Is REL being more 'musical' than SVS a myth, or is there some real science behind this?

[jsilvela]

You only have one sub? In that case if you have bad nulls then no EQ will ever fix it.

And, as said a few times, entry level MultiEQ is not good for subs.

haha, yes, only one sub, and I think a second one (not a REL) will be my next audio improvement.
But first I want to get the hang of my one sub, and learn to integrate it better.

I do realize EQ cannot fill the nulls, but MultiEQ was:
- failing miserably at damping a very powerful mode @39Hz
- completely over-amplifying LF to the point I felt I was inside a gangsta vehicle

My sub with "by-ear" integration sounded much much better than integrated by the Denon.
I imagine more people may have had similar experiences.

 

[jsilvela]

Two subs really are the minimum. They do not need to be identical: as Archimago showed you can use a very big one for extension and a cheap and small one to smoothen the response.

I saw you post that previously, and had already gone to the Archimago post.
And other posts from you about Kef Kubes being good for money in the EU.
I have been pretty persuaded by what you say on both counts, I just want to learn more before getting a second sub.
Thanks!

 

[bodhi]

- completely over-amplifying LF to the point I felt I was inside a gangsta vehicle

That is strange. Do you mean that the response in listening position was bonkers?

With one sub, even with the XT32, other positions than then measured one can have huge peaks with the obvious heard and felt side effects as the algo tries to smooth things out by boosting the dips. Adding second sub immediately fixes this for the most part, smoothing the overall response in room and considerably less EQ is required.

There is not much more to learn about the subject. The overall difference in adding a second sub in your situation is very clear and audible, no need to AB test and ponder if it is worth it. So, just get any sub you can get your hands on that you don't need to commit to buying and hear for yourself. If that is not possible you might try REW room simulator and see how 1 and 2 subs behave in your room theoretically. For me the simulation was spot on.

 

[jsilvela]

That is strange. Do you mean that the response in listening position was bonkers?

With one sub, even with the XT32, other positions than then measured one can have huge peaks with the obvious heard and felt side effects as the algo tries to smooth things out by boosting the dips. Adding second sub immediately fixes this for the most part, smoothing the overall response in room and considerably less EQ is required.

There is not much more to learn about the subject. The overall difference in adding a second sub in your situation is very clear and audible, no need to AB test and ponder if it is worth it. So, just get any sub you can get your hands on that you don't need to commit to buying and hear for yourself. If that is not possible you might try REW room simulator and see how 1 and 2 subs behave in your room theoretically. For me the simulation was spot on.

You can see the graphs on this post in the "what's the fuss" thread: https://audiosciencereview.com/foru...subwoofer-specs-and-brands.40468/post-1437346

Since then I've been doing a bit of experimentation with placement, and am getting much smoother audio generally, according to REW.
I will try Audyssey again in the next days.

The second sub is making good sense, and I do think I will go for it. I have my eye on a Kef Kube.
But I really do want to make some tests and get the best from what I have, before springing.

 

[bodhi]

You can see the graphs on this post in the "what's the fuss" thread: https://audiosciencereview.com/foru...subwoofer-specs-and-brands.40468/post-1437346

Since then I've been doing a bit of experimentation with placement, and am getting much smoother audio generally, according to REW.
I will try Audyssey again in the next days.

The second sub is making good sense, and I do think I will go for it. I have my eye on a Kef Kube.
But I really do want to make some tests and get the best from what I have, before springing.

You said there that you had dynamic EQ enabled for the problematic response and it looks like it. Check your measurements with Dynamic EQ disabled and I think you'll find it's fine.

With music, Dynamic EQ with 0 offset is too much bass. I have it at -15dB for music and 0 for movies and it's fine.

Basically you have two options: let Audyssey flat out the sub response and trust Dynamic EQ with offset to do it's thing. It works for me.

Second option is to create your own low shelf and disable Dynamic EQ. When you do it by ear it works best only for the volume you are using while testing and perceived bass becomes too much or too little when changing volume (Fletcher–Munson curve).

 

[jsilvela]

You said there that you had dynamic EQ enabled for the problematic response and it looks like it. Check your measurements with Dynamic EQ disabled and I think you'll find it's fine.

With music, Dynamic EQ with 0 offset is too much bass. I have it at -15dB for music and 0 for movies and it's fine.

Basically you have two options: let Audyssey flat out the sub response and trust Dynamic EQ with offset to do it's thing. It works for me.

Second option is to create your own low shelf and disable Dynamic EQ. When you do it by ear it works best only for the volume you are using while testing and perceived bass becomes too much or too little when changing volume (Fletcher–Munson curve).

