|
Post by megash0n on Sept 13, 2020 19:23:39 GMT -5
You know...i did see they were higher at one point, but I thought you were still going back and forth leveling them all. Once you started doing this, I skipped about 10 seconds of the video and just assumed you were making them all level. Thanks for pointing that out. I do believe I'm even more confused now. Just for clarity on what you are doing, are you saying that, when measuring, your subs are potentially 10 db louder, then Dirac is potentially doing the same thing it is doing to mine but yours ends up being equal because you had the built in 10 db extra? I hope that makes sense. If I max my amp gain, mine measures 10 db higher on the volume screen, but basically equal on the measurements screen. Here are a couple screen grabs of the subs SPL and speaker SPL during the volume calibration. This is where you can see that the subs are ACTUALLY 10dB higher than speakers. Notice in the first photo, which is after all the other speakers have been leveled, that all the sliders are showing the same (within about 1dB) level "according to Dirac". Again, I'm using the Dirac sliders to adjust everything the same which is about -22dB on every slider including the subs. SPEAKERS: 67dB; ย ย ย ย ย ย ย ย ย ย ย ย ย ย ย ย ย SUBS: 77dB. View Attachmentย ย View AttachmentDuring the Measurement sweeps, the speakers and subs play about the same SPL. SPEAKERS: 84dB; ย ย ย ย ย ย ย ย ย ย ย ย ย ย ย ย ย SUBS: 84dB. View Attachmentย ย View Attachmentedit: I let Dirac do what it wants without any intervention on my part. Then, I would say you are in effect getting the same as me. You're subs start higher. You can see that, if your subs were actually the same on the volume screen, you'd be getting the issue I have. Agree or no? Edit: when I purposely make my subs 10db louder on a meter, Dirac volume calibration shows this as well whereas yours shows it is even when it isn't. We both seem to have odd issues.
|
|
|
Post by bluescale on Sept 13, 2020 19:28:58 GMT -5
My concern is that it will attenuate all speakers down to the subwoofer because of this.. And then, another 10db because it supposedly does this for headroom. If I crank my sub amp gain all the way up, which makes the subs play with the rest, my overall volume is about 10 db higher afterwards. Meaning, I will watch a movie at -15 instead of -5. If it weren't for countless bugs for so long, I might not question it, but any time something doesn't make sense, I ask questions. โบ I donโt think thatโs happening, but I do get your concern. Level wasnโt my issue, but my last Dirac run was clearly not right. And the G3P have had so many issues I donโt trust anything until I can verify it. Sometimes I feel like the XMC-2 is gaslighting me.
|
|
|
Post by megash0n on Sept 13, 2020 19:32:30 GMT -5
My concern is that it will attenuate all speakers down to the subwoofer because of this.. And then, another 10db because it supposedly does this for headroom. If I crank my sub amp gain all the way up, which makes the subs play with the rest, my overall volume is about 10 db higher afterwards. Meaning, I will watch a movie at -15 instead of -5. If it weren't for countless bugs for so long, I might not question it, but any time something doesn't make sense, I ask questions. โบ I donโt think thatโs happening, but I do get your concern. Level wasnโt my issue, but my last Dirac run was clearly not right. And the G3P have had so many issues I donโt trust anything until I can verify it. Sometimes I feel like the XMC-2 is gaslighting me. I know it is happening to me. With all other things being equal, if I leave my sub where Dirac thinks it needs to be, I'm at a volume of -5 for loud watching. When I increase my sub gain by 10 db before measuring, I'm at a -15 for the same level of volume.
|
|
|
Post by bluescale on Sept 13, 2020 19:43:33 GMT -5
I donโt think thatโs happening, but I do get your concern. Level wasnโt my issue, but my last Dirac run was clearly not right. And the G3P have had so many issues I donโt trust anything until I can verify it. Sometimes I feel like the XMC-2 is gaslighting me. I know it is happening to me. With all other things being equal, if I leave my sub where Dirac thinks it needs to be, I'm at a volume of -5 for loud watching. When I increase my sub gain by 10 db before measuring, I'm at a -15 for the same level of volume. Iโm a little confused. Even if you increase the sub gain prior to running Dirac, wouldnโt Dirac just pull it back down to where it thinks it needs to be?
