Best sub EQ under $400 - Page 3

Is there a particular way to test if the anti-mode makes a difference that I can hear, as opposed to testing by graphs? Is there a test tone, or a particular song which would help me determine if the anti-mode has made a difference?

Yeh, set it up and listen to your favorite bass music or movie. Then, turn it off and listen again.

Bill

The best testing material is something that sounds horribly boomy and will wake up the most dominating modal resonances in the particular room before the 8033 Anti-Mode is used. To make comparing by listening easier, 8033 has "bypass" button, which can be used to toggle corrections on and off. This way it is easy to hear the difference, especially in the most problematic rooms and situations. You can also use test tones (sine waves) or sweeps to notice that the peaking room modes have vanished.

Of course using room response measurements where possible, it is quite easy to verify that the device works in all cases. But in the end of the day, the most important thing probably should still be the fact it makes audible difference, not the measurements themselves. The final benefit of using 8033 is hence dependent of how bad the situation was without it. The worse and more boomy the room was before, the bigger and more noticeable difference is achieved with Anti-Mode technology. For example if the room has a 15dB peak with really long decay time at around 45Hz, the difference will be easily the most concretic audible improvement to ones system by far. On the contrary, if you have a room with 4 feet thick mineral wooling on every wall and surface (basically anechoic chamber), probably the difference is not so audible at all, as there were hardly a problem within the bass reproduction to begin with. But as it happens in real life listening rooms, the the latter case would be extremely rare to even most dedicated HT and Hifi enthusiasts.

Email from UPS today that the 8033B has shipped from Finland and I can expect delivery on the 9th.

Bill

Mine came this morning. Shipped on Wednesday from Finland!! Amazing even with allowances for the time zones.

I won't get mine until Monday, so let us know how it works for you.

Thanx, Bill

I will not get to it for, at least, a week, so don't hold your breath.

Bill, I too am interested in your results, as you know I have 3 3 T and MBM and I know you have mutiples of both. Your experience and how to deal with more than 1 sub will be interesting.

As usual thanks for blazing a trail.

Tracy

More info on the operation of the 25 & 35 LIFT from TEL79 that he sent to me:

The liftings are somewhat user-dependent preferences. If your subs are ported and tuned to low frequency (~20H) and they are capable of high power outputs and excursions, you can use lift 25 which actually lifts 20Hz about 6-8 dB. This will give very deep bass, and some prefer it. However, if your subs have reflex tuning above 20Hz lift 25Hz is probably more than they can handle, so you might try Lift 35Hz which lifts 30Hz. The default setting is still flat response (no lifting), which works for any subwoofers.

There are some graphs I posted in this thread earlier, but the final response graphs of course depends on your system and room. You can try out which sounds the best for you. Whatever lifting you choose, it has no effect on the most important thing that 8033 does, which is the room EQ and removing standing waves etc.

Also may want to try to place your subwoofers in to the corners of the room. Without 8033 corners are useually the worst spots to place subwoofers as they give the most peaking modes, but with 8033 they are actually the best places as 8033 will take down any resonacne by digital domain counter modelling. Within corners your subwoofers will get the benefit of smaller 3D sound radiation angle, and they will reach lower frequencies and deeper bass.

This sounds like the best EQ for any sub.

Bill

Just ordered one myself. Looking forward to what this can do!!

I'm just wondering if I would still be able to use my Audessey calibration for the rest of my speakers with the 8033? I know it's possible, but how would the 2 interact for the subwoofer? As far as I know, there isn't a way to deactivate Audessey just for the sub...so when calibrating the sub with the 8033, Audessey would not be factored into it's measurements, but when you played something with Audessey on, it will be applying it's own (all be it minor) sub correction along with the 8033's correction...I'm wondering if this will throw off the 8033 calculations? if anyone who has used both can comment, please do. Thanks, Bill

we need a north american web site. That is still in euros :-)
does it ship from europe in which case S&H would cost a lot

I was wondering how it corrects the decay time. I was under the impresion that decay time/reverb can be best correctted by using room treatments (bass traps/absorbers)

I have used BFD (DSP1124P and FBQ2496). I dont like these for the fact that they add significant distortion when passed a full range signal. The distorsion is not as much when passed the low frequency alone but I am a bit hesitant to use them because of their impact on sound quality.

