Home Audio Listening Test: DIY Scanspeak vs. DIY HiVi Swans bookshelves (56k careful)

[ErinH]

I live near a very awesome builder; Scott Selvidge. He does a lot of 'on the side' builds and does so to a superb quality. I've been talking to him about an HT DIY with Scanspeak drivers and he asked me to do a listening test of some bookshelves he has. One incorporates a 7" revelator and dome scanspeak setup. The other used Swans drivers, and I honestly have no idea which is which. The big deal here is that the Swans are a very small fraction of the price of the Scans he builds.
Here's Scott's website:
http://www.schilacoustics.com/home.html

The HiVis weight about 19lbs/each. The scans weigh about 54lbs each. The scans have granite top and bottom, about 1" thick. If you check this link you'll see a cutout image of how Scott build's the Scans. I believe he said the Hivis use a pre-made cabinet.

I put the bookshelves ~13" on each side of the TV, and about 20" from the wall, simulating a real listening environment for me.

I did about an hour session and am really amazed how good the swans sound. The scan's lower end is more robust, but overall it's very hard to tell any differences myself. Additionally, the Swan uses a 5" driver, where the Scan set uses a 7" driver, so there's a solid explanation for the Scan sounding more robust. There were a couple songs that seemed to bring out a bit of shreekyness, I'm thinking in the 2khz-3k range, but first listen was very pleasant.
I tried a range of music from Peter Gabriel, Coldplay, Michael Jackson, Natalie Merchant, and a few others here and there. I even put it through the paces on a couple of my fave rap songs and both sounded phenomenal. A couple times I even thought the Swans sounded better. Overall I'm biased toward Scan and they seem to be a tad more transparent but so little I'm not even sure it's worth mentioning. The biggest difference was the Scans simply had more impact in the chest. Short of that, I really haven't noticed anything that's significantly different.

I'll keep doing some ear tests throughout the week; had to stop for now.

My wife said she really likes the look of the Scans; she said if she could make the hivis in a darker wood like the Scans she'd take those.

Oh, and my back hates those Scans.

I attached a few plots for your enjoyment. File names are labeled according to the Frequency range I've focused on.

 

[ErinH]

RTA Results:

Scanspeaks are in Blue, Swans are in Green.

20hz to 20,000hz:



20hz to 500hz:


500hz to 5,000hz:


5,000hz to 20,000hz:

 

[ErinH]

Now, what hurts is the price difference. The Scans retail for $2400/each. The Swans retail for $500/each.

FWIW, here's what I believe are the drivers for the Swan bookshelf speakers:
5" midbass/midrange
http://www.madisound.com/catalog/product_info.php?manufacturers_id=136&products_id=1507

Tweeter:
http://www.madisound.com/catalog/product_info.php?manufacturers_id=136&products_id=1492

Look at those prices... goes to show you that it's all in the install...

Doesn’t this make you want to pick up a jigsaw, grab some MDF and get to work on your own DIY set? I mean, just think, can you imagine what real quality you’re getting out of sets that cost $2k for a tower? How critically are you listening to your gear to notice the difference the price point makes? Granted, I’ve heard the $500 Klipsch towers at BB that absolutely are horrible, but think of what you can do with a little DIY initiative. Just sayin’.


Looks like I’ve found my home audio gear, now.

 

[Swerd]

Very nice write up, with photos and FR curves too! Yes, well-designed DIY speakers can be as good as commercial retail speakers from a B&M store costing some 3 to 5 times as much as the DIY parts.

It is clear from those curves that those two speakers share similar response across the midrange where the crossover is. Not surprising because they had the same designer. Over all, they look very smooth.

I'm guessing, but it looks like the cabinet for the HiVi speakers is sold by Parts Express, and is also available in a cherry color and glossy black finish.

For the much larger price, the ScanSpeaks will have lower bass response, as you noted, and also greater power handling than the HiVis. Even at $2400 a pair those ScanSpeak speakers are a good deal compared to commercially available 2-ways using those same drivers.

 

[ErinH]

I agree on all points.

I should note that the Scan uses a 7” driver, while the Swans are 5” drivers. That obviously means more low end. If the Swan had used a 7”, I doubt I would have been able to tell much of a difference at all.

Yes, the cabinets, I believe, are indeed PE pre-fabs. I think they run about $120/pr, or so.

I’m sure with a dead silent envioronment and some really good ‘reference’ music, one could pick out the Scan, but in my living room test, which was as real world as it gets, I couldn’t tell enough to notice if I had been blindfolded.

