Umc quick setup

[snodog]

I can try and rerun it when my neighbors arent home, its no problem at all. I would like to see what others say about the settings as well, I will also take some pictures of my room, I will borrow a camera from work. It is just a bit strange to me because from where my couch is to the front speakers there is nothing but open space. It seems like if there was a problem it would be due to sound waves bouncing off walls or something and the mic reading that vs maybe if the volume was a bit lower it wouldnt be as sensitive to the oddities? Not sure though as I have no experience with this. Im not saying there is anything wrong either btw only that its strange there can be such a difference in the same speakers.
I do like it a lot, I wish there was a very good balanced multi-channel surround mode that would work as a universal setting for me for say mp3s and tv. That is the only thing I dont care for and find myself changing inputs frequently as audio quality seems to change from song to song on mp3 format and such. Still trying to get it dialed in just right.

[ksuvet]

klinemj:

31.5 63 125 190 250 500 1K 2.2 4.5 9 18
L +4 -2 +1 +3 +4 0 +1 +1 -3 -4 -2
R +5 0 +2 +5 0 0 +1 +1 -2 -5 -4


Crossovers for both set at 250 hz. Speaker spec = 35hz-25khz.
Hope that helps. I would have posted a table, but I'm not sure how the table button works.

Edit: Well, the formatting on my typing doesn't show the same on the actual post. Sorry it doesn't line up better.

[Nemesis.ie]

Regarding re-running, another run can be done into another of the 3 memory slots.

[klinemj]

ksuvet...note that you have a 4 dB jump going from the 500 to the 250 on the L and a 5 dB jump going from the 250 to the 190 on the right. To me, this means that Emo-Q heard your speakers down by 4 and 5 dB at those frequencies, so it reco'd the big bump right there.

Note that you also have 3 dB jumps up higher...looks like Emo-Q ignored those. I bet it's programmed to ignore results above some frequency.

If so and given the data you showed & based on how Lonnie explained Emo-Q's x-over choice, I would have expected an x-over setting of 190-250. I bet if we could see the full curve behind the bar chart you'd see exactly why 250 was chosen.

It appears to me Emo-Q did its job, and something is creating the result you are seeing. It could be your room OR it be any circuitry between the pre/pro and your speakers...or, it could be the mic and how it's positioned.

Bu, it looks to me like the software did what Lonnie said it is programmed to do.

Other's thoughts on this?

Mark

[venaka]

Feb 17, 2010 0:42:31 GMT -5 ksuvet said:

klinemj:

31.5 63 125 190 250 500 1K 2.2 4.5 9 18
L +4 -2 +1 +3 +4 0 +1 +1 -3 -4 -2
R +5 0 +2 +5 0 0 +1 +1 -2 -5 -4


Crossovers for both set at 250 hz. Speaker spec = 35hz-25khz.
Hope that helps. I would have posted a table, but I'm not sure how the table button works.
This is good to see,
what type of speakers do you have, bookshelves or towers?
Edit: Well, the formatting on my typing doesn't show the same on the actual post. Sorry it doesn't line up better.

[venaka]

Feb 17, 2010 12:11:24 GMT -5 venaka said:

Feb 17, 2010 0:42:31 GMT -5 ksuvet said:

klinemj:

31.5 63 125 190 250 500 1K 2.2 4.5 9 18
L +4 -2 +1 +3 +4 0 +1 +1 -3 -4 -2
R +5 0 +2 +5 0 0 +1 +1 -2 -5 -4


Crossovers for both set at 250 hz. Speaker spec = 35hz-25khz.
Hope that helps. I would have posted a table, but I'm not sure how the table button works.
This is good to see,
what type of speakers do you have, bookshelves or towers?
Edit: Well, the formatting on my typing doesn't show the same on the actual post. Sorry it doesn't line up better.

what type of speakers do u have , tower or bookshelves??
Interesting

[ksuvet]

Swan Diva 6.1

[klinemj]

ksuvet...

