This AI stuff is really obnoxiously cheesy and hyperbolic. And like most of my experiences with "AI," it quite often gets a LOT wrong. The reason for this is the source material... AI doesn't "know" what is correct, it only knows what is most commonly repeated, and if that information is wrong then its output is wrong. It also doesnt actually "know" anything about whatever subject its working on, it only has a vast library of source material to mashup its responses from. Like all things computerized, its the old adage, "garbage in, garbage out"
"AI doesn't "know" what is correct, it only knows what is most commonly repeated." - like 99.9% of humans, actually. We mostly just repeat what we hear, especially if it's what we *want* to hear.
The AI might be technologically impressive, but I don't think it helps your video in any way. I usually come to your videos so that you, the acoustics expert, can break down concepts for me, not for an AI to try rambling in podcast form to summarize a paper.
Yeah, same here I'm afraid And I find that style of podcasting unlistenable (with all the excessive back-and-forth agreement between the presenters) I had to skip over most of the AI babble
Prob is tho fellas in 5 years if your not using AI in your business, especially this particular one, which is the opinion of the current & former CEO of google btw AND the one in which they are both putting the most attention too, U will fail. Now if you guys just work for someone and don’t own a business and are not entrepreneurs, then who cares u like what u like and don’t wanna hear this from Jesco and that’s perfectly fine but the point is him trying it was/is a smart business move and shows just how in tune he is . Let me ask u both this question what should he have changed about his presentation that would have made it “Watchable”?? the fact u felt the need to just complain helps nothing and no one. Critiquing is one thing, but with no feed back you’re just criticizing and nobody wants to hear that BS. Respectfully of coarse …
I really liked the format, and hope you do something similar in future. As a layman, it’s difficult to get into the science and understand the papers. The AI did an incredible job of translating the jargon into something I could understand easily, with some surprisingly impressive analogies. The issue with AI (and with random forum threads online), is that the summaries can be wrong, misleading, or cherry-picked. Having you, an expert, essentially act like a referee, allows me to get the data simplified for me, whilst being confident in its accuracy. I hope you do this again with other topics, particularly ones that debunk commonly believed myths!
Great Video Jesco! I agree with you on all fronts here and super interesting to think more critically about ITD gap in smaller home studio rooms. Also, never had seen NotebookLM - super crazy and kinda scary thinking AI could create a full media content channel pretty soon.
Tbh I would have taken away a lot more If YOU had presented us the details of the LEDE paper. The AI was incoherent and frankly wasted everybodys time with lots of talking but very little content.
About 1 minute before the end the author compares line-of-sight signal alone w/out reflections vs. line-of-sight + spectrally correct, properly delayed, moderately high-level reflections. The author smartly posits whether the latter has any advantage vs. the former in a control room setting. Psycho-acoustics expert/recording engineer Dr. David Griesinger has scientifically compared the two and found that the latter increases listening pleasure with no downside, but I don't know his opinion for work purposes in a control room. DG states that the latter (above) maximizes detail, immersion and envelopment. Dr. Floyd Toole states that the latter increases pitch sensitivity. I understand how and why line-of-sight signal alone may be ideal in a control room. OTOH...for a few months I have used a unique modular loudspeaker with a clean, undisturbed line-of-sight signal + late-onset reflection assist (spectrally correct, properly delayed, variable level.) IMO even someone listening to music for work in a control room would only benefit from using such a loudspeaker and I know of no downside. Imaging, spatial qualities and lack of proximity effect are all improved. Re. ideal bass performance, Dr. Earl Geddes, Todd Welti of Harman Labs and numerous others including Dr. Robert E. Greene have long promoted the distributed bass array as the ideal solution to eliminate bass modes in three-dimensional space, before they propagate (smooth bass throughout the room except for light build-up in the corners.) Solving bass modes is otherwise virtually impossible because of the mathematical relationships between the room's boundaries and long bass wavelengths. Even if one lowers the peaks, filling in a 9+ dB dip is impossible with gain; EQ tends to