detectable difference, for a group of listeners; enough to be statistically valid.

 

* This would deeply disturb those of an objectivist bent; and at least one of those people  would assemble an matching rig - and measure, measure, measure. Leaving no stone unturned, he would unearth some anomaly, which he could link back to the mobo - perhaps, "The DAC that was used is poorly engineered; it reacts to a type of noise I can make this mobo produce!" - and be triumphant; he had demonstrated that the posted results were meaningless ...

 

* The objectivists would feel good, because their thinking had not been impacted; some of the subjectivists would despair - "Here we go again!!"

 

* And only a few individuals would realise, that this frantically digging objectivist had in fact demonstrated exactly what the other side always work with: that components always have flaws and weaknesses, and it's a balancing act, a journey, to assemble a combo of bits that overcome all the bottlenecks, to an acceptable degree.

 

* Nothing would be learnt. No progress would be made. And the world would keep turning ...

 

You're making a very large assumption here....that the mobo variation makes a statistically significant detectable difference. I don't think that it would, but that is what the test would be set up to do. 

 

Second, you say "this would deeply disturb those of an objectivist bent". It wouldn't deeply disturb them - the whole point of a double blind trial is to provide objective evidence - where there is a difference, or not. And if someone can find out what is causing that difference, then that's a win-win isn't it! 

 

The fact remains, that when the results matter, this is how it is done. You can't argue with that. So if we really want to find out the answers, this is how, no matter how cumbersome to set up and how much certain audio manufacturers wouldn't like it. 

 

 

[extracampine]

20 hours ago, fas42 said:

IME, everything you do in the electrical environment of a higher resolution setup makes an audible difference. For the simple reason that most designers and engineers assume that what they have done to avoid this, is "good enough". And this immediately unleashes the enormous tweaking, and snake oil industries we currently have ...

 

Well, that's your opinion and not what we were discussing here. We're talking about the objective stance, and double-blind controlled trials.

 

20 hours ago, fas42 said:

Whether a particular mobo does or doesn't is pretty irrelevant - it's just a grain of sand on the beach of all the things that can matter. I can alter how my current system sounds by making almost absurdly silly changes to how things are set up - and the reason for that is there is an extremely fine balance required for optimum SQ - disrupt that balance, and it's easy to hear the effect.

 

It's just as relevant as any other component is. Indeed, it's $550 of relevance, which is a lot to some. And whether "all the things that can matter" actually do matter, well, you'd have to do some double blind controlled trials to see if they do.....

 

20 hours ago, fas42 said:

It's hard to do the DBT well enough to get meaningful results.

 

Back to the "it's difficult" argument.... It's not prohibitively difficult - you just get 2 systems that are the same, except for the mobo, and compare them - in a controlled way in keeping with the trial protocol of course.

 

20 hours ago, fas42 said:

And there is a powerful need in many objectivists to find any loophole which means they don't have think deeper about what's going on.

 

I'm not sure what you mean by this. When we have the results of the trial, either they are statistically significant, or they are not. So we have an answer to our question either way.

 

20 hours ago, fas42 said:

There have already been attempts to do this sort of thing more rigorously; and the backlash by those who, say, belong to the AES has been very strong. With the desire to push it under the carpet.

 

Yes, all throughout history there have been backlashes to progress. When it was first suggested that our Earth is not at the centre of the universe, there was a backlash. So what?

 

20 hours ago, fas42 said:

When an attitude has built up over many decades, like this, it will take a great deal of effort to break it down  - no single experiment has got a hope of getting somewhere, no matter how well it's done ...

 

A single experiment will yield results on what it was designed to show, if done properly. That is all we are talking about here. If there are people with an "attitude" (presumably who don't understand scientific trials), that would be their problem. And of course, like in other fields, these trials would be repeated, and other trials would be done, to build up an "evidence base".

[extracampine]

Interesting discussion.

 

12 hours ago, fas42 said:

Changing the configuration of a system, any sort of system, audio or otherwise, in any sort of environment, is highly likely to alter its behaviour in some area. Which may be easy or hard to detect, using the human senses, or measuring devices.

 

But not necessarily likely to lead to reliable audible differences, hence the need for a DBT to answer that question properly.

 

12 hours ago, fas42 said:

It seems only in the audio world is there a bizarre need to prove that this can happen, using such methods as DBTs - for items which don't meet the approval of those with an objectivist leaning.

 

No, like I said, in any field where it actually matters (take developing medications as an example) there is a need to prove it can happen. It's not bizarre at all. And if you're looking at parting with hundreds or thousands of $ for audio equipment then it's not bizarre either that we would want to know if something is better or not. 

