SME Posted August 23, 2016 Author Share Posted August 23, 2016 I read through this thread with awe and jealousy! Unbelievable setup and execution! Top notch driver and amplifier selections and that custom DSP! I can't imagine how amazing your audio experience will be when finished. I wish i could do this for my home, but unfortunately I will have to get rid of my current HT setup in a couple of months and I don't know when I can have a HT again. How are you planning to setup the nearfield subs? I currently have a nearfield ported UM15 behind my couch. Whlle it is pretty awesome for movies, I find it distracting for music because it sounded a bit muddy and it felt unnatural to feel the couch vibrate with the music but not my chest or my clothes. Thanks! I have a curved sectional sofa where each seat is approximately angled toward the center of the screen. It has a total of 7 back seat cushions and is about 10 feet wide at the extremes. The center of the sofa is up against a wall that extends with openings on both ends. I need to post a picture. The near-field subs (a total of four) will be distributed behind the sofa. They will be the same height as the back of the sofa so that they function as end-tables. Each will consist of a dual-opposed pair of 10" drivers in side-firing configuration and a 6" down-firing port tuned in the 30-45 Hz range. I haven't yet decided on the exact tune, and I plan to finalize the choice after they are mostly built. Quote Link to comment Share on other sites More sharing options...
SME Posted August 24, 2016 Author Share Posted August 24, 2016 The SEOS-15 baffle openings are a real pain to route out, even ignoring the time and materials spent to recreate the cut-out template, but they are done now. The result looks very clean, except for a slight vertical alignment issue between cutouts. I turns out I needed to offset the inner cut-out because of the 10 degree slant on the face. Thankfully, everything else fits together very well. Assembly is in progress now, and I expect to start gluing tomorrow. If all goes well, I may have a new center channel as soon as this weekend. That means I need some of the new drivers I have on order to ship ASAP, so I have something new to build. Quote Link to comment Share on other sites More sharing options...
Ricci Posted August 24, 2016 Share Posted August 24, 2016 Thanks! I have a curved sectional sofa where each seat is approximately angled toward the center of the screen. It has a total of 7 back seat cushions and is about 10 feet wide at the extremes. The center of the sofa is up against a wall that extends with openings on both ends. I need to post a picture. The near-field subs (a total of four) will be distributed behind the sofa. They will be the same height as the back of the sofa so that they function as end-tables. Each will consist of a dual-opposed pair of 10" drivers in side-firing configuration and a 6" down-firing port tuned in the 30-45 Hz range. I haven't yet decided on the exact tune, and I plan to finalize the choice after they are mostly built. What's the thought process behind tuning to 30-45Hz and not lower? 1 Quote Link to comment Share on other sites More sharing options...
SME Posted August 24, 2016 Author Share Posted August 24, 2016 What's the thought process behind tuning to 30-45Hz and not lower? It's a matter of compromise and the fact that this build is aiming for sound quality over quantity. The front room subs will blow these out of the water in terms of sheer output, but seat-to-seat response variation is a big issue above 40 Hz. At the same time, the locations behind the sofa perform poorly in the 20-45 Hz range. With the room EQ I am planning to implement, I might still be able to do something useful with the behind-the-seats subs in that range, but it's much less important than 45 Hz and above. I also debated going sealed using drivers with a bit more motor and mass like the AE AV15 for more ULF, but it wasn't worth the trade-offs. I decided that a D.O. configuration should be a priority so that they can be used as tables without things vibrating off, and that requires a side-firing configuration to fit at all. Then there is the issue of minimizing depth so that I don't have to pull my sofa any farther forward. There's only 11 feet between the front wall and the partial wall behind the sofa. This is why I opted for 10" instead of 12" or 15" drivers, which would offer more output. The ULF increase from adding four D.O. 10" sealed boxes would be paltry (with all the subs playing the same signal), but the mid-bass gain for going ported is quite substantial (with different subs playing different signals). Then there's the question of where to tune. I want to play down to at least 45 Hz. Lower is better for better integration / phase matching and more room EQ capability, but I lose more output in the 45-60 Hz range and the port resonance gets lower too. Then there's also the issue of fitting the port in the box while keeping turbulence to a minimum, and I don't know whether I'll see any effective length increase by down firing the port at whatever distance from the floor I ultimately settle on. In the end, the practical differences between a 30 Hz and 45 Hz tune are pretty minor, and I will wait until they are built before optimizing this choice. For that matter, there's nothing stopping me from tuning each box a little bit different as each will get its own amp/DSP channel. Quote Link to comment Share on other sites More sharing options...