Thanks for the info about what your fronts are. But note: it's not what the MFG-er says your speakers will do that determines the x-over setting Emo-Q chooses. If you have not yet watched the video from Lonnie on how Emo-Q sets x-over level, give it a look. Then, in light of that, what do you think about my post above re. Emo-Q's interpretation of the data?

venaka & others, also...your thoughts on the data in light of Lonnie's video?

Thx,
Mark

[snodog]

I dont think Ill be re-running the EmoQ again. I just would like to better understand the gains and frequencies. Is there an informative page on this that is comprehensive enough and easy to understand for anyone. Maybe its just me or maybe its the microphone but the levels on the speakers are just too up and down from one to the other for me to be satisfied with the auto calibration part.

[ksuvet]

I watched the video previously, but just went back and reviewed it now. My situation may be one like he described with the "frequency gap." Here is the entire data set. Note the ? for SBR which doesn't exist, so I entered that data into my SL, for better or worse.

 	31.5	63	125	190	250	500	1k	2.2	4.5	9	18
L	4	-2	1	3	4	0	1	1	-3	-4	-2
C	10	7	5	1	-2	-2	1	-3	-4	-6	-6
R	5	0	2	5	0	0	1	2	-2	-5	-4
SR	10	4	-2	-1	-2	-2	0	2	0	-3	-1
SBR?	10	6	-2	-3	-4	-4	-3	2	0	-2	-1
SW	-4	-2	-2	-4	-3	0	3	1	2	4	10

Crossover settings were 250hz for LCR, 130HZ for surround, 75hz for SW. The SW and surrounds make sense. However, I don't know why C wasn't between 125 and 190hz. Looking at 31.5-125hz on L&R, it looks like the crossover should be around 60hz, which I where I would naturally set it. All that said, I'm not the person to analyze this intelligently, so I could easily be missing something.

[venaka]

Feb 17, 2010 0:42:31 GMT -5 ksuvet said:

klinemj:

31.5 63 125 190 250 500 1K 2.2 4.5 9 18
L +4 -2 +1 +3 +4 0 +1 +1 -3 -4 -2
R +5 0 +2 +5 0 0 +1 +1 -2 -5 -4


Crossovers for both set at 250 hz. Speaker spec = 35hz-25khz.
Hope that helps. I would have posted a table, but I'm not sure how the table button works.

Edit: Well, the formatting on my typing doesn't show the same on the actual post. Sorry it doesn't line up better.

ive found your settings are pretty similar to mine.But emo gave me 100 for the front and 110 for center.
All I know is that if you change the crossovers, your eq must be changed as well..
I set my own eq and saved it on setting two..Took me a long time, set them to zero and replayed the same song, (Hell Freezes Over, Eagles DTS)..And guess what, at the end the only difference was the 31.5 and 63 were like an extra 4 gains..

Conclusion,
just rum emo-q and then tweak to your likings.

[johnnyg]

Feb 17, 2010 18:14:28 GMT -5 venaka said:

All I know is that if you change the crossovers, your eq must be changed as well.

I don't think that is correct. The way I understand it, EmoQ generates an EQ curve for each speaker to flatten its response throughout the entire range. Changing the crossovers will not affect that other than shift signal between speakers.

[venaka]

Not so much on eq, but I felt speaker levels needed changing with crossover changing..
If I would drop crossovers to 40 I would also drop speaker levels.

[dfdo]

ksuvet, snodog: I had the same issue. Emo-Q adjusted the x-over for the front speakers (B&W CDM2s) to 250Hz and crazy EQ levels. Sounded like crap. Only the distance was correct. So I ended up setting everything manually Now the system sounds quite decent.

[snodog]

dfdo, how did you know to adjust the IQ levels? Did you just adjust them all to zero first and go from there, I have way too much bass every damn thing I have on the bass is ridiculous

[klinemj]

ksuvet...as I look at your data, the front and centers both make sense per the video. Front per prior explanation, for the center...look at the 250/190 data...-2 to +1...a 3 dB delta. That's the first 3 dB delta it found as you sweep down from some likely upper limit that is programmed. So, that's likely why it chose the 250 instead of the next lower time it found a 3 dB delta.

