|
|
Post by geebo on Jul 16, 2020 12:51:16 GMT -5
Harmon curves are โTarget curvesโ, essentially what you want your response to look like after Dirac is through. These are alternates to the generic house curve that Dirac supplies, I link them in the 2nd post. I also suggest where to keep them in my โbetaโ post above. So you do this by dragging the curve like in Dirac Full ? to where it would match these Harman Curves, this whole new Dirac is going to be an experiment, my issue of always running it for my XMC-1 was trying to get a quiet household, with a 5 year old and 9 year old this is far and between from happening haha. Edit: I see now the links to the curves, so whats the point in running Dirac if you can just load a curve and upload it to the processor ? Just as in the full version of Dirac for the XMC-1 there is an option to save or load target curves. You simply download the curves then load them to your set of measurements. You cannot just load the target curves into the processor. You have to load them into Dirac after you've run a set of measurements. The target curves are not filters but rather what you want your filters to be. |
|
|
|
Post by creimes on Jul 16, 2020 12:58:28 GMT -5
I haven't done enough research and such, it's like a whole other world haha.
|
|
|
|
Post by marcl on Jul 16, 2020 12:59:13 GMT -5
Cool! If I have trouble sleeping I can put it on loop :-) Seriously .... cool to have these videos coming along! The basics are important. I do think that rather than watching the measurements several times it would be useful to talk about the following: - More detail on setting the input level ... is the target -30db? I think it was -24db in Dirac 1.0 - How to decide on the measurement pattern, and how to space each measurement from the others - How to use the "curtains" - How to add and delete control points to customize the target curve - Are there recommendations or limitations on the steepness or amplitude of curves and adjustments? Limitations on flattening nulls in the response? - Emphasize that when the filters are created, you are only seeing a predicted - not actual - resulting response Iโll echo Lsc โs comments regarding trial and error, but would also say that though some of us have a head start, the collective โweโ will be learning much of this together (and you are already in the front row). Iโll also say that along with the Emo and Dirac setup guides and videos, I liked the Mehau links I provided, he had some good processes for setting mic and speaker levels that were helpful when the โone size fits allโ approach didnโt work. I understand! I wrote the questions mostly for the benefit of others. I did many, many dozens of calibrations with the XMC-1 and tried some very unconventional things like using Dirac as a crossover and compensating for large mains running in parallel with subs. From what i see 3.0 works just like 1.0 with a new UI. It would be helpful to get input from Dirac on any limitations on boost, cut and Q. |
|
|
|
Post by jjkessler on Jul 16, 2020 20:54:52 GMT -5
I am working on installing my RMC-1L (upgrade from XMC-1) where besides running DIRAC, I never truly dialed in my two SVS PB-12 Ultra subs 70% home theater, 30% stereo/music listening).
If connected to the two non-LFE sub outputs on the RMC-1L and configured as dual mono, how does DIRAC view those two subs? As a summed single mono sub? What amount of corrections can Dirac make (time alignment, PEQ,ETC)? What Iโm struggling with is would a MiniDSP along with DIRAC make sense and what exactly does that buy me? I understand neither can totally correct poor physical sub placement and all room dynamics.
What do I let the MiniDSP handle and what do I let Dirac handle for the best integration into my room (20x31x9) subs and L R set up at 1/4 width ? Listening area is four theater chairs about 14โ from the front speakers
|
|
|
|
Post by AudioHTIT on Jul 16, 2020 23:04:19 GMT -5
If connected to the two non-LFE sub outputs on the RMC-1L and configured as dual mono, how does DIRAC view those two subs? As a summed single mono sub? Iโll answer this question and let someone else take the mini-dsp. Dirac sees each sub as a separate channel and speaker and equalizes it on its own without regard to what the other is doing, thatโs how mine are currently setup, but I plan to Y them both into the center output, as soon as I get time to pull my cabinet out. Then it has no choice but to see the two as a single device and equalize the group. |
|
|
|
Post by markc on Jul 17, 2020 3:40:58 GMT -5
No, you will download the firmware when itโs officially released. No, the Emo mics are โgroup calibratedโ, the file still improves the accuracy. You can download DL anytime, though you canโt do anything without the firmware, so probably best to wait. Not even group calibrated. That would produce a calibration file for each group. There is one single EMM-1 calibration file. Keith answered this before. This EMM-1 calibration file is the result of a group testing of a set of Emotiva mics back in 2015. Subsequent production EMM-1 microphones were then batch tested to make sure that the sample conformed closely enough to this single calibration file. It's possible this batch testing is not even done anymore. Either way, the calibration file is the average of the results of a specific set of original production microphones which set the standard. (The file is exactly the same as the one supplied in 2015 which is unsurprising if the new devices are manufactured to the same standard using the same components) All subsequent microphones are assumed to conform to this standardised calibration if a sample microphone from that production batch conforms to the calibration file within acceptable variance. |
|
|
|
Post by markc on Jul 17, 2020 4:02:48 GMT -5
|
|
|
|
Post by tngiloy on Jul 17, 2020 5:53:10 GMT -5
Knowing that the 'calibrated' mic in the Emotiva box is not individually calibrated I highly recommend getting a mic from Cross Spectrum Labs. There may be other outlets that do calibrated mic sales, but that's where I got mine and I recommend them highly. The solutions I got with the UMIK-1 mic from CSL compared to the one in the XMC-1 box was night and day better. YMMV, www.cross-spectrum.com/If any of you decide to get a real calibrated mic, you might want to do it soon. With Dirac 3.x soon to be here for the RMC-1/XMC-2 there may be a run on these and they may be back-ordered. Just saying. |
|
|
|
Post by AudioHTIT on Jul 17, 2020 6:12:33 GMT -5
I'd start with the Dirac Live for Emotiva Processors User Manual (Now on version 1.3 rather than the 1.1 version linked to on the first page of this forum (needs updating)) ... Fixed! Thanks. |
|
|
|
Post by AudioHTIT on Jul 17, 2020 7:05:38 GMT -5
Knowing that the 'calibrated' mic in the Emotiva box is not individually calibrated I highly recommend getting a mic from Cross Spectrum Labs. There may be other outlets that do calibrated mic sales, but that's where I got mine and I recommend them highly. The solutions I got with the UMIK-1 mic from CSL compared to the one in the XMC-1 box was night and day better. YMMV, www.cross-spectrum.com/If any of you decide to get a real calibrated mic, you might want to do it soon. With Dirac 3.x soon to be here for the RMC-1/XMC-2 there may be a run on these and they may be back-ordered. Just saying. I added the Cross Spectrum Labs link to the 2nd post, though theyโve gone through the July 14th batch it sounds like they get, test, and release more regularly. Thereโs also a link to get the UMIK-1 directly from mini-DSP, they also come with individual calibration files (on-axis and 90-deg), though not to the same frequency extremes โ miniDSP is 20Hz to 20KHz ยฑ 1dB, CSL is 5Hz to 25KHz no tolerance given. A question about the CSL calibration file(s), they only mention an on-axis file, most Dirac measurements are taken with the mic pointing at the ceiling, because of this Iโve always used the 90-deg UMIK-1 calibration file. Does CSL include an off axis file as well? This also begs the question, what happens when your mic pointing at the ceiling happens to be on-axis with a ceiling speaker? The measurement canโt be as accurate. |
|
|
|
Post by frenchyfranky on Jul 17, 2020 7:15:16 GMT -5
Knowing that the 'calibrated' mic in the Emotiva box is not individually calibrated I highly recommend getting a mic from Cross Spectrum Labs. There may be other outlets that do calibrated mic sales, but that's where I got mine and I recommend them highly. The solutions I got with the UMIK-1 mic from CSL compared to the one in the XMC-1 box was night and day better. YMMV, www.cross-spectrum.com/If any of you decide to get a real calibrated mic, you might want to do it soon. With Dirac 3.x soon to be here for the RMC-1/XMC-2 there may be a run on these and they may be back-ordered. Just saying. I added the Cross Spectrum Labs link to the 2nd post, though theyโve gone through the July 14th batch it sounds like they get, test, and release more regularly. Thereโs also a link to get the UMIK-1 directly from mini-DSP, they also come with individual calibration files (on-axis and 90-deg), though not to the same frequency extremes โ miniDSP is 20Hz to 20KHz ยฑ 1dB, CSL is 5Hz to 25KHz no tolerance given. A question about the CSL calibration file(s), they only mention an on-axis file, most Dirac measurements are taken with the mic pointing at the ceiling, because of this Iโve always used the 90-deg UMIK-1 calibration file. Does CSL include an off axis file as well?
