Thanks for the responses, gentlemen. And thanks, Andrew (Drewan) for providing your measurements for comparison.
Kotjac, I suspect that you're right. Tomorrow I'll probably make a quick experimental measurement with the speakers in their normal position, with the mic on a boom, to see if that glitch occurring at 0.7 ms after the direct sound arrival goes away, or at least improves.
Andrew, your interpretations are correct.
Bruce, I'm not entirely sure either of how to explain the wild looking group delay plots. However there are several things that leave me feeling encouraged about them:
1)All three of our group delay plots (mine, the one Andrew posted, and yours that were taken by Larry/DEQXpert last year) look similarly wild. (And mine appear to be the best of them, actually, when viewed on the same vertical scale :-))
2)Despite the wild appearance, as we know Andrew achieved outstanding results **with no smoothing.**
3)Applying 100% smoothing to my GD plot (and Nyal had recommended to me that I use 100% smoothing) results in essentially a perfect straight line above about 2.7 kHz, with the wiggles getting progressively larger below that frequency, and exceeding just a few ms only below around 500 Hz. And I suspect I won't be correcting much below 500 Hz anyway (I'll be addressing that region in the room correction process), and I also suspect that a major contributor to the group delay anomalies below several hundred Hz is the path length difference between the measurement mic and my woofers, relative to the path lengths from mic to tweeters and mids. Which by virtue of simple geometry will be much less of an issue at normal listening distances than at the 3 foot mic distance.
Bruce, regarding your question about the step response, the handoff between mid-ranges and woofers looks to me to be reasonable, and not particularly dissimilar to many step response plots JA has provided in Stereophile reviews. Aside, that is, from the aforementioned glitch at about 0.7 ms which as mentioned may be due to a mic stand reflection. And, again, the delay between the initial arrival of the woofer outputs and the outputs from the other drivers is presumably contributed to by the differences in path lengths from the close up mic position.
Best regards,
-- Al
|
An update: I received my new DEQX HDP-5 from dealer AcousticFrontiers.com a few days ago, slightly ahead of the schedule for this production run which DEQX had projected back in March.
Installing the DEQX component in the system was straightforward, although time-consuming in my particular case because I cant access that or some of the other components in my system from the rear when they are in position, and also because of reconfiguration of many of the interconnections in the system that was necessary (as described in the next paragraph). But once that was done I was immediately able to listen to music via both speakers and headphones, using the DEQX in its as delivered bypass mode configuration.
I then read through the user manual and the software/calibration manual, and installed and browsed through the calibration software on one of my laptops. As he does with all purchasers of such products, Nyal Mellor of Acoustic Frontiers then provided me with a free person-to-person webinar/software walkthrough via the Internet, which took 70 minutes and was highly informative and helpful. One basic point which was made clear was to avoid over-correcting, which can be tempting due to the power and flexibility of the calibration software. Or if one does want to try an aggressive set of corrections, to also create a more conservative set, store them in different profiles (which can be selected between at the push of a button on the remote), and compare the resulting sonics.
I should mention that insertion of the DEQX into my system has involved changing several things at once. I have replaced my Classe CP-60 preamp with it. Due in part to the limited number of analog inputs provided on the HDP-5 (one balanced, one unbalanced), I am connecting the AES/EBU output of my CDP to the corresponding digital input of the HDP-5, rather than connecting the CDPs analog outputs to the preamp as before. My other critical source is phono (unbalanced), which I am connecting to the HDP-5s unbalanced analog inputs. I am connecting two relatively non-critical analog sources (tuner and cassette deck) through a DB Systems line-level switchbox to the HDP-5s balanced analog inputs, using XLR-to-RCA adapters. I am connecting my Squeezebox, which I and my wife just use for relatively non-critical Internet radio listening, to the HDP-5 via an Analysis Plus optical cable, rather than via analog as before. (My research seemed to indicate divided opinion as to whether the Squeezeboxs Toslink output or coaxial digital output is preferable, and given that and our relatively non-critical use of it I figured it would be best to use Toslink and avoid any possibility of ground loop issues or coupling of electrical noise into the HDP-5).
Thus far I have spent several hours with both speakers and headphones assessing its transparency in bypass mode (i.e., with no corrections or calibrations applied), on both LP and CD, with mostly classical recordings (chamber, symphonic, and operatic), but also with some recordings from various other genres. On some recordings I noted little or no difference compared to my previous setup. On some I noted slight to moderate improvements in resolution of fine detail in complex passages, and in some cases also slight expansion of the soundstage. I did not perceive any loss of transparency of any kind on any of the recordings.
Ill mention also that as might be expected sonics when listening to Internet radio are vastly improved as a result of using the DEQXs DAC rather than the one in the Squeezebox, despite the severe bit rate compression of the stations my wife and I tend to listen to.
I will post further updates in the coming weeks as I proceed with the calibration/correction processes, but they will be somewhat slow in coming due to other things I will be occupied with in the near future.
Regards,
-- Al
|
05-16-15: Bifwynne
Al, please clarify/confirm what you are reporting. At this point, I think you are saying that your DEQX is being used in bypass mode ... no signal corrections yet. If so, presumably there should be little or no impact on the signal as it passes through the gizmo, which is pretty much what I am getting from your post. Yes, that's correct, Bruce. But a logical concern in using the DEQX in conjunction with a phono source would be the transparency of the A/D and D/A conversion processes that it inserts into the signal path, and perhaps also the transparency of its digital volume control function. Some of the experienced DEQX users posted comments earlier in the thread attesting to that transparency, but obviously before getting into calibrations and corrections I wanted to confirm that with my own system, recordings, and ears. As well as assessing the differences resulting from replacement of my preamp with the DEQX, and replacement of analog interconnections to my digital sources with digital interconnections, and substitution of the DEQX's DAC function for their internal DACs. Best, -- Al |
Hi Bruce,
Yes, I've been following the ARC KT-150 upgrade threads to some extent, particularly including your posts. Glad it appears to be working out well.
And yes, at this point I'm satisfied with the transparency of the DEQX, and I'm ready to move on to the measurement and calibration/correction processes. As I indicated, though, progress may be a bit slow in the coming weeks, due to other activities. Also, before moving the speakers to the center of the room for the measurements (outdoors measurement being a non-starter in my case, as it was in yours), I'm planning to first dry run the entire process with the speakers in their normal position. That way I won't risk moving the speakers to the center of the room and then back, finding that I've screwed up or at least not done things optimally, and then having to move them again. Even though the bases and footers I have the speakers on only raise them a bit less than 5 inches above the floor, I don't relish lifting/sliding their 108 pound weight off of those bases, and subsequently raising and maneuvering them back on.
Best,
-- Al
|
Andrew (Drewan77) & Bombaywalla, thanks for your kind comments. And Bruce (Bifwynne), thanks for the suggestion about putting pillows on the floor for the speaker measurements. That is also suggested in the manual, and I'm planning to do it. I should have some time tomorrow to start playing with measurements, although I suspect I won't have anything meaningful to report for a few days or more.
Regarding the manual, for my HDP-5 the 38 page user manual and the 143 page calibration ("installers") manual were on the calibration software CD. Similar if slightly earlier versions of those manuals can be downloaded from the DEQX site. (Click on the "owners" tab near the top of the home page, then "upgrades," and then scroll to the bottom of the page that appears).
I printed out the manuals, 3-hole punched them, and put them in a loose-leaf binder, which for lengthy documents such as these I find preferable to viewing a pdf on a computer (unless I want to use the pdf reader's "find" function to look for a specific term).