I have added corner bass traps to my room. Though the FR as not changed significantly except the peak at 142Hz the boominess is not there as much. I sit right at the back wall. Attached is the FR before and after bass traps. it is not in log scale. never mind.

Like I said the peaks and nulls have not improved but the decay time has probably improved. I dont know how to make the waterfall plots to see if the decay time/reverb has improved but I can hear the difference.

Now my question is: will the 8033 improve that nasty curve at my listening position

Please ignore that black frame. That is not part of the bass trap.

So does this 8033 replace all the bass traps. I dont care if the bass response in positions other than listening position is not corrected. All I care about is the bass response at listening position for 2ch music. So if 8033 is for one position only it doesnt bother me as it is automatic unlike manual parametric EQ like BFD. If I change the listening position then just by one click of a button I get it corrected instead of having to measure the FR and then use REW to make the filters etc. Also it has RCA pins unlike the 1/4th inch pins on the BFD.

I dont mind spending $400 if I can get rid of all the bass traps (OC 703 crap)
in my living room.

I really doubt it. First, room modes are annoying at frequencies up to 200Hz and few of use would use a sub crossover that high. Second, electronic eq from a single point is very specific and, even with one listening position, it cannot be in all the places your two ears will be (unless you listen in mono with your head in a vise).

This may be dandy for a sub but it is not panacea.

I really wonder how the decay time could be controlled electronically when it is basically due to physical phenomenon i.e room interactions. Unless the room characteristics are changed by adding treatments etc., how will the decay time/reverb change?

So what does 8033 do different from Audyssey. I have tried Audyssey and it was just pathetic. I have absolutely no control as to what I can do. Audyssey is pretty much a one button click process except that I get to pick multiple locations. Other than that there is really no user interaction and what I got back from Audyssey was just terrible.
I only want the bass EQ-ed but Audyssey will EQ the full range and the result is pathetic. I have got much better results using the BFD than Audyssey. But I prefer passive methods like bass traps/absorbers than active EQ.

If the room is treated then there is really no need to do any active EQ. But the main draw back is dealing with materials like OC 703 which may pose health issues in the long term. Commercial products like Realtraps are affordable only by ball players.

200Hz equates to about 1.5 meters or five feet. That kind of room is not very comfortable to seat both the listener and the sub


Yes, EQ from a single point is very specific with high frequencies. But exactly, Anti-Mode 8033 IS for a sub, not for high frequencies. It also isn't a floor-wax or a dessert-topping. Also, it has an area-calibration mode, so you can increase the usable listening area.

Anyway, you, mcallister and bsoko2 can test it yourselves and let us know what the pluses and minuses are.

Decay times and response behaviour have a sort of connection, as they are just different domain aspects of linear but mixed phase system. Using treatment improves both a bit, but usually not enough. If you have 15dB peak at 40-50Hz, it would take excessive trapping and treatment to get rid of it.

It is much more effective to use both treatment and DSP, as the DSP is only thing going to make dramatic improvements on such resonances. They are not in competition, they complement each other. Anti-Mode 8033 uses special AMFs to modify decay times on the reconances. It's basically a method of introducing anti-phasing sound to counter mode (hence the name Anti-Mode). It will make decay slightly (not miraculously) faster compared to some normal EQs, especially in subwoofers placed in to corners. The response will also get flat as a result of these anti-phasing filters and iterative AM algorithm.


Most definitely, that curve is basically a text book example of what 8033 is designed to correct.

You can use Audyssey for the speakers, but it's a bit questionable if it will improve the system or actually make it worse. Due to tests, Audyssey in even some more expensive receivers (~2000e) does not even touch the response of the subwoofer, only speakers. Using EQ to higher frequencies is not particulary effective, as the response and sound characteristic are changing very position dependent. Even inches away from measurement point will make dramatic changes. Forcing a flat response up from 200Hz may result in sound image that sounds colored and simply wrong. I would recommend to use the Audyssey only to measure speaker distances and volume, not EQ or "Dynamic EQ" which only adds more distortion to the signal (dynamic loudness). This is what people using both Audyssey and AM has concluded to do, and it works great.


In Finnish Hifi-world magazine (issue 2/2008) Audyssey, YPAO and MCACC were tested and also compared to 8033. The quote from the magazine was as directly taken from www.dspeaker.com ("For example the Anti-Mode 8033 tested in previous issue, was compared to these systems, completely superior tool for adjusting the subwoofer")

Do you think that modes are a problem only at the fundamental frequency of the primary room dimensions?


