-well remember that Toole’s book is a synthesis of work from many different researchers and research projects.
Yes, it also includes his own research (and particularly Olives), but that research is at least in part “built” on the work of others.
Caution: most subjective testing is flawed (at least in some respect, if not many respects) - so accepting the conclusions as fact can sometimes lead to an erroneous outcome (though general conclusions from good research is likely to be more accurate than not.) Of course even objective testing has similar problems with a range of methods on how to obtain data and how to interpret it.
Yes, it also includes his own research (and particularly Olives), but that research is at least in part “built” on the work of others.
Caution: most subjective testing is flawed (at least in some respect, if not many respects) - so accepting the conclusions as fact can sometimes lead to an erroneous outcome (though general conclusions from good research is likely to be more accurate than not.) Of course even objective testing has similar problems with a range of methods on how to obtain data and how to interpret it.
Last edited:
Ehm?
ALL scientific research and testing is a matter of statistics, with a certain margin of error and only valid within a certain context.
So when people immediately accept any conclusions as direct facts, they're simply already making a mistake to begin with.
Understanding the context of any research and experiment is absolutely key and mandatory, whatever the outcome is.
There are several things in Floyd's book that are skimmed over or very simplified.
Things like system design and low frequency response just to give one simple examples.
But that doesn't mean that the entire book is garbage?
That was also not the point of his research.
I sometimes get a feeling that for some people it's either black or white.
But that's not quite how these things work and go.
That also goes both ways btw, for (quite) some people (on a specific forum) this seems to be the holy bible that can't be touched and it's a total sin to have any type of feedback or critique on.
In that case I would suggest looking up the definition from science again and start from the beginning.
I totally blame all the dishonest incompetent review magazines for the amount of bad speakers out there. I havent read one except TNT who didnt rate products overly positive to keep getting adds and free samples from distributors.
If only a few magazines had made competent and honest reviews thru the ages, we would have had much better products. I used to gob it up, but the little knowledge ive gotten through experience and through sites like this made me realize how naive and stupid I was trusting anything they write.
The only usefull review I remember was from the aforementioned TNT-audio who made me buy the XTZ A100D3. It was actually every bit as good as they wrote😊.
Cheers!
If only a few magazines had made competent and honest reviews thru the ages, we would have had much better products. I used to gob it up, but the little knowledge ive gotten through experience and through sites like this made me realize how naive and stupid I was trusting anything they write.
The only usefull review I remember was from the aforementioned TNT-audio who made me buy the XTZ A100D3. It was actually every bit as good as they wrote😊.
Cheers!
Start following Erin's reviews.
Also, be critical about it on a fair way, he would love that!
But omg, some of the things that come by, I sometimes literally wonder if there was actually ANY audio/sound engineering involved.
He already has. Starting around the 35 minute mark he uses a number of DIY designs to illustrate pitfalls of different full range drivers:
He could do better, for me the complete lack of distortion measurements gives pause. But it's fascinating to read criticisms here flatly and explicitly contradicted in almost every video, raised by people deeply deluding themselves that by doing so they're 'defending science'. It's wild and depressing to watch.
1/ The FRs i have seen are measured on IEC baffles not infinite baffles, so the good stuff doesn’t happen until 39 minutes
2/ the hump (W8-1808) is th e transition from pistonic reponse tranitioning to chaotic response. Chaotic response is controlled resonance. One peak is not an issue, a train of them is. The Audio Nirvanas are good examples of resonance trains. And he did a SUper 12, no impedance curve
3/ he takes a lot of time describing bipole dip. The same happens with a box on a wall. As the width of the box increases relative to the depth the depth of the dip decreases, a bipole is ideally wider than it is deep.
His comments re a cap on the back driver of a bipole is incomplete. One wants a low pass on the back driver, which means the drivers need to be wired in series with a big shunt cap.
4/ the Eikona did not measure as bad as i thot it would.
5/ Unity Audio uses the Solen Airbourne wood cone with whizzer. The one i did up for Cal with full enable sounds quite decent.