Content-wise, IMO the manuals are informative and reasonably well done. (Perhaps it was a different story some years ago when Psag and other long-term users purchased their units). Although the online session I had with Nyal Mellor was certainly a valuable supplement, in part because of suggestions he made that were in the direction of greater conservatism in the corrections than the calibration manual would seem to suggest.
Bruce, a question for you: When the DEQXpert people calibrated your speakers, how far did they end up placing the microphone from them? And if you know, how many milliseconds after the direct sound arrivals did they place the point at which subsequent arrivals were windowed out?
The reason I ask relates to the relatively large physical spacing between some of your drivers, which based on pictures I've seen I suspect is around 3 feet between the lowest of the four woofers and the tweeter. On my speakers, also, the two woofers are a significant distance (about 15 inches) above and below the two tweeters, which in turn are about at listening height.
The reason I started thinking about that is it occurs to me that the greater the physical separation between drivers, the greater the distance should be between the speakers and the measurement microphone, which in turn (assuming the speakers are not measured outdoors) will necessitate shortening the duration of the measurement window (prior to arrival of the first reflections), which in turn will raise the minimum frequency that should be corrected and/or reduce the accuracy of the corrections.
The reason I'm envisioning for that is not related to off-axis dispersion of the drivers, since the mic is placed at the level of the drivers which presumably have the narrowest dispersion (i.e., the tweeters). What I'm envisioning is that with the mic placed at tweeter level, the closer it is to the speaker the greater the difference will be between the distance from mic to tweeter and from mic to other drivers. And if the drivers are widely spaced, the amount of that path length difference will be significantly different than the difference between those path lengths as they exist at the listening position, due to the shallower angle between those drivers as viewed from the listening position.
In other words, it seems to me that if drivers are spaced relatively widely, and the mic is not moved correspondingly further away from the speakers during the speaker calibration process (with the downside of shortening the "window," and hence the accuracy and/or low frequency extension of the corrections), the speakers may be corrected for a timing error that won't exist at the listening position.
I've done some geometric calculations for the 15 inch distance between the woofers and the tweeters on my speakers. At a 4 foot measurement distance the path length differential between the distances of the mic to the tweeters and the woofers is 0.18 feet. At my 11.5 foot listening distance that differential is only 0.06 feet. The difference between those differences is 0.12 feet, corresponding to a propagation delay at the speed of sound of about 0.11 ms (milliseconds). Which would seem to mean that the DEQX will correct for a 0.11 ms timing error that won't exist at the listening position, if my speakers are measured at a distance of 4 feet, and a somewhat larger error than that in the case of your speakers.
The planes of the baffles on my speakers, btw, are such that the woofers are mounted a little forward of the tweeters and mid-ranges, presumably to help with time alignment. But that is unrelated to the point I am describing.
Also, to provide a bit of perspective on a 0.11 ms timing error, that would be readily perceivable on the step response graphs JA provides with his speaker measurements in Stereophile, those graphs having a time scale of 1 ms per major division. One of the purposes of those graphs being to provide some idea of the time coherence or lack thereof of the speaker.
Apologies for the long-windedness of this post, but I hope it is clear, and that is the background for my question about the measurement distance the DEQXperts chose to use with your speakers.
Best regards,
-- Al
|
Thanks everyone for the good responses. I guess part of the answer to the issue I described, about the possibility of correcting a non-problem in the case of large speakers that can't be measured from an optimal distance due to reflection constraints, is that under such conditions speaker corrections would (or at least should, per Nyal's (AcousticFrontier's) recommendations) be performed only at frequencies above the point where the woofer(s) are likely to be significantly rolled off. For example, the crossover point of the woofers in Bruce's (Bifwynne's) speakers are indicated as being at 230 Hz, with a 12 db/octave rolloff above that point. (I don't know what the corresponding figures are for my speakers, as they aren't published and haven't been measured as far as I am aware). Bombaywalla, thanks for your inputs as well. As you aptly stated, there are always tradeoffs. Re your last post, though, undoubtedly the measurement they had Bruce perform at the listening position was for room correction, not speaker correction, room correction generally being done with DEQX only at frequencies below around 200 Hz or so, where room effects predominate. Speaker correction, including time alignment, would have been performed at the 36 inch distance he mentioned, and only at higher frequencies as I indicated. On another note, would anyone have any comments on the possibility of surrounding the measurement microphone during the close-up speaker measurements with two of these (four panels total, surrounding the mic on three sides). Acoustic specs are here, and look impressive. Or, alternatively, a mic baffle such as this one, which is apparently made of the same material as the large panels. Best regards, -- Al |
Well, hearing one vote in favor and none opposed I've decided to order the large acoustic panels I referenced in my previous post, for purposes of shielding the mic from reflections during the speaker calibration measurements.
The one slight concern I've had about doing that is the possibility that the panels might in themselves cause some low level reflections, that would arrive at the mic a millisecond or two after the direct sound. But hopefully not, and even if that were to occur to some small degree I'm thinking it could probably be minimized by some re-positioning of the panels. And although I would normally be hesitant to spend $340 on something that may end up being used only once, I'm guessing that the investment will provide benefits that are essentially permanent.
Andrew (Drewan77), thanks again for your always valuable perspectives. I find it interesting and somewhat surprising that they extended the window of Bruce's speaker calibration as far out as 24 ms (about 17 or 18 ms after the direct arrival), given that significant reflections occurred earlier. Makes me a bit less worried about the reflections I may end up with.
Best regards,
-- Al
|
Bruce, the only thing I can add at this point to the good responses that have been provided to your questions is to mention that in contrast to speaker correction, DEQX allows you to perform room correction adjustments on the fly, in real time, by inserting and/or dragging adjustment points on the computer screen while you are looking at the measured frequency response plots on that screen and while you are listening. Which is a neat and I believe pretty much unique feature.
You can also click a button which inverts the room correction curve, then insert and/or drag correction points so that the inverted curve lines up with the most significant peaks and dips in the measured frequency response ("most significant" based on the combination of magnitude and width), then re-invert the correction curve and assess it sonically. Also a neat feature.
In doing that you would of course not want to risk messing up the real .mzd file that is in use. The manual describes a procedure for creating a duplicate file that can be played with, but it seems unnecessarily roundabout. I've found that a simpler procedure is to copy the .mzd file to a different folder, then change its name and copy it back to the original folder. Then double-click its icon to open it with the DEQX software, or else open the DEQX software first and use "file/open."
When you're done you can then re-upload the original file to the DEQX, if you want to.
Best,
-- Al
|
Andrew (Drewan77), thanks again for your always excellent inputs. A couple of minor clarifications to your post just above, if I may: Sound waves travel at different wavelengths/speeds from the lowest/slowest to the highest/fastest frequencies so what you are aligning to is a subset of all. Although the velocity of a sound wave in air does indeed vary as a function of frequency, as I understand it the amount of that variation is small enough to be negligible for practical purposes. See the graph near the lower right corner of this paper, where it can be seen that even under the worst case condition (0% humidity) a frequency of 10 Hz is less than 0.03% slower than a frequency of 20 kHz. At a listening distance of 3 meters, that would result in a propagation delay difference of less than 0.003 milliseconds between those two extremely different frequencies. With a single full range speaker containing passive crossovers, or a 2-way, 3-way measured accurately at once (ie without requiring subs), then an appropriate DEQX processor will do everything for the user and it automatically becomes time and phase coherent. But of course only to within a degree of accuracy and over a range of frequencies that are constrained by the duration of the correction window and by reflections that may be captured within that window. Thanks again, though, for another excellent post. Best regards, -- Al |
Acoustic panels are not one of my areas of expertise either, Bruce. But if you want to DIY at a low price, the suggestions Bombaywalla made earlier of Owens Corning 703 or 705 seem worth looking into. They can be ordered via Amazon, where lots of user comments and Q&A's are provided.
