SVS Ultra Bookshelf Speaker Review

[richard12511]

Thanks to Amir for this review, but I think his ears were in a bad mood today. These seem like a good option and reasonable value for a home theater.

I've started to wonder how much mood and hormones(like endorphins) might affect these reviews. Objectively these are one of the best speakers we've ever seen. Not quite Genelec/Neumann tier, but somewhere around Revel(M105/M106) and KEF(R3) tier.

It's interesting to compare this review to the review of the Revel M55XC. This SVS measures excellently(top 5%), yet sounds bad. The Revel M55XC measures terribly(bottom 10%), yet sounds excellent(golf panther tier).


PS.

It's kinda funny how this site has changed my views of audio science. It's changed my views in the opposite way I expected it to. After reading the Harman threads on AVS and then reading Toole's book, I was absolutely convinced that I could base future purchasing decisions 100% on spinorama measurements, and that would guarantee that I got great sound. I was also convinced that you could (with near certainty) predict which speaker should be preferred based on spinorama alone. I found this site after reading Toole's book, and I was so happy; other people who see audio the same way I do!!!

I expected this site to reinforce my beliefs in the science of Toole/Olive, but it's kinda done the opposite. I was the guy over on AVS constantly arguing with the subjectivists who said "you can't tell if a speaker sounds bad just by looking at measurements", or "you can't tell if a speaker is good just by looking at measurements", and it frustrated me to no end how much they seemed to ignore the established science. However, in an odd twist of fates, this site - Audio Science - has kinda, in a way, proved the subjectivist right. Spinorama measurements really aren't sufficient to characterize the quality of a loudspeaker(as Toole had led me to believe). It really is possible to have a speaker with an excellent spinorama that sounds bad(SVS Ultra), and it really is possible to have a speaker with a terrible spinorama that sounds excellent(Revel M55XC).

Something that Amir said in that Revel thread really resonated with me, and really made me start to put more weight into his subjective opinion, and that was he has huge incentive to get the subjective listen right. With this speaker, he listened to it after seeing the measurements, so he had a HUGE bias to make him believe this speaker sounded great, and yet he was able to ignore that bias and be honest with what he heard. I would still rather him listen before measuring, but in this case, the fact that he measured first actually makes me believe the subjective impression more.

Anyway, reviews like this (where the subjective and objective are so at odds) are the most interesting to me, and I think they have the best potential to advance our understanding of the science.

 

[Thomas_A]

A bit of the same problem in the 2-4 kHz vs 1-2 region off-axis as the GR research. Should be evident with female voices/tenors.

 

[amirm]

It's kinda funny how this site has changed my views of audio science. It's changed my views in the opposite way I expected it to. After reading the Harman threads on AVS and then reading Toole's book, I was absolutely convinced that I could base future purchasing decisions 100% on spinorama measurements, and that would guarantee that I got great sound. I was also convinced that you could (with near certainty) predict which speaker should be preferred based on spinorama alone. I found this site after reading Toole's book, and I was so happy; other people who see audio the same way I do!!!

I expected this site to reinforce my beliefs in the science of Toole/Olive, but it's kinda done the opposite. I was the guy over on AVS constantly arguing with the subjectivists who said "you can't tell if a speaker sounds bad just by looking at measurements", or "you can't tell if a speaker is good just by looking at measurements", and it frustrated me to no end how much they seemed to ignore the established science. However, in an odd twist of fates, this site - Audio Science - has kinda, in a way, proved the subjectivist right. Spinorama measurements really aren't sufficient to characterize the quality of a loudspeaker(as Toole had led me to believe). It really is possibly to have a speaker with an excellent spinorama that sounds bad(SVS Ultra), and it really is possible to have a speaker with a terrible spinorama that sounds excellent(Revel M55XC).

Spinorama is still super important and reliable indicator of performance. It is the "Olive Score" that is causing the confusion and what is at odds sometimes with my listening impressions.

We have a single number that takes multiple variables. By definition then you can get the same number with many variations of the underlying parameters. Those speakers can't sound the same.

