Thank you so much for your understanding and continued support! I truly appreciate all your comments and questions. Please know that while RUclips isn't my main job, I always strive to engage with you as much as I can. If you'd like me to prioritize your questions, feel free to use RUclips $super thanks. I'll make sure to address those questions in my upcoming videos as a priority. Thanks again for being a part of my community!
Quite possibly the best summary of factors we need to consider in our homes. You prove that technical discussions brought down to a basic level can be educational for home theater enthusiasts. Well done! Keep it coming. I see RP22 is a free download. That is great news too.
10:02 “You gotta listen loud”… 😎👍 I’m always trying to explain that. Sure my neighbours wouldn’t agree sometimes 😂 (I do my best to be reasonable and have isolated the walls as best I can)
Hi Matthew, great channel thank you! Question: In your cinema how have you maintained the room within a room sound isolation but with in-wall speakers? Do these not compromise the wall? Have you had to make a third room in a room?
I'm new to your Channel, you look like you have so much information. And I also see you have back "Surround speakers" which no one really talks about but is very important to get the complete surround experience. 👍🏽✌🏽
Thanks. I like to think I know a thing or two. I professionally design home theaters for a living. I am a CEDIA instructor. I am well respected for my knowledge. I certainly don’t know everything (or even close). I enjoy sharing what I know to help the passionate enthusiasts in our industry. As for rear surrounds. My view is that a good 5.1 with side surrounds is great. But I like a 7.1 or 9.1 better. In a dedicated room like mine, no reason not to do it.
Great video! Had to rewatch a few times...so much info. Thank you! Can you do a video on ETC, Impulse Response interpretation? I believe many of us aren't sure what to address and what to ignore.
when i got my dsp into my system, i really could see how high pass my main speakers affected how loud they could play(Arendal 1961 monitors). as Poe said, 80hz highpass on those i reach the max excursion, so i set it at 180hz 12db slope. dual 5" woofers on those speakers in sealed boxes. if anyone wonder ported speakers wont help, those usually have port tuning below 60hz.
Yeah the port tuning would need to be closer to the crossover point. It’s also the fact that your receiver or amp is free to power a more limited bandwidth. This too is helpful. Many receivers would struggle to provide enough power.
When doing in-room frequency response measurements, should you do one speaker at a time plus subwoofers or front left AND right speakers simultaneously plus subwoofers?
First of all, thanks, Matt, for providing us with informative guidance! There's never one solution to a perceived acoustic issue; there are always a myriad of factors in play, and it can become increasingly overwhelming to grasp. I'm really keen on RP32 - is there a target release date? My question is: With all loudspeakers targeted for 105dB SPL, realistically, with immersive material, that target SPL for overhead speakers can become costly. In your opinion, what is the 'lowest max SPL' overheads should aim to achieve? My complication is that, I can see my 4 overhead speakers clipping at reference levels, and considering either an upgrade, or adding an additional 2 (for 6 overheads). I'm trying to understand if the extra 2 overheads will contribute to the overall SPL output of my overheads. It might sound backward in my thinking, but because the overheads play object-orientated (therefore not a discrete channel), it seems I wouldn't need 105dB SPL from each overhead...what are your acoustic acrobatic thoughts? lol
For some reason I thought I covered this. But tops actually dont absolutely need to hit that. Because of how music can and is being recorded/mixed, I think it’s still useful. But they are targeting I believe 102dB rather than 105. Also remember you sit much closer to them.
As for listening to the reference, I think that the people who can't stand this standard are people who have speakers that distort too much at these high volumes. When I got speakers capable of doing more than the reference without audible distortion, everything took sense. I'd never been able to listen up to the reference before because my speakers had too much distortion and therefore hurt my ears, but now that's no longer the case
Very nice video! When we are discussing these target SPLs do we mean the levels that are reached with a full range sweep (how long?) or burst test? 85dB with “peaks” of 105dB sounds like short bursts which may avoid compression seen in longer testing. Also there’s a Dolby Atmos Specifications doc for large theaters that specifies SPL targets of 105 for low freq (excluding LFE), 101 for mids, 98 for highs, and 92db for ultra highs which kind of matches your suggestion that 105 doesn’t mean 105 at all frequencies.