Also, although at higher price points and not DIY, you may want to look into the suggestion Nyal made earlier of the HF versions of some of the RealTraps.
The Clearsonic panels I ordered, btw, will be arriving here on Tuesday. I may not be able to report results until the following week, though, as I'll be occupied with other things during most of the coming week. Also, I've been envisioning that when I move the speakers toward the center of the room for the close-up measurements, in my particular room surrounding the mic with the panels is likely to be more beneficial than surrounding the speakers with them. But I may try it both ways.
Best,
-- Al
|
|
|
Hi Bruce, Here are the benefits of DEQX as summarized in the calibration software manual: Correcting full range passive (traditional Hi-Fi) speakers plus room correction:
Anechoic Frequency-response calibration.
Anechoic Phase-response/Timing-coherence calibration.
Integrate subwoofer/s with time-domain and/or Parametric EQ room correction.
Multiband Parametric EQ for real-time preference and room EQ adjustment.
Improved imaging and sound-staging.
Improved frequency-response accuracy.
Improved timing coherence.
Three band tone control including a fully parametric-EQ band with 99-memory (remote control).
Additional features if using DEQX-Active crossover option (available on some models):
Stereo 3-way active crossovers: 6dB/octave to 300dB/octave.
Steep linear-phase filter option.
Automatic timing/phase alignment between up to 3-way speakers (or 2-way plus subs).
Increased loudness.
High dynamic resolution (reduced distortion) due to drivers operating in linear operating zone.
Reduced crossover distortion - reduced comb filtering.
Improved natural dispersion - reduced unwanted on-axis driver beaming of high frequencies. I'd imagine that the main benefit DEQX can provide with respect to driver distortion would be in cases where subs are being used (as in your case), or in biamped or triamped setups. In those situations DEQX could introduce very sharp filter slopes that would keep some of the drivers from having to deal with frequencies that are out of their comfort zone. And it could do that without the adverse timing and phase effects that would result if that were done in speaker crossovers or elsewhere in the analog domain. With a single-amped speaker that is being driven full-range, such as in my case, I'd imagine that any benefits to driver distortion would be incidental, due to relatively minor "second order" effects. An example perhaps being taming a frequency response peak in the bass region that is contributed to at that frequency by harmonic distortion of a lower frequency. Regarding Magico, as you've no doubt seen in various threads here and elsewhere, they tend to be controversial. While they seem to do pretty much everything right on paper, and a lot of folks love them, some find them to be unmusical. Also, the impedance characteristics of the S5 shown in the SoundStage measurements you linked to do not inspire confidence in their tube-friendliness. While their impedance magnitude is relatively flat across most of the spectrum, as you've mentioned in the past, it is around 3 ohms in a good part of the bass region, and that low magnitude is combined with a fairly severe -45 degree capacitive phase angle around 50 Hz. That said, though, I suspect that your amp could handle them better than most tube amps, due in part to their relatively low output impedance and their very substantial power supply. Best regards, -- Al |
Update: Today I performed the close-up speaker measurements. I moved the speaker being measured close to the center of my living room/listening room, with the Persian rug you can see in my system description photos folded back roughly in half. That resulted in the front baffle of the speaker being located 10.5 feet from the large window on the front wall (i.e., the wall behind the speaker). The speakers left side was located about 5 feet from the fireplace on the left. Its right side was located about 7 feet from the right wall, but just 4.5 feet from the large mahogany antique radio/phono console you can see along the right wall in my system description photos (the thing that looks like a Chippendale-style bureau, which Im pretty certain weighs in the area of 350 to 400 pounds, and which I therefore wasnt about to move. Its shipping weight in the large wooden crate in which it was sent to me about 20 years ago was 567 pounds!). The distance to the rear wall, the middle third of which is an opening to another room, was around 11 feet. I positioned the mic at a height of 30 inches, which placed it exactly between the heights of the two closely spaced tweeters on my speakers. When I measured each speaker, I moved the other speaker a few feet to the side of its normal position, to get it as far away as possible from the speaker being measured. I placed pillows on the rug between the speaker being measured and the microphone. I closed all the windows so that the mic wouldnt pick up bird chirping and other outdoor sounds, and I turned off all noise-making appliances that were within earshot. Fortunately a noisy rainstorm didnt start until a few hours after I finished with the measurements. I made a total of 9 measurements on each speaker, three different distances (2.5 feet, 3 feet, 3.5 feet), times three setup conditions (no acoustic panels, two large double-section acoustic panels surrounding the microphone on three sides and projecting somewhat forward of it, and the same acoustic panels re-positioned to surround the speaker on three sides and project somewhat forward of it). Each of the 9 measurements of each speaker consisted of a 2.4 second sweep repeated 9 times, with the sample rate set to its 96 kHz default setting (2.4 seconds being the longest sweep time choice that appeared to be offered at the 96 kHz sample rate). I adjusted the volume such that peaks were indicated in the DEQX software as being 102 to 104 db, which my speakers and amp can handle comfortably (even at listening distances). I wore a professional quality hearing protector (the kind designed for use with outdoor power equipment) during the measurements. The software indicated a confidence level of about 38 db on each of the 18 measurements. All of this took a bit more than 4 hours, most of which of course involved moving and replacing the speakers, and setting up the mic, computer, acoustic panels, etc. After viewing the impulse response plots corresponding to each of the measurements (on a much larger screen than my laptop provides), as I had suspected the ones taken with the acoustic panels surrounding the microphone were clearly the best, with the ones taken with the panels surrounding the speaker also being significantly better than the ones taken with no panels in place. The 2.5 foot measurements looked a bit better than the 3 foot measurements, which in turn looked a bit better than the 3.5 foot measurements. But Im inclined to go with the 3 foot measurements due to the concern I expressed earlier in the thread about timing artifacts resulting if very close-up measurements are used, given the somewhat wide spacing of the drivers on my speakers. The next step will be to generate speaker corrections in the software, utilizing those measurements. Ill probably create two sets of corrections, one conservative and one somewhat aggressive, and upload them to separate profiles on the DEQX. Ill then evaluate the resulting sonics with a variety of recordings, before attempting any room corrections. My next significant update is likely to be several days or more from now. Finally, Ill mention that Nyal (AcousticFrontiers), from whom I purchased the HDP-5, was kind enough to contact me the other day to see how I was doing with it, and graciously offered to review and comment on my DEQX project file. Ill probably take him up on that in the coming days or weeks. Best regards, --Al |
Thanks for the nice comments, guys.
Bruce (Bifwynne), I've sent you an .mzd file, containing all of the measurements and also my first speaker calibration attempts, which I'll start to assess sonically tomorrow.
Best,
-- Al
|
Roscoe, no, I didn't take any pix while making the measurements.
Upon re-reading the description I provided of the setup, I would just add that I of course moved the chairs that are visible in my system description photos out of the room, and I mounted the measurement mic on a small professional mic stand which allowed it to be placed at the 30 inch height I mentioned.
