Salk SoundScape 8

[AcuDefTechGuy]

I'll ask Jim to post the off-axis stuff...

I am not qualified to rebuttal to this, so I'll pass it along to Jim and Dennis and see what they have to say.

It would be nice for all fans of speakers to see more SS measurements since there are no 3rd party measurements.

Are the parts (drivers, crossovers, cabinets, etc) on the Revel & KEF cheaper and lesser quality than the SS?

 

[AcuDefTechGuy]

...you haven't heard them.

Doh!

As my 9 yr old daughter would say, "Yeah.............. I guess you're right."

My rebuttal: Oh, yeah, well..........I do like the shiny red cabinets.

 

[Nuance AH]

^ LOL! I just hope no one is taking us serious here, as this is all just in good fun. This hobby is too subjective to listen to other people's opinions.

I really did ask for the rest of the measurements to be posted, though. These are literally brand new speakers, so it may take a little time, though.

It would be nice for all fans of speakers to see more SS measurements since there are no 3rd party measurements.

You bet!

Are the parts (drivers, crossovers, cabinets, etc) on the Revel & KEF cheaper and lesser quality than the SS?

I doubt it, but your argument has about as many holes as mine so I went for it.

 

[Swerd]

We now wait for the Salk Nation rebuttal.

This is too tempting to resist.

1) Besides the rounded edges, it is rather a very square shaped speaker. Some people will argue it is not the best shape for "optimum" sound reproduction (not I since I don't know much about speakers building theories ). Grant & other can chime in on speaker shapes. Like shapes of Salon2, KEF Reference, B&W Diamond....

Salk Nation Response
This is a funny argument (that the SS8 is very square shaped) because there are those (Is ADTG one?) who complained about the odd shape of the upper module of the SS10/12. They objected to the “Thomas the Tank Engine appearance”.

A simple response is that the SS8 looks rather square because the front baffle must be wide enough for those 8" woofers. It also is less expensive to construct a boxy-looking cabinet than to build the complex shape of the SS10/12. The SS8 is also a lot less expensive than the Salon2, KEF Reference, B&W Diamond, etc. More importantly, it is less important for a speaker’s sound just what shape a cabinet has than it is for that shape to be present and accounted for when the crossover is designed. The front baffle width, whether the tweeter is off-center or not, how much the cabinet edges are rounded over, and whether the cabinet edges are parallel or not, are all features that can affect some part of the speaker’s sound. If they are present when the raw unfiltered behaviors of the various drivers are measured, a good crossover designer can equalize the minor undesirable traits those cabinet features may or may not make to the overall sound.

2) It has a single on-axis measurement. One. The jury will not look favorably on an $8/9K speaker with just one measurement when a $1.6K Philharmonic speaker has like 10 times that?

Salk Nation Response
This one has been driven into the ground – repeatedly. How many manufacturers, other than Salk or Philharmonic provide even one good on-axis frequency response curve? It will probably be easy to obtain some off-axis measurements, just ask.

It is more difficult to measure the effect of the partially open-backed midrange driver. As you vary the stuffing, how does this change soundfield depth?

3) There is not a single, not even one 3rd party measurement or review.

Salk Nation Response
We would all like to see some reliable 3rd party measurements. Considering that the SS8 was very recently announced – there is one owner at present – 3rd party measurements and reviews will have to wait.

I can make a personal observation here (more like a Dennis Murphy Fanboy Response, than a Salk Nation Response). I have seen Dennis make enough frequency response measurements to believe that his methods and tools are quite reliable and honest. Dennis probably believes this too, but he will never say so in public. His ability to make such excellent designs depends entirely on his ability to reliably and accurately measure speaker performance. He would love for 3rd party measurements to confirm what he measures himself.

4) Some folks believe the BG Neo8 is better than the Accuton (like Bob).

Salk Nation Response
Some folks believe that Bose is better, that bright shiny metal-domed tweeters sound bright, that speakers horns all sound like cupped hands around a driver, etc. There is no accounting for what “some folks believe”.

It is more important that a driver be properly used in a speaker design so that distortion or break-up noise are not included in the pass band, and that the crossover allow for good off-axis dispersion at the high end of a speakers pass band. If I were a smarter Salk Nation Spokesman, I might comment about in-phase or out-of-phase behavior of two drivers through their crossover range. If Dennis cares, he can chime in.

