|
|
Post by fbczar on Oct 23, 2022 23:30:14 GMT -5
Thanks, I see it now. I have been using the HK/Limage setup without Dirac pending repair on one of my subs and it sounds fantastic. Obviously Dirac handles distance settings, but how do you handle distance settings without Dirac with your Rooze setup? Now THAT is a question I struggled with for over a year! If I wasn't using Dirac I could use REW to look at the impulse responses, and I would know which peak to use as the primary for distance and ignore the others. I could enter the Distances right in the processor (roughly 1ms per foot) and tweak until it was perfect. Same issue for those who use the up-firing "Dolby Enabled" Atmos speakers. But with Dirac, it can easily be fooled into using a weaker but earlier peak as the reference for the distance, and that would make the distance wrong by anywhere from one to three milliseconds. So ... first thing I realized was with the speakers 45 degrees to the wall in the Rooze setup, the tweeters had to be on the edge closest to the wall so there's as little as possible difference between the actual location of the tweeter and the reflection. In my case now that's less than a foot (i.e. less than 1ms). Then there was the issue of some of the sound from the rear wave dispersing around enough to be seen by Dirac, and having it flip the polarity of the speaker. That was a big problem and for some time every time I did a Dirac calibration I had to reverse the speaker wires after. Okay so, I added the 2ft wood baffles to the leading edges of the speakers - the edges aiming at the MLP - and they made it harder for the rear wave to get around and also blocked reflections of the rear waves from opposite speakers which were causing imaging issues. Like I said, I chased this a long time! I even used temporary absorbers placed across the edge of each speaker like a T to block the weaker waves just for the MLP measurement since Dirac uses that to determine the distance. That all worked pretty well. But now my final solution after moving the speakers more toward the front wall is to do a narrow pattern calibration ... just 7 points in a 3ft cube around the MLP. The impulses are a little sloppy because Dirac still sees somewhat dispersed reflections off the side walls, but the alignment is correct because there's a definitive large peak for Dirac to align to. LCR are aligned precisely on each other at 0. View AttachmentWell, it seems it is good I have retired. I will need the extra time for Dirac! Of course, the HK/Limage position for Magneplanar speakers (Those interested should Google it) may be totally different compared to your Rooze positioning. I think I may try your final solution first or maybe right after a standard measurement session. I do need to learn a lot more about impulse measurements so I can tell what all those reflections are up to. Right now, without Dirac, the system sounds exceptionally nice so I can always fall back on the status quo. 😊 |
|
|
|
Post by marcl on Oct 24, 2022 6:19:23 GMT -5
Now THAT is a question I struggled with for over a year! If I wasn't using Dirac I could use REW to look at the impulse responses, and I would know which peak to use as the primary for distance and ignore the others. I could enter the Distances right in the processor (roughly 1ms per foot) and tweak until it was perfect. Same issue for those who use the up-firing "Dolby Enabled" Atmos speakers. But with Dirac, it can easily be fooled into using a weaker but earlier peak as the reference for the distance, and that would make the distance wrong by anywhere from one to three milliseconds. So ... first thing I realized was with the speakers 45 degrees to the wall in the Rooze setup, the tweeters had to be on the edge closest to the wall so there's as little as possible difference between the actual location of the tweeter and the reflection. In my case now that's less than a foot (i.e. less than 1ms). Then there was the issue of some of the sound from the rear wave dispersing around enough to be seen by Dirac, and having it flip the polarity of the speaker. That was a big problem and for some time every time I did a Dirac calibration I had to reverse the speaker wires after. Okay so, I added the 2ft wood baffles to the leading edges of the speakers - the edges aiming at the MLP - and they made it harder for the rear wave to get around and also blocked reflections of the rear waves from opposite speakers which were causing imaging issues. Like I said, I chased this a long time! I even used temporary absorbers placed across the edge of each speaker like a T to block the weaker waves just for the MLP measurement since Dirac uses that to determine the distance. That all worked pretty well. But now my final solution after moving the speakers more toward the front wall is to do a narrow pattern calibration ... just 7 points in a 3ft cube around the MLP. The impulses are a little sloppy because Dirac still sees somewhat dispersed reflections off the side walls, but the alignment is correct because there's a definitive large peak for Dirac to align to. LCR are aligned precisely on each other at 0. View AttachmentWell, it seems it is good I have retired. I will need the extra time for Dirac! Of course, the HK/Limage position for Magneplanar speakers (Those interested should Google it) may be totally different compared to your Rooze positioning. I think I may try your final solution first or maybe right after a standard measurement session. I do need to learn a lot more about impulse measurements so I can tell what all those reflections are up to. Right now, without Dirac, the system sounds exceptionally nice so I can always fall back on the status quo. 