Plus, I can get a dual 7” Swan cabinet L/R/C for less than one of those Scan cabinets. It’s pretty much a no-brainer now.

The cool thing is that the builder wanted me to test them first before deciding to dump so much money on a scan DIY. Now, I’m very glad he did.

 

[Shock]

Swan make amazing speakers. Not only do they sound superb but the styling and attention to detail are second to none. Who here would say no to a set of Swan 2.3B's!?

Everytime i look at them i get a halfy......

 

[R-Carpenter]

Before you lay out money for Scans, look at Dayton Reference series drivers.
Don't get me wrong, Scans are the most neutral, lowest distortion drivers on the market. The quality is second to none, but.... For HT set up they just don't sound as involving and punchy as Dayton Reference. I have 2 custom towers with 2k worth of SS in each pair and a relatively simple Dayton based towers sound way better on HT. Music is another story.
Hi Vi are good drivers for the money but not my personal preference. Kind of in the middle of the road.
The graphs look strange. No 2 speakers, especially with different sets of drivers should sound that similar. I also noticed a flat 20 Hz extension. It may be just a way RTA measures but a 5 inch driver should start a slow roll off around a 100 Hz. A 7 inch Revelator rolls of around 70hz if I remember correctly.

 

[ErinH]

The graphs are always a bit awkward on the low end. But, to verify the broad range, I compared to smaartlive and arta program results. They were all identical from 100hz and up. TrueRTA is a bit weird for me and I haven't taken the time to truly get to the bottom of that issue. :/
As for rolloff, you know that it's all dependent on the enclosure. I haven't bothered to guess the internal volume of the enclosure and model it. I might tomorrow if I get bored.
I also need to check into default weighting true uses.

anyway... ...

I would've said hivi's aren't good had I not sat down and done listening tests. I'm now very surprised. There's no way I'm going to buy the Scans; I'll go with the Swans.

As for Daytons, I've never been a fan at all. In every setup I've heard them in they seem to be tinny and a bit dull. That's in three different setups; one being a car. I take that as a characteristic of the drivers at that point. Had it been one setup I wouldn't have let that be my determining factor.

It's unbelievable how great the Swans sound compared to the Scans. If i were in an anechonic chamber, listening to classical music, I may have been able to make which was which, but in my listening tests I couldn't really. And definately not enough to warrant the nearly 500% price delta. I like the 'warmth' Scans seem to have, and the Swans have the same characteristic.

 

[WmAx]

In regards to the enclosures on those Schilacoustics link, since there was a statement along the lines that these were impressive. Yes, they 'look' impressive, but as far as I can tell, this construction is not going to yield any appreciable improvement over any other average hi-fi speaker cabinet(which means it produces substantial acoustic output from the cabinet panels).

-Chris

 

[WmAx]

Before you lay out money for Scans, look at Dayton Reference series drivers.
Don't get me wrong, Scans are the most neutral, lowest distortion drivers on the market. The quality is second to none, but.... For HT set up they just don't sound as involving and punchy as Dayton Reference. .

Actually, I would say in some instances, Dayton has superior drivers. The RS150 for example; I don't know of SS that is overall as good across the same bandwidth considering all aspects of performance. Also, I don't believe SS has any mid directly comparable with the excellent RS dome midrange. As far as regular woofers go, the RS Dayton subwoofers are likely superior to the SS offerings based on the motor and compliance measurements via Klippel analysis.

Dayton really has some amazing drivers available for dirt cheap these days.

But, to be frank, if one really knows what they are doing, they can get transparent performance from lower quality drivers than any of these being discussed. It all about the specific implementation and working around the weakness of any particular piece of hardware.

-Chris

 

[WmAx]

I’m sure with a dead silent envioronment and some really good ‘reference’ music, one could pick out the Scan, .

Actually, people assume bigger differences than really exist in most instances, in regards to driver differences alone. I have put a lot of research into this area, including extensive FR compensated randomized double blinded testing using recordings in a semi-anechoic tunnel. One can get a driver/crossover set up, in this environment, to sound so close to the original signal/recording, that comparing the recording and original signal on a reference grade playback unit/set will be extremely subtle difference(using nothing special in regards to drivers)! But, do realize, an anechoic environment actually masks some things contrary to what many assume, and of course, this type of comparison only deals with a single response axis. The only point being that really, the main issue is implementation of drivers, not the cost/ultimate quality of them(which is viewed out of perceptual proportion relative to the most improtant things most of the time based on my observation).

-Chris

 

[R-Carpenter]

Hehe, how much lower can you get then $35 for RS150? My own HT is based on this driver.