mitchco Posted June 15, 2017 Share Posted June 15, 2017 View full article STC 1 Accurate Sound Link to comment
mitchco Posted June 17, 2017 Author Share Posted June 17, 2017 @mr_bill both loudspeakers have almost identical frequency responses, but delivered in very different packages. As mentioned before, speakers are hard to review as it comes down to ones personal preference. What I am trying to do is verify the speakers advertised performance against a neutral reference or as close as one can get. For my own personal preference, both speakers have too much high frequency energy coming at my ears and a bit too much of a bass bump. As @GrahamJohnMiles has indicated the "type" of bass response of the Gold's is audibly different than traditional box speakers that some may like and others, like Graham, don't like. I mentioned in the article as well that the bass quality is different, I can't describe it, but I don't think I would describe it in quite the way Graham did :) In the case of the 600 XD's, I believe the latest firmware upgrade provided a way to reduce more of the top end, which I did not get a chance to audition, but that would be welcome to my ears. @coot no worries, I found the documentation as different as the speakers. @firedog my preference is for high efficiency speakers, usually large 15" woofers with compression drivers mated to large waveguides, like the JBL 4722, which is as fugly as one can get from a visual perspective. The JBL M2's or JBL 4367 are delivered in a more visual appealing package, but again more along the lines of my own personal preference. Which you can see is far removed from the two speakers I have reviewed so far. The 600 XD's are a competent design and perform exactly as advertised. The Golds are as different as a design as one can get, both visually and sonically. Toning down the bass bump and top end would make them sound more subtle in my opinion. But they do have excellent dispersion characteristics because of the shape of the cabinet. I would love to review the Kii Three's Accurate Sound Link to comment
mitchco Posted June 18, 2017 Author Share Posted June 18, 2017 As indicated in the article, there is too much bass and treble, which is reflected in the comments here. The most telling chart is the frequency response comparison between my reference/preference and the Golds. There is a +10 dB SPL bass boost from about 20 Hz to 80 Hz with the Golds. Perceptually to our ears, that sounds roughly twice as loud as compared to my reference. So I can understand the comments of thumpy bass, which may be more problematic in some rooms compared to others. From 5 kHz to 20 kHz, the Golds have approximately +5 dB SPL more high frequency output as compared to my reference/preference. As indicated in the article, too bright for me. And understand for some, including myself, can sound fatiguing. However, both my reference and the Golds measured frequency response fall within the range as indicated by Dr. Toole's (and Sean Olive's) research on subjectively preferred steady-state room curves looking at the chart in the article. The difference in curves are that the Gold's fall in the "untrained listeners" curve and my reference, that has been DSP'd falls in the "trained listeners" curve. I call the untrained listeners curve the "candy" curve, which after too much, one wants to leave, as @GrahamJohnMiles did (novelty). I hope Devialet reads the review and readers feedback. I use DSP already on my reference system, and given that the Devialet speakers are DSP'd, it should not be a big deal to dial back the bass and treble to fit more with Toole's "trained listeners" curve and likely a better fit for folks that don't like the candy curve. Or at least have two or three user selectable curves. Of course, one can DSP/eq the speakers to one's preference, but it would be nice, like the 600 XD's, to be able to tailor fit the bottom and top of the frequency spectrum to better fit one's listening preference using controls that came with the speakers. Further, I feel if Devialet fine tuned the time alignment of the bass drivers with the mid and high frequency drivers, would also better integrate the transient response of the Golds. Given the excellent dispersion characteristics of this loudspeaker, reducing the acoustic output at the frequency extremes would realize its full potential. Accurate Sound Link to comment
mitchco Posted June 24, 2017 Author Share Posted June 24, 2017 On 6/23/2017 at 8:06 AM, Archimago said: Thanks Mitch for the wonderful review! Appreciate the time to get all those measurements; not easy... Definitely difficult to get a grasp on the sound at the Vancouver Audio Show last year! A buddy recently commented to me that he was interested in getting these but I know he'll be wanting them attached to his big screen TV and using them for movie purposes. You commented on the slight audio lag and voice sync - presumably due to communication overhead and FIR filter latency... Did you by chance estimate how many milliseconds that was? I suppose one would have to use a computer or blu-ray player with audio compensation features to get it spot on. I'd certainly be curious if Devialet publishes any information on this and whether latency changes in a 4-speaker multichannel configuration. Also, you're using the Dialog which I gather allows more connectivity options. Does this device also decode home-theater multichannel CODECs? AC3+/DTS? I see there's no HDMI input so I assume lossless surround like DTS-HDMA and TrueHD are out? Can it handle multichannel 5.1 FLAC over the ethernet? Hi Arch, thanks for the kind words. Yes, I was going to use the acoustic timing reference feature in REW to determine the milliseconds delay, then work got in the way and ran out of time. I could not find any info on home-theater/multi-channel type configurations. Not supported it seems. However, your friend should contact Devialet directly to get the definitive answer. Accurate Sound Link to comment
