Any reason not to go to DEQ2496?

[Otto]

Hi guys,

I've been considering moving from the BFD1124p to the DEQ2496. I've read a number of the "what about the DEQ" threads.

In general, the 1124 has been good to me, but I'm looking for a couple features that I think can be had in the DEQ2496.

I enjoy listening to 2-channel music, and I've been using "bypass mode" for this. In bypass mode, both my mains and my sub receive the same, presumably unaltered, analog signal -- no bass management is applied. My mains are rated to 30 Hz, and they roll off somewhere above that, of course. I want to use the DEQ as a low pass filter, such that I can integrate my sub correctly with my mains (when they are running full range). I would use the "shelf filter", perhaps in conjunction with the graphic EQ to achieve a low pass filter. Any reason why this wouldn't work?

The shelving filters are offered at 6 dB or 12 dB per octave. And I believe the graphic EQ has a range of +/-15 dB. Does anyone know if these add when they are used concurrently? I think they should, but I have no experience with this unit.

I also like the concept of the dynamic filter application. Has anyone used this? I suppose I would have to take measurements at varying output levels and work to understand what type of response to set for those listening levels.

Has anyone felt limited by the bandwidth appliable to the PEQ? The BFD offers 1/60 octave steps while the DEQ "only" offers 1/10 octave steps. I don't use many filters with a really tight bandwidth, but sometimes it comes in handy -- probably because I'm trying to get my FR too flat.

I saw that there is a way to limit the amplitude range when using the RTA, but does anyone know if there's a way to limit the frequency range on the RTA? I think not; their FFT is probably set up to do the 61 bins in a one-way-only fashion, but that would be cool if you could see 61 FFT bins from 10 Hz to 100 Hz. Guess that's why we have REW.

Any comments from users? Does it sound different from the BFD?

I like the increased dynamic range.

I like the display better.

I'm OK to enter filters by hand.

[terry j]

Otto, I didn't answer before cause I didn't fully understand your questions, and thought I could get clues from the answers of others ha ha. But, for someone as helpful as you towards others to have your question float lonely in the ether simply meant I must expose my ignorance for the whole world to see!! ha ha ha.

I have absolutely no experience with either of the other two BFD's, so unfortunately I can't directly compare the sound of the units. My gut feel is that the deq would be cleaner and more transparent, but having just written that I have no idea why that should be so, maybe it has been written elsewhere and I have a vague memory of it, but as I say I personally wouldn't know that one way or the other. Humph, maybe YOU will be the person to clear that up for us??

I have mucked about with integrating a sub with mains as you describe above, and yes it does work. From memory though, I didn't particularly use the shelving filter, don't recall exactly why not but think it was kinda due to the slope on the shelving filter was shallow or something. Anyway, by combination of geq and peq was able to achieve a relatively sharp lowpass to the sub. My mains are flat to 29 hz as measured from the listening position, and so it was simply an experiment to see whether or not I needed a sub for the last little bit down low. The experiment by the way was a little indeterminate, if the sub was between the mains (I'm on a wooden floor) then it seemed to be on the same joist as my seat and the rumble was often not in synch with the mains (ie the rumble did not integrate well), and if the sub was behind me I could tell it was behind me. Both were horrible. Anyway, that lead to another feature of the deq that you may not have considered yet, integrating it as I did and as it seems you might, allows the sub to be further away than the mains, as the signal to the sub can be delayed by the deq and therefore time aligned.

Dynamic steup I haven't played with, quite frankly the manual sucks (for me) and my background knowledge is not sufficient to work it out.

In practice, the few systems I have used it in to equalize the bass the ;limited bandwidth of the filters has not been a problem...there is always the first time I suppose!

I think what you where asking about fft bins was it would be good if all the 'sliders' on the unit where concentrated down low?? As you say, they are not. But with the combination of what is available with both the geq and peq in the area of interest a pretty good job can be done.

If you where over this side of the lake I would gladly lend you mine to evaluate, but a bit impractical methinks as it stands.

My gut feel. As you will only be using it for below 50 hz say, then I think a lot of what you pay the money for will not be utilized, so you may be better off using the more high quality unit of the bfds that are available. This of course is leaving aside the question of "is the deq more high quality than either of the bfd's", an answer I can't give as mentioned before.

From memory, you have a set of Vandersteens?? I know nothing about them, but if you were to use the deq then it is of course possible to use them on the mains as well, and correct the whole shebang. There is a limit to the quality of the system that you would use the deq on, wheteher that limit is reached with the vandersteens I don't know.

By the way, you wouldn't use the automated functions of the deq down low, simplt run sweeps in REW and tweak manually, then run sweeps again.

I am more than happy to run tests on specific questions if you wish, have been tempted to do that a couple of times in the past, need to take it all over to the other system and set it up. Could run the unit straight thru the rew setup to check it's electrical response, as Brucek has done with mics etc. That for example will tell us whether the boosts add as you asked above.

Thats enough for now, let me know if that has helped or not and whether you would like me to do whatever tests that would help. I am not super clever on that kind of stuff, but if it is the fashion of 'monkey see monkey do' then I guess I would be a pretty good monkey!

[Otto]

Hi terry,

Thanks for your response. It sounds like you've had some success with the device overall.

I also have a gut feeling that the DEQ might have a cleaner sound to it, but I'm not sure why I feel that way. It may be hard to tell when EQing sub levels only. I did use a peaky gain filter the other day, and there was terrible audible ringing, even though my sub measured flat. I forgot to look at the waterfall plot; it would probably have shown what I was hearing. It wasn't subtle; it was bad. So I wonder if these two devices implemented different types of digital filters such that they would have a different sound. I guess I may end up being the guy to try to rate the differences between the two.

I think you're right about the shelf filter being a little shallow, and also correct that you can make a resonable cutoff by using it in conjunction with the GEQ and/or PEQ.

Yeah, the Vandersteens are probably revealing enough so that I wouldn't want to try to EQ them. Of course, once I get the DEQ in place and everything running, there would be no harm in doing a test to see if it improves things over the full frequency spectrum.

No tests in mind at the moment, but if you come up with any ideas, go for it!

Guess that's it! Time for work...

[brucek]

I did use a peaky gain filter the other day, and there was terrible audible ringing, even though my sub measured flat.

So if you had a peaky gain filter engaged and it measured flat in the room, then at the point of the peaky filter, the room was at a dip of equal intensity. Doesn't matter which device you use, adding gain like that is not a good idea. You may well have run out of headroom in the DSP.

Of course, once I get the DEQ in place and everything running, there would be no harm in doing a test to see if it improves things over the full frequency spectrum.

I think you already know the results of that experiment..

brucek

[Otto]

Definitely agreed on both points.

On the former, I had applied a small gain, and it didn't do much, so I kept turning it up (knowing I really wouldn't leave it that way). I guess I was a little surprised to hear that ringing, though. I don't think the DEQ would do a better job with that type of filter...

[Wayne A. Pflughaupt]

Otto,

The shelving filters are offered at 6 dB or 12 dB per octave. And I believe the graphic EQ has a range of +/-15 dB. Does anyone know if these add when they are used concurrently?

If your asking of one will affect the properties of the other, the answer is yes.

Has anyone felt limited by the bandwidth applicable to the PEQ? The BFD offers 1/60 octave steps while the DEQ "only" offers 1/10 octave steps.

Not a problem. There is no good reason to equalize a sub with filters smaller than 1/6-octave.

Any comments from users? Does it sound different from the BFD?

With subs, there’s really no way that it could. Any audible difference would be with the mains.

I agree with Terry - as long as this is for the sub, there's no functional reason to pay the extra money. If you're smitten with the display - that's another thing altogether.

Regards,
Wayne

[Otto]

as long as this is for the sub, there's no functional reason to pay the extra money.

Hi Wayne,

Yeah, it's for sub only. The thing I'm looking for is a clean way to make a LPF, so that I can disengage bass management from my preamp. I think I could do this more effectively with the DEQ as opposed to the BFD because of the GEQ and shelf filters in the DEQ. Whaddya think?

[brucek]

The thing I'm looking for is a clean way to make a LPF

Just make an LPF with some BFD filters.

brucek

[Otto]

Yeah, I've played around with that. I just wanted something a little more. I don't know, in the end it'll probably net me only a little extra to go to the DEQ. We'll see; still on the fence.

[Wayne A. Pflughaupt]

Otto wrote:

Hi Wayne,

Yeah, it's for sub only. The thing I'm looking for is a clean way to make a LPF, so that I can disengage bass management from my preamp. I think I could do this more effectively with the DEQ as opposed to the BFD because of the GEQ and shelf filters in the DEQ. Whaddya think?

A shelving filter doesn’t act like a low- or high-pass crossover filter, where response is 12 dB down an octave out, 24 dB down two octaves out, and so on. It’s merely another type of EQ filter, and as such its action is fully dependent on the amount of boost or cut you apply. It works like this: From flat response with a 0 dB reference line, if you apply a 6 dB cut, response drops 6 dB, and then is flat again below that point. In other words, it “plateaus” response, not endless drops it.

As such, describing a shelving filter having a 6- or 12-dB slope, as Behringer does, is just wrong. For instance, how can you have a true 6-dB slope if the filter is adustable +/- 15 dB?? I dunno, maybe the DEQ locks down the amount of boost or cut when you engage the shelving, or maybe the 6- or 12-dB reference has to do with how steep the “hinging” action at the center frequency is. The manual doesn’t say much about it either way.

Regarding mimicking a shelving or low-pass filter with the BFD or DEQ, you may find that difficult to accomplish.

Recently I decided that EQing my sub flat wasn’t getting it for me, so I thought I’d try my buddy brucek’s famous house curve filter to get me about a 6 dB or so reduction at 100 Hz. Well, it didn’t go too well.

I established a reference 32 Hz SPL reading, to make sure my filter wasn’t making any changes at that point – it was only supposed to affect above that. But nothing I tried worked the way I needed it to. I first started with brucek’s recommended two-octave 366 Hz filter cut 15 dB, but that didn’t get it. I forget exactly what it was, but I think that it wasn’t getting me enough cut at 100 Hz. Deepening the cut, when I got what I wanted at 100 Hz, I found that it had affected all the way down to 32 Hz – my reference SPL level had dropped. I trued moving the frequency up and down, different (mostly tighter) bandwidth settings and cut values. I eventually gave up and totally re-equalized for the house curve. I guess the problem was that according to brucek’s predicted filter response picture (in my house curve sticky thread), the house curve filter only gets about a 4 dB reduction at 100 Hz. If you need something different, you’re on your on. You’ll pretty much need a real-time display to figure it out (or whatever that nifty program is that brucek created that picture with!).

Based on that, I’d hate to try to figure out a shelving filter without something like TrueRTA to let you see exactly what’s going on. Keep in mind too, that by the time you achieve a 12 dB slope, you’re probably going to end up with a fairly tight filter, and it will only be a maximum of 12 dB down at the center frequency. Going deeper, or trying to achieve a steeper slope is going to require an even tighter filter. This is especially problematic since you want to hinge at 30 Hz or so: You’re probably going to find that by the time you get to 70 Hz or so, everything will be back up!

Anyway you cut it, Otto, you just can’t create a decent low- or high pass filter with an equalizer, especially at the low frequency you want to. If that’s ultimately what you want, I suggest just getting a real crossover. Behringer has some cheap ones. I might suggest moving up to something like the Rane AC-22

Regards,
Wayne

[brucek]

You’ll pretty much need a real-time display to figure it out (or whatever that nifty program is that brucek created that picture with!)

But you already have a real time display program to figure it out.

Start REW and turn on only Target and Filters and start creating whatever filters you want to track a target. When you're done enter them into your BFD and you're done. You just do it sitting at your computer without hooking anything up to your equipment.. It's as dynamic as you can get. WYSIWYG.

