SMS-1 Question

[Vaughan100]

I have a question regarding the SMS-1 that I was hoping someone could help me answer.

I have two presets, one is my own equalized setting and the other is the unaltered, raw response. I would like to calibrate both presets to the same level ie I want 73-74 dB when using pink noise to set my subwoofer flat with my mains on both presets.

How would I do this ? Because as soon as I eq my response I find that I must reduce the output by a couple dB's which inadvertently affects my raw unaltered response because then I need to increase the level to match the 73-74 dB level.

I just would like to compare apples to apples in terms of output level for both presets. Help would be appreciated.

Thanks.

[brucek]

I have two presets, one is my own equalized setting and the other is the unaltered, raw response. I would like to calibrate both presets to the same level ie I want 73-74 dB when using pink noise to set my subwoofer flat with my mains on both presets.
.

I don't quite understand how you can have a preset that is the raw response unaltered, but I know little to nothing about the SMS, so I probably just don't understand something it does.

Anyway, your idea is not too practical. Subwoofer pink noise is usually band limited from 30Hz to 80Hz. If the raw measure had a high peak in that area, the energy provided by the peak would result in a lower calibration than is reasonable. This is the reason for equalization, so you can have smooth response across the subwoofer band.

brucek

[Vaughan100]

There are six presets with my SMS-1. Preset 6 is the default "raw" response. So I can go back and forth between the normal uneq'ed response and the altered eq'ed response "preset 5".

I just want to be able to calibrate both presets so that I can switch between the two and maintain an accurate comparison (in terms of level).

--Regards,

[Vaughan100]

But you won't get an accurate comparison in terms of level regardless, except at the one point at which you match them. That's the whole point of EQ, after all.

That's the thing. In order for there to be fair comparison, I need to level calibrate both presets to the same level.

Here are some new graphs I worked on today. Before I post the graphs, just a bit of information. I had to position my subwoofer about half a meter out from the corner. Setting on AVR is small, 80 hz crossover. Crossover disabled on the SMS-1.

Software version 2.1. 15 hz subsonic limiter and a 6 dB slope used. Unfortunately I don't have the luxury to place the subwoofer where I would like but I have to work with what I've got I guess.

This is my unequalized graph (I have maintained the same settings as before in the menu :



This is my equalized graph :



This graph is equalized except I closed the door directly behind me :



Then, what I did was I took my unequalized graph and calibrated so that my subwoofer was producing 73 to 74 dB's using pink noise on AVR. This gave me a level setting on the Velodyne as 37.

Due to input distortion that might be a problem with the SMS-1 (according to Home Theater Shack) I took my sub level and increased it to -10 (it's lowest setting).

This is the uneq'ed graph but calibrated flat in term of level :



I think I may have increased the graph levels slightly (about 81-82 dB's more or less)

And finally, my eq'ed graph (preset 5) calibrated using pink noise on AVR. In order for me to reach 73-74 dB's, the SMS-1 volume level of 15 was perfect.



Listening to music, I find that bass sounds a fuller than before and sounds great on some music but boomy on others (more so than it was before). It sounds full but then on some music it almost sounds overly boomy.

Movies suffer from the same problem as before. In the dojo sequence in the Matrix, Neo falls down hard on the mat. This has 20 hz material in it.

Unequalized, I can clearly perceive a greater depth to the low frequencies. Neo falls down hard several times within the sequence and there is deep bass there. In this mode, the hits had power and depth.

The equalized results were again, really disappointing. It's almost as if the results from the graphs do not reflect what I'm hearing/feeling. I increased bass across the range of frequencies. I should be getting increased perceived level of bass but it's the complete opposite. I get audibly less impact, or extension on the same scenes.

Why on earth would this be ?

I'm really not sure why this is but I might not be doing something right (which is definitely a possibility). I suspect that the subsonic limiter is kicking in more than it should on the SMS-1.

Thoughts, suggestions would be most welcome.

--Regards,

[brucek]

Why on earth would this be ?