Best regards,
-- Al
|
Caution: Very long post :-) Well, it seems that my forward progress has taken a step backward, although the step backward has resulted in discoveries that will hopefully be beneficial in the end, to me and to others who may read this. After evaluating four different speaker correction profiles on numerous recordings, all based on measurements taken with the acoustic panels I purchased placed around the measurement microphone, Ive concluded that even though those measurements, when viewed on impulse response plots over a reasonable time scale (e.g., 30 ms), looked considerably better than the measurements taken with no panels and with the panels placed around the speaker being measured, the panels were doing more harm than good. And I say that even though there were no points on the impulse response plots at which reflection amplitudes appeared to look greater with the panels placed around the mic than in the other plots, and at the great majority of points reflection amplitudes looked significantly smaller with the panels placed around the mic compared to the other two cases. As will become clear later in this post, what I probably should have done was to place the panels up against the nearest reflective surfaces (the fireplace on the left and the large antique radio/phono on the right, as seen in my system description photos), rather than surrounding the mic with them on three sides, at a fairly close distance. Ill first say that one of those four profiles (one of the first two I tried, with the impulse response truncation window terminated 7.2 ms after the initial sound arrival, and corrections only performed between 600 Hz and 10 kHz), on most but not all recordings sounded distinctly better than bypass mode and than the other three profiles I tried (all of which had the window terminated about 18 ms after the initial sound arrival, with corrections over a somewhat broader range of frequencies). However I noted especially on the last two profiles I tried that the image was shifted considerably to the right, even though bypass mode (as always) was perfectly centered. And that right channel boost (the difference being in the area of 2 to 6 db depending on frequency, and occurring primarily between about 200 Hz and 1 kHz) could be clearly seen when the correction profiles were viewed in the DEQX software, with the profiles for the two speakers placed on the same graph and the scale of the vertical axis suitably adjusted. I also found while doing further experiments with the software that the volume difference between the two channels increased dramatically in proportion to the duration of the truncation window, which really puzzled me at first, and also increased in proportion to the distance of the measurement microphone from the speaker (meaning also that it increased as the distance between the mic and the panels behind it became smaller). And the volume difference was worse for correction profiles created from the measurements made with the panels surrounding the mic than for correction profiles created from the measurements made with the panels surrounding the speaker. And, as mentioned above, I found that the issue occurred primarily (although not exclusively) in the area of 200 Hz to 1 kHz, especially around the middle of that area, with the volume difference varying considerably at different frequencies. After a lot of study of frequency response plots, impulse response plots, and step response plots, I concluded with a fair amount of certainty that the cause of the different corrections for the two speakers was that I didnt have the panels placed in precisely the same locations when I measured the two speakers. The reason for the slightly different placements being that since I was making measurements with the panels surrounding the speaker as well as with the panels surrounding the mic, and it happened that I did the measurements with the panels surrounding the speaker last, I moved the panels aside when the first speaker being measured was moved away from the center of the room, and the second speaker was moved into that position. Upon very close examination, the consequences of that can be seen in terms of slight differences between the timing of the wiggles of the impulse response measurements for the two speakers in the area of about 3 ms after the initial sound arrival, and can also be especially seen in the form of a roughly 1 db difference occurring at that same instant between the step response plots of the two speakers, in the cases of the measurements taken at 3 and 3.5 foot distances which I used for the correction profiles. So, I wondered, if the issue was being caused by reflections from the panels occurring just 3 ms or so after the direct sound arrival, why would the consequences of those reflections in the correction profiles get worse as the truncation window was extended much further out in time, for instance from 7 ms to 18 ms and beyond? Im not totally certain, but I believe the answer to that is inherent in the mathematics of the Fast Fourier Transform, some variation of which I assume is what the software uses to convert between the time domain (impulse and step responses) and the frequency domain (frequency responses). Now if I were to redo the measurements while making a point of placing the panels at precisely the same locations for both speakers, I could evidently eliminate the inter-channel differences in the corrections. However, the fact that slight differences in panel placement caused dramatic differences in calibration profiles between the speakers would seem to say that even if I were to achieve identical profiles for both speakers, that identical profile would reflect (pun intended) significant adverse effects of the panels. So for that reason, in addition to the effort that would be involved in re-measuring the speakers, Im not planning to do that. Instead Im now planning to simply try some correction profiles that Ill create based on the measurements Ive already taken with no panels in place. If those dont work out well, then Ill consider re-measuring the speakers, with the panels much further from the mic and probably placed against the reflective surfaces I mentioned earlier. Also, if I were to try to correct the inter-channel differences using the equalization capabilities of the DEQX, given the extensive variations of those differences as a function of frequency I suspect that the effort would be extremely time-consuming, and would probably result in a less than ideal set of complementary colorations. So although my efforts have had a bit of a setback here, its probably a good thing that I didnt make a point of placing the panels in exactly the same positions for the measurements of the two speakers. If I had done so I probably wouldnt have discovered any of this, and I would very conceivably have ended up deriving less benefit from the DEQX than I hopefully will, eventually. So, undaunted, I shall persevere and carry on. Due to various unrelated upcoming activities, my next significant update will probably be in about a week. Meanwhile, just using the DEQX in either bypass mode or with the one correction profile I mentioned as being superior to the others, provides (despite a bit of channel imbalance in the case of the latter) a modest but notable improvement on what I previously had. Best regards, -- Al |
Andrew (Drewan), thanks very much for providing the info in your system description thread. As I indicated there, it provides a good quantitative perspective for DEQX newbies such as myself on what is possible in the speaker calibration process. Mapman, thanks also. Regarding DEQX's recommendations for how to approach speaker calibration, on the one hand the manual states: The microphone should be mounted about 2 ft to 3 ft (or 60 cm to 1 metre) from the speaker and lined up with the tweeter axis. Best results will be obtained if you use a microphone stand with a boom to support the microphone. If larger speakers are used it may be necessary to mount the microphone further from the speaker, in order to reduce the angle between the drivers and the microphone (and any
associated off-axis effects). However, moving the microphone further from the speaker will potentially increase measured room effects and /or reduce the period of the measurement before the first room reflection, hence the reason for the suggested distance of 3 feet. Better results may be obtained by measuring the speaker outside, provided that a quiet location is used. One method, particularly useful for large speakers, is to rest the speaker on the ground, pointing up at a 45 degree angle. The microphone should still be lined up with the speaker's
driver axis, i.e. it will be pointing down at the speaker at a 45 degree angle. When measured with this method, the speaker measurement can be similar to that obtainable in an anechoic chamber. But on the other hand they state: The longer the delay for the first room reflection to arrive relative to the time it takes for direct sound to reach the measurement microphone, the more bass information can be gleaned. A minimum target would be 3 milliseconds, which will limit measurement usefulness to frequencies above about 400Hz. Doubling this to 6 milliseconds for example, achieves an additional octave of bass resolution: down to about 200Hz, which is the lower correction limit used for this case. In this example, if you attempt to correct below 200Hz phase distortion will be introduced to the filter, which will make the resulting sound dull and flat - worse than the uncorrected speaker. The 3 and 6 ms 400Hz/200 Hz guidelines seem very aggressive (optimistic) compared to inputs that were provided to me by Nyal, and that have been cited by others earlier in this thread, and that were used by the DEQXpert people when they set up Bruce's (Bifwynne's) system. (They terminated the "truncation window" about 17 ms after the direct sound arrival in his case, as contrasted with the 3 and 6 ms figures cited in the manual. Also, the corresponding figure in Drewan's case was about 18 ms). So it seems that given the great number of variables that are obviously involved DEQX is understandably not able to provide a specific quantitative feel for the degree of compromise that will be introduced at various points along the continuum between ideal measurements taken under essentially anechoic conditions, and points that are significantly less than ideal. 06-07-15: Drewan77