6/ first gen CHR-70. It got better in later versions. Not a big fan.
7/ he then touts his LGK, but as a WAW bipole. The stored energy signature is VERY good.
Note that the spectal decays used periods as time axis not mSec.
dave
2/ the hump (W8-1808) is th e transition from pistonic reponse tranitioning to chaotic response. Chaotic response is controlled resonance. One peak is not an issue, a train of them is. The Audio Nirvanas are good examples of resonance trains. And he did a SUper 12, no impedance curve
3/ he takes a lot of time describing bipole dip. The same happens with a box on a wall. As the width of the box increases relative to the depth the depth of the dip decreases, a bipole is ideally wider than it is deep.
His comments re a cap on the back driver of a bipole is incomplete. One wants a low pass on the back driver, which means the drivers need to be wired in series with a big shunt cap.
4/ the Eikona did not measure as bad as i thot it would.
5/ Unity Audio uses the Solen Airbourne wood cone with whizzer. The one i did up for Cal with full enable sounds quite decent.
6/ first gen CHR-70. It got better in later versions. Not a big fan.
7/ he then touts his LGK, but as a WAW bipole. The stored energy signature is VERY good.
Note that the spectal decays used periods as time axis not mSec.
dave
For (absolute) free, highly unlikely. (..though maybe if it sparks his curiosity.)
As part of the price for the re-work and parts - he has. Maybe he will continue to do so.
-so it sort of depends on what you mean by “free”.
..though perhaps the “sticking point” was “DIY”?
As part of the price for the re-work and parts - he has. Maybe he will continue to do so.
-so it sort of depends on what you mean by “free”.
..though perhaps the “sticking point” was “DIY”?
Watch this video and understand why many consider GR 'research' to be largely a hoax
A couple of tidbits in the video are that Danny at GR 'research'...
1) spends 20-30 minutes designing crossovers for the speakers he is 'upgrading'. Bear in mind that the original designer of those speaker, probably spent many hundreds of hours or maybe thousands, to voice them. But for Danny it only takes 20 minutes. And he doesent even have to listen to them 🙂
And.....
2) He never listens to the 'upgraded' speakers! In any case, not both speakers, as he always receives only one (1) speaker from the customer.
I have been building speakers for almost 20 years. For me, it takes between 6-18 months to complete a project. But, maybe Danny knows something I don't? And, for that matter, something that no one else knows.....
Feel free to watch the video yourself.
A couple of tidbits in the video are that Danny at GR 'research'...
1) spends 20-30 minutes designing crossovers for the speakers he is 'upgrading'. Bear in mind that the original designer of those speaker, probably spent many hundreds of hours or maybe thousands, to voice them. But for Danny it only takes 20 minutes. And he doesent even have to listen to them 🙂
And.....
2) He never listens to the 'upgraded' speakers! In any case, not both speakers, as he always receives only one (1) speaker from the customer.
I have been building speakers for almost 20 years. For me, it takes between 6-18 months to complete a project. But, maybe Danny knows something I don't? And, for that matter, something that no one else knows.....
Feel free to watch the video yourself.
I don't know much about GR research or Danny. I also don't really care to much.
But to just respond to a few things.
I know people as well as companies who spend an awful lot of time, but the end result is and sounds still bad.
I don't know what you mean with "completing a project"?
If that means incl preliminary research, orderering parts, woodwork, assembly, measurements and tweaking in just your spare time?
Sure
But otherwise if it's just measurements plus an upgrade, I find that a bit much (no offense).
But I agree 20 minutes is maybe a little to short.
If I would estimate it for a client and there is REALLY a rush on it, I would estimate 2 working days if you want it also at least some reasonable guarantee on it. But that is really bare minimum, no additional investigation on other issues.
Far from ideal and definitely not the best you can get out of a system.
If a client is in uber rush and doesn't care about any fine details. I could do it in a day or a couple of hours.
Assuming that everything is ready and a measurement setup and other needed tools are accessible immediately.
Obviously no guarantees here.
Realistically, a couple of weeks is usually fine. Although, you never work on one single project at a time.