😊 Yes I timed my retirement just right ... for when I upgraded to the XMC-2, upgraded to Atmos, and got Dirac again a few months later ... then down the rabbit hole! I don't think the HK Limage presents any problems with using Dirac any way you want. The reason is that you have the opposite circumstance from the Rooze ... you have absolutely no toe-in, while the Rooze has nothing BUT toe-in. The most conventional setup has the speakers either woofers inboard with little or no toe-in, or woofers outboard with just a few inches toe-in. But in both conventional and Limage, the tweeters are aimed pretty much right at the listening area so even a fairly wide Dirac measurement pattern will work just fine ... because, Dirac uses just the FIRST measurement at the MLP to determine the distance setting. Yes, it would be good to become familiar with the REW Impulse Response measurement so you can see the timing. |
|
|
|
Post by fbczar on Oct 24, 2022 12:22:52 GMT -5
Well, it seems it is good I have retired. I will need the extra time for Dirac! Of course, the HK/Limage position for Magneplanar speakers (Those interested should Google it) may be totally different compared to your Rooze positioning. I think I may try your final solution first or maybe right after a standard measurement session. I do need to learn a lot more about impulse measurements so I can tell what all those reflections are up to. Right now, without Dirac, the system sounds exceptionally nice so I can always fall back on the status quo. 😊 Yes I timed my retirement just right ... for when I upgraded to the XMC-2, upgraded to Atmos, and got Dirac again a few months later ... then down the rabbit hole! I don't think the HK Limage presents any problems with using Dirac any way you want. The reason is that you have the opposite circumstance from the Rooze ... you have absolutely no toe-in, while the Rooze has nothing BUT toe-in. The most conventional setup has the speakers either woofers inboard with little or no toe-in, or woofers outboard with just a few inches toe-in. But in both conventional and Limage, the tweeters are aimed pretty much right at the listening area so even a fairly wide Dirac measurement pattern will work just fine ... because, Dirac uses just the FIRST measurement at the MLP to determine the distance setting. Yes, it would be good to become familiar with the REW Impulse Response measurement so you can see the timing. Thanks for the advice. Relative to the HK/Limage placement have you any idea why the high frequencies are not rolled off? I was under the impression that high frequencies from the Ribbon Tweeter would be more pronounced with the speaker pointing directly at the MLP, but in my room, without toe-in, the 3.7i's play high frequencies without compromise. I keep seeking a disadvantage to the Limage setup, but I can't find one. |
|
|
|
Post by marcl on Oct 24, 2022 12:37:58 GMT -5
Yes I timed my retirement just right ... for when I upgraded to the XMC-2, upgraded to Atmos, and got Dirac again a few months later ... then down the rabbit hole! I don't think the HK Limage presents any problems with using Dirac any way you want. The reason is that you have the opposite circumstance from the Rooze ... you have absolutely no toe-in, while the Rooze has nothing BUT toe-in. The most conventional setup has the speakers either woofers inboard with little or no toe-in, or woofers outboard with just a few inches toe-in. But in both conventional and Limage, the tweeters are aimed pretty much right at the listening area so even a fairly wide Dirac measurement pattern will work just fine ... because, Dirac uses just the FIRST measurement at the MLP to determine the distance setting. Yes, it would be good to become familiar with the REW Impulse Response measurement so you can see the timing. Thanks for the advice. Relative to the HK/Limage placement have you any idea why the high frequencies are not rolled off? I was under the impression that high frequencies from the Ribbon Tweeter would be more pronounced with the speaker pointing directly at the MLP, but in my room, without toe-in, the 3.7i's play high frequencies without compromise. I keep seeking a disadvantage to the Limage setup, but I can't find one. The ribbon tweeters have narrower dispersion than the midrange panel ... but not that much. And you're sitting far enough away that the angle is not that great. Imagine what I thought about pointing the tweeter at the wall a few inches away and listening only to the reflection! But it worked (albeit with a VERY hard reflective side wall), and of course with some compensation from Dirac. Also I forgot to mention ... the way these programs measure Impulse Response (which is used to align timing), they are much more sensitive to the higher frequencies. That's why I was focusing on the line of sight to the tweeter. In the first year that I listened with the Rooze, I put the speakers back to conventional setup twice and it was immediately evident to me and my wife that the Rooze sounded better in terms of depth and soundstage, and no sound quality compromise. Your observation seems consistent with ours. |
|
