|
|
Post by hsamwel on Sept 23, 2021 12:14:58 GMT -5
There are always going to be limitations - and differences between microphones.
No microphone has precisely identical response at 0 degrees and 90 degrees. And, even if an expensive microphone provides calibration curves for both, there is no way to enter that information into Dirac. And that's still only two specific directions.
It is certainly possible to use time windowing to get an accurate measurement of the sound coming directly from the speaker - being received at 90 degrees.
And most calibrated microphones are rated for accuracy at 90 degrees and provide a 90 degree calibration curve.
However the "room response" is a combination of sound reaching the microphone from all directions. There is no way to differentiate how much of what sound is arriving from which direction. And, even if there was, it would be of limited usefulness without a full 3D plot of the microphone's response. And that would be useless unless there was some way to enter it into the calculations.
In reality the software was designed to be used with "common calibrated microphones" - like ours or the Umik... So, baked into its operation, there are assumptions about the directional response of the microphone it will be used with...
Therefore it might actually produce less accurate overall results if used with a microphone that performed significantly differently than what it expects...
If the sound hits the microphone at the wrong angle it was calibrated to you’ll get different results. That’s why it’s so important to place the mic straight up for each measure point. Also all microphones used for home usage, like UMIK-1 or UMIK-2 are not industry spec. You can’t expect them to measure as correct as a $5000 microphone. Dirac does as good as it can with the gear you use with it, and also how the gear is used. Sometimes it gets it wrong and you’ll have to redo the measurement. If users follow Diracs recommendations they will get good results more often than not.. 100% repeatability is impossible with the kind of gear we’re using. Also most of us don’t have robots or lasers to place the mic at exact positions. We also measure in homes. This means something may have been moved or added between measurements. I have done about ten complete measurements and all but one sounded great. Although all of them looked about the same in the frequency graphs. I’d say Dirac can repeat its measument pretty good as long as the speaker or room doesn’t do anything really odd. Even with the cheap gear we’re using. Just get the first mic position dead center and pointing straight up. Yes of course.. I agree on most of what you write.. But Dirac doesn’t calibrate based on the quality of the mic or better with any one mic. It calibrates based on what data it recieves from the mic. A better mic would give, not better measurements but rather more accurate. This makes Dirac’s changes just a little different depending on the quality of mic used. Do we, normal users, need a better mic than UMIK-1? Probably not! That’s why Dirac recommends it as a good enough mic to use. Example your own mic, while ok, measures worse (less detailed) than UMIK-1 in the low and high end. This result of input data is what Dirac base its calibration on. On some measurements the small differences between these two mics can make a pretty big difference in percieved sound quality. So yes, IMO a better quality mic would measure even more correct and yield better results. Most mics (EMM-1 and UMIK-1) measure a drop in the highs even though there aren’t IRL.. Because of this many boost their highs to make it ”flat”. The thing is, you should NEVER boost anything in the high or low end because the mics we use don’t measure accurately enough in those regions. You can actually hurt your speakers if played loud enough because of the boosting. |
|
|
|
Post by marcl on Sept 23, 2021 12:31:49 GMT -5
I just installed 3.1.2, Windows version. I loaded my last project created with 3.1.1, and it took a couple minutes and recalculated filters. Not sure if there will end up being a difference in the resulting filters (I may run REW later) but just FYI it did recalculate ... which it does not always do with each new version.