|
|
|
Post by ttocs on Sept 13, 2020 20:23:18 GMT -5
Here are a couple screen grabs of the subs SPL and speaker SPL during the volume calibration. This is where you can see that the subs are ACTUALLY 10dB higher than speakers. Notice in the first photo, which is after all the other speakers have been leveled, that all the sliders are showing the same (within about 1dB) level "according to Dirac". Again, I'm using the Dirac sliders to adjust everything the same which is about -22dB on every slider including the subs. SPEAKERS: 67dB; SUBS: 77dB. During the Measurement sweeps, the speakers and subs play about the same SPL. SPEAKERS: 84dB; SUBS: 84dB. edit: I let Dirac do what it wants without any intervention on my part. Then, I would say you are in effect getting the same as me. You're subs start higher. You can see that, if your subs were actually the same on the volume screen, you'd be getting the issue I have. Agree or no? Edit: when I purposely make my subs 10db louder on a meter, Dirac volume calibration shows this as well whereas yours shows it is even when it isn't. We both seem to have odd issues. My subs are the same as speakers according to the volume screen. Dirac reports the level to be the same as all the rest of the speakers, so I leave that alone. It's only in reality that a SPL meter will show the SPL to be different, BUT, I don't change anything in Dirac, I just use the sliders the way Dirac intends. Dirac knows what it wants to do and I let it. I do not use an SPL meter for anything other than information about "How Things Work", which is the title of a book my Aunt gave me when I was 14. The levels that you see in the video and photos stay that way and are not altered. I don't adjust the sub to be louder. It doesn't matter that it isn't the same room SPL, what matters is that the slider is in the green at the same level as all other speakers in the green as calculated by Dirac for Dirac's purposes. I don't agree that we both have odd issues. The measurement tests I've done, and will continue to do, are simply because I have a desire to know how things work, and, if I change one thing - how does that affect things downstream? I get good results using Dirac with its defaults, in this regard I'm blissfully ignorant.
|
|
|
Post by megash0n on Sept 13, 2020 20:37:14 GMT -5
Then, I would say you are in effect getting the same as me. You're subs start higher. You can see that, if your subs were actually the same on the volume screen, you'd be getting the issue I have. Agree or no? Edit: when I purposely make my subs 10db louder on a meter, Dirac volume calibration shows this as well whereas yours shows it is even when it isn't. We both seem to have odd issues. My subs are the same as speakers according to the volume screen. Dirac reports the level to be the same as all the rest of the speakers, so I leave that alone.ย It's only in reality that a SPL meter will show the SPL to be different, BUT, I don't change anything in Dirac, I just use the sliders the way Dirac intends. Dirac knows what it wants to do and I let it. I do not use an SPL meter for anything other than information about "How Things Work", which is the title of a book my Aunt gave me when I was 14. The levels that you see in the video and photos stay that way and are not altered.ย I don't adjust the sub to be louder. It doesn't matter that it isn't the same room SPL, what matters is that the slider is in the green at the same level as all other speakers in the green as calculated by Dirac for Dirac's purposes. I don't agree that we both have odd issues. The measurement tests I've done, and will continue to do, are simply because I have a desire to know how things work, and, if I change one thing - how does that affect things downstream? I get good results using Dirac with its defaults, in this regard I'm blissfully ignorant. I don't know. I have a ticket open with Dirac. I'll see what they say tomorrow.
