TRUE TARGET by Magic Beans Audio | Presentation | Q&A | Demo

[@joentell]

For the more intermediate and advanced folks out there, here's a more technical explanation. I think by trying to simplify it, a lot has been lost. It's tricky to explain because they're used to thinking of target curves as something that is static and pre-determined prior to measuring. We determine the room's transfer function by measuring the near field and MLP response. If we know the delta, that will tell us the room's response, or what we're calling RSS Room's Sound Signature. If we correct the MLP measurement to follow the RSS, that will inherently correct the nearfield response of that speaker to something very close to a flat response. It differs from simply taking the inverse of the nearfield response in that it takes into account the MLP measurement and how the bass integrates with the room in the nearfield and therefore doesn't over correct the bass by improperly trying to make the bass completely flat NF. This method also smoothly accounts for the transition region since it's based on the room's transfer function. Because research has shown that people prefer a speaker that performs flat anechoic with a smooth off-axis, we can attempt to make a speaker flat with the limitations of the speaker in mind. The response of that "ideal" speaker is different in different rooms and also dependent on the MLP location. That's why I say that the target isn't pre-defined and is determined by the speaker and room interaction. From there, we can create the target response curves we call the True Target curves as either a target curve or or as filters that create the target curve. Does this all make sense or is there a different way I can explain this to make it easier to understand?

[@adamjj85]

This helps. So if you take Channa's as an example, the true target had a dip in the low mids, say 100-300hz. How would you explain the logic for having the dip? Is that uncorrectable room anomalies or something else?

[@joentell]

@@adamjj85 it's uncorrectable. It's like trying to overpower a null at his MLP, which it probably is.

[@Redliger54]

​ @joentell I've been reading your conversation to try and understand exactly what you are trying to say in the video, and I have to agree a little with ukspawn on how your wording can make it confusing. I wanna support you, but its hard to understand when you on speak only your terms, and not on a confused person's terms of understanding. Especially for a beginner like me. So if you could please use the quoted terms the way I use them, I think it might clear up some misunderstanding. I really appreciate you taking the time to respond to us btw! Since I'm a beginner in this hobby, my understanding of "target curve" is the frequency response I should be hearing at the MLP. (I might be wrong) For what I understand of your conversation is that your use of "target curve" is the eq needed to make the speaker sound like the perfect speaker at the MLP (am I wrong?) I think you're saying that "True target" or as I understand it "the eq correction curve" is determined by correcting for the RSS at the MLP ( am I getting this?) Because this relies on variables such as the Nearfield speaker measurements and the room transfer function, you are saying we can't just correct to harman in room for every speaker, and that the "target curve" or frequency response we hear at the MLP needs to be different for every speaker b/c of its position within the room and how it interacts with it. (hope i got it) If you've ever seen Initial D, there's a driver called god hand who doesn't follow the correct racing line, but is still faster than everyone else. The way he explains it is that a racing line is not the goal, but simply the result of good tire management. I think that is similar to what you're trying to say if I am understanding you correctly. Harman In room response is the traditional racing line, but True Target may not produce this "ideal line," but still makes the speakers in the system sound ideal because its based on other scientific principles other than matching the harman target. (is that correct?)

[@adamjj85]

@@joentellOk that makes sense. So my interpretation is that this is presenting the true response at the MLP taking into account uncorrectable room anomalies. So while someone may have a traditional house curve like Harman set as a "target" wit their room EQ, that may not be 100% what they end up with due to the uncorrectable anomalies. That being said, wouldn't the true target be the same result as something like Dirac or other Room EQ would present anyways when trying to correct to the Harman curve for example? I'd be interested in testing it out to see. Or are you saying the main benefit is that true target knows what the speaker is doing vs. the room and it can better craft the correction filters vs. auto EQ trying to correct things it can't and maybe doing more harm than good?

[@joentell]

@@adamjj85 I agree with almost all of what you said, except for the part about Dirac or Audyssey providing similar recommended target responses. Those attempt to first EQ the MLP measurements to a flat response, then they overlay a generic recommended target on top of that. This has been the issue I've had with them since the beginning which is they leave it up to the user to know what their target curve should be. Most people don't know and so they end up using the default recommendation. If they're lucky, it will be close enough, if not it's considerably off but hopefully the parameters set in the app disallow correction for a bad fit. For the EQ that it incorrectly applies, it ends up sounding unnatural at times and doesn't make efficient use of the system's power.

[@TechnoDad]

Oh yeah! Wish I was there for this event. Looks like it was a good time. It's time to UNLEASH THE BEANS!! LOL!

[@angelajudetv]

RELEASE THE BEANS!!!

[@TrueLies23]

Excited for this. To be honest, I've held off buying MEQX for my 4700 because I just haven't felt that I'm missing anything to justify the price point. Many products have caused me to approach with curious skepticism. The way you've been able to convey comparisons and results to us has me really impressed and excited for the future of this product and for you. I know a lot of people are going to see value in the product. You, your mind, the way it works, what it has achieved and will achieve, that is of way more value than any one thing you produce, whether it's this app, SACT, anything. I just wanted to make sure that I took a minute to appreciate you for you, and not just your contributions to the audio community or what you have to offer the world. Keep it up!

[@joentell]

This is one of the coolest things I've heard someone say to me. Thank you. I'll be showing my wife for sure! Thank you.

[@m4nc1n1]

MultiEQ-X is 100% worth it. Been doing HT for 30 years and using Marantz products for 15 of them. Definitley worth it just to delete midrange compensation! lol. That should of never been the default.

[@TrueLies23]

@@m4nc1n1 I'll be dropping several hundred dollars all at once I'm sure: MEQX SACT MBTT MiniDSP XYZ, 123, now I need it all, what have you guys done to me 😫 😂😂

[@JohnTwo1]

@@m4nc1n1 multieq mobile phone app can remove that also.

[@BrodeyDoverosx]

DIRAC incorporates the room when you use the full recording positions option. That’s why you take all the different points, even those where no one will sit.

[@joentell]

They use various positions to ensure they aren't overcorrecting for any single position.