Sure but there are modes outside of the sub range in most rooms. In fact, the range from 100-200Hz has more of an effect on coloring the sound than does the low bass range.

A lot of conclusions (about many products) are based on less than successful use of them. Note the long 'official' thread on Audyssey to show the intricacies of getting it right. Similarly, I am wrestling with the ARC in an Anthem D2 and it, too, is far from 'plug-and-play.'


This is worse for bass and a poor argument for not EQing higher up. At higher frequencies, the density of modes, spatially and in frequency, is higher, making positioning less critical than for bass (or mid-bass).


Well, I have not had a chance to play with the 8033 yet but I believe most people who have the opinion that Audyssey or other EQ products, used correctly(!), colors the sound have simply adapted to the colorations of their room/speaker and base their preference on that adaptation. Now, if you have a well-designed and acoustically treated room, as well as good, well-positioned speakers, the effect may be minimal but it always helps. (Again, when done right.)

You are right, I didn't think of that as I was only thinking about subwoofers.

You would then need to have eq above the crossover frequency with the main speakers too, but still limit the correction to the 'bass range' to avoid problems with the locality issuescaused by the correction (which isn't a problem with the mono subwoofer channel).

Doing EQ for the higher frequencies is OK as long as you dont try to make too narrow high-Q correction filters based on one point. At this side of the Schroder radius, modal peaks and nulls can have bandwidth of only few hertzes (no smoothing), and the center frequency is altered by same amount when microphone is moved a few inches to any direction. They would correct the response very locally. Automatic EQ algorithm for higher frequencies must be done with different approach to obtain successfull results. You can use for example shelving filters and wideband bandstops, and the measurements can be taken from many points to get better view of the situation (and avoid early reflection errors). The algorithm computes smoothed versions of responses and combines the multiple point data in different way compared to bass range, where also the absolute phase information is important (for algorithm).

The measurement routine is also different for higher frequency automatic EQs to work properly. You can measure the speaker itself with very short MLS burst, speaker with early reflections with a bit longer one, and late reverberation decay characters (waterfall time curves) of the room with long windowed MLS or sweep. If done right, automatic EQ can indeed improve higher frequencies too. The effect might not be as obvious as it is in the bass range with severe modal problems. I guess I'm personally favoring treatment over EQ on the higher frequencies.

But it is true, that even the best systems can and will fail if the user don't know how to use them.

In the test (Finnish Hifi-world 2/08) I mentioned between YPAO, Audyssey and MCACC, it was Audyssey (in Onkyo TX-SR875E) and MCACC (in Pioneer VSX-LX70) that resulted in better response measurement-wise, but they were considered subjectively worse. YPAO (in Yamaha RX-V3800) made only slighter adjustments, and was subjectively considered best by the professional testing crew. This supports my opinion of EQing higher response, optimising the result for a standard MLS or sweep measurement is not the best option. It should also be noted, that none of them corrected the subwoofer response properly, where the most benefits of PEQ would have been.

Just wanted to clarify that this is not just the bass response. I have not set the cross over high. My sub (Sunfire True Sub Junior) can only go upto some 100Hz. Its specs are 22-100Hz. The cross over is set at 90Hz. I have an old Yamaha RX-V795a and it has just one cross over at 90hz. No options.

The FR is for the mains and sub combined. Mains are set to small.

The nasty peaks beyond some 90Hz or so are mainly due to the mains.

I have one important question. Does 8033 accept full range signal or does it acept only below 80Hz. I guess the SVS EQ only accepts 80Hz and below no matter what I feed. If that is the case with 8033 then it will be useless to me
because I have some nasty peaks over 80Hz that I need to tame down.
I can only do this by sending full range signal as my receiver has fixed cross over at 90Hz and also my sub can only play up to some 90Hz.

Like I said before I can feed the BFD FBQ2496 full range and use its 20 filters to get a decent response up to 200Hz. But the FBQ2496 causes significant distortion to the high frequencies even if they are not EQ-ed. Just passing the full range to the BFD causes distorsion to the signal. The distorsion is not noticable at low frequencies. It is not an audiophile quality devide. Mainly mean for PA systems I guess.

So how does 8033 handle full range signal?