I am actually talking with the supervising engineer at Skywalker Sound about this. So I think it’s all a bit complicated. The actual spec comes from SMPTE and refers to the idea that pink noise should achieve 85dB at -20dBU. So you take pink noise and play it back at -20dBU. Levels should be 85dB. As a results, voices tend to then be around 80-90 dB in the movie. That means we have 20dB of dynamic range. But in practice it won’t mean the movie is going to sustain output at 105dB during loud scenes. During loud action scenes, peak SPL will actually far exceed 105dB for short times. But that isn’t really that meaningful. As for how this applies to speaker choices. We don’t know. That hasn’t been thought through as well as it needs to be. A speaker that can do 120 dB with a burst test might max out at 90dB with a sustained long term wide bandwidth noise signal. I am still exploring this topic and will come back with a more succinct answer. The individual I contacted feels that certain types of max SPL tests which achieve at least 102dB (at the seated distance)would be sufficient. I am still waiting for an answer on what types of stimulus signals these would be. I suspect it’s going to be shaped pink noise and M noise.
Very good video, looking forward for more information on this topic in the future. i have a question about reference Level, after calibration, the AVR at 0dB is supposed to be Reference level. but how do you know all sources are playing at reference level? for example streaming services like Netflix, hulu, disney. and could the streaming device also affect this? so how do you know you are watching a movie at reference level?
There is no reason it shouldn’t be. They would need to be playing with the digital signal level for that to be true. You could try testing it I suppose. But no way to send a test signal to Netflix. You could send test signals to airplay or load onto RUclips.
Don't you think using Var Smoothing is better because then you can see all the dips and peaks in the low end? Because using 1/6 smoothing you could have a giant null due to modes or SBIR and you wouldn't know because it wouldn't show it. You might not hear a huge narrow dip at 50hz but I think once you fix it you'll notice you suddenly have more punch in the bass.
All of these approaches could be helpful. It just depends on what you are doing. Variable smoothing was designed by John to use for equing. It’s not a psychoacoustic smoothing that matches what your ear detects. That’s actually what psychoacoustic and ERB smoothing was developed for. Adam Pelz likes to use 1/6 as a way to get an idea of what is going on. Those high Q dips are just not that important in evaluating it.
Hi can you tell me what app do u use for spl ? I always thought 105db is per speaker not all of them. The ref is 105db is that peak or continous? Because if you play 105db continous the peak can reach 115db for main. Thanks
I don’t use an app to measure that stuff. But I do use NIOSH SLM on my phone for quick and dirty SPL checks. It’s the best and most actuate SPL app. I never said 105dB was for all speakers. It’s per speaker. Like you said. It’s a peak value but as I have noted elsewhere, calling it transient peak is a bit misleading. The 105 is a theoretical max because of 20dB of headroom on a -20dBu signal giving 85dB. As my friend at Skywalker sound pointed out, that pink noise as prescribed for the test can only achieve -12dBu max and so wouldn’t achieve 105dB. The hard issue is equating this to speaker performance. How loud does a speaker need to play with what type of content to achieve the max SPL needed for accurate movie reproduction? It’s a question nobody is answering correctly at the moment. We need to think this through, which we are doing with CEDIA.
I have found with my system it doesn’t start getting fun and dynamic till I’m within 5db from reference level 0 being reference and starting at -5db any lower than -5db the dynamics are gone
I suspect it’s not that useful. I would think the receivers power supply would be the limiting factor in most cases. But if you have extra channels. Give it a try.
I am probably not listening correctly. Did he just say to use an SPL meter to set each speaker to 108db instead of the usual standard 75db? This would be based off setting the volume to 0db on the receiver when adjusting db levels.
