damonrpayne Posted July 20, 2005 Share Posted July 20, 2005 Just messing around on the web, I found quite a bit of software that can generate test tones and save them as a .WAV file, which I could then write to CD and play through an audio system. I assume one of the goals of proper room treatment is to have a room that is flat from 20hz-20khz. Since I have found software that can generate test tones on my PC, I'm sure I can find something that can record the sound from a microphone placed at a listening position and display the frequency of the sounds recorded for the duration of the test. Given these two things graphed together I could determine where the bumps and flats were, and have a very scientific way of tuning a room. The only possible flaws I see in this method are 1) It assumes my equipment is perfectly flat 2) I may need expensive microphone/software to do a good job? Can anyone recomend any particular software or equipment? Quote Link to comment Share on other sites More sharing options...
J.4knee Posted July 20, 2005 Share Posted July 20, 2005 Yes ideally you would need a stable consistent tone reproduction by your source equipment, a quality mic and RTA. You could get a pseudo-acceptable response using a SPL meter and recording your readings but you'd be there for quite a while unless you stepped your test frequencies and that would only offer a crude examination of your environment plus you'd have to take into account the limitations of you meter. I would expect an RTA with a pink or white noise generator would be easier than generating tones across the spectrum. Quote Link to comment Share on other sites More sharing options...
dragonfyr Posted July 20, 2005 Share Posted July 20, 2005 This is aimed at all the recent threads aimed at tuning rooms! I applaud the fact that many are recognizing the role of the room in the performance of the system! This is terrific! BUT...there needs to be a better understanding of just what is involved and how to proceed. And unfortunately, for all the recommendations being made, most are so non-specific as to merely have lots of folks spending alot of time, effort and money to simply 'change' things! And unfortunately, that is as complimentary as I can be. I would like to see more REAL improvements made! So, to answer the question above regarding what you will have after all the effort and money and time, aside from additional character,is that the net result will be of little use. Unfortunately, for all of your time and effort you will NOT have useful information from which to tune a room! The fundamental information that is necessary does not lie in the frequency domain! Unfortunately, you will have a very nice graph of the composite response of the super-positioned signals. But you will have NO clue as to that the component parts ARE!!! Moreover, an RTA and EQ or other tools cannot fix this complex resultant waveform. How many times will we have to go around on this!? Rather you must be able to identify and analyze each of the component signals that combine to give you the waveform you will have spent so much time and effort to measure. And these are measured in the time domain. Welcome to the analytical tools of MLS (Maximum Length Sequence analysis) or TDS (Time Domain Spectroscopy) and the real world measurement tools of MLSSA or TEF. In order to rectify and 'tune' the room, the individual signals are 'manipulated'. And now we enter into the realm of time domain analysis to identify just how these signals are to be dealt with. And simply trying to absorb them or diffuse is not the answer!!! (Besides, how many of you realize that you only want to treat a portion of the frequency band of each reflected signal? - and how many know the frequency bandwidth that the particular absorptive or diffusive material you are using effects, and how? Do you know if those frequencies are even the problem frequencies or if those particular reflections are a problem? And how do you determine the proper balance of all the various treatments? Gee. One almost wonders why some bother to invest in the proper tools to analyze this !? And I already had an expensive RTA! Sarcasm? Heck No. At this point its exasperation. ;-) And for all of you who will complain that a 25 word solution was not proposed, it was! Use the proper tools and know the proper way to address the acoustics within the time domain. Unfortunately, it doesnt come in a kit or a box! I can just see this general room tuning approach applied to crossover design! I can just see ALK saying, well the original crossover had 3 capacitors, so I will improve it by adding 2 more! And make them polypropylene! Yeah, thats the key! Poly! None of those Teflon jobs! And values? Hey, Who needs stinkin values!? After all, this one is bigger then the original! And it will sound different! Think thats absurd!? Well, welcome to the world of home brew room tuning! As that is what is being done here and in other threads with arbitrary application of absorption and diffusion based on emotion and trying to eliminate first order reflections with a mirror and materials that you do not even know are effective for the frequencies contributing to a problem. despite A LOT of work and effort!! You know, you can save a lot of work and achieve the EXACT same net effect literally!, but simply wrapping your head with some cotton bating and holding it in place with an ACE bandage! Laugh, but this is LITERALLY correct! Just doing something because you understand that something should be done is not enough! To be effective you must do the CORRECT thing! And I am not trying to belittle genuine efforts! But we have addressed the fundamental problem with RTA and EQ solutions ad nauseum! You cannot EQ non-minimum phase signals! You cannot correct for the effect of reflected out of phase signals with an EQ! And I will make one other comment. In the other room acoustics thread a suggestion was made that if you were going to spend >=$1500 to have your room tuned, that you should call a professional!!!! Indeed!! If you are going to spend that to have it analyzed, PLEASE call me! I will TEF your room and completely analyze it with both TDS and MLS!!! Come on folks!! Where do you guys get these prices???? Has anyone actually investigated this!? $1500!!! Perhaps, if the guy has to fly in from <?xml:namespace prefix = st1 ns = "urn:schemas-microsoft-com:office:smarttags" />Sydney, Australia! Quote Link to comment Share on other sites More sharing options...
DrWho Posted July 20, 2005 Share Posted July 20, 2005 ---------------- On 7/20/2005 3:56:59 PM dragonfyr wrote: ...threads with arbitrary application of absorption and diffusion based on emotion and trying to eliminate first order reflections with a mirror and materials that you do not even know are effective for the frequencies contributing to a problem... And I will make one other comment. In the other room acoustics thread a suggestion was made that if you were going to spend >=$1500 to have your room tuned, that you should call a professional!!!! Indeed!! If you are going to spend that to have it analyzed, PLEASE call me! I will TEF your room and completely analyze it with both TDS and MLS!!! Come on folks!! Where do you guys get these prices???? Has anyone actually investigated this!? $1500!!! ---------------- Ummm, thanks for indirectly addressing the person that wrote that comment...and then even misinterpreting it! If you only have $200 to spend on treatment, then you certainly can't justifity spending $600 to have some dope professional come in and tell you what you need to do. We all have budgets and I'm certainly not going to waste the majority of my budget just to find out what I need to do...not to mention there will be a significant level of doubt in what the professional has to say anyway (issues with taking the data, interpreting data, finding the proper products, implementing them correctly, etc etc....) Btw, the mirror trick does work and you even said yourself that the time based measurements support similar conclusions. To be honest, I don't fricken care what method one uses to do anything, but all that matters is the end result. And you certainly aren't going to tell me that I don't hear an improvement (and I don't even care if it's technically correct as long as it sounds good). Your analogies really suck, but if randomly adding 2 more capacitors made it sound better, then by all means add them! All that said, how bout you go build a room using all your fancy toys and then I'll build another room using "emotional" techniques and then we can compare how different they sound. ANYWAYS, none of this is relevant to this thread and I really don't welcome these indirect slanderings being carried out into other threads. so back to this thread... _________________________________________________ ---------------- On 7/20/2005 3:01:23 PM damonrpayne wrote: Just messing around on the web, I found quite a bit of software that can generate test tones and save them as a .WAV file, which I could then write to CD and play through an audio system. I assume one of the goals of proper room treatment is to have a room that is flat from 20hz-20khz. Since I have found software that can generate test tones on my PC, I'm sure I can find something that can record the sound from a microphone placed at a listening position and display the frequency of the sounds recorded for the duration of the test. Given these two things graphed together I could determine where the bumps and flats were, and have a very scientific way of tuning a room. The only possible flaws I see in this method are 1) It assumes my equipment is perfectly flat 2) I may need expensive microphone/software to do a good job? Can anyone recomend any particular software or equipment? ---------------- The problem with using solid tones and a mic of any kind is that you have to deal with standing waves. It's actually quite fun to do if you ever have the chance...when listening to a solid tone, moving your head even an inch could mean the difference between ear piercing loud and silence. It actually happens at evey frequency, just at different spots in the room for different frequencies. This alone makes the test tone method ineffective. Another approach is to bust out an RTA and use pink noise (you can get a behringer ECM8000, a mixer and free RTA software for a total of $100). Pink noise is random noise so you will no longer have issues introduced by standing waves. The downside to an RTA is that you can't distinguish between the direct and reflected signal, which may or may not be of concern. We pretty much know that all our electronics are for the most part flat and we can even look up the response for our speakers...or even better use the RTA to test it ourselves in a boundary free environment and then cancel out any flaws with the measuring equipment. Compare this against your in room response and you start to see the effects caused by your room. The only thing lacking with this approach is the ability to measure how the room is affecting the sound over time. The room is a complex system of all sorts of reflections and an RTA isn't going to be able to tell you the delays which would then allow you to go back and figure out where the problems are... This is where the TEF comes into play. One thing to note though is that the TEF must take longer samples as the frequency gets lower, which means at the lower frequencies it's coming closer to providing the same information as your RTA (not a big deal though considering the wavelengths are huge too). I would say the difference between using an RTA and a TEF is more akin to a loudspeaker designer using published specs instead of measuring the actual driver that got shipped to his door and then choosing not to account for differences of cabinet volume due to loudspeaker and port displacement (aka, minor 1dB differences in performance). It's been my somewhat limited experience that the differences between using a TEF and RTA to fine tune have resulted in extremely similar outcomes. The key to good measurements is knowing how to interpret the results and to trust your ears. I like to think of measuring as something we do to attach a number to what we hear...so if your measurements (pick any of the methods) indicate that you should change something that you don't hear, then don't change it. Quote Link to comment Share on other sites More sharing options...
PrestonTom Posted July 20, 2005 Share Posted July 20, 2005 Good Luck with an interesting project. As you point out, you are assuming that the components are spectrally flat. This is probably not a bad assumption, except for the speakers (and how they are set up in the room). The probelm is a tricky one (that is why I think it is an interesting one). Although I don't agree with the tone of voice, DRAGONFYR is quite correct in anticipating some of the problems - and I whole-heartedly agree that it much more than an issue of getting an EQ unit in the circuit. Dr Who has a more optimistic approach and I appreciate him taking the time to share his knowledge, since I enjoy learnining about this also. The basic issues are that in a frequency domain perspective, temporal aspects of the reproduction are being ignored (by necessity and by assumption). In this regard a direct sound & its reflection are not distiguished but rather they are simply summed and the result is standing wave and comb filtering. Simply trying to equalize your way out of this is not a good solution. To a certain extent, attentuating the reflections or making the direct sound more "directional" might help but it is incomplete. The time domain solutions are an alternative. In this case, a primary sound and its reflection are not highlighted as a standing wave or comb filtering, but rather as an echo and reverberation (I am using a crude approximation). The solutions may be similar (non-reflective surfaces, traps, placement of chairs and speakers etc), but as was pointed out the key is determining whether an anomaly in the output of the RTA, MLS etc is a big deal or not. That is, which "dip" needs to be fixed. As I understand things, that relation is not well-understood so a bit of experimentation will be required (not all reflections are deleterious to "good sound"). The other difficulty with the time domain approaches is that they are much more recent inventions (compared to Fourier which has been around for a long time). I think less intuition and common tricks of the trade have been developed. Good luck and let us know how it goes, -Tom Quote Link to comment Share on other sites More sharing options...
