RESULTS OF MY ACOUSTICAL ROOM TREATMENT Introduction: Until recently I had not given consideration to acoustical room treatment, as I was naively under the impression that loudspeakers had by far the greatest influence on a system’s sound. My opinion on this has been emphatically changed, as I have found myself on a very steep learning curve the past few weeks! I therefore thought it would be helpful to share my results in case others are considering pursuing a similar route with regards to room treatment. Motivation: When I recently removed the dominating wardrobe and stack of unused loudspeakers from my room in an attempt to create a more symmetrical and spacious listening environment, the quality of the acoustics took an absolute nose-dive. Obviously the excess furniture clutter was doing an effective job at keeping the sound relatively tight and well balanced, as after its removal the room became unbearably boomy, echoey and smeary. I was absolutely devastated by the change for the worse, but was determined not to revert to the former layout. After much online research I contacted GIK Acoustics for advice on how to treat my room with acoustical products to restore its previous sonic character, and after an extremely helpful consultation I placed an order for a few of their absorption panels. Regretfully I did not take any in-room acoustical measurements prior to removing all of the excess furniture clutter, however I have documented each stage since, from the room being completely absent of treatments through to its current treated state. The Room: The listening room is a standard cuboid-shaped space of dimensions 13’7” long by 12’5” wide by 10’8” high. It features a single bay window behind the listening position (but offset to one side). All four walls are parallel (or as parallel as to be expected from a Victorian build!), and are plaster over stone construction. It has a suspended wooden floor that has been sheeted with plywood and carpeted with underlay, and a flat ceiling of traditional lathe and plaster construction. The system is located centrally on the long wall (opposite the window), and thus plays across the shorter room dimension. The loudspeakers are positioned 5’4” apart, 4’1” from the side walls, and 2” from the front wall (all measurements taken from front edge of loudspeaker cabinet and on tweeter axis). The loudspeakers are mildly toed-in and are a parallel distance of 7’5” from the listening spot (7’10” if measured diagonally). The listening spot is central and 3’ from the rear wall. The System: Source: Digital lossless library of mixed resolutions (16/44 through 24/192). Transport: Mac Mini using Audirvana/iTunes. DAC: Schiit Bifrost asynchronous 32/192 USB. Amplification: Yamaha CR-1000 Natural Sound Stereo Receiver. Loudspeakers: Tannoy Lancaster Monitor Gold 15” LSU/HF/15/8 (w/upgraded crossovers, wiring and cabinets). Cabling: Digital Interconnect (Basic 1m Belkin USB 2.0 cable); Analogue Interconnects (1m Van Damme XKE instrument cable with NYS373 REAN phono plugs); Loudspeaker Cable (Van Damme Studio Blue 2x2.5mm). The Treatments: I have only included some of the purchased treatments in the results and analyses that follow; namely the GIK 244 Broadband Bass Trap (a 2’ x 4’ x 4” panel typically used to treat the first reflection points), and the GIK Tri-Trap (a 2’ x 4’ x 16.5” wedge bass trap used to treat the room corners). I also purchased the GIK Monster Bass Trap (a 2’ x 4’ x 6.5” broadband panel typically used to treat the lowest frequencies) to specifically address my room’s peaks and nulls at the lower end of the scale, however I have yet to find a location in which this product reliably alleviates these issues. I have also still to install EPS diffusor panels, thus the room is by no means the finished article yet. The Measuring Equipment: Mic: Behringer ECM8000. Mic Pre-amp & ADC (analog-to-digital converter): Apogee Duet FireWire 24/96. Software: Room EQ Wizard. The Results: Key for all graphs: Red = untreated room. Blue = four 244 panels at first reflection points (two per side wall). Green = four 244 panels at first reflection points (two per side wall) and four Tri-Traps (one per corner). Yellow = four 244 panels at first reflection points (two per side wall) and eight Tri-Traps (two per corner). (N.B. I have many more results recorded as I repeated the tests by varying the listening position forwards and backwards a few inches from my usual spot. It would however be superfluous to upload all of the results, and in fact the general picture does not change significantly, so for the purposes of this write-up I am focusing on a single listening position, 36” from the rear wall.) Analyses & Discussion: The objective is to obtain as uniformly flat a frequency response and RT60/waterfall as possible, and to reduce the RT60/waterfall to an acceptable target reverberation time, given the inherent practical limitations of the room and the system. Initially I was disappointed to find that the frequency response (Figs. 1A, 1B & 1C) had not flattened out as much as I hoped it would post-treatment, but perhaps I had unrealistic expectations. Of course, it depends on how much “smoothing filter” you apply to the graphs. With 1/48th octave smoothing the post-treatment response still looks horrendous, but with 1/3rd octave smoothing (which is typically what loudspeaker manufacturers quote in their anechoic tests) the response doesn’t look too bad. I am not sure which smoothing filter correlates most closely with what we actually hear, so I have included 1/48th, 1/12th and 1/3rd graphs to give as complete a picture as possible (although I think 1/12th strikes the best compromise between detail and trend). Most evident from the frequency response overlays is the positive effect installing the 244 panels at the first reflection points on the side walls had. There was a significant gain in output of mid and especially high frequencies (an average boost of around 6dB), bringing their respective SPLs up closer to those of the low frequencies and thereby creating a flatter response. It is interesting that the installation of absorption treatments had the effect of boosting upper MF and HF output, as I would have intuitively expected the opposite result. This suggests that there must have been significant first reflection cancellation occurring across these frequencies in the untreated room. There was an audible improvement in focus, detail and clarity following installation of