I think you misheard. Reference level is 85dB. So you set it to 85dB with a -20dBU signal and the receiver set to whatever its reference level is (like 0 on many). That then means that a 0dBU signal would, in theory, give you 105dB output. In practice it’s not that simple. But that’s what I was referring to. 108dB is for headroom and isn’t about setting levels. You set the levels to achieve 85dB at a given reference volume level. But many receivers or processors can go positive from 0 allowing higher SPL than reference. And if your speakers have 3-4 dB of headroom, you could then achieve that in transient peaks. 108dB is 3dB of headroom above reference.
@@Ab-hc3xm it was actually a really good catch that deserved clarity. I need to maybe do a video on this topic. I just talked to the head engineer at Skywalker sound about this and he pointed out that in practice these signals would never achieve 105dB. And transients could actually be quite a bit higher than this. But actual max SPL would be a lot lower.
I myself listen close to reference levels when watching movies alone. The problem is when the wifey watches movies with me she doesn't like it coz she finds the sudden explosions or gunfire too loud and scares her 😆
This isn’t uncommon. If you have any dynamic range compression, you might try using it. When you turn it down, it doesn’t change the dynamic range, but it does mean that the voices are too quiet and the more quiet noises are buried in the noise floor. Dynamic range compression is needed to fix this. It’s a dirty word amongst most enthusiasts but shouldn’t be. After all, you aren’t listening as the artist intended when you turn it down. It’s already wrong. So compressing it is just compensating for that.
Thank you so much for your understanding and continued support! I truly appreciate all your comments and questions. Please know that while RUclips isn't my main job, I always strive to engage with you as much as I can. If you'd like me to prioritize your questions, feel free to use RUclips $super thanks. I'll make sure to address those questions in my upcoming videos as a priority. Thanks again for being a part of my community!
Quite possibly the best summary of factors we need to consider in our homes. You prove that technical discussions brought down to a basic level can be educational for home theater enthusiasts. Well done! Keep it coming. I see RP22 is a free download. That is great news too.
Thank you! It is free. That is by design. This is not best practices for the special few. We want everyone to have this information.
Very helpful. Thank you for your time!
You covered it and them some! THANK YOU 🍻
10:02 “You gotta listen loud”… 😎👍
I’m always trying to explain that. Sure my neighbours wouldn’t agree sometimes 😂
(I do my best to be reasonable and have isolated the walls as best I can)
Hi Matthew, great channel thank you! Question: In your cinema how have you maintained the room within a room sound isolation but with in-wall speakers? Do these not compromise the wall? Have you had to make a third room in a room?
Thank you for your question. I’ll record a video to answer your question in full when I get back from CEDIA training in Chicago
Thanks, Matt. Excellent information. I've learned so much from your videos and they are greatly appreciated! Tom in Colorado.
That was BANGING!
I'm new to your Channel, you look like you have so much information. And I also see you have back "Surround speakers" which no one really talks about but is very important to get the complete surround experience. 👍🏽✌🏽
Thanks. I like to think I know a thing or two. I professionally design home theaters for a living. I am a CEDIA instructor. I am well respected for my knowledge. I certainly don’t know everything (or even close). I enjoy sharing what I know to help the passionate enthusiasts in our industry.
As for rear surrounds. My view is that a good 5.1 with side surrounds is great. But I like a 7.1 or 9.1 better. In a dedicated room like mine, no reason not to do it.
Well done 👏🏼 👍🏼 thank you for the info and demo 👍🏼✌🏼🏴🇬🇧
Glad you enjoyed it
Great video! Had to rewatch a few times...so much info. Thank you!
Can you do a video on ETC, Impulse Response interpretation? I believe many of us aren't sure what to address and what to ignore.
Sure. Will be doing these more in the future.
when i got my dsp into my system, i really could see how high pass my main speakers affected how loud they could play(Arendal 1961 monitors).
as Poe said, 80hz highpass on those i reach the max excursion, so i set it at 180hz 12db slope. dual 5" woofers on those speakers in sealed boxes.
if anyone wonder ported speakers wont help, those usually have port tuning below 60hz.
Yeah the port tuning would need to be closer to the crossover point.
It’s also the fact that your receiver or amp is free to power a more limited bandwidth. This too is helpful. Many receivers would struggle to provide enough power.