 

12 hours ago, fas42 said:

So, if you make a change to your system, in any area, and you think it sounds better, then you're shortchanging yourself unless you do a comprehensive DBT to confirm this?

 

No, of course not. Individuals are free to do what they like with their own systems. The point is that the fact that you think it sounds better could quite possibly be due to placebo/expectation bias/etc - so other people can not make any conclusions from it. This is fine for you in your own home if you're not bothered about that - but when we're talking about the consumer market and expensive purchases, then many will want reliable information to guide them. 

 

12 hours ago, fas42 said:

The "So what?" is that progress in thinking, and real changes how things are done is severely hindered for long periods of time, unnecessarily. Historically, you wait for those who have "bad thinking" to drop dead - because movement occurs in the youngest generation, who are not handicapped by "set in concrete" ideas .

 

I'm not entirely sure on your point; you had said that there had been attempts to do this sort of thing already (DBT) and that the backlash from certain groups (e.g. AES) has been strong. I had countered by saying so what if there is a backlash, it's all in the name of progress. You seem to be agreeing with me on this one.

[extracampine]

14 hours ago, fas42 said:

What do you want to 'prove'? That a mobo that happens to generate lots of extra electrical noise, as compared to one that is less prone, may cause audible changes for a particular combination of audio components? Or that the mobo that is the subject of this will cause a change for any audio rig? Or that this mobo may cause a change is some setups?

 

Prove is not the correct word, as you can never prove something 100% - instead you can show something to be highly likely, with good statistical power and small confidence intervals. What exactly you want to show would be up to the person who designs the study. You could aim to show any of those things that you suggest. In terms of our discussion, your first suggestion would fit best - whether the motherboard causes any audible changes for a particular combination of audio components. You raise a good point - ideally you would conduct this study with a few different combinations of audio components, in case there is any spurious result. 

 

14 hours ago, fas42 said:

The first objective is silly - it will always be possible to find a really noisy motherboard, that will impact some really badly implemented components.

 

Well, I wouldn't be so sure. The concept of electrical "noise" is often spoken about on audio forums, yet the evidence seems to suggest that this doesn't audibly affect the SQ. A DBT like we were discussing would answer this question. 

 

14 hours ago, fas42 said:

To me, this is all in the realm of, say, buying a car that won't break down over a really bad bit of road - we go to great lengths, in a scientific study, to find a car that doesn't break down ... now, is that meaningful in the grand scheme of things?

 

Yes, very. If a (valid) study is done to find a car that doesn't break down, that would be meaningful. The chances of that car breaking down when you are driving it would be significantly less - but never zero, which links in to my point above about the word "prove".

 

14 hours ago, fas42 said:

It's the "something is better" angle that's everything here - does it have to be better for absolutely everyone who may possibly consider buying it, or does it only have to be better for a few individuals?

 

Now you're talking about statistics and statistical power. This depends how certain you want to be on your results. The more people you include in the study, the more powerful it will be and the more valid the results will be. So, again similar to my point about the word "prove", it would highly unlikely be "better for absolutely everyone" - nothing ever is, that is an unrealistic goal. So for example, statistically it might be better for 95 out of 100 people, which is a vast majority when it comes to audio!

 

14 hours ago, fas42 said:

If a medication 'works' for only 50% of those who try it, is that a product a scam, because so many people have wasted their money?

 

Well, medications don't just work or not work; it's not as simple as that. They increase or reduce the risk of things from happening. So you could, for example, be taking a medication to reduce your risk of heart attack (e.g. if you were obese and a smoker) - and it might reduce your risk from 70% to 50% - and you could still have a heart attack. But that doesn't mean that the medication hasn't worked; it has still reduced your risk. The question is how certain can we be that it reduces the risk from 70% to 50%. To draw an analogy with audio, if we can be highly certain that a component increases the SQ from 60% to 70%, say (to use some arbitrary measure of SQ) then that information is useful. 

 

14 hours ago, fas42 said:

Reliable information? In the audio world? ... You've got me rolling on the floor with that one ...

 

Yes, isn't that the whole point of this discussion? We require more studies such as DBTs to obtain this reliable information!

 

14 hours ago, fas42 said:

Yes, progress will happen ... what irks me is that the pace of such is so slow - luckily, the manufacturers of the raw ingredients of audio systems have not been standing still, in the meantime; meaning, that very low cost setups are vastly better in key areas than they were say 30 years ago.

 

You may well be right, I've not really looked into that. You'd have to do a DBT... 😂