|
|
|
Post by idea53 on Sept 14, 2020 6:33:13 GMT -5
You know...i did see they were higher at one point, but I thought you were still going back and forth leveling them all. Once you started doing this, I skipped about 10 seconds of the video and just assumed you were making them all level. Thanks for pointing that out. I do believe I'm even more confused now. Just for clarity on what you are doing, are you saying that, when measuring, your subs are potentially 10 db louder, then Dirac is potentially doing the same thing it is doing to mine but yours ends up being equal because you had the built in 10 db extra? I hope that makes sense. If I max my amp gain, mine measures 10 db higher on the volume screen, but basically equal on the measurements screen. Here are a couple screen grabs of the subs SPL and speaker SPL during the volume calibration. This is where you can see that the subs are ACTUALLY 10dB higher than speakers. Notice in the first photo, which is after all the other speakers have been leveled, that all the sliders are showing the same (within about 1dB) level "according to Dirac". Again, I'm using the Dirac sliders to adjust everything the same which is about -22dB on every slider including the subs. SPEAKERS: 67dB; SUBS: 77dB. During the Measurement sweeps, the speakers and subs play about the same SPL. SPEAKERS: 84dB; SUBS: 84dB. edit: I let Dirac do what it wants without any intervention on my part. I notice that you measure dB(C) and I guess that is the answer to your question. dB is not a unit it is just a relation between two numbers and can be used for any relation imaginable. In acoustics we talk about dB SPL (Sound Pressure Level) were the number is related to the hearing threshold of a "standard" person at 1 kHz. To this we also add that the hearing threshold is different at different frequencies and also the sensitivity for equal loudness of sounds. Theerfore there exist different frequency weighting functions to be able to measure a single number dB SPL and compare that to how loud we humans percieve the sound. A-weighting at low levels, B for mid level and C for high levels. But it has turned out to be common practice to use the A-curve for most measurements since that also correlates rather good with how our ears get damaged by sound. I suspect that Dirac has set up the process using A-weighted sounds and when you compare a low frequency signal measured with A and C-weighting the numbers can easily differ by 10 dB.
|
|
|
Post by marcl on Sept 14, 2020 8:59:35 GMT -5
Here are a couple screen grabs of the subs SPL and speaker SPL during the volume calibration. This is where you can see that the subs are ACTUALLY 10dB higher than speakers. Notice in the first photo, which is after all the other speakers have been leveled, that all the sliders are showing the same (within about 1dB) level "according to Dirac". Again, I'm using the Dirac sliders to adjust everything the same which is about -22dB on every slider including the subs. SPEAKERS: 67dB; SUBS: 77dB. During the Measurement sweeps, the speakers and subs play about the same SPL. SPEAKERS: 84dB; SUBS: 84dB. edit: I let Dirac do what it wants without any intervention on my part. I notice that you measure dB(C) and I guess that is the answer to your question. dB is not a unit it is just a relation between two numbers and can be used for any relation imaginable. In acoustics we talk about dB SPL (Sound Pressure Level) were the number is related to the hearing threshold of a "standard" person at 1 kHz. To this we also add that the hearing threshold is different at different frequencies and also the sensitivity for equal loudness of sounds. Theerfore there exist different frequency weighting functions to be able to measure a single number dB SPL and compare that to how loud we humans percieve the sound. A-weighting at low levels, B for mid level and C for high levels. But it has turned out to be common practice to use the A-curve for most measurements since that also correlates rather good with how our ears get damaged by sound. I suspect that Dirac has set up the process using A-weighted sounds and when you compare a low frequency signal measured with A and C-weighting the numbers can easily differ by 10 dB. Some added info .... can't measure subs with db A, right? sites.google.com/site/cliftonwindeducation/home/understanding-sound/understanding-sound-terms/dba-vs-dbc"Notice that "C" weighting is appropriate when measuring very loud sounds. A Rock Concert is an example of when it should be used."
|
|
|
Post by megash0n on Sept 14, 2020 9:19:08 GMT -5
For those who are interested, I reached out to Dirac to get a better understanding of what is needed for DLBC. I'm sure this is high-level, but good information nonetheless.
Flavio Fellah commented:
Yes, it's not a trivial task as one would expect... the internal bass management on the RMC-1 has to be disabled when the DLBC filter is active on the unit. The RMC-1 needs to include an individual processing chain for the DLBC which includes the Linkwitz-Riley filters, allpass filters, gains, and delays which is only active when the Dirac Live Bass Control filter is active. For this chain to work properly individual channels for the subs are required, the internal Bass Management on the RMC-1 must be disabled and Dirac takes control of the crossover. I imagine that it will take time but that's a question for Emotiva to answer.