Ricci Posted August 24, 2016 Share Posted August 24, 2016 Are you concerned about cancellation of the main subwoofers output at the seats from the nearfield subs at / near the high pass? Quote Link to comment Share on other sites More sharing options...
3ll3d00d Posted August 24, 2016 Share Posted August 24, 2016 Are you concerned about cancellation of the main subwoofers output at the seats from the nearfield subs at / near the high pass? I would think he's using an appropriate FIR filter to deal with that, at least that's how I handle it Quote Link to comment Share on other sites More sharing options...
SME Posted August 24, 2016 Author Share Posted August 24, 2016 Are you concerned about cancellation of the main subwoofers output at the seats from the nearfield subs at / near the high pass? Yes. For a given HPF cutoff, the lower the tune the less aggressive the signal processing needs to be to retain acceptable phase match. But the devil is in the details, so I prefer to remain flexible until it's time to do the actual integration. Quote Link to comment Share on other sites More sharing options...
SME Posted August 24, 2016 Author Share Posted August 24, 2016 I would think he's using an appropriate FIR filter to deal with that, at least that's how I handle it Correct. And see above. Quote Link to comment Share on other sites More sharing options...
3ll3d00d Posted August 24, 2016 Share Posted August 24, 2016 what are your main subs going to be? Quote Link to comment Share on other sites More sharing options...
SME Posted August 24, 2016 Author Share Posted August 24, 2016 what are your main subs going to be? A great question, especially given that I can finally answer it as of today: (4) Sealed 21" Modified Focusworks GUJ21V2 as two dual-opposed pairs. Quote Link to comment Share on other sites More sharing options...
lowerFE Posted August 25, 2016 Share Posted August 25, 2016 You mention you want to improve the seat to seat variation of >45Hz. However, in my nearfield sub experiments, even 1 feet away (which is closest distance from the nearfield sub's woofer to my ear) the FR is still very much affected by room modes. Have you ever experimented with nearfield subs in your room? Quote Link to comment Share on other sites More sharing options...
SME Posted August 25, 2016 Author Share Posted August 25, 2016 Yes I have. I am running two Hsu MBM-MK2 behind my sofa, one on each side, right now. These units are down-firing, so distance to the ear is closer to 1 meter than1 foot. Anyway, the term "near-field" is inappropriate because you can't really sit near-field to both (or all four) MBMs at the same time. They do have a cleaner response, on average, for much of the frequency range compared with the mains and subs up front. The responses at the MLP look very clean from 100-200 Hz at the MLP without DSP, and so things are probably pretty close to near-field there. Below 100 Hz, room effects dominate their responses. I say room effects because early reflections play a role at least as big. Aside from a few frequencies that stick out like sore thumbs, most of my response ills shift around with different listening positions. Because my room is asymmetric, the responses are very dissimilar, no matter where you sit on the sofa. Some interesting patterns emerge though. A particular sub or MBM tends to have good control over some frequencies and poor control over others. Where there is good control, there is usually fairly strong output and fairly even coverage across the seats. Where there is poor control, coverage is erratic with some seats being very hot and some being in nulls, and unfortunately, the peaks and nulls often shift with location. A room EQ strategy that emerged from these observations is to try to produce most of the bass from subs or MBMs with the most control. If two or more subs/MBMs have good control for a particular range of frequencies, then they tend to combine well together and yield responses that are better than with either alone. However, a sub or MBM with poor control will frequently combine poorly with others and can even do more harm than good to the response of a sub or MBM with good control. So instead of sending the same signal (apart from different crossovers and delays) to each sub or MBM, I opted to use independent EQ to further improve integration, and a lot of the manual optimization process I followed consisted of directing signal to subs/MBMs with more control. It took some iteration to get the filters right, and I did everything using simulations against measured data to avoid having to constantly remeasure everything at multiple listening positions, but in the end, I got a big improvement in response smoothness and seat-to-seat coverage vs. just feeding everything the same signal (except delays and crossovers) and running Audyssey. The waterfalls improved a lot too, and the difference in sound quality was very apparent. Basslines