|
|
Post by pepar on Feb 15, 2016 12:02:26 GMT -5
New user here. Lots of home theater and system setup experience, and a Audyssey Pro user.
I used my Cross Spectrum-calibrated Dayton EMM-6, a 50' balanced mic cable (as an extension) and then the XLR-to-USB adapter cable. Win 7 auto-installed the C-Media driver, but I needed to select the (Intel?) HD Audio mic to get any sound into the speaker level and I/O level screen. I loaded the 1/3 octave 90 degree cal file and followed the instructions and when I was done all speaker levels were into the middle of the green with none exceeding it. On to the measurements. BTW, I use the mic pointed up at the ceiling.
Nine positions on the "sofa," i.e. my main listening position and the single seats left and right of it. (This is my rear row of two rows of three seats each.) That completed, I made some minor target curve edits, mostly a 1dB boost in LCR at 80Hz. (BTW, my spkrs system is M&K S-150-based.). They subsequently were given a 90Hz crossover so the rise gave them a bit more fullness at the low end.
After listening to my "system setup" go to movies, I had trimmed my subs down to -8.5dB and -11dB.
I then moved to my final go to system balance check content, Fourplay's "Chant." Here, at this song, I discovered -12dB is the max. So my -9dB and -12dB were all I could do.
Anyone have any thoughts on why I needed to cut the sub levels by so much? My 50' extension cable? My mic orientation and/or choice of cal file??
|
|
|
|
Post by Percussionista on Feb 15, 2016 15:33:17 GMT -5
<abbr>...</abbr> Anyone have any thoughts on why I needed to cut the sub levels by so much? Well, taking a clue from the manual from my own SVS PC-12-plus subwoofer. The bass management screen on the sub itself is used to set all the parameters for the sub before one even gets into AVPs and such. The manual suggests various settings to start with, and for "volume" it suggests starting at -20db. Now there's a big hint! I kind of glossed over this when I first started setting up my sub, and actually had it set at only -5db. The low bass was frightening in movies with big bass hits at that level, and my wife was particularly annoyed ;-) I kept fiddling with the UMC-1 levels for the sub but it was really hard to tame. Finally I RTFM, and saw the initial volume suggestion. So, all I'm saying is that perhaps your sub's 0db level is really hot and needs to start at a much much lower level and then use your measurements to dial it in from there. |
|
|
|
Post by pepar on Feb 15, 2016 15:37:34 GMT -5
Thanks. On the Dirac room correction software, all channels were comfortably within the green meter zone on the levels and output screen before proceeding with the measurements. All of my subs' additional "features" had been bypassed, defeated or otherwise disabled from Day 1.
|
|
|
|
Post by millst on Feb 15, 2016 16:52:37 GMT -5
Emotiva recommended that we not use XLR extension cables with the mic due to attenuation. I have used one before without incident, but it was nowhere near 50ft. Since then, I've switched to their recommended approach of using a USB extension.
-tm
|
|
|
|
Post by pepar on Feb 15, 2016 17:46:08 GMT -5
Yes, the 50-footer goes for tonight's efforts. Somebody "on another forum" posted they use a 12-foot extension with no issue. But then someone else pointed out that a too long mic cable would likely attenuate the highs and not the lows.
Interestingly, Emotiva tech support told me that my subs' gain is too high. When i pointed out that there were in the green, they said they didn't need to be and could be just below it. That's a head-scratcher as I would think that Dirac would simply set them even higher to compensate.