the first reflection panels, plus a noticeable difference in tonal character. However the actual audibility of the changes when listening to music was not as significant as the frequency response overlays suggest. I can only assume that this is because there were still long reverberation times (as shown in Figs. 2 and 3B) favouring the upper bass frequencies. As is evident from these two graphs, installation of the first reflection treatments had little impact in neither reducing nor flattening out the room’s ringing below 1kHz, and this mirrored what I heard. However, the subsequent installation of a quad set of Tri-Traps (one installed in each corner of the room) did have a significant impact in reducing ringing, and the effect this time was both measurable and audible. As shown in Fig. 2, reverb times reduced on average by 1/3rd to around 500ms, and more importantly flattened out considerably across the audible range above 100Hz (excluding the remaining peak at ~300Hz). [N.B. The observant will notice from the waterfall graphs (Figs. 3A, 3B, 3C & 3D) that there hasn’t been as much improvement below 90Hz, and very little improvement below 50Hz. The dimensions of my room give rise to a fundamental mode of ~42Hz, and frequencies this low are notoriously difficult to treat unless heavy artillery is thrown at them. I have been advised that I would either require Tri-Traps 3’ in width instead of the standard 2’, or Soffit traps (cuboids of size 4’ x 16.5” x 16.5”), and unfortunately due to layout restrictions my room cannot accommodate such massive structures!] This time there was a more noticeable audible improvement in focus, detail, clarity and imaging accuracy following installation of the four Tri-Traps. The sound was tighter and snappier, and dynamics had improved too. The difference in the system’s tonal presentation was also more evident, presumably because the now shorter decay times in the upper bass frequencies were no longer masking the attack of the higher frequencies. Ignoring the potential for anecdotal bias, this was probably the closest my system/room sounded to the way it did before it was cleared of the excess furniture, but now the sound had improved focus. The final installation stage was to add another four Tri-Traps (stacking them atop the existing ones), taking the total to eight. This has in my opinion had the biggest audible impact in changing both the tonal presentation of the system and ambience of the room, and by much more than the graphs suggest. In fact, before even playing any music the whole “feel” of the room changed profoundly the instant the final trap went up; it was quite eerie actually! As shown in Fig. 2, reverb times reduced on average by a further 20% to around 400ms. The room sounds much “deader” after this last stage of treatment and, as a result, nuances and cues in favourite tracks that I had never noticed before have now been unmasked for the first time. While this is most definitely a welcome positive, other aspects of the change in sound required more time to accustom to and ultimately appreciate. Initially it seemed that this latest stage of treatment had robbed the music of its ambience a little too much, making it sound too focused and sterile and not as spatially enveloping and emotionally involving. However having now lived with this sound for a few weeks, it has become the “norm” to my ears, and listening to music in less-damped listening rooms now sounds sub-optimal. Tonally, the system sounds markedly more different now than at any stage before. In fact if I was blindfolded I would swear that I was listening to a completely different setup; that is how pronounced the change has been. It now sounds significantly crisper in the upper frequencies, and the previously soothing Tannoy lower mid-range “bloom” seems to have receded and been replaced with a thinner, cleaner, and more piercing upper-MF presence. The closest tonal presentation I can liken it to is that of an outdoor reproduction system that is (presumably) devoid of the usual colourations caused by an indoor listening space. It is indeed as if a blanket of colouration has been lifted from the system, such has been the gain in transparency and clarity. Another major benefactor of the treatment has been the system’s dynamics, which are now simply to die for! I have literally been startled on numerous occasions by the clarity, attack and scale of the crescendoes in favourite tracks that I did not realise had such great SPL range. Last but not least, I have noticed a significant lowering of the system’s/room’s noise floor post-treatment, which makes it possible to listen at lower SPLs than before whilst retaining impressive amounts of resolution. Conclusions: The above results underline the importance of a two-pronged approach in the use of absorption products in acoustical room treatment; treatment of early reflection points to restore a flatter in-room frequency response at the listening position, and treatment of room corners to reduce and smooth out reverberation times. The results show that - in my situation at least - both phenomena greatly influence the ability to hear a system’s true resolving capabilities and thus obtain accurate sound reproduction. It is clear from my findings that the facility to control these significantly enriches the listening experience. Further Thoughts: It is evident - especially from Fig. 1A - that bass nulls still persist post-treatment (e.g. at ~100Hz). Going forward, this is something that will require further investigation and treatment. More generally, it will be interesting to listen to - and measure the effects of - adding diffusion treatments to the room. Finally, it would be good to discover what the optimum or recommended reverberation time actually is for a domestic listening environment for 2-channel HiFi stereo, as this would aid in determining whether the current amount of treatment in my room is adequate. I have searched many online threads but cannot actually find a specific target value, only the recommendation that a room dedicated to Music should possess a longer reverb time than one dedicated to Home Theatre. Speaking of which - and on a final positive note - dialogue through my Plasma TV’s built-in speakers has never sounded better than it does now. It really is a novelty and joy being able to watch telly and no longer need to strain to decipher muddy, unintelligible speech!... Feedback: Your thoughts on any of the above are most welcome.