I did read the comments on the Dynamic EQ on that other thread, and tried Audyssey without it.
The results were also bad. The LF band came down yes, but there was a wide dip in 40-60 Hz, and listening test showed my bass had vanished.

I did see with REW that at my listening position, the response had a tall room mode and a deep null. Playing with location, I am managing to get something with lower peaks and shallower nulls.
I speculate that if MultiEQ is low-res for LF, maybe it was latching on the null or to the room mode, and coloring everything.
Once I get something smoother with REW, I will retry Audyssey.

Basically you have two options: let Audyssey flat out the sub response and trust Dynamic EQ with offset to do it's thing. It works for me.

I did not know you could do that!
The Dynamic EQ so volume changes follow Fletcher-Munson is looking damn brilliant if I get it working. I'm coming around to AVR's, I have to say.

 

[Willem]

I use dynamic eq/loudness in my RME ADI-2 DAC and it is very useful when you llay at lower levels. However, for measurements purposes one should obviously turn it off.
I am about to embark on a new subwoofer and eq project using MSO. I will use a miniDSP 2x4 HD to eq the subs and for the 80-250 Hz range I will try to apply the MSO filter values to the ADI-2 to equalize the main speakers. But perhaps a simple REW curve will be better for that - who knows?

 

[jsilvela]

I keep coming to this issue: the REL exception if you like.
For all the passionate argumentation for and against, there's little raw data on REL in general.
The only proper measurement I've seen of a REL was this one https://www.audioholics.com/subwoofer-reviews/rel-ht-1510-sub/conclusion
It's for a model in the HT line.
What I saw there clearly looked like the FR of a single driver. Not very flat at any point. More like a parabola / bell curve.
I understood that to be "the REL way" i.e. no DSP applied to the driver, just concentrate on the driver itself, the LP filters and the finish/cabinet.
This appealed to me then.

Been peaking at the multi-sub idea, and the reality that for small rooms, the region <100hz is dominated by discrete modes, and your flat anechoic FR is going to get messed, and is going to be very spatially uneven.
If eventually you are going to do multi-sub, and smooth out via top-level DSP, then I don't see much point to having DSP on each individual sub.
Though having DSP on the sub I guess makes it an easier "building block" for your AVR.

Floyd Toole's book basically says that at low frequencies it's all about getting output "out there".
So, in a way, if what you want are >1 "dumb" individual subs and higher level calibration via mic + DSP,
wouldn't it be preferable to have the sub basically be a driver?

I was eyeing a Kef Kube 10b, and downloaded the instruction manual to see if the DSP had an "off" switch for "just give me the raw driver"... nope.

 

[jsilvela]

Quoted from the article https://www.audioholics.com/subwoofer-reviews/rel-ht-1510-sub/conclusion

This leads to one of the HT/1510’s strengths: its time-domain performance. This sub has extremely low group delay across the entire bass frequency range, and that means it doesn’t allow the reproduced sound to last much longer than the input signal at all. Bass notes will not linger longer than they are supposed to with the HT/1510. This has always been an important performance criterion for REL Acoustics, but I don’t regard it as a serious problem so long as it is kept under a certain level, and most competently engineered subs are able to keep it below a level where it could become an audible problem.

And here again I wonder, after reading about the relative low sensitivity of the human ear to low frequencies (relative to the sensitivity for mid-high frequencies, I mean).
In the range 30Hz - 100Hz, is there a preference for flatter frequency response?
I know from the Toole book and ASR in general that anechoic flat on-axis (aka "neutral") is statistically preferred in blind tests.
Are there blind tests specifically concentrating on the low frequencies?

 

[Willem]

SVS are currently offering the older SB1000 and SB2000 (i.e. without the filters of the pro versions) at much reduced prices, in at least some countries. As you write, if you are using something like MSO/2x4HD, the filters are unnecessary.

 

[jsilvela]

SVS are currently offering the older SB1000 and SB2000 (i.e. without the filters of the pro versions) at much reduced prices, in at least some countries. As you write, if you are using something like MSO/2x4HD, the filters are unnecessary.

I've been watching the Geddes video you recommend, and then looked at an MSO video tutorial.
The comp/math person in me finds it all very interesting.
But something is bugging me.

You start with >1 subs, which may or may not have individual DSP activated.
Then you'll use a miniDSP to feed each sub an individual filter generated by MSO. So, 1 or 2 rounds of DSP per sub.
Then, your AVR will "hear" the above as a sub with a much flatter response, spatially and sonically.
And it will apply its crossover filters to integrate with the mains+center speakers.
So, potentially 3 rounds of DSP, minimum 2, per sub.