sounded cleaner and tighter. Bass transients became much more audible. tactile, and subjectively powerful. Many transient details emerged that I'd never heard before in familiar content. The bigger surprise is that the bass sounded cleaner in the rest of the room too. There was much less bass build-up near the boundaries, less mud and overhang on notes, and so on. I originally implemented this sub EQ configuration using a MiniDSP 2x4. My LFE channel response looked great, but I still had a lot of problems integrating the subs with the mains, even with sub distance optimizations. I found my mains were screwing up the sub response at frequencies as low as 40 Hz, even with a 100 Hz crossover on a 55 Hz tuned ported speaker. So later, I added MiniDSP OpenDRC-ANs for to customize processing for the mains, which yielded another big performance gain. At that point, I abandoned Audyssey completely. However, I figured out quickl I could do a lot better still. Now I am running all these old filters on my custom DSP system, which will enable me to explore even more powerful room EQ with longer FIR filters and matrix processing. In the "surrounded by bass" approach, bandwidths are allowed to overlap and crossovers don't really exist. Instead, responses are shaped to yield optimal results in the listening area subject a defined set of headroom constraints. This overlapping approach must be constrained as frequency increases to avoid localization, but it may be possible to use these techniques to achieve more accurate localization cues at the seats than without. Thus, the region of correction could extend well into the mid range. The multiple side-firing D.O. configuration provides something like a line-array to counter-act some of the interference patterns that would otherwise be seen for higher frequencies. It remains to be seen just how high I can get away with using the MBMs, but the drivers themselves are perfectly capable of playing cleanly to 1 kHz and above. I should note that the predefined headroom constraints as part of room EQ are critical. IMO, a fundamental flaw in room EQ systems like Audyssey and Dirac is that they are operating in the dark with regard to headroom. Educated guesses about the capabilities of the system from raw measurements are no more than educated guesses, and this may be part of the reason that Dirac Unison (which does support integration of multiple sources) doesn't seem to have found applications beyond cars, where equipment specs and acoustics are tightly controlled. But anyway, this is getting more into the details of where this project is going once all my speakers are built. Quote Link to comment Share on other sites More sharing options...
SME Posted September 5, 2016 Author Share Posted September 5, 2016 My center channel is done and installed on its shelf above the TV, but things have been crazy with plumbing issues and getting ready for family visit. I've only tried it with a surround music sample and a video game. I tried music with PL2x, and quickly recognized just how much PL2x screws up the tonal balance of the sound. The surround and depth effects in the music are actually better with just two speakers than with PL2x. Anyway, from what I've heard, it sounds very good, but the one thing that bothers me is that the dialog seems to originate more distinctly from above than I'd like. That surprises me because the horn is only a few degrees higher than the horns on the left and right are. Perhaps elevation is easier to hear after it passes some threshold? Or maybe the ear and brain get some height information from the woofer, which is much higher. Another possibility is that it'll sound a lot better once the ceiling reflection is treated. The ceiling reflection is by far the worst reflection, even though is consists of content only below 2 kHz and is hard to see on the ETC. One thing I opted to do is initially set the DSP to be identical to the other front speakers, except for the inter-driver delay as the woofer and horn are inverted and somewhat higher in elevation. I even used the same gain. So far, I really like the results, but I really need a lot more time to evaluate. Maybe tomorrow when family is here, someone will want to watch a movie or something. Other than that, I'm not happy with the performance of the shelf brackets I bought and will want to rebuild it with better brackets. The speaker is maybe 70 lbs total, and the shelf f is 9" deep, and the ends of the brackets sag maybe a couple degrees or so. They also have a narrow natural resonance around 2.5 Hz. Sadly, *first* set of brackets I bought with actual cross braces had other design flaws that made them unusable. One would think that a competent shelf bracket should not be hard to find at a hardware store, but I'll have to keep looking. The worst part is having to redo the shelf which was a pain. Pictures soon. Quote Link to comment Share on other sites More sharing options...