At the risk of repeating myself (apologies), are you absolutely sure that there is no chance of taking an outdoor speaker measurement? Pretty much. Just moving my speakers off and then back onto the bases and footers I have them on, which elevate them less than 5 inches, involved non-trivial effort and careful logistical planning, as they weigh 108 pounds each. To get them to a location outdoors that is both level and not close to reflective surfaces they would have to traverse several steps and approximately 50 feet, and then I would have to provide some non-reflective means of tilting them back, to minimize ground reflections. Also, less significant additional complications would be that I would have to run about 50 feet of speaker cable (my amplifier also weighs more than 100 pounds), and a 40 foot or so XLR extension for the mic. Well before deciding to purchase the DEQX I had realized from the comments earlier in the thread and from reading the DEQX literature that as a practical matter I would be unlikely to realize its full potential. But as long as it ends up providing me with a substantial improvement over what I had I won't be disappointed. As Mapman aptly said just above, "it might be a dilemma for which one practically just has to accept some kind of compromise." Best regards, -- Al |
Psag and Mapman, thanks very much. Obviously I don't have to finalize any decisions just now, and we'll see how things develop in the coming weeks. But while I certainly value the experience-based inputs from Drewan and Psag, and while it's certainly clear that realizing the full potential of DEQX requires an essentially anechoic measurement, what doesn't seem clear (and probably can't be, given how many application-specific variables are involved), is the degree of compromise that will result in any given situation from doing the measurement indoors. Obviously Drewan and Psag are strong advocates of outdoor measurement. On the other hand, though, in this post by Forrestc, who also seems very experienced, indoor measurement is described as "by no means a deal breaker." And in Kal's (Kr4's) review in Stereophile the speakers he calibrated on his own, and I presume also the other pair he used, which were calibrated remotely by DEQX, were done indoors. The pair he did himself, with good results, were done with the impulse response truncated at only 5.5 ms after the direct sound arrival, and with the correction performed down to 200 and 150 Hz in the two profiles he created, with the latter even being slightly preferred! Perhaps one relevant variable influencing the degree of compromise resulting from indoor measurement, btw, is how much correction is needed by the particular speakers that are involved. In that regard I've noted that the impulse and step responses I've measured on my speakers, during the first few tenths of a millisecond or so and with no panels near them or the mic, seem to me to look pretty good. Relative, that is, to the step response plots I've seen JA present in Stereophile in conjunction with reviews of other floor-standing speakers which do not use first order crossovers. So we shall see. Thanks again for your inputs. Best regards, -- Al |
Ptss, no, my background is in EE, although way back when I certainly had my share of math courses, and a few that touched on philosophy. I also have a law degree, although I haven't practiced as such.
Best regards,
-- Al
|
Thanks very much, Pete, for the nice words and the interesting comments.
An update: I've created and assessed three more speaker calibration profiles, this time based on the measurements I took with no acoustic panels in place. (As I had previously described, the visibly cleanest and most reflection-free measurements had been the ones taken with panels surrounding the measurement microphone, but I found that very early arriving reflections from the panels themselves caused significant differences in the calibration profiles for the two speakers, which worsened in proportion to the duration of the truncation window. The same held true for the measurements taken with panels surrounding the speakers, to a slightly lesser degree but coupled with increased pickup of room reflections. Recall, btw, that I did not have the panels in precisely the same positions when I measured the two speakers, which fortuitously led to me discover the significance of the early arriving panel reflections).
Also, I should mention that when I performed the measurements even though I had pillows and a folded back (double thickness) Persian rug on the floor between the speaker and the mic, the waveforms captured by DEQX for all of the measurements clearly show a significant floor reflection (as can be determined based on its arrival time, beginning a little under 3 ms after the direct sound arrival). There is also evidence of a ceiling reflection arriving a little under 8 ms after the direct sound arrival, although it appears likely to consist mainly of low frequencies that I'm not addressing in the speaker calibrations.
Although all three of the new profiles sounded reasonably good, ultimately I was not satisfied with any of them. Most notably on classical solo piano, where the presentation had a slightly mechanical quality, and a slight loss of definition in the mid-treble, that were not present in bypass mode.
Given all of that, what I've decided to do is to order a third double-section acoustic panel and then re-do the speaker measurements. This time I will place one panel up against the fireplace you can see along the left wall in my system description photos; one panel up against the antique radio/phono console on the right; and one panel on the floor between the mic and the speaker being measured (positioned once again in the center of the room). In my circumstances it is reflections from the sides and the floor, and possibly the ceiling, that appear to be the main concerns. I believe that moving the speaker being measured to the center of the room gets them far enough away from the front and rear walls to allow those reflections to be windowed out without significant compromise to the calibrations.
Apologies for the slow progress to those who may be awaiting the conclusions of this adventure. I haven't even begun to address room correction yet. But from my perspective the slow progress has not been bothersome, as what I have in bypass mode and with one or two of the correction profiles is already better than what I had previously (which in turn I was pretty satisfied with). And it's been an interesting learning experience.
Best regards,
-- Al
|
|
Bruce, yes, thanks for sharing these things with me. I'll look forward to the measurements and/or pics, in part because I'm curious as to what kind of room configuration they would consider to be too small for your speakers.
Roscoe, so far I've tried 7 different speaker calibrations, based on my initial set of measurements (which I will be re-doing from scratch as described above). The lower frequency limits I've used in those calibrations have ranged from 300 to 600 Hz, depending on where the truncation window was terminated.
Nyal had advised that I start out very conservatively with respect to that parameter, even as high as 900 Hz if the end of the truncation window would have to be as close as 5 ms after the direct sound arrival (with other rule of thumb combinations being suggested such as 600 Hz/10 ms, and 350-400 Hz/15 ms, if the truncation window could be extended to those points without encompassing major reflections). Those are obviously more conservative limits than the ones suggested in the manual, but he also suggested doing additional trials calibrating to progressively lower frequency limits, and trying to identify the point beyond which the sound starts to worsen.
In setting the limits I kept those rule of thumb guidelines in mind, while also, as I created each calibration, going back and forth between the screen on which the truncation window is set, and the following screen which depicts the resulting frequency response. In doing that I tried to set the lower frequency limit of the calibration above the frequency at which significant differences started to become apparent in the frequency response plot as I varied the duration of the truncation window.
Also, at the other end of the spectrum Nyal had advised that I should avoid monkeying very much with the speaker's natural rolloff at high frequencies. And in a separate conversation I recently had with Alan Langford of DEQX, (regarding an unrelated matter involving a very minor quirk I had noticed and reported in the operation of the touchscreen that is provided on the HDP-5, about which they were extremely responsive and will be resolving shortly with a couple of changes the user can easily incorporate, mainly a firmware update), he suggested limiting the high frequency limit of speaker calibration to 10 kHz. Given also that my particular speakers have a rise of a few db between 10 and 20 kHz, according to the DEQX measurements, followed by a rolloff above 20 kHz, I certainly plan to follow their advice on the subsequent calibrations. Although the ones I've done so far have had various limits ranging from 10 to 23 kHz.
As you'll realize from my previous recent posts, I can't really say much at this point from my listening experiences as to which of those limits (both at the low end and at the high end) work best with my speakers, because the anomalies caused by non-optimal placement of the acoustic panels when I did the initial measurements affected the middle frequencies and mid-treble too greatly for those other limits to matter very much.
BTW, during the course of the conversation with Mr. Langford he indicated that as a general rule of thumb outdoor calibration often tends to be less necessary for box-type speakers than for other kinds.