Is all this necessary? Is the difference between flat and non-flat response in LF so noticeable?

 

[Willem]

I don't think that with MSO into a miniDSP 2x4HD one also needs filters on the subs, so that is why I htink money can be saved there. I have no experience with Audyssey and the like, but I would be surprised if there are real benefits adding it to the subs' equalization. There should be for the other speakers, however.

 

[rcstevensonaz]

I don't think that with MSO into a miniDSP 2x4HD one also needs filters on the subs

This has been a key question in my mind — I asked it a while back on this (or another) thread, but I don't recall anyone giving a definitive answer. So put differently, is there processing capabilities in an SVS-1000/2000 Pro (or other similar priced line of subwoofers) that cannot be replicated by a miniDSP 2x4HD and/or miniDSP DDR-24?

I have no experience with Audyssey and the like, but I would be surprised if there are real benefits adding it to the subs' equalization

This has also been one of my questions. If the AVR has multiple subs hanging off of the same subwoofer output (i.e., a Y-cable), then I am sure Audyssey not be able to achieve the same "taming of nodes" vs. subwoofer-based DSP since it cannot address each sub individually. So in a world where AVRs do not have 2 to 4 sub outs and before Audyssey XT32 really came of its own, it seems that subwoofer-based DSP could be a huge benefit.

But, what about Audyssey MultiEQ-XT32 with an AVR that has separate sub outs for each subwoofer? E.g., 2 sub out for dual subwoofers or 4 sub out for tri- or quad subwoofers? Does that remove the need for subwoofer-based DSP when the line out to each subwoofer is directly controlled?

 

[jsilvela]

[...] So in a world where AVRs do not have 2 to 4 sub outs and before Audyssey XT32 really came of its own, it seems that subwoofer-based DSP could be a huge benefit.

Good point there. But also: in the context of ASR readership, would a miniDSP 2x4 connected to the AVR's sub-out not be much better than having DSP on the subs?
Unless the sub DSP can get a room reading either via calibration mic or via PEQ created by REW, the DSP in the sub will not address room modes.

And, if using the sub-out of an AVR ... can the lowpass filter for the sub be defeated?

[...] I have no experience with Audyssey and the like, but I would be surprised if there are real benefits adding it to the subs' equalization.

Are you doing everything with a miniDSP or similar? If using an AVR, are you bypassing the low-pass filter?
At least in my AVR, I think as soon as you set a crossover, there are filters applied both to the mains and to the sub.
If there were a way to bypass the AVR's low-pass, I think that would be useful to integrate the REL's strongly non-flat FR.

 

[rcstevensonaz]

But also: in the context of ASR readership, would a miniDSP 2x4 connected to the AVR's sub-out not be much better than having DSP on the subs?
Unless the sub DSP can get a room reading either via calibration mic or via PEQ created by REW, the DSP in the sub will not address room modes.

Whether Audyssey, Dirac, miniDSP2x4 DSP, or built-in subwoofer DSP is the better system for adjusting timing, phase, PEQ, etc. for each subwoofer is an interesting question. But I think that is out of scope for this thread.

The ability to get a "room reading" and to set the DSP parameters is assumed for all methods. Whether that be via automated software (Aydyssey, Dirac), manual measurement and configuration (REW), an application provided by the subwoofer (e.g., SVS's smartphone app), or tuning by ear.

And, if using the sub-out of an AVR ... can the lowpass filter for the sub be defeated?

For some subwoofers (such are REL), just use the LFE input. The subwoofer assumes LFE channel is already bass-managed by the upstream AVR / source and disables the LFE filter.

Otherwise, just set the low-pass on the subwoofer to its highest setting (often 120Hz or higher). Because the DSP system will cutover frequency will be lower than the subwoofer's highest setting, it is essentially defeated because it won't ever be engaged.

Are you doing everything with a miniDSP or similar? If using an AVR, are you bypassing the low-pass filter?
At least in my AVR, I think as soon as you set a crossover, there are filters applied both to the mains and to the sub.
If there were a way to bypass the AVR's low-pass, I think that would be useful to integrate the REL's strongly non-flat FR.

This is part of what I'm still trying to figure out as well. I have a Denon X-4400H, Audyssey MultiEQ-XT32, and two separate (but still mono) sub outs. Since I only have dual REL T/5i subs right now, I'll just Audyssey software for the full management of mains and subs. However, I'm not sure what would be the best approach when I later add a larger 3rd subwoofer such as an SVS SB-2000.

