Gareth James wrote on Sat, 03 February 2007 12:48 |
Elliot, the TH-215 died also? According to the specs its 1400w continuous, 2800w program not 2000? Again I'm suprised the box died when the amp wasn't even clipped or near clipping. Unless there is something I'm not seeing I wouldn't have thought either of the two Danleys would have been pushed to breaking point on that amp that far from clipping...something doesn't seem right. |
Scott Hibbard wrote on Sat, 03 February 2007 15:03 |
Paul, The Quakes were loud and pounding. They were obviously driven to their absolute limit, they did die trying. Two dead drivers. Did we kill the Quake drivers? For whatever reason, I don't remeber that?! I know we cook the EM 215 drivers... ScottH |
Tim Morin wrote on Sun, 04 February 2007 12:33 |
I agree, we would also be intersted in a shootout of this type. the problem will be that most of these are mostly about loud and painful. I have read threads about the NY sub shootout and most of the fuss is who's cabinets were the loudest. I have not seen any threads on the musicality of the subs as that is truly what determines which of these manufacturers will be on a major tour or install in the future. just my 2 cents. |
Michael 'Bink' Knowles wrote on Sun, 04 February 2007 12:57 |
Very nice photos, Will. By far the best capture of the venue and the atmosphere. I'm surprised by the very small amount of hearing protection in evidence. Most of the testing was reasonable volume, eh? -Bink |
Tim Morin wrote on Sun, 04 February 2007 11:33 |
I agree, we would also be intersted in a shootout of this type. the problem will be that most of these are mostly about loud and painful. I have read threads about the NY sub shootout and most of the fuss is whose cabinets were the loudest. I have not seen any threads on the musicality of the subs as that is truly what determines which of these manufacturers will be on a major tour or install in the future. just my 2 cents. |
Michael 'Bink' Knowles wrote on Sun, 04 February 2007 12:57 |
snip… Most of the testing was reasonable volume, eh? -Bink |
Ivan Beaver wrote on Sun, 04 February 2007 13:58 |
We have setup lots of blind tests between full range boxes. It can be difficult to set a level playing field. The low end response in particular (and overall flatness of the response) is a tough puppy to be "fair" with. What we have decided is the best way (we are open to other opinions however) is to put full range pink noise into each one and set the level as close as possible using a C weighted SPL meter. We use a custom switcher that I designed/built that allows for individual adjustments over the cabinets and then allows for bypassing each level control so sensitivity differences can also be heard. When you listen to them you will hear (with different source material) that some boxes sound louder than others and on other tracks the levels seem more matched. The boxes with the most lowend will sound warmer and will skew the "sound" of the upper freq. When you have the extra extension on the low end end the higher freq don't seem as harsh. So if you say to highpass all the cabinets at say 100Hz, you are unfairly robbing performance of the cabinets that are capable of truly full range response. You would be surprised how many cabinets that claim to be full range really aren't-despite what the specs say. When you have a switcher and multiple boxes setup next to each other, you can really hear all sorts of differences between boxes. The little details really stand out. A shootout would be fairly hard thing to do-especially in a small area as you would have to have at least 2 of each box (to check for arrayability and how well they play together) and be able to switch easily between them. I think there would have to be some strict guidelines setup as to what will and what won't be done during the comparisons. I think you would have to set some guidelines as to different types of boxes that should be compaired to each other- It would be really telling however. |
Tom Danley wrote on Mon, 05 February 2007 10:48 |
Hi In my personal opinion (not DSL’s ), there are some highly polished, highly respected acoustical turds floating in great numbers out there. You can see / hear / identify them, or not, when doing side by sides. |
Quote: |
Lastly, I would again echo David at his suggestion that this kind of thing be done in a larger space, Ideally this would be outside or a stadium etc where measurements would be or nearly are half space and at a distance which removes the source size / box size errors which appear near field. Also, this lets people move farther away back to where ones ears are more linear and more able to hear some speaker issues (where as local noise is more audible up close). Anyway, I’ll cut this off here and get back to work. Best, Tom |
Tim McCulloch wrote on Sun, 04 February 2007 18:42 |
Hi Mark- The Bassmaxx Triples had been playing when the neighbor across the street came over with an attitude and threatening to call the police........... Paul Bell had anticipated complaints from neighbors in the spaces above Club Rebel....... Tim Mc |
Mark "Bass Pig" Weiss wrote on Mon, 05 February 2007 15:40 |
One would think they had to have heard ALL of the testing, but then if the club's own sound system played every night, wouldn't the neighbors complain about that too? Or was this testing THAT much louder that it penetrated both the building, crossing, what I would imagine is a very wide street (120' perhaps for typical downtown Manhattan thoroughfares?) and then had to penetrate the building across the street. Now if both buildings were supported on a number of common members that span across the street and can conduct vibrations, then I guess that could explain it, but I thought all NYC buildings were built on pylons that went directly into bedrock below them. If so, then the vibrations had to come via other paths. Given my own experience with Bassmaxx subs, I suppose it is certainly quite possible in NYC. It would have been interesting to see a SPL reading at the 5th floor apartment, both indoors and out. I'm guessing it had to be well over 90dB in the 40-60Hz range. Nonetheless, it was a very interesting anectdote to read about, which added a unique flavor to this event. |
Tom Danley wrote on Mon, 05 February 2007 16:48 |