|
|
|
|
Post by davidl81 on Sept 23, 2021 12:54:03 GMT -5
I just installed 3.1.2, Windows version. I loaded my last project created with 3.1.1, and it took a couple minutes and recalculated filters. Not sure if there will end up being a difference in the resulting filters (I may run REW later) but just FYI it did recalculate ... which it does not always do with each new version. I read this real quickly and thought you had just installed Windows 3.11 OS and I was thinking “I guess retro operating systems are a thing now” 🤣🤣 |
|
|
|
Post by marcl on Sept 23, 2021 12:59:17 GMT -5
I just installed 3.1.2, Windows version. I loaded my last project created with 3.1.1, and it took a couple minutes and recalculated filters. Not sure if there will end up being a difference in the resulting filters (I may run REW later) but just FYI it did recalculate ... which it does not always do with each new version. I read this real quickly and thought you had just installed Windows 3.11 OS and I was thinking “I guess retro operating systems are a thing now” 🤣🤣 Hahahaha the first Windows I ever saw (didn't use) was 2.0. I did a lot of work with a power plant process control system that started on 3.0 with a 25MHz 486 PC. The marketing guys wanted screens to open within 2 sec. We joked that we didn't use a stopwatch, we used a calendar. I wonder how long it would take to calculate Dirac filters on one of those? Whoops ... they only had a 40MB hard drive so couldn't load the installer even!  |
|
|
|
Post by marcl on Sept 23, 2021 13:41:17 GMT -5
There is now an option in the cheeseburger menu to "Disable delay/gain". So I wonder what that does for us? Disabling gain, we could adjust Levels manually. But I assume if delay compensation is disabled we would have no way to adjust it ... unless the Distance settings made when not in a Dirac slot would become the defaults. I'm betting this is a curveball to Emotiva KeithL any idea what this means to us, if anything? |
|
|
|
Post by okjazz on Sept 23, 2021 14:19:16 GMT -5
There is now an option in the cheeseburger menu to "Disable delay/gain". So I wonder what that does for us? Disabling gain, we could adjust Levels manually. But I assume if delay compensation is disabled we would have no way to adjust it ... unless the Distance settings made when not in a Dirac slot would become the defaults. I'm betting this is a curveball to Emotiva KeithL any idea what this means to us, if anything? When the delay/gain get disabled, what would Dirac be then useful for? |
|
|
|
Post by marcl on Sept 23, 2021 14:35:00 GMT -5
There is now an option in the cheeseburger menu to "Disable delay/gain". So I wonder what that does for us? Disabling gain, we could adjust Levels manually. But I assume if delay compensation is disabled we would have no way to adjust it ... unless the Distance settings made when not in a Dirac slot would become the defaults. I'm betting this is a curveball to Emotiva KeithL any idea what this means to us, if anything? When the delay/gain get disabled, what would Dirac be then useful for? With delay and gain compensation disabled, Dirac would still do frequency response and impulse response correction. Other processors allow you to manually adjust delay and gain after Dirac creates filters and loads them. We can adjust gain, but delay is disabled when using a Dirac EQ slot. |
|
|
|
Post by okjazz on Sept 23, 2021 15:09:04 GMT -5
When the delay/gain get disabled, what would Dirac be then useful for? With delay and gain compensation disabled, Dirac would still do frequency response and impulse response correction. Other processors allow you to manually adjust delay and gain after Dirac creates filters and loads them. We can adjust gain, but delay is disabled when using a Dirac EQ slot. I got confused when I read it initially. I thought by disabling the gain, it would disable also the gain correction at specific frequencies, which you are calling frequency response. I think it might be more useful if delay/gain toggle are separated instead of having them combined as proposed in the latest Dirac software. What would someone do if s/he want to disable one versus the other? More control is always better. |
|
uwe
Minor Hero
RMC-1
Posts: 20
|
Post by uwe on Sept 23, 2021 15:33:49 GMT -5
When the delay/gain get disabled, what would Dirac be then useful for? With delay and gain compensation disabled, Dirac would still do frequency response and impulse response correction. Other processors allow you to manually adjust delay and gain after Dirac creates filters and loads them. We can adjust gain, but delay is disabled when using a Dirac EQ slot. I suspect this is a way of testing the influence of the gain and delay elements in the case of phase and gain artifacts. But you could also outsource this to external DSP, maybe also in combination with MSO. |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Sept 23, 2021 15:51:34 GMT -5
|
|
|
|
Post by marcl on Sept 23, 2021 15:52:26 GMT -5
With delay and gain compensation disabled, Dirac would still do frequency response and impulse response correction. Other processors allow you to manually adjust delay and gain after Dirac creates filters and loads them. We can adjust gain, but delay is disabled when using a Dirac EQ slot. I got confused when I read it initially. I thought by disabling the gain, it would disable also the gain correction at specific frequencies, which you are calling frequency response. I think it might be more useful if delay/gain toggle are separated instead of having them combined as proposed in the latest Dirac software. What would someone do if s/he want to disable one versus the other? More control is always better. I appreciate the job Dirac does to get the channel gain pretty close, but I always have to tweak a couple db here and there anyway. As for delay, Dirac hits it within much less than 1ms in most cases. I have a particularly unusual situation with my fronts and they sometimes can be off by 1-2ms. It's a very unique situation so I don't know why many others would want to manually adjust delay. |
|
|
|
Post by leonski on Sept 23, 2021 16:00:43 GMT -5
What is the RESOLUTION of the human ear/brain system? I suspect 1ms well below 'distinguisible'/
Conventiona Magnepan Wisdom says to put speakers 5 feet away from the front wll so you can 'tell' the reflection from the first arrival.
