What Does Audyssey REALLY Do? Testing the Denon X4700H!

[@joentell]

Buy the Denon X4700H bit.ly/DenonX4700H I asked who was interested in a similar video comparing DIRAC to Audyssey. Here it is! kzbin.info/www/bejne/oKa1p6FvgaeVg80

[@FURognar]

Fantastic video Joe. This is the kind of content I am looking for.

[@roccobruno8027]

Incredibly thorough ... thank you for the detailed work. I can now understand why many audio specialists recommend only using room correction up to 500Hz.

[@mhnoni]

What is the reason?

[@joentell]

Who is interested in seeing the same test with DIRAC Live on my NAD T778?

[@danielmiller469]

Yeah I definitely think it would be cool to be able to compare the direct live equalisation versus the Odyssey equalisation

[@keepingupwiththejones2933]

Yes please 🙏 😍

[@justplainlawr]

Yes absolutely. Shout out to Joe! This was a great video! Way to go peeling away some of the layers and mystery about what audyssey actually does! This video was great!

[@FURognar]

I am absolutely interested. Planning to build a dedicated theater in the future and need to decide between an Audyssey system or a Dirac system. Videos like these will help immensely.

[@jasonrobinson4274]

Joe, another test you should do if possible, using both the Denon and the NAD is the changes the upmixer has on the signal. I have already tested this on Denons and DSU is 2-3db louder than Neural X in the bass region and Auro3D is 10db louder, making Auromatic practically useless, especially if you already have house curves or using DEQ. Since Auromatic is mainly duplicating the channels and adding its own reverb when upmixing, the more channels you have, the louder the bass gets. NAD doesnt have Auro, but at least the other tests can also be performed. So use a L+R signal from REW and measure using the upmixers. SpareChange mentioned in one of his movie reviews that Neural X neutered his bass on his Trinnov, so he uses DSU, but he also doesnt really use Auromatic, so not sure if these anomalies also apply there.

[@HarryM2022]

Great video Joe as usual. At the beginning you showed the difference between before and after Audyssey and said it was quite similar. Do correct if I am mistaken but having the Umic-1 straight in front of the speaker you eliminate the room issues so not much to correct. It would be great to see the difference (I would expect to be substantial ) doing the measurement from the listening position.

[@joentell]

There was actually a lot to correct because of the reflections from nearby objects and placement near a wall affects the bass.

[@mladia]

@@joentell Yes. But what if you put the U-mik pointed straight up exactly where the Audyssey mic was?

[@joentell]

@@mladia I'm guessing it will be more similar. I may try to do more tests about how it time aligns and level matches. Also, I could see how well they align the phase between speakers and subs

[@BenjiGator23]

Great video! From what I've heard directly from Audyssey though (through some of Sound United webinars) is that MultEQ XT32 really needs to rely on the 8 mic positions in order for it to really shine. Even if the various mic positions are only a few inches away from the MLP. It would be interesting if you can do the same exact test but by using all 8 mic positions at the varying positions (lets say 6 to 12 inch radius from the MLP) instead of the 3 mic positions in a stationary position, and by leaving the UMIK-1 just at the MLP.

[@rynev3392]

Great technical video, I’ve been curious about this for a while now…… I just realized that I wasn’t subscribed, but I changed that, keep up the good work

[@MRPC5]

Great info! One thing I think I've noticed is that when you limit the eq range in the app, it doesn't limit the range of dynamic eq or the various sound modes, and all of them seem to bump up the treble or suck out some mids, so there's no way to stop Audyssey/the receiver from making content brighter than intended. Unless I'm mistaken which is definitely possible.

[@joentell]

I didn't notice that, but I did notice that adjusting the bass in the tone controls doesn't adjust the bass evenly below 1khz where it starts. For example, if you have a sub, you would expect it to affect the main speakers and the sub equally. It doesn't. It only applied to the main speakers which threw off the balance with the sub. Interesting stuff.

[@C_O_N_C_E_P_T]

You should do a video on what dynamic volume does also. With it on and dynamic eq being off.

[@MrBrandchr]

Hi @Joe, did you find the reason for the frequency increase in the high midrange?+ I can see it as well in my measurements, for some reason Audyssey totally overcompensates something in this area. If i cut off Audyssey below (e.g. 500hz) everything sounds well.

[@missing1102]

Yes. This measurement is accurate. I think I watched a video on audioholics that said to adjust the frequency correction to under 500hz. That was a good noticeable improvement for me but I still don't like the way Audyssey sounds with music. I may pay somebody to change the target curve on my phone if I reinstall the Denon I have.

[@jrep88]

Good video. I recently purchased a SR 7015. I was going to do the same measurements myself. Looks like the curve editor does what it is suppose to.

[@joentell]

It seems to for the most part.

[@jrep88]

@@joentell Definitely better than the old school EQ bands

[@maximiliangockel]

This is GOLD, thanks Joe!