When doing in-room frequency response measurements, should you do one speaker at a time plus subwoofers or front left AND right speakers simultaneously plus subwoofers?
First of all, thanks, Matt, for providing us with informative guidance! There's never one solution to a perceived acoustic issue; there are always a myriad of factors in play, and it can become increasingly overwhelming to grasp. I'm really keen on RP32 - is there a target release date?
My question is: With all loudspeakers targeted for 105dB SPL, realistically, with immersive material, that target SPL for overhead speakers can become costly. In your opinion, what is the 'lowest max SPL' overheads should aim to achieve?
My complication is that, I can see my 4 overhead speakers clipping at reference levels, and considering either an upgrade, or adding an additional 2 (for 6 overheads). I'm trying to understand if the extra 2 overheads will contribute to the overall SPL output of my overheads. It might sound backward in my thinking, but because the overheads play object-orientated (therefore not a discrete channel), it seems I wouldn't need 105dB SPL from each overhead...what are your acoustic acrobatic thoughts? lol
For some reason I thought I covered this. But tops actually dont absolutely need to hit that. Because of how music can and is being recorded/mixed, I think it’s still useful. But they are targeting I believe 102dB rather than 105. Also remember you sit much closer to them.
As for listening to the reference, I think that the people who can't stand this standard are people who have speakers that distort too much at these high volumes. When I got speakers capable of doing more than the reference without audible distortion, everything took sense. I'd never been able to listen up to the reference before because my speakers had too much distortion and therefore hurt my ears, but now that's no longer the case
Very nice video! When we are discussing these target SPLs do we mean the levels that are reached with a full range sweep (how long?) or burst test? 85dB with “peaks” of 105dB sounds like short bursts which may avoid compression seen in longer testing. Also there’s a Dolby Atmos Specifications doc for large theaters that specifies SPL targets of 105 for low freq (excluding LFE), 101 for mids, 98 for highs, and 92db for ultra highs which kind of matches your suggestion that 105 doesn’t mean 105 at all frequencies.
I am actually talking with the supervising engineer at Skywalker Sound about this.
So I think it’s all a bit complicated. The actual spec comes from SMPTE and refers to the idea that pink noise should achieve 85dB at -20dBU. So you take pink noise and play it back at -20dBU. Levels should be 85dB. As a results, voices tend to then be around 80-90 dB in the movie. That means we have 20dB of dynamic range. But in practice it won’t mean the movie is going to sustain output at 105dB during loud scenes.
During loud action scenes, peak SPL will actually far exceed 105dB for short times. But that isn’t really that meaningful.
As for how this applies to speaker choices. We don’t know. That hasn’t been thought through as well as it needs to be. A speaker that can do 120 dB with a burst test might max out at 90dB with a sustained long term wide bandwidth noise signal.
I am still exploring this topic and will come back with a more succinct answer. The individual I contacted feels that certain types of max SPL tests which achieve at least 102dB (at the seated distance)would be sufficient. I am still waiting for an answer on what types of stimulus signals these would be. I suspect it’s going to be shaped pink noise and M noise.
@@PoesAcoustics Thanks for trying to get us closer to a real answer on this topic, it's interesting stuff for sure.
Very good video, looking forward for more information on this topic in the future.
i have a question about reference Level, after calibration, the AVR at 0dB is supposed to be Reference level. but how do you know all sources are playing at reference level?
for example streaming services like Netflix, hulu, disney. and could the streaming device also affect this?
so how do you know you are watching a movie at reference level?
There is no reason it shouldn’t be. They would need to be playing with the digital signal level for that to be true. You could try testing it I suppose. But no way to send a test signal to Netflix. You could send test signals to airplay or load onto RUclips.
@@PoesAcoustics I find myself adjusting the volume by 10-15dB depending on which streaming service/disc I'm watching.
Don't you think using Var Smoothing is better because then you can see all the dips and peaks in the low end? Because using 1/6 smoothing you could have a giant null due to modes or SBIR and you wouldn't know because it wouldn't show it. You might not hear a huge narrow dip at 50hz but I think once you fix it you'll notice you suddenly have more punch in the bass.