SME Posted September 5, 2016 Author Share Posted September 5, 2016 As a point of amusement: I had a friend over tonight who spent some time on the sofa listening to music that was playing. He chose a side seat, away from the MLP. He is familiar with these speakers. I gave him a demo shortly after I got the front left and right running, and I also play music when we hang out in the adjacent dining room and play games. Anyway, after listening for a while, he commented that he thought the center channel really helped tie the sound stage together and then asked me how long it took to integrate them. I responded by telling him he should go walk up real close to the center channel to listen to it closely. It wasn't until his head was inches away that he realized it wasn't playing anything. Quote Link to comment Share on other sites More sharing options...
maxmercy Posted September 5, 2016 Share Posted September 5, 2016 My center channel is done and installed on its shelf above the TV, Pictures soon. You will need a projector if you are sinking this much time and energy into the sound system. JSS Quote Link to comment Share on other sites More sharing options...
SME Posted September 7, 2016 Author Share Posted September 7, 2016 You will need a projector if you are sinking this much time and energy into the sound system. Yes I will. It has been my plan all along after the sound is done. Indeed, with the new center channel up and running, the TV seems way too small. Perhaps it's the big dynamics or the pinpoint front stage imaging falling in the wrong place, but I can't help but imagine a screen that spans the front left/right speakers that somehow got shrunk down. The sound-stage is huuge, extending both wide and deep. I have these calibrated to around 83 dBC (albeit with roll-off in the horn section). Last night, we watched Guardians of the Galaxy, which was quite comfortable to listen to at "-3". I love the dynamics in this one, and it's one of those movies that both has good sound and is fun to watch. It's something like my third time through, but it was the first time I heard every word of dialog clearly. Today I watched "Samsara" at "0", and it sounded great. A lot of the music in it has very strong lower mid content that made the cabinets of my Hsu speakers buzz like crazy. My surrounds did a bit of that, unfortunately, but the speakers I built just filled the room with rich sound. And tonight, my wife and I watched the Studio Ghibli "Howl's moving Castle". I opted to try the DVD to see if it had a more dynamic track than the BD. I knew it was a good sign when I saw a cinema preview for "Cars". This movie sounded great at "0" also. The dialog projected forward a bit but was very natural and crystal clear. I'm not Japanese, so I can't evaluate intelligibility and I refuse to listen to a English dub track. The score sounded like it was playing live on wider, more distant stage. This movie isn't really a subwoofer movie, but there were plenty of strong tactile moments in the upper bass / low-mids from the house noises in the movie and some explosions. I do think the surrounds were holding things back. A lot of people say surrounds don't matter much, and I agree that they don't matter *as much* as the front stage, but I think they can still detract quite a bit if their performance is not up there. My new speakers leave the Hsu HB-1 surrounds in the dust, despite their 92 dB/2.83V rated sensitivity. I can hear it, and I can also see the level indicators for them on my Motu A16 display shooting up above the half-way mark quite a lot too. But I'll have to wait as I planned to do all the subs first. The surrounds will likely be the most difficult build by far, so I'm saving them for last. The Acoustic Elegance TD6H drivers I plan to use are ordered at least. Speaking of subs, the 8 Acoustic Elegance TD10X for my behind-sofa mid-bass units arrived today. I'm going to hold off on building those until my 4 x 21" subs are done. Those drivers just shipped today, so I'll be working on those boxes next. Quote Link to comment Share on other sites More sharing options...
