Can someone recommend me a room mode calculator?

Started by stevesdata on 17 November 2007. 39 replies. In the Library under Measurement and tools.

Originally posted at johnlsayers.com, topic 9767.

I have the Acoustic Calculator from mhsoft.nl but it seems to spit out incorrect results for decmal places eg. I enter 4 and its correct. I enter 4.2 and it spits out incorrect results : ( Love the rest of the calculator btw just this bug... I could really do with double checking all the room modes I have calculated from scratch in excel. :?
Hi, I just wanted to link to the one liked here in the forum and standard referred by Steve/Knightfly (which he entered as sticky here), and point via the Google link there are lots, to show the OP there wasn't a reason to be stuck on that single one he had trouble with. It was meant neutral. Bob's is good, Jeff's is good, John's is very good and special, and I didn't link to my own either, which you both didn't mention as well. I will complete that topic I referred with current links of these others you mentioned here (and some more).
Eric_Desart wrote:
Bob's is good, Jeff's is good, John's is very good and special, and I didn't link to my own either, which you both didn't mention as well.
Eric, sorry buddy - I don't believe I ever used yours.......... didn't even knew it existed - would love a link...... Rod
rod gervais wrote:
dbl post
apologies I should have checked the other links first :roll:
Rod: It's an Excel file: http://forum.studiotips.com/large_files ... ulator.zip Note: this file is designed for Windows and doesn't work well (better: rather bad) on a MAC. I rarely link to my own stuff, unless presenting something specific. In fact I have an improved version of this, but lost by private circumstances. These pictures were auto-generated by this Excel file (without aid of any graphic program, direct from within Excel): You can't generate these pictures from the downloadable one, but the looks are roughly similar.
External image, not preserved — original: http://img295.imageshack.us/img295/848/roommodes04ik7.gif
Somebody is going to change the electronics of old disks, hoping to make them accessible again. It's annoying: I have a disk of 250 Gig lying in front of me, full of data/own written software, still not accessible.
Apologies I feel a bit silly should have searched better. Oddly enough I searched google before and didnt find much :? These are great resources to test the one I made which simply calculates all modes and is in metric up to 3 - axial, tangential, oblique. I will post it if anyone interested but its very basic compared to some of these fantastic examples. Many thanks :)
Eric_Desart wrote:
It's annoying: I have a disk of 250 Gig lying in front of me, full of data/own written software, still not accessible.
Eric, do you know the format of the disk? I have had no small luck at retreiving data off hard drives (depending on format) and possess quiet a few utilities devoted specifically towards that purpose (I have had part of my work include Unix System Manager, Windows NT system manager - OS2 systems manager - all within networked invironments - and ported our system over from OS2 to Windows NT with the (successful) conversion of our accounting system along the way. I am a certified A+ computer technician....... did this a lot until I got too busy - now the company has full time staff on board. Rod
myfipie wrote:
Just to add to the others. http://www.realtraps.com/modecalc.htm Glenn
Reminder:
Eric_Desart wrote:
Please look for these wich calculate ALL modes (Axials, Tangentials, Obliques). Explained here: http://forum.studiotips.com/viewtopic.php?p=5570#5570
Glenn the one you referred doesn't. That calculator emphasizes the unimportance of non-axials, while simultaneously advising ratios from 'unspecified' Authors/sources, which are based and calculated by using non-axials, which is a contradiction in itself. The graph I entered in this thread, are calculated on the use of all modes. It are, amongs others, ratios from this Author, which are used as adviced ratios in the calculator you referred. Also explained here: http://www.johnlsayers.com/phpBB2/viewt ... ght=#64080 And I linked to a neutral Google search: There are lots calculating them all. The original Knightfly sticky link I referred in this thread, and the ones later linked by others also calculates them all. The one you refer does not belong in this series. Rod, I'll come back to that disk (but thanks for this gentle gesture).
Erik, :D It is Glenn not Glen. :lol:
Glen the one you referred doesn't.
I knew this. Glen(N)
Glenn, I was still editing (post crossed), and adjusted it (was just a simple typo). If you knew (which I know you do), you should know it doesn't belong in the series discussed here. Hence my comments were more meant to prevent confusion for others. You have 2 only-axials (incl. the one you refer here) and no all-modes calculators referred in a sticky in a forum you moderate. But that's your choice for whatever reason.
I was still editing (post crossed), and adjusted it (was just a simple typo).
In the future I will be sure to wait more then 45 min and several edits before pointing it out. :roll: :lol: Glenn
myfipie wrote:
I was still editing (post crossed), and adjusted it (was just a simple typo).