5) Big giants like Harman & KEF have spent millions in research & development, while Salk uses OEM DIY drivers that anyone and their neighbors and high-school students can use. Salk does have one weapon that neither Harman nor KEF has -- Dennis Murphy Crossovers.

Salk Nation Response
Dennis does a lot more than just design the crossovers. He chooses which drivers are used so that crossovers can be properly executed. And there is nothing wrong with OEM DIY drivers. In fact, one reason why the big companies avoid using them is so they can claim their speakers use “custom made drivers”. That way no clever DIY builder can clone them.

 

[Dennis Murphy]

This is too tempting to resist.

I think we need to design a flag for Salk Nation. But I'm too busy nearing completion of a nuclear device. Anyhow, I would have been glad to take some off-axis plots, but a big truck drove up yesterday and carted away a ton (literally--exactly one ton) of old mule cabinets I've used over the users to design crossovers. The SS8 was the last to go. I knew this would happen. I would not be surprised if there was a little flare at the bottom of the tweeter range--I think I already mentioned that. I added a tuning resistor to help compensate for that in the on-axis response, but I doubt that the measurements will be perfect. Merely better than most of the measurements I've taken of other speakers.

 

[AcuDefTechGuy]

I vaguely recall someone mentioned that some of the high-end speakers parts were not really so high-end. Maybe some caps or something?

Is it inappropriate to ask what the ratio is on actual cost vs speaker price?

If I put in $2K parts, is it fair to sell for $4K?

We have a good idea when it comes to KEF, Revel, B&W, Paradigm, etc. The mark-up is probably 3 times cost?

I think Dennis mark-down his speakers as charity.

 

[GranteedEV]

o

5) Big giants like Harman & KEF have spent millions in research & development, while Salk uses OEM DIY drivers that anyone and their neighbors and high-school students can use.

The opposite also true. Harman/KEF limit themselves to where their research has gotten them and might actually ignore important concepts. That means no RAAL Ribbon Tweeters, no Open Back mids, no Open Baffle Dipoles...

Plus I can't Salon2s in my choice of custom finish.

And a few people have noted that the KEF Blades sounded forward and bright at CES. Gene has noted that JBL's Bi-Radial horn sound colored.

B&W has all the research money in the world, yet TLS Guy thinks his speakers produce tighter bass.

Focal makes some of the most ostentatious speakers out there, yet no measurements to back up the performance exist. Some even say they're not very good products.

How about Paradigm... do you looove Paradigm because they make their own drivers with lots of research money?

 

[Swerd]

I vaguely recall someone mentioned that some of the high-end speakers parts were not really so high-end. Maybe some caps or something?

Please don't start a discussion of "high-end caps" . Believe me, you really don't want to go there.

We have a good idea when it comes to KEF, Revel, B&W, Paradigm, etc. The mark-up is probably 3 times cost?

It may be much more. It's not uncommon for a retail dealer's mark up to equal the distributor's price. So the final retail price might easily be 5× the manufacturing cost.

And that is for moderately priced speakers, let's say in the $200 – $2000 per pair price range. For higher priced speakers, $2,000 – $20,000 and up, I wouldn't try to guess.

 

[Nuance AH]

B&W has all the research money in the world, yet their speakers still suck.

I fixed it for you.

 

[DS-21]

***As for the measurements, I'll ask Jim to post the off-axis stuff, just for you, even though you're "hatin'" on his stuff.

Dr. Murphy already commented on the poor directivity control at the bottom of the tweeter's passband. See below:

***I'm pretty sure they would show a slight elevation in the lower highs significantly off axis horizontally (40 degrees and more), due to the superior horizontal dispersion of the tweeter.***

We just disagree as to the relevance of that acknowledged problem. Which is fair enough. And depending on how high the mid/ribbon crossover is, it may in fact be sonically a very small problem at best. (The lower the crossover, the more deleterious a "mushroom cloud" polar pattern is.)

As an aside, I do agree completely with what Dr. Murphy wrote in the rest of that paragraph: "I'm not convinced that it's worth building in a saddle in the on-axis response in order to achieve a mathmatically perfect summation of the polar response of a speaker."

