|
Post by ttocs on Jan 17, 2021 13:47:58 GMT -5
Is anyone else getting Subwoofer inverted phase in Dirac? Of the 3 brands of subs I've got, it's only the Martin Logan Dynamo subs that measure correctly as positive phase as seen in the Dirac Impulse screen, but Dirac then inverts the phase. It does not matter what settings I change in the sub, including inverting the phase of the sub itself, Dirac will ALWAYS force it negative. It happens with Center Sub output, Left Sub output, with and without miniDSP, multiple phase settings on the sub, with sub's crossover enabled or not, using the LFE input on the sub or the Line In input, it's all the same result. Negative. I'm stumped. If the sub measures Positive, Dirac changes it to Negative. If the sub measures Negative, Dirac keeps it Negative. This shows Dirac flipping the impulse to negative. In the end it really shouldn't matter since the subwoofer outputs are only used for LFE, so it won't affect any other channel. But still . . . . .
|
|
|
Post by 5channels on Jan 17, 2021 14:08:26 GMT -5
|
|
kcl
Seeker Of Truth
Posts: 4
|
Post by kcl on Jan 17, 2021 18:29:34 GMT -5
New XMC-2 owner and just finished my first Dirac setup. This is with the latest firmware 2.1 and Dirac s/w version for their website
Unfortunately I also got the image shift problem to the right. Specifically when playing CDs the voice that is supposed to be cantered sounds like it comes from the right speaker. Whatโs strange is this is happening when both in direct or reference stereo modes, which the DIrac filter is not supposed to be active. Switching to the empty user slot fixes the image problem.
With movies I donโt sense there is image shift problem with the Dirac filter engaged, but hard to tell for sure. Iโm also not hearing much improvement over not using the filter. So perhaps there is a problem with the Dirac version. Iโll try to redo the measurements again to see if there is a difference.
So far a little bummed about this Dirac experience!
|
|
|
Post by bender6502 on Jan 17, 2021 18:35:57 GMT -5
This does sound like something has gone wrong with the calibration. Do you use a UMIK mic or the Emotiva EMM-1? What settings for volume did you set in Dirac? Mic sensitivity and main volume.. Too low volume is not good and can make some issues you have. High volume is better but makes it uncomfortable during calibration. Dirac will tell you when you set volume too high and start to clip. Do you have screenshot of your filter stage in Dirac? Showing both calibrated result and measured! Including the curtains. Do you use Harman or Dirac curve? Thereโs nothing wrong with you front stage. I donโt know the specs of your front speakers but you could try toe in them slightly towards MLP. Especially when you have them near side walls. This is probably not a hardware thing, itโs either with XMC/EDNIB software or Dirac software. You could try Dirac 3.0.5 or 3.0.9. These versions worked fine for sure. I finally got a chance to recalibrate today with the UMIK-1 and that solved the center stage issue. No more diffused center and back to proper imaging. Now, I can move forward with experimenting with different curves, etc. I'm currently using default Dirac curves. I was skeptical that UMIK-1 will solve this issue. I still don't understand how this could solve the problem. Do I have a defective Emotiva EMM-1? In the end, I used Dirac 3.0.11 to measure and export the filter. Thanks for suggesting this, I would've never guessed that mic was the problem.