[@Fr0stkeule]

Does that mean Audyssey is bad at measuring? I need a new device either between 8805A or an arcam mv 41. I would only take the arcam mv 41 because of dirac. What do you think of MultEQ-X is the an alternative to dirac.

[@joentell]

I offer a service for manual calibration for MultEQ-X, so I'm biased. I do believe I can get superior results with MultEQ-X, but it's a tedius process. Dirac is good if you just want a good, quick auto-calibration. Here's the link to my services if you're interested. joentell.com/booking

[@wktek48]

Great review, thinking of buying the audyssey multi eq x app with the denon mic for my denon 3500h.

[@m4sta2k]

Reason there is no sound on flat settings is a bug. You need to first save the settings file and then go back and transfer it over by going back to main menu after you have done that you will then have sound on flat settings

[@joentell]

Oh I see. I eventually figured it out, but just messing around, but it's good to know the specific thing that needs to be done to fix it. Thank you!

[@Antimonkat]

Only thing that kind of confuses me about this video is, room corrections job is to take the room out of the equation and compensate for its effects for a more expected response, but with your mic positions, you've kind of already done that, so yeah this helps us visualize what the different curves look like, and what midrange compensation and the roll offs change, but doesn't really show what, or more importantly, how effective audyssey is at doing its main function, compensating for the room. I love the idea, I just wish you did a real world measurement, maybe as a follow up as you now have the near field measurements as an "ideal" now do the seating position before and after audyssey and see how different it is from near field before (how much and in what way the room is affecting the response), and then see how close it can get you back to the near field measurement with compensation.

[@joentell]

I know what you're saying and it makes sense to use it in a more practical situation to see if it can compensate for room interactions by comparing the correction in the listening position to the nearfield measurement. Maybe I could do it in a future video because I also wanted to look at how they level match and time/phase align with subs. It does take a while to make these. If you were willing or could do the test, I would love to watch it.

[@Antimonkat]

@@joentell yeah, for what it is worth, i do plan on doing a KZbin series on the build out of my theater including things like audio measurments with REW before and after acoustic treatments, it's not going to be great with just a gopro and internal audio, but I want it documented for myself and friends and family anyway, and if i get a few extra views on th side, cool.

[@joentell]

@@Antimonkat very cool!

[@Mike_v_E]

what is the difference between the pink noise measurement and a frequency sweep measurement? And when do you use which?

[@peaceandrelaxationwithgodscrea]

Excellent video well thought out and informative

[@mladia]

So if I use the the app Curve Editor I should use Reference because Flat is not going to take the changes? Does this affect Bass EQ I did with MiniDSP 2x4HD?

[@joentell]

Yes, only Reference reflects the changes. BEQ is something done externally and Audyssey has no idea what the MiniDSP is doing.

[@mladia]

@@joentell Thank you. Your tests are invaluable for us. Do you know what smoothing Audyssey uses in it's graphs? It looks like it is 1/12 smoothing right?

[@Flatsteff]

Hey :) Cool Video, but is this a good way to do it? Audysseys job is "to remove the Room and the Problems, that the Room brings into your Setup" . If you set up all mics so close, because you " don´t want to measure the room", it cant work , like it should, or am i so wrong?

[@joentell]

The idea is too see what it does. I'm pretty familiar with the inner workings of Audyssey and get to talk with the people at Audyssey to confirm. This is not a perfect test, but good enough to get an idea.

[@bommareddybrijendrareddy9301]

Does the AVR sound better in an acoustically treated home theater room without Audyssey calibration sir?

[@UnitedRedDevil13]

If I’m using Klipsch rp600m speakers as my main R L, should i also cap the MultEQ frequency range to 500hz on those as well? Thanks!

[@UnitedRedDevil13]

Recommend flat over reference? Thanks in advance!

[@joentell]

Yes only because I don't like what Auto-EQ does above 500hz.with proper manual EQ, it is possible to correct full range using the MultEQ-X PC app. I recommend flat if you plan on using the Audyssey Editor mobile app as that is the profile that allows you to use a custom target curve.l

[@UnitedRedDevil13]

@@joentell thanks a lot. I’m super new to all this so learning as i go. Using a denon s760h and i did buy the app. Would you also recommend turning off all midrange compensation? So far i have dynamic volume off but dynamic eq on. Set it to FLAT. thanks again!

[@joentell]

@@UnitedRedDevil13 kzbin.info/www/bejne/f4CoqZ2Flp6qbK8

[@Sky1]

I spent hours getting my bass just right. Then I ran Odyssey and it completely nulled out 30 hertz. So what do I do now?

[@JakubSvoboda23]

Very nice job in doing this video, as always! I have X2600H DVB...

[@adamjj85]

Great testing Joe! I do not get the same weirdness at 2-4k with Audyssey Reference. My reference and flat comparison only has a slight change in roll off above 6k as expected.