Clearly, the proper answer is not to hack up the crossover, but to physically control the directivity of the tweeter at the bottom of its passband such that it matches that of the next driver down at the top of its passband. That's what Revel does, with the EOS waveguide. That's what KEF and Tannoy do, with (respectively) the Uni-Q and Dual Concentric drive-units. That's what Danley Sound Labs does, with the Synergy Horn.

The opposite also true. Harman/KEF limit themselves to where their research has gotten them and might actually ignore important concepts. That means no RAAL Ribbon Tweeters, no Open Back mids, no Open Baffle Dipoles...

One of those is a IMO flavor of the month at best, and not a concept. Those of us who have been around for a while have read the exact same nonsense about Ravens and then LCY's, only to realize upon hearing them for ourselves that they're a bunch of very expensive hype and mediocre sound. Ribbons are a fad that will probably always wax and wane.

And IMO they don't do open back stuff because it often hurts more than it helps, outside the bass. Especially when the midrange's rear wave is firing through a tunnel.

Though OB bass a la Gradient/Celestion ca. 1985/Linkwitz/John K does allow one to use fewer modal region sources, albeit at the cost of less placement flexibility.

Even in the aspect their proponents claim to be OB's forte, imaging, you may recall a blind test conducted by David L. Clark with his Detroit-area audio club (SWMWTS or something like that) that found no significant differences in apparent source width between a Linkwitz Orion, a Behringer B203something, and a DIY omnidirectional design, when all were EQ'ed.

 

[Nuance AH]

Clearly, the proper answer is not to hack up the crossover, but to physically control the directivity of the tweeter at the bottom of its passband such that it matches that of the next driver down at the top of its passband. That's what Revel does, with the EOS waveguide. That's what KEF and Tannoy do, with (respectively) the Uni-Q and Dual Concentric drive-units. That's what Danley Sound Labs does, with the Synergy Horn.

Gotcha. Although, Revel's Salon2 doesn't stay flat off axis in the same range Dennis was talking about, right? There's a rise at 3KHz. Or am I misunderstanding?

45, 60 and 75 degrees off axis:

Average of five measurements (up, down and left and right):

For comparison, here's the Philharmonic 2, which uses the RAAL ribbon tweeter:

On-Axis:

40 Degrees Off-Axis:

50 Degrees Off-Axis:

 

[DS-21]

Gotcha, though I'm not sure how it applies to an off-axis measurement. Revel's Salon2 doesn't stay flat off axis in the same range Dennis was talking about, right? Or am I misunderstanding?

No, like any good controlled-directivity speaker the big Revel progressively rolls off off axis, while keeping the same general response shape through the mids.

With the Revels, the pattern holds until about 10kHz, where things start narrowing due to the size of the tweeter's piston. From Stereophile:

In my experience, I can tell at a glance if there's a chance I'm going to like a speaker or not by looking at such a graph. If it displays a mushroom cloud midrange*, like the horrible-sounding and overpriced Magico speaker whose horizontal polars follow (again from S'phile):

then there is simply no chance I am going to like it. If the horizontal front hemisphere measures smoothly, like the Revel above, or the KEF 201/2 below (S'phile again for internal consistency):

then it's probably worth a listen.

That said, there's a good body of research that suggests off-axis holes in the response are little cause for concern. While that may be (read: probably is) true, the same is certainly not true for peaks of the type I'm discussing.

*"Mushroom cloud" is the term I use, because that's what a spectrograph of the response would look like.

 

[Nuance AH]

No, like any good controlled-directivity speaker the big Revel progressively rolls off off axis, while keeping the same general response shape through the mids.

With the Revels, the pattern holds until about 10kHz, where things start narrowing due to the size of the tweeter's piston. From Stereophile:

In my experience, I can tell at a glance if there's a chance I'm going to like a speaker or not by looking at such a graph. If it displays a mushroom cloud midrange*, like the horrible-sounding and overpriced Magico speaker whose horizontal polars follow (again from S'phile):

then there is simply no chance I am going to like it. If the horizontal front hemisphere measures smoothly, like the Revel above, or the KEF 201/2 below (S'phile again for internal consistency):

then it's probably worth a listen.

That said, there's a good body of research that suggests off-axis holes in the response are little cause for concern. While that may be (read: probably is) true, the same is certainly not true for peaks of the type I'm discussing.

*"Mushroom cloud" is the term I use, because that's what a spectrograph of the response would look like.