Here's a two second job creating a 30Hz low pass filter target tracking REW's 24dB/octave 30hz target. I think it tracks the target fairly close. I only went up to 300Hz - I figure the sub is down naturally by then....

LPF.jpg

brucek

[Otto]

Thanks for the discussion guys; it's the kind of thing I'm looking for.

About the shelf filter: I think it's lame that it's not well defined in the Behringer manual. When I read it, I didn't question the definition of "shelf filter". It's not somthing that I've really run across before, but the 12 dB per octave implied to me that the slope continued infinitely in whatever direction. So, Wayne, you're saying that the shelf filter has a slope that starts as some frequency f, and is applied for over some frequency range ([/i]r[/i]), at which point the remainder of the frequency spectrum (ideally, until infinity) remains simply cut by some amount, G. How big is r? An octave? Something else? If r is one octave, then the absolute value of G will always be either 6 or 12 dB.

I guess it doesn't really matter too much, but I'm disappointed to hear that it doesn't work the way I thought it would.

I think I will consider a "real" crossover. Any suggestions for one that works down to, say, 10 Hz? With a cost of < $300? And a variable phase control? I'll check out the Rane later today.

As to making LPFs with the BFD, I agree with brucek that it's possible. I've played with it to some degree of success. I'll continue to poke around at real crossovers and such. Perhaps I'll go back this weekend and see if I can do better than what I have. FWIW, my sub will play >> 300Hz, so I'm up in the 2 kHz range when I'm quashing that sub output. Still, it seems to work OK. I guess I just wanted a definitive LPF on that thing.

Thanks!!!!

[brucek]

I guess I just wanted a definitive LPF on that thing.

OK, I'm thinkin' that you're just looking for a new toy (not that there's anything wrong with that).

The BFD with about 6 filters makes a 'definitive real' LPF. That leaves 6 filters to play around with for eq. Unless you buy an analog crossover, the new toy will employ the exact same DSP math that the BFD uses to create a filter set.

I use this LPF trick on my second system. It's an old analog processor that has a full range mains and sub out that I require a very low LPF on the sub to integrate properly with the sub. The sub LPF on its own amplifier doesn't go anywhere near low enough for my liking. I load the LPF filters in the upper 7 to 12 BFD slots and then don't touch them after that. Any eq is done using the lower filters from 1-6....

brucek

[Wayne A. Pflughaupt]

I stand corrected, guys – thanks.

Otto wrote:

So, Wayne, you're saying that the shelf filter has a slope that starts as some frequency f, and is applied for over some frequency range (r), at which point the remainder of the frequency spectrum (ideally, until infinity) remains simply cut by some amount, G. How big is r? An octave? Something else? If r is one octave, then the absolute value of G will always be either 6 or 12 dB.

Here’s what a shelving filter looks like:

All I can figure out with Behringer’s 6/12 dB thing is that one (presumably the 12 dB) would maybe make the transistion between where it starts and levels out sharper or more “sudden.”

I think I will consider a "real" crossover. Any suggestions for one that works down to, say, 10 Hz? With a cost of < $300? And a variable phase control? I'll check out the Rane later today.

Not sure what you mean by “works down to.” I probably led you astray recommending a “real” crossover – I forgot that it’ll probably be hard to find one with a crossover frequency lower than 40 Hz, since most pro subs are pretty much done by then, so the BFD might be your best bet after all.

Still, it can’t hurt to look. I have these brands in my “Pro Audio Manufacturers” Favorites: Applied Research & Technology (aka ART), Ashly, Behringer, Carvin, dbx, ElectroVoice, Furman, Peavy Phonic, PreSonus TDM, Yamaha. There is also the “DJ brands” (read low-end) like Nady, DOD, Pyle, Samson

Regards,
Wayne

[terry j]

was mucking around a bit today, and perhaps this data will be of use to some.

First of all, the sub I quickly threw together was the thor of seigfreids, I was hoping to get flat to 20 in room but fell a little short. Like Otto, my mains 'only' go flat to 30, and I was curious about the last little bit below. The following graph shows the raw, and then eq'd response to about as low as I could get it.

raw sub, plus flat to 20.jpg

(hope I can make sense of my poor notations!!!!)