mitchco Posted June 30, 2017 Author Share Posted June 30, 2017 @input username here Vertical directivity is hard to get right, which results in vertical lobing issues like in Figure 1a in this Linkwitz-Riley article from Rane. My understanding is that the Kii THREES use 4th order Linkwitz-Riley (LR4) filter slopes on all drivers and therefore should have the vertical directivity pattern as depicted in Figure 2 in the Rane article. This Audioexpress article has a directivity measurement of the Kii THREES in Figure 3. But it is horizontal only I believe. That measurement display of directivity is based on Dr. Earl Geddes excellent paper on Directivity in Loudspeaker Systems. Thanks for the tip. I hope to listen and measure a pair soon. Accurate Sound Link to comment
mitchco Posted June 30, 2017 Author Share Posted June 30, 2017 @Kal Rubinson Kal looking forward to Stereophile's September issue. It will be interesting to see if Stereophile's measured frequency response of the Kii THREES is the same as the Audioexpress measurement I linked above i.e. flat to 20 kHz anechoic response. My JBL constant directivity cinema loudspeakers also measure flat to 20 kHz in half space. I find them too bright out of the box, as I have of any loudspeaker that is calibrated flat in an anechoic or half space. In-room measurements of these type of calibrated loudspeakers do not match the preferred listeners curve (I.e. downward tilt) as graphed on page 17 of Toole's paper that I referenced in my article. They all measure flat to 20 kHz, even at 9ft listening distance... Note the JBL's also spec both vertical and horizontal directivity, plus directivity index supplied in the link. Would be nice to see similar measurements for the Kii THREE's. I don't have the facility, and I am too lazy to take advantage of Earl's Polar Map service. DSP design can now shape the ideal polar response to the listener. Good times ahead. Accurate Sound Link to comment
mitchco Posted May 22, 2019 Author Share Posted May 22, 2019 Hi @Em2016 yes it would be very easy to use digital room eq to tone down the bass and treble. But I was hoping the speaker would have some onboard capability like the D&D 8c or the Kii Three to contour the tone. Given that this is a DSP type speaker already, it seems like (and still is) an oversight not to have simple tone contouring available to the user, who may not have digital eq or a measurement system... asdf1000 1 Accurate Sound Link to comment
mitchco Posted May 24, 2019 Author Share Posted May 24, 2019 Hi @Em2016 yah, I don't remember any high frequency distortion when listening or measured. I am away from my main computer until Sunday, but will see if I have some measurements comparing the Kii Three and D&D 8c with the Phantom Golds and post up some distortion charts. Cheers, Mitch asdf1000 1 Accurate Sound Link to comment
Popular Post mitchco Posted May 25, 2019 Author Popular Post Share Posted May 25, 2019 @Em2016 Here I overlaid the Kii THREE's, D&D 8c's and the Phantom Gold's left and right distortion measurements. As noted, my equilateral triangle is about 9ft and each of these were calibrated for reference level playback (i.e. 83 dB SPL): As can be seen, all three speakers are well behaved from a distortion perspective at reference listening level. I would not take the absolute distortion numbers as gospel as these were not performed at 2.83v @ 1 meter. I also limited the the low frequency bandwidth to 500 Hz as below that I believe at that distance the room comes into play more than the speakers (could be true for all frequencies and the distortion is just the noise floor of my room). Also, my measurement mic is not the best for taking distortion measurements... To understand why the distortion plot stops at 10 kHz: https://www.roomeqwizard.com/help/help_en-GB/html/graph_distortion.html But the point is from a relative comparison perspective, all three speakers are super smooth and none of them sounded distorted in any way to my ears. Like I say, I don't recall at all, for any of these speakers sounding distorted in the top octave. Wrt the soundstage measurement deviation from linearity at 90 dB, I could no speculate what or why that is. The Gold is a DSP speaker, so there could be a limiter in the amp circuit and have nothing to do with the tweeter for all we know. Cheers, Mitch ShawnC and asdf1000 1 1 Accurate Sound Link to comment
mitchco Posted May 27, 2019 Author Share Posted May 27, 2019 Hi @Em2016 I do feel the Golds are bright sounding, at least to my ears, from a neutral frequency balance preference. But still sound smooth as I don't hear any high frequency distortion. Nothing new on the list other than I upgraded my dual Rythmik L12 subs to F18's and very happy with the result: https://www.avsforum.com/forum/113-subwoofers-bass-transducers/1214550-official-rythmik-audio-subwoofer-thread-1223.html#post57390652 asdf1000 1 Accurate Sound Link to comment
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now