[@adamjj85]

I will echo Justin from Harman's confusion and some others in the comments. Admittedly, I still don't understand the argument of needing a different target curve for each room. Perhaps it is still a terminology issue. The correction and filters would vary for sure, but the end result (target) should be the same in each room no? Otherwise, we have no benchmark for everyone to shoot for to reach "reference" sound that studios are tuned to. And also to reach what Toole had proven most listeners prefer, elevated bass, a smooth midrange, and sloping treble.

[@joentell]

If you aren't understanding, I definitely need to make another video explaining this concept. Let me write a comment and pin it.

[@joentell]

Ok, I just pinned a comment. Let me know if you get it

[@thamaesta]

This is awesome. You know what would be even cooler. Maybe the next step. That you can use binaural mics for calibration.😮

[@joentell]

I have an idea for that already. It's very useful for certain types of calibration. I'll save that for the future.

[@EveryDayJay415]

Very nice way of explaining things! That engineer from Harman had a good point about not using 1000s of filter’s that you already know won’t work for everyone. Sometimes keeping things simple is better then complicated methods. “Keep it simple stupid”

[@joentell]

I think that was Brandon, the custom integrator. It's true though. Not only will a generic target not work for everyone, it will waste power because it might try to fix unfixable issues.

[@EveryDayJay415]

@@joentell yep, similar to trying to over correct a car that’s losing control, sometimes you gotta just let it do its thing and not fight physics. In this case the room is the slippery road and our speakers are the sports car.

[@joentell]

Exactly. You know how it is. When the back end comes loose, you sometimes have to turn the wheel in the same direction as the slide to reset the balance before counter-steering.@@EveryDayJay415

[@joentell]

And yet another analogy. If you go off-roading, you can't expect it to ride flat like a limo. Sure, it can absorb some bumps and dips to smooth things out, but on some of those big bumps, you just gotta let it fly. LOL!

[@stingber]

Hi Joe, I totally understand your logic and it make sense. I look forward to try it in the beta testing. Excited to find out what would be my target curve for my system 😃👍📻

[@lbudt29]

Once you get this established, then you can work on widening the sweet spot or make it mobile, when you walk around the room.

[@joentell]

This can be done by adjusting how wide of a space you measure in during the MLP measurements

[@rf95-n9s]

If I’m understanding right, I think this method is throwing the idea of a “target” or “house curve” all together. After all, it is true that Toole never thought the Harman curve was really something that should be applied to all speakers. It was just the observable response of preferred anechoically neutral speakers within actual room. It seems like the main point of magic beans is to differentiate the causes of the deficiencies of the sound. By measuring nearfield and at the MLP, you’re able to tell with the nearfield measurements the problems with the speakers themselves, and then with the farfield/MLP measurements, you’re able to tell what the room itself is doing. With this additional information, you can make smarter decisions about what and how to EQ. I still don’t fully understand how the final “target” is exactly calculated, but it’s starting to make more sense after listening, thinking, and reading some more. It’s taking the additional nearfield measurements and how they compare that makes the main difference.

[@joentell]

It's a breath of fresh air to find someone who understands and has read the full context of the Harman curve. I have a pinned comment with more details and another video with TechnoDad showing how it works. I also made another recent video on @dailyhifi explaining it in greater detail.

[@jmxtra]

Brilliant approach!! adjusting system (amplifier, speaker and room) via true target curve/filters, so that at the listening position you receive an "accurate" system output response for any input/source material. So a source signal input is always modified by the playback system(amp, speakers, room, listening position, etc); so the true target curve will ensure your output from your system matches the input to the system. I'd love to try out your software on my setup if possible, thanks. John

[@joentell]

It's available at magicbeansaudio.com

[@urbanlarsson8252]

I can see a considerable high 60Hz mains hum peak on the laptop. How is that treated as you cannot hear it but the mic seems to be electrical sensitive to it and adds that to the reference as being part of the rooms acoustics?

[@Audio_Simon]

Joe, I want to make sure I fully understood your approach. As I understand it, the goal is to avoid flattening the speakers' frequency response entirely, as people enjoy the sound character of their own speakers. That's presumably why they purchased them! Instead of aiming for a generic target curve, you take measurements both near the speakers and at the listening position. By comparing these, you can separate the speakers' intentional sound characteristics from unwanted room coloration. Then, you apply EQ selectively to address the room issues while preserving the speakers' natural tonality as much as possible. This avoids over-equalizing and allows you to improve clarity without fundamentally altering the intended sound. Please let me know if I have that right or if I'm missing anything.

[@joentell]

I've pinned a more in-depth explanation. I've also explained here on our Daily HiFi Podcast kzbin.info/www/bejne/iJeshZJnfK6ZmsUsi=WoH1buKirHdE44Bb

[@Audio_Simon]

@joentell Thank you I did see the pinned message but wanted to see if my more wordy explanation lined up. I'll watch the podcast!

[@trauma50disaster1]

@@Audio_Simon why would we assume near field sound is anything close to what the engineer designed to be heard at 10 feet or mlp?

[@wktek48]

Will follow this, always looking for other ways to calibrate my 5.1.4 system.

[@hdmoviesource]

Me and my brother watched this in full today, trying to understand this new method. My current EQ that Im using on my system is the best Ive ever got my system to sound. It was measured at the MLP only. Could this method be better? Only testing it will know. From what I saw of the demo, the demo system looked to have a target curve of +10db from treble to bass. The harman target curve I use is an 8db rise from treble to bass, but that's in my room. I'd be very interested to see what this app does to my speakers, and the sub. My subs are using a 6db rise from 80hz to 30hz and its the best Ive ever got it to sound I've used 8db on my subs before, but this sounds even better, and its less bass, technically. Im already thinking about ways to improve the app, but I need to understand what its trying to achieve first. I want to know if corrects for nearfield, or try's to mirror the nearfield response in the MLP? I also don't know whether its correcting the nearfield at all? Or just using it to cross-correlate the results to MLP? This is fun, and BTW, I actually use a similar method to calibrate my subs with, and its 100% based on the results of the original measurement, and not based on random numbers. This is why this method is intriguing.