As for my status at this point, I've just received the third acoustic panel I ordered the other day, but most of this week I'll probably be too tied up with various family and other obligations to perform the next set of measurements. Hopefully I'll get them done within about a week.
Best regards,
-- Al
|
Good question, Roscoe, as I should have clarified that statement further.
If I position the window cursor close to the direct sound arrival, say within 3 ms or so to cite an extreme example, and then look at the frequency response on the next screen, there will be a downtrend in the frequency response below something like 500 Hz, roughly speaking. Presumably that occurs because the short duration of the window doesn't make it possible to sufficiently capture low frequency information. (Consider that 3 ms is less than one cycle of all frequencies that are below 1/0.003 = 333 Hz).
If the window cursor is then moved way out, to say 20 ms or so, the frequency response shown on the next screen will show a substantial rise in bass response at many low frequencies, with perhaps a narrow dip or two at certain frequencies.
Intuitively it seemed to me to make sense to set the lower frequency limit of the calibration around the inflection point below which that decrease or increase in bass response starts to occur for the chosen window duration.
This is all consistent with a comment Nyal had made to me that a common mistake is to correct to too low a frequency relative to the duration of the truncation window, the result being an excessive bass boost.
And consistent with various statements in the calibration manual, the subsequent room corrections should be a more suitable means of addressing the bass region, or at least most of it.
Best regards,
-- Al
|
Interim update: The plan I previously mentioned to re-measure my speakers has been delayed a bit, in part because last week a sudden and particularly violent thunderstorm resulted in damage to the power supply of my HDP-5. But to make a short story shorter, DEQX responded in an incredibly prompt and supportive manner, and I was back in business within a few days.
I now expect that I'll have time to perform the measurements sometime next week.
Kudos to DEQX!
Best regards,
-- Al
|
Bruce, no, as it turned out I didn't get a chance to re-do the speaker measurements this week. And I've got a number of other things scheduled for the next several days. So it might be another week or so before I'm able to do the measurements.
Meanwhile, though, the HDP-5 continues to please just as a preamp and DAC upgrade, used in either bypass mode or with the one fairly conservative speaker correction profile I indicated early on was superior to the others.
Best regards,
-- Al
|
Hi Bruce, The Quatro CT does seem intriguing to me, with respect to your circumstances. Presumably there would be no need with it to go to the trouble of doing speaker calibrations with the DEQX, and of course doing room corrections is a much simpler process that does not involve moving the speakers. Also, I would think you could do away with the sub you are using. So there would be some nice simplifications, albeit for significant $. On the other hand, though, JA's measurements of the earlier version of the Quatro raise some caution flags in my mind, starting with the sensitivity measurement that is significantly lower than specified. Best regards, -- Al |
I finally found the time today to re-do my speaker measurements, as I had previously indicated I was planning to do. A link to photos of the measurement setup and screenshots of some of the measurements is provided further down in this post. Once again I moved the speaker being measured to the center of the room. However, instead of positioning the large acoustic panels I had purchased ( Clearsonic S5-2D's) close to either the microphone or the speaker, as I had previously done (which created early reflection problems from the panels themselves), I positioned one of them in front of the fireplace on the left, and one in front of the large antique radio/phono console on the right. I also positioned a third panel, which I purchased subsequent to performing those initial measurements, on the floor, under the mic stand and extending to the front of the speaker. It pretty much eliminated a significant floor reflection that was evident in the earlier measurements, despite pillows having been placed on the floor at that time. The panels on the sides were located 50 inches from the nearest side of the speaker. The front baffle of the speaker was 10.5 feet from the wall and window behind it. The nearest objects in front of the speaker were for the most part around 7 or 8 feet away, and not directly in front of the speaker or directly behind the mic, although some objects off to the side were as close as 5 or 6 feet. I measured each speaker from distances of 30, 36, and 42 inches. The 30 and 42 inch measurements were performed with the speaker grilles removed. The 36 inch measurements were performed both with and without grilles. As it turned out, though, the 36 inch measurements looked essentially identical with and without grilles. I've uploaded to flickr.com three photos of the measurement setup, as well as a set of screenshots of the 36 inch measurements sans grilles. Each of the screenshots includes measurements of both speakers, superimposed, and with no smoothing. You can see all of this here. Looking at the impulse response screenshots, the small wiggles occurring at around 14 ms (about 8 ms following the direct sound arrival) I'm pretty certain are a reflection from the ceiling. What strikes me as perhaps the most significant concern, that I see initially, is the spike in the impulse response occurring just 0.7 ms after the direct sound arrival. I'm not sure what that is, although based on its arrival time I suppose it figures to be a baffle reflection. In the coming days, or more likely weeks, I'll be creating and sonically assessing a number of correction profiles. When I settle on one that seems best, I'll then address room corrections, which I haven't addressed at all so far. Apologies for the slow progress to those who may be interested in the conclusions I'll ultimately reach. Best regards, -- Al |
P.S. to my post just above: When I said ... What strikes me as perhaps the most significant concern, that I see initially, is the spike in the impulse response occurring just 0.7 ms after the direct sound arrival. I'm not sure what that is, although based on its arrival time I suppose it figures to be a baffle reflection. ... I'm thinking that perhaps it also could have been a reflection from the mic stand. Can any of the experienced DEQX users offer an opinion as to how serious an issue that spike, and the subsequent brief ringing, might be? Best regards, -- Al |
If you haven't already seen it, take a look at the comments about DEQX on pages 3 and 4 of the recent "Sloped Baffle" thread. (Expand the contents of all of the posts on each page if necessary, by clicking on the date of the first response on each page, and use your browser's "find" function (usually under the "edit" menu) to search the page for "DEQX"). I've read through all of the the writeups at their website, and I certainly find it to be an intriguing product, especially given that it can in many cases serve as a preamp as well as providing sophisticated room and speaker correction. And I'll say that my BS meter never budged above zero while reading through those materials. Regards, -- Al |
^ No question about it!
Best,
-- Al |
Congratulations, Bruce! Glad to hear that the time and effort you've put into this assessment has paid off.
Best,
-- Al
|
|
Thanks for the invite, Bruce. I'll definitely keep that in mind. I bought the rep's demo, which was a PreMate. So I have DAC capabilities ... to be explored and understood in the coming days and weeks. Question: Have you or will you be trying the DEQX for vinyl listening with your ARC preamp removed from the signal path? Best, -- Al |
Bruce & Lloyd, A couple of comments that were made earlier by some of the experienced DEQX users who have been making outstanding contributions to the thread:
08-28-14: Drewan77
DEQX is also an excellent preamp, completely neutral and very analogue sounding to my ears (I play a lot of vinyl) and also contains very good DACs (only bettered by a new Graham Slee product called the Majestic. I had previously used a Chord 64 and the DEQX DAC was much more musical, the Slee even more so).
08-28-14: Psag
Another amazing aspect is that the preamp section is utterly transparent, to my ear, with an analog source.
[Although Psag noted on 9-18 that "For CD and vinyl, I continue to use my tubed preamp."]
09-17-14: Drewan77
I used to use DEQX for both analogue/digital, now only analogue (although it contains an excellent DAC, I now use something even better). The two analogue inputs are configured as below
1) RCA Phono analogue input: Turntable/phono amp
2) XLR Balanced analogue input: balanced connection to Graham Slee Majestic DAC/preamp. This has an analogue input that I use for SACD/second TT, plus multiple digital inputs (coax, optical, USB) for the CD transport/digital streamer/laptop etc.
I also use the DEQX analogue volume control. Overall there is no trace of anything 'digital' in the sound I hear which is a testament to the smoothness of the DEQX processor. You have to hear it to believe me of course!