[@Macroproxy]

Do they both roll off more than your speaker does on its own? Is "Roll off 1 and 2" in the app the same as "reference" and "flat" in the AVR settings? I'm trying to figure out how to use EQ goal to keep it flat through 20k, at least on axis or at the MLP, then I can just boost bass a bit with sub levels and EQ on that end.. I thought the "flat" setting would be actual flat as possible, not just a different roll off level.

[@adamjj85]

@@Macroproxy Everyone's speakers are different, but flat is the closest you'll get to true flat and does NOT roll off more than my speakers in my case. Reference definitely has a steeper roll off of about 3 db or so. I did not test Roll off 1 and 2 but I use 1 as it looks more natural.

[@adamjj85]

@@Macroproxy Also, if you don't want to touch your speakers high end, then you can use a curtain with the app and exclude above the frequency of your choice.

[@Macroproxy]

@@adamjj85 Yeah, I was doing this before and left it only under 500Hz. At the time there weren't any peaks or dips anywhere at least in the front LR as I didn't test all of them with REW. I'm pretty far off axis to my Atmos/height speakers, so it would probably be nice to boost those in the high end, but I suppose that would have to be done manually... If even a good idea.

[@joentell]

Yes, I typically recommend limiting Audyssey to 500hz or wherever the transition frequency is in your room. Problem is it doesn't make corrections that you may want above that. Using DIRAC, I let it run full-range as I feel it does a better job overall.

[@TylerStout]

Really interesting video! A lot of the same stuff I've found plus some stuff I hadn't thought about or tried. I would love to see this for more room correction suites. Maybe you can get the Yamaha guy you had on the daily hifi to send you a A6A which would be pretty similar to the Denon in terms of specs. I would say the better way to do this would be to get a laser tape measure and place the mic a set distance away (like 1m) and swap the mic out along with using the straight up cal for the umik. I'd wager to bet that the rise in response is due to a reflection off of the Audyssey mic/stand.

[@joentell]

The smoothness of the rise and the width of it tells me it's more of a microphone calibration thing. Reflections are typically blips, but you could still be correct. I'm only guessing.

[@TylerStout]

@@joentell Sorry I wasn't clear was talking about the big blip around 2500hz. Agree the with you about the other stuff being a calibration thing. I keep meaning to import my miniDSP cal and the CSA cal files into excel and compare them as It would be interesting to see how off the UMIK is from the factory cal. I know the CSA cal goes higher and lower than the miniDSP one.

[@joentell]

@@TylerStout Oh yeah, let me know! I'm interested

[@TylerStout]

@@joentell Hey so I have them in a Google Sheet. Not sure if I can post links here but I can send you the sheet or upload to the charts to imgur and send you them that way. I couldn't get the charts labeled exactly how I'd want but they are in logarithmic scaling. Honestly there's more of a difference than I was expecting. For the 90 deg cal there's about a 4-5db difference below 100hz and about a 1-2db difference 400hz and 8000hz and then a similar story for 0 deg but some extra wonkiness at around 9khz.

[@juanmillaruelo7647]

Great stuff, Joe! :-) Thanks for sharing your research work with us. A doubt cropped up: Is the target chosen for room compensation the same for stereo music and for home theater? Does it matter? (HT: plenty of dialogue that shouldn't be drowned out, etc)

[@joentell]

Yes same target curve. I've heard people say they recommend Reference for movies and Flat for music.

[@stevenmichael7770]

so wait - when you turn off the midrange compensation it ADDS the dip? I thought taking it off would result in a flat response and turning it on would cause the dip. Either they're backwards, Joe's backwards, or I'm backwards!

[@joentell]

No, I might have misspoken. Turning on midrange comp, adds the dip.

[@JuanJose-wt5yj]

Thanks for the video. Do you have tests with the Audyssey mic on listening point vs Umik-1 with REW ?. I have had 3 Audyssey microphones and all 3 had treble drop, one began to drop between 5-6 Khz, another between 7-8 Khz and another between 9-10 Khz. I say this because when passing Audyssey in our listening point, if by default it marks a drop in highs that is not real (checked with Umik-1 + REW), then Audyssey will apply an equalization in those highs that they do not really need. I had a Denon X4000 and sold it because of Audyssey's poor work on high frequencies. Since it applied an equalization that my speakers did not need. Now with the app at least we can tell it to equalize up to the frequencies we want, not before. I hope I have explained myself well Written with Google translate Greetings

[@joentell]

I understood. I didn't do testing with it in my listening position.

[@melberg9098]

thx for this vid Joe

[@joentell]

You're welcome!

[@brydon10]

Great one Joe.

[@Steve30x]

I need something like that for the DENON DHT-S216 soundbar I got because for some very strange reason Denon thought it's a great idea to not allow you to reduce the bass from the "subwoofers" as Denon calls them 😂 enough. