|
|
Post by nickwin on Apr 29, 2015 9:07:51 GMT -5
I dont think this is actually a problem. The only channel that is suppose to get the 10db boost is the LFE channel. This works because the LFE channel is recorded 10db down, after the 10db boost in the pre/pro it's back to where it should be. Bass routed from the mains should have the same gain as the mains, it should not be 10db hot. You can see from the graphs posted on this thread that it's functioning as intended, with routed bass being level with the main, and LFE signals being 10db higher.
If you calibrate your spl levels so that the sub is 10db hot, LFE will be 20db hot, 10db more than it should and routed bass would he 10db higher than the signal from the main.
|
|
|
|
Post by unsound on Apr 29, 2015 9:32:36 GMT -5
I dont think this is actually a problem. The only channel that is suppose to get the 10db boost is the LFE channel. This works because the LFE channel is recorded 10db down, after the 10db boost in the pre/pro it's back to where it should be. Bass routed from the mains should have the same gain as the mains, it should not be 10db hot. You can see from the graphs posted on this thread that it's functioning as intended, with routed bass being level with the main, and LFE signals being 10db higher. If you calibrate your spl levels so that the sub is 10db hot, LFE will be 20db hot, 10db more than it should and routed bass would he 10db higher than the signal from the main. I think you misunderstood my post. I lowered the LFE level and boosted the subwoofer level by the same amounts. So, the LFE output remains as is. However, boosting the sub increases the SPL or the routed bass to match up with the SPL of the fronts above crossover. If I didn't boost the sub, the measured SPL below crossover was about 5 db down from the SPL above crossover in my case. This definitely wouldn't be the case for those people with speakers where the output drops rapidly below crossover. For those with capable towers, depending on the room interaction, Dirac may be applying a strong cut to the mains below crossover. When this is applied to the sub, the output is lower than desired. I hope that clears up what I did. |
|
|
|
Post by novisnick on Apr 29, 2015 9:39:16 GMT -5
I dont think this is actually a problem. The only channel that is suppose to get the 10db boost is the LFE channel. This works because the LFE channel is recorded 10db down, after the 10db boost in the pre/pro it's back to where it should be. Bass routed from the mains should have the same gain as the mains, it should not be 10db hot. You can see from the graphs posted on this thread that it's functioning as intended, with routed bass being level with the main, and LFE signals being 10db higher. If you calibrate your spl levels so that the sub is 10db hot, LFE will be 20db hot, 10db more than it should and routed bass would he 10db higher than the signal from the main. I think you misunderstood my post. I lowered the LFE level and boosted the subwoofer level by the same amounts. So, the LFE output remains as is. However, boosting the sub increases the SPL or the routed bass to match up with the SPL of the fronts above crossover. If I didn't boost the sub, the measured SPL below crossover was about 5 db down from the SPL above crossover in my case. This definitely wouldn't be the case for those people with speakers where the output drops rapidly below crossover. For those with capable towers, depending on the room interaction, Dirac may be applying a strong cut to the mains below crossover. When this is applied to the sub, the output is lower than desired. I hope that clears up what I did. I'm sorry for not reading the whole thread, forgive me, sometimes it's the simple things. Are the subs in phase with each other? This is huge as I'm sure most know but often forget to check. Please forgive if this has been checked/done. |
|
|
|
Post by nickwin on Apr 29, 2015 9:44:33 GMT -5
I dont think this is actually a problem. The only channel that is suppose to get the 10db boost is the LFE channel. This works because the LFE channel is recorded 10db down, after the 10db boost in the pre/pro it's back to where it should be. Bass routed from the mains should have the same gain as the mains, it should not be 10db hot. You can see from the graphs posted on this thread that it's functioning as intended, with routed bass being level with the main, and LFE signals being 10db higher. If you calibrate your spl levels so that the sub is 10db hot, LFE will be 20db hot, 10db more than it should and routed bass would he 10db higher than the signal from the main. I think you misunderstood my post. I lowered the LFE level and boosted the subwoofer level by the same amounts. So, the LFE output remains as is. However, boosting the sub increases the SPL or the routed bass to match up with the SPL of the fronts above crossover. If I didn't boost the sub, the measured SPL below crossover was about 5 db down from the SPL above crossover in my case. This definitely wouldn't be the case for those people with speakers where the output drops rapidly below crossover. For those with capable towers, depending on the room interaction, Dirac may be applying a strong cut to the mains below crossover. When this is applied to the sub, the output is lower than desired. I hope that clears up what I did. My point is that you shouldn't have to raise the subwoofer level. If things are functioning properly the routed bass will be even with the mains signal and the LFE will be 10db