Having reviewed the information provided at the DEQX site pretty thoroughly, I see no reason whatsoever to question their observations, putting aside dogmatic biases some may have against the A/D conversion which the DEQX performs internally when provided with an analog input. What happens to my ARC Ref 5 SE??? That was the point to the question in my previous post. Based on everything that has been said in this thread, and on the info provided at the DEQX site, it seems to me to be highly conceivable that you may get results that are as good or better with it removed from the system than you are now getting with it in the system, albeit perhaps with some slight further tweaking of the DEQX settings. In which case you could sell the ARC preamp and perhaps more than recover the amount you paid for the DEQX. Best regards, -- Al |
Excellent post by Bruce (Bifwynne), IMO. Like Cerrot, in general I too am biased in favor of minimizing what is in the signal path. But digital signal processing can do amazing things these days, that often are either not possible in the analog domain, or that cannot be achieved in the analog domain without significant tradeoffs. In this case, those tradeoffs begin with the fact that limiting one's choice of speakers to those that are time coherent rules out most of the speakers that are on the market. And for various reasons, electrostatics such as Cerrot uses are not for everyone.
In any event, putting its time correction feature aside, DEQX seems like a promising candidate in its price range just for its room correction, preamp, and DAC capabilities.
One minor correction to Bruce's post: An Nth order crossover rolls off at 6N db/octave, so 3rd order = 18 db/octave.
Best regards,
-- Al
|
Bombaywalla, I'm pretty certain that none of the DEQX products, or at least the current ones, include phono stages. Given that, and also given the considerable level of expertise and experience that is evident in Drewan's posts, and also given that RCA connectors are sometimes referred to as "phono connectors," I would expect that the word "phono" in his statement was simply intended to distinguish the unit's RCA input from the balanced XLR input referred to in his next sentence.
Best regards,
-- Al |
Thanks, Kal! I'm looking forward to reading it.
Best regards,
-- Al |
Just got my copy of the December Stereophile. Kal, thanks much for your characteristically thorough, nuanced, and excellent review. Thanks also to Drewan, Psag, Forrestc, and Bruce for your comments on your DEQX experiences. I'm sold, and I expect to order one sometime this winter (don't want to do it now for unrelated reasons). In my case it would be an HDP-4, in part because I want the three sets of outputs it provides. Regarding Roscoe's mention of the jitter measurements, and the slight misgivings JA expressed about some of the other measurements, those all involved noise and spurii that were so far below the levels of the test signals (in nearly all cases considerably more than 100 db below, at any individual frequency), that IÂd be surprised if they had any audible significance. Plus the manufacturer's response to the review indicates that the jitter performance of the current design has been improved by the addition of a "very low-noise power supply regulator." 11-12-14: Bombaywalla
Bruce,
good to read that you continue to like your purchase of DEQX & that you've come over to the side of time-aligned speakers. :-) Glad you recognize & hear what time-alignment can do for music playback - I feel that all my posts weren't all in vain...... at least one person listened & benefited. :-) So I expect that to become at least two persons! Thanks :-) Regards, -- Al |
Thanks for the nice words, guys, and especially to Andrew for the kindly offer.
I'll most likely be ordering the HDP-4, together with the DEQX/Earthworks M23 calibration kit, in January or February. As a technically oriented person, I'll try to implement all of the procedures myself, at my own (slow and deliberate) pace. I suspect that will extend over at least several weeks before I either declare the profiles and settings to have been finalized, or decide that I need to take Andrew up on his kindly offer and/or utilize the DEQExpert service.
I'm planning on using the HDP-4 in place of, rather than in series with, my existing preamp (a Classe CP-60), which receives inputs from five different sources, and provides outputs to three different destinations.
My two most critical sources are CDP and phono. Pending possible revision during my listening tests, I'm planning on connecting a digital output of my Bryston CDP to the HDP-4 via AES/EBU. I'll connect the output of my phono stage (actually, the phono section of a vintage Mark Levinson ML-1 preamp, accessed via tape outs) to the HDP-4's unbalanced analog input.
For the less critical sources, I'll connect a digital output of my Squeezebox Touch (which I only use for internet radio) via Toslink. I'll connect the outputs of my vintage tuner and Tandberg cassette deck (I still have occasion now and then to play some musically and sonically excellent classical cassettes from way back when on the Connoisseur In Sync Label) to a mechanical switchbox that will select between them, with the output of the switchbox connected to the HDP-4's balanced analog input via RCA-to-XLR adapters.
I'll be connecting one of the HDP-4's three sets of outputs to my VAC power amp, one to my STAX headphone amp, and one to the cassette deck (although I can't recall the last time I ever recorded anything with it). Obviously the outputs to the Stax and the Tandberg will be configured for bypass mode.
Thanks again. Best regards,
-- Al
|
Thanks for your further comments, gentlemen.
Psag, even if I do end up removing the Classe CP-60 from the system, I don't envision selling it in the foreseeable future. It's much too good a performer, IMO, to be selling it for what I suspect it would bring. (I paid $1350 for it about 6 years ago, when I believe it was something like 8 years old). So I'd keep it as a backup, or possibly use it in a second system.
But regarding keeping it in the system along with the DEQX, my instinct is generally to have as little in the signal path as necessary. Also, as you can tell from the photos in my system description, my setup can't readily accommodate both units physically. But we'll see, of course, how the sonics work out with the DEQX installed in the configuration I described.