Okay, I completely understand that. Thanks.

 

[Swerd]

… In my experience, I can tell at a glance if there's a chance I'm going to like a speaker or not by looking at such a graph. If it displays a mushroom cloud midrange*, like the horrible-sounding and overpriced Magico speaker whose horizontal polars follow (again from S'phile):

then there is simply no chance I am going to like it.

Is the "mushroom cloud midrange" that you describe, the peak centered somewhere between 3 and 4 kHz as shown for the Magico speaker?

 

[Dennis Murphy]

Okay, I completely understand that. Thanks.

Just a few comments. There may be advantages to wave guides and horns, but they can also introduce colorations. I'll be working with a controlled directivity design shortly, and I'll have more to say then. As for the actual mesurements, note that the Revel Salon demarcations are 10 dB, so that peak in the off-axis measurements off axis at 1500 Hz is twice as bad as it looks compared with the more revealing 5 dB demarcations. Also note that the elevated portion of the Philharmonic response curve 50 degrees off axis amounts to about 2 dB, and that the on-axis response is countured to counter this to some degree. Finally, I would be very surprised if the ear processes the late arrival reflected sound from the far-off axis energy as it does the early arrival and on-axis waves. I still regard this subject as important but far from settled.

 

[DS-21]

I'll be working with a controlled directivity design shortly, and I'll have more to say then.***

As someone who's learned much from you since the old MAD days - though I still wish you guys had taken a harder line on multichannel amp ratings - I'm excited to read that!

 

[Dennis Murphy]

As someone who's learned much from you since the old MAD days - though I still wish you guys had taken a harder line on multichannel amp ratings - I'm excited to read that!

I understand your viewpoint. We would have taken a harder line if we (1) had more solid evidence for just what that harder line should be, and (2) we were convinced that the AV market was broken. Compared with the Republican nomination race, it's doing pretty darn well.

 

[GranteedEV]

Just a few comments. There may be advantages to wave guides and horns, but they can also introduce colorations. I'll be working with a controlled directivity design shortly, and I'll have more to say then.

I would love to see your take on a SEOS-12 design

Finally, I would be very surprised if the ear processes the late arrival reflected sound from the far-off axis energy as it does the early arrival and on-axis waves. I still regard this subject as important but far from settled.

I imagine that would depend on the room.

 

[DS-21]

I understand your viewpoint. We would have taken a harder line if we (1) had more solid evidence for just what that harder line should be, and (2) we were convinced that the AV market was broken. Compared with the Republican nomination race, it's doing pretty darn well.

Hehe, fair enough. Now that I know more about how executive agencies actually propose rules, and how then then get litigated with varying standards of review and levels of deference based on how they were made in the first play, I definitely have more sympathy for the "market's not broken" position.

And not gonna touch the last comment.

 

[Dennis Murphy]

Just a few comments. There may be advantages to wave guides and horns, but they can also introduce colorations. I'll be working with a controlled directivity design shortly, and I'll have more to say then. As for the actual mesurements, note that the Revel Salon demarcations are 10 dB, so that peak in the off-axis measurements off axis at 1500 Hz is twice as bad as it looks compared with the more revealing 5 dB demarcations. Also note that the elevated portion of the Philharmonic response curve 50 degrees off axis amounts to about 2 dB, and that the on-axis response is countured to counter this to some degree. Finally, I would be very surprised if the ear processes the late arrival reflected sound from the far-off axis energy as it does the early arrival and on-axis waves. I still regard this subject as important but far from settled.

Just one more comment, or really a question. And this does not apply to concentrically designed mid-tweet configurations like the KEF. But even if you use a waveguide or some other means to control directivity, if the tweeter is above the mid (3-way design) or woofer (in a 2-way design), the vertical off-axis response will still show an increasing suckout in the crossover region. That's just due to the changing relative flight times as the relative distance of the tweeter and mid (or woofer) change as you move off the vertical axis). And that will overlap with the area of tweeter flare in the horizontal off-axis response. It's not clear to me how all of this comes out in the wash. Depending on the room, and how the ear processes vertical off-axis information vs. horizontal off-axis information, I can see where the power response might even be smoother for a speaker that doesn't have special wave guides to control directivity. I'm not claiming this is the case, but it just seems to be that whole subject is really complicated.