Tonite, tonight.
|
|
KeithL
Administrator 
Posts: 9,929
|
Post by KeithL on Feb 16, 2016 0:49:52 GMT -5
Let me clarify a few things.... 1) The connection between the microphone and its preamp module is analog, and the output of the microphone isn't very low impedance. Therefore, adding an extension cable between them, which may significantly increase the capacitance of the original cable which the microphone was calibrated with, may cause a roll off in the high end. (Since cables vary, we can't say for sure how much effect a given cable would have; however, because extending the wire would increase capacitance, it would almost certainly cause a loss of high end and probably wouldn't affect the bass.) 2) It is also possible to use a USB extension cable between the USB encoder module and the computer. While using a cable that's too long there might cause the microphone to stop working, or to work erratically, it won't affect the frequency response. (So, if you put a USB extension between the encoder and the computer, as long as it works without errors, the accuracy will be maintained.) 3) The procedures in the instructions were designed to help you set levels in Dirac Live that will work well for most users in most rooms. However, those settings determine the levels at which the test itself is run; they do NOT affect the calibration that Dirac Live generates. Therefore, carefully matching those levels before running the test isn't required, and varying them slightly in order to avoid errors won't affect the results of the calibration. If, even after setting the levels as instructed, you see red vertical lines when the test strip is run, or if you get a message about "too many clipped samples" afterwards, then simply try a level a few dB lower. (If your levels are too low you will see any one of several different errors that indicate that the software was unable to distinguish the test signal from the background noise.) Dirac Live doesn't control the levels associated with the microphone gain and output level on your audio card, and those are the ones you are setting when you make those initial settings. Yes, the 50-footer goes for tonight's efforts. Somebody "on another forum" posted they use a 12-foot extension with no issue. But then someone else pointed out that a too long mic cable would likely attenuate the highs and not the lows. Interestingly, Emotiva tech support told me that my subs' gain is too high. When i pointed out that there were in the green, they said they didn't need to be and could be just below it. That's a head-scratcher as I would think that Dirac would simply set them even higher to compensate. Tonite, tonight. |
|
|
|
Post by pepar on Feb 16, 2016 7:49:24 GMT -5
Thanks, Keith. So, ...
1) and 2) - extend the USB stage of the cable and not the balanced mic portion.
3) wow, that give me a lot more understanding of what I am doing and why I am doing it. So I really only need to get the speaker outputs nicely above the background noise level without clipping. Am I reading you correctly that Dirac doesn't set channel levels? Does it set distances?
Jeff
|
|
|
|
Post by pepar on Feb 16, 2016 11:43:00 GMT -5
Am reading the Dirac manual cover to cover and already found where I had missed something important. Apologies for not doing my due diligence.  |
|
|
|
Post by mrg1999 on Oct 26, 2017 14:35:30 GMT -5
If, even after setting the levels as instructed, you see red vertical lines when the test strip is run, or if you get a message about "too many clipped samples" afterwards, then simply try a level a few dB lower. (If your levels are too low you will see any one of several different errors that indicate that the software was unable to distinguish the test signal from the background noise.)
I'm getting the "too many clipped samples error" The clipped samples are coming from the sub (Ultra Sub 12). I've dialed down the output level to about 8 or 9 o'clock, lowered both the gain and output levels on the Dirac software. The other speakers are barely reaching the green area during the recording. I'm wondering if lowering the low pass filter point on the sub could help to correct the clipping?
|
|
KeithL
Administrator
Posts: 9,929
|
Post by KeithL on Oct 26, 2017 14:44:27 GMT -5
That errors does simply mean that the signal level has exceeded clipping. The solution is to turn down the gain on the sub - even more. As long as you don't get an error that indicates that the signal is too low, then you're fine. I would definitely do that before altering the frequency response by changing the filter. If, even after setting the levels as instructed, you see red vertical lines when the test strip is run, or if you get a message about "too many clipped samples" afterwards, then simply try a level a few dB lower. (If your levels are too low you will see any one of several different errors that indicate that the software was unable to distinguish the test signal from the background noise.)
I'm getting the "too many clipped samples error" The clipped samples are coming from the sub (Ultra Sub 12). I've dialed down the output level to about 8 or 9 o'clock, lowered both the gain and output levels on the Dirac software. The other speakers are barely reaching the green area during the recording. I'm wondering if lowering the low pass filter point on the sub could help to correct the clipping?
|
|
