Originally Posted by kbarnes701
Originally Posted by sdurani
Originally Posted by kbarnes701
It's all very interesting but the purpose of this thread, as stated by the Thread Starter in the first post or two, is to help newcomers to REW to get it up and running and then provide them with basic advice on interpretation of their graphs, with more advice on how to improve their room, using REW as their guide.
The purpose of this thread, as stated by the Thread Starter, has always been two parts:
Originally Posted by jevansoh
The first step was introducing folks to a new and easier way to take measurements and getting everyone on the same page with how to display and interpret the measurements themselves.
I believe we've accomplished that.
The second part has always been Choosing an acoustical model. Folks need to start doing this.
With that goal in mind, discussions about room reflections are on topic (even necessary) in a thread where folks need to choose an acoustical model. So cut 'em a little slack maybe?
Agreed about the objectives. But endless back-and-forth bickering between various 'experts' isn’t very helpful, IMO, to the people who really need this thread (like me for example) in order to get a grip on the basics. AFAICS there has been almost zero information posted on how to choose the acoustic model, but a lot of discussion about which expert is 'right' and which is 'wrong'. This is, of course, seen through the prism of my own level of expertise. Your expertise is far greater than mine and so, looking at the posts through your prism, the discussions probably make all sort of sense. But the stated aim of the thread is to help people who are just starting - the various experts in the thread provide most value, IMO, if they try to help newcomers rather than engaging in esoteric discussions among themselves. There are plenty of threads for expert discussion and disagreement already. Just how I see it.
EDIT: cases in point from the last page or two:
Originally Posted by jkasanic
I'm pretty sure I have a lot of work to do wrt these reflections but I'm stuck on this "pick a room model"
or bust mentality. How should I know which model will fit my preference if I've never experienced either one?!
Clearly, I've done enough research now to be considered dangerous and I comprehend the basic philosophical differences between the various models but this doesn't mean I understand
how picking one model over another will sound in my room. Given the cost involved, picking the wrong model could potentially have a significant impact financially. I'd prefer to take some of the guesswork out of it so I'll start by treating my room for the high gain early reflections which by most accounts everyone agrees are bad but once that's complete, where do I go?!
Originally Posted by AustinJerry
Regarding picking a model, and after hearing the discussions on this thread, I have come to the conclusion that the LEDE/RFZ model is the one to work towards. I think we should have two objectives: taming the modal response as measured by the waterfall, and reducing early reflections (<20ms) to -15db or better, depending on the level of effort you want to expend. For those of us with non-dedicated rooms, we may never get to the more advanced objectives of the model, I.e. re-introducing diffuse reflections past 20ms.
The LEDE/RFZ model sounds like a better fit for me because listening to music is a very high priority.
Originally Posted by jlpowell84
Jerry, are these models and their philosophies in full descriptive theory in this thread prior or elsewhere? I would love to learn about them.
In picking a room model, since it's highly doubtful one will have the opportunity to try out more than one or even visit rooms designed for LEDE, NE, Ambechoic, etc and spend any quality time there, you have to take a bit of a leap of faith.
I wouldn't let not knowing which model to choose stop you from getting started.
Up to a point, most all models have many/most things in common.
No matter which model you choose, you need to do the following:
First you should get the best speaker and listener placement figured out. This should be done with no treatments and no EQ. You want the best response possible while also achieving symmetry and still allowing for proper placement of surround/surround back speakers if you intend on focusing more on multichannel music/movies, but at a minimum, the L/R speakers should be placed optimally combined with the best MLP.
Optimally means you should focus on the best overall frequency response to begin with (remember, this is before treatments and EQ), along with the soundstage width you prefer, making sure to keep the height, width, and front wall to front of speaker distances all different. You should pick a spot with as few reflections, as little SBIR, and as few nulls as possible, but mainly focus on frequencies above the crossover assuming you'll be using separate subwoofers. Note, even with using external subwoofers which will have their own separate placement, remember a crossover isn't a brick wall filter and the speakers and subs have to work with each other, not against each other. Also, common crossovers are 80hz which is still well below the transition frequency and below specular reflection territory in all rooms so Modes/Nodes/SBIR is still very important, and shouldn't be dismissed altogether thinking that the subs and their placement will fix all the problems. Everything has to be designed/placed to work together.
Next, you will need to integrate your subs using all the known and often discussed methods to integrate them with your main speakers.
Then you can start introducing treatments, starting with bass trapping. The reasons you want to introduce bass trapping first is because most of the bass traps folks in this thread will use are going to be broadband and will themselves alter the frequency response and specular reflections. If you added reflection panels first and other broadband acoustical panels to get the room tamed and get to the decay levels you want to achieve, then start adding bass traps, you'll quickly find you're making your room too dead and will have to remove and probably also change the position of many panels. You can save money (buying less panels) and time by focusing on bass traps first.
Only after you have the frequencies under 300hz tamed as much as possible for an even decay rate and relatively flat frequency response with no major nulls and if possible no major peaks (but no matter what, you need to get rid of the nulls via placement of subs, speakers, and MLP plus bass traps as no amount of tweaking or EQ will help you here) should you move forward.
At this point, (again, this is for any/all acoustical models) you want to take a look at your efforts and see how they've all paid off in REW, doing another before/after test. You should already see a lot of improvement at this point, but when looking at the ETC, which is what we focus a lot on in determining the different acoustical models and how the corresponding ETC should look for each model, you need to make sure that all the high gain early reflections up to around 20ms are tamed to -20db or below and ALSO (this is where a lot of people fail) that there are no spikes or masses of dips in that response from the direct signal to about 20ms. You may have all your reflections under -20db out to 20ms but it should also look smooth and be densely populated. If you have sparse reflections with peaks and the peaks are at -20db while most of the rest of the response is -25db or even -30db, this isn't good either. You want the ETC to be smooth, densely populated, and no major peaks/nulls out to around 20ms.
Only at this point do you really need to be concerned with implementing an acoustical model.
By the time you've accomplished all the common goals listed above, you'll hopefully have a better idea of how the changes you make alter the sound of the room and your listening experience and enjoyment, plus in your down time when not physically working on the room or making measurements, hopefully you'll have had time to research the different acoustical models more.
By the time you get to the point of really needing to nail down how you'll finalize your room, you'll probably realize where you'll want to take it and how you'll accomplish it.
So go ahead and get started, and be sure to share your results. There is no reason to wait until you're 100% sure whether you're going with LEDE/NE/Ambechoic, etc, before treating your room and making it sound 10 times better than it does now. You'll learn a lot in the process and have a much clearer understanding by the time you're to the last step and as long as at that time you are able to make a decision on a model and stick with it, you won't have to undo or redo anything so you'll have had just that much more time to enjoy your system and all the improvements along the way.
Hope this helps.