|
|
|
Post by megash0n on Jan 17, 2021 19:35:36 GMT -5
Is anyone else getting Subwoofer inverted phase in Dirac? Of the 3 brands of subs I've got, it's only the Martin Logan Dynamo subs that measure correctly as positive phase as seen in the Dirac Impulse screen, but Dirac then inverts the phase. It does not matter what settings I change in the sub, including inverting the phase of the sub itself, Dirac will ALWAYS force it negative. It happens with Center Sub output, Left Sub output, with and without miniDSP, multiple phase settings on the sub, with sub's crossover enabled or not, using the LFE input on the sub or the Line In input, it's all the same result. Negative. I'm stumped. If the sub measures Positive, Dirac changes it to Negative. If the sub measures Negative, Dirac keeps it Negative. View AttachmentThis shows Dirac flipping the impulse to negative. View AttachmentIn the end it really shouldn't matter since the subwoofer outputs are only used for LFE, so it won't affect any other channel. But still . . . . .ย I didn't text back and forth, but my subs showed inverted as well. I purposely inverted the XLR cable going to my sub amp due to the whole "Emotiva inverted amp thing". All the amps and gear in this room is Emotiva with the exception of the Behringer amp and the subwoofers. I am %80 sure that I then swapped the speaker wires feeding each sub, and it still showed inverted in Dirac. It is quite possible Dirac is inverting to make the crossover... Wait.. Never mind, Emotiva's implementation doesn't know anything about the crossovers.
|
|
|
Post by ttocs on Jan 17, 2021 21:37:03 GMT -5
Is anyone else getting Subwoofer inverted phase in Dirac? Of the 3 brands of subs I've got, it's only the Martin Logan Dynamo subs that measure correctly as positive phase as seen in the Dirac Impulse screen, but Dirac then inverts the phase. It does not matter what settings I change in the sub, including inverting the phase of the sub itself, Dirac will ALWAYS force it negative. It happens with Center Sub output, Left Sub output, with and without miniDSP, multiple phase settings on the sub, with sub's crossover enabled or not, using the LFE input on the sub or the Line In input, it's all the same result. Negative. I'm stumped. If the sub measures Positive, Dirac changes it to Negative. If the sub measures Negative, Dirac keeps it Negative. View AttachmentThis shows Dirac flipping the impulse to negative. View AttachmentIn the end it really shouldn't matter since the subwoofer outputs are only used for LFE, so it won't affect any other channel. But still . . . . . I didn't text back and forth, but my subs showed inverted as well. I purposely inverted the XLR cable going to my sub amp due to the whole "Emotiva inverted amp thing". All the amps and gear in this room is Emotiva with the exception of the Behringer amp and the subwoofers. I am %80 sure that I then swapped the speaker wires feeding each sub, and it still showed inverted in Dirac. It is quite possible Dirac is inverting to make the crossover... Wait.. Never mind, Emotiva's implementation doesn't know anything about the crossovers. I'm with ya on the crossover idea, except that it's only the Martin Logan subs that are forced to a negative impulse by Dirac, not the Sumiko or Rel subs. I asked Dirac, so we'll see what Flavio says.
|
|
|
Post by ttocs on Jan 17, 2021 22:51:39 GMT -5
The accessory boxex did not include anything but a mic, stand bracket and USB cable with the Emo interface. Pretty much the same stuff that I had used to implement Dirac with my XMC-1. I just watched an Emo Dirac setup video that featured what looked like an ethernet switch, cables and what looked like a Raspberry Pi. None of those items came with my XMC-2. Given this most recent info, I am led to believe that there might actually be a change in Dirac implementation on the horizon or whoever packed up my box really blew it. To date, Emotiva has been shipping the Dirac Kit separately.
|
|
|
Post by ttocs on Jan 18, 2021 9:08:40 GMT -5
Dirac Support responded to my inquiry about Subwoofer Phase. They said that it's the very first peak that is chosen to represent the phase. In my case this is true. The first tiny peak is what was chosen. Dirac will change this first peak to positive every time, and there is no way to change this behavior. I always thought the phase is reflected by the first-largest peak. Same sub in both plots. One is inverted phase.
|
|
|
Post by marcl on Jan 18, 2021 9:23:11 GMT -5
Dirac Support responded to my inquiry about Subwoofer Phase. They said that it's the very first peak that is chosen to represent the phase. In my case this is true. The first tiny peak is what was chosen. Dirac will change this first peak to positive every time, and there is no way to change this behavior. I always thought the phase is reflected by the first-largest peak. Same sub in both plots. One is inverted phase. View Attachment View AttachmentIt makes sense that they would use the first peak. But as I found, if there is a small first peak that is not really what represents the direct sound then you have to manually flip polarity to make it right. So do your Sumi subs and ML subs both have a small first peak? If so, and if Dirac flips one and not the other, then they didn't answer your question, right?