Best regards
|
|
|
Post by marcl on Sept 14, 2020 9:54:44 GMT -5
OPERATIONAL DEFINITIONS, MEASUREMENTS AND OTHER TORMENTS I have an XMC-2; and, have had Emotiva Pre/Pros 'since the beginning.' There were certainly pre/pro issues at the very beginning. I've been following this thread religiously since its inception. One of the great problems in technical discussions is transparency in the matter of 'operational definitions', i.e. exactly what are you doing to obtain the results you see? Have you heard the good joke "Thinking is like hard work, only not as satisfying"? Maintaining the rigor of an operational definition of a test methodology is hard. Sometimes you actually have to write down or otherwise archive data for later inspection. More hard work. Coming from a university research psychology background, I began my 'semi-anechoic' acoustic measurements in 1993 or so using Bill Walso's "IMP/MLS" computer-based system. Then through Walso's "LAUD" and "Praxis" measurement systems. Now of course "REW" blows most everything else away - and 'free' makes it so much more special (I've donated). Along the way I created "The Bohlender-Graebener RD-75 Dipole Baffle Study" which was later part of my now-defunct "Acoustic Line Source Research" website. The Dipole Baffle Study measurements were done in a huge gymnasium and a large auditorium (for suspended-in-air anechoic measurements). I fully concur with the notion that it is optimal to start evaluation of an 'in-room' frequency response with an anechoic (or semi-anechoic) frequency response - which is very difficult to obtain in real life. Putting my money where my mouth is so to speak, I am just finishing a back yard remodel 'Summer of Pandemic' combination taijiquan court and loudspeaker measurement area. The 'first reflection' (time of flight) is always the problem for gaining adequate resolution in an anechoic measurement. In the huge gymnasium sound reflections were quite evident - they just took longer to arrive, and hence allowed for greater frequency resolution of the FFTs deriving the frequency response. Ahem ... you could 'see' impulse reflections in the data from microphone stands. It takes a while to learn to interpret. Have you heard the reputed Oscar Wilde comment that 'It isn't so much what a talking dog says, what matters is that the talking dog says anything at all.' (There's also the Ben Johnson's take on female preachers.) What seems amazing is that Dirac (and the others) can make much sense of in-room 'chirp' measurements at all. Letting others know 'exactly' your operational definitions is necessary for good communication and reliable replication of methodology. The thread seems to be working its way through repeatability measures. It isn't widely discussed, but the 'faith-based' assumptions defined by philosophers of science go something like this: 1. It is assumed that there is order in the Universe. 2. It is assumed that humans can ascertain that order. 3. It is further assumed that humans can understand the discovered order in the Universe. [Have you paid any attention recently to either the micro or macro universe? Folks are having a real hard time making sense of, for instance, "dark matter" and "dark energy." Best to be humble in one's ignorance.] I have observed in this thread that there are significant attempts by others to ascertain the 'order' in the Dirac methodology. So let me cheer you 'explorers of the unknown Dirac World' onward. Since I am comfortable getting some other projects done before 'going all Dirac" I will benefit from your ongoing work. Bravo! Free is good as I said. So please write clearly to your 'peanut gallery'(me et. al.) audience. Be good audiophile scientists. This Dirac system (and others) didn't exist in 1993 when I started loudspeaker measuring. Things have come a long way since then. A very long way! Sadly the difficulty in keeping track of 'the variables' never goes away. Yours in science, John Old School SCIENCE! I enjoyed reading this, John. It made me recall the guy I bought my first pair of Magnepans from in '77 ... Dr Clay Barclay. The "affable neurosurgeon" as he was also known around Philadelphia. He had an engineer working for him designing a measurement device that he called the "BADAP". I forget exactly what the acronym stood for, but basically it was a box with a small B&W video screen and it pounded out an impulse into the room and reported on the results. I never had the opportunity to see it in action. Somehow I assumed audio measurements in my own room would be too costly and so never looked for a way to do it until I discovered REW in 2012 ... while waiting for my second pair of Magnepans to arrive. I just looked in the folder and I have saved 682 REW files since 2012. Each speaker movement, each added bass trap, each Dirac calibration ... alas, without as comprehensive documentation as I now wish I had. Regarding Dirac, yes I see lots of folks trying to speculate on how it works, many clearly never having read the available technical discussions from Dirac. Many assume it's just IIR filters like everyone else uses. If that were to be the case, and in order to drop the levels enough to correct to a flat target without any boost, it would have to lower my system's levels by nearly 25db. It clearly isn't doing that, and yet it is able to lift some of the nulls without distortion (which would be measurably evident in REW). Dirac says - albeit without spilling the proprietary secret sauce - that a combination of IIR and FIR filters are used and that all-pass filters are used, and indeed the Emotiva Dirac Settings backup file reveals many, many dozens of biquad coefficients and FIR taps. Makes no sense to second guess them on how this all works ... they have kept the secret for 20 years. I quote Floyd Toole a lot because he is certainly among the foremost authorities on speakers playing in rooms, and because he writes his books and lectures in terms that are accessible, AND because he has spoken specifically on the topic of the legendary Harman Curve ... which he is mostly responsible for describing. I'm interested in your thoughts on Toole's work and the Harman Curve. My interpretation from all I've read in his books and other sources is that the curves are not intended as EQ targets, but rather as being representative of the response in a room that correlates with speakers that have flat anechoic response and (equally important!) uniform off-axis response. The curves are not intended to be used the way some folks are using them with Dirac. Though of course ... you pay your money for Dirac. you can make whatever curve you want. That said, I think misinterpretation of the use of target curves can lead to some tail-chasing and blaming Dirac for unpleasant results. Looking forward to your ongoing observations ....
