Subwoofer Comparison

[beaRA]

Well when I said 'bass management system' I was referring to automated systems eg Audyssey, Trinnov Optimizer, Dirac Bass Control, etc. I'm not super familiar with the Geddes approach, I've only read a couple of short papers from him. What I read indicated that it would be unsuitable for multichannel audio and since I don't want two completely different signal chains in my system I discarded it.

I did some recent Googling and I couldn't actually figure out one source that fully explains Geddes on subwoofers. There are various short papers written by him, and some videos, and they all seem to contain small contradictions in methodology and requirements. For example in one of the early papers he says that EQing the mains based on room data should never be done(at what frequency range? My room contains issues at 100-500hz that actually exceed the low bass problems, and EQing them audibly fixes a lot..). But then in this video he states that not being able to EQ mains was a limitation imposed by the customer's setup...?

Geddes states that the mains should take the full signal and he doesn't seem to address the possibility of them being overloaded as a result. Some other sources indicate he assumes large, high output mains. In multichannel the LFE signal is way too loud for it to be realistically mixed into the mains unless yours have the same output as true subs. Also, as he notes in the video, the actual levels that each of the bass sources will play is unintuitive and calculated by his computer program. But there's no way to predict it in advance, and it varies by huge amounts! How can you compensate with a single gain setting if one of your small subs is tasked to play 20hz at a high level? The page I linked above states that you could use subs of different size by placing the smaller ones closer to the listening position, and that may be true on average, but it ignores the fact that the filters could easily render small listening distances differences irrelevant.

Finally, I think you need to send your measurements to him to actually get the real solutions? According to the MSO page listing approaches, "Despite discussion in that thread spanning twelve years, he's never described specifics of how he computes the per-sub filter parameter values."

This all seems really complicated, confusing, and lacking information. MSO supports an optimization strategy using the Geddes approach(in the most high-level sense) and it addresses various additional complications including the dissimilar subwoofer issue. They have you high-pass your mains if they're not "of closed-box type, with high power handling and an anechoic low-frequency cutoff of 80 - 90 Hz".

Dirac claimed that their Bass Control would eventually be able to use mains as supplementary bass sources, but as far as I know that isn't working yet. In theory it should be possible for software to programmatically measure your mains limitations and tailor the resulting matrix of EQ solutions for each bass source to match.

Honestly my long term plan is to just get a processor with Dirac

I agree with you that this approach seems far more complicated than the Welti approach of symmetrical placement in a sealed rectangular room, but it does seem more flexible for weirdly shaped rooms. I never got the impression that he thinks this would not be appropriate for multichannel content. He actually talked about using this approach for his own home theater. For others looking for a good summary, @Matthew J Poes did a video on the AVNirvana channel. I also found this article from miniDSP helpful.

 

[Sancus]

I think this article is a good practical explanation of how to setup subs the Geddes way with some of the explanation as well.

OK, but this leaves out the actual hard part which is the optimization of each step. Doing it manually isn't going to produce the best result. Apparently Geddes advised this in some of his earliest posts, but then later said you need a computer program to do it and started joking about the manual procedure, "Depending on the number of available degrees of freedom this could be an arduous task.(That’s why I developed a computer program to do it!) But heck, DIY’s and audiophile’s will do anything for the best possible sound. Right!?"

I agree with you that this approach seems far more complicated than the Welti approach of symmetrical placement in a sealed rectangular room, but it does seem more flexible for weirdly shaped rooms. I never got the impression that he thinks this would not be appropriate for multichannel content. He actually talked about using this approach for his own home theater. For others looking for a good summary, @Matthew J Poes did a video on the AVNirvana channel. I also found this article from miniDSP helpful.

The problem is that I didn't see how you could run the summed bass signal through mains. It can be as high as 123dB SPL, which is going to make any normal speaker unhappy at <120hz. The key piece of knowledge is the strategy isn't designed for any random speakers, it's for his sealed compression driver speakers with huge woofers. It isn't optimal for any random speakers, especially ported ones. You have to high-pass at 80-100hz. I still think this could be problematic with some speakers frankly, as the LFE goes up to 120hz, but I'm not sure since I don't actually know what typical SPLs are above 100hz in the LFE channel. In any case, there is a sort of crossover being created here by the low passes and high pass, it's just not a normal one.

He actually addresses this in the video you linked and says "I want as many LF sources as possible... I have 15 inch woofers in the mains why not use them?" at about 24:00. He doesn't advocate using this for small or ported mains anywhere I've seen. He also doesn't say that running the signal through the mains is a REQUIREMENT, only that it works best if you have 15" woofers lol.