So in practice that quickly becomes a couple of months. You just don't spend all those months, 8 hours a day on one single project.
But to just respond to a few things.
The amount of hours is a pretty poor measure for sound quality in my opinion.
I know people as well as companies who spend an awful lot of time, but the end result is and sounds still bad.
I don't know what you mean with "completing a project"?
If that means incl preliminary research, orderering parts, woodwork, assembly, measurements and tweaking in just your spare time?
Sure
But otherwise if it's just measurements plus an upgrade, I find that a bit much (no offense).
But I agree 20 minutes is maybe a little to short.
If I would estimate it for a client and there is REALLY a rush on it, I would estimate 2 working days if you want it also at least some reasonable guarantee on it. But that is really bare minimum, no additional investigation on other issues.
Far from ideal and definitely not the best you can get out of a system.
If a client is in uber rush and doesn't care about any fine details. I could do it in a day or a couple of hours.
Assuming that everything is ready and a measurement setup and other needed tools are accessible immediately.
Obviously no guarantees here.
Realistically, a couple of weeks is usually fine. Although, you never work on one single project at a time.
So in practice that quickly becomes a couple of months. You just don't spend all those months, 8 hours a day on one single project.
BTW, I now watched a little of the video.
Depending on the type of speaker, but for your generic standard book shelf ported speaker design, you most definitely are fine with just a gated response for measuring.
It depends on how predictable a design or manufacturer is. But I do agree that a nearfield stitched method, plus measuring distortion etc is way better.
But that companies spend "hundreds and thousands of hours on R&D", isn't saying much. On itself for just a speaker filter, that is a total BS argument.
As a generic thought, sure fine, but like I said before, your daily tasks involve a lot more than just one speaker design, let alone one filter design.
What people seem to totally forget here, is that if you're in the position like Danny is, you don't have to care about a brand new design from scratch.
You just get an already existing design to work on, so any flaws or other problems are already given.
For passive filters you also can't tweak the low end anymore.
So it's only out of interest for the total bigger picture to know what the low end does. But it can't be changed anymore in a passive design.
Getting a single speaker without listening is a bit tricky.
Although, I myself, have enough experience to make that work to some extent.
I would only NEVER do that without spending some time with the client if he/she likes that as well.
Or in other words, let the client be the ears and me just only focus on the filters.
I have done that in the past a few times actually. Since I can't get an impression on the sound and give direct feedback, it's not ideal.
Although, if a client is experienced enough in listening (and knowing the right words), it's doable.
But it's always doing that together.
Just creating a filter straight out of measurements and calling it done wouldn't be my choice.
Mostly because you have no clue what sound the client likes (in their room).
Depending on the type of speaker, but for your generic standard book shelf ported speaker design, you most definitely are fine with just a gated response for measuring.
It depends on how predictable a design or manufacturer is. But I do agree that a nearfield stitched method, plus measuring distortion etc is way better.
But that companies spend "hundreds and thousands of hours on R&D", isn't saying much. On itself for just a speaker filter, that is a total BS argument.
As a generic thought, sure fine, but like I said before, your daily tasks involve a lot more than just one speaker design, let alone one filter design.
What people seem to totally forget here, is that if you're in the position like Danny is, you don't have to care about a brand new design from scratch.
You just get an already existing design to work on, so any flaws or other problems are already given.
For passive filters you also can't tweak the low end anymore.
So it's only out of interest for the total bigger picture to know what the low end does. But it can't be changed anymore in a passive design.
Getting a single speaker without listening is a bit tricky.
Although, I myself, have enough experience to make that work to some extent.
I would only NEVER do that without spending some time with the client if he/she likes that as well.
Or in other words, let the client be the ears and me just only focus on the filters.
I have done that in the past a few times actually. Since I can't get an impression on the sound and give direct feedback, it's not ideal.
Although, if a client is experienced enough in listening (and knowing the right words), it's doable.
But it's always doing that together.
Just creating a filter straight out of measurements and calling it done wouldn't be my choice.
Mostly because you have no clue what sound the client likes (in their room).