Usually when someone prefers their unequalized response it's a result of the peaks that are present before EQ that gives the impression of impact at those 'single notes'. If after listening to the unequalized response for a while, then dial down the EQ until you like the sound.

If you're concerned about the SMS not representing a true response in its graphs, then measure the system with REW.

Due to input distortion that might be a problem with the SMS-1 (according to Home Theater Shack) I took my sub level and increased it to -10 (it's lowest setting).

Yeah, I might not get too aggressive with that. You're trading overall dynamic range and S/N ratio for possible momentary frequency dependant distortion at very high levels......

brucek

[Vaughan100]

Usually when someone prefers their unequalized response it's a result of the peaks that are present before EQ that gives the impression of impact at those 'single notes'.

That's the thing, Brucek. I had quite a few dips and nulls before I eq'ed. I basically raised the levels a lot inbetween 50-80 hz to solve that issue. I also raised output at 25-40 hz or so a little bit too so the end result is that I should be getting more deep bass output and not less.

But I seem to be getting less perceived deep bass. I popped in Revenge of the Sith, Chapter 3, and when that star destroyer flies past the screen, with uneq'ed response, the flyby was powerful and deep. With eq'ed response, that deep powerful flyby was just meh.

I don't understand why this is. The only thing I can think of is the subsonic filter being overly aggressive on the SMS-1. I don't know.

--Regards,

If after listening to the unequalized response for a while, then dial down the EQ until you like the sound.

Yeah, I might not get too aggressive with that. You're trading overall dynamic range and S/N ratio for possible momentary frequency dependant distortion at very high levels......

What would you advise me to do ?

[brucek]

What would you advise me to do ?

You've lowered the sub output trim on your receiver to its lowest setting. You're hoping to compensate for that by turning up your subwoofer amplifier. That's fine for the louder levels, but what about the softest levels that are now lost in the noise when they come from the receiver and so the SMS ADC doesn't have any signal to work with. That definition is lost.

If we assume Ilkka's graph of the SMS is valid for all SMS's - and I can't imagine it is - It would appear that at moderate input levels the SMS is great. So, if you have a 'normal' level feeding the SMS, then normal levels down to the softest levels pass through and are digitized properly and all is well. Only at the highest input levels would there appear to be any distortion. The highest test levels Ilkka used in that test was 1.5 voltRMS. Believe me, most consumer receivers are lucky to output anything above 1 voltRMS at full volume. Do you operate your system at full volume?

brucek

[Vaughan100]

So are you saying that I should decrease my sub gain volume and increase the sub level in my AVR to get the flat 73-74 dB's ? Or to decrease the SMS-1 volume and increase the level in my AVR to reach the same level ?

In order for me to reach the 81-82 dB level on the graph I simply increased my master volume to -9 (from -80 to +10). I obviously used the same -9 MV on both presets. Based on this, with MV at -9, with preset 5 which is my custom EQ, I increased the levels of the large dips and had to increase slightly, the levels at 20 hz straight through to around 50 hz.

As you can see from the graphs, the levels inbetween 50 to 80 hz have been increased considerably. That at least should be giving me far greater bass punch.

Would increasing my sub level (in AVR) to get the same level 73-74 dB make that much of a difference and if I were to re-eq, should I first increase the sub level in AVR and perhaps lower the SMS-1 main volume or just decrease the gain on the sub (which is approaching about 30% or so).

Thanks !

--Regards,

[brucek]

You're making this far more complicated than it is.

Bypass the SMS, zero all your receiver trims, run your receiver test tones and balance all your speaker and sub levels with the trims.

Turn on the SMS and equalize the sub.

Once equalized adjust the sub amp to rebalance the sub to the other speaker levels.

brucek

[Vaughan100]

Bypass the SMS, zero all your receiver trims, run your receiver test tones and balance all your speaker and sub levels with the trims.

Alright. I understand that.

Turn on the SMS and equalize the sub.

But as soon as I turn on the SMS-1, it too has it's own volume control. If I use volume 15 which is what I am currently using, then I have to calibrate using all three variables. I need to do a balancing act between the sub level in the AVR, the SMS-1 volume control and the gain control on the sub.