In the future I will be sure to wait more then 45 min and several edits before pointing it out. :roll: :lol: Glenn
I saw it while entering my second edit, which was relative short after composing my post, and the other ones I made later. I made a typo OK? And nothing more OK? I apologize to you and the rest of the world for making typos, my poor language knowledge, bad grammar, poor spelling and slow typing with 1 to 2 vingers. I was primary busy with calculators in my mind and post. Which I thought to be the topic and content of your post was about, not the spelling of your name. My comments were worth more, and going deeper (as you know), than a spelling discussion.
I made a typo OK? And nothing more OK? Nag nag ....
Honestly found it more funny then anything else. Glenn
Eric_Desart wrote:
That calculator emphasizes the unimportance of non-axials
Not to open a can of gummy worms here, but maybe I can clarify a few points. Glenn Kuras just emailed me for my opinion:
I need to wrap my head around this: L. W. Sepmeyer: 1965 - 1:1.14:1.39. This is one of world's most common used ratio. When one judges those room ratios based on the Axial modes alone one can only conclude that the modal distribution of this ratio looks terrible. However L.W. Sepmeyer counted on the non-Axial modes in order to optimize the modal energetic distribution. The same is valid for his other ratios. M. M. Louden: 1971: Best ratio as per Louden - 1:1.4:1.9 Another of world's most common used ratios. When studying lots of Louden's room ratios, even the best ranked (he published 125 ratios in decaying order), one will notice that without the non-Axial modes, only judged on the Axial modes, lots of those common used and worldwide accepted ratios should be rejected resulting from an ugly looking distribution. Care to comment on this?
I'm sure that non-axials account for something. Jeff Szymanski once explained that in theory tangentials are 3 dB lower than axials (half as strong), and obliques are 6 dB lower (one quarter as strong). But non-axials do not happen at the very low frequencies that axials do. The first tangential mode starts almost an octave higher than the first axial mode, and oblique modes start even higher still. So by the time you get high enough in frequency for most of the non-axials to kick in, you're well into the range where bass traps can easily tame them. The main reason I don't show non-axials in my ModeCalc program is because it's simply Too Much Information for most people. If you include all the modes you'll end up with 100 frequencies, which is impossible to deal with. I'm sure you've seen those lists showing modes every 1 to 3 Hz. WTF is anyone, let alone mere mortals, supposed to do with all that? Worse, the importance of the modes is not made clear in most of those calculators, so the forest is hidden by the trees. If you look at only the axial modes you can see much better where the strongest resonances are. My ModeCalc goes further by highlighting modes that are close together with taller lines. IMO that is more useful than showing all non-axial modes which are 3 or 6 dB less important. I could even argue that in many case modes can mostly be ignored - though obviously you don't want square or cube or even multiples like 10 by 20 etc. Most of the people I deal with are not building $250K rooms in a 35 by 45 foot shell. So the room size and shape must be ergonomic and practical as much as acoustically valid. If the choice is between having a "perfect" ratio or having a room that can accommodate your keyboards and guitar amps and guest seating, I'll take the latter every time. And then treat the hell out of it with bass traps. Which is needed anyway even with an "ideal" mode distribution. Finally, as we now all know after my RO post Room modes, calculated versus measured, the modes you predict are not necessarily the modes you'll end up with. So you carefully plan all those modes that are precisely 2 Hz apart, then build the room, and all that precision is now out the window. At least at the lowest mode frequencies which matter the most because they're hardest to trap. --Ethan
Ethan I am in no way saying ANYONE is 100% right or wrong but (Eric I think you have some great points on the subject) Ethan I think has at least started questioning things to find new answers. I will say I found Rods comment VERY interesting and explained a great deal of questions I had on the subject. Rod wrote,
Ethan, This is exactly the same (although different) as modeling walls for sound isolation. All of the math - all of the theories - all of the experience - and when it comes right down to it - the ONLY accuracy exists with the actual testing performed after construction. I know it seems like this shouldn't be - but it is. All of the mode calculations are based on a rom with a perfect boundry - nothing escapes - and they are also based with the sound emitting from the surface of the boundry. Then we walk into a room - in most cases wallls against the building exterior are a totally different density and construction than the interior - some with brick - concrete - perhaps plywood (or exterior grade gyp) with stucco over the exterior of that. Our mode calcs know nothing of this - nothing of the construction - nothing of the techniques involved in that construction. We also do not produce the sound at the boundry - as it is assumed within the calculation. We step inside - slide that speaker away from the wall - create a distand for the sound to travel that the calculator never took into account - a reflection off the wall behind that was never anticipitated by the mathematical equation. And in the end - we are suprised to find a variance? I (for one) would be surprised to find extreme accuracy - By the way my friend - for what it's worth - I don't think (when we deal with sound) that we can look at it as percentages - I thought last night about Jeff's observation that it's only a 20% when viewed in relation to the nearest mode you expected - and would be about 6% veiwed in relation to another frequency....... He's right - we can't view it in that respect - if it isn't consistent then it skews the results. The long and short of it is this - modeling is good for leading us in a direction that will cause us the least problems down the road, but in the end - the only thing that's accurate is the final testing. Sincerely, Rod
Glenn
myfipie wrote:
I will say I found Rods comment VERY interesting and explained a great deal of questions I had on the subject.