Hi Protocol is an important part of measuring or comparing anything. As Ivan said “When the "rules" change mid stream, it makes it a little unfair to the earlier "competitors". In this case, he refers to the request to “crank it up” which came right after our boxes had been auditioned. This and the recorded peaks may lead to impressions like Mark’s “Sounds like comparing the space shuttle takeoff to a bunch of 747s taking off” when they never even blasted the TH-215’s. |
Pascal Pincosy wrote on Mon, 05 February 2007 18:55 |
Hi Tom, I posted a similar comment earlier but it seemed to have gotten lost in the shuffle to this new thread. Extrapolating the SPL readings, one would see that your TH-215's recorded a peak that was >2dB louder than specs on your web site. Not a very accurate comparison considering room gain and all, but still. Perhaps I'm not understanding the methodology used to apply power to the speakers. Who made the determination on how much power was applied to each cabinet and why were some boxes pushed hard and not others? Either way, the measurements will speak for themselves. Easy enough to calculate peaks for each cabinet from the sensitivity readings. Tom, maybe you could annotate the thread with updated specs for the TH-215's power-handling so that one could calculate peak output with the new numbers? |
Pascal Pincosy wrote on Mon, 05 February 2007 23:55 |
Perhaps I'm not understanding the methodology used to apply power to the speakers. Who made the determination on how much power was applied to each cabinet and why were some boxes pushed hard and not others? |
Wil Davis wrote on Sun, 04 February 2007 17:15 |
Hey Chaps! I had a wonderful time in NYC; it was nice to meet you all, and great to get a chance to hear some amazing gear. I also learned a helluva lot; overall a great experience, thanks again to you all, esp. Paul & Co. and Bennett & Co. and Dave… and Ivan… and Mark… and John… and Mike… and Eliot… and… Here are some of my pics, not quite as newsworthy as those of Paul's Uncle Ray, but they might give those of you who couldn't make it, an idea of what it/we looked like: http://www.k1wd.com/misc/stuff/NYCShootOut2007/ So, when's the next one? - Wil |
Ivan Beaver wrote on Tue, 06 February 2007 07:13 |
We had two DVM's neither of which had great resolution. They varied in the voltage readings by 1.5dB. |
Wayne Parham wrote on Wed, 07 February 2007 15:27 |
Are the measurement datasets online anywhere? I looked through all the posts and didn't see much in the way of hard data. Is it being organized before placing it online or am I overlooking it somewhere? I think I saw that you used a reference voltage, 2.83v. Did you measure impedance charts? I assume you measured SPL, but did you measure distortion? If so, what power (or voltage) level? For distortion, did you use a sweep or discrete sines and filter for harmonics? Did you filter out the fundamental to find THD+N, or maybe instead filter for specific harmonics? For frequency and impulse response, did you use a maximum-length sequence, a sweep or what? |
Wayne Parham wrote on Wed, 07 February 2007 15:27 |
<snip>… How far away were the speakers from the walls? How far was the microphone placed from the speakers? What was the layout? I know you were indoors, but I'm just wondering what the setup was. Sorry for so many questions, just trying to see what you guys did. Not looking to armchair quarterback at all, just looking for information to know how to compare your data with mine and others. |
Tom Danley wrote on Fri, 02 February 2007 14:14 |
Hi Elliott The TH-215’s have a new drivers in them too, these should be.... Tom Danley |
Tim McCulloch wrote on Thu, 08 February 2007 19:47 |
Ivan and & I discussed the drivers, and Tom can offer more output, but higher F3, or go lower with less output, driver dependant. It might be an "option" thing... or for the sake of identical performance, Tom may pick a driver. HTH Tim Mc |
Michael Hedden Jr.wrote on Thu, 08 February 2007 19:50 |
The drivers we used at the shootout are indeed standard issue. We will change the spec sheet to reflect that shortly. By the way for the power handling crowd this provides us with 2400 watts RMS, 4800 watts continuous, 9600 watts peak on the only sub at the shootout essentially flat from 60 to 30 Hz. Mike Hedden |
Tim McCulloch wrote on Thu, 08 February 2007 11:47 |
The indoor measurements of the Danley TH-115 will be compared with outdoor measurement of the TH-115. Mark Seaton is hoping to derive some "correction" from the differences, and that correction will be applied to all speakers, as they were all measured in the exact same spot, same mic position (it was taped to the floor), and same input voltages. Regarding mic distance.. that was measured. I believe Mark has that information as well, but IIRC it was slightly over 1 metre. Tim Mc |
Wayne Parham wrote on Thu, 08 February 2007 14:31 | ||
Very good, thanks. With the microphone distance known, SPL can be calculated relative to 1 meter at whatever power or voltage levels were used. That gives us a reference. Good idea on the indoor/outdoor correction scheme. But didn't I see that the drivers used at the NYC shootout were different than those used in the published (outdoor) measurements? I guess that could be solved by measuring outdoors again with the new drivers to get a new outdoor baseline. But then again, didn't the TH-115 drivers overheat? Seemed like I read they smoked. |
Michael Hedden Jr. wrote on Thu, 08 February 2007 19:50 |
The drivers we used at the shootout are indeed standard issue. We will change the spec sheet to reflect that shortly. By the way for the power handling crowd this provides us with 2400 watts RMS, 4800 watts continuous, 9600 watts peak on the only sub at the shootout essentially flat from 60 to 30 Hz. Mike Hedden |
Michael Hedden Jr. wrote on Thu, 08 February 2007 21:12 |
For the record the only drivers that smoked at the shootout were from EM. so even more impressive was that a single TH115 was responsible for the first official noise complaint! |
Maqc Kerr wrote on Fri, 02 February 2007 03:04 |
We got smoke out of the TH115. I'd say that was pushing it to the max. |
Paul Bell wrote on Thu, 01 February 2007 17:41 |