The surrounds don't make sense @ 130 - but...there is a more complete response curve behind all this from which the x-overs are set, so what we're looking at is just some discrete points & the truth would lie in the full curve we can't see. If we saw that, there may some odd spike that triggers the 130 choice.

On the SBR...I saw elsewhere that there may be some mislabelling in the software that would explain what you see & would say it's possibly actually the SL as you note.

To johnnyg...yeah, I think you are right. Changing xover levels only affects what goes out of the pre/pro to the speaker.

Now...here's the curious point: why are some many folks getting 250 hz at the front cutoff (which seems way high...)? If each one did what we just did and found the same thing...then of the choices that could explain it (room, pre/pro signal path, speakers, cabling, and mic)...it would seems statistically unlikely that with all the different rooms, speakers, and cabling out there that one of those three things results in identical results of 250 hz front x-over. For each, there are 2 things in common: the UMC-1 and the mic.

This leaves me wondering...could the UMC "think" it's sending out a particular set of frequencies to the front as test tones, but maybe it's not really doing that & is actually providing less dB at a frequency near 250 (so when it compares what the mic says it hear to what it thinks it sent out, it thinks there's a drop)? Or...could the UMC's pre-provided mic have a "hearing problem" and it's not hearing the tone sent out in the ~250 hz range? Either of these would explain an artificial low point near 250 that would trigger a 250 hz x-over on the front.

If others with 250 hz front x-over reco's from Emo-Q would similarly compare their front EQ settings, maybe we're onto something that could help...

At least that's my thoughts, but I'm a Chemical Engineer...not an electrical one or a programmer...

Mark

[snodog]

LOL well we arent mixing ammonia and bleach here but I like what you are saying. I also think the mic needs to be re-examined or like I said a bit ago, the volume level is ridiculous upon testing. It is unlikely that everyone's speakers would be set at 250hz.

[loopinfool]

Feb 17, 2010 18:52:33 GMT -5 klinemj said:

Now...here's the curious point: why are some many folks getting 250 hz at the front cutoff (which seems way high...)?

Do a search for "speaker floor bounce frequency" and you'll see why. I posted about this phenomenon here before. It's really common with tower speakers, and I believe Emo-Q is both too sensitive and should be looking for the lowest dip when setting the crossover frequency.
It looks like those surrounds fall pretty fast below 125Hz, so I can understand that crossover setting. I'd probably try something lower, like 100Hz.

You can try sticking carpet squares or pieces of foam on the floor a couple of feet in front of your L+R speakers. It might this issue, but you'll need to leave them there for the correction curves to match properly.

- LoopinFool

[Mischief]

something I am seeing a lot of is that people see a rated low end frequency for their speakers and expect to hit it. It depends greatly on your room and placement of the speakers. Many speakers say they can hit 35 hz but that is under very specific conditions.

Most speakers need several feet from the back wall and the same on the sides to reach anywhere near max bass extension. In addition, if you have a nice cabinet between your speakers, you are cutting it down even further.

Stuffy couch, just a little more, carpet, take off another few hz.

My point is that it is important to know your system, get a calibration disc and run the sweeps to verify what each speaker can really do. What many people are trying to do is the equivalent of trusting their cruise control to steer their car.

[klinemj]

loopinfool...great lead...that very well could explain the routine 250 hz front x-overs...very well might not be anything at all w/the UMC-1 or the mic. I had not heard of that effect before, but it is always useful to recall that the published speaker frequency responses are from anechoic chambers...

I do like the fact that Emo-Q does allow us to take its initial settings and adjust from there...those who choose can just drop the x-over point to whatever they like to hear and be done w/it. Ideally, in a future upgrade (not an urgent one, IMHO), they would make some programming enhancements as you note (looking for lowest dip, or perhaps lowest sustained dip that doesn't recover, etc...). Software is both a great and a horrible thing...it will do precisely what you program it to do, even if that's not what you really meant for it to do...

And mischief...your cruise control doesn't steer your car? Hmm...I thought mine did. Maybe that explains the wrecks I had before I quit using it...