 

[js_k0914]

You appreciated the upper bass emphasis of Revel M16.
Now why you didn't like SVS Ultra's?
Even it has less upper bass than M16.
The M16's response is far more warm characteristic.
I think it will help to listening test that you attach the room response where you listen at.

 

[muslhead]

This is a review and detailed measurements of the SVS Ultra 2-way Bookshelf speaker. It is on kind loan from a local member and costs US $999 for a pair on Amazon including Prime shipping.

The sample I received has wood grain but what is on sale is glossy finish:

View attachment 76198

There are dual binding posts in the back:
View attachment 76199

There is not a lot of clearance to usncrew/screw the binding posts even for my narrow fingers.

Measurements that you are about to see were performed using the Klippel Near-field Scanner (NFS). This is a robotic measurement system that analyzes the speaker all around and is able (using advanced mathematics and dual scan) to subtract room reflections (so where I measure it doesn't matter). It also measures the speaker at close distance ("near-field") which sharply reduces the impact of room noise. Both of these factors enable testing in ordinary rooms yet results that can be more accurate than an anechoic chamber. In a nutshell, the measurements show the actual sound coming out of the speaker independent of the room.

I used over 800 measurement point which was sufficient to compute the sound field of the speaker. Measurement axis is the tweeter center.

Temperature was 78 degrees. Measurement location is at sea level so you compute the pressure.

Measurements are compliant with latest speaker research into what can predict the speaker preference and is standardized in CEA/CTA-2034 ANSI specifications. Likewise listening tests are performed per research that shows mono listening is much more revealing of differences between speakers than stereo or multichannel.

Spinorama Audio Measurements
Acoustic measurements can be grouped in a way that can be perceptually analyzed to determine how good a speaker is and how it can be used in a room. This so called spinorama shows us just about everything we need to know about the speaker with respect to tonality and some flaws:

View attachment 76200

Response is generally smooth but deviations flat on-axis are quite broad. This means that more musical notes hit them and hence much more audible than "high Q" narrow variations. Emphasis is from a few hundred hertz to 2 kHz and then above 8 kHz or so.

Sensitivity is higher than typical reaching as high as 90 dB at some frequencies. Given the core business of SVS is subwoofers, perhaps the design goal was for these speakers to play loud and if that means deviating from flat on-axis response, so be it.

Interestingly, important early reflections add up to a smooth response:

View attachment 76201

Putting the two together we get what we already saw in on-axis response analysis:

View attachment 76202

We have simultaneous "warmth" and "brightness" depending on how you draw the trend line.

Impedance is on the low side as it is typical of this size speaker:

View attachment 76203

I did not try to optimize the waterfall display so here it is with default parameters:

View attachment 76204

As it is, it shows persistent resonances at a number of frequencies. Of course if I raise the display floor, they can be reduced or eliminated in the measurement.

Driver Responses
I put the microphone very close to the tweeter, woofer and port and recorded a full sweep:
View attachment 76205

It seems that the boost in upper frequencies to midrange is due design of the crossover and possibly the resonance in the port. The former could have been designed out so one has to assume it is intentional. Here is the crossover:

EDIT: looks like above is for their tower speakers, not bookshelf that I tested.

Lots and lots of parts in there although all inductors are iron core making me wonder if they will saturate.

Speaker Radiation Pattern
Let's analyze how the sound radiates horizontally relative to tweeter axis:

View attachment 76206

Focusing on the red line, ideally that would be horizontal and not have the ups and downs. The woofer is getting directional as we approach the 2 kHz frequency whereas the tweeter is at the other extreme, with wider directivity so we have a discontinuity. Beamwidth widens, then narrows, then widens again. Not a good thing in general.