SME Posted September 15, 2016 Author Share Posted September 15, 2016 (edited) I've been working on and off on optimizing the center channel. It took a while to notice the flaws because I don't get as much opportunity to watch movies, but it was definitely too bright and pretty uneven in the low end. I've been able to get some improvement with EQ adjustments, and today I did some acoustic experiments to get insight into that side of things. As far as EQ, the high frequency section requires little attention. I had to boost output some in the 1-2 kHz region, and I recall seeing this change in response depending on the presence of a continuous baffle on one or both of the top and bottom of the horn. I pretty much overhauled the low frequency section. I had hoped that by using the same signal shaping to the center that I use to the left and right that I would achieve the same balanced decay / room energy" profile and sweet sound. But that is not what happened. The acoustic environments are quite different, so they have different profiles. My re-balance still left the bass anemic, but I would rather avoid boominess and excessive bass energy in the room and correct the acoustics instead. On the acoustical side of things, I ran some experiments to assess the effects of changing the acoustics in different ways. It helps to know that my center channel is about 9" and is placed, horn below woofer, about 1" off the back wall. The woofer is about 2 feet from the ceiling. The horn is a a bit less than 3 feet from the ceiling. The 55" flat screen TV screen is placed nearly flush with the front of the horn and right below it. Yes, the poor TV gets to act as a baffle extension. I will post a picture one of these days. The tests I did included: Treatments on the wall behind the speaker on one side. I only tested one side for convenience knowing that whatever improvements result are likely to be twice as good with both sides treated. I tested with absorption by holding a 2'x4' OC703 panel horizontally oriented and adjacent to the horn and woofer of the speaker, approximately flush to the front. I repeated this test but instead of the absorber panel, I used used a sheet of BB plywood, about 3'x4'. I tested absorption placed on the equipment racks that the TV sits on and lie directly below the center channel. They are about 4 feet below the woofer. I tested absorption held approximately in the location my geometric calculations suggest will treat the 1st reflection point on the ceiling. This required me to hold the absorber while standing on a step ladder to the side of the speaker. I tested the 2'x4' OC703 panel oriented horizontally and placed on the rear wall on top of the existing absorber panels. The rear wall is about 20" behind me, and the existing panels are 6" thick, offset from the wall by 2", and extend about 18" above the MLP. My results from these tests can be summarized in one rather ugly plot: These are all frequency response measurements at the MLP with 1/24th octave smoothing. I measured the horn and woofer separately with all crossovers in place. They are gated to 9 ms, a time I chose because it is acoustically quiet (no energy arriving at this time) and because it excludes later side-wall and secondary reflections that are not really part of the experiment. Unfortunately, the short window and 4th order 110 Hz crossover causes the bass response shown to be somewhat under-developed, but the trends are entirely consistent with results seen with longer windows, so we'll have to ignore the extra bit of droop. The measurement of experimental conditions is represented by a different colors, representing each of the tests. The tests 1-5 are respectively color-coded orange, red, cyan, green, and blue. Each test had a baseline measurement, which is plotted in gray. For the most part they overlap into a rather thick dark line. I plot all of them because the acoustic environment can and often does change as a consequence of the tests themselves. When doing baseline measurements, I try to put my body in the same posture as I will be in when hold the absorber or plywood sheet. To me, the most obvious change came with the ceiling absorber (green lines, test 4). A single 2x4 OC703 panel is enough to make a dramatic improvement in the low mid-range response. It does reduce the bass (< 200 Hz) a bit, but there are better ways to get more bass. Unfortunately, my measurement has a narrower ripple in both the horn and woofer responses that wasn't captured in the baseline. That it is narrower says it is due to energy arriving later. I'm fairly certain this is a secondary reflection involving my body and the TV as I stand on the ladder to hold the absorber. My attempt at posture matching was not very good, and I believe I pivoted more for balance when holding the absorber. As