Then, taking my needs as an example, I aimed for a roll off from 30 hz. First up I used the geq of the deq 2496 to cut the bands at 40, 50 etc etc by the maximum amount, namely -15 db.


sub to 20 plus -15 above 40.jpg

As we can see, and as can be expected, the signal drops by (only) 15 db. In fact, whilst the sweep is in progress I can still hear it past the cutoff point we wanted.

And in any case I don't think using the geq in this manner yields a satisfactory low pass filter, unless -15 db is acceptable.

Next I looked up Bruces recipe for a low pass filter. Bit of a pain really, had to enter all the figures into the filters panel, then convert them to the form used by the deq! By the way, the maximum cut allowed in the deq peq section is -15, so it is not an exact match of Bruces but nonetheless the best we can get.

flat to 20 plus bruce low pass.jpg

The graph using Bruces recipe is the darker one, and strangely enough, the result is worse than by just using the geq function.



So, it seems that Waynes take on the 'shelf' filter is correct, the frequencies in question are simply attenuated by 15 db all the way up, which is not the same as a low pass function.

OF COURSE, it is perfectly possible that I did it all wrong so let's bear that in mind!!! ha ha ha.

Otto, seeing these results I feel for what you want to do the DCX 2496 may be more suitable???, it will after all give a genuine low pass function with peq as well.

Give me a little while and I'll test that as well, I have one here I need to nut out and this is as good a project as any to learn on eh??

I know you were looking for a second hand unit Otto, but when new the two are comparable in price??

sorry, looking at the graphs I first aimed for a cutoff of 40 hz, it was only after seeing Bruces formula that I realised I should have gone for thirty (to compare). Forgot to save the second set of measurements, can redo it if needed. So the cutoffs don't exactly match, but hope there is enough here to tell the story.

[Otto]

Hi terry,

terry j wrote:

First of all, the sub I quickly threw together was the thor of seigfreids,

You've got me! What the heck does that mean?

The following graph shows the raw, and then eq'd response to about as low as I could get it.

So did you boost around 20 or 25 Hz to get that frequency response reasonably up at 20 Hz? Seems like a lot of boost. Any problems with clipping or weird sounds?

Then, taking my needs as an example, I aimed for a roll off from 30 hz. First up I used the geq of the deq 2496 to cut the bands at 40, 50 etc etc by the maximum amount, namely -15 db.... As we can see, and as can be expected, the signal drops by (only) 15 db. In fact, whilst the sweep is in progress I can still hear it past the cutoff point we wanted.

I like it so far.

Mine also goes rather high. When using the BFD to make an LPF, I also had to quash response at rather high frequencies -- up to 2 or 3 kHz, IIRC.

And in any case I don't think using the geq in this manner yields a satisfactory low pass filter, unless -15 db is acceptable.

Agreed, especially if you can still hear it at a "reasonable" level.

[quote]Next I looked up Bruces recipe for a low pass filter. Bit of a pain really, had to enter all the figures into the filters panel, then convert them to the form used by the deq! By the way, the maximum cut allowed in the deq peq section is -15, so it is not an exact match of Bruces but nonetheless the best we can get.

Did you use a 4th order target slope? I think that's what brucek was doing, and it's a reasonable approach. That said, I took a slightly different approach as well: I left the mains on for all measurements, and just tweaked the BFD as needed to get a decent FR, fully integrated with the mains from the get-go. It's a very iterative process, but my wife really loves the sweep tones. (Actually, she doesn't mind).

That is a strange one. I would not have expected it to bump back up like that...

One idea -- before applying any PEQ, would it be useful to use on of the shelf filters? That's pretty much what you've done by setting all the GEQ gains to their minimum (i.e., max attenuation). Adding another group of global cuts via a shelf filter should push things down farther, even if it's "only" 12 dB or whatever.