only smooth a peak at one location of the room while making response worse at another location. Siting 4 subs @ strategic locations throughout the room adds 4 new randomized bass modes to the room's fixed bass modes, thus minimizing the effects of the latter (4 fixed modes vs. 4 fixed + 4 randomized modes.) Griesinger promoted a method to further smooth the distributed array's modes called Phase Quadrature, identifying 2 subs each on the L and R sides. The ideal phase for the entire sub array is 90 degrees to maximize smoothness in the pass band (shared range of the mains and subs; less then 90 = lumpier bass, more than 90 risks power loss at the lowest F.) Griesinger suggests setting the phase of 2 subs on one side of the room @ 45-degrees and the remaining 2 subs on the opposite side @ 135-degrees. The average is still 90-degrees, but the asymmetrical phase also maximizes smoothness below the pass band. (For the record, the Double Bass Array may work even better than a properly tuned Distributed Bass Array, but the former is about 4x costlier than the latter and requires building 4 subs each into the front and rear walls, 8 subs total. The 4 front subs are in-phase with the music program; a Double Bass Array has the 4 rear subs 180-degrees out of phase and delayed for a period equal to the room's length. Bass from the front array travels to the rear of the room where it disappears due to the rear array's inverted phase.) I spent about twice as much on acoustic treatments ($750 on professional plans alone, acoustic soffits and corner traps) vs. the cost of the distributed bass array and the latter's performance incinerated the former's. There's a reason a particular acoustics expert calls full range speakers, "a failed experiment."
I enjoy your content. Tricky subject with no firm answers, but I've followed a lot of your advice and, at the very least, you've helped me avoid wasting money or making things worse!
This was a fascinating use of AI. While it had its moments, I thought the format in which it presented its interpretation was engaging and personable. And the topic was, of course, always an interesting discussion. Kudos on a great experiment!
Please please don't do this 'AI podcast' format again 🙏🏻 I find that style of wafflecast insufferable. I can only listen to a podcast if the guest speaker talks concisely, and is allowed to speak without interruption. That's what is so good about your channel Jesco so please, don't put us through that again! (Interesting subject matter, ofc!)
For the next Q&A Knowledge Bomb: I’m building my own desk for my studio, and I had an interesting thought I’d like you to consider. The desk will be a standing desk, and it’s about 4 ft from the front wall. I’ll need a way to hide cables. Rather than just putting a piece of wood on the back, what if I made it out of acoustic insulation that was covered on both sides by fabric? Would this have any benefit at all?
What if the back wall can be taken out of the equation entirely-ish. I positioned my studio so the French doors are centered in the back wall, about 5 meters away. I have massive gobos covering the doors that can be moved, but I can also open the patio doors all the way, creating about 3.5 square meter opening. Does bass just flow right on out the hole int he wall? Does it make any difference? The only reason I have this option is because I live quite remote in the mountains so the only noise is wind and the occasional little beastie wandering about. I did this mostly for options during the recording process. Most significantly, would this change the acoustics of the room enough to confuse my learning of the room for mixing? Or maybe another arrow in the quiver for mixing listening options? Thank you for your time sir
You’re an FN genius for using notebook for this. Dude I’d love to collaborate with you on some interesting treatment ideas and designs I’ve come up with thru trial and error in my own space. It’d be too difficult to get my point across thru email so would you be interested in working out a call or zoom or something and just for a few minutes of your time? Look at it like this, if u don’t like my ideas, you can tell me to get lost idiot, an that’s always fun, or I’m on to something and we potently give acoustics a much needed update /upgrade then we go down in history for changing the game!! and for me ,well, I hit the drawing board BUT continue to watch your vids an learn no matter what you think…Frankly it’s a win win for both of us either way!!!😂😂😂 don’t mind my corny attempt at being funny, can’t help it when I get excited about an idea🤷♂️😂Thanks Jes my friend appreciate all you do
18:40 Because, like you have said, speakers don’t have enough mass to decouple properly. Would it help to put something heavy on them to weigh them down? Or would that create more resonances or other problems? Thanks for the video btw! Great content!