|
|
|
Post by megash0n on Jan 18, 2021 9:42:24 GMT -5
Dirac Support responded to my inquiry about Subwoofer Phase. They said that it's the very first peak that is chosen to represent the phase. In my case this is true. The first tiny peak is what was chosen. Dirac will change this first peak to positive every time, and there is no way to change this behavior. I always thought the phase is reflected by the first-largest peak. Same sub in both plots. One is inverted phase. View Attachment View AttachmentIt makes sense that they would use the first peak. But as I found, if there is a small first peak that is not really what represents the direct sound then you have to manually flip polarity to make it right. So do your Sumi subs and ML subs both have a small first peak? If so, and if Dirac flips one and not the other, then they didn't answer your question, right? Interesting. Would you recommend that, if one finds that Dirac reversed due to a "small peak", that one should then reverse polarity after measurement? This seems to make sense, but the follow-up is that, if they use that first peak, then are they also doing timing alignments based on that as well? If so, then this seems to be a bigger issue than polarity.
|
|
|
Post by ttocs on Jan 18, 2021 9:44:50 GMT -5
Dirac Support responded to my inquiry about Subwoofer Phase. They said that it's the very first peak that is chosen to represent the phase. In my case this is true. The first tiny peak is what was chosen. Dirac will change this first peak to positive every time, and there is no way to change this behavior. I always thought the phase is reflected by the first-largest peak. Same sub in both plots. One is inverted phase. View Attachment View AttachmentIt makes sense that they would use the first peak. But as I found, if there is a small first peak that is not really what represents the direct sound then you have to manually flip polarity to make it right. So do your Sumi subs and ML subs both have a small first peak? If so, and if Dirac flips one and not the other, then they didn't answer your question, right? Here are two Dirac plots of the same sub Sumiko with different polarity. Dirac chose the first peak in both and made that positive, so it flips it with the same approach as the ML subs. The Sumiko subs have a 12" driver and a 12" passive radiator. The ML subs have only a 12" driver and are much more precise. edit: oops. I forgot to mention these are for a Sumiko sub.
|
|
|
Post by marcl on Jan 18, 2021 9:58:28 GMT -5
It makes sense that they would use the first peak. But as I found, if there is a small first peak that is not really what represents the direct sound then you have to manually flip polarity to make it right. So do your Sumi subs and ML subs both have a small first peak? If so, and if Dirac flips one and not the other, then they didn't answer your question, right? Interesting. Would you recommend that, if one finds that Dirac reversed due to a "small peak", that one should then reverse polarity after measurement? This seems to make sense, but the follow-up is that, if they use that first peak, then are they also doing timing alignments based on that as well? If so, then this seems to be a bigger issue than polarity. Exactly! It can be a bigger issue than polarity. But I think it's rare. Dirac should align subs on the smaller first peak because all subs will have this. But if you have another speaker with a small first peak and Dirac aligns on that, it could be a problem. In either case though, Dirac should normalize polarity based on the first peak. Polarity and time alignment ... two different things ... both need to match up. I had a situation (due to my unique reflected front channel setup) where Dirac aligned the fronts properly but it flipped polarity every time. So after a calibration I crawled back there with my allen wrench and reversed the speaker wires. I have since corrected the problem with repositioning and use of temporary baffles for the MLP measurement. BTW, I wonder if the fact that the Sumi subs have passive radiators is causing some affect on the impulse response and resulting in Dirac seeing them differently from the ML subs.
|
|
|
Post by ttocs on Jan 18, 2021 10:03:14 GMT -5
BTW, I wonder if the fact that the Sumi subs have passive radiators is causing some affect on the impulse response and resulting in Dirac seeing them differently from the ML subs. oops. I just updated the post above to reflect that the plots are for a Sumiko sub. Dirac treats the Sumiko and ML subs the same, using the first peak as the impulse.
|
|
|
Post by jjkessler on Jan 18, 2021 11:18:12 GMT -5
Finally got around to attempting to run Dirac on my RMC-1L and am running into issues.