Marci uses the ROOSE (ROOZE?) setup with is a respected alternative setup.....IF you have the room
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Sept 23, 2021 16:23:47 GMT -5
I got confused when I read it initially. I thought by disabling the gain, it would disable also the gain correction at specific frequencies, which you are calling frequency response. I think it might be more useful if delay/gain toggle are separated instead of having them combined as proposed in the latest Dirac software. What would someone do if s/he want to disable one versus the other? More control is always better. I appreciate the job Dirac does to get the channel gain pretty close, but I always have to tweak a couple db here and there anyway. As for delay, Dirac hits it within much less than 1ms in most cases. I have a particularly unusual situation with my fronts and they sometimes can be off by 1-2ms. It's a very unique situation so I don't know why many others would want to manually adjust delay. Hi Marcl, Isn't this exactly what REW allows? I refer to the timing alignment tool being separate than the frequency or EQ window. If I am understanding correctly the user may time align the speakers first and then export that information to their desired DSP. Once time alignment is nailed down then the EQ window can be used to generate filters which adjust gain, Q, etc.... I have found my time alignment is seemingly independent of any EQ. In other words nailing down timing alignment I can then forget about it and work with EQ. I just think it curious why someone that is obviously fluent in REW might use Dirac? If the user is double checking Dirac and observing that various adjustments are off by ms, db, etc and going back to correct through the use of REW then why Dirac? |
|
|
|
Post by marcl on Sept 23, 2021 17:34:01 GMT -5
I appreciate the job Dirac does to get the channel gain pretty close, but I always have to tweak a couple db here and there anyway. As for delay, Dirac hits it within much less than 1ms in most cases. I have a particularly unusual situation with my fronts and they sometimes can be off by 1-2ms. It's a very unique situation so I don't know why many others would want to manually adjust delay. Hi Marcl, Isn't this exactly what REW allows? I refer to the timing alignment tool being separate than the frequency or EQ window. If I am understanding correctly the user may time align the speakers first and then export that information to their desired DSP. Once time alignment is nailed down then the EQ window can be used to generate filters which adjust gain, Q, etc.... I just think it curious why someone that is obviously fluent with REW might use Dirac? If the user is double checking Dirac and observing that various adjustments are off by ms, db, etc and going back to correct through the use of REW then why Dirac? Lots of overlapping stuff here so let's see if I can concoct a concise answer ... #1 - Dirac uses a proprietary mix of filters that are more accurate and effective than REW's generation of IIR PEQ biquads. #2 - Dirac does Impulse Response correction, which REW can't do. #3 - If you use a miniDSP to handle multiple subs (in lieu of DLBC) then yes the REW alignment tool is useful for aligning the subs to each other. #4 - If we had access to the delays after Dirac calibration and could override the Dirac settings, we might find some benefit in using the REW alignment tool to tweak the alignment of each speaker to the subs at the specific crossover frequency of that speaker. As has been said ... "hitch the engine to the caboose" ... at the crossover. But the crux of the biscuit is that REW and Dirac are totally different things. |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Sept 23, 2021 18:01:37 GMT -5