higher. If you boost the routed bass so that its no longer a full 10db down from the LFE, your running your routed bass hot (or your LFE cold). It sounds like you are actually experiencing another problem. Most likely a phase cancelation around the crossover which you wouldn't experience on the LFE channel, that would make the routed bass seem weak relative to the LFE channel and could make you think that you need to boost your routed bass. The bottom line is that the LFE channel is suppose to be 10db above all the other channels, including the routed bass from them. Do you have the ability to take some measurements with REW? I bet your getting a big null at the crossover. Does your sub(s) have a phase knob? EDIT: keep in mind that you don't need to calibrate the LFE 10db high, if your using the XMCs internal test tones the subwoofer level compensates for this, so set them all even and the LFE will be 10db hot just like it should. |
|
|
|
Post by unsound on Apr 29, 2015 10:03:31 GMT -5
My point is that you shouldn't have to raise the subwoofer level. If things are functioning properly the routed bass will be even with the mains signal and the LFE will be 10db higher. If you boost the routed bass so that its no longer a full 10db down from the LFE, your running your routed bass hot (or your LFE cold). It sounds like you are actually experiencing another problem. Most likely a phase cancelation around the crossover which you wouldn't experience on the LFE channel, that would make the routed bass seem weak relative to the LFE channel and could make you think that you need to boost your routed bass. The bottom line is that the LFE channel is suppose to be 10db above all the other channels, including the routed bass from them. Do you have the ability to take some measurements with REW? I bet your getting a big null at the crossover. Does your sub(s) have a phase knob? I'm not sure I understand why you think the routed bass will be even with the mains level. From what we've seen, the subwoofer is sent the filters from the main. Tony even thinks these are then combined with the filters from the sub. How can we be sure that the resulting SPL below crossover will match the mains above crossover? I did all these measurements with REW. I don't see any phase-related issues at all. If I change the phase, I can see a huge null develop between 50 Hz and 80 Hz (my crossover). EDIT: My LFE was 10 dB higher than the SPL of mains above crossover. However, it was 15dB higher than SPL below crossover when feeding REW a FL or FR only signal. |
|
|
|
Post by unsound on Apr 29, 2015 10:07:51 GMT -5
I'm sorry for not reading the whole thread, forgive me, sometimes it's the simple things. Are the subs in phase with each other? This is huge as I'm sure most know but often forget to check. Please forgive if this has been checked/done. Yes, I did check for phase. I definitely had a phase issue to begin with. The situation I'm describing is after getting the phase to line up. However, I have run enough measurements to make me dizzy. I will recheck everything at some point! |
|
|
|
Post by nickwin on Apr 29, 2015 10:46:22 GMT -5
My point is that you shouldn't have to raise the subwoofer level. If things are functioning properly the routed bass will be even with the mains signal and the LFE will be 10db higher. If you boost the routed bass so that its no longer a full 10db down from the LFE, your running your routed bass hot (or your LFE cold). It sounds like you are actually experiencing another problem. Most likely a phase cancelation around the crossover which you wouldn't experience on the LFE channel, that would make the routed bass seem weak relative to the LFE channel and could make you think that you need to boost your routed bass. The bottom line is that the LFE channel is suppose to be 10db above all the other channels, including the routed bass from them. Do you have the ability to take some measurements with REW? I bet your getting a big null at the crossover. Does your sub(s) have a phase knob? I'm not sure I understand why you think the routed bass will be even with the mains level. From what we've seen, the subwoofer is sent the filters from the main. Tony even thinks these are then combined with the filters from the sub. How can we be sure that the resulting SPL below crossover will match the mains above crossover? I did all these measurements with REW. I don't see any phase-related issues at all. If I change the phase, I can see a huge null develop between 50 Hz and 80 Hz (my crossover). EDIT: My LFE was 10 dB higher than the SPL of mains above crossover. However, it was 15dB higher than SPL below crossover when feeding REW a FL or FR only signal.Something is definitely wrong. It should be 10db exactly or very close to it. This is simply because Dirac gain matches your mains to your sub. Like ansat said, the routed bass gets the gain structure of the mains. Let me flip this around, why would you assume routed bass would not be level matched to the mains? Yes the EQ filters from your mains speakers could throw that level off to some degree, but because EQ filters are a combinations of boosts and cuts, it shouldn't effect the overall level very much. Look at ansats graph, even taking into account the filter stacking issue (first graph), his routed bass is still essentially even with his mains, and the LFE is still 10db above that just as it should be. Also, you can get around the filter issue by simply moving the lower curtain of your main up to the crossover