This is less revealing in our heatmap graph of the same:

View attachment 76207

And here is our vertical performance which is less critical than horizontal:
View attachment 76208

Speaker Distortion Analysis
Let's start with Klippel system distortion measurement at 86 dB and 96 dB SPL @ 1 meter:

View attachment 76209

My threshold is 0.5% at 96 dB SPL and the SVS Ultra more or less meets that above 100 Hz which is good. Here it is at an absolute level:

View attachment 76210

Switching to Audio Precision with our deep dive into distortion we have the following results (these tests are being refined):

Dynamic amplitude compression is very gradual and never falls off the cliff:
View attachment 76212

This should let the speaker play quite loud without bottoming out as small speakers typically do.

Distortion at 100 Hz tone is reasonable which we could tell from previous Klippel measurements:
View attachment 76211

Sweeping the full audible band at five different levels we get:

View attachment 76213

I am not trusting the above graph much as it seems to show distortion his higher than lower levels??? I did have to re-calibrate my system so maybe something went wrong there.

Here is our 32-tone test simulating "music" at two playback levels:

View attachment 76214

Distortion sadly increases where our hearing is most sensitive (2 to 5 kHz).

Here is IMD distortion versus level:

View attachment 76216

Subjective Listening Tests
My instant reaction to the sound of SVS Ultra was that it sounded "warm." The upper bass frequencies that exaggerated give an immediate pleasant impression. Alas, that quickly went away and while I could hear the appeal of it, it caused everything I played to have the same characteristic which was not right. At the same time, female vocals had very sharp extensions that were almost painful to listen to. This would come and go of course as the singing went along.

I played with EQ but after a while I gave up. Strangely no matter what I did, I could not get rid of the brightness in vocals. I did bring down the upper bass and lower mid-range level and that mostly helped but then it exaggerated the highs.

On positive front, the SVS Ultra can play really loud with good bass capability. I could get it to start to bottom out but that was at every high playback levels.

Conclusions
Having looked at the measurements first before listening, I thought they would either sound good "out of the box" or do so with just a bit of EQ. That did not happen. Try as I did, I could not like the sound with or without EQ. Yes, I could see the appeal of boosted upper bass in giving "warmth" to the speaker. But too much of that was well, too much of a good thing. I am puzzled where the sharpness was coming from seeing how I could not tame it. Perhaps directivity error was causing this.

I can't recommend a speaker that doesn't sound good to me and that is where I stand with SVS Ultra. I suspect measurement score will be good, making me look bad. So be it!

-----------
As always, questions, comments, recommendations, etc. are welcome.

We have a ton of harvest from our garden. Now need to preserve them by canning them and need money to buy those supplies. Otherwise we won't have anything to eat during winter. So please donate what you can using: https://www.audiosciencereview.com/forum/index.php?threads/how-to-support-audio-science-review.8150/

Similar response to what i experienced and why i sold mine

 

[Sancus]

It's interesting to compare this review to the review of the Revel M55XC. This SVS measures excellently(top 5%), yet sounds bad. The Revel M55XC measures terribly(bottom 10%), yet sounds excellent(golf panther tier).

This is only true if you assume score == spinorama quality. However, reading the actual measurements, this doesn't seem to be true at all. This speaker's directivity is a mess compared to the Revel M55XC.

There seem to be various ways that speakers can end up with a good Olive score despite having poor directivity(E: I think the Harbeth was the first big one), and it's an open question whether these speakers actually sound good or not.

Looking at this speaker's horizontal directivity I certainly wouldn't buy it at $500/channel nor consider it to be a particularly good option.

 

[amirm]

You appreciated the upper bass emphasis of Revel M16.

What? That is lower bass that is higher, not upper bass. When I say upper bass, I mean a few hundred hertz. Not 100.

 

[js_k0914]

What? That is lower bass that is higher, not upper bass. When I say upper bass, I mean a few hundred hertz. Not 100.

Here is "Upper bass" that people mostly say.

 

[Racheski]

Having looked at the measurements first before listening, I thought they would either sound good "out of the box" or do so with just a bit of EQ.

This was probably discussed ad-nauseum, but is there a reason why the listening tests are done after the measurements are taken instead of before?

 

[xykreinov]

On ASR we get extensive data and can reach our own conclusions...