This can be confusing.

Once equalized adjust the sub amp to rebalance the sub to the other speaker levels.

But then how do I compare results using two different presets ? In order for me to do this, I have to increase the volume on the SMS-1 a lot (for unaltered response) to match the mains (using the same sub level in the AVR) compared to the EQ'ed result in which case I must lower the SMS-1 a lot and maintain the same sub level in the AVR.

I can save results for both presets so that they have achieve the 73-74 dB output using the pink noise in my AVR. That is the goal. To get an apples to apples comparison, otherwise how would I compare results ?

It seems complicated but I'm not sure that it's not supposed to be complicated.

Thanks.

--Regards,

[brucek]

If I use volume 15 which is what I am currently using, then I have to calibrate using all three variables

Don't use volume 15 (whatever that is). Use a level that is basically unity gain and equalize. Then touch up with the sub amp volume.

But then how do I compare results using two different presets ?

Bypass the SMS if you want to hear it unequalized - it must have a bypass button. If you used a unity gain on the equalized preset then the overall volume will be about the same between the two.

brucek

[Vaughan100]

What does unity gain mean ?

--Regards,

[Vaughan100]

If you used a unity gain on the equalized preset then the overall volume will be about the same between the two.

I'm not sure I understand. I can bypass the SMS-1 completely by using preset 6 but then the overall levels using pink noise are far lower than normal (using identical SMS-1 volume levels). To calibrate both presets to equal levels require different SMS-1 volume levels.

Remember that I have to juggle between all three variables to arrive at the level of 73-74 dB's.

--Regards,

[brucek]

What does unity gain mean ?

It means 1:1 ratio between input and output. If I feed a volt in, I get a volt out.

You set all the frequency controls to zero and the preset gains and the master gain to some neutral position to cause the unequalized preset to simply pass the signal through unaffected. That's what I call bypassing the SMS.

Then follow the standard routine I outlines above and zero all your receiver trims, run your receiver test tones and balance all your speaker and sub levels with the trims.

Now all the speakers and sub are at the exact same level and the sub is unequalized with signal passing through the SMS.

Now equalize the sub using one of the other presets. The overall level will end up around the same level as the unequalized level. IF not tweak that presets individual volume control

You're done.

[Vaughan100]

I think I understand. I'll try that out.

Is there a reason why my SPL meter is still moving back and forth using pink noise ? It shouldn't be doing that because my FR is about as flat as it's going to get, right ?

I mean, can my frequency response get any better ? If so, please advise me (based on the graphs) on what I should do to improve.

Thanks for your advice Brucek ! Much appreciated.

--Regards,

[brucek]

Is there a reason why my SPL meter is still moving back and forth using pink noise ? It shouldn't be doing that because my FR is about as flat as it's going to get, right ?

An SPL will always move around quite a bit when reading pink noise - perfectly normal.

Your response chart is quite flat - couldn't get much better.......

brucek

[Wayne A. Pflughaupt]

I get audibly less impact, or extension on the same scenes.

Why on earth would this be ?

Possible because the bumps you had before at ~ 40 Hz and 25 Hz are no longer there:

And possibly because you’ve equalized for flat response with no room curve (although the complaint there is usually “thin” sounding bass, not what you’re describing).

Is there a reason why my SPL meter is still moving back and forth using pink noise ? It shouldn't be doing that because my FR is about as flat as it's going to get, right ?

SPL meters always move with pink noise.

Do we really know anything about this processor? The specs at Velodyne’s site are pretty worthless. Are the AD/DA converters any good?

Regards,
Wayne

[Vaughan100]

Wayne, what I don't understand is that I increased the levels across the board. In graph 2, you clearly see that the response from 25 hz and up has been increased to produce a more uniform, flat response.

I mean, from 25 hz and up the levels are 73-74 dB's on the graph. I've increased the levels to match the speaker levels on the graph which is about 80 dB's or so. So I should be getting much improved low end output compared to before.