So is LEDE design opposite of FTB design? We would like to know much about FTB design. It will be great if you share your opinions about FTB design. Thank you.
Hello I have a great question for you and would love your input since you know treated rooms well and have experience mixing music . I was debating on treating my room or using vsx headphones . What is your input on the studio emulations in place as I'm sure you have probably tried these out at one point ? I'm listening through them right now and they do sound like a treated room where I can hear details very well . Looking forward to your input !
The idea of Initial Sound Delay Gap is about psychoacoustic grounding in the studio as well as perceiving the acoustics of the recording environment. Completely dead studios clearly do work but the environment is unnatural.
Hey Jesco Big Fan ; Quick question. If you had to chose one over the other, which one would you choose? Because of the room shape, I only have 2 options. 1. Equal lateral triangle at 55 inches on a 2.1 system. But my sub is a bit closer to a side wall than the 1/4 rule. Also one speaker is closer to a side wall than the other. But I haven’t an equilateral triangle at 55”. 2. My speakers are equal distance from back and side walls , and sub is in correct place. But from tweeter to tweeter is aprox 77.7” and from speaker to listening position is aprox 46” or so. So the equilateral triangle is not proper. I haven’t a ton of treatment , all inspired by you. I’ll send pics. In pic it’s set up as #2. Also I have the Neumann kh 80’s and kh750 with Dsp. I assume the triangle is most important as Dsp may fix the offset timing by the side walls. But I’d love to hear your opinion on this matter. Thank you so much. Bill Ibe
Interesting video. Re your comment of maximizing the gap, and supressing the room...don't you think that a totally dead room is unpleasant to work in? Isn't that why we include diffusion?
Oh they can help very much, especially the Trinnov. But not in isolation.... You have to remember that there are really several components to a correctly controlled room. The proper placement of speaker and listening position in the room. Acoustic treatment, and then room correction. Trinnov is not a magic box that can fix everything all by itself, it fist needs the other elements to be done right, but once they have been done, Trinnov can take care of a lot of the rest, especially for things like phase correction and soundstage.
Was really looking forward to this video but I’m not in to listening to a computer pretending to be human in a podcast talking about a complex, technical and somewhat subjective subject that I’m also very passionate about. I’ll skip the video and read the papers instead.
Don't the vast majority of consumer listening environments range from less than ideal to hot garbage? Does getting the "optimal" mixing environment really matter if the consumer isn't going to be experiencing the finished result as intended? Isn't a "perfect" listening environment only slightly better than a "good" one? I understand the idea that home studios aren't ideal for mixing because of early reflections, standing modes etc, and that real issues can arise from mixing in an environment like that, but isn't "pretty good" going to result in a mix that sounds good in most environments? My favorite music isn't music that is perfectly balanced and nicely packaged for every conceivable environment. In fact, a lot of it sounds mediocre to downright bad (music recorded and mixed in the 50s, 60s, 70s and 80s). I still love the music. The mix, unless it's horribly off, isn't really that critical. Balance is important, sure, but do we look at a Picasso and say "you know, that blue he chose it a bit strident - I would have painted that in a better lit room"?
NO NO NO to AI voices. Please never again! 🤐 They disconnect you & your company from us - the audience of audio folks that you're trying to appeal and present too. The equivalent for me as a V.O. listening to an AI voice, is probably you being told - "Room treatment = waste of time - just use room correction, or an online mixing/mastering service instead" 🤮
Jesco I really really like you and your channel, but if I have to listen to your pdf thing one more time Im climbing up the walls, I think we all know it by now you dont have to mention it again and again and again and again ^^
This AI stuff is really obnoxiously cheesy and hyperbolic. And like most of my experiences with "AI," it quite often gets a LOT wrong. The reason for this is the source material... AI doesn't "know" what is correct, it only knows what is most commonly repeated, and if that information is wrong then its output is wrong. It also doesnt actually "know" anything about whatever subject its working on, it only has a vast library of source material to mashup its responses from. Like all things computerized, its the old adage, "garbage in, garbage out"
"AI doesn't "know" what is correct, it only knows what is most commonly repeated." - like 99.9% of humans, actually. We mostly just repeat what we hear, especially if it's what we *want* to hear.
