Well in the lecture I linked he seemed to change his mind stating a 110dB target for realistic dynamic range. And anyway to me this is getting into subjectivity and missing the underline thing that is going on. I am more concerned with the skeleton. The fact that you can move your head and it's not stuck in a vice. The way the sound seems to fill the room and have no perceived deviations in the balance. This is what seems to be common to these speaker types.
Yeah if you are already running PC on a Mac Dan then there a lot of free dsp ideas out there that imo can merc the analog counterparts - try building an analog circuit that just delays a signal. You can set them up to be variable and tweak them at the listening position.
Last edited:
Thats what I like about Mr. Linkwitz, his goals and requirements keep changing but his designs still linger on unchanged. There is simple no way that the Orion will do 110 dB - period. And in the old days he never mentioned Constant Directivity as a requirement, but now he does. Except that the Orion is NOT CD and never has been. Hence, it would be great if his designs kept up with his ever changing requirements.
The Orion is not a "bad" speaker don;t get me wrong, but as any many people who have posted reviews on my site will tell you they can't compete with a design that considered 110 dB and CD as design requirement from the start.
I'm an Orion builder+owner and engineer who harbors suspicions on dynamic range.
This weekend curiosity and internet prompting led me to get my Radio Shack SPL meter out while listening to the _Take 5_ on my _Time Out_ CD. That read 85dBC which is 5-10dB shy of anecdotal reports on live small jazz group levels (Internet searches were surprisingly uninformative and I'd love real numbers from some one who has recorded live jazz). After spending 18 months living in Seattle within comfortable walking distance of Dimitrou's Jazz Alley and Tula's I'd say drum kit and upright bass lack their live physical impact at that level but with much more it ceases to be pleasant on a pair of Orions in a live-ish room.
I also fed it to Octave which is GNU's partial Matlab clone. The delta between RMS and peak power was 16.3dB on the left side and 18.4dB on the right.
Assuming the channels were adding uncorrelated (with set in the left channel and piano in the right that seems reasonable) that would be 82dB average coming from each speaker. I don't know how much I gain back from my current room's reverberant field, although that's likely to become over 87dB at the speaker which adds with 18.4dB on the right side to 105.4dB at the speaker which is enough to reach tweeter xmax at the wrong frequency and is definitely beyond the 96dB level Seas used to quote amplifier like distortion on their old data sheets.
There are much (> 30dB peaks beyond RMS) more dynamic recordings, bigger rooms (I might have 2500 cubic feet), deader rooms, and larger listening distances (7' is nice) which would all make things worse.
You could easily listen at "not over 85dB" and sill exceed 105-115dB at the speakers.
Last edited:
That depends on your metric.
Directivity is relatively monotonic compared to most conventional speakers for published measurements from all three categories (Earl has fine grained measurements on his web-site, Linkwitz has Beethoven measurements available via Stereophile with spot numbers on the Phoenix on his web-site, and I'd throw the Revel Salon 2 from Harman with Stereophile measurements into Floyd's bucket)
The Linkwitz Pluto is very well behaved in that regard and does something right, although it's crippled dynamically (I have Pluto+ in my bedroom).
Last edited:
Yes, thats true, but as I said, Linkwitz designs don't agree with what he preaches. Floyd's do (or JBLs do, Floyd isn't really a designer) and I see this as logical given his perspective. But short of the importance of VER Floyd and I agree on everything else.
Proving that Linkwitz assumptions were way off the mark.
I've measured 110-120 dB in a Theater.
As I said, I'd much rather have the option of as much dynamic range as I might need that to assume and be forced to live with less.
Measurements of the Beethoven speaker designed by Siegfried and sold by Audio Artistry. It's a 4-way, dipole through 2Khz, actively bi-amplified. 1998 Stereophile Loudspeaker of the Year implying they reviewed it and made some measurements.
Polar response through +/- 135 degrees at 5 degree increments horizontally (you do have to guess at what's in the valleys on either side of 90 degrees):
from
Stereophile: Audio Artistry Beethoven loudspeaker system
Last edited:
Boy, that was a while back. All the gear, including microphone, dates back to first buying MLSSA some time around 1991 or so. I wanted to see what tweeters were really doing, as well as nailing down - and actually being able to see - the effects of cabinet diffraction.