It appears this is where the max attenuation of the BFD (-48 dB, IIRC) is more powerful and useful than that of the DEQ2496. When I hit those signals at 1 kHz and 2 kHz with overlapping (BW = 120) filters, they're inaudible and unmeasurable from the listening position.

I've thought of it...

[quoteI know you were looking for a second hand unit Otto, but when new the two are comparable in price??

Yeah, I kinda figured that. Even watching them on eBay, the price really does approach the new price. I'll probably just get a new one from Amazon if I do it (got a $100 Amazon gift certificate lying around).

sorry, looking at the graphs I first aimed for a cutoff of 40 hz, it was only after seeing Bruces formula that I realised I should have gone for thirty (to compare)... So the cutoffs don't exactly match, but hope there is enough here to tell the story.

Yeah, your rolloff is very steep compared to a 4th order filter (at least in the 30 - 40 Hz region). Looks like about 24 dB per 10 Hz from 30 Hz to 40 Hz, and looks like you're about 6 dB down in the mid-30s.

Did you do anything with it as far as integration with the mains? Any listening?

Good post!

Thanks!

[terry j]

HI otto

re the thor, it's from the linkwitz site, simply a peerless 12 inch in a sealed box. Had one lying around doing nothing, so whipped up a box real quick. Obviously he uses a Linkwitz transform on it, me? just threw in eq !!

Yes, was quite a bit of boost needed down AT 20 and 25 hz, funnily enough I didn't seem to get much help when i threw it against the wall, wonder why??

Re listening impressions, I much prefer it not being in the system quite frankly. I can get my mains very nicely flat to 30 hz, put the sub in and the response goes all screwy. ????

So, the reason for the little experiment in the end was to answer some of the questions that came up in this thread re shelf filters etc.

By the way, it is still all set up if anyone has specific requests so to speak, I will do my best to get them answered he he he.

I didn't play them at anything approaching loud levels, but my gut feel is that i would be hearing strange noises that would worry me

To be honest, I do feel the attenuation provided would not be enough to completely stop extraneous noises appearing, and the graphs seem to show that. As I said, I could quite clearly still hear the sweep as it went on up past the 'cutoff' point, and whilst it may not be particularly audible with music, it nonetheless is still there and so must count as unwanted 'noise'.

It may not have been clear, but the 'formula' of Bruces I used was the one he posted earlier in this thread. It doesa seem that some of the other BFD's do have deeper cuts available, which of course would lead to better results.

This particular sub i used did not seem to have a very high FR, in any case i ran a splitter feed of my mains bass, which are cut off at 300 hz, yet the sub did not seem even close to going that high. And you are getting signals way above those frequencies!!

Didn't fully understand some of your post, but that's Ok because if you have a specific request that you'd like me to do then we'll give it a go, eg if you want me to throw in a shelf filter...but I thought that was basically what I did by putting maximum cut on all frequencies above 30 hz. Hope this response made sense, it's late over here and with a couple of beers under the belt.

[tomacco]

Otto wrote:

Hi guys,

I've been considering moving from the BFD1124p to the DEQ2496. I've read a number of the "what about the DEQ" threads.

In general, the 1124 has been good to me, but I'm looking for a couple features that I think can be had in the DEQ2496.

I enjoy listening to 2-channel music, and I've been using "bypass mode" for this. In bypass mode, both my mains and my sub receive the same, presumably unaltered, analog signal -- no bass management is applied. My mains are rated to 30 Hz, and they roll off somewhere above that, of course. I want to use the DEQ as a low pass filter, such that I can integrate my sub correctly with my mains (when they are running full range). I would use the "shelf filter", perhaps in conjunction with the graphic EQ to achieve a low pass filter. Any reason why this wouldn't work?

The shelving filters are offered at 6 dB or 12 dB per octave. And I believe the graphic EQ has a range of +/-15 dB. Does anyone know if these add when they are used concurrently? I think they should, but I have no experience with this unit.

I also like the concept of the dynamic filter application. Has anyone used this? I suppose I would have to take measurements at varying output levels and work to understand what type of response to set for those