The AI might be technologically impressive, but I don't think it helps your video in any way. I usually come to your videos so that you, the acoustics expert, can break down concepts for me, not for an AI to try rambling in podcast form to summarize a paper.
Yeah, same here I'm afraid
And I find that style of podcasting unlistenable (with all the excessive back-and-forth agreement between the presenters)
I had to skip over most of the AI babble
Prob is tho fellas in 5 years if your not using AI in your business, especially this particular one, which is the opinion of the current & former CEO of google btw AND the one in which they are both putting the most attention too, U will fail. Now if you guys just work for someone and don’t own a business and are not entrepreneurs, then who cares u like what u like and don’t wanna hear this from Jesco and that’s perfectly fine but the point is him trying it was/is a smart business move and shows just how in tune he is . Let me ask u both this question what should he have changed about his presentation that would have made it “Watchable”?? the fact u felt the need to just complain helps nothing and no one. Critiquing is one thing, but with no feed back you’re just criticizing and nobody wants to hear that BS. Respectfully of coarse …
I really liked the format, and hope you do something similar in future. As a layman, it’s difficult to get into the science and understand the papers. The AI did an incredible job of translating the jargon into something I could understand easily, with some surprisingly impressive analogies.
The issue with AI (and with random forum threads online), is that the summaries can be wrong, misleading, or cherry-picked. Having you, an expert, essentially act like a referee, allows me to get the data simplified for me, whilst being confident in its accuracy.
I hope you do this again with other topics, particularly ones that debunk commonly believed myths!
Great Video Jesco! I agree with you on all fronts here and super interesting to think more critically about ITD gap in smaller home studio rooms. Also, never had seen NotebookLM - super crazy and kinda scary thinking AI could create a full media content channel pretty soon.
I like the idea of applying LEDE to live rooms more than control rooms. I think that RFZ works better for control rooms.
Tbh I would have taken away a lot more If YOU had presented us the details of the LEDE paper.
The AI was incoherent and frankly wasted everybodys time with lots of talking but very little content.
It was fine and very enjoyable. 🙄
About 1 minute before the end the author compares line-of-sight signal alone w/out reflections vs. line-of-sight + spectrally correct, properly delayed, moderately high-level reflections. The author smartly posits whether the latter has any advantage vs. the former in a control room setting. Psycho-acoustics expert/recording engineer Dr. David Griesinger has scientifically compared the two and found that the latter increases listening pleasure with no downside, but I don't know his opinion for work purposes in a control room.
DG states that the latter (above) maximizes detail, immersion and envelopment. Dr. Floyd Toole states that the latter increases pitch sensitivity. I understand how and why line-of-sight signal alone may be ideal in a control room. OTOH...for a few months I have used a unique modular loudspeaker with a clean, undisturbed line-of-sight signal + late-onset reflection assist (spectrally correct, properly delayed, variable level.) IMO even someone listening to music for work in a control room would only benefit from using such a loudspeaker and I know of no downside. Imaging, spatial qualities and lack of proximity effect are all improved.
Re. ideal bass performance, Dr. Earl Geddes, Todd Welti of Harman Labs and numerous others including Dr. Robert E. Greene have long promoted the distributed bass array as the ideal solution to eliminate bass modes in three-dimensional space, before they propagate (smooth bass throughout the room except for light build-up in the corners.) Solving bass modes is otherwise virtually impossible because of the mathematical relationships between the room's boundaries and long bass wavelengths. Even if one lowers the peaks, filling in a 9+ dB dip is impossible with gain; EQ tends to only smooth a peak at one location of the room while making response worse at another location. Siting 4 subs @ strategic locations throughout the room adds 4 new randomized bass modes to the room's fixed bass modes, thus minimizing the effects of the latter (4 fixed modes vs. 4 fixed + 4 randomized modes.)