Best regards,
-- Al
|
Andrew (Drewan77), awesome! Thanks! 11-19-14: Timlub
Hi Guys, I have read several post, but not near all, this may have been addressed. I have no doubt that the DEQX could be a game changer, but it would require Biamping, triamping etc. Tim, my hope and expectation, based on the inputs from several DEQX users in this thread, and on Kal's review, and on the comments on time coherence by Bombaywalla and others in this thread and the recent "sloped baffle" thread, is that that is not the case. My Daedalus Ulysses speakers are not even biwirable or biampable, having just a single pair of terminals. Designer Lou Hinkley doesn't want users doing those things, and in the process risking introducing compromises to what he has worked hard to achieve. I simply hope and expect that the sophisticated time alignment and room correction features of the DEQX, and possibly a little bit of additional equalization here and there, will enhance the already good sound I am getting from them to a significant degree. As well as perhaps also providing an upgrade to my preamp, and maybe even to the DAC and analog sections of my CDP. As Kal said in the conclusion of his review: It made my very good speakers undeniably better, smoother and cleaner, and endowed them with a bigger soundstage. It mad dense complicated music easier to resolve, and all music more of a joy to hear. Best of luck with the Mini-DSP, btw. I am not familiar with it. BTW, I'll add to my previous comments about the system configuration in which I intend to use the DEQX that I do not envision adding subs in the foreseeable future. A majority of Ulysses owners do not use them with subs, although Lou offers what is apparently quite a good (and somewhat expensive) passive sub that is designed to mate with them. But the Ulysses are rated down to 28 Hz +/-2 db, which is good enough for me. And that rating seems consistent with evaluations I have performed using test tones, as well as listening to well recorded organ music, etc. John, LOL! :-) I wonder what the "three amigos" would have to say about that. (To the others, that's an inside joke; don't ask). Come to think of it, though, I don't want to know :-) Best regards, -- Al |
Tim, Andrew, Denis, thanks for the excellent inputs. Points taken. 11-19-14: Timlub
When building speakers, we time drivers [that] are not time aligned on the frontal plane by adding padding/baffle step compensation to change the delivery of the tweeter and mid hitting your ear at the same time. This is built into the crossover. An External device cannot change that as far as true timing speed is concerned. Hi Tim, You may already be aware of this, but just to be sure the DEQX processing divides the spectrum into thousands of segments, adjusting the delays of each of them such that time coherence is attained even in the presence of a passive crossover within the speaker that is higher than first order, which would normally make time coherence impossible. See the graph and the references in the text to group delay in this writeup at the DEQX site, as well as the HDP-4 brochure and their FAQ writeups. Best regards, -- Al |
|
11-22-14: Bifwynne
... my "redbook" ARC Ref CD-8 CDP is hooked up to the PreMATE via two modalities: (1) analogue through my ARC Ref 5 SE preamp in "normal" analogue fashion and (2) digitally via the digital output of CD-8 directly into the PreMate's XLR input. In the later case, my CDP is acting merely as a transport and the PreMATE is the DAC. Hi Bruce, At the risk of adding further complication to an already very complicated assessment and adjustment process, I'll mention that I would keep in mind the possibility that the sonics you hear from the analog and the digital outputs of your CDP MIGHT be affected, at least slightly, by having both of those outputs connected into your system simultaneously. Having multiple paths in the system through which signal return (ground) currents can flow, some of them associated with digital circuitry and some with analog circuitry, raises a caution flag in my mind about the possibility of one affecting the other, at least slightly. And also the possibility of issues arising from ground loops involving those signal return connections and the AC safety ground wiring of the three components. And, finally, the possibility of low-level "crosstalk" within the DEQX between the digital and analog inputs, if and when signals are provided to both at the same time. Best, -- Al |
11-22-14: Lewinskih01
Al, since I'm posting I wanted to throw in a comment directed to an earlier post of yours. In Acourate, the treble, mids and bass need to be in different channels for the software to be able to time align them. I believe the setup you were planning with DEqX had just one channel for right and one one for left (from a digital processing point of view). I would double check DEQX would allow you to time align the drivers in such a setup. Thanks, Lewinski. Your interpretation of my intended application is correct. And the very question you raise had in fact occurred to me some time ago. But the writeups at the DEQX site make very clear, as I interpret them, that in addition to being able to time align independently powered drivers, their present-generation processors can also restore time coherence (to a good approximation) within and throughout however much of the audible spectrum can be reproduced by whatever each of its output channels is providing a signal to. Otherwise, for one thing, that processing would be worthless with respect to correcting the coherence issues of in-speaker passive crossover networks that are higher than first order (6 db/octave). Also, I don't doubt that were I all mixed up about that, Drewan, Psag, et al, would have pointed that out to me when I described my intended system configuration. Best regards, -- Al |
A'gon audio pal here :-)
Roscoe, congratulations on the purchase! As Bruce (Bif) indicated, I'm looking forward to reading of your experiences.
My plan is to purchase the recently released HDP-5 in March or thereabouts, although the HDP-4 would no doubt serve my purposes just as well. But I find myself unable to resist the added coolness of the touchscreen display on the -5.
Enjoy! Best regards,
-- Al
|
Thanks very much, Roscoe. Yes, I noticed that the HDP-5 has the "low noise" switching power supply, presumably similar if not identical to the one used in the PreMate, rather than the linear supply used in the -4. Just speculating, but perhaps the reason is simply related to internal real estate, with the touch screen and its associated circuitry taking up too much space to allow the presumably larger linear supply to fit.
As far as I've been able to tell from the website the only difference between the -4 and -5 besides the ones you cited is that the -5 includes the USB interface module as standard, within its $1K higher price, rather than as a $500 option.
Enjoy! Best regards,
-- Al
|
Thanks very much, Drewan. I've been keeping in mind the kindly offer of assistance you had made earlier in the thread.
Best regards,
-- Al
|
An update re my situation: I had indicated a while back that I was planning to order an HDP-5 around this time. Last night I contacted Nyal Mellor of AcousticFrontiers.com, who is a DEQX dealer having extensive specifically relevant expertise. (He is also an Audiogon member, btw, as "acousticfrontiers," as well as being a participant in various other forums). He indicated that DEQX is currently implementing some revisions to the initial HDP-5 design, which will be incorporated in a production run that will be available in the USA approximately mid-May. So I'll be waiting until then to place my order. Nyal also pointed out that he has some DEQX-related videos on YouTube, which I haven't yet watched but I certainly will. Best regards, -- Al |
Hi Bruce,
The revisions weren't specifically indicated, and it appeared that Nyal was just relaying information he received from DEQX which in turn didn't go into a lot of detail. But it sounded like the revisions were most likely just bug fixes specific to the HDP-5. I suspect they had no applicability to the more mature models.
Best,
-- Al
|
Unsound, as far as I am aware DEQX does not make power amps, class D or otherwise. And I've seen no indications that HDMI inputs are or will be incorporated in any of their products. Regarding 24/192, in JA's measurements writeup that was presented in Stereophile in conjunction with Kal's review of the PreMate it was stated that: The USB input accepted 24-bit data with sample rates ranging from 44.1 to 192kHz, including 88.2 and 176.4kHz. The TosLink input accepted data with sample rates up to 96kHz, and the AES/EBU input sample rates up to 192kHz. However, the DEQX Calibration app (v.2.93), running on a Windows XP machine, indicated that 192kHz data were downsampled to 96kHz, and 176.4 down to 88.2.... Though the PreMate will accept 176.4 and 192kHz datastreams, it appears that it downsamples them to half those rates before the data are presented to the DSP section, then finally the DAC. So no, there is no true bypass.... Running powerful DSP at 4Fs sample rates is very consuming of resources, so this compromise is not uncommon. It is likely that the benefits of the DSP correction outweigh the potential drop in sound quality due to the downsampling. Best regards, -- Al |
Hi Roscoe, In my initial email to Nyal I said: ...if you should happen to be aware of any issues which might result in the performance of the HDP-5 being inferior to that of the more mature HDP-4 I would of course appreciate being apprised of them. I realize that the HDP-5 utilizes a switching power supply while the HDP-4 has a linear supply, but I assume the design is well implemented and that is not an issue. Neither of his subsequent responses made reference to that question. His many posts that I've seen at various forums are confidence inspiring, and I feel safe in assuming that if he was aware of any negatives about the new design he would have mentioned them. Best regards, -- Al |
Thanks, Pete. I appreciate your comment, and the nice words. And Bombaywalla, thanks also for your good comments.
I would place myself somewhere in the middle ground between being overly trusting and overly cynical, and I try to judge each case on its own merits without bias in either direction. As you (Pete) observed, I tend to be pragmatic. In this case, adding to my confidence is that based on the exchange I had with him I'm pretty certain that Nyal has not yet sold or possessed any HDP-5's. Assuming that is the case, he would most likely not be in a position to respond either way to my question.
And in fact in looking at the websites of the various dealers listed at the DEQX site I had found that none of the ones in the USA make any reference to the HDP-5 being available. I found just one dealer, in the UK as it happens, indicating it as being in stock (presumbly from the first production run, bugs included).
And several additional factors add to my confidence: First, the PreMate which Bruce (Bifwynne) purchased, and that Kal and John Atkinson reviewed, uses a switching power supply. Perhaps some or all of the earlier models that are being used by the more experienced DEQX users who have been participating here do as well. Second, as I indicated what I will be purchasing will be from the second production run, which is explicitly stated to be incorporating some needed "corrections." And, finally and perhaps most importantly, Nyal offers 30-day return privileges, less only two-way shipping.
Thanks again. Best regards,
-- Al
|