such, it's hard to see here but the lower part of the horn response (where its vertical directivity has fallen off) is also improved in the 1-2 kHz range. The addition of a panel above the existing treatments on my rear wall (blue lines, test 5) also appears to improve mid-range response somewhat but not nearly as much as the ceiling treatment. One tricky question is whether this improvement would be seen if I had a ceiling panel installed because that secondary reflection hits the back wall up a bit higher. Hmm, maybe I should wait on changing anything here. The addition of a panel on my racks (cyan lines, test 3) has minor effects on the response. My body was out of the room for this one. The changes involve mostly details, indicating effects on reflections that involve the ceiling too. Interestingly, this one effects the horn in the 1-2 kHz region. It turns out that the vertical response is quite strong 90 degrees below the horn due to the baffle extending effect of having the TV below it. FWIW, my room design does not accommodate any acoustic treatment on top of the racks, but I still want to know what effect it may have. I mention test 1 and 2 last as they involve similar experiments. Note that I did not measure the horn with the plywood baffle because that measurement was a real pain and I had no reason to believe it would make a difference. The speaker is about 18" wide, and so holds some directivity down pretty low, like 200 Hz or so. Below there, sound begins to diffract around the edges, causing a reduction in output at the seats (baffle step effect). As sound begins to diffract around the edge, a new inverted wavefront is initiated that causes interference and low frequency loss (baffle step effect). The sound that diffracts around the edge then reflects back causing additional interference but restoring the bass below the point at which the wave fronts are too close together to interfere. This creates a suck-out in the upper bass that complements the low mid-range suck-out due to the ceiling. Using absorption kills the front wall reflection but not the baffle step effect, at least until the frequency is too low to be absorbed. The plywood baffle, on the other hand, eliminates both the diffraction and reflection leading to an even stronger bass response. Both treatment methods improve the bass directivity, which leads to tighter, punchier bass and less room boom, but the plywood baffle provides more headroom. From these results, I am planning to treat the ceiling ASAP. I know I will want to do something to the front wall, adjacent to the speaker, but I haven't decided on what combination of baffling, absorption, and diffusion to use. That region is also a 1st reflection point for the surrounds, so I may want to opt for absorption or diffusion or a hybrid combination, but with diffusion, I may actually get most of the bass back because the average depth of the diffuser would be about half as much. The problem definitely deserves more detailed study, and I'm confident I will find a solution that provides the bass I want. The addition of rear wall absorption appears a bit more dubious at this point. I need to treat the ceiling and then see what happens. I'd love to be able to make that surface diffusive, especially if there's a lot of diffusion at the front. The best part about all this is that my measurements are quite consistent (at least qualitatively) with what I predicted when I did this design. So far, everything appears to be on track. Edited September 16, 2016 by SME Quote Link to comment Share on other sites More sharing options...
SME Posted September 20, 2016 Author Share Posted September 20, 2016 I added some more pictures to the first post, one of the front of the room with the new center channel installed and two more of the rear of the room to illustrate what I'm working with here. I really love the look of the center channel right up above the TV. It almost looks like it's built-in. Wouldn't it be cool if TVs came with built-in speakers like that? Quote Link to comment Share on other sites More sharing options...
3ll3d00d Posted September 20, 2016 Share Posted September 20, 2016 Anyway, from what I've heard, it sounds very good, but the one thing that bothers me is that the dialog seems to originate more distinctly from above than I'd like. That surprises me because the horn is only a few degrees higher than the horns on the left and right are. Perhaps elevation is easier to hear after it passes some threshold? have you considered that this is what your eyes are telling you? I don't recall the details off hand but IIRC the chapter on localisation in https://auditoryneuroscience.com/topics/book-previewwent into some detail on how the visual system influences this. Quote Link to comment Share on other sites More sharing options...