Holding the mike by hand - arm out, body sideways, peering at the fast-refresh display - was a joke, not repeatable at all. Just screwing the mike on the stand, treating it like a PA mike, was little better, with lots of clutter above 8 kHz, and the time domain a mess. I couldn't see the effects of grill-cloth reflections (they exist and are measurable), cabinet diffraction, or the difference between mediocre tweeters and the really good ones.
It was only when Mike Spurlock and I collaborated on the new stand I could finally see what I knew was there. Nice crisp impulse-response reflections from cabinet edges, making the difference between different edge radii apparent for the first time, and it was obvious which tweeters were storing energy and which ones weren't. Back then, the Scan-Speak 9000 was not just the flattest, but also the cleanest in the time domain - the notorious sticky damping probably had a lot to do with that. The newer tweeters don't seem as good from I can tell.
I also used a very fast sampling rate of 120 to 140 ksamples along with a gentle Bessel or Butterworth lowpass filter to avoid overshoots and ringing in the digital domain. MLSSA has the unusual feature of user-tunable lowpass filters, in the analog domain, that were ahead of the A/D converters. These days, you can just use a 192/24 soundcard with the analog filters fixed at 80 kHz.
The absence of time-domain clutter also has the nice side effect of removing most of the need for some kind of frequency smoothing - really clean time data, of course, results in smooth-looking FR data. If there was any "secret" to the Ariel, it wasn't in the design, which didn't have anything all that novel in it, but avoiding the fashionable audiophile drivers that didn't measure all that well in the time domain.
P.S. On dynamic range, I'm pretty much in agreement with Dr. Geddes. The Radio Shack sound meter doesn't begin to measure brief transients, which can easily be 10~15 dB higher than what you see on the meter. When those transients are "squashed" by medium-to-low efficiency loudspeakers, it's not immediately obvious the dynamic range is compromised, particularly if you don't regularly listen to high-efficiency speakers. Once you get accustomed to high-dynamic-range sound, though, it's hard to go back. The sad thing is that this kind of dynamic range was the norm in the Fifties - the problem, of course, was much higher levels of coloration, and really pretty awful bass response.
Last edited:
But today we can have the high dynamic range without coloration and with multiple subs the bass is excellent. I just don't get the nostalgia craze. Nothing that I have had in the past can compare with the speakers that I have today.
The only "trick" to the Spurlock Stand is the plastic C-shaped sleeve between the 1/2" steel rod and the 1/2" microphone body.
Mike started with a 6" length of plastic pipe with a 1/2" internal diameter, sliced a narrow slit on a table saw (using a guide, of course), softened the whole thing in a low-temp oven so with the gap widened just ever so much, it would slide nicely over the 1/2" steel rod. After waiting several hours for it to cool and firm up again, he then used a sander to create very smooth, gently curved edge on the front part of the sleeve, so the whole thing has an aerodynamic appearance when seen in close-up.
In practice, it is very easy to use. The slit is just wide enough to accomodate the exit of the microphone cable, and the internal diameter of the sleeve is a precision slip-fit for the microphone and steel rod. The rod could just as easily be aluminum or brass, of course - what matters is that it is the same diameter as the body of the microphone. I dress the mike cable by spiraling it around the rod, then down the mike-stand.
The benefit of the sheer length of the rod is the diminished and delayed reflection of the rest of the conventional mike-stand. Actually, I've never been able to "find" the reflection of the stand by itself, even though I know it's out there somewhere around 5~6 mSec or so. Since inverse-square-law attenuation applies to weak mike-stand reflections as well as the things we want to measure, the length of the rod is probably helping a lot.
Mike started with a 6" length of plastic pipe with a 1/2" internal diameter, sliced a narrow slit on a table saw (using a guide, of course), softened the whole thing in a low-temp oven so with the gap widened just ever so much, it would slide nicely over the 1/2" steel rod. After waiting several hours for it to cool and firm up again, he then used a sander to create very smooth, gently curved edge on the front part of the sleeve, so the whole thing has an aerodynamic appearance when seen in close-up.