Yes, the first gentleman who wanted to know what was going on was a NYC EPA inspector. Some weeks ago, I posted notices throughout the building so the many recording studios would be aware of our event. I included "DB levels of 145..." on the notice. One neighbor, who hates club Rebel, made an appointment for the inspector to be here during the shootout. He was impressed with what we were doing and found no problems or people complaining about the sound. The across the street neighbor heard the bass and ran over to scream. We had been doing the extended listening of the Trip cabinets so we all took a little break. This was the only noise complaint we got. |
Michael Hedden Jr. wrote on Thu, 08 February 2007 21:49 |
Gee I wasn't aware the data was sealed! Can I help it if I own a TEF, Mark Seaton is an old friend and Ivan Beaver and I work together? By the way I also supplied the Earthworks M30 and calibrator for the shootout which we sent off to Larson/Davis to be checked and recalibrated just so it would be spot on for the shootout. I'll be glad to send anyone the files that would like them but you have to have the TEF software to view them. I'm sure Mark is wondering what he got himself into as it is no small task to put all this data into a screen capture that doesn't look like a mess. But the great thing about good data is it becomes an arbiter and you'll see it isn't just a claim it's a fact most of the subs start rolling off between 50-40 Hz and are 5-15 dB down @ 30 hz while the TH215 is indeed basically flat. Mike Hedden Danely Sound Labs, Inc. |
Mark Seaton wrote on Thu, 08 February 2007 23:36 |
Ryan Lantzy wrote on Fri, 09 February 2007 13:45 |
I'm not sure this is the final format for how the rest of us will see the measurements but I have a suggestion. That graph is very dense and hard to see some of the colors/traces in certain areas. Additionally some of the lines are thinner than they are in the key so I think the image compression messes with the actual shade of color. Would it be possible to get more detailed images? I would not insert them into the messages here however because they make the thread VERY hard to read on a smaller screen. (And I'm working on a 19" at 1280x1024!!) Another suggestion would be to some how export the data from TEF in an "X-Y" plot of some sort. That way we could load it up in our own viewer and overlay whichever traces we wanted to. Being able to select which traces you want to view is a nice feature which I'm guessing you can do in TEF. I'm just wondering bitmaps/JPEGs/GIFs are the only means of export from TEF or if you can export the raw "data" and then import into Excel for example? Or is there some freeware X-Y plot viewer we could all import the data to and use. |
Mark Seaton wrote on Thu, 08 February 2007 22:36 |
To give a little perspective to all the data aquired, (and so no one is wondering if it exists) here is a quick display of the 2.8V measurement for all 14 subs measured. Do remember that 2 were powered, so a 2.8V label is meaningless, although we kept the subs at somewhat respective levels. Yes, that's 14 subwoofers, and you're only looking at 1 of 7 measurements for each. If anyone is interested in seeing the data, Ivan or I can e-mail you the nearly 2.0MB of 98 separate measurements. Again, in current form the files are only viewable with the TEF software. |
Wayne Parham wrote on Fri, 09 February 2007 10:15 |
The response curves for all the systems look very similar. Are the boxes that were measured indeed that much alike, or are the measurements so similar mostly due to room influence? I'm afraid it might be the latter. |
Quote: |
Boundary conditions would smooth horn response, and room modes would contribute peaks and nulls. I'm concerned that the measurements are heavily influenced by the room, and that they're mostly the room's sonic signature. Maybe your idea of making a comparison with outdoor measurements will let you come up with a conjugate filter that you can apply. That's a good idea, maybe it will shed some light. But I'm concerned that there is just too much room influence to extract useful data from these graphs. |
Pascal Pincosy wrote on Fri, 09 February 2007 04:31 |
I can take a quick look at the Danley web site to acertain that the TH-215 is mostly flat down to 30Hz. First off, none of us at Danley Sound Labs wishes to be jerks so I apologize if my post comes off that way. It is very difficult to read intent into something typed because words have different meanings. In fact because words mean different things I wish to elaborate on that just a bit. In the beginning days of Danley Sound Labs many folks including frequent posters on this forum wished us well but cautioned that the real chore would be in stemming the marketing tide that is so prevalent in industry. The marketing hype they where talking about being defined as obvious and intentional exaggeration, an ingenious or questionable claim, method, etc., used in advertising, promotion, or publicity to intensify the effect or swindle, deception, or trick. We used to call this lying or cheating but that was back when we as a society actually had absolutes but we are beyond that now... Anyhow, this isn't good and when a company makes claims that simply don't stand up to the light of truth it should be exposed. Danley Sound Labs isn't interested in hype or deceit. Evidence of this can be found in the fact that we publish the entire sensitivity frequency response of all our products. The hype types give you some number with no referenced frequency and in many cases don't supply any measured data. Fact is if all manufacturers followed our suit you could get the data you are waiting on right now and it would be very easy to compare product to product. We tell the way our subwoofer data is gathered (ten meters outside, half space, 28.3 volts) so that you can easily test/verify our data. We are one of a minority of manufacturers with EASE and CLF data gathered by an independent lab and the only company that has independently gathered data of a full range array. We are sinful humans and as such fall short of perfection but our desire, commitment, and goal is to be the reference standard in the loudspeaker industry in not only word but in deed. Mike Hedden Danley Sound Labs, Inc. |
Wayne Parham wrote on Fri, 09 February 2007 11:37 |
