Value of subwoofer amplifier power ratings?

Mr Wolf

Well-known Member
I may get around to getting a UMIK-1 and have a play with REW at some point but I suspect that will bring up all sorts of inadequacies which will need cavernous amounts of cash to sort out!!
That's certainly what happened in my case - less than 2 weeks lapsed between me cracking open the UMIK-1 and dropping £2.4k on a pair of new subs LOL.

So yes, ignorance can be bliss (and less expensive) but, that said, I've never looked more forward in my life to watching movies with LFE heavy soundtracks as the new subs have taken things to a whole new level of enjoyment. Me and the Mrs watched Terminator Genisys last night on Blu-Ray and, while the film was OK, the bass effects were simply incredible on 7.1 TrueHD.
 

D1gita1

Active Member
I agree which is the whole point of this thread.

Here's a bit more info as I just found a chart I recorded from the test which was about 3 months ago - thought I had deleted it otherwise I would have included in the first post.

View attachment 1510875

The green line is the 500W 12" sealed SVS SB-2000, the blue line the Velodyne 750W 15" ported sub. Each line represents the maximum SPL output before compression set in somewhere along the FR line. I can't remember but I think I had the 80Hz crossover enabled in the AVR.

Room size is 5.68m x 5.34m x 2.38m. Both subs were in the same corner location with the UMIK-1 at the MLP. All EQ was disabled although both subs have DSP, the Velodyne manual stating that it has a 15Hz HPF.
Quick off the top off my head thought looking at this.....

You look to have a big 20Hz dip caused by the room. If the Velodyne port is tuned around 20hz and the hpf killing everything under 15hz (I'm not sure how aggressive the hpf slope is) then that's a big piece of the puzzle. The Velodyne driver is probably already rolling off at 25Hz Hz, right were the room looks to be so cancelling the frequency range. The sealed sub having no real hpf picks back up below the dip, slightly offsetting the cancellation, while the Velodyne looks all but finished at that point. That 250 extra Watts of power will do nothing to solve that dip.
 
Last edited:

Dobbyisfree

Well-known Member
Me and the Mrs watched Terminator Genisys last night on Blu-Ray and, while the film was OK, the bass effects were simply incredible on 7.1 TrueHD.

That's part of the point in having a decent AV setup... "What did you think of that film dear?" "Dunno, it didn't have a story, but the sound and visual effects were amazing."
I'd say at least half of the films we've watched on Netflix we would've given up at a maximum of 20 minutes in if we were watching them on TV sound. :)
 

Dobbyisfree

Well-known Member
I’ll be interested to see how you get on.

Not the 4 subs yet but just some info for you.
So, the total opposite of what is normally shared but kind of on the subject of power as in (pathetic) two REL Quakes quoted as only 100w each. Each 6cm from front wall inbetween the front floorstanders, symmetrically either side of MLP by 80cm, room is 6m wide and 5.5m length (Audyssey said they were each 5.1m from MLP).
It's a shame that the Audyssey app doesn't show the subs individually when (apparently) they are being individually EQed.
One is set to "depth" and one to "slam" which was in the hope of them spanning a wider range.
@AndreNewman using REW it said they would have a peak at 30Hz!
Oh and they are just sat on the carpet, no spikes in (yet) or anything else and nothing on top pushing them down.
1621494348147.png
 

vm1451

Active Member
As that graph is relatively simple, could you explain what I'm looking at? Obviously volume against frequency but what does it mean? 0db is reference volume? The two graphs are before and after EQ?
 

Dobbyisfree

Well-known Member
Obviously volume against frequency but what does it mean?

That is what the Audyssey app tells you. Yes, the green is before and the red is after.

As to what the levels are, I don't know, but I would like to know!

As far as I'm aware it's not telling anything useful about output levels.
 

Conrad

Moderator
Red is predicted after. They’re generally not super accurate.

The levels are deviation from the calibrated level. If you’re calibrated to 75dB then 0 is 75, -5 is 70dB, +10 is 85dB etc.
 

vm1451

Active Member
Ta!
 

vm1451

Active Member
Red is predicted after. They’re generally not super accurate.

The levels are deviation from the calibrated level. If you’re calibrated to 75dB then 0 is 75, -5 is 70dB, +10 is 85dB etc.
Ah, so 0 can float depending on what the user set it to, i. e. It's not fixed from graph to graph? The downward slope is 'roll off?
 

Dobbyisfree

Well-known Member
I'm probably wrong but I don't believe you can change the setting, I think that the Denon AVR-X calibrate at 75dB? Conrad will know as he's omniscient!

Below, for example, is one of the surround speakers which I think is also showing to the 75dB as calibrated level?

P.S. Reasonable response for an ancient Gale Gold Monitor that was £28 the pair inc postage!