In practice, it is very easy to use. The slit is just wide enough to accomodate the exit of the microphone cable, and the internal diameter of the sleeve is a precision slip-fit for the microphone and steel rod. The rod could just as easily be aluminum or brass, of course - what matters is that it is the same diameter as the body of the microphone. I dress the mike cable by spiraling it around the rod, then down the mike-stand.
The benefit of the sheer length of the rod is the diminished and delayed reflection of the rest of the conventional mike-stand. Actually, I've never been able to "find" the reflection of the stand by itself, even though I know it's out there somewhere around 5~6 mSec or so. Since inverse-square-law attenuation applies to weak mike-stand reflections as well as the things we want to measure, the length of the rod is probably helping a lot.
Last edited:
Thank you Earl. I have been saying this for years. I don't understand the obsession with imaging in the sense of placing instruments in a 3-d filed for a large symphonic work. As you point out it doesn't happen. Even listening to a violin solo, close your eyes and what I hear is a spacious, dis-joined sound field. I suspect some of this has to do with the lister's distance from the source, at least for the solo. For full orchestras there is just too much scattering from the member of the orchestra, the schell and/or room to localize a single instrument.
Earl, how do you explain the Gradient Revolution - a loudspeaker with less dynamic range than the Orion - being a statistical draw with your Summa's, under your own (controlled) test conditions at your house?
Doesn't that imply that you experienced some sort of mental "imaging" with your eyes open? We should not forget that listening to live music is a sighted (?) experience, while audio reproduction is blind. So I believe that audio recordings need to have a higher degree of separation of individual instruments and voices to make up for the missing visual experience. Don't tell me that you find listening to a live act more pleasant with your eyes closed than looking at the performers.
Actually, when I attend a symphony I often close my eyes to enhance the aural experience. If one sense is shut down the other senses become more acute. Don't you close your eyes when critically listening at home? Aren't audiophiles notorious for listening in the dark? 😎
If I attend a rock concert, then the visual experience is much more a part of the performance. Of course, there is no image to discuss with such highly amplifier sound, unless you want pin point location of the PA towers. 🙂
I think there is a difference between a sound filed that creates a general illusion of a 3-d space and one that locates individual instruments with artificial pin point accuracy. But imaging in the pin point sense is highly dependent on the recording, given quality speakers correctly set up. If that is what the recording engineer attempted to create, a good speaker will reproduce it, artificial or not.
I have also had a surround system since the '80's. I don't always turn it on, and I am sure die hard 2-channeler would consider it heresy, but it goes a long way to recreating what I hear at a live event.
A standard piston speaker like the ones used by Toole and Olive result in a indirect sound field that is very different from the one generated by your speakers with the strong toe-in. These are important differences.
Looking at the recommended ITDG there are recommendations from 2ms to 15ms. There's not much scientific data to justify any of these recommendations.
Best, Markus
We would need to look at polar data ±180° Then it would be pretty obvious how different the indirect sound field is. This kind of data should be standard for every loudspeaker manufacturer.
Best, Markus
Did he call it CD? I never heard him say the word. His hypothesis is that the reflections should be exactly the same in terms of spectral content. Only 3 types will do this omni, dipole, cardioid. In the lecture I linked he criticized the dipole for not being an efficient use of energy calling it an acoustic short circuit. It seems to me that he might like his omni better or think that it is more versatile or easily accessible. And I guess he was talking about 1 meter with the 100+ dB. So maybe he is more concerned with nearfield. I will say that objectively he has a better shot at the dynamic range making his speakers bi-amped and active.
But yeah no doubt reviews have said as much. His compress earlier than yours.
Last edited:
Sure I do. As long as there is no visual reference to what comes from the loudspeakers, there is not much value in gazing at the walls. 😉
But is it just me: When I look at the singers in a choir and see the sopranos open their mouth, I image the soprano voices coming from that soprano block(?). Regardless, whether my ears actually can locate the direction of the sopranos or not.
This sense of locatebility should be in the pure audio experience too. A choir that is reproduced as a large mass without partitions would not be High Fidelity in my opinion.
I don't need pin point. But I do at least need the first violins sitting on one side of the conductor and the second violins on the other. Fifth row, but not 50th row.
- Status
- Not open for further replies.