Mark, let me start off by saying in no way to I want to come across as a Monday morning quarterback. No matter how much influence the room had, I think what you guys did was great. Kudos to you, to Paul, to everyone involved. I think your efforts in measuring and compiling the data are worthwhile and appreciated. I don't know if everyone reading these threads knows how much work is involved in what you're doing. You spent several hours setting up and making the measurements, and then double that amount of time converting them to a readable form. Maybe triple the time, considering you are trying to make a conjugate filter to compensate for room influence and level setting. So I applaud you, very much so. That said, I still think the room influence may have swamped the data. I've measured horns outdoors, and I can say with absolute certainty that the differences in response between basshorns are huge. Take a look at the response charts taken at the two outdoor Prosound Shootouts, for example. Each of the horns look entirely different. There's a lot of difference, even in some cases between horns that are close to the same size, having similar construction. Take the Cerwin Vega and Tuba horns, for example. Similar size, totally different response. The Cerwin Vega didn't go all that deep, but response was pretty flat. The Tubas were more peaky. Then there was my 12Pi and David Lee's BASSMAXX boxes, similar in size but completely different in response. So I think the room had a great deal of influence in the datasets captured in NYC. Prosound Shootout 2006 Results |
Aaron McQueen wrote on Fri, 09 February 2007 17:59 |
Wayne, I think there are some plot x and y axis scaling differences between your plot and the one Mark posted. |
Wayne Parham wrote on Fri, 09 February 2007 12:36 |
The chart above has 2dB/division, but the lines between divisions are closer together. Yours shows 5dB/division. Disregarding everything above 100Hz, there is still a trend of uniformity in the NYC measurements that isn't there in the outdoor measurements in Tulsa. Looks like the room causes notches at 50Hz and 70Hz, and peaks around 45Hz and 85Hz. Other than that, deep bass response in NYC is smoother than the outdoor measurements, either from post-processing or the number of samples or possibly from boundary reinforcement. |
Michael 'Bink' Knowles wrote on Fri, 09 February 2007 13:33 |
One way to try and get rid of room influence on the displayed data is to add all the data together and take the mean to make an artificial "room response curve." If one were able to completely flatten the mean curve then each subwoofer as compared to the mean would show how it responds differently. The caveat here would be the artificiality of the way the data is manipulated. Still might be interesting as a point of comparison. -Bink |
Michael Hedden Jr. wrote on Fri, 09 February 2007 12:11 |
History aside we are already working on your idea and will have a section of our site dedicated to it shortly. Thanks, Mike Hedden Danley Sound Labs |
raj sookraj wrote on Fri, 09 February 2007 19:09 |
Since voltage was constant, will the graph be adjusted to correspond to 1w/1m for each sub or pair of subs? If so, how is that going to be done? At Minimum impedance? Average impedance of a certain frequency range?? Impedance at 60hz? |
Michael Hedden Jr. wrote on Fri, 09 February 2007 15:59 |
Danley Sound Labs isn't interested in hype or deceit. Evidence of this can be found in the fact that we publish the entire sensitivity frequency response of all our products. The hype types give you some number with no referenced frequency and in many cases don't supply any measured data. Fact is if all manufacturers followed our suit you could get the data you are waiting on right now and it would be very easy to compare product to product. We tell the way our subwoofer data is gathered (ten meters outside, half space, 28.3 volts) so that you can easily test/verify our data. |
Jeff Permanian wrote on Sat, 10 February 2007 00:04 |
Obviously, the room had its effects at 50 and 80 hz. |
Ryan Lantzy wrote on Fri, 09 February 2007 13:45 |
I'm not sure this is the final format for how the rest of us will see the measurements but I have a suggestion. That graph is very dense and hard to see some of the colors/traces in certain areas. Additionally some of the lines are thinner than they are in the key so I think the image compression messes with the actual shade of color. Would it be possible to get more detailed images? I would not insert them into the messages here however because they make the thread VERY hard to read on a smaller screen. (And I'm working on a 19" at 1280x1024!!) |
Mark Seaton wrote on Fri, 09 February 2007 13:36 |
If people are going to get their panties in a bunch over a peak in the response around 70-85Hz and a few other wiggles and aren't capable of extracting what is meaningful from the differences in the curves, I'll leave them for others to play with an convert. Personally, I think readers here are a little more intelligent than that. |
Pascal Pincosy wrote on Fri, 09 February 2007 20:06 |
Hey Jeff, do you have outdoor measurements of your Growler that you could post? If so, could you overlay the indoor measurements taken and post the results? |
Mark Seaton wrote on Fri, 09 February 2007 10:39 |
The JTR Growlers have a simple and smooth enough curve that would make this easier, with the caveat that temps have been rediculous cold here and I need to take another measurement to match the measurement parameters for direct subtraction in the TEF software. |
Pascal Pincosy wrote on Fri, 09 February 2007 20:06 |
Hey Jeff, do you have outdoor measurements of your Growler that you could post? If so, could you overlay the indoor measurements taken and post the results? |
Ivan Beaver wrote on Sat, 10 February 2007 09:03 |
Here is a overlay of the inside shootout TH115's (a pair standing up) (BLUE) and outside (RED), with the same configuration. The levels are off due to several factors. The mic distance was 4' at the shootout and not 1M and the voltmeter used to set the 2.83V was of unknown accuracy. But that is not what is important here, but rather the basic shape of the curves. The Outside measurements of the TH115 were taken at 10M and 200 watt (28.3V) into the pair-the same drive(maybe?) as at the shootout. I also included a single TH115 (GREEN)outside under the same conditions but with 100W input (28.3V) just for reference. |
Michael Hedden Jr. wrote on Sat, 10 February 2007 20:05 |
We have another sub that will be shown at the NSCA show that has similar sensitivity but is flat to 20 Hz. |
Quote: |
Three TH115s is really what I'd compare to the Trip due to both boxes rolling off below 40 Hz. <snip> Three TH115s have similar impedance and overall box volume to the Trip. |
Paul Bell wrote on Sat, 10 February 2007 20:21 |
PSW via Bennett covered the beer, thanks guys! |
Wayne Parham wrote on Sat, 10 February 2007 22:46 |
Can you please post charts without smoothing applied to them? |
Quote: |
...I'll spare the widescreen mode... JTR Growler single ground plane vs. dual at Rebel |
Gareth James wrote on Sat, 10 February 2007 15:15 |
Just been trying to make a little sense out of the graph. I'm looking at the responses and then factoring in the voltage/impedance differences between cabinets. |
Ivan Beaver wrote on Sun, 11 February 2007 17:12 |
We did not do impedance measurements. Mark and I decided to use the constant voltage input method. The problem with this method is that the cabinets with a lower impedance will be dissipating more wattage than those with a higher impedance. For example lets compare the Tripp which has a rated impedance of 3 ohms as compared to the TH115 which has a rated impedance of 8 ohms. We applied 2.83 V to both. Assuming a constant impedance (which neither is) that would relate to 1 watt for the TH115 and 2.67watts for the Tripp, which is a 4.3dB difference. Now bear in mind that we were actually measuring 2 cabinets so it would be 2 watts for the TH115 and 5.34 Watts for the Tripp. Again the difference (in db) is the same. So when Mark issues the responses you can attempt to do a 1 watt comparison between the two boxes by subtracting 4.3dB from the Tripps response at a particular freq to directly compare to the TH115. Oh course the various 4 ohm boxes (most were) would be a different comparison, but you can do the math to figure the difference. If you try to use the impedance and voltage for 1Watt method, you have to make a lot of decisions regarding what the actual impedance is. Is it the rated impedance from the manufacturer or the lowest impedance measured or the average impedance and if you use average, over what freq range? This opens up a whole new can of worms for people to argue over what is "right". |
Tom Danley wrote on Sun, 11 February 2007 20:00 |
Hi Mac Impedance matters only as much as Voltage and is unavoidable when Wattage matters, even in your context ?when I drive my amp this hard, how do these speakers compare?" Impedance governs how much you can actually deliver to each speaker when your amp reaches Voltage clip or Current limiting with a low impedance. Impedance governs how many speakers you can put on each amp and determines how many boxes will get you the most possible sound out. In this case here too, speakers have power ratings, those are determined by the system efficiency, efficiency is the ratio of radiated acoustic power to all the losses so impedance is part of this. All of these are directly related to the engineering issue of getting the most out with the least cost per size per dollar etc. When used as an engineering spec, efficiency and power capacity can be use to very roughly estimate the speaker?s maximum output or output at any input power. Again, this is based on 1W 1M efficiency, NOT Voltage sensitivity. Lastly, consider a somewhat more extreme case of two imaginary speakers, both reach a maximum output of 135dB, both are driven with 56Volts. One is an 8 Ohm load, the other is .8 Ohms. One takes 3920Watts, the other 392 Watts. There is nothing wrong with constant voltage measurements but to have any connection to Power or efficiency, the load the Voltage was delivered into is unavoidably a key part. Best, Tom Danley |
Tom Danley wrote on Sun, 11 February 2007 22:05 |
They went to the trouble to calibrate a mic to have accuracy within a fraction of a dB, they poked around in the room to find a ?smoothish? spot re: room effects. I was simply suggesting in that spirit, that having some vague idea of how many watts plus or minus a few dB were being delivered, might be also interesting to some of the more technical, when reading the TEF curves. |
Dave Dermont wrote on Sat, 10 February 2007 15:58 |
I’d like to thank Jeff Permanian for thinking of the small venue type guys, and bringing out the Growler. I especially liked how two of them stacked were the perfect height for adding a short high-output “combat audio” box like a TX-4 or U-15. To balance this glowing review, I’d like to add that I think “Growler” is a really stupid product name. Dave "occasional subwoofer user" Dermont |
Tim Coyle wrote on Sun, 11 February 2007 22:58 |
I think the Outline boxes were spread across 4 channels |
Tim Coyle wrote on Sun, 11 February 2007 23:15 |
Mac - quite possibly so- I guess it was the frenchies that I was thinking of - though the outline did not have a pass-thru. I left my notepad at home, and all those front-loaders look alike - except when they've got cup-holders. peace- |
Quote: |
...I was aware of what they were doing...and their limitations as well... |
Michael 'Bink' Knowles wrote on Mon, 12 February 2007 02:12 | ||
Since I wasn't there I'm guilty of being an armchair quarterback when I say it would have been interesting to plot each speaker's distortion performance. If each speaker were driven to a certain set percentage of distortion (as measured further back so as to lessen the impact of port noises and rattling wheels) at a fair number of frequency points then we would have an idea of how loud each cabinet could go while staying faithful to the signal it was receiving. A test like this would have taken at least a week. -Bink |
Tom Danley wrote on Mon, 12 February 2007 10:35 |
