SVS Ultra Bookshelf Speaker Review

[amirm]

The revel is better, but not that much better. SVS also doesn't have the same design facilities that HK does.

They do some extent. Not in-house but they frequently test their speakers in anechoic chamber:

I really think they were shooting for higher playback levels and compromised elsewhere to get there. Or as the video says, subjectively they thought deviating from research-based response.

 

[amirm]

How this relates to Amir truly disliking the SVS speaker and loving the Revel gear I don't know.

A correction: I don't "truly dislike" SVS. If I had, it would get a headless panther but it did not. It does a lot of things well by playing loud with very low distortion and general tonality that is good. I just couldn't get over female vocals and high frequency sharpness that would come and go. So I did not recommend it.

 

[Beave]

I read somewhere that these were designed by Mark Mason, who used to work at PSB. And I think he used the NRC anechoic chamber both while at PSB and for testing of these SVS speakers.

 

[Beave]

I wonder how the 3-way SVS Ultra Center would do in testing. With a 4" midrange in between the 6.5" woofers and the 1" tweeter, it should have better dispersion than the 2-way bookshelf speaker.

 

[AverageJoeAudiophile]

They do some extent. Not in-house but they frequently test their speakers in anechoic chamber:

I really think they were shooting for higher playback levels and compromised elsewhere to get there. Or as the video says, subjectively they thought deviating from research-based response.

It seems natural as they're primarily known for HT.

Same would likely go for shops like HTD. When I measured their Level 2 bookshelves, they were all manners of F'd up. But people in the HT world hold them in high regard.

 

[amirm]

I read somewhere that these were designed by Mark Mason, who used to work at PSB. And I think he used the NRC anechoic chamber both while at PSB and for testing of these SVS speakers.

Interesting. This reminds me that I should test the one PSB speaker I bought.

 

[spacevector]

I did not offer one because I could not find filters that fixed the issues I had with the speaker. That's not to say you can't improve it. You can by filling in the mid-range with EQ and perhaps pull down the few hundred hertz to 2 kHz peaking. In other words, make the on-axis response more flat.

Thanks I meant member submitted EQ that you apply and try. I remember you liking it on few past speakers. There is an EQ thread - I will post there to solicit some. Do you still have the speaker and able (willing) to try some filters?

 

[amirm]

Thanks I meant member submitted EQ that you apply and try. I remember you liking it on few past speakers. There is an EQ thread - I will post there to solicit some. Do you still have the speaker and able (willing) to try some filters?

I still have it. Will return it sometime this week though.

 

[daftcombo]

Definitely. There is a standard trick in active little speakers to on purpose add harmonic distortion that the speaker can create, versus the bass notes that it can't.

I am finding that when I do a sweep and the speaker hugely distorts to the point where the sound it produces has nothing to do with the low frequency note, a high pass filter makes a big difference. In cases like SVS where that didn't quite occur, then it becomes tricky and subjective testing is necessary to evaluate if it is a good idea or not.

Very interesting indeed. A few ideas:
1) Using steeper slopes (if possible 192 or 96 dB / oct) to cut the lower bass to help identifying the distorting frequencies without removing othe bass just higher; I know Roon is limited to 36 dB / oct but perhaps you can add several low shelf filters at the same frequency to create a steeper slope.
2) Could you perhaps identify distortion issues running the test you do with blue bars showing harmonics but at lower frequencies?
3) Once identified, the distorting frequencies could be cut and be replaced by EQed up frequencies of the harmonics, then a listening test to confirm the same is heard?

 

[amirm]

I don't have the time to play with this speaker much more. Maybe with the next speaker I can add the filter to it. Out of box Audio Precision only has sharp filters so I can't match what Roon is doing. I can include custom filters though.

 

[MZKM]

It seems natural as they're primarily known for HT.

Same would likely go for shops like HTD. When I measured their Level 2 bookshelves, they were all manners of F'd up. But people in the HT world hold them in high regard.

I own an older version of the Level 3 bookshelves (no AMT tweeter and the woofer is black with a plastic phase plug), and it isn’t all that great other than being able to get loud, when I took in-room measurements it was pretty poor, huge boost centered at 1kHz but spanning 300Hz-2kHz, it was +8dB at 1kHz (I got these as a 5.1+receiver bundle off Craigslist for cheap, plus they are in a nice cherry wood finish). I am very interested how their current L3 bookshelves with the AMT tweeter and new woofer perform, especially so with the towers which use a dome midrange and 8” woofers.

 

[Maiky76]

I don't have the time to play with this speaker much more. Maybe with the next speaker I can add the filter to it. Out of box Audio Precision only has sharp filters so I can't match what Roon is doing. I can include custom filters though.