|
|
|
Post by idea53 on Sept 14, 2020 15:36:09 GMT -5
I notice that you measure dB(C) and I guess that is the answer to your question. dB is not a unit it is just a relation between two numbers and can be used for any relation imaginable. In acoustics we talk about dB SPL (Sound Pressure Level) were the number is related to the hearing threshold of a "standard" person at 1 kHz. To this we also add that the hearing threshold is different at different frequencies and also the sensitivity for equal loudness of sounds. Theerfore there exist different frequency weighting functions to be able to measure a single number dB SPL and compare that to how loud we humans percieve the sound. A-weighting at low levels, B for mid level and C for high levels. But it has turned out to be common practice to use the A-curve for most measurements since that also correlates rather good with how our ears get damaged by sound. I suspect that Dirac has set up the process using A-weighted sounds and when you compare a low frequency signal measured with A and C-weighting the numbers can easily differ by 10 dB. Some added info .... can't measure subs with db A, right? sites.google.com/site/cliftonwindeducation/home/understanding-sound/understanding-sound-terms/dba-vs-dbc"Notice that "C" weighting is appropriate when measuring very loud sounds. A Rock Concert is an example of when it should be used." A rock concert might actually be a bad example of when the C-filter should be used since the A-filter give a better representation of the level of risk of hearing loss. So which filter to use always depend on the purpose of the measurement not the actual level of the sound. The C-filter is mainly used instead of Linear (i.e. no filter at all) since it cuts the lowest octave of sound and you will get less problem with infra sound in your measurement. But A-filter give a better (but really not good) representation on how loud you percieve the sound to be. That is if you want to use a single number representation of your sound. I prefer to at least see a 1/3-octave spectrum to "understand the sound".
|
|
|
Post by marcl on Sept 14, 2020 15:44:53 GMT -5
A rock concert might actually be a bad example of when the C-filter should be used since the A-filter give a better representation of the level of risk of hearing loss. So which filter to use always depend on the purpose of the measurement not the actual level of the sound. The C-filter is mainly used instead of Linear (i.e. no filter at all) since it cuts the lowest octave of sound and you will get less problem with infra sound in your measurement. But A-filter give a better (but really not good) representation on how loud you percieve the sound to be. That is if you want to use a single number representation of your sound. I prefer to at least see a 1/3-octave spectrum to "understand the sound". Yes but ... in this discussion of the relative levels of subs vs other speakers, you really have to use C-weighted or the sub levels will be inaccurate compared to the others. BTW, I heard Parliament Funkadelic at Newport Jazz Festival a couple years ago and 40ft from the stage it was 110dbC!