I am not clear on your other point though. You said ?the 2,6 ohm Bassmaxx may have been at a slight disadvantage because while the Powersoft K10 can deliver 2000W/ch into 8 ohms and 4000W/ch into 4 ohms, it can only deliver 6000W/ch in 2 ohms.? A few thoughts, the impedance of 3, 8 Ohm drivers in parallel is around 3 Ohms, this cabinet would then have a nominal impedance of about 3 Ohms. If you look at Wayne?s web site, he had measured the impedance of the Boxes he tested including this one. http://audioroundtable.com/ProSpeakers/messages/372.html This would limit the maximum delivered power then to only about 5KW, being in between 2 and 4 Ohms. I do not see how when Voltage referenced, how that puts a 3 Ohm load at a ?disadvantage? when compared to an 8 Ohm box (which receives ? 4.3 dB less power at the same voltage) or the other 4 Ohm boxes however, the only load which would get more power at a fixed Voltage would be a 2 Ohm load. Ivan and Mark debated where to put the test boxes in the room to avoid issues like the stage in the beginning, then they were trying to correlate the outdoor measurements of the th-115 to the indoor ones. Best, Tom Danley |
Mac Kerr wrote on Mon, 12 February 2007 10:41 |
It may be possible to somewhat eliminate the effects of the room by creating some transform based on measurements taken under different conditions, but since the various speakers filled the room side to side between 20% and 60% I'm not convinced that the transform will apply equally to all speakers. |
Mac Kerr wrote on Mon, 12 February 2007 10:41 |
This discussion has already gone on way too far, but my comment was meant to reflect the fact that the difference between the 8 ohm boxes and the 4 ohm boxes was nominally 3dB, while difference between the 4 ohm and 2.6 ohm boxes would be slightly less since the amp no longer follows the doubling of power for halving of impedance behavior. My understanding was that the "shootout" was primarily a listening experience. The fact there were measurements taken would quantify what we were hearing, and perhaps expose differences that we were not hearing. It may be possible to somewhat eliminate the effects of the room by creating some transform based on measurements taken under different conditions, but since the various speakers filled the room side to side between 20% and 60% I'm not convinced that the transform will apply equally to all speakers. What little of the data I have seen, and what we saw live, did support what we heard as far as low frequency extension, and ability to create high SPL at those very low frequencies. Since I believe this was primarily a listening experience, I will let you know how I think the various products fared. This is only one man's opinion. I think there were 3 categories of speakers when all was said and done. The group of speakers that were generally outclassed, and didn't rate a close listen on day 2. Some of these just didn't have enough output, some didn't have enough low frequency extension, and some were small enough to not have either. All of the speakers that we listened to on day 2 had some feature of interest. I felt there were 5 speakers that stood above the others by having good low frequency extension, good output power, and sounded good at all levels. In no particular order, for me these were, the 2 Danleys, TH115 and TH215, the 2 EM Acoustics, the EMS-215 and the Quake, and the traditional front loaded double 18 Outline Subtech-218. Unfortunately I was out of the room getting lunch while the Bassmaxx Z5000 was first being listened to, so I never heard it at reasonable levels. At the very high levels I heard for the last minute or so of listening I could hear only the roar of a jet engine. This was also true of the very high level listening of the X3C. I don't know whether it was an interaction with the room at those levels, or if my ears were shutting down, even though I was wearing hearing protection, but the extreme levels produced by the 2 Bassmaxx cabinets did not sound sound good to me. I had to leave the room during that part because it was painful to me. The X3Cs were interesting because they sound good at reasonable levels, and went very low, and got very loud. I just can't say whether they sounded good at those high levels. Another speaker that was very interesting was the JTR Growler. Because of it's size it can't compete box to box with the bigger speakers, but it had very impressive low frequency extension, and sounded very good. With the size, and apparent price point it is a very impressive sub. The last interesting speaker was the 6 18" powered, active cardioid, ADR sub. It too sounded very good at reasonable level, but at very high started to roar. The active cardioid feature seemed to work. My opinion, it's worth at least what you paid for it. Mac |
Ivan Beaver wrote on Sun, 11 February 2007 02:32 |
Actually the quote that was said was that a PAIR of Trips was about the same cubic volume as 3 TH115's. |
raj sookraj wrote on Mon, 12 February 2007 16:54 |
I corrected the data in this spreadsheet according to the manufacturer's website. If any numbers are incorrect, please post it. It's arranged from lowest to highest impedance, with other useful information... |
Michael Hedden Jr. wrote on Tue, 13 February 2007 03:35 |
Well don't know if the horse is dead yet but he doesn't appear to be moving much but lets try this again. The TH115 does have a sensitivity of 110 dB @100 Hz and an average sensitivity of 103 dB from 39 Hz- 80 Hz. A sensitivity must be referenced to a frequency so let's take the the Bassmaxx Trip which has a listed sensitivity of 109 dB but unfortunately no frequency of reference. Based on measurements found at http://www.audioroundtable.com/ProSpeakers/messages/372.html it would appear that the Trip's 109 sensitivity is referenced to a pair of Trips @ 75 Hz. This same data also shows the pair of Trip 's sensitivity @ 45Hz is 103dB which is actually 2 dB less than a single TH115 at the same frequency. A single Trip @ 45 Hz is actually 7 dB from a single TH115 @ 45 Hz so it is really easy to see Pascal why three TH115's would do just fine against two Trips. |
Tim Coyle wrote on Mon, 12 February 2007 03:58 |
Well.... I've not had time until recently to keep up with the thread, so this is my first 2 cents. So huge thanks to Mark for dealing with all the measurements, and trying to correlate them quickly. Thanks to Elliot for the off-site pictorial coverage - even though I notice some fire equipment in the EM photos And thanks Bennett & Paul for beer & pizza, without which I would not have stayed around to learn anything. And I did learn a couple of things (some a bit late). |
raj sookraj wrote on Tue, 13 February 2007 00:54 |