3ll3d00d Posted September 20, 2016 Share Posted September 20, 2016 btw I have just got my speakers up and running, atm I'm running them flat as I haven't had time to do any sort of in room measurements. I was slightly surprised to find that I don't find them forward/bright at all. What this says about the room and/or my hearing is TBD Quote Link to comment Share on other sites More sharing options...
SME Posted September 21, 2016 Author Share Posted September 21, 2016 have you considered that this is what your eyes are telling you? I don't recall the details off hand but IIRC the chapter on localisation in https://auditoryneuroscience.com/topics/book-previewwent into some detail on how the visual system influences this. Yes! I think you are right here. I've noticed also that the images seem to originate from a higher point when listening in stereo too. Just seeing that center up there makes my brain go "sound from up there". It does seem a lot better since I adjusted the EQ, bringing the high frequencies down a bit more, or maybe I'm just more used to it. btw I have just got my speakers up and running, atm I'm running them flat as I haven't had time to do any sort of in room measurements. I was slightly surprised to find that I don't find them forward/bright at all. What this says about the room and/or my hearing is TBD When you say flat, do you mean flat anechoic, in half-space, or in-room at the MLP? Anyway, we'll see if your opinions change with more time and more listening. I started out with mine pretty close to flat in-room, and I was quite happy with what I heard at first. The speakers are so much better than what they replaced that they sounded great to me, even though in retrospect they were tonally biased way too bright. Indeed, the target curve I used with my old speakers was unbalanced on the bright side as well, but it was hard to hear what was wrong because of the other problems with the speakers. I also lacked a good method for iterating on the shape of the curve. It's much harder to assess whether a change is beneficial without being able to do quick A/B tests. If you go look at the first post, I updated my crossovers/response curve there to what I last measured. You'll note right away that the center looks much worse than the other two. See a few posts above for my investigation of the acoustic causes of that. The weird thing about my current responses is that it kind of looks like the horn, the woofer, and subs are all fairly flat but are running at different levels. Yet, what you see sounds very balanced to me, and I believe most others would agree if they heard it. I arrived at my results for the left and right after a lot of time spent listening and iterating on the shape. Well-recorded classical music provides some of the best test material, and at this point, good classical recordings sound truer to life on my system than I have ever heard on *any* speaker system in my memory. I can only imagine how much better it will sound when I get the acoustic problems fixed and re-do it all with optimized FIR filters instead of the rather crude biquad filters I'm using now. And yeah, I was kind of surprised to see the treble looking quite flat from 1.5-7 kHz after adjusting by ear, but I probably shouldn't be. I expect the measured woofer response will change a lot once the left and right speakers are installed to their final locations, and I've treated the major acoustic problems. It'll be interesting to see what kind of signal shape changes I need to make for them at that point. Quote Link to comment Share on other sites More sharing options...
3ll3d00d Posted September 21, 2016 Share Posted September 21, 2016 I mean anechoic flat, there is the usual mess caused in room though I haven't had time to measure except to quickly verify levels. I thought I'd live with it as is for a bit and then get into EQ'ing to give me a better handle on how it actually sounds in itself. What are the diffusers made out of btw? also why place some above the speaker, I didn't catch what the rationale was there. Quote Link to comment Share on other sites More sharing options...