It's interesting he tried Multi-Sub Optimizer and didn't like it because it sets delays(which just makes no sense to me, there are subs with long DSP delays that you need to compensate for if nothing else), but well this video is 4 years old. And it produced quite some discussion about the validity of some of his claims. Honestly, I find his descriptions of how to do the optimization step for each sub to be so vague, contradictory, and changing in every source about them that it makes it almost impossible to understand.

There should be a paper describing how his software worked and how he advocates, at the current time(!), the exact way to optimize the frequency response for each sub. But there isn't. He seems to be using his own knowledge and various different pieces of software at different times to achieve it. Which I'm sure works just fine for him, but it isn't a general procedure you can follow for any setup.

As far as the Welti paper goes, it states the matrix of positions that were determined to be optimal in a rectangular room but it doesn't require you to use those positions or you can't get any benefit or anything like that. And it also advocates for a software-based solution. I still think Multi-Sub Optimizer is the way to go for a freeware solution. It doesn't lock you into any particular number of subs or positions either.

 

[beaRA]

OK, but this leaves out the actual hard part which is the optimization of each step. Doing it manually isn't going to produce the best result. Apparently Geddes advised this in some of his earliest posts, but then later said you need a computer program to do it and started joking about the manual procedure, "Depending on the number of available degrees of freedom this could be an arduous task.(That’s why I developed a computer program to do it!) But heck, DIY’s and audiophile’s will do anything for the best possible sound. Right!?"



The problem is that I didn't see how you could run the summed bass signal through mains. It can be as high as 123dB SPL, which is going to make any normal speaker unhappy at <120hz. The key piece of knowledge is the strategy isn't designed for any random speakers, it's for his sealed compression driver speakers with huge woofers. It isn't optimal for any random speakers, especially ported ones. You have to high-pass at 80-100hz. I still think this could be problematic with some speakers frankly, as the LFE goes up to 120hz, but I'm not sure since I don't actually know what typical SPLs are above 100hz in the LFE channel. In any case, there is a sort of crossover being created here by the low passes and high pass, it's just not a normal one.

He actually addresses this in the video you linked and says "I want as many LF sources as possible... I have 15 inch woofers in the mains why not use them?" at about 24:00. He doesn't advocate using this for small or ported mains anywhere I've seen. He also doesn't say that running the signal through the mains is a REQUIREMENT, only that it works best if you have 15" woofers lol.

It's interesting he tried Multi-Sub Optimizer and didn't like it because it sets delays(which just makes no sense to me, there are subs with long DSP delays that you need to compensate for if nothing else), but well this video is 4 years old. And it produced quite some discussion about the validity of some of his claims. Honestly, I find his descriptions of how to do the optimization step for each sub to be so vague, contradictory, and changing in every source about them that it makes it almost impossible to understand.

There should be a paper describing how his software worked and how he advocates, at the current time(!), the exact way to optimize the frequency response for each sub. But there isn't. He seems to be using his own knowledge and various different pieces of software at different times to achieve it. Which I'm sure works just fine for him, but it isn't a general procedure you can follow for any setup.

As far as the Welti paper goes, it states the matrix of positions that were determined to be optimal in a rectangular room but it doesn't require you to use those positions or you can't get any benefit or anything like that. And it also advocates for a software-based solution. I still think Multi-Sub Optimizer is the way to go for a freeware solution. It doesn't lock you into any particular number of subs or positions either.

I'll agree with you again that it's probably not a good idea for most people to use their mains as additional low frequency sources. The basic concepts still apply to someone trying to integrate 3 asymmetrically placed mismatched subs. The full optimization is certainly vague and it seems like Geddes found something that works for him but never managed to summarize it completely into a procedure that others could follow. MSO seems like a more complete realization of the mode manipulation approach that is actually usable for enthusiasts.

The Welti approach relies on specific placement to reduce spatial variation and global EQ to flatten the response. If we refer to Investigation #4 in the paper, the common single subwoofer placements are evaluated in configurations 1, 2, & 3. There are several multi-sub arrangements (4, 5, 8, 9, 14, 15, 16, 17) that provide little benefit or even worse consistency. This finding indicates that the approach actually does require you to use a very specific set of placements to see any benefit. He does suggest that more work should go into exploring the "multiple channel multiple receiver equalization technique" and it seems like MSO is filling in that gap now.

 

[richard12511]

At one time it was quite long, and it is in a enormous room. After re-organizing the furniture my listening distance has gotten very short so I'm starting to debate ditching the Funk and going to a much smaller pair of subwoofers as I never have the occasion to turn it up anymore... and since the Mouse House has taken everything over, movies that actually call for a lot of <30hz output have gotten very few and far between. Basically it feels like I have a 2,000 horsepower dragster parked in my garage, but I only need a commuter car.

To clarify I think that, their entry level subs aside, JTR does have an enormous advantage in performance-per-dollar. It's not even close. The RS1 is the only one that I thought that maybe it isn't true. And I did want to draw some attention to where a lot of your money goes when you buy a Funk, which the spreadsheet & your original post didn't really acknowledge.