- Firmware is 2.1 - Submitted my Dirac registration / MAC address some time ago to Emotiva - Double and triple checked to interface box set up, seems correct and ledโs lit up When I go into the RMC-1L speaker menu to select Dirac, I see โNot Availableโ as the only option - At the laptop, Dirac does not see my RMC-1L as a device
Any ideas on what to try next?
|
|
KeithL
Administrator
Posts: 9,945
|
Post by KeithL on Jan 18, 2021 11:46:52 GMT -5
The NOT AVAILABLE message has NOTHING to do with the license. (If the license wasn't active you would get an error message from the Dirac Live software after it connects to the processor.)
The NOT AVAILABLE message means that the processor is unable to communicate with the Network Interface Box over your network. This can happen because: - the processor hasn't gotten a valid IP address (or isn't connected to the network) - the Network Interface Box isn't working right - there's something else odd going on with your network
Assuming that your Raspberry Pi arrived in good condition... The next thing to check is the indicator lights on the FRONT of the Raspberry Pi. The large red LED should light when you power it on and remain lit. There is also a small green LED that sits on the edge of the big red LED. The small green LED should flicker for several seconds... then settle down to an occasional flicker. IF THE GREEN LED FLASHES IN A PATTERN OR FOUR FLASHES AND THEN A PAUSE THEN THE PROGRAM CARD IS BAD. (If this is the case then let us know and we'll send you another microSD program card to try.)
Assuming this is all good the next thing to check is the IP (network) address on the processor itself. You will find this on the Information and Network menu screens. If your processor has a network address of 0.0.0.0 then this is DEFINITELY NOT A VALID IP ADDRESS. If the address starts with 62. then the address is almost certainly NOT VALID. If the address starts with 192. or 10. then it may or may not be valid.
The most common cause of the processor not getting a valid IP address is a temporary fault in the DHCP client on the processor. The easiest solution, which usually works, is to reset the DHCP client in the processor.
To do this... Go into the Setup | Network Menu on the processor... UNCHECK the check box for DHCP... Completely exit the Setup Menu... Now go back into the Setup | Network Menu... And RECHECK the DHCP check box... Again exit the Setup Menu... Turn the processor fully off and on again (rear panel power switch)...
This will usually successfully reset the DCHP client. If this works, the NIB will be able to connect to the processor, and you should be good to go...
Finally got around to attempting to run Dirac on my RMC-1L and am running into issues. - Firmware is 2.1 - Submitted my Dirac registration / MAC address some time ago to Emotiva - Double and triple checked to interface box set up, seems correct and ledโs lit up When I go into the RMC-1L speaker menu to select Dirac, I see โNot Availableโ as the only option - At the laptop, Dirac does not see my RMC-1L as a device Any ideas on what to try next?
|
|
|
Post by ttocs on Jan 18, 2021 11:51:51 GMT -5
Finally got around to attempting to run Dirac on my RMC-1L and am running into issues. - Firmware is 2.1 - Submitted my Dirac registration / MAC address some time ago to Emotiva - Double and triple checked to interface box set up, seems correct and ledโs lit up When I go into the RMC-1L speaker menu to select Dirac, I see โNot Availableโ as the only option - At the laptop, Dirac does not see my RMC-1L as a device Any ideas on what to try next? Check the IP Address for the RMC-1L in Menu:: Information. If it has an IP and is on the same network segment as the Emotiva Dirac Network Interface Box (EDNIB), then the RMC-1L should be able to enable the checkbox for Dirac. If there is no IP address, check that DHCP is enabled. Then, the quickest way to get it working is to disconnect the power to the EDNIB, turn off the rear power switch for the RMC-1L for more than one minute, then switch back on, wait for it to boot up completely to On, then reconnect the EDNIB. 192.168.1.xxx is not the same segment as 192.168.0.xxx.