All of these approaches could be helpful. It just depends on what you are doing. Variable smoothing was designed by John to use for equing. It’s not a psychoacoustic smoothing that matches what your ear detects. That’s actually what psychoacoustic and ERB smoothing was developed for.
Adam Pelz likes to use 1/6 as a way to get an idea of what is going on. Those high Q dips are just not that important in evaluating it.
Great info Poe!
Thanks!
what are some decent speaker options for 1000 or less each? I doubt I need to much output at the lower volumes I use for movies.
Hi can you tell me what app do u use for spl ? I always thought 105db is per speaker not all of them. The ref is 105db is that peak or continous? Because if you play 105db continous the peak can reach 115db for main. Thanks
I don’t use an app to measure that stuff. But I do use NIOSH SLM on my phone for quick and dirty SPL checks. It’s the best and most actuate SPL app.
I never said 105dB was for all speakers. It’s per speaker. Like you said. It’s a peak value but as I have noted elsewhere, calling it transient peak is a bit misleading. The 105 is a theoretical max because of 20dB of headroom on a -20dBu signal giving 85dB. As my friend at Skywalker sound pointed out, that pink noise as prescribed for the test can only achieve -12dBu max and so wouldn’t achieve 105dB.
The hard issue is equating this to speaker performance. How loud does a speaker need to play with what type of content to achieve the max SPL needed for accurate movie reproduction? It’s a question nobody is answering correctly at the moment. We need to think this through, which we are doing with CEDIA.
I have found with my system it doesn’t start getting fun and dynamic till I’m within 5db from reference level 0 being reference and starting at -5db any lower than -5db the dynamics are gone
Yes this is not unique to your system. It’s just how it is. It’s why we need to listen louder to really appreciate what we have.
@@PoesAcoustics yes agreed, louder and cleaner
What’s your thoughts on biamping off an avr?
I suspect it’s not that useful. I would think the receivers power supply would be the limiting factor in most cases. But if you have extra channels. Give it a try.
I am probably not listening correctly. Did he just say to use an SPL meter to set each speaker to 108db instead of the usual standard 75db? This would be based off setting the volume to 0db on the receiver when adjusting db levels.
I think you misheard.
Reference level is 85dB. So you set it to 85dB with a -20dBU signal and the receiver set to whatever its reference level is (like 0 on many). That then means that a 0dBU signal would, in theory, give you 105dB output. In practice it’s not that simple. But that’s what I was referring to.
108dB is for headroom and isn’t about setting levels. You set the levels to achieve 85dB at a given reference volume level. But many receivers or processors can go positive from 0 allowing higher SPL than reference. And if your speakers have 3-4 dB of headroom, you could then achieve that in transient peaks. 108dB is 3dB of headroom above reference.
Ok, thank you for the clarification. I'm a novice at best when it comes to these things 😅
@@Ab-hc3xm it was actually a really good catch that deserved clarity. I need to maybe do a video on this topic.
I just talked to the head engineer at Skywalker sound about this and he pointed out that in practice these signals would never achieve 105dB. And transients could actually be quite a bit higher than this. But actual max SPL would be a lot lower.
ty!
22Hz in the intro measured with Audio Tools for iPhone and the iTestMic2 🙂
I myself listen close to reference levels when watching movies alone. The problem is when the wifey watches movies with me she doesn't like it coz she finds the sudden explosions or gunfire too loud and scares her 😆
I had the volume issue with my SO to. I got here Loop Earplugs that just reduce the sound by 15db and she has no issues now.
@@robotag95Good idea.
This isn’t uncommon. If you have any dynamic range compression, you might try using it. When you turn it down, it doesn’t change the dynamic range, but it does mean that the voices are too quiet and the more quiet noises are buried in the noise floor. Dynamic range compression is needed to fix this. It’s a dirty word amongst most enthusiasts but shouldn’t be. After all, you aren’t listening as the artist intended when you turn it down. It’s already wrong. So compressing it is just compensating for that.