[@joentell]

Check out my pinned comment if you want to understand in more detail

[@hdmoviesource]

Thanks Joe.@@joentell

[@Saturn2888]

Oh, this is what you were talking about with target curves! I thought this was a manual Dirac Live competitor.

[@hdmoviesource]

When can I buy it Joe? Whats the price going to be? Can I add my own target curve? Can I download the app, find the app on my phone with my PC, and edit the target curve folder to be what I put in?

[@joentell]

Pricing is still not set. It will be competitive. You can add your own changes in MultEQ-X, but the purpose of the app is to give the True Target for each speaker. You can season to taste later. It also depends on what you're using. With Dirac, it draws a target curve that you can adjust by adjusting the points.

[@BenjiGator23]

Joe, thanks for the explanation and can’t wait to try it out on my system. A couple of questions come to mind so i hope you can assist. Can this do anything to multisub setups like phase alignment between the two subs together, as well phase alignment between the subs and main speakers? Also, what volume level should the AVR be when calibrating? Do we need to adjust to 75 db on the master volume?

[@StylianosPourgoures]

@joentell I think I started to understand this after some reading around Dr. Toole's opinions on target curves and Shroeder's Frequency (transition frequency). So basically if I understand correctly the combined data of the near field and the MLP will allow the app to determine the transition frequency of the room and use the ideal filters to EQ above and below that in a way to fix room problems below with broad and narrow filters and possibly fix speaker problems with broad filters (as long as your speakers are reasonably flat in the listening window) without EQing them in a way that adds resonances and degrading the speaker. Narrow filters above the transition frequency should only be use to correct speakers based on anechoic data and not MLP measurements. Correct me if I'm wrong.

[@joentell]

One thing I would say is I do think an in situ nearfield measurement is useful considering that nearby surfaces can act as the speaker's "waveguide" and should be accounted for. An anechoic measurements of that speaker won't do that.

[@jankdmtg6995]

The concept sound somewhat like RoomPerfect from Lyngdorf. I remember hearing Lyngdorfs correction early on, and it had a specific issue - it made every speaker sound like Lyngdorf; so to speak. Well, since TDAI-3400 they have implemented a new strategy, where they let the speakers keep their signature, while removing room interaction. And it is pretty awesome, but also quite expensive. This has led me down a path of trying to recreate Lyngdorf, but DIY. Steinway-Lyngdorf use open baffles, so I deviced an open baffle with low diffraction, minimal phase errors and wide sound dispersion. It works really well, but I need to find the right room correction software, and this sounds like just the right thing (and I have applied for the beta).

[@BobaDominasian]

So basically are you trying to maximize the speaker's performance in the room that it's in?

[@lycancatt4248]

there is litterally 0 point in knowing what a speaker in a room can do, as every room is different. the harmen dude has a really good point that how you get to a certain target doesnt matter at all, and coming from his background as well where it has to work all the time wherever a product lands, that makes total sense

[@joentell]

Thanks for commenting. I'm working on another video that will help you understand the method in more detail.

[@hiramb7956]

I agree with the concept that each room may require its own custom "target" / correction. Generic curves do not fit all Rooms. Yet when we change the speakers, or seat positions, we should change the Filters or EQ, but not the Room Target, right? Can you clarify. thx.

[@joentell]

The target will still change because the room effect on the speakers will change at the listening position

[@kewlbug]

God bless Justin and his patience.

[@JohnSmith-qi6co]

I get it that there must be a target curve that’s best for (1) me and (2) my room and (3) my equipment all together. In a perfect universe, let’s say I try my 100 best tracks against a million different target curves and I finally find the best. While I believe you that it might work but what reason is there the target curve MB comes up with will be close to that great one I’m in search of. Why would that be true? Thanks

[@nedkulic2164]

Joe I really can't wait mate till it's out..... I'm looking forward to doing the Beta test as I have a 9.2.6 Setup 😊

[@joentell]

Did you sign up for the beta? If so, you should have received an invite to purchase it.

[@nedkulic2164]

@@joentell Hi Joe... I did sign up and all I got was Thanknyou email.... I subscribed to your early release on your website. The one with the count down clock! Is there another website for the beta version? Ned

[@joentell]

@@nedkulic2164 if you signed up for the public beta, you should've received the invite. Email [e-mail] and we will be able to see double-check. Include your first and last name in the email.

[@bayapaya892]

Can i run magic beans straight from my MARANTZ 7706 OR DOES IT RUN FROM THE COMPUTER 🖥?

[@joentell]

From the computer or mobile device

[@ROYALTYMB]

I recall watching a video by Audioholics where they mentioned that Auto EQ could be significantly more effective if it allowed the input of spinorama data for the specific speakers being used. This would enable the receiver's Auto EQ to adjust the frequency response to match the speaker's natural curve, rather than attempting to conform it to a predetermined house curve. if I'm understanding this right that's what this is trying to do.

[@joentell]

I have some current videos about target curves that specifically addresses this. I don't think EQ'ing to the speaker's natural curve is the best method. What happens if your speaker's natural frequency response isn't very good, but the directivity shows that it is eq'able. Why not make it better than the natural response?

[@mandybettinger6423]

What curve would you use to make the speaker sound good?

[@joentell]

@@mandybettinger6423 flat on-axis nearfield taking directivity into consideration above the transition region and a bass rise that follows the room's actual response.

[@mandybettinger6423]

Generally you're told when taking MLP measurements that you're supposed to only EQ to the room transition frequency. if I'm understanding correctly, what this does is uses a MLP measurement to correct up to room transition frequency that will follow what the room is doing to the speaker, and a near field measurement for frequencies above the room transition. Doing this would nullify the negative effects of EQ above room transition frequency. Thanks for your patience in explaining this. I am really interested.