I corrected the data in this spreadsheet according to the manufacturer's website. If any numbers are incorrect, please post it. It's arranged from lowest to highest impedance, with other useful information. |
Pascal Pincosy wrote on Mon, 12 February 2007 22:55 |
Mike, as mentioned earlier on this forum, those outdoor measurements were taken with the neo drivers. The NY shootout was done with the ferrite drivers, which have a significantly better response in the low end, as seen in the NY measurements. |
Michael Hedden Jr. wrote on Tue, 13 February 2007 17:28 |
I agree with you Wayne. What we are comparing is our outdoor measurements taken at ten meters to the ones done similarly at your Tulsa show. These should correlate well. |
Michael Hedden Jr. wrote on Tue, 13 February 2007 11:28 |
I agree with you Wayne. What we are comparing is our outdoor measurements taken at ten meters to the ones done similarly at your Tulsa show. These should correlate well. |
David J Lee wrote on Tue, 13 February 2007 15:46 |
If you couldn't make it to the shootout and/or if you want to hear a bunch of big sound systems with big subwoofer arrays, Trips and Deuces included, come to Miami on the 23rd and 24th of March for the Winter Music Conference and Ultra Fest. (Contact me for details.) It's a great opportunity to hear lots of big line arrays and subwoofers in the same place at the same time. (Meyer, VDosc, Nexo, BASSMAXX, DAS and more.) If you can't make it, want to do your own shootout?!? It's the only way to be sure... |
Wayne Parham wrote on Tue, 13 February 2007 11:13 | ||
I have no doubt that a change of drivers would make a change in response. However, I don't think there is any way to compare data from the indoor measurements in NYC with data from the outdoors measurements in Tulsa. The environmental differences are just too great, and it impacts response and overall SPL. SPL is greater indoors because reflected energy is contained and response shape is modified because of room gain and standing wave modes. |
Michael Hedden Jr. wrote on Tue, 13 February 2007 16:06 |
I agree that overly processed data isn't a good thing but I don't know anyone that would consider 16% overly processed. Many manufacturers present data with 33-50% smoothing and you are correct use a big enough crayon and you can really smooth the curve. |
Michael Hedden Jr. wrote on Tue, 13 February 2007 16:06 |
A much more significant point than post processed smoothing is the actual frequency resolution of the data. Danley's subs are measured with a frequency resolution of 3.5 Hz and displayed with 8192 data points. So for comparison's sake what is the frequency resolution and how many data points where used in the Tulsa shootout data? |
Quote: |
...the Trip’s THD was 39.5%, Fwiw, the TH-115’s was 14% and the TH-215’s was 8.8%... |
Tom Danley wrote on Tue, 13 February 2007 20:14 |
Hi Bink While the spectra is unknown in this case, normally with a woofer or simple mechanical oscillatory systems, the third harmonic is usually the dominant one. It is true that with music (but not non-harmonically related sounds), that even harmonics are more tolerated being musically related intervals, (even preferred in some cases) compared to odd and the second has to be pretty high to be audible. Without masking sound (such as in a movie scenes where there is LF but little else) , Dolby labs found that because of ones ears decreasing sensitivity, that one could hear lf distortion ?too easily? down low. For example (one data point I remember) with a 20Hz sine wave (worst case), a third harmonic of only 7% was judged to be equally loud as the fundamental as a result of your ears being so much less sensitive at 20Hz as they are at 60Hz. Thus they concluded, making subwoofers with a low cutoff AND inaudible distortion was a very daunting task. Best, Tom |
Tom Danley wrote on Tue, 13 February 2007 19:23 |
Wayne, a couple things, the more you stretch out the frequency axis, the smoother any curve looks. Your curves cover from 20-20K, nice for a full range speaker but at least the top 5 octaves are pretty much undesirable so far as a subwoofer plot. Also some Smoothing is actually desirable for two reasons. First, the “grass” and tiny steps in your measurement are not part of the subwoofers output. These will generally go away up close where you have a better Signal to noise ratio or add smoothing. For example, in your posted curve, your noise immunity appears to fade away around 80-85dB, the grass is noise, not the speaker but being uncorrelated noise, would be effectively smoothed. |
Tom Danley wrote on Wed, 14 February 2007 01:23 |
The 28 Volt harmonic distortion differences are interesting and a trend likely adding to the impression of the Trip’s extreme loudness at very high powers. At 28 Volts (at 262 Watts into the Trip, well down from the 6KW program rating), at 32Hz where harmonics mostly fall unmasked in the subwoofers own frequency range, the Trip’s THD was 39.5%, Fwiw, the TH-115’s was 14% and the TH-215’s was 8.8% |
raj sookraj wrote on Tue, 13 February 2007 16:35 |
Did we record the nominal load shown on the Powersoft amps for all the subs? Also, can anyone explain how they were wired to the amps. I think someone said that most pairs of subs were on one channel. |
Ivan Beaver wrote on Wed, 14 February 2007 06:26 |
I thought I would post a shot of a 10% smoothed response and a non smoothed response, just so you can see the "big difference" it makes. The reason it looks smoother than your measured responses is because of the resolution used. We used a 16.6 second sweep that used 1024 samples over a much narrower bandwidth, resulting in a 2.5Hz resolution. This was the measurement taken at the shootout and is of a TH115. I used the same freq bandwidth as you did, but the trace does not go as high, because we did not sweep it as high as you did. The TEF preserves all the data and has no smoothing when taking the measurement. This allows you to display the data any way you want to later on, as done here. |
Mac Kerr wrote on Wed, 14 February 2007 10:31 |
The lack of small detail in the smoothed plot doesn't misrepresent the speaker's response, nor does the small detail in the un-smoothed plot change the overall impression of the response. ...but the final purchasing decision should be made with your ears. Mac |
Mac Kerr wrote on Wed, 14 February 2007 10:31 |
