joz Posted May 16, 2019 Share Posted May 16, 2019 Just when I thought I could look forward to a cheap deqx you’ve gone and liked it! 1 Link to comment Share on other sites More sharing options...
almikel Posted May 16, 2019 Share Posted May 16, 2019 8 hours ago, Satanica said: You've done the in room subwoofer time alignment? Time aligning multiple subs with each other is not too hard. Time aligning a sub with a woofer is harder - I've done my head in on this issue - and I don't have a good answer/solution... ...from what I understand, using the peaks based on the impulse response between woofer and sub and aligning them in time (by applying the appropriate delay, usually to the woofer) won't achieve time alignment...or good summing between woofer and sub. The process of aligning peaks in the impulse response is so simple - bummer it doesn't produce the best outcome. I'm unsure of the best approach to achieve the best integration between woofer and sub. Mike Link to comment Share on other sites More sharing options...
Satanica Posted May 16, 2019 Share Posted May 16, 2019 3 hours ago, almikel said: Time aligning a sub with a woofer is harder - I've done my head in on this issue - and I don't have a good answer/solution... ...from what I understand, using the peaks based on the impulse response between woofer and sub and aligning them in time (by applying the appropriate delay, usually to the woofer) won't achieve time alignment...or good summing between woofer and sub. The process of aligning peaks in the impulse response is so simple - bummer it doesn't produce the best outcome. I'm unsure of the best approach to achieve the best integration between woofer and sub. It's the DEQX method as in their documentation to line up the first major impulses. Why do you think it's not optimal? Link to comment Share on other sites More sharing options...
almikel Posted May 16, 2019 Share Posted May 16, 2019 @Satanica, basically because they're not perfect impulse responses resulting from a linear phase system. https://bobmccarthy.com/phase-alignment-of-subs-why-i-dont-use-the-impulse-response/ and another SNA thread where it was discussed cheers Mike Link to comment Share on other sites More sharing options...
Satanica Posted May 17, 2019 Share Posted May 17, 2019 (edited) 3 hours ago, almikel said: @Satanica, basically because they're not perfect impulse responses resulting from a linear phase system. https://bobmccarthy.com/phase-alignment-of-subs-why-i-dont-use-the-impulse-response/ and another SNA thread where it was discussed cheers Mike If you use the DEQX method they will be time aligned. Your link is regarding phase alignment, not the same thing. I don't think it's important as you really want to concentrate on the overall frequency response of your system, in this instance bass. And remember that this is only bass we are talking about where our ears are (apparently) not particularity sensitive to timing issues. If there are really big phase issues to worry about it will show up in the frequency response. Edited May 17, 2019 by Satanica Link to comment Share on other sites More sharing options...
almikel Posted May 17, 2019 Share Posted May 17, 2019 19 hours ago, Satanica said: If you use the DEQX method they will be time aligned. Your link is regarding phase alignment, not the same thing. I don't think it's important as you really want to concentrate on the overall frequency response of your system, in this instance bass. And remember that this is only bass we are talking about where our ears are (apparently) not particularity sensitive to timing issues. If there are really big phase issues to worry about it will show up in the frequency response. time alignment, phase alignment and smooth frequency response at crossover are all the same thing (in this context). Lining up the impulse response peaks between woofer and sub will result in too much delay of the mains (assuming distance to LP is the same etc etc). It's a reasonable place to start, but then 0 delay is also a reasonable place to start. Somewhere between the 2 is where optimal summing between sub and mains will occur. Mike Link to comment Share on other sites More sharing options...
Mike13 Posted May 18, 2019 Share Posted May 18, 2019 Does anyone know if non windows/Mac based steamers are compatible with DEQX through USB? My SOtM SMS 200 only has a USB out. The DEQX website has a windows USB driver, but I don’t see how this is would apply, unless you are using a player on your PC and use the USB out of that. Link to comment Share on other sites More sharing options...
Satanica Posted May 18, 2019 Share Posted May 18, 2019 (edited) 3 hours ago, almikel said: Lining up the impulse response peaks between woofer and sub will result in too much delay of the mains (assuming distance to LP is the same etc etc). I presume you mean lining up the first major impulse response peaks of sub-woofer(s) and mains. 3 hours ago, almikel said: It's a reasonable place to start, but then 0 delay is also a reasonable place to start. Somewhere between the 2 is where optimal summing between sub and mains will occur. I understand what you're saying but I think the first major impulse is the best place to start and will result in the best overall performance if you do nothing else. Have you confirmed the frequency response of your system specifically through the region where your sub-woofer and mains overlap? You could do this with something like REW. Edited May 18, 2019 by Satanica Link to comment Share on other sites More sharing options...