1621497316174.png
 

Conrad

Moderator
Ah, so 0 can float depending on what the user set it to, i. e. It's not fixed from graph to graph? The downward slope is 'roll off?
I think Audyssey calibrates to 75dB at -20 (which I think works out to be reference) but it might float a bit. If you have a higher gain amp, or the gain higher on a sub, it'll end up with a larger negative trim to try and bring that speaker down to 75dB.

Yes, the downward slope is a roll off, but to be expected from small subs in a large room.

As the before graph is pretty accurate, a tedious but useful way to use it would be to measure, then adjust, then measure again, saving the graphs each time. Then you can compare the before responses and see which have the fewest nulls. Audyssey is fine taming peaks, but it's not great for time alignment so giving it the best before response to work with will likely lead to the best outcomes.
 

Mr Wolf

Well-known Member
Not the 4 subs yet but just some info for you.
So, the total opposite of what is normally shared but kind of on the subject of power as in (pathetic) two REL Quakes quoted as only 100w each. Each 6cm from front wall inbetween the front floorstanders, symmetrically either side of MLP by 80cm, room is 6m wide and 5.5m length (Audyssey said they were each 5.1m from MLP).
It's a shame that the Audyssey app doesn't show the subs individually when (apparently) they are being individually EQed.
One is set to "depth" and one to "slam" which was in the hope of them spanning a wider range.
@AndreNewman using REW it said they would have a peak at 30Hz!
Oh and they are just sat on the carpet, no spikes in (yet) or anything else and nothing on top pushing them down.
View attachment 1514981
That red "predicted" response line looks more like wishful thinking by Audyssey as there's no way it's going to be able to EQ out that null in the 40-55Hz region like it's predicting.

I have a similar sized room (5.7mx5.4m) and get a very similar shaped measured response (with UMIK-1/REW) to your green line with a pair of subs up front between the centre channel speaker and the mains. The only way to address that null was to put at least one sub at the rear of the room.
 

Dobbyisfree

Well-known Member
Read your post and thought I would try using my calibrated ears :)

With amp at -25dB master volume used a Youtube 100-0Hz in 5Hz steps.

So, joking aside, no where near a perfect test BUT you are right (as usual) slight quietening for 50 and 55 and louder at 30 too (compared to adjacent tones in both cases).

New setup (TV and setup moving 90 degrees to other wall) will be done in the next 13 weeks (in time for sofa deliveries) and subs will be in corners.
 

AndreNewman

Active Member
@AndreNewman using REW it said they would have a peak at 30Hz!
Slight misunderstanding, I said a Quake in MY ROOM has a peak at 31Hz, as does a Monolith. As do all my speakers, the peak is the resonant node between front and back walls 5.4m apart.

The peaks in your room will be different, unless your room is the same size and shape and construction as mine, probably not...


Oh and they are just sat on the carpet, no spikes in (yet) or anything else and nothing on top pushing them down.
View attachment 1514981

Looks like a peak at ~40Hz and another at ~65Hz if my guess at the scale is right. Not sure why the correction has a downward tilt, opposite of what most people want, unless you asked it to do that?

I have no experience of Audessy, this is where these graphs come from?
 

Dobbyisfree

Well-known Member
I meant that I used the room sim in REW and it said a peak at 30Hz.

Thanks for pointing out the tilt, that confused me too so I went back to the curve editor... So, I had tweaked in curve editor to try and boost between 100 and 120Hz. What it does then is not "boost that above 0dB" but it takes it downwards if you see what I mean? Sorry, hard to explain. Easier to show like this:

1621504857296.png


I've put the target curve back to flat and sent the file back to the AVR, you can see the difference.
Yes, Audyssey XT32 creates these graphs but there is a curve editor to tweak your target curves.

Thanks again Andre, every days is a school day!
 

Dobbyisfree

Well-known Member
Not that you're probably that interested in Audyssey but you did ask 🤣

You can load the *.ady file into a program called Ratbuddyssey to see the measurement, then it looks like this... and shows you below 20Hz:

1621505487988.png
 

AndreNewman

Active Member
I meant that I used the room sim in REW and it said a peak at 30Hz.
Ah ok, good.

I found the REW room sim to be scarily accurate.

Thanks for pointing out the tilt, that confused me too so I went back to the curve editor... So, I had tweaked in curve editor to try and boost between 100 and 120Hz. What it does then is not "boost that above 0dB" but it takes it downwards if you see what I mean? Sorry, hard to explain. Easier to show like this:

View attachment 1515044

I've put the target curve back to flat and sent the file back to the AVR, you can see the difference.
Yes, Audyssey XT32 creates these graphs but there is a curve editor to tweak your target curves.

Thanks again Andre, every days is a school day!
That makes sense, I thought it was an odd thing to have.

As said already, that after looks a bit optimistic.

In my experience you just need to knock off the big peaks and it will sound hugely better, too much correction is as bad as no correction.
 