I don't see any peaks in that response curve and I never brought anything in the deep bass range down. I increased. Apparently the SMS-1 is extremely accurate in it's measuring otherwise I wouldn't have bought one.

--Regards,

[Wayne A. Pflughaupt]

I guess I need to enroll in a “how to get your point across” class.

The broad bump you had at 40 Hz would have lent itself to perceived impact, as would the 80 Hz region an octave above it (that’s the sweet frequency for that kick drum “chest thump” you get at concerts, I’m told by professional sound mixers). Flatten them out = less perceived impact. Which is actually more accurate, because you were hearing an exaggerated effect before.

Same with the 25 Hz. Removing that bump there, perceived extension goes with it.

I don't see any peaks in that response curve and I never brought anything in the deep bass range down.

But raising everything above it (between 25-40 Hz) has the same effect. I.e., if you boost the area between 25-40 Hz, to flatten response, or cut at 25 and 40 Hz, to flatten response, you end up with the same thing either way – flattened response!

The problem probably is, you’re just not used to hearing accurate response. It can take some getting used to.

Again, I suggest moving from flat response to a house curve – let’s see how you’d like that, while you still have some hair left. A lot of people think it sounds more accurate, even if it doesn't "look" accurate.

Regards,
Wayne

[JohnM]

It would also be better to cut the peaks then raise the overall level rather than boost the response either side of the peaks. You are applying a lot of boost which causes extended decay times at the frequencies you boost, making the sound more boomy.

[Vaughan100]

Alright things are much better today. I did what a member suggested and I reversed polarity on the SMS-1. Huge difference. That huge suck out before from 80 to 120 hz has been almost transformed.

I haven't used too much boost this time around. What I did was, I increased my sub level from -10 to -7 and I raised my gain control on my sub. When I did the graph I made sure that my speaker level hit 80 and I matched this for the subwoofer.

I noticed that, yes, the 20-25 hz content is around 3-4 dB's higher than it is unequalized. After this, I popped in some music. Now MUCH better. Everything sounds more balanced and fuller.

The funny thing is that now, my EQ'ed results are not that different from my unEq'ed results (there is a difference, don't get me wrong, but it's not as substantial as I made to make in previous attempts). The main culprit was flipping polarity from + to -.

Watching films, the deep bass is still there but just not quite as strong as after equalization which I understand.

I will post updated graphs today to confirm.

John, if I'm shooting for an 80 dB graph level so that everything inbetween 20 to my crossover frequency is flattish to 80 dB then all I need to do is to cut or boost to achieve that result, right ?

What do you mean by raising the overall level ? If I cut 20 - 30 hz and then boost the rest until around 70 hz in which case I then cut, is that not achieving the same result ? This time around, I almost had to cut 80 hz down all the way. This time around things were different.

The level calibration I'm starting to get confused (and believe me, I am) with because if I achieve the flat result using a baseline volume on SMS-1 at 19 (for both unequalized and equalized graphs) then go and raise the levels using pink noise in AVR afterwards, I end up increasing the levels by 2-3 on the SMS-1 volume so the levels then become 21 on the SMS-1.

This is what I need to achieve 73-74 dB's for the subwoofer level (both in equalized and non equalized level). BUT....but and this is what I'm getting confused with... if I THEN go to the SMS-1 graph display, will my bass then not be out of proportion because I just raised the volume by 2-3 to give me the volume of 21 ?

This is my stumbling block.

Help would be appreciated.

--Regards,

[Vaughan100]

Alright, some new graph's.

Unequalized :



Equalized :



My last attempt in class :



Tell me what you think ?

[Sonnie]

You can place the <img> tags around your link instead of <url> tags and the images will be visible within your post...

[Vaughan100]

I thought I tried that and nothing happened ! Gee whiz. .

I tried the same thing over at Audioholics and nothing happened. I just put [img] "picture file"[/img]

Is that right ?

--Regards,

[Sonnie]

Yep... or click on the icon and type or paste in the URL of your image and click OK.