Satanica Posted May 18, 2019 Share Posted May 18, 2019 11 minutes ago, Mike13 said: Does anyone know if non windows/Mac based steamers are compatible with DEQX through USB? I don't know, but I think DEQX can help and they are known to be good with support. Link to comment Share on other sites More sharing options...
almikel Posted May 18, 2019 Share Posted May 18, 2019 15 minutes ago, Mike13 said: Does anyone know if non windows/Mac based steamers are compatible with DEQX through USB? I would have thought so - you've got a HDP Express II right? It should support it... ...I have the older model HDP3, which doesn't have USB audio in - I use a USB to SPDIF converter Mike Link to comment Share on other sites More sharing options...
almikel Posted May 18, 2019 Share Posted May 18, 2019 12 minutes ago, Satanica said: I presume you mean lining up the first major impulse response peaks of sub-woofer(s) and mains. yes 38 minutes ago, Satanica said: I understand what you're saying but I think the first major impulse is the best place to start and will result in the best overall performance if you do nothing else. Have you confirmed the frequency response of your system specifically through the region where your sub-woofer and mains overlap? You could do this with something like REW. The first major peak is the way it's aligned now - I haven't mucked with anything for ages, and I didn't change the settings after the discussion that lining up the first major impulse is not the best approach... 23 hours ago, Satanica said: I don't think it's important as you really want to concentrate on the overall frequency response of your system, in this instance bass. And remember that this is only bass we are talking about where our ears are (apparently) not particularity sensitive to timing issues. If there are really big phase issues to worry about it will show up in the frequency response. agreed 47 minutes ago, Satanica said: Have you confirmed the frequency response of your system specifically through the region where your sub-woofer and mains overlap? You could do this with something like REW. or DEQX using the "room measurement" process... ...I haven't done any measurements or tweaks to config for a long time - last time there were no significant "suck outs" between sub and mains - with time alignment based on aligning the first major peak...and the room response was OK...and the system is sounding great. ...I've been planning to move the DEQX output gain jumpers a notch, so it's about time I did a full calibration. cheers Mike Link to comment Share on other sites More sharing options...
almikel Posted May 18, 2019 Share Posted May 18, 2019 1 hour ago, Satanica said: Have you confirmed the frequency response of your system specifically through the region where your sub-woofer and mains overlap? You could do this with something like REW. Good call on REW if looking at phase - I've never gleaned anything useful from the DEQX phase graphs...I've gotten motivated to do some measurements /reconfig... Mike Link to comment Share on other sites More sharing options...
Satanica Posted May 18, 2019 Share Posted May 18, 2019 59 minutes ago, almikel said: or DEQX using the "room measurement" process... ...I haven't done any measurements or tweaks to config for a long time - last time there were no significant "suck outs" between sub and mains - with time alignment based on aligning the first major peak...and the room response was OK...and the system is sounding great. Correct me if I'm wrong but the DEQX Room Measurement does not measure the mains and Sub-woofer(s) together but rather separate? 1 hour ago, almikel said: ...I've been planning to move the DEQX output gain jumpers a notch, so it's about time I did a full calibration. Yep, I've done that so that my power amps can potentially be driven to full power. The out of the box gain setting is quite conservative. Link to comment Share on other sites More sharing options...
almikel Posted May 18, 2019 Share Posted May 18, 2019 24 minutes ago, Satanica said: Correct me if I'm wrong but the DEQX Room Measurement does not measure the mains and Sub-woofer(s) together but rather separate? it's been a while since I fiddled...just starting a measurement session now...I thought you could run a room measurement with everything...I may be wrong... ...I've lost my TSA doco for the jumper settings - you don't have it handy by any chance? cheers Mike Link to comment Share on other sites More sharing options...
almikel Posted May 18, 2019 Share Posted May 18, 2019 12 minutes ago, almikel said: ..I've lost my TSA doco for the jumper settings - you don't have it handy by any chance? found it - cheers 1 Link to comment Share on other sites More sharing options...
Mike13 Posted May 18, 2019 Share Posted May 18, 2019 For room measurement it measures the mains and speakers separately. 1 Link to comment Share on other sites More sharing options...
almikel Posted May 18, 2019 Share Posted May 18, 2019 5 hours ago, Satanica said: Correct me if I'm wrong but the DEQX Room Measurement does not measure the mains and Sub-woofer(s) together but rather separate? 5 hours ago, almikel said: I thought you could run a room measurement with everything...I may be wrong... just ran some measurements - you're right - sub and mains are separate measurements... 6 hours ago, almikel said: The first major peak is the way it's aligned now Turns out not correct - I'm running zero delay between mains and sub currently...tricky to remember settings made months ago ...did some REW measurements also - crossover between mains and sub is 50Hz and nothing funky is happening FR or phase wise at crossover (50Hz). Below is left mains + sub 70Hz dip moves in freq when adding right mains - likely SBIR. cheers, Mike 1 Link to comment Share on other sites More sharing options...