|
|
|
Post by bluescale on Sept 14, 2020 22:11:53 GMT -5
I plan to observe that, but I'm pretty sure I already know that the subs play lower during the sweeps. I don't think that's a problem, though as the level during the sweeps doesn't matter. For example, if you run a measurement, and then run into clipping on the next measurement, you can go back to the level setting screen, drop the level of the clipping speaker, and then continue your measurements without having to remeasure your previous sweeps. I haven't thought through (or researched) the *why* of how that works, but it does. My concern is that it will attenuate all speakers down to the subwoofer because of this.. And then, another 10db because it supposedly does this for headroom. If I crank my sub amp gain all the way up, which makes the subs play with the rest, my overall volume is about 10 db higher afterwards. Meaning, I will watch a movie at -15 instead of -5. If it weren't for countless bugs for so long, I might not question it, but any time something doesn't make sense, I ask questions. โบ I was hoping to post this last night, but I've had a hell of a time getting REW installed and running on this newer laptop. For whatever reason, ASIO4All won't work with the Nvidia GTX 1070. I finally ended up using the Java drivers. Anyway, the end result is that regardless of the sub level during measurements Dirac will normalize the volume level of the subs relative to the other channels. That's what I expected, but you indicated that you're getting different results based on sub level. My full bandwidth sweep taken at -25dB on the XMC-2 resulted in an average spl of 94.7dB per REW. Averages are a bit deceptive, obviously, but it's a slightly more nuanced Harman curve that goes from 84.5dB at 20khz to 94.5dB at 16hz. The takeaway from all this is that I have plenty of headroom after running Dirac, even though the sub levels play softer during the calibration sweeps.
|
|
|
Post by megash0n on Sept 14, 2020 22:23:48 GMT -5
My concern is that it will attenuate all speakers down to the subwoofer because of this.. And then, another 10db because it supposedly does this for headroom. If I crank my sub amp gain all the way up, which makes the subs play with the rest, my overall volume is about 10 db higher afterwards. Meaning, I will watch a movie at -15 instead of -5. If it weren't for countless bugs for so long, I might not question it, but any time something doesn't make sense, I ask questions. โบ I was hoping to post this last night, but I've had a hell of a time getting REW installed and running on this newer laptop. For whatever reason, ASIO4All won't work with the Nvidia GTX 1070. I finally ended up using the Java drivers. Anyway, the end result is that regardless of the sub level during measurements Dirac will normalize the volume level of the subs relative to the other channels. That's what I expected, but you indicated that you're getting different results based on sub level. My full bandwidth sweep taken at -25dB on the XMC-2 resulted in an average spl of 94.7dB per REW. Averages are a bit deceptive, obviously, but it's a slightly more nuanced Harman curve that goes from 84.5dB at 20khz to 94.5dB at 16hz. The takeaway from all this is that I have plenty of headroom after running Dirac, even though the sub levels play softer during the calibration sweeps. But, all channels are attenuated to the sub level is what I'm saying. Whether you have volume knob left or not isn't the issue. I don't understand why my sub is level on the volume calibration screen, but then plays 10db less when measuring. Something seems off. If I purposely turn my sub amp up 10 db prior to measuring, I dont lose this additional 10 db across the board. Due to my issue, I losing roughly 20 db. I understand why we lose the first 10. Dirac is taking 10 away so they arent actually boosting a frequency.
|
|
|
Post by bluescale on Sept 14, 2020 23:05:26 GMT -5
But, all channels are attenuated to the sub level is what I'm saying. Whether you have volume knob left or not isn't the issue. I don't understand why my sub is level on the volume calibration screen, but then plays 10db less when measuring. Something seems off. If I purposely turn my sub amp up 10 db prior to measuring, I dont lose this additional 10 db across the board. Due to my issue, I losing roughly 20 db. I understand why we lose the first 10. Dirac is taking 10 away so they arent actually boosting a frequency. What Iโm saying is that I donโt lose that extra 10 dB when I leave the sub level prior to measurement. Either youโre doing something differently than me, and Iโm not understanding, or we are seeing different results. As far as I can tell, the only difference between my our testing is that I donโt use my sub amps to adjust the levels, since I have them matched using the miniDSP. For one measurement I adjusted the level down to match the mains in the Dirac volume matching screen. In another I bumped the level down in the miniDSP. In a third run I didnโt adjust the volume at all and left it about 15 dB hotter than the main speakers. It didnโt matter. Dirac put forward the same results every time.
|
|
|
Post by ttocs on Sept 14, 2020 23:21:34 GMT -5
]I don't understand why my sub is level on the volume calibration screen, but then plays 10db less when measuring. Something seems off. If I purposely turn my sub amp up 10 db prior to measuring, I dont lose this additional 10 db across the board. Due to my issue, I losing roughly 20 db. I understand why we lose the first 10. Dirac is taking 10 away so they arent actually boosting a frequency. Are you only using the sub slider as the gauge for making the sub volume adjustment? Or are you using an SPL meter? If you are using the meter, then ditch the meter and ONLY use the Dirac slider gauge to get the sub adjustment into the green so it shows EXACTLY the same level as all the other speakers on their slider gauges. When I set my sub I only used the Dirac screen to make the setting be the same -22dB level as the other speakers. Trust the Force Luke.