I actually bought the Funk because it was, for what I needed, by far the best subwoofer available at the time that was measured on data-bass. Pre-dated the Rythmik F18 which is absolutely the direction I would go in today. Your spreadsheet is a huge help to the community - thank you for the good work.

I have gone down this road a few times on the forum and apparently nobody shares my opinion that less distortion, is always better - regardless of any audibility threshold. 0 is always the target. Things may be inaudible in a steady sweep but may become audible on music material. I hope more insight into IMD is done, standard tests are developed, etc. so we can explore that a bit more in the future.

I avoid talking to Nathan because the conversations tend to be so expensive so no, I was operating the assumption that the 18.0E is equivalent to the 18.0. Yes the cost has come down but I assume it's because the drivers have been made for quite some time and all the tooling is paid off, apprentice in the shop can build the square cabinets to standard, maybe everything is fixtured, the brand is doing better as a whole and pushing more volume, etc

Also I apologize for the general tenor of my posts. Under a lot of stress right now and I've gone off my Forum-Talk Baseline of "quite dickish" to "complete asshole." I'll be taking a bit of a break here until I can get my head on straight.

No worries. To be honest, I didn't even realize the 18.0E was a separate thing. If it really does equal the 18.0, that does represent a much better value than Funk subs usually are(speaking strictly of performance). How many do you have? Maybe you could send one to @hardisj and we could see . Seems like it would really cut into Funk's business if the $2,900 sub was identical(performance) to the $4,300 sub, though I'm sure many would still be willing to pay the upcharge for the curved cabinet(which is the main difference I see).

 

[temps]

No worries. To be honest, I didn't even realize the 18.0E was a separate thing. If it really does equal the 18.0, that does represent a much better value than Funk subs usually are(speaking strictly of performance). How many do you have? Maybe you could send one to @hardisj and we could see . Seems like it would really cut into Funk's business if the $2,900 sub was identical(performance) to the $4,300 sub, though I'm sure many would still be willing to pay the upcharge for the curved cabinet(which is the main difference I see).

I've just got the one sub. I am fortunate, my basement is somehow quite flat <100hz across the whole listening position so I couldn't ever convince myself to go for multiples.

It's the older 18.0SE... so maybe that's more representative of the 18.0E than the 18.0 measurements are? It'd be a bugger to ship. I could probably measure it myself, if I ever got the time. I have everything I need except a way to provide ~4,000 peak watts of clean power whilst parked in an empty field LOL so I could use some suggestions there. My truck generator is only good for about 400 watts.

 

[sweetchaos]

Improvements:
- added links that can be ordered from PowerSoundAudio.co.uk to 'my notes' section (of the spreadsheet) for the corresponding PSA subs

 

[richard12511]

I've just got the one sub. I am fortunate, my basement is somehow quite flat <100hz across the whole listening position so I couldn't ever convince myself to go for multiples.

It's the older 18.0SE... so maybe that's more representative of the 18.0E than the 18.0 measurements are? It'd be a bugger to ship. I could probably measure it myself, if I ever got the time. I have everything I need except a way to provide ~4,000 peak watts of clean power whilst parked in an empty field LOL so I could use some suggestions there. My truck generator is only good for about 400 watts.

From looking at the CEA specs of the older vs newer models, the main difference seems to be that the newer models are much better down low. The 2013 version actually didn't pass the 10Hz and 12Hz sweeps(too much distortion to meet the spec).

 

[bogart]

@sweetchaos many thanks for maintaining this. It's just great. It's probably talking me out of getting a v1 Revel B110 for lack of data on how it performs, so... money saved!

 

[richard12511]

@sweetchaos many thanks for maintaining this. It's just great. It's probably talking me out of getting a v1 Revel B110 for lack of data on how it performs, so... money saved!

Where do you live? Revel makes SOTA speakers, but not subwoofers. There are many much better options at equivalent price. @sweetchaos datasheet is basically all you need to pick the best sub for your budget and situation . Let us know if you have questions.

 

[CtheArgie]

Hi "Posters"! Sorry for the very (possibly) naive question. I have a pair of wonderfully sounding EVE Audio SC 207. In my smallish office they sound terrific (and REW measure very well). I may have to move them to a larger room and there the bass will not be sufficient. So, I wanted to add a subwoofer to them but a/pair of SBs with high pass filters. I am then most likely to have to choose with pro powered subwoofers. I can have one SB, that is easy, but if I decide to have two smaller ones, how do I connect them? Do I use one as left and connect only the left channels and similar in the right? I have not found instructions to do so. Most appear to be only for 2.1 and I may prefer to go 2.2 for better sound in the room. Does anyone have any suggestions or comments? Most appreciated!

 

[sweetchaos]

Hi "Posters"! Sorry for the very (possibly) naive question. I have a pair of wonderfully sounding EVE Audio SC 207. In my smallish office they sound terrific (and REW measure very well). I may have to move them to a larger room and there the bass will not be sufficient. So, I wanted to add a subwoofer to them but a/pair of SBs with high pass filters. I am then most likely to have to choose with pro powered subwoofers. I can have one SB, that is easy, but if I decide to have two smaller ones, how do I connect them? Do I use one as left and connect only the left channels and similar in the right? I have not found instructions to do so. Most appear to be only for 2.1 and I may prefer to go 2.2 for better sound in the room. Does anyone have any suggestions or comments? Most appreciated!

2 scenarios I can think of:
1. If your subwoofer doesn't allow daisy chaining, like the JBL LSR310S...the setup should be similar to this:

This shows a single subwoofer connected to studio monitors. Now imagine the same picture with 2 subs. In which case, you'll be running a left signal to 1st subwoofer, and running a right signal to a 2nd subwoofer. In summary, you're running mono signal to each sub and then to each studio monitor.

2. However....If your subs allow daisy chaining, like the Neumann KH 750 sub can. Then you could run stereo signal to Neumann KH 750, and then daisy chain the stereo signal to the 2nd subwoofer (or 3rd,4th, if needed).

I hope someone else can confirm this works. This is strictly from my understanding of how studio subs work, since I only had home theater subs previously.

 

[sweetchaos]

Monoprice is having their “halfway to Black Friday” sale: 15% off site wide for today only, Thursday May 13. (In the U.S.) I’m sure some products are excluded from the offer. Use code HALFWAY2BF.

View attachment 129517

Monoprice subs are 15% off today!

 

[bogart]

Where do you live? Revel makes SOTA speakers, but not subwoofers. There are many much better options at equivalent price. @sweetchaos datasheet is basically all you need to pick the best sub for your budget and situation . Let us know if you have questions.

Like @sweetchaos I am in Canada. SVS seems to be our most affordable option here. I have an old Acoustic Energy Aesprit 308 sub (lots of blanks on the spreadsheet for that guy!) that is probably close to 20 years old now. It’s fine with a pair of Revel F206, but I definitely think about replacing it with a better pair of something.

 

[pozz]

Where do you live? Revel makes SOTA speakers, but not subwoofers. There are many much better options at equivalent price. @sweetchaos datasheet is basically all you need to pick the best sub for your budget and situation . Let us know if you have questions.

To be clear, we don't know how good Revel subwoofers are. There are no available measurements, unfortunately.

 

[pozz]

Like @sweetchaos I am in Canada. SVS seems to be our most affordable option here. I have an old Acoustic Energy Aesprit 308 sub (lots of blanks on the spreadsheet for that guy!) that is probably close to 20 years old now. It’s fine with a pair of Revel F206, but I definitely think about replacing it with a better pair of something.

Consider Arendal Sound. Canada has a treaty with Norway, so no duties on import.

 

[richard12511]

To be clear, we don't know how good Revel subwoofers are. There are no available measurements, unfortunately.

True, though I do think we can make reasonable estimates based on driver size and available power.

For example, the Revel B112v2, with 1 12" woofer and 1000 watts. That's $700 more than the Rythmik FV25, which gives you 2 15" woofers and 1800 watts. Also only $400 less than the JTR RS2, which gives you 2 18" woofers and 4000 watts. I don't see any way the B112 could compete with either of those subs.

 

[pozz]

True, though I do think we can make reasonable estimates based on driver size and available power.

For example, the Revel B112v2, with 1 12" woofer and 1000 watts. That's $700 more than the Rythmik FV25, which gives you 2 15" woofers and 1800 watts. Also only $400 less than the JTR RS2, which gives you 2 18" woofers and 4000 watts. I don't see any way the B112 could compete with either of those subs.

It's as you say in terms of price. Same goes for Genelec subs compared to others available, not considering other things. If we had actual data we could compare to the enclosure volumes to output, which would give the best indication (not that I see Revel winning that competition). Main thing that interests me is that I imagine the Revel drivers would show good performance elsewhere, since they design them internally using a Klippel system.

 

[sweetchaos]

Improvements:
- Added CEA-2010-A data for SVS 3000 Micro, provided by Brent Butterworth

 

[Kachda]

Improvements:
- Added CEA-2010-A data for SVS 3000 Micro, provided by Wirecutter

Can you post the link to the wirecutter review - I can't find it on their website. Thanks.

 

[sweetchaos]

Can you post the link to the wirecutter review - I can't find it on their website. Thanks.

https://docs.google.com/spreadsheets/d/1KjlXBWMPAHgU_V9DAHSdIZ_ke0MxwH6ekt3vTIlYOJU/edit#gid=0