[@joentell]

@@mandybettinger6423 well said

[@BuffSquadBigBenni]

Love the concept, and much respect to you Joe, but I still think the explanation of the target curve is kinda vague (maybe I am just stupid, that is very plausible). I acknowledge what you are saying, that the target curve is dependant of the room + speaker interaction, so that you don't force a strict/absolute target curve over your setup, but instead "mould" what you have into a somewhat optimal target curve. But there must be some kind of built in desired target curve or guideline, in your "true target" - what is this desired target? I know many factors are taken into consideration, but I never heard you specific state what the desired target curve in your app is.

[@joentell]

I think people are caught up in thinking there actually IS a pre-defined target response at the MLP. The app determines the True Target based on the parameters of the speaker and the room. We can't possibly know prior what the optimal response is for speakers in a room. We can for headphones, because it's mostly a static environment that's relatively consistent from one person to another. I'll try to use a simplified example. How many watts do you need for a speaker to play at an 85dB reference level? It's not possible to say without knowing the speaker sensitivity, whether it's indoors or outdoors, and how far you are from the speaker. So, if we don't know something as simple as the wattage without knowing those parameters, how can we know how to change the signal going to an unknown speaker so it has a proper response at the MLP? Unlike volume, we don't know what the desired response of different speakers are in different rooms and at different distances. We only know the desired response we want from the speaker itself. In an ideal room, having a perfectly flat, neutral speaker with great directivity is preferable. Unfortunately, we don't often have an ideal room. We have nearby reflections and imperfect speakers.

[@BuffSquadBigBenni]

@@joentell Alright I hear you. But in order for your app to calculate the target response, you must have programmed some guidelines into it? Unless it really is some kind of magic and the app calculates a response for us, without any form of instruction from its maker :D You must have programmed some kind of favorable curve into the app, or how else should it know how what to correct for? To me it sounds like you want to keep it a secret, as some kind of proprietary info. And that is completely fair, I get it, but then that is just what you have to communicate to the potential customers of the app. Looking at the commentaries to this video, it seems I am not the only one confused about these things. Again, no disrespect, just curious like yourself. I also do manual calibrations and always want to know the knitty gritty of things :)

[@joentell]

@@BuffSquadBigBenni I get that you're genuinely curious. The method itself determines the output curve. I don't know what the output curve is prior because I don't know the speaker or room response prior. The goal is to optimize the speaker Nearfield (flat if possible) and smooth the response of the bass farfield (if possible.) We also want a smooth transition that follows what's possible for the speaker based on what the room is doing to the response. So the parameters are based on figuring out what is outside of the correctable range in order to not overcorrect something that isn't correctable in the first place. Does that help at all? To simplify let's just assume I want to do something that most people calibrating nearfield studio monitors aim to do which is to make the speaker flat nearfield. What will the response be from 8ft away? I have no idea. It depends on the room.

[@BuffSquadBigBenni]

@@joentell I think so. Thanks for taking your time to explain. Good luck with the app, I think it will be popular! I purchased the SACT this black friday, and it did wonders for my setup even though I would classify myself as an advanced user. Wish you the best. Are Chana part of your Magic Beans project, or is this your own venture? I think you're a great team.

[@joentell]

@@BuffSquadBigBenni it's my own independent venture, but Channa is a good friend and we help each other out whenever we can.

[@AllroyAllroy]

When the app is released, what will we need to implement it, are only specific receivers able to be used, do we need a mic?

[@joentell]

We recommend UMIK-1. We will try to support as many models as possible. Some AVR's are simply too limited in what they allow a user to do.

[@ashikmusic7364]

I would like to buy now but thinking about how can I export it to cinema 50

[@saschawehde8072]

Hey Sir, can you tell me if the moving microphone method (near field + MLP) works only with the 0 degree file (method used in the Techno Dad video -> you calibrated his System) or can one use the 90 degree file for both as well (when using this method in REW)? Thanks

[@phagetfish]

I have 8 seat positions and Audyssey seems to create a 'bubble' around the MLP. How does Magic Beans Audio work for multiple seating locations? From Channa and above demos its seems to work a target for the MLP only? I have taken Audyssey a step further with the subs by using Multi-Sub Optimiser to get a more balanced bass response across 6 of the 8 seats, with the MLP having the best response.

[@joentell]

This won't replace MSO, I use that also. I'm sure you already know that the more you optimize for more seats, the less optimal for a single seat. But using this method, you're free to move the mic to a wider area to get a larger spatial average. As much as you can cover in 10 seconds

[@kuronekosan2830]

How to use it with equalizer apo for 5.1.2 ? Please make a tutorial on this

[@erikherbig8868]

I have a question about measuring subwoofers. How would you go about measuring a subwoofer that has the ports in the back and the speaker cone is not visible?

[@joentell]

For our current implementation of this app, the only thing that matters for subs is the response at the MLP. At those frequencies the subs are inseparable from the room.

[@sritri]

So when MB comes out, I would need to buy MB and probably Mic and accessories, preferably SpatialCD, "and" a $200 non-transferable licence from Audessey MultiEQ-X if am using a Denon AVR ?

[@joentell]

You will need a UMIK-1 or other calibrated mic. The Spatial Toolkit makes it easier, but you can use test tones from REW for up to 7.1 over HDMI. You just need to switch speakers over to the accessible channels if you have over 7.1. MultEQ-X is highly recommended because it allows you to exclude the measurements and corrections made by Audyssey so we can control corrections 100%. You can also use the $20 Audyssey Editor app to draw the curves or even use the manual graphic EQ per speaker, but those methods are far less precise.

[@AmazonasBiotop]

What I think I might have understood. You do moving mic capture at sweet spot (or how large you want it to be).instead of number of points. That is almost potato or poteto. The "only" difference is the near field measurement of the speaker. So you know exactly what comes out of the speaker. And you know exactly what arrives at the sweet spot. So you then you know what the difference between what come out of the speaker and what is arriving to the sweet spot. Then you know the what the room is doing.. That is that difference. So with that information of what that SPECIFIC room is doing then you should know: * What decay time each frequency has that the room dictate * What is the the room modes from the rooms dimension. * what the speaker frequency response is and what it can't do. Actually we could evolve the product so that it gave feedback to the user. That in HIS room with HIS speakers: * He could move the sweet spot back and forth (bass null/peak) * when we know what frequency range has most reflections for this specific room. Then a absorber of specific thickness that target that specific range. * If speaker need a little more bass or less bass. By moving closer/further away from or to wall or corner. In other words a interactive setup tool for that room and those speakers (and electronics). As a way to evolve and doing something unique and create a niche that may be needed when today we don't have that help in our home.. 😅❤

[@Masterslessons]

Logic son

[@joentell]

We have other very cool and innovative things similar to what you mentioned. I think it will blow people's minds.