|
|
|
Post by megash0n on Jan 18, 2021 12:05:16 GMT -5
Finally got around to attempting to run Dirac on my RMC-1L and am running into issues. - Firmware is 2.1 - Submitted my Dirac registration / MAC address some time ago to Emotiva - Double and triple checked to interface box set up, seems correct and ledโs lit up When I go into the RMC-1L speaker menu to select Dirac, I see โNot Availableโ as the only option - At the laptop, Dirac does not see my RMC-1L as a device Any ideas on what to try next? also verify you actually have a link light on the network connection on the back of the processor.
|
|
|
Post by jjkessler on Jan 18, 2021 12:40:42 GMT -5
I followed the DHCP reset steps, no success
On the Dirac module, I get 4 quick green flashes that repeat
Taking a step back
I took everything off my GigE switch and isolated the RMC-1L,, the Dirac Interface box, and the hardwired internet connection to just the Emotiva provided switch. Status lights all active at the Emotiva switch
Reset my internet gateway for 1 min as well as the RMC using the same DHCP reset steps
Dirac live still showing as unavailable on the RMC preset
IP address at the RMC is showing 10.0.0.2 Gateway is 10.0.0.1
At the laptop, no issue logging into Dirac but, no devices found (underway I need the RMC Dirac to be enabled)
|
|
|
Post by megash0n on Jan 18, 2021 12:48:08 GMT -5
I followed the DHCP reset steps, no success On the Dirac module, I get 4 quick green flashes that repeat Taking a step back I took everything off my GigE switch and isolated the RMC-1L,, the Dirac Interface box, and the hardwired internet connection to just the Emotiva provided switch. Statue light all active at the switch Reset my internet gateway for 1 min as well as the RMC using the same DHCP reset steps Dirac live still showing as unavailable on the RMC preset IP address at the RMC is showing 10.0.0.2 Gateway is 10.0.0.1 At the laptop, no issue logging into Dirac but, no devices found (underway I need the RMC Dirac to be enabled) Log into your router to verify both addresses for the Raspberry Pi and the RMC. From your laptop, see if you can ping both successfully. If you can, I would recommend unplugging the power from the Pi, waiting a moment, then plugging it back up. There are other recommendations here on how to shut it all down, but assuming you have network connectivity (ping) to both devices, my only guess is the Dirac software Emotiva has on the Pi isn't running or it isn't firing its services because it isn't "seeing" communication on the port it is expecting to. I'd call their support number immediately and ask for help troubleshooting. I'm not sure if they will, but they do have the ability to log into that Pi and verify it is working correctly. (Over a Zoom or something meeting)
|
|
|
Post by ttocs on Jan 18, 2021 12:48:23 GMT -5
I followed the DHCP reset steps, no success On the Dirac module, I get 4 quick green flashes that repeat Taking a step back I took everything off my GigE switch and isolated the RMC-1L,, the Dirac Interface box, and the hardwired internet connection to just the Emotiva provided switch. Statue light all active at the switch Reset my internet gateway for 1 min as well as the RMC using the same DHCP reset steps Dirac live still showing as unavailable on the RMC preset IP address at the RMC is showing 10.0.0.2 Gateway is 10.0.0.1 At the laptop, no issue logging into Dirac but, no devices found (underway I need the RMC Dirac to be enabled) Please disconnect the EDNIB and try the approach I posted above. Everyone who has tried that method has had 100% success, so long as there is not a problem physically or electrically with any of the components. This won't change anything about getting the MAC address to Dirac, but it will allow the RMC to get an address. The EDNIB must be disconnected first. This is simple and only takes two minutes.
|
|