Griesinger promoted a method to further smooth the distributed array's modes called Phase Quadrature, identifying 2 subs each on the L and R sides. The ideal phase for the entire sub array is 90 degrees to maximize smoothness in the pass band (shared range of the mains and subs; less then 90 = lumpier bass, more than 90 risks power loss at the lowest F.) Griesinger suggests setting the phase of 2 subs on one side of the room @ 45-degrees and the remaining 2 subs on the opposite side @ 135-degrees. The average is still 90-degrees, but the asymmetrical phase also maximizes smoothness below the pass band. (For the record, the Double Bass Array may work even better than a properly tuned Distributed Bass Array, but the former is about 4x costlier than the latter and requires building 4 subs each into the front and rear walls, 8 subs total. The 4 front subs are in-phase with the music program; a Double Bass Array has the 4 rear subs 180-degrees out of phase and delayed for a period equal to the room's length. Bass from the front array travels to the rear of the room where it disappears due to the rear array's inverted phase.)
I spent about twice as much on acoustic treatments ($750 on professional plans alone, acoustic soffits and corner traps) vs. the cost of the distributed bass array and the latter's performance incinerated the former's. There's a reason a particular acoustics expert calls full range speakers, "a failed experiment."
I enjoy your content. Tricky subject with no firm answers, but I've followed a lot of your advice and, at the very least, you've helped me avoid wasting money or making things worse!
This was a fascinating use of AI. While it had its moments, I thought the format in which it presented its interpretation was engaging and personable. And the topic was, of course, always an interesting discussion. Kudos on a great experiment!
Hey jesco, you were testing the AVAA, how did it go
Please please don't do this 'AI podcast' format again 🙏🏻
I find that style of wafflecast insufferable. I can only listen to a podcast if the guest speaker talks concisely, and is allowed to speak without interruption.
That's what is so good about your channel Jesco so please, don't put us through that again!
(Interesting subject matter, ofc!)
For the next Q&A Knowledge Bomb:
I’m building my own desk for my studio, and I had an interesting thought I’d like you to consider. The desk will be a standing desk, and it’s about 4 ft from the front wall. I’ll need a way to hide cables. Rather than just putting a piece of wood on the back, what if I made it out of acoustic insulation that was covered on both sides by fabric? Would this have any benefit at all?
great format
What if the back wall can be taken out of the equation entirely-ish. I positioned my studio so the French doors are centered in the back wall, about 5 meters away. I have massive gobos covering the doors that can be moved, but I can also open the patio doors all the way, creating about 3.5 square meter opening. Does bass just flow right on out the hole int he wall? Does it make any difference? The only reason I have this option is because I live quite remote in the mountains so the only noise is wind and the occasional little beastie wandering about. I did this mostly for options during the recording process. Most significantly, would this change the acoustics of the room enough to confuse my learning of the room for mixing? Or maybe another arrow in the quiver for mixing listening options?
Thank you for your time sir
You’re an FN genius for using notebook for this. Dude I’d love to collaborate with you on some interesting treatment ideas and designs I’ve come up with thru trial and error in my own space. It’d be too difficult to get my point across thru email so would you be interested in working out a call or zoom or something and just for a few minutes of your time? Look at it like this, if u don’t like my ideas, you can tell me to get lost idiot, an that’s always fun, or I’m on to something and we potently give acoustics a much needed update /upgrade then we go down in history for changing the game!! and for me ,well, I hit the drawing board BUT continue to watch your vids an learn no matter what you think…Frankly it’s a win win for both of us either way!!!😂😂😂 don’t mind my corny attempt at being funny, can’t help it when I get excited about an idea🤷♂️😂Thanks Jes my friend appreciate all you do
18:40 Because, like you have said, speakers don’t have enough mass to decouple properly. Would it help to put something heavy on them to weigh them down? Or would that create more resonances or other problems? Thanks for the video btw! Great content!
So is LEDE design opposite of FTB design? We would like to know much about FTB design. It will be great if you share your opinions about FTB design. Thank you.