8033 is meant for only subwoofer, so you cant EQ mains with it. The device has a built-in lowpass at D/A side which will filter out above 160Hz. The range of correction is limited to 16-144 Hz, as this is the region when Anti-Mode algorithm can be used. Above that, AM type automatic algorithm can not be used because of phenomena relating to modal density and Schroder radius.

Unfortunately in your case, the 8033 can not help above to fix above 90 Hz if the cross over is fixed to this frequency.

My LFE cross over is fixed at 150hz, so I expect help with that.

Bill

Agreed.


I won't tell anyone.


Well, I do not have access to that report nor have I used any of the EQs except Audyssey, and that in other equipment. The mic Onkyo uses for this is, imho, a poor design that makes getting correct measurements very tricky. OTOH, when it comes to subjectivity, that's a bigger can of worms.

Thanks


It's true, subjectivity plays a big part at higher frequencies. It's difficult if not impossible to make a full-band auto-EQ that would satisfy everyone. Like you thought earlier, it's even probable that the listener gets accustomed to the anomalities and coloration of the room, and the corrected sound balance then feels too weird. In the bass range there is much better consensus of what sounds right, or at least what sounds wrong. Nearly all people appreciate getting rid of the boomy modal peaks.

This is one of the differences when designing automatic algorithms to clean bass range vs. other freqs. In the bass, you can actually counter-model the room accurately with DSP, even if the best results apply with local nature. But at the higher frequencies the accurate response in both frequency (and digital z-domain) and time domain is actually changing even more rapidly from point to point. However, what virtually remains the same on higher frequencies is the "envelope" of the spectrum, which can be tweaked to get better response. I assume this is the property you meant when you stated that the response at higher frequencies has globally more consistent nature. There, at higher the best DSP-correction model is stochastic, compared bass frequencies where exact z-domain countering can be used.

There are some semi-stochastic approaches to bass EQ as well in the previous equipment, for example Velodynes SMS-1. It uses heavy smoothing (1/3 or 1/2 oct) in the measurement it can plot to TV. However that is also perhaps the main reason why it's automatic EQ algorithm falls kind of short. That and 1Hz resolution it uses which is simply not enough. You'd get better results with it (or much better with BFD etc.) when you manually set the filters with computer based measurement software such as Room EQ wizard, because that way you get closer to the exact counter-model of the room. Also the price in Europe for SMS-1 was maybe too high for wide commercial success (>800e). SMS-1 had it's own good properties like the ability to use it as cross over and the remote controller.

With 8033 the motivation was to make precise auto-EQ (improved spectral resolution, 24 filters, 4-phase iterative measurement, wider area compensation) which everyone can use and the price would still be the same as with manual PA EQs + microphone + accessories. I think this is one of the main reasons 8033 became so popular immediately after it was released (first production lot was sold out in 2 days), and the fact no one likes boomy modes.

I do not think we are disagreeing much about any of this. Ideally, all would be resolved with physical acoustical treatments which, afaik, have no downsides, other than cost and bulk. Short of that, one needs to handle low frequencies somewhat differently from the way we deal with high(er) frequencies. The only complication I was trying to emphasize is that subwoofer EQ is constrained by the fact that our upper limit for reasonable crossover frequency (to avoid obvious localization) is an octave or two lower than the Schroder limit. That leaves, to me, a glaring gap in the most critical range, one that affects voice timbre. I do not have a good off-the-shelf solution for that, other than a Rives PARC for each pair of main channels and a sub-PARC (or 8033 or other bass EQ) for the woofer(s).

Yep, I know what you mean.

It is possible that maybe one day in the future there will be consumer priced full-band EQ/Cross-over device from VLSI Solution's DSPeaker product series. But for now, theres the 8033 Anti-Mode for subwoofer and a couple of other new DSP products. The 8033AM seems to be helping a lot of people struggling with bad bass response. As far as I know, it is also the only plug 'n play auto-EQ in it's price range.

As with any EQ, the improvement on 8033 depends on how bad the initial situation was. In really good listening rooms the improvement is noticeable but subtle, whereas in other rooms the improvement is unbelieveable. As the old say goes: If it's not broken, don't fix it.

Here's one room from the days of 8033s development era, which was definitely broken. The CSD graphs f-axis has linear scale, because it suited much better for development purposes back then. The sub was placed in corner (best position for 8033 algorithm)

The room before:


and the same room after automatic calibration of 8033


I just hope that some customers of 8033 can have as 'bad' room as that was, so that they can have the same feeling that I had when I listened the difference.