[@trauma50disaster1]

we need a/b double blind testing from independent listeners.

[@jmxtra]

Hello Joe, didn't get a chance to check, however I suspect systems that test/measure system response at listening position will automatically adjust amplifier etc response for accurate system response: ie.flat input-output system response. Anyway, I would love to test your software if possible, John

[@joentell]

The software is available for purchase at magicbeansaudio.com

[@iqfav7884]

Hi Joe I have a Denon x4300h. I don't have multiq x. I know you said that you can export the filters to a 10 band eq. The x4300h only has a 9 band eq for each speaker. Will magic beans still work ?

[@joentell]

Yes it will work. You just exclude the first EQ

[@mikeortiz2139]

I think the use of the word "target" is incorrect. The Target is the Frequency Response of a speaker at the MLP. The Nearfield Frequency Response is your "Source". And the Room Sound Signature is really the Speaker Correction Curve. So roughly this: Target = Source + S.C.C.

[@joentell]

Nearfield response + correction + room tranfer function = target MLP response + correction = target

[@RXP91]

I've always thought that now we have all this Kippel and Anachoic data you should be able to correct the speaker to be perfectly flat too. I actually experimented this with on my LS50's but could barely hear the difference so didn't think it mattered much with good speakers.

[@joentell]

To be fair, the LS50's are already tuned well to begin with. KEF is heavily measurement-based and they know how to make great speakers.

[@stackoverflow8260]

Dirac already does impulse response correction..basically correcting the speaker first and then comes "room eq"?

[@Masterslessons]

No as well, shown on the graphs. But you better believe Dirac and others are watching and will steal this type of advancement. Happen to me with Dbox. I made the Dbox" literally and it was a kit for everyone for your computer chair only needing your chair not the legs and it worked with a box I made with a washing machine and Y axis in early 2000s. I would say protect your intentions and utility patent everything.

[@joentell]

We don't do impulse response and phase correction at the moment. This is designed to work with Dirac and other room calibrations.

[@Masterslessons]

@@joentell dude Joe one word for Atmos speakers. Artison Masterpiece and Sketch remember those? You need to demo those but not using them as LCR but as Auro Or Atmos Speakers. Many reasons why I can talk all day..

[@Masterslessons]

If you remember back in the days you can use them to replace a center channel as the speakers had 4 cabinets inside split and had the lower speakers and side and Left and Right while the center used parts of both speakers. Well using them as Auro speakers" imagine that, you can now replace the VOG speaker not the (center) and use a other pair and add the ((top center)) by having another pair of top mounted Artisons up front working its magic imagining a top center. No one else has done this and or even mentioned it but I thought I would bring it up. Also each speaker has a rear" tweeter that expands the sound stage " so when these bad boys mounted above on the ceiling it adds even more crazy effects.

[@stackoverflow8260]

@joentell that's a start.. but Joe, personally, I feel like I need more explanation about why your method is a better approach. I have always been a proponent for Placement -> Treatments-> DSP based FR/IR/Phase correction. I am looking forward to playing with your EQ though. In fact, here's what I'm trying to do to see if DLBC is working at all.. use miniDSP SHD for a xover at 50hz and 2 subs. Using REW, take individual L and R measurements and do an L+R measurement and calculate predicted L+R. Now, see if L+R is greater than or equal to predicted L + R. Also, inspect FIR at various frequencies up to 50hz and note any phase differences. Redo the same without miniDSP SHD but using DLBC and Integra setting the xover at 50hz. Hopefully, DLBC will have less phase differences when compared to the other case, and thus, L+R should be close to predicted L+R

[@mattgiunt]

So when is Magic Beans For Dummies coming out. You know we are out there and dont want to miss out.

[@joentell]

We've made the app easy to use for people with various levels of experience. From beginner to advanced.

[@mattgiunt]

I was able to follow along until you got to Audessey things since I own a Yamaha 2080 and never saw that calibration before. I think this app is going to be epic. It probably would make sense to put out short videos showing how to take the curve and input it into the big receiver players like Yamaha,Anthem, Arcam and any others Ive missed. I got great results from SACT disc after the learning curve.@@joentell

[@Prerich45]

Joe, prerich here, is Sean Olive involved with this also? This is kinda cool, I'm looking forward to this and I'm going to sign up as a beta tester...I hope I'm selected!

[@joentell]

No. I have never spoken to Dr. Sean Olive. I would like to one day eventually. I have commented on some of his FB posts, but I don't recall if he's every replied to any of those. 😁

[@jamespark1855]

as a professional sound systems engineer I am utterly confused what the product is trying to achieve and why it’s better

[@joentell]

I have a video in more detail on the @dailyhifi channel. I've also pinned a comment with an explanation.

[@ChadAV69]

I think you could've stopped at "your brain knows what to expect in terms of a curve in a certain spot in a certain room, anything different sounds weird". If you're sitting right next to a wall and you EQ it so the bass is less, it's going to sound weird due to all of the other psychoacoustical things going on in the room. It's not going to sound right, it's just going to sound like the speaker has a bass problem. Another example is say you're in a 17x12 room, you're sitting about 9 feet from the front wall and the speaker is 18 inches off the front wall. It's going to have a natural bass buildup due to being close to the wall. If you lower the bass, it's going to sound like it doesn't have enough bass even though it measures "appropriately". I think this is because your brain can hear that the rest of the sound (not bass) is interacting with the room in a way that indicates that the speaker is next to the front wall. It KNOWS there should be more bass. If you EQ the bass down, it sounds weird. I really think you're on to something here, Joe. I've been experimenting using your ideas just using REW and Audyssey and this is really something else. What I've been doing so far is I make everything over the transition frequency flat from a nearfield measurement, then I take MLP measurements and I just set the target curve to follow the natural bass rise of the speaker. I have never had a better sounding system. The only thing I am not sure of is how to get the proper subwoofer response curve since there's so many modal things that influence the measurements depending on subwoofer placement etc.. I am experimenting with following the curve of the center speaker like if it rises at 3db/oct I do that with the subwoofer. I haven't been able to do it yet but I will try it.