point which will give the sub ONLY the LFE filters. That said if you look at ansats second graph you can see that removing the mains filters from the sub by moving the curtain barely changed the overall level of the routed bass. Do your mains have a lot of cut applied below the crossover? Thats the only situation I can think of that would have a big effect on the routed bass level. Even then 5db seems excessive. Do you get the same routed bass level from all of your mains? If you are really sure that you have ruled out phase cancelations at the crossover maybe you just have a bad set of measurements or filters? Have you had this same behavior with filters created from a different set of measurements? If you post a graph with your main full range, main with xover, and LFE channel overlaid that might give us more insight. |
|
|
|
Post by unsound on Apr 29, 2015 11:08:32 GMT -5
Something is definitely wrong. It should be 10db exactly or very close to it. This is simply because Dirac gain matches your mains to your sub. Like ansat said, the routed bass gets the gain structure of the mains. Let me flip this around, why would you assume routed bass would not be level matched to the mains? Yes the EQ filters from your mains speakers could throw that level off to some degree, but because EQ filters are a combinations of boosts and cuts, it shouldn't effect the overall level very much. Look at ansats graph, even taking into account the filter stacking issue (first graph), his routed bass is still essentially even with his mains, and the LFE is still 10db above that just as it should be. Also, you can get around the filter issue by simply moving the lower curtain of your main up to the crossover point which will give the sub ONLY the LFE filters. That said if you look at ansats second graph you can see that removing the mains filters from the sub by moving the curtain barely changed the overall level of the routed bass. Do your mains have a lot of cut applied below the crossover? Thats the only situation I can think of that would have a big effect on the routed bass level. Even then 5db seems excessive. Do you get the same routed bass level from all of your mains? If you are really sure that you have ruled out phase cancelations at the crossover maybe you just have a bad set of measurements or filters? Have you had this same behavior with filters created from a different set of measurements? If you post a graph with your main full range, main with xover, and LFE channel overlaid that might give us more insight. I will try and work on this today (unlikely) or tomorrow (more likely). |
|
|
|
Post by unsound on Apr 30, 2015 14:09:20 GMT -5
EDIT: False Alarm. It is indeed a phase issue as you guys were suspecting. I was so sure I'd fixed the phase issue because that's the first thing I thought. I even posted about it in another thread! But, then, I was messing around with 4 subs and forgot that the one I'd actually fixed the phase issue with isn't the one that I have currently connected. Aaaargghhh! But, at least, we're all good now. Thanks for your help. Ok, so here are my measurements. Most of the captions should be self-explanatory, but let me explain anyway: The below 4 are for FL input from REW 1. FL set to Large and no Dirac 2. FL set to Large with Dirac applied 3. FL set to small (80Hz) and with Compensation. The Compensation being a -4.5dB on the LFE Level in Preferences and a +4.5 dB on the subwoofer level in Dirac levels. 4. FL set to small (80Hz) and without Compensation. Please note that this is about 5dB below #2. With compensation, the #3 curve is close to the #2 curve. The below 2 are for LFE input from REW 5. Sub with Compensation. Same compensation as above. 6. Sub without compensation. As you can see #5 and #6 are overlapping and my net result for LFE is unchanged. However, without compensation the difference between the LFE and mains goes up to about 16dB in spots and, on average, is much more than 10 dB. I do think that 4.5 dB was a little aggressive and maybe 3.5 dB would server better.  |
|
|
|
Post by nickwin on May 1, 2015 9:05:26 GMT -5
When your adjusting phase, are you doing it on individual subs (phase aligning them to each other) or to all of them as a whole (phase aligning them as a whole to the mains)? Or did you do both?
You really shouldn't have to do the compensation your doing, but it does look like its fixing the problem. Have you tried a new set of measurements? To me it looks like a bad filter set. If you redo it make sure you pull your mains curtains up to the crossover point if you haven't done that.
Its kind of weird that in your graph adding 4.5db gain to the sub raised the 20z output almost 10db! Yet only 5db at 70z. I actually had similar behavior on my last filter set. The more boost I have the sub the more pronounced my house curve became. The only thing I can think of that would cause this is cancelation around the crossover, but maybe there is something else going on.
EDIT: This is actually looking very similar to the behavior I'm getting. I had to add 2db to my sub to get my routed bass even with my mains at the crossover, but doing so raised my 20z output up about 5db. I ended up toning down my subs house curve to compensate for that.
|
|
|
|
Post by nickwin on May 1, 2015 9:07:46 GMT -5
Considering we still haven't heard peep from Emotiva about this routed bass issue I have to think they are looking into it... And probably found something that they are looking into further. Hopefully we will hear soon.
|
|
|
|
Post by unsound on May 1, 2015 9:21:35 GMT -5
When your adjusting phase, are you doing it on individual subs (phase aligning them to each other) or to all of them as a whole (phase aligning them as a whole to the mains)? Or did you do both? You really shouldn't have to do the compensation your doing, but it does look like its fixing the problem. Have you tried a new set of measurements? To me it looks like a bad filter set. If you redo it make sure you pull your mains curtains up to the crossover point if you haven't done that. Perhaps you missed the Edit section at the very beginning. I'm actually down to 1 sub now and it was a different sub than the one I had adjusted phase for which is why I was seeing what I was seeing. That's taken care of now. When I incorporate my 2nd sub, this is what I'll likely be doing: 1. Use REW to align the phase between the 2 subs 2. Designate the subs as mono in the XMC and use a splitter to them. 3. Run Dirac 4. Check phase response with mains and center. To me, this is where it gets tricky. Trying to get the subs to stay in phase with each other and the front 3 speakers. |
|
|
|
Post by nickwin on May 1, 2015 9:29:09 GMT -5
When your adjusting phase, are you doing it on individual subs (phase aligning them to each other) or to all of them as a whole (phase aligning them as a whole to the mains)? Or did you do both? You really shouldn't have to do the compensation your doing, but it does look like its fixing the problem. Have you tried a new set of measurements? To me it looks like a bad filter set. If you redo it make sure you pull your mains curtains up to the crossover point if you haven't done that. Perhaps you missed the Edit section at the very beginning. I'm actually down to 1 sub now and it was a different sub than the one I had adjusted phase for which is why I was seeing what I was seeing. That's taken care of now. When I incorporate my 2nd sub, this is what I'll likely be doing: 1. Use REW to align the phase between the 2 subs 2. Designate the subs as mono in the XMC and use a splitter to them. 3. Run Dirac 4. Check phase response with mains and center. To me, this is where it gets tricky. Trying to get the subs to stay in phase with each other and the front 3 speakers. Ok, so the last graph you posted was before you "fixed" the problem? Did it just go away when you connected the right sub? I read your edit, but I thought the graph was up to date. |
|
|
|
Post by unsound on May 1, 2015 9:48:17 GMT -5
Ok, so the last graph you posted was before you "fixed" the problem? Did it just go away when you connected the right sub? I read your edit, but I thought the graph was up to date. Yes, the graph was posted before I realized I was using the uncorrected sub. After I got the phase of that sub matched up, the SPLs were looking just like they should without any compensation! I posted the edit after that. |
|
|
|
Post by nickwin on May 1, 2015 9:58:15 GMT -5
Ok, so the last graph you posted was before you "fixed" the problem? Did it just go away when you connected the right sub? I read your edit, but I thought the graph was up to date. Yes, the graph was posted before I realized I was using the uncorrected sub. After I got the phase of that sub matched up, the SPLs were looking just like they should without any compensation! I posted the edit after that. Cool, glad you figured it out. Did the behavior where turning up the sub made the very low end of the routed bass raise more than the upper end go away? I have a similar situation and I'm wondering if I still have a phase issue. |
|
|
|
Post by unsound on May 1, 2015 11:11:26 GMT -5
Yes, the graph was posted before I realized I was using the uncorrected sub. After I got the phase of that sub matched up, the SPLs were looking just like they should without any compensation! I posted the edit after that. Cool, glad you figured it out. Did the behavior where turning up the sub made the very low end of the routed bass raise more than the upper end go away? I have a similar situation and I'm wondering if I still have a phase issue. I didn't check that as I didn't raise the sub. I attributed that to the phase issue because once the crossover has heavily suppressed bass from the mains (farther away from the crossover), the bass from the subs isn't getting canceled out any more. At least, that was my reasoning for the shape. |
|
|
|
Post by nickwin on May 1, 2015 14:38:05 GMT -5
Cool, glad you figured it out. Did the behavior where turning up the sub made the very low end of the routed bass raise more than the upper end go away? I have a similar situation and I'm wondering if I still have a phase issue. I didn't check that as I didn't raise the sub. I attributed that to the phase issue because once the crossover has heavily suppressed bass from the mains (farther away from the crossover), the bass from the subs isn't getting canceled out any more. At least, that was my reasoning for the shape. My reasoning was the same. Nick |
|
|
|
Post by nickwin on May 4, 2015 9:15:14 GMT -5
I'm still hoping Emotive will chime in and help up figure out what were seeing here.
|
|
|
|
Post by unsound on May 4, 2015 9:26:28 GMT -5
I'm still hoping Emotive will chime in and help up figure out what were seeing here. A call to Emotiva might not be a bad idea, especially if you ask for @keithl. If he responds here, it'll be great cos everyone will be able to see the response. |
|
|
|
Post by nickwin on May 4, 2015 9:52:56 GMT -5
I'm still hoping Emotive will chime in and help up figure out what were seeing here. A call to Emotiva might not be a bad idea, especially if you ask for @keithl. If he responds here, it'll be great cos everyone will be able to see the response. Iv talked on the phone and emailed with two different Emotiva techs already and they couldn't tell me exactly what's going on. I think this one needs to go strait to Lonnie or Ray, if anyone knows exactly how the xmc handles routed bass they will. |
|
