This of course opens up the interesting point that the overlap in personality types between people doing business degrees and Engineering degrees may not be that wide and so the training material and approach is likely to need to be completely different to have any chance of success. And that's before you take into account that a percentage of Engineering undergrads chose that course as it minimises the required interaction with other humans!
We sold class D amplifiers to Sony for a number of years. They asked us to retrofit one of their products with our demo board which we did and then took it back to them at their audio labs in Shinagawa, Tokyo. They set the product up through some speakers and about 7 or 8 of their engineers sat in the chairs and listened intently for about half an hour and then gave us a critique.
All smart guys (EE's) and the lab was kitted out with B&W (Nautilus), ML, Accuphase, Luxman etc - so they did in the course of their product development listen to all sorts of really high end stuff.
When a Japanese guy tells me the engineers listened and they (dis)liked this or that, I usually believe them. Same procedure BTW at Pioneer and Panasonic.
Why do you mention especially business degrees at this point? (I've to admit to think that philosphy of science would be of interest to nearly anyone, though)
I'm sorry, but I'm completely at a loss wrt the last sentence.....
Jakob: Some of the labs I have worked in we joked could have been aspergers research centres. This is not being degrading to engineers but many of them are further along the scale that has autism* at one end and whatever counts as 'normal' at the other. I've worked with experts in filter design who would forget to take their pyjamas off before getting dressed for work. Compare this with the wannabe CEOs doing business degrees (that Mark noted get taught about bias).
So the (intended humerous) point was that you have 2 collections of people with one group having the extreme of being trainee sociopaths and the other group having the extreme of people who talk to their spectrum analysers. Even outside the extremes you have very different outlooks so how do you get the same concept to sink into both of them.
*Edit: not up to date with the latest research so apologies if I don't even have the correct current term for this.
So the (intended humerous) point was that you have 2 collections of people with one group having the extreme of being trainee sociopaths and the other group having the extreme of people who talk to their spectrum analysers. Even outside the extremes you have very different outlooks so how do you get the same concept to sink into both of them.
*Edit: not up to date with the latest research so apologies if I don't even have the correct current term for this.
@ Bonsai,
the reactions to such informations (of course often anecdotal, but first hand) illustrate the fact that information processing is often mainly bias/belief driven.
If a manufacturer does not (or is beliefed to do not) any subjective listening it is taken as evidence for honesty/scientific approach -and of course taken at face value.
If a manufacturer is relying (also) on subjective listening it _must_ marketing blurb. (Because no serious manufacturer would do it that way)
Some typical examples to find in the recent posts.......
Next step, if evidence for the inclusion of subjective listening is overwhelming, will be to conclude that it is only done for marketing purposes (but surely they do not believe in that).
One would think it is obvious, that some logical inconsistenicies/fallacies are the basis of this kind of conclusion routine........
the reactions to such informations (of course often anecdotal, but first hand) illustrate the fact that information processing is often mainly bias/belief driven.
If a manufacturer does not (or is beliefed to do not) any subjective listening it is taken as evidence for honesty/scientific approach -and of course taken at face value.
If a manufacturer is relying (also) on subjective listening it _must_ marketing blurb. (Because no serious manufacturer would do it that way)
Some typical examples to find in the recent posts.......
Next step, if evidence for the inclusion of subjective listening is overwhelming, will be to conclude that it is only done for marketing purposes (but surely they do not believe in that).
One would think it is obvious, that some logical inconsistenicies/fallacies are the basis of this kind of conclusion routine........
Point taken (and agreed) that different presentation styles would generally help people to fully grasp the ideas. (Universities seem to think that people have to adopt to their presentation style, though - at least in Germany 🙂 ), but as this subtopic was related to the "engineers" I was wondering why of all fields the business wannabe CEOs were mentioned.
Ok, ok, conveying of the ideas might face different "hurdles" but generally it shouldn't be that different from let's say mathematics.
Last edited:
Just my inability to quote Mark's post and your response together where he mentioned how was currently being taught this.