Hi Marcl, Isn't this exactly what REW allows? I refer to the timing alignment tool being separate than the frequency or EQ window. If I am understanding correctly the user may time align the speakers first and then export that information to their desired DSP. Once time alignment is nailed down then the EQ window can be used to generate filters which adjust gain, Q, etc.... I just think it curious why someone that is obviously fluent with REW might use Dirac? If the user is double checking Dirac and observing that various adjustments are off by ms, db, etc and going back to correct through the use of REW then why Dirac? Lots of overlapping stuff here so let's see if I can concoct a concise answer ... #1 - Dirac uses a proprietary mix of filters that are more accurate and effective than REW's generation of IIR PEQ biquads. #2 - Dirac does Impulse Response correction, which REW can't do. #3 - If you use a miniDSP to handle multiple subs (in lieu of DLBC) then yes the REW alignment tool is useful for aligning the subs to each other. #4 - If we had access to the delays after Dirac calibration and could override the Dirac settings, we might find some benefit in using the REW alignment tool to tweak the alignment of each speaker to the subs at the specific crossover frequency of that speaker. As has been said ... "hitch the engine to the caboose" ... at the crossover. But the crux of the biscuit is that REW and Dirac are totally different things. Just clarifying, By overlap I'm thinking 1 and 2 go together and the solution is to use another program such as Rephase which could be used with REW's acoustic timing measurements to produce these desired filters? Seems like the REW/Rephase option might be a "bug free" alternative? |
|
LCSeminole
Global Moderator 
Res firma mitescere nescit.
Posts: 20,850
Member is Online
|
Post by LCSeminole on Sept 25, 2021 18:10:10 GMT -5
DiracLive v3.1.2 now available.
|
|
|
|
Post by krauley on Sept 30, 2021 12:11:29 GMT -5
on dirac software page, what is the difference between diraclive-latest and diracliveprocessor-1.4.15 win64. I using a windows machine so obviously not interested in anything Mac.
|
|
|
|
Post by ElectricKoolAid on Sept 30, 2021 12:19:19 GMT -5
diraclive-latest.exe is correct. It's under the very first heading at the top of the page "Dirac Live" which underneath states:
Supported Brands: ARCAM, AudioControl, BRYSTON, DATASAT, EMOTIVA, FOCAL, Integra, JBL, Lexicon, Lucie, miniDSP, MONOPRICE, NAD, ONKYO, Pioneer, quadral+, ROTEL, StormAudio, THETA DIGITAL, WISDOM
|
|
|
|
Post by krauley on Sept 30, 2021 12:25:35 GMT -5
diraclive-latest.exe is correct. It's under the very first heading at the top of the page "Dirac Live" which underneath states: Supported Brands: ARCAM, AudioControl, BRYSTON, DATASAT, EMOTIVA, FOCAL, Integra, JBL, Lexicon, Lucie, miniDSP, MONOPRICE, NAD, ONKYO, Pioneer, quadral+, ROTEL, StormAudio, THETA DIGITAL, WISDOM Thank you for the reply and is the other software a suite of programs similar to REW for those without the processors mentioned above? just curious as to what it is but i dont want to d/l and install just to appease my curiousity.  |
|
|
|
Post by ElectricKoolAid on Sept 30, 2021 13:14:58 GMT -5
You're welcome. Yes, I believe that version is just the stand alone software that runs on Windows. If so, I tried that out a few years ago when the offered a 30 day free trial of it. It's exactly the same as Dirac Live on the Emotiva Processor, just runs as software on your PC. I used to run an HTPC so that was the easiest way for me to test out Dirac Live at the time since not many AVRs/PrePros had it. It's also used by music producers who need a room calibration software for music production I believe.
|
|