You may have tried something similar already.
But if you still want to try here is an EQ.
Score bumped to 6.0.

01=[  58.0,  0.00,  1.00,... High-Pass
     158.5, -0.75,  1.10,... PEQ
     405.0,  1.00,  1.20,... PEQ
     520.0, -0.80,  3.00,... PEQ
     834.0, -2.00,  2.00,... PEQ
    3570.5, -0.80,  4.00,... PEQ
   17826.0  -1.50,  1.50,... PEQ
    1];

 

[justcheeze]

This is a great review, confirms my suspicions of the Ultras sounding bright and in some cases, harsh.

Amirm, if you thought the Ultras are annoying at the top end, you would rip your ear drums out and drown whatever its left in the lobes with mud if you heard the Primes. Those things are on another level of harshness.

 

[MZKM]

This is a great review, confirms my suspicions of the Ultras sounding bright and in some cases, harsh.

Amirm, if you thought the Ultras are annoying at the top end, you would rip your ear drums out and drown whatever its left in the lobes with mud if you heard the Primes. Those things are on another level of harshness.

Prime Pinnacle Tower measurements, the retained energy off-axis around 4kHz I’ll for sure make it sound bright.

 

[Kachda]

Interesting. This reminds me that I should test the one PSB speaker I bought.

Oh please do. I listened to the imagine t3 once and I thought they were amazing. Just too expensive for my wallet. Their cheaper bookshelves are also well reviewed so it would be interesting to see their spin.

 

[Dj7675]

Here we go, a speaker with a high preference score and no recommendation. Someone please explain this to me.

Post 151 @amirm I thought explained it just fine. One can choose to ignore the listening tests (nothing wrong with that), but I think they are interesting... in particular the ones that don't necessarily match the Olive Score. It may lead to a better understanding of the incomplete nature of the Olive Score and possibly to a better understanding of what other aspects of speaker measurements are important. The Olive Score is interesting but it doesn't appear to me that it automatically is the be all and end all number that determines if a speaker is good or not. It just isn't there.

Post 151
"Once more, there is no mismatch between my impressions and measurements. This speaker does NOT have ruler flat on-axis response, nor proper directivity as the research indicates are signs of excellent design and highest listener preference. This is what the measurements show and is in agreement with me potentially not preferring the speaker.

The mismatch is with the single value olive score. So no, it does not follow that I am automatically wrong. The CEA-2034 specifications would have just had the Olive score and none of the measurement graphs if all that we cared about was the Olive score.

This is on top of the fact that we have found issues with the documentation/computation of Olive Score. Since we are the only ones who is trying to compute and use it, there is no way to verify where we stand (not fully anyway).

Finally the role of distortion was supposed to be follow on research which was never completed or published. This is my acuity in hearing simulated distortion from speakers: https://www.audiosciencereview.com/forum/index.php?threads/distortion-listening-test.8152/ "

 

[richard12511]

Most subjective reviews seem to mention these speakers being a bit bright.

 

[KaiserSoze]

I wonder how the 3-way SVS Ultra Center would do in testing. With a 4" midrange in between the 6.5" woofers and the 1" tweeter, it should have better dispersion than the 2-way bookshelf speaker.

The use of a separate midrange is smart in a center channel speaker because it permits the two woofers to be crossed to the midrange at low enough frequency (500 Hz in this case) to avoid most of the destructive interference between the two woofers, in the laterally off-axis response. With this center channel speaker the distance separating the two woofers is a little more than a foot and almost exactly equal to one-half wavelength at 500 Hz. 90 degrees off axis there will be a suck-out at the crossover frequency and spreading into a response dip covering much of the midrange. For me, the need to orient main stereo speakers horizontally would be a non-starter, with or without the missing midrange in the off-axis response. With some similar speakers the midrange and tweeter are mounted on a circular plate that can be rotated, which I think is very smart because a center channel speaker then becomes a perfectly acceptable R/L main speaker. To my way of thinking it is very dumb for manufacturers to make 3-way center channel speakers without putting the midrange and tweeter on a circular plate.

 

[CDMC]

I'm really surprised and appalled by some of the responses I've read here. I can tell there are some strong opinions about these SVS speakers and the ability of spin data to predict loudspeaker preferences.

I'd like to chime in and say that in my opinion, the replies here that criticize Amir's listener observations are very anti-science. Like it or not, the auditory impressions provided by a reviewer are data. They're the same type of data we would get from the gold-standard test (a blinded A/B listening comparison under otherwise identical conditions), except without all of the laborious measures to eliminate bias. A good scientist never dismisses data simply because it doesn't "agree" with his/her prevailing theory. Like it or not, Amir's listening impressions ARE data - and should be treated as such.