When I did a foundation economics course at school the teacher pointed out that one part of the syllabus for university entrance involved some fairly chewy calculus, but the students were not told about this until after they had mastered the technique otherwise they would have been scared off trying. Not to say that there are not a lot of economists who are brilliant mathematicans but pointing out there are hurdles in many fields. And as a species we are a diverse bunch, which is how we got here in the first place!
MFB topology uses inverting configuration, which means CM performance isn't critical. In complex crossovers (not just cascades of S/K) several opamps function in different role, which means each role asks for different opamp. Some opamps are also very good for HF but unusable for LF (and vice versa).
I’m assuming something like 500-2k to eliminate the room?
Thanks George,
So this would be ‘full bandwidth’ pink noise generated?
And not to sound stupid but how exactly do you implement changes to ‘match’ the amps >imp vs freq to the speaker?
Of course, and this is why DIY is useful, however if you are making a product it has to have a USP, and with an audio product it's often the sound, and then that needs to be sold.
Ahh.....am a bit slow (?today?)
Have to memorize, don't quote a complete post (Markw4's in this case) if you just want repond to the first sentence...... 🙂
And yes diverse we are.
Yes, that's right, you knew I knew that. 🙂 But it still matters. Read the data sheet and all the specs, you will see that it covers all the bases better than most.
The only thing it can't do is drive a really low Z load, which shouldn't be an issue in this case.
T
Thanks Jakob,
I’m looking at getting a RME adi-2 dac that has quite a bit of dsp based adjustment which includes 5 band peq. Between that and measuring fr at lp I should be able to come up with my own version of rew.
I’m not sure if I remember your ‘peqs’ references.
I’m still sticking to my guns that flat is not where it’s at.....I think it will be nice at least to see input (RME has a 30 band spectral analyzer) vs what’s happening @ lp.
@ mountainman Bob,
the links related to PEQS:
User's Handbook
https://tech.ebu.ch/docs/tech/tech3287.pdf
and
Sound examples
https://tech.ebu.ch/files/live/sites/tech/files/shared/testmaterial/PEQS_FLAC.zip
the links related to PEQS:
User's Handbook
https://tech.ebu.ch/docs/tech/tech3287.pdf
and
Sound examples
https://tech.ebu.ch/files/live/sites/tech/files/shared/testmaterial/PEQS_FLAC.zip
Looks like a good way to train your ears......assuming your system is accurate to start with!
That should be an interesting study to make, for someone like you, Richard, because you seem to be well balanced and well equipped between measurements and listenings.
Comparing the few very low distortions OPAs, like the LM4562 VS AD AD797 VS strait line.
Doing the same with CFAs like THS3092.
Then comparing the best VFA VS the CFA vinner (If any).
Why this ? I don't believe the major difference between the two topologies is about the slew rates, As most of the modern OPAs are fast enough, but their compressive VS expansive behaviors.
We could make 2 testing boards with 4 OPAs each, a 1/10 attenuator followed by a Gain of 10 non inverting, followed by gain1, inverting + attenuator 1/10, followed by an gain 10 inverting, followed by a gain 1, non inverting.
Could be complicated as often CFAs require lower feedback impedances and some compensation caps for VFAs
Getting out of this topic for some days is refreshing. Reading-it back revealing.
People, here, are all situated on a social line between very few that believe in Magic and Voodoo and those who believe that measurements tells everything in audio, no listening out of ABX etc.
I noticed that the people situated between the middle and the right of this line are the most intolerant aggressive and impolite ones. Why ? Pseudo science makes people frustrated ?
And I asked myself a question. The designers that are at the right of this line do not listen to their changes at each step of their work ? Or do they listen music via LTSpice and co ? And, if they do listen on their bench, do-they really set an abx process at each listening to ensure any progress ? Honestly !
Oh, BTW, Evenharmonics, could-you stop with your "Matched level" obsession ?We all practice-it as a basic requisite (since more 50 years for me) !
Last edited:
Hi Richard thanks for the pic and your review.
What was the cost of the mod to your DAC-3 ?.
Dan.
- Status
- Not open for further replies.