[@danielwander605]

So im stopping at 16:27, and ive flipped through a bunch of comments where people seem to get it and are excited. However, I still do not get what the goal is. A nearfield measurement will give a decent idea of what the speaker does anechoic. The MLP measurement will give you the speaker and the room. Then what? Most people pick a target curve they know they prefer or try different ones to find their preferred one. They then are EQing the combination of the speaker and the room to make it as close to their preferred target. So, what's the inherent flaw in this method? The only thing I can think of is you're trying to apply EQ to counteract the room to make the speaker sound as close to its anechoic response as possible. But essentially that's just a target curve chosen based on the natural response of the speaker. Maybe that's preferred by some people in some circumstances but that's not something you would need an app to do. So, I'm quite confused still.

[@joentell]

The issue with what most people do is they are arbitrarily picking a target curve. It's not based on anything related to their particular speaker or room's acoustic properties. You are almost there with your understanding of how it works. If you know the effects of the room and you use that as your target for your MLP measurement, the end result is something similar to a correction of your speaker to a flat response. That's the simple answer. There's a reason why we do it this way instead of simply inverting the nearfield response to make it flat. We want to take into account the bass response at MLP and as well as account for the transition region. It wouldn't make much sense to try to "remove the room" by making the MLP response sound like the NF response. That would probably sound pretty bad and isn't possible to do because the room will do what it wants.

[@danielwander605]

@@joentell I think that helps. So if I were to simplify it then I would say: it uses the near field and MLP measurements to determine a preferred target curve based on some formula you've come up with. I never looked at target curves as arbitrary though. I always considered it preference, but it's not universal, which I think is your point. So the room, the speakers, the source material, and personal preference can determine the "preferred" target curve for each specific setup. In the end personal preference is the most determining factor I think. Still, if the app produces some kind of baseline tailored for each specific setup then you can tweak to taste from there I guess.

[@hideopalescent]

It’s a bit funny that Joe subscribes more strongly to Floyd Toole than an actual harman employee 😅

[@joentell]

I don't know about that. I talk to Justin and I don't think he would say that. I can show him your comment and ask him if he agrees though. 😁

[@joentell]

​​@@ukspawn666ok. I will. Update: I sent him a screenshot and asked him to comment.

[@hideopalescent]

@@joentell cool... Reason I say this is because what you said makes sense to me and is sorta Toole 101. There are some room interactions we don't want the room correction to touch because a) it's impossible to fix with EQ and b) our brain/ear adapts to it quickly anyways. There are other good reasons for a target curve to vary as well... We expect to hear more relative bass in a small room versus a very large room... A speaker design that is not the JBL M2 (omni in low freq and quickly narrowing dispersion in high) e.g. omnidirectional, dipole, cardioid, line source, etc. is going to have different room response than the Harman curve and it wouldn't be "wrong". Just a couple off the top of my head. Anyway.. this is really cool stuff you're doing. Thank you!

[@joentell]

@@hideopalescent YAYYYY! Someone who understands what I'm saying! This makes me very hopeful! I pinned my comment with a more detailed explanation of how it works and why it works.

[@danielmiller469]

One question will this work on the mini dsp

[@joentell]

100%

[@FrankieKennethL]

Does this integrate w Anthem ARC in any way ? I didn’t see them as an option when you showed exports

[@joentell]

Can you please let me know what Anthem ARC supports? Does it support PEQ or graphic EQ import?

[@FrankieKennethL]

@@joentell unfortunately neither. I am not aware of any ARC Genesis option that allows imports nor user level PEQ.

[@joentell]

@@FrankieKennethL how can you adjust the target curve with it?

[@FrankieKennethL]

@@joentell there is a tilt slider that users can use to designate plus of minus of the curve and has roll-off parameters that users can adjust for and either subtle or aggressive roll-offs. I’m not smart enough to know how to explain it better. Lol

[@joentell]

@@FrankieKennethL oh, that doesn't sound very specific. It sounds like they don't allow the user to design their own target so you have to rely on their calibration method.

[@cnhhnc]

Magic Beans? Good name there!

[@joentell]

Lol. It was a joke and I ran with it.

[@jeffg3595]

I feel like everyone using industry terms to talk above “curves” and “targets” is overcomplicating things. This is taking data from YOUR speakers/system, YOUR room, and YOUR MLP, using that data along with DSP to optimize your space referencing known “preferred” playback parameters like the Harmon curve for enthusiasts. Optimizing the playback for your system within your environment is the True Target

[@trauma50disaster1]

but how does the near field info matter for designed sound changes that are supposed to occur at distance, based on the the speaker design? Just because near field sound is different than 10 feet away doesn't make it wrong, the engineer designed the speaker knowing the characteristics at 10 feet.

[@rafaelsmith681]

I have to admit....watching this video confused me. I think...could be wrong...that this app is trying to make each speaker sound the same at the MLP as it does nearfield. I could care less what we call it target, correction, etc...what is confusing is what is its final "goal"...which again I think is making each speaker sound the same as MLP as it does nearfield....which I am not to sure is what I really want. If that is not what its doing then we need a better explanation as to what "response" the app is trying to create for each speaker at the MLP....and if it will be unique to each speaker which again I am not sure is something we want. i.e what is it going to look like if I were to to do a REW measurement after using the app. I have been using MultiEQ-X simply as a tool to apply EQs...but I do all the measurement/EQ/Target work with REW (using harmon curve as target) and my system has never sounded better.