Dobbyisfree

Well-known Member
This seems an appropriate enough thread to post this question on. Where subs are located symmetrically about the MLP, using SubEQ HT, has anyone compared results with:
(a) Connecting two symmetrical subs to one output to
(b) Connecting each individually?
To use multiple subs, option (a) may be the only option available to people.
So question to subwoofer experts, if you only have Sub EQ HT to do EQ would you rather have:
(1) Four subs, two at the front on one sub output and two at the back on the other sub output. Or
(2) Two subs, each connected separately at the front. Or
(3) Two subs, each connected separately at the rear.

And @Conrad all four would be ported 🤣
 

Conrad

Moderator
This seems an appropriate enough thread to post this question on. Where subs are located symmetrically about the MLP, using SubEQ HT, has anyone compared results with:
(a) Connecting two symmetrical subs to one output to
(b) Connecting each individually?
To use multiple subs, option (a) may be the only option available to people.
So question to subwoofer experts, if you only have Sub EQ HT to do EQ would you rather have:
(1) Four subs, two at the front on one sub output and two at the back on the other sub output. Or
(2) Two subs, each connected separately at the front. Or
(3) Two subs, each connected separately at the rear.

And @Conrad all four would be ported 🤣
Assuming you don't have a minidsp in the equation and it's just sub-out - Y splitter - sub(s) then:

a) this should be fine, but you lose the benefit of time alignment, you'll only gain headroom and any smoothing that the sub responses achieve with the same set of EQ applied. If the responses are very different then you could introduce lots of issues. Symmetrical to you doesn't mean equal responses, is the whole room symmetrical? I'm guessing not assuming it has one door. Furniture is usually not symmetrical either.

b) if the sub-eq in the AVR has the capability to time align then this is the preferred option, assuming you're not giving up channels.

As for preference:

1) four subs is tough to time align but I'd rather have two than a pair. You haven't considered them in left/right pairs, any reason why not? Again, if the fronts have a similar response and the rears have a similar response (magnitude and phase) then consider them one sub. This is actually what I do.

2 and 3) completely impossible to say for definite, too much depends on the room. One isn't always better than the other. The sub response is a function of the sub and listener's placement in the room, and the room itself.

As a good rule of thumb best placement options are usually: corners, 1/3, 1/4 and 1/2 wall lengths. And these can be mixed. So you might find that a sub 1/4 along the left wall from the front and 3/4 along the right wall from the front might work best.
 

Dobbyisfree

Well-known Member
Assuming you don't have a minidsp in the equation and it's just sub-out - Y splitter - sub(s) then:

a) this should be fine, but you lose the benefit of time alignment, you'll only gain headroom and any smoothing that the sub responses achieve with the same set of EQ applied. If the responses are very different then you could introduce lots of issues. Symmetrical to you doesn't mean equal responses, is the whole room symmetrical? I'm guessing not assuming it has one door. Furniture is usually not symmetrical either.

b) if the sub-eq in the AVR has the capability to time align then this is the preferred option, assuming you're not giving up channels.

As for preference:

1) four subs is tough to time align but I'd rather have two than a pair. You haven't considered them in left/right pairs, any reason why not? Again, if the fronts have a similar response and the rears have a similar response (magnitude and phase) then consider them one sub. This is actually what I do.

2 and 3) completely impossible to say for definite, too much depends on the room. One isn't always better than the other. The sub response is a function of the sub and listener's placement in the room, and the room itself.

As a good rule of thumb best placement options are usually: corners, 1/3, 1/4 and 1/2 wall lengths. And these can be mixed. So you might find that a sub 1/4 along the left wall from the front and 3/4 along the right wall from the front might work best.

Thank you so much. Yes, the SubEQ HT time aligns each of the subwoofer outputs. So, I think I will go for it with the front and rear pairs, as their time alignment and response should be similar. Totally take on board what you say about the room though and that you can never really call them a pair.

I should've said, sorry, they are symmetrical left/right to the MLP but not front and rear. The rears would be closer to MLP. Hence, suggestion that the pairs would be split front and rear.

Very valuable words from you though, thank you, so like I said I will try as discussed then run Audyssey again and see if the sub sound seems more controlled with just the front subs individually aligned and EQed.

Cheers.
 

Dobbyisfree

Well-known Member
I'm not even sure if any of you on this thread use Audyssey but this response from them may be of interest regardless:

 

The latest video from AVForums

AVForums Movies Podcast: Which is the best decade for horror movies?
Subscribe to our YouTube channel

Latest News

Hisense TVs get Disney+ on VIDAA smart OS
  • By Andy Bassett
  • Published
Astell&Kern launches entry level SR25 MKII portable music player
  • By Andy Bassett
  • Published
LG rumoured to release 97-inch OLED TV in 2022
  • By Andy Bassett
  • Published
What's new on Blu-ray (UK) for Halloween 2021
  • By Mark Costello
  • Published
AVForums Movies Podcast: 20th October 2021
  • By Phil Hinton
  • Published

Full fat HDMI teeshirts

Support AVForums with Patreon

Top Bottom