Right, "best" is relative in the way I said it. We reached our own conclusions on all the other speakers with good spinoramas and poor performance in listening, too. I was just pointing out that, of them, this one is the fanciest on paper.

It is interesting because if you read Dr. Toole's book and research, poor directivity is the #1 failure in speaker design. There is not a single Revel speaker that doesn't use a waveguide for this reason. Yet, somehow the formula is designed in a way that still scores such speakers highly.

Huh, really interesting. With that in mind, I've been curious about what you think of magneplanars, electrostats, etc. for awhile, considering their very small optimum listening window (which is low directivity, no?). I'd be really interested in a review of them.

 

[MZKM]

This was probably discussed ad-nauseum, but is there a reason why the listening tests are done after the measurements are taken instead of before?

He wants to hear what the measurements produce, rather than listen and later find what in the measurements resulted in it.

It’s like knowing you have a bug in your code on line 40 that you are trying to fix, rather than knowing your code doesn’t work and finding where the bug is.

However, this of course can make one hyper focus on an area that you might think is of concern and thus hear it worse than it actually is, or think it’ll sound amazing but it only sounds decent and thus you become disappointed.

 

[jhaider]

Dang, this is the best objectively measuring speaker that Amir doesn't subjectively like, isn't it?

Objectively these are one of the best speakers we've ever seen.

How do you guys figure? It has a huge midrange dispersion disruption. Just look at the horizontal polars.

IMO, focus on axis response, listening window, and horizontal polars, along with the DI from the Spinorama. PIR is interesting but not very useful.

 

[Davelemi]

I use these on my desk and they sound fantastic to me there, even better if you blast em straight at your ears.

In the far field though, I've never been able to make them sound correct...

Well said! I initially had these set up near field and felt they sounded pretty good, but once I moved them into a bigger room they were terrible. I sold them soon after.

 

[Promit]

The subjective impressions are consistent with what I got over several years of owning and tweaking with a set of SVS Prime Towers. Same tweeter in those, I’m told.

 

[js_k0914]

And I wonder why the fundmental response of THD part is different to measurement Frequency response. To measure THD the fundmental response is very important. Often it is far different to on axis FR. So, especially the THD of bass part is not relible.

 

[amirm]

Huh, really interesting. With that in mind, I've been curious about what you think of magneplanars, electrostats, etc. for awhile, considering their very small optimum listening window (which is low directivity, no?). I'd be really interested in a review of them.

I am getting a magnepan LRS for review. Will see whether the measurement system can capture their soundfield.

On your general question, they can sound really great on some kinds of music. I just don't care about them sounding that way on everything you listen to. Images are always tall, and sound always diffused. In Harman double blind tests, I voted the Martin Logan down substantially because of the latter.

 

[amirm]

And I wonder why the fundmental response of THD part is different to measurement Frequency response.

Standard distortion measurements are in-room. I can convert them to anechoic (take out in-room reflections) but that takes extra effort so I don't always do it.

 

[js_k0914]

Standard distortion measurements are in-room. I can convert them to anechoic (take out in-room reflections) but that takes extra effort so I don't always do it.

Oh I understand. Thank you for the answer.
Always I have thankful all the review and your effort.

 

[b1daly]

It is interesting because if you read Dr. Toole's book and research, poor directivity is the #1 failure in speaker design. There is not a single Revel speaker that doesn't use a waveguide for this reason. Yet, somehow the formula is designed in a way that still scores such speakers highly.

What is the intuition about the problems with mismatched directivity between woofer and tweeter? Does the sound change tonal quality in unpredictable ways as listening position changes?

 

[Racheski]

He wants to hear what the measurements produce, rather than listen and later find what in the measurements resulted in it.

It’s like knowing you have a bug in your code on line 40 that you are trying to fix, rather than knowing your code doesn’t work and finding where the bug is.

However, this of course can make one hyper focus on an area that you might think is of concern and thus hear it worse than it actually is, or think it’ll sound amazing but it only sounds decent and thus you become disappointed.

I'm not following. I'll start another thread if I feel the need....