dragonfyr Posted July 20, 2005 Share Posted July 20, 2005 ---------------- On 7/20/2005 4:44:12 PM DrWho wrote: One thing to note though is that the TEF must take longer samples as the frequency gets lower, which means at the lower frequencies it's coming closer to providing the same information as your RTA (not a big deal though considering the wavelengths are huge too). I would say the difference between using an RTA and a TEF is more akin to a loudspeaker designer using published specs instead of measuring the actual driver that got shipped to his door and then choosing not to account for differences of cabinet volume due to loudspeaker and port displacement (aka, minor 1dB differences in performance). It's been my somewhat limited experience that the differences between using a TEF and RTA to fine tune have resulted in extremely similar outcomes. ---------------- Nope, they don't! The TEF, using tracking filters, is completely immune to such interference regardless of the wavelength. And spin it HOWEVER you want! the RTA cannot provide information regarding the individual component signals in the time domain as TDS or MLS analysis does. PERIOD! And yes, I was aware of who made the suggestions, but as I was not interested in trying to deal with egos and only wanted to deal with some of the suggestions as representative of many of the oft mentioned ideas I refrained from mentioning names. So much for trying to deal with the ideas themselves! And the refernce to the relationships between signals not being well understood..fortunately they are very well understood! But unfortunately instead of exploring these new models and using the proper tools, we persist in turning the same flawed method on its side and assuming that now it will work, or now upside down, or if we squint and hold our head thus... Folks, my frustration is that despite some genuine interest, we do not progress to the newer time based paradigm and the tools and systems theory that has emerged. We simply keep trying to hammer with the wrench and debating how best to hold the wrench rather then adopting the proper tools and practicees!! All the RTA in this case tells you is that comb filtering is present. Duh! Isnt this a foregone conclusion!? There, I saved you the money and effort. And simply treating the first or second order reflections as linear systems is incorrect. Heck, if you want to do this, simply get a CAD program and run ray trace functions from point to point! Hmmm? I wonder if anyone has ever tried that!? And I wonder why that isn't widely used!? We can easily determine the reflective paths. Further analysis of this limitation is precisely why simply trying to use the shotgun approach, be it with a mirror or a CAD program, is not widely employed in serious room tuning. Some emerging real tools will allow folks to actually resolve some of the issues that continue to be rehashed repeatedly. But the RTA is not and will not be the tool to do it! Use it to roughly identify the location of room mode nodes...but don't try to use it for that which is not designed. And I don't care how good you can hammer with a wrench! But I suspect we will continue to see it! And as with cables, if you spend the money and exhaust the effort, I suspect that you will hear a difference. Unfortunately your RTA will be as useless in determining the proof of performance of your modification as it was in identifying them! Gee, if only tuning the room were as important as those fancy cables so many have! Moreover, if one does not have the resources or the knowledge to address the issue, spending the equivalent of a cheap CD player or turntable ($400-600 for a named pro!! and even that is NOT necessary!) (And certainly NOT $1500 folks!) in a world where folks routinely spend orders of magnitude more for their cheapest component is a worthwhile expense for something that will easily make the 2nd largest contribution to the quality of the playback of your system next to the speakers themselves. <?xml:namespace prefix = o ns = "urn:schemas-microsoft-com:office:office" /> And DOCTOR who?, Since you are the one to press this, I HAVE tuned some major rooms and studios! Some of us are not new to the time based paradigm and have been actively involved with it's development for some time! Quote Link to comment Share on other sites More sharing options...
damonrpayne Posted July 20, 2005 Author Share Posted July 20, 2005 OK, I need to re-read the responses and digest them, but a couple of comments: 1) I would personally like to avoid using an EQ of any kind. First of all it has the potential to introduce more noise into the system. Second of all, I just don't want to, the room is the problem and I'm just researching room treatment strategy. 2) I should have been more clear about my goal: The Scientific Method. This is why I mentioned frequency sweeps as my first idea for recording the room. For example: -Frequency sweeps indicate a 7db bump @ 9000 hertz at my listening position -Look into room treatments that are effective in this range -Apply room treatments, re-do freq sweep -Lather, rinse, repeat Quote Link to comment Share on other sites More sharing options...
dragonfyr Posted July 20, 2005 Share Posted July 20, 2005 As I have alluded, there are a number of rather amazing techniques that have the potential to greatly aid the folks who do not have the resources nor the desire to acquire a TEF or MLSSA and then learn how to use it. Nevertheless, I will mention that these techniques are actively being developed and refined as I speak, and therefore I will try to present more regarding them as they become more mature in nature. However, 'til then, may I suggest checking out the ETF5 software from Acoustisoft. Now this will still require the proper interpretation - but I would be glad to try to assist anyone who is brave enough to work with this. It does offer reasonable MLS sampling as well as some other useful tools. And it presents a good selection of tools at a reasonable price. <?xml:namespace prefix = o ns = "urn:schemas-microsoft-com:office:office" /> Quote Link to comment Share on other sites More sharing options...
colterphoto1 Posted July 20, 2005 Share Posted July 20, 2005 Thanks for your very insightful and readable post DrWho! I always enjoy your views on audio engineering. Those college courses are paying off. Michael 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.
Note: Your post will require moderator approval before it will be visible.