The lack of small detail in the smoothed plot doesn't misrepresent the speaker's response, nor does the small detail in the un-smoothed plot change the overall impression of the response. |
Wayne Parham wrote on Wed, 14 February 2007 13:57 | ||
The problem is when you compare a smoothed response curve of one speaker with an unsmoothed curve from another speaker. Someone proposed comparing a response curve of their subwoofer that had been post-processed for 1/6 octave smoothing to a response curve of another subwoofer that had not been smoothed. The difference is pretty significant. |
Tom Danley wrote on Wed, 14 February 2007 16:10 |
Hi Pascal, Wayne Perhaps you did not read my post carefully, I said I have converted the data for people to look at / use / view without a TEF machine. |
Quote: |
You and David pushed the idea of “how loud” the Trip goes using very vague specifications not backed up by measured curves yet you squeal “cherry picking” when after some considerable time I used actual measurements to counter your specific argument. |
Quote: |
Pascal, clearly you have a personal issue with us /me /our company but it would also seem to be based on little first hand experience or actual acoustic data. Our data is repeatable, done in a proper way and some of it is even taken by an independent acoustic measurement company, so where is the beef? |
Mac Kerr wrote on Wed, 14 February 2007 13:07 |
There are more compelling reasons why this data cannot be compared to your non-smoothed data from Tulsa. These plots can only be compared against each other. The data will quantify what we heard or didn't hear at the listening session, in the same room as the measurements were made. |
Tom Danley wrote on Wed, 14 February 2007 15:07 |
Given how much animosity measurements and data seem to cause perhaps and the difficulty putting it in context, it would have been better to only listen to them. |
Tom Danley wrote on Tue, 13 February 2007 19:23 |
Hi Dave, Pascal, Wayne. I have done a data conversion of some of the data into more user-friendly files which I will send on to Mark, Ivan and Mac shortly. |
Quote: |
Unfortunately inside isn’t like outside, there are a lot of room effects and a room full of unpowered speakers is not even the same as an empty room. Since it wasn’t at a meter or in half space, the 1w data was pretty uninteresting so I didn’t do anything with it. I took the 28V measurements and converted them to Text files (freq, mag, phase) and Jpg files for each box. The text files can be loaded into many different programs for further doodling. Since most of the boxes were 4 Ohm loads (and wanting to minimize the work), I up and down converted the measured levels by the appropriate amount depending if it were a 8, 4, 3 or 2 Ohm nominal load and saved these as “power-normalized”(Relative to the 4 Ohm boxes which are not changed) curves as JPG files for each. This way the level is shifted to simulate an equal power into each box’s nominal load Z. Without seeing each color, I figured two curves per graph would be safe so I also paired the boxes up by size ( by box volumes on the spread sheet) and plotted power-normalized responses in pairs, saved as JPG images. I also converted the THD files for each box at 28 Volts to Jpg files. David, the curve below is what one would get in that room, in that space with each of these three speakers had they been driven at the same power. |
Quote: |
This would reflect the power response in that location in that room but not much else. As you can see, the electro acoustic efficiency and frequency response is similar between the Trip and TH115 while the TH215 is less efficient but has a lower corner frequency. |
Tom Danley wrote on Wed, 14 February 2007 10:10 |
Hi David, why not hire someone with a TEF to measure your products so people can see how they perform? Also, are we your real competition, do you really want a shoot out with us to see how just loud or low “one box” can go? Tom Danley |
Tim Morin wrote on Thu, 15 February 2007 18:49 |
dave, when you said that only you and Tom can play on this level, did you mean as manufacture or as engineers? either way seems quite offensive to assume that |
Michael 'Bink' Knowles wrote on Mon, 19 February 2007 15:22 |
That middle chart looks like your relative -6dB point is down at 44Hz. Your +0/-6dB points (aka +/-3dB points) are 44-92Hz outdoors in half space for one box and about 43Hz to more than 120Hz for two together. Am I reading it right? Smooth. -Bink |
Ivan Beaver wrote on Tue, 17 April 2007 07:23 |
Tom Danley had sent results to "The powers that be" on this forum quite a while ago so they could post. But for some reason they have not. I don't know why or why not. If someone doesn't post in awhile I will post them. But right nowit is not my place. |
Mac Kerr wrote on Tue, 17 April 2007 11:36 |
]I am one of those people. and I am not willing to post results of a shootout where the plots were matched by one of the participants. I am not impugning the integrity of anyone, but would any manufacturer be happy knowing that publicly posted shootout results were modified in any way by their competitor? Originally Mark Seaton was going to make some comparison plots and post those. What I have are individual plots that Tom Danley adjusted to to make the power levels comparable. There are 72 individual files, and we do not have an ftp site for transferring that much data. The data would be easier to compare if some comparisons were made, and the data should be looked at by someone without a horse in the race. Mac |
Mac Kerr wrote on Tue, 17 April 2007 11:36 |
I am one of those people. and I am not willing to post results of a shootout where the plots were matched by one of the participants. I am not impugning the integrity of anyone, but would any manufacturer be happy knowing that publicly posted shootout results were modified in any way by their competitor? Originally Mark Seaton was going to make some comparison plots and post those. What I have are individual plots that Tom Danley adjusted to to make the power levels comparable. There are 72 individual files, and we do not have an ftp site for transferring that much data. The data would be easier to compare if some comparisons were made, and the data should be looked at by someone without a horse in the race. Mac |