The other thing I'd like to point out is that there are apparently some individuals who think they can eyeball spin charts and make magical predictions that Speaker A will sound better than Speaker B with 100% certainty. As far as I know, there is NO agreed upon and objective way to convert spin charts to preference predictions, and NO evidence that one's ability to "eyeball" a series of spin charts is superior to Olive's regression formula, which we know uses a deliberate SUBSET of spin data, requires complex math, AND only explains 74% of the variability in listener preferences in a closed set of 70 speakers. So, the notion that spin data is a highly reliable way to gauge a speaker's sound quality without listener validation is unfounded and unsupported by evidence - unless someone has something to share.

In reading through this thread, I am struck by a few things:

1) Somehow, Amir's listening skills have been heavily questioned. Last I checked, he never claimed to be "goldenear". He does claim to be trained listener and has provided substantial evidence as to the ABX listening testing and training he has done as well as his attendance at Harmon events. What we do know is that Harmon's testing showed that professional audio reviewers were not the best listeners.

2) Amir has been very clear, he had hoped that the measured results would correlate better to what is heard in his subjective tests, but in fact, they do not reliably inform us of how a speaker will sound in the room.

3) Constant attacks about bias towards Harmon products. Yes, there is probably is some for a few reasons. 1) He sells Harmon and is very clear that he is a dealer of their goods. We don't know if he sells it because a) he believes in their products so decided to sell them, or b) gets a superior profit off their products so pushes them. I am just guessing, it is the former, as if he were purely profit driven, he wouldn't be spending all this time on this site and having spent $100,000 for a Kippel to measure and provide free reviews.

4) Complaints about subjective reviews. Amir doesn't claim his subjective impressions are anything but his subjective impressions. The difference between his subjective impressions and professional audio reviewers is 1) his are direct, 2) he doesn't dress them up in BS, and 3) he is abundantly clear about his financial interests in companies and connections with them.

5) Claims of a Harmon bias. There is a Harmon bias on this site, but I don't believe it is because of profit, but available data. There is very limited data available about speaker measurements and their correlation to perceived sound. The data that exists was nearly all created by the years of research of Toole/Olive, first for the NRC, then for Harmon. No other private company seems to have any interest in making their data available (we know that Dynaudio has extensive testing facilities, but they don't make their research available, they don't even offer spinorama measurements of their products).

Most of all, don't forget that Amir has given us a wealth of data and a place to share and discuss it without being attacked. It seems to me that warrants at least some basic courtesy towards our host.

 

[patate91]

In reading through this thread, I am struck by a few things:

1) Somehow, Amir's listening skills have been heavily questioned. Last I checked, he never claimed to be "goldenear". He does claim to be trained listener and has provided substantial evidence as to the ABX listening testing and training he has done as well as his attendance at Harmon events. What we do know is that Harmon's testing showed that professional audio reviewers were not the best listeners.

2) Amir has been very clear, he had hoped that the measured results would correlate better to what is heard in his subjective tests, but in fact, they do not reliably inform us of how a speaker will sound in the room.

3) Constant attacks about bias towards Harmon products. Yes, there is probably is some for a few reasons. 1) He sells Harmon and is very clear that he is a dealer of their goods. We don't know if he sells it because a) he believes in their products so decided to sell them, or b) gets a superior profit off their products so pushes them. I am just guessing, it is the former, as if he were purely profit driven, he wouldn't be spending all this time on this site and having spent $100,000 for a Kippel to measure and provide free reviews.

4) Complaints about subjective reviews. Amir doesn't claim his subjective impressions are anything but his subjective impressions. The difference between his subjective impressions and professional audio reviewers is 1) his are direct, 2) he doesn't dress them up in BS, and 3) he is abundantly clear about his financial interests in companies and connections with them.

5) Claims of a Harmon bias. There is a Harmon bias on this site, but I don't believe it is because of profit, but available data. There is very limited data available about speaker measurements and their correlation to perceived sound. The data that exists was nearly all created by the years of research of Toole/Olive, first for the NRC, then for Harmon. No other private company seems to have any interest in making their data available (we know that Dynaudio has extensive testing facilities, but they don't make their research available, they don't even offer spinorama measurements of their products).

Most of all, don't forget that Amir has given us a wealth of data and a place to share and discuss it without being attacked. It seems to me that warrants at least some basic courtesy towards our host.

The thing is that there's simple method that can be applied to avoid few biaises.

This is supposed to be a science oriented forum.

Subjective impression are not mandatory.