SME Posted September 21, 2016 Author Share Posted September 21, 2016 I mean anechoic flat, there is the usual mess caused in room though I haven't had time to measure except to quickly verify levels. I thought I'd live with it as is for a bit and then get into EQ'ing to give me a better handle on how it actually sounds in itself. What are the diffusers made out of btw? also why place some above the speaker, I didn't catch what the rationale was there. So then I gather your crossover uses some kind of baffle step compensation? Anyway, take your time. One thing to keep in mind is that imbalances usually become more pronounced with higher playback levels. A bit of excess brightness might sound quite nice at a low playback level, but then your ears complain when you turn it up loud. This is often even more of an issue if the album uses a lot of compression, and of course, clipping is a lot harsher sounding on a brighter system. The diffusers are made of polystyrene, which makes them relatively cheap but not very robust. The pair of diffusers on the front wall are there to spread out some of the reflection coming from the back wall. It made a significant positive difference with my old speakers, which were nowhere near as clean in off-axis response. I'm not sure they are as necessary with the new ones. The only reason I put them above the speakers is to get them out of the way of the direct sound. Originally, they were attached to the top half of the absorbers that are adjacent to the left and right speakers, but in that location, they definitely do interfere (in a bad way) with the direct sound from the horns. With them on top of the horns, they don't measurably interfere. In the long term, I'll probably construct diffusers out of wood for all the walls and move the polystyrene ones to the ceiling. Quote Link to comment Share on other sites More sharing options...
3ll3d00d Posted September 21, 2016 Share Posted September 21, 2016 no baffle step compensation as the woofer is loaded by the floor so I knew that that would add something but I found it hard to get a reliable reading on exactly what it would add. My best guess is that it would add more than baffle step under 150-200Hz hence I decided to ignore it and rely on EQ to compensate instead. I know what you mean re letting it sink in over time. Personally I find back to back A/B a bit of a fruitless exercise, I find that something either sounds wrong pretty immediately or an itch develops over the course of a few weeks. If either happens then it's a signal to drill into what is going on, if not then I leave things alone. This tends to lead to bursts of activity followed by long periods of stasis (with respect to the configuration), last time was after the subs changed ... took ~3months to dial in them in but then that configuration was then left unaltered til now (~18months later). Quote Link to comment Share on other sites More sharing options...
SME Posted September 22, 2016 Author Share Posted September 22, 2016 no baffle step compensation as the woofer is loaded by the floor so I knew that that would add something but I found it hard to get a reliable reading on exactly what it would add. My best guess is that it would add more than baffle step under 150-200Hz hence I decided to ignore it and rely on EQ to compensate instead. I know what you mean re letting it sink in over time. Personally I find back to back A/B a bit of a fruitless exercise, I find that something either sounds wrong pretty immediately or an itch develops over the course of a few weeks. If either happens then it's a signal to drill into what is going on, if not then I leave things alone. This tends to lead to bursts of activity followed by long periods of stasis (with respect to the configuration), last time was after the subs changed ... took ~3months to dial in them in but then that configuration was then left unaltered til now (~18months later). Now I remember you saying that. if you have EQ capability, that is the way to go for dealing with low mids and bass because these frequencies are so dependent on room and placement. I'm pretty sure I won't be touching my front left and right for a while. Things seemed to just click after the last 1/4 dB change I made. I don't think they are perfect yet, but they sound better than ever before. I think I might leave the EQ this way until I raise and invert those speakers to be the same height and orientation as the center. And I can't do that until the new subwoofers are done because they and their platforms will be stacked on top of the subs. I've been reading the JBL M2 thread on AVSForum and relating a lot to the comments about the listening experience with those speakers. I wish someone measured polars on those because I'd be interested to know how the pattern compares to the SEOS, apart from the 120x120 degree in the M2 horn vs. 90 x 40ish in the SEOS. Either way, the SEOS-based designs and M2 clearly occupy an elite tier of speakers by virtue of their smooth off-axis response coupled with dynamic capability. IMO, there's not really a "best" speaker among these elite few, because application specific considerations apply. I do believe the M2 design excels because of its optimized DSP configuration and that a similarly optimized SEOS based design could compete with or even beat the M2 in many applications. I know in my room, I have more than enough high frequency energy with a 90x40 horn, so I suspect M2s would be too bright and would need to be EQed down even more. And anyway, they wouldn't physically fit if I wanted a retractable big screen. Still, I'd love to hear M2s some day in a room that is appropriate for them to compare what I have achieved with what appears to be the most well-regarded (by those whose opinions mean anything to me) commercial speakers available Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.