[@joentell]

I've pinned a comment with a more technical explanation

[@gufiaito69able]

Hey joe will this work with the htp-1?

[@joentell]

I have one in for review. It definitely should. It works with Dirac.

[@dwayneharris1438]

Would magic bean audio stop me from buying 1400 dollars worth of panels for my HT.

[@joentell]

It shouldn't. That's a good place to invest in your system's sound.

[@chiefrushhh]

Hopefully it's not too expensive

[@saulmcgill-ws5hy]

I must say I've never seen anyone take a measurement by waving it in front of the speaker before?

[@joentell]

kzbin.info4MN1qPUrPP4?si=jRJfJfxaVMAXMW_n

[@Masterslessons]

Joe imagine this. A Small 7-11 inch tablet setup or cellphone with micro 5.1 setup maybe piped Type C to a baby 5.1 speaker setup smaller than any energy take 5. Something even smaller just for the concept alone. I would like worlds first smallest functional home theater. Pocket Theater" I'm not trolling. There are communities of small PCs micro cars 1:18 scales I think it would pretty nice a little baby 5 inch subwoofer being calibrations on the beans

[@joentell]

Yes. I think there's room for all of these ideas!

[@EveryDayJay415]

This is almost exactly what I’ve been dreaming of build for my RV even made a few prototypes

[@davidhoover2446]

I’ve been tuning PA systems for about 15 years. There are so many different parameters. Even when I’m using Smaart - it might look good, but sound like junk. There is not a piece of equipment that can replace a professional tuning a system by ear. Phase of eq and crossovers or linear phase crossovers or a speakers not taking EQ and sucking the dynamics out of the system because you pushed the phase change too much, etc. you really need. I’m sure one day algorithms can get close, but the speaker to room interaction can be heard and calculated by the ears and a great result can be achieved.

[@joentell]

The goal of psychoacoustics is to close the gap between what we can measure and what we humans perceive. When you say it looks good in Smaart, what do you mean specifically? Do you have a specific target curve you aim for each time?

[@lycancatt4248]

this was my take away too, a pa/dsp tuner here by trade and it seems like this is taking what we do and trying to make it universal, and it simply can't be. everyone has a target curve, especially if they've had a recent audiology report so applying this much to the signal is likely robbing dynamics and realism from the sound.

[@peterphan227]

I'm totally confused. Not a good way to sell a product if you can't explain it. The more you explained, the less I understood. I actually got more confused.

[@joentell]

I explained that this was an impromptu presentation. Sell a product? It's not available just yet, so I don't have anything to sell at the moment. What part don't you understand? I'm happy to explain it. It seems that others are confused about it also, so I'm here to help clarify.

[@Masterslessons]

I don't want Dirac to steal this and mix an A.I.function to adapt for all the people in the listening position and giving this technology a boost x10 you better patent pending this and get scripts for A.I to function extensions. Looks like it's all going that way and fast. Like how Getting VR setups done and configured, the audio with calibration and be simplified and mainstreamed. This is scary

[@Masterslessons]

Sony DAV-IS10/BM2 DVD Home Theater System in my closet needs serious calibration. I love it's micro foam factor. Please help 🙏

[@joentell]

I offered this to Dirac when they were here a year ago. They weren't interested and I understand they're busy with the stuff they're working on.

[@westsmith8987]

Joe, not that important, but the analogy the guy brought up about going to New York, perhaps this is a better way to use that, say you want to go to NYC. If you go from LA you have to take a certain route. But if I want to to NYC here From Chicago. I need different directions to get to the “desired” place. Magic Beans is the GPS that maps out the best to to the destination from your starting location. Just popped into my head when I was watching

[@ffdfdfdfdfdfdfdf]

This is hilarious

[@joentell]

In what way? Just seeing if you're trolling or if you have something of value to add.

[@texmuphy68]

The confused guy is all about theoretics, perfect targets and measurements and not about what sounds good to your ears.

[@joentell]

I try to tell people that acoustics and psychoacoustics are different fields of study. Closely related, but psychoacoustics deals with how humans perceive sound.

[@MilGrip76]

Too many noobs starting at step 1, don't waste time with shaping your descriptive language for every person. Thr whole point of educating is giving others the tools to know why you are doing what you are doing.

[@LearnYDF]

Let me break it down as to what JoeNTell is trying to achieve here to the best of my knowledge from watching this video. True Target seems to be aiming to replicate your speaker's Sound Signature near field and achieve that same sound signature on your Main Listening Position (MLP). So what it does it take the Frequency Response of the recording of the near field of your speaker's sound signature, then sees the difference in the frequency response of the MLP. Then adds or subtracts in a PEQ way to achieve the same Frequency Response as what the Sound Signature of your speaker looks like near field. So if you have horrible sounding speakers, it is trying to replicate that same Frequency Response or the Sound Signature of the speaker from it's Near Field and make it sound like Near Field at the MLP. It completely misses the point of why people use Dirac or Audyssey or Anthem. There, the goal is so that the MLP achieves the target curve (be it Flat [Reference] or the Harman Curve). It does this by trying its best to fix the 2 flaws; that is your room gain AND your speaker's Frequency Response (Sound Signature/Flaw). The problem is, it is very hard to physically tune a speaker to produce the entire frequencies at the same decibels. That is why you get those peaks and troughs in the Frequency Response which are basically the flaws or weakness of your said speaker. This is where the Room EQ softwares we have on AV Receivers try to fix. The goal is to fix both your speaker's flaws AND the Room Gain so that ultimately, your MLP will get as close to the Reference Sound which is Flat as possible or following the Harman Target Curve for example. If it does impulse response, even better. It will balance out the time it takes for the sound to reach your ear at the same time no matter how far the left speaker is to the MLP as compared to the right speaker or the back speaker. But for JoeNTell's True Target, sadly, it is trying to just give you the Target Curve of your speaker's Sound Signature (Frequency Response) near field and apply is to the far field MLP. Meaning what you'll end up getting is a very messy sound signature of your speaker due to the inherent flaws in most speakers with its peaks and troughs. That is why he mentions it won't sound the same as the system in the other room. Because as mentioned, it is just trying to replicate your speaker's sound signature. And not aim to fix it. Ultimately, best way to describe and give an example is this: Dirac/Audyssey/Anthem - Target: Reference (Flat Target Curve) = 0 - 0 - 0 - 0 - 0 MLP (speaker's Far-Field Frequency Response + Room Gain) = 7 - 2 - 5 - 4 - 1 Correction Filter = -7, -2, -5, -4, -1 To get final sound output at MLP (fixed with Correction Filter) to be = 0 - 0 - 0 - 0 - 0 Joe N Tell's True Target - Target: Speaker's Near-Field Response Frequency Curve = 3 - 4 - 7 - 2 - 5 MLP (speaker's Far-Field Frequency Response + Room Gain) = 7 - 2 - 5 - 4 - 1 Correction Filter = -4, +2, +2, -2, +4 To get final sound output at MLP (fixed with Correction Filter) to be = 3 - 4 - 7 - 2 - 5 Again, like I said, the goal of the other sound system's Room EQ is to get your sound to be as close to Reference Sound (Flat) as possible or to your Target Curve. Thanks to the flexibility of these softwares. The goal of Joe N Tell's True Target is to copy/replicate the good and the bad (Flaws) of your speaker's Near-Field Frequency Response and apply it to the MLP Far-Field so that the Far-Field MLP sound matches that of its Near-Field readings. Which is far from ideal and especially since Reference sound for a majority of speakers out there is impossible because again, it is very hard to make a speaker that can physically be Flat in its volume throughout the entire Frequency Response. That is why we need DSPs to help balance out what needs to be increased or lowered for every part of the frequency via PEQ to help the speaker achieve that Reference Sound and not try to replicate a speaker's 'Sound Signature' which is just a glorified way to name a speaker's Flaw in it's Frequency Response.

[@joentell]

I appreciate your thorough response, but I have to say that you are wrong. It simply does not do what you've said. What you've stated there is a horrible idea and nearly impossible to really do. If I understand what you think this is doing, you're saying that I am trying to undo the room's influence to get a speaker's nearfield response to have the same as the farfield response. You cannot undo the room's influence on the sound in the way you've described. Frankly, even if you had an ideal speaker with a perfect response, you would not want it to sound exactly the same at the MLP. It would sound like you're using the speakers as headphones.

[@LearnYDF]

@@joentell Ah... Then I look forward to your proper video where you can give a proper explanation and introduction on how it works. Apologies for not getting it. I am just trying to make sense of what True Target is trying to do based on the explanation and examples in the video which seems to be a little confusing. I love your usual speaker reviews. Their explanations there are easier to follow so I hope you'll come out with an official video to give justice to your baby here as well. Looking forward to it!

[@joentell]

All good. If I were to over-simplify the process, we try to make the nearfield measurement of the speaker flat/accurate/neutral and we account for the bass response using the MLP measurement, where the bass and room interaction are inseparable. That's a simplified version. We can determine the transition region and gradually transition from one method to another without using any arbitrary calculations. @@LearnYDF

[@marktebbutt8052]

Hi from the UK. Following this with interest. Think the issue for me is a need to understand 'the how' MB achieves good sound. Which is traditionally whatever target curve is ones preference. My method is to buy speakers which I like the sound signature of. Position them as optimally as possible and employ room treatment to get them as close as possible to their Spinorama data. Then use REW to correct any correctable room influence and EQ to my preferred target curve. All the heavy lifting done via REW and applied using MultiEQ-X. So, I have a definite starting point, end target and process. This is what I would like to understand with MB. There must be a target/goal and if the rooms signature is acoustically bad surely so will the end result be? Also if the initial NF speaker measurement is influenced by the room or objects the starting point for the MB process is off from the start? I am sure there are explanations for us A/V process driven folks not just 'it works, it's quick & sounds better'.

[@m4nc1n1]

This is pretty much REW in an app

[@joentell]

These are all valid questions. I think your current method of EQ using REW seems reasonable and thought out. If the room is acoustically bad, that really needs to be addressed via physical means of treatment. Side note: I have Dirac ART, and I still recommend physical room treatment. Bass can be EQ'ed to have a more ideal response, assuming you have the headroom and enough subs to fill any major dips. I think you already know that. The higher frequencies can also be made more ideal assuming you have a speaker with a good directivity index. The tricky part is transitioning between the two methods of EQ where the bass+room are inseparable, and the higher frequencies where we can hear an audible difference between on-axis direct sound that arrives first, vs off-axis reflected sounds that come later. Blending the two methods of EQ is what makes this different, since we can seamlessly transition between them based on the transition region in your room. The question is what target is your preference? Research has shown, the preferred target response at your MLP is likely whatever the response is if you were to put a speaker that measures flat anechoically, with a smooth off-axis response. That's why I mentioned that "speaker correction" is a better term for what we're doing rather than "room correction." We don't simply make the speaker flat nearfield though because that's ignoring the fact that the bass is EQ-able. Does that clarify things somewhat? If not, feel free to ask away.

[@marktebbutt8052]

Appreciate your time taken Joe to respond, thank you. Obviously until it's in hand and am able to test personally there will always be questions but will follow your MB journey with interest. Be very curious to see feedback from the beta testers especially the Reverend, (I have applied for Public testing). I do hope there will be the provision to trial MB before purchase? I know there will be great interest on the AVforums community here in the UK if MB does prove to be Magic!! I am an avid advocate, promoter and offer technical help of MultiEQ-X/REW on our UK forum so hopefully Magic Beans will also slot into the A/V tool box 🤞🏻😁.

[@joentell]

@@marktebbutt8052 reverendslim has it on his system now and early reports look promising. 😁 He was one of the first to try it. I chose him because I expect him to be critical of the app. I expect nothing less than the honest truth. I'll keep an eye out for your name. There are lots of sign-ups.

[@marktebbutt8052]

@@joentell cheers Joe 👍🏻👍🏻👍🏻.