marten Posted May 19, 2019 Share Posted May 19, 2019 Mike, how far are your subs from your mains? You say currently there's no electronic delay set to time-align your mains with subs. What do you think would happen if you should artificially introduce a path length delay between mains and subs - eg. separate mains and subs by a 2m (or more) path length difference to the mic position. If you re-run the exact same REW test showing FR and phase plot, would you expect any anomaly to show in that 50Hz region? . Or perhaps, in that low freq range, room standing waves dominate so much it's impossible to detect anything useful in the FR about delayed wavefront arrivals? Link to comment Share on other sites More sharing options...
almikel Posted May 19, 2019 Share Posted May 19, 2019 single sub is about 1m closer to LP than left main speaker I would expect to see some ripples in the FR and phase response, but maybe room modes are too dominant...I can try it tonight...with a bit of luck the mic may even be in the same spot still Mike Link to comment Share on other sites More sharing options...
davewantsmoore Posted May 20, 2019 Share Posted May 20, 2019 23 hours ago, marten said: What do you think would happen if you should artificially introduce a path length delay between mains and subs - eg. separate mains and subs by a 2m . Or perhaps, in that low freq range, room standing waves dominate so much it's impossible to detect anything useful in the FR about delayed wavefront arrivals? Yes. 50Hz waves are 6.8m long. Moving the subwoofer by 1m is 14% of a WL.... ie. quite insignificant. If you moves the subwoofer by 2m, then you would start to enter the territory where you might notice something small (but still fairly in significant). It is a much better approach to put the woofer where ever the room response works best .... and then if you need to adjust the timing, to do that by delaying a channel. Link to comment Share on other sites More sharing options...
marten Posted May 20, 2019 Share Posted May 20, 2019 If room modes can cloak the FR/phase plot effect of inserting a sub vs mains path length difference, I guess an anechoic test would be needed to properly investigate. In a typical living room, does it help at all if you keep the test SPL very low, or do room modes excite to an equal extent for low vs high SPLs ? Link to comment Share on other sites More sharing options...
davewantsmoore Posted May 20, 2019 Share Posted May 20, 2019 42 minutes ago, marten said: do room modes excite to an equal extent for low vs high SPLs ? In general. Yes. ... that being said, you usually need to run relatively high SPL to get a high SNR. Testing should typically be just loud enough that being in the room is uncomfortable. Quote I guess an anechoic test would be needed to properly investigate. Yes - although due to the extreme measurement errors possible.... it is better to simulate these things. Or at least simulate AND measure (to ensure your measurement isn't too far different from what is expected). You can still test in room .... the modes just may make it more difficult to observe what is happening. Link to comment Share on other sites More sharing options...
marten Posted May 20, 2019 Share Posted May 20, 2019 (edited) Thanks Dave, instructive as always. On the "simulate things" - what options do we have for simulating sound arrival at the listening position from two non-coincident sources (mains plus sub) ? Edited May 20, 2019 by marten Link to comment Share on other sites More sharing options...
almikel Posted May 20, 2019 Share Posted May 20, 2019 10 hours ago, almikel said: single sub is about 1m closer to LP than left main speaker I would expect to see some ripples in the FR and phase response, but maybe room modes are too dominant...I can try it tonight...with a bit of luck the mic may even be in the same spot still Mike mic was moved but as Dave said, @50Hz the wavelength is long - and too late in the evening to pull out the calculator to work out % of wavelength for differing delays in the DEQX... ...the FR and phase was pretty good without adding delay to mains or sub for the crossover at 50Hz - I'll leave it as is. Link to comment Share on other sites More sharing options...
almikel Posted May 20, 2019 Share Posted May 20, 2019 (edited) 1 hour ago, marten said: On the "simulate things" - what options do we have for simulating sound arrival at the listening position from two non-coincident sources (mains plus sub) ? In a DEQX context I still think the first major peak of the impulse of sub and woofer is the guide. If your sub lags your woofer, then treat the difference in the peaks as the maximum time delay for the woofer. Start at 0 sub (edit) mains delay and looking at FR and phase in REW, slowly increase delay on the woofer in DEQX to get a smooth FR and phase in REW. If the sub leads the woofer in the impulse response (likely sub is much closer to the LP than the mains), same process, but add delay slowly to the sub and monitor FR and phase in REW You could probably do it all in REW (but adjusting delay in DEQX) - I'm just more used to using the DEQX for finding the impulse response peaks. REW is better to see how drivers are combining and looking at phase. Mike Edited May 20, 2019 by almikel typo Link to comment Share on other sites More sharing options...
Recommended Posts