Yes Glenn, as Rod wrote, "the ONLY accuracy exists with the actual testing performed after construction" which of course is the whole reason I started that 14-page thread over at RO. However, I do disagree with this:
Rod wrote, I don't think (when we deal with sound) that we can look at it as percentages
I disagree with Rod (and Jeff) about percentages not being the right way to assess modes, for the reason I explained in that thread. Musical notes and octaves and frequencies are always expressed as percentages. (Logarithms, really, which is the same concept.) That an FFT algorithm happens to use a fixed Hz bandwidth is a side effect of the algorithm, not a design goal. FFT algorithms have nothing to do with the subject at hand anyway. A musical Bb note is always some percentage higher than the A note below, regardless of which octave you consider. And modes and their bandwidths (Q) etc are also always expressed as a percentage. Or they should be. I realize that Everest and some others talk about mode bandwidths in terms of Hz rather than Q which is a percentage. But Q is indeed the correct way to view audio bandwidths, whether for an EQ filter or in acoustics. --Ethan
Ethan Winer wrote:
Finally, as we now all know after my RO post [url=http://www.recording.org/modules.php?name=Forums&file=viewtopic&t=26568]
Uh :Dbbc rd 1992-09
Ethan, I don't have the time for another pi$$ing match, but I would like to address this...
Ethan Winer wrote:
The main reason I don't show non-axials in my ModeCalc program is because it's simply Too Much Information for most people. If you include all the modes you'll end up with 100 frequencies, which is impossible to deal with. I'm sure you've seen those lists showing modes every 1 to 3 Hz. WTF is anyone, let alone mere mortals, supposed to do with all that? Worse, the importance of the modes is not made clear in most of those calculators, so the forest is hidden by the trees. If you look at only the axial modes you can see much better where the strongest resonances are. My ModeCalc goes further by highlighting modes that are close together with taller lines. IMO that is more useful than showing all non-axial modes which are 3 or 6 dB less important.
IMO, there are HUGE flaws in this line of thinking. But the main one is: Information is deliberately being left out. Regardless of what you, the programmer, believes to be "important," you cannot censor information because you believe it benefits the user. Because you have deliberately censored the information is exactly the reason Eric is always quick to point it out and criticize it. And rightly so, IMO. You cannot be in all rooms and cannot possibly know what information is or is not important. The possibility is and always has been that non-savvy users could obtain results from your calculator and make serious room design errors. That possibility will always exist, regardless of your personal opinions and philosophies on room acoustics. ... ... Of course, from another perspective, if you don't place any importance on room modes then why even bother with a calculator :?:
OK, so maybe I have a little time... I couldn't let this go:
Ethan Winer wrote:
I'm sure that non-axials account for something. Jeff Szymanski once explained that in theory tangentials are 3 dB lower than axials (half as strong), and obliques are 6 dB lower (one quarter as strong).
I never explained any such thing. As I have explained before, the 3/6 dB difference is only relevant to modes of different types that occur at the same frequency. And it is theoretical. And it would be additive, whereby an axial and tangential mode at the same frequency could still combine to have an overall amplitude higher than each mode taken on its own. The actual amplitude of a mode is dependent on the characteristics of the room and the relative positions of the source and receiver in the room. Further, I have seen at least one software developer include "weighting" schemes for different mode types, adjustable by the user. Some authors (Bonello comes to mind) have offered up reasoning for this in their writings. But their work is often dated. From a 2007 perspective, I cannot say that I agree with this line of thinking since the modal response of a room can be predicted much easier these days, thereby removing any need to "weight" anything. In my experience, this calculated/predicted response often agrees well with the measured response. All the best,
Ethan Winer wrote:
A musical Bb note is always some percentage higher than the A note below, regardless of which octave you consider.
except in the case when a nearby frequency mode or combing filter effect causes the note to appear to become sharp or flat(ter). :?