This also begs the question, what happens when your mic pointing at the ceiling happens to be on-axis with a ceiling speaker? The measurement canโt be as accurate. I really like this question. |
|
|
|
Post by 5channels on Jul 17, 2020 7:29:10 GMT -5
|
|
|
|
Post by marcl on Jul 17, 2020 7:37:03 GMT -5
If connected to the two non-LFE sub outputs on the RMC-1L and configured as dual mono, how does DIRAC view those two subs? As a summed single mono sub? Iโll answer this question and let someone else take the mini-dsp. Dirac sees each sub as a separate channel and speaker and equalizes it on its own without regard to what the other is doing, thatโs how mine are currently setup, but I plan to Y them both into the center output, as soon as I get time to pull my cabinet out. Then it has no choice but to see the two as a single device and equalize the group. Picking up the miniDSP ball .... If your subs are nearly equidistant from the listening position you can use the Y cable and adjust the gain with the knob on each sub so they are equal. Then Dirac will see them as one sub and correct the sum of their output and align them with the other speakers. If you have subs more than a foot or so different distances from the listening position you can connect the sub output to a miniDSP2x4 with the Advanced plugin. This allows you to individually set the time alignment and levels for up to four subs. You get the subs aligned to each other and then Dirac does the overall correction. You can use Room EQ Wizard to measure the subs individually while connected to the miniDSP (convenient Mute button for each in miniDSP as well as polarity inverter). In the REW data you look at the impulse response for each sub and determine the exact value of the time delay, up to 7.2ms. Sometimes response can be further tweaked with small adjustments to the time alignment value since its effect is not exactly the same at all frequencies. Attachments:
|
|
|
|
Post by cwt on Jul 17, 2020 7:40:52 GMT -5
A question about the CSL calibration file(s), they only mention an on-axis file, most Dirac measurements are taken with the mic pointing at the ceiling, because of this Iโve always used the 90-deg UMIK-1 calibration file. Does CSL include an off axis file as well? This also begs the question, what happens when your mic pointing at the ceiling happens to be on-axis with a ceiling speaker? The measurement canโt be as accurate. Found this reference Bruce melp242.blogspot.com/2019/12/understanding-microphone-calibration.htmlinteresting calibration conclusions for sure |
|
|
|
Post by marcl on Jul 17, 2020 7:53:47 GMT -5
I am working on installing my RMC-1L (upgrade from XMC-1) where besides running DIRAC, I never truly dialed in my two SVS PB-12 Ultra subs 70% home theater, 30% stereo/music listening). If connected to the two non-LFE sub outputs on the RMC-1L and configured as dual mono, how does DIRAC view those two subs? As a summed single mono sub? What amount of corrections can Dirac make (time alignment, PEQ,ETC)? What Iโm struggling with is would a MiniDSP along with DIRAC make sense and what exactly does that buy me? I understand neither can totally correct poor physical sub placement and all room dynamics. What do I let the MiniDSP handle and what do I let Dirac handle for the best integration into my room (20x31x9) subs and L R set up at 1/4 width ? Listening area is four theater chairs about 14โ from the front speakers More specifics on miniDSP .... MiniDSP has PEQ on the two inputs and also on each of four outputs. Some suggest knocking down the big peaks with these filters and letting Dirac do the fine tuning. I tried both ways and found no reason to go through the extra effort with the miniDSP PEQ. Dirac has much more sophisticated filters and the result is excellent. The key is to adjust the Dirac target curve to correct at the level of the lowest null, since it's nearly impossible to lift a null caused by reflections. I was able to get near flat measurements this way. Once you get the best results with Dirac, if necessary you could make fine adjustments in the miniDSP. BTW .... in my current setup, for movies I use subs connected to the miniDSP driven from the Center sub output LFE-only; for music I split the Left sub output (bass management only) with a Y and send one side to the miniDSP where a crossover sends only <50Hz to the subs. The other side of the Y goes to Magnepan woofer modules that do a better job on bass management from 50Hz upward. When we get Dirac, I'll leave the setup the same and let Dirac correct each sub output on its own. |
|
|
|
Post by frenchyfranky on Jul 17, 2020 7:59:10 GMT -5
Very interesting this interview, for you AudioHTIT , he answers to the above question about microphone pointing ceiling speakers at the 47:00 minute. |
|
lokyc
Minor Hero
Posts: 71
|
Post by lokyc on Jul 17, 2020 8:11:03 GMT -5
I have a cross spectrum labs emm6 analog mic and it comes with 0, 45 and 90 deg calibration files.
Shame about the bundled one being poorly calibrated.
Regarding subs, does Dirac not support multiple subs? I have 2 stacked on top of each other, toying with idea of separating them to either side of MLP. For space reasons i sit right against back wall. No sure if that would help anything.
|
|
|
|
Post by jjkessler on Jul 17, 2020 8:17:45 GMT -5
What I noticed with my dual subs and the XMC-1 with Dirac is they were both independently measured however, when running together in real world movie viewing, the output was too high. I really think they needed to be measured as both subs were one and plan on trying the Y splitter idea, running both subs from a single output as set up my RMC-1L
|
|
klinemj
Emo VIPs 
Official Emofest Scribe
Posts: 15,089
|
Post by klinemj on Jul 17, 2020 8:19:41 GMT -5
Shame about the bundled one being poorly calibrated. I wouldn't say the bundled mic is "poorly" calibrated. It's batch calibrated like pretty much every mic provided with almost all pre/pro's and AVR's. But, many prefer the individually calibrated mics - and I am one of them. Since you already appear to have one, you are set. Mark |
|
lokyc
Minor Hero
Posts: 71
|
Post by lokyc on Jul 17, 2020 8:34:16 GMT -5
Shame about the bundled one being poorly calibrated. I wouldn't say the bundled mic is "poorly" calibrated. It's batch calibrated like pretty much every mic provided with almost all pre/pro's and AVR's. But, many prefer the individually calibrated mics - and I am one of them. Since you already appear to have one, you are set. Mark Hmm. That's a good point. I was using Acourate PC based DSP until Dolby Vision got me back to pre/pros. Acourate required the mic and output to be the same interface or else a real pain with ASIO64. Looking at the video, it looks like Dirac can use separate interfaces for input and output. Yay! Can keep my mic! (I hope.) |
|
