I wanted to comment on this point as I missed it the first time around.
This is absolutely true and I have felt this same exact thing myself very often. This idea that "It can't hurt.", "Everything is important" and "optimize everything" is a real detriment to any discussion. Loudspeaker design is all about tradeoffs and these extreme points of view just do not get us anywhere. I have often said how much more impressed I am with a Honda than a Ferrari, because the Honda shows a lot more engineering. Its a great car that lasts a long time and sells for very little. Its not too hard to make a car when cost is no object.
Likewise I am not impressed by a "good" speaker that costs $20,000. That's just not necessary and it is this kind of thing that has seriously hurt audio. The idea that if you haven't done it all you haven't done enough. The consumer just goes - Oh well then there just isn't any point in getting involved.
Ranking of priorities is a must. You must know what is important and what is not and how the different factors rank. All-or-nothing just isn't a good approach to anything let alone audio.
Engineers in other fields would use that as the basic definition of "good engineering". Sadly it has disappeared from our business.
For the most part I think this just reflects the current state of the market. Customers either want the cheapest possible product or they are chasing after a luxury good and want to be convinced that cost was no object and the product couldn't be improved upon (ego gratification?). A middle ground of "good value" is sadly missing.
This is part of why I admire the designs of Henry Kloss (KLH, Advent, Tivoli).
Regards,
David
Read up on Ken Kantor's "Magic Loudspeaker", as this is a pretty good description of it.
For the most part, Earl's preference for a more directional system in a more lively room, with careful aiming to control the early reflections, is also seeking the same effect.
I think there is research consensus that later reflections from wide horizontal listening angles can give a good sense of envelopment that 2 channel 2 speaker reproduction generally lacks. The key seems to be that early reflections, especially from angles near the loudspeakers sources (and especially if only seperated by a vertical angle) need to be minimized. If the early field is fairly dry and the later field is busy and includes much energy from wide, low interaural correlation angles, then this gives a good result.
I play with artifical reverb from time to time with a 4 speaker setup. 2 normal speakers are at a typical +- 30 degrees position, The 2 reverb speakers are well to the sides at more like +-60. I like the effect and it tends to draw the sound out of the speakers. Not night and day, but generally a good thing.
With two speakers only, in a conventional setup, people are generally forced to chose either clarity from high directivity and very little room effect, or more room effect but with a loss of clarity with more omni speakers.
More choices would be nice.
David S.
Hold on - not true! There are many many really good commercial speakers that don't cost a fortune, and manufacturers who focus on the relevant. If we disregard sub 60Hz even some 500$/€ for a pair gives fantastic returns (gain/cost ratio). If we want really fullrange and even output from 30Hz up - it gets difficult but perhaps 1200 for a pair.
The problem lies on how to recognize a good speaker! The pricetag really doesn't tell that! As we (informed) know there are several factors and measurements needed for evaluation. And we must set our personal priorities and remember the room environment intended for the speakers!
I respect the efforts of Soundstagenetwork and Stereophile (this decade) on publishing relevant measured data of speakers from all price categories. Our national hifi-magazine never publishes data of a speaker costing more than 2000€
Amen, Earl.
But the trouble comes in what we choose to pay attention to, and what we ignore.
That's why "rank order" is critical. You pay attention to those things that rank high and ignore those that don't.
I think it is better to extract the 'reverb' from the stereo signal itself and use that and that only instead of artificial additional reverb.
Three or more speakers are of course better than two for stereo reproduction.
Two speaker stereo is dead
Pano
As I said, for the most part the entire panel of "experts" noted above were in agreement on about 80-90% of the issues. I listed the ones that were not in agreement. Neither you nor anyone else need to accept what "we" think, but as a group "we" represented a huge amount of knowledge. Writing off those "opinions" would be very arrogant.
I think that Dave outlined the different positions on the undecided 15%.
As I said, for the most part the entire panel of "experts" noted above were in agreement on about 80-90% of the issues. I listed the ones that were not in agreement. Neither you nor anyone else need to accept what "we" think, but as a group "we" represented a huge amount of knowledge. Writing off those "opinions" would be very arrogant.
I think that Dave outlined the different positions on the undecided 15%.
I was recently re-reading this part of the book and it made me wonder if the average DIYer can meet the targets found in Olive's study. I'm almost convinced that with outdoor measurements coupled with modeling, you could pull it off.
Can anyone here identify some of the highest ranked speakers in Toole's and Olive's testing? I just want to look at how they look, driver complement and hazard the crossover topologies and so on. Basically, try to make a cheapo clone of the design
I actually did say that it is the last few % that makes for the kind of system that we all seek here. But as such, I see no point in arguing the points that "the expert" all agree on.
When we talk about directivity and room interactions those are areas that we do not all agree on and it is those areas that make for the differences that we hear in "good" designs. One can get to "good" (80-90%) pretty easy these days, but "great" takes something more - that last few %.
Can anyone here identify some of the highest ranked speakers in Toole's and Olive's testing? I just want to look at how they look, driver complement and hazard the crossover topologies and so on. Basically, try to make a cheapo clone of the design
Coincidentally, they are mostly Harman products. The thing is that the devil is in the details. The Harman products tend to be refined designs - attention to details - things that require substantial measurements and the ability to control the design. Its possible for a DIY to do this, but it is not as easy as just looking at the "driver compliment" and the crossover topology and copying it. That's a common DIY fallacy.
re. More choices
The issue is being stuck with 2 channels.
"5 channels = home theater = bad"
"Music must be 2 channels."
With this thinking we are always stuck with some manner of compromise: you take higher directivity for clarity, I take wider dispersion for spaciousness, etc. Neither is the total solution. Only with more channels can we chose a wider pallet of sound spaces and recreate real acoustic spaces with the requisite reflections from all directions and with large hall timings.
David S.
The issue is being stuck with 2 channels.
"5 channels = home theater = bad"
"Music must be 2 channels."
With this thinking we are always stuck with some manner of compromise: you take higher directivity for clarity, I take wider dispersion for spaciousness, etc. Neither is the total solution. Only with more channels can we chose a wider pallet of sound spaces and recreate real acoustic spaces with the requisite reflections from all directions and with large hall timings.
David S.
Well, if there is indeed more to it, and maybe there really is, then it would be good start to at least know what these products are. Anyone?
I've looked through SoundStage Network measurements and apart from the KEF Reference 201/2 and Revel Ultima Salon, I don't see any other measurements that resemble the ones Toole or Olive show. Where are these loudspeakers, especially the affordable $1200 ones?
Last edited:
Ummm... no, no, no. Have you seen the on-axis response?
And no, I don't want to EQ the on-axis just because the off-axis follows the on-axis. The driver responses themselves are poor.
We are talking smooth driver responses to start with. Only then can you achieve the targets set down in Olive's study.
And no, I don't want to EQ the on-axis just because the off-axis follows the on-axis. The driver responses themselves are poor.
We are talking smooth driver responses to start with. Only then can you achieve the targets set down in Olive's study.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.