|
|
|
Post by bluescale on Sept 14, 2020 23:40:41 GMT -5
]I don't understand why my sub is level on the volume calibration screen, but then plays 10db less when measuring. Something seems off. If I purposely turn my sub amp up 10 db prior to measuring, I dont lose this additional 10 db across the board. Due to my issue, I losing roughly 20 db. I understand why we lose the first 10. Dirac is taking 10 away so they arent actually boosting a frequency. Are you only using the sub slider as the gauge for making the sub volume adjustment? Or are you using an SPL meter? If you are using the meter, then ditch the meter and ONLY use the Dirac slider gauge to get the sub adjustment into the green so it shows EXACTLY the same level as all the other speakers on their slider gauges. When I set my sub I only used the Dirac screen to make the setting be the same -22dB level as the other speakers. Trust the Force Luke. Youโre misunderstanding his point. If you level match the sub using the slider, itโll match on the level calibration screen, but the sweeps that play have the subs play 10dB louder than the other speakers. I experience that also, but I donโt believe this is a problem. Heโs seeing results that make him question that.
|
|
|
Post by ttocs on Sept 15, 2020 1:16:57 GMT -5
Are you only using the sub slider as the gauge for making the sub volume adjustment? Or are you using an SPL meter? If you are using the meter, then ditch the meter and ONLY use the Dirac slider gauge to get the sub adjustment into the green so it shows EXACTLY the same level as all the other speakers on their slider gauges. When I set my sub I only used the Dirac screen to make the setting be the same -22dB level as the other speakers. Trust the Force Luke. Youโre misunderstanding his point. If you level match the sub using the slider, itโll match on the level calibration screen, but the sweeps that play have the subs play 10dB louder than the other speakers. I experience that also, but I donโt believe this is a problem. Heโs seeing results that make him question that. Yes, the Volume Calibration adjustment sweeps for the subs play at a different room SPL as measured with a handheld meter, which is the way it should be and doesn't matter. The important part is to set all speakers and subs to register the same on the sliders. The measurement sweeps will then all play within a 2 or 3 dB of each other. I'm only trying to verify how he is going about making the adjustments. When he adjusts the subs, is he looking at the green gauge on the sliders and matching the subs to all other speakers? Or doing something else?
|
|
|
Post by bluescale on Sept 15, 2020 2:52:45 GMT -5
Youโre misunderstanding his point. If you level match the sub using the slider, itโll match on the level calibration screen, but the sweeps that play have the subs play 10dB louder than the other speakers. I experience that also, but I donโt believe this is a problem. Heโs seeing results that make him question that. Yes, the Volume Calibration adjustment sweeps for the subs play at a different room SPL as measured with a handheld meter, which is the way it should be and doesn't matter. The important part is to set all speakers and subs to register the same on the sliders. The measurement sweeps will then all play within a 2 or 3 dB of each other. I'm only trying to verify how he is going about making the adjustments. When he adjusts the subs, is he looking at the green gauge on the sliders and matching the subs to all other speakers? Or doing something else? In my testing the subwoofer level didnโt seem to make much difference. I tried matching it and playing it at a higher level. The end result was the same.
|
|
|
Post by hsamwel on Sept 15, 2020 5:20:51 GMT -5
This is basically what I do following the other sticky guide. One thing that is interesting is that it discusses how it will or will not add an additional 10db for subs depending on your setup. This is not happening for me. Regardless, my point is that, when you go to the measurements section, all speakers, besides the center sub, play at least 10 db louder than they were set on the volume calibration screen. ttocs has confirmed this as well. I've opened a ticket with Dirac to understand if this is a bug or on purpose because it doesn't make sense. Now, I feel there is something to setting the mic properly instead of just going with 100 . I think the sweet spot for me is around 95. I thought it was better to lower the mic sensitivity and instead up the volume to get better readings. I think even Dirac mentions this on their help. I have run Dirac cals on 100%, 86% and 81%. Ended up using 86, but 81 worked fine as well.
|
|