Hello I have a great question for you and would love your input since you know treated rooms well and have experience mixing music . I was debating on treating my room or using vsx headphones . What is your input on the studio emulations in place as I'm sure you have probably tried these out at one point ? I'm listening through them right now and they do sound like a treated room where I can hear details very well . Looking forward to your input !
The idea of Initial Sound Delay Gap is about psychoacoustic grounding in the studio as well as perceiving the acoustics of the recording environment. Completely dead studios clearly do work but the environment is unnatural.
interesting topic but i think i prefer more succinct summaries of source materials instead. That lengthy podcast feels like wasted time to me.
Strewth! 🙄
Hey Jesco Big Fan ;
Quick question. If you had to chose one over the other, which one would you choose?
Because of the room shape, I only have 2 options.
1. Equal lateral triangle at 55 inches on a 2.1 system. But my sub is a bit closer to a side wall than the 1/4 rule. Also one speaker is closer to a side wall than the other. But I haven’t an equilateral triangle at 55”.
2. My speakers are equal distance from back and side walls , and sub is in correct place. But from tweeter to tweeter is aprox 77.7” and from speaker to listening position is aprox 46” or so. So the equilateral triangle is not proper.
I haven’t a ton of treatment , all inspired by you. I’ll send pics. In pic it’s set up as #2.
Also I have the Neumann kh 80’s and kh750 with Dsp. I assume the triangle is most important as Dsp may fix the offset timing by the side walls. But I’d love to hear your opinion on this matter. Thank you so much.
Bill Ibe
Bringing speakers out a meter from the wall is the hi-fi thing to do to increase soundstage depth. The AI 🤖 picked up on that
Interesting video. Re your comment of maximizing the gap, and supressing the room...don't you think that a totally dead room is unpleasant to work in? Isn't that why we include diffusion?
Have you done anything comparing tools like Trinnov, ARC, and Sonarworks? Or do you think they even help?
Oh they can help very much, especially the Trinnov. But not in isolation.... You have to remember that there are really several components to a correctly controlled room. The proper placement of speaker and listening position in the room. Acoustic treatment, and then room correction. Trinnov is not a magic box that can fix everything all by itself, it fist needs the other elements to be done right, but once they have been done, Trinnov can take care of a lot of the rest, especially for things like phase correction and soundstage.
Was really looking forward to this video but I’m not in to listening to a computer pretending to be human in a podcast talking about a complex, technical and somewhat subjective subject that I’m also very passionate about. I’ll skip the video and read the papers instead.
Don't the vast majority of consumer listening environments range from less than ideal to hot garbage? Does getting the "optimal" mixing environment really matter if the consumer isn't going to be experiencing the finished result as intended? Isn't a "perfect" listening environment only slightly better than a "good" one? I understand the idea that home studios aren't ideal for mixing because of early reflections, standing modes etc, and that real issues can arise from mixing in an environment like that, but isn't "pretty good" going to result in a mix that sounds good in most environments? My favorite music isn't music that is perfectly balanced and nicely packaged for every conceivable environment. In fact, a lot of it sounds mediocre to downright bad (music recorded and mixed in the 50s, 60s, 70s and 80s). I still love the music. The mix, unless it's horribly off, isn't really that critical. Balance is important, sure, but do we look at a Picasso and say "you know, that blue he chose it a bit strident - I would have painted that in a better lit room"?
I’m not a fan of the AI. Otherwise great video.
A wee bit too much phatic discourse from Jane. It was a cool idea but I honestly enjoy your speaking voice more.
NO NO NO to AI voices. Please never again! 🤐 They disconnect you & your company from us - the audience of audio folks that you're trying to appeal and present too.
The equivalent for me as a V.O. listening to an AI voice, is probably you being told - "Room treatment = waste of time - just use room correction, or an online mixing/mastering service instead" 🤮
Jesco I really really like you and your channel, but if I have to listen to your pdf thing one more time Im climbing up the walls, I think we all know it by now you dont have to mention it again and again and again and again ^^