But your situation is different since you only have Audyssey XT and I'm curious what others will suggest. My instinct is that you would still let the AVR do the low-pass filter between the mains and sub out. And then you would use a miniDSP to manage the subwoofer integration. You lose the ability to adjust the mains with the miniDSP software, but there have been a lot of posts that suggest not messing around much with the upper frequencies anyway.

 

[jsilvela]

For some subwoofers (such are REL), just use the LFE input. The subwoofer assumes LFE channel is already bass-managed by the upstream AVR / source and disables the LFE filter.

I'm not sold on using the LFE on the REL. The one solid measurement I've seen for a REL is this https://www.audioholics.com/subwoofer-reviews/rel-ht-1510-sub/conclusion
My measurements on the T/5x suggest similar: a parabola/bell-curve whose maximum increases and moves right as the crossover setting increases. The gain at say 20 Hz moves little with different xover settings though.
I've measured the LFE input to have similar response to the low-level input with crossover at the max position.

The result of using the LFE with Audyssey is that Audyssey perceives the REL to be at a high volume, and thus applies a negative gain to it (-9 dB in my case). This has the effect of lowering the lowest frequencies of the sub.
In contrast, if I set the crossover lower, say to around 70-80 Hz, Audyssey does a better job. The sub is detected at a lower volume. The AVR does not apply negative gain, and relative to the LFE input, the lower frequencies in the sub are thus higher.

The REL having such a bell-shaped FR makes it more difficult to integrate automatically, which is why I'd be interested in using the low-level and disabling the LP filter on the AVR.

 

[Earthbound]

I’ve been looking at Rel due to their Nuetri/speaker level connection option. I have a SET amp with no subwoofer option. What other brands provide this option? Rythmik specifically states that it is not compatible with this type of amp.

 

[jsilvela]

I’ve been looking at Rel due to their Nuetri/speaker level connection option. I have a SET amp with no subwoofer option. What other brands provide this option? Rythmik specifically states that it is not compatible with this type of amp.

There are two parts to this:

1 - being able to take "high level" connections. Other brands have this too. Not unusual. E.g. Kef has subs with speaker level inputs. I believe SVS has some models too, though I don't have specifics. But the high-level connections are usually considered "legacy". Only REL uses them as their recommended input.

2 - integration with mains. This is usually considered to require a high-pass filter of the main speakers. I.e. if you'll be setting 80Hz as your crossover frequency, you set that on the sub. Your amp would hopefully apply a high-pass filter to your bookshelf/floorstander mains so they roll off at 80Hz. Your amp probably does not offer that.

The "REL way" is to ignore point 2, and integrate "by ear" to your existing amp, without high-pass filtering your mains.
Though pound-for-pound the REL will have less output and less extension than others, they do cater exactly to this kind of use case.

Luck!

 

[DonH56]

You have to be careful with high-level connections to subs. On most balanced/bridged amplifiers, and some tube amps (like my old ARC D-79), the (-) speaker output is not ground but driven (active amp output stage or transformer winding/tube coupling). That means if you connect to a grounded high-level input on a sub you can destroy the amp, or at least send it into protection and shut down mode.

A better way to add a sub would be to add a crossover before the power amp, splitting off the low frequencies to the (line level) sub input, and higher frequencies to the SET (or whatever) amp. That way you reduce the load on the high-frequency amp and speakers. Easier integration and better sound.

 

[jsilvela]

You have to be careful with high-level connections to subs. On most balanced/bridged amplifiers, and some tube amps (like my old ARC D-79), the (-) speaker output is not ground but driven (active amp output stage or transformer winding/tube coupling). That means if you connect to a grounded high-level input on a sub you can destroy the amp, or at least send it into protection and shut down mode.

A better way to add a sub would be to add a crossover before the power amp, splitting off the low frequencies to the (line level) sub input, and higher frequencies to the SET (or whatever) amp. That way you reduce the load on the high-frequency amp and speakers. Easier integration and better sound.

Ah, good points. In the case of @Earthbound who is considering a REL:
The REL high-level cable supplied with the sub has 3 wires. Two "hot" ones to connect to the (+) speaker terminal for Right and Left speakers.
And a "neutral" wire that should be connected to the (-) speaker terminal in one of the channels *for an AB amp*.
For the balanced/bridged topology, the instructions specify to leave the neutral wire disconnected, exactly for the reason you mention.

Got to say, REL get a lot of flack for doing less per-dollar, but they very much have people in mind with this kind of use case.
With the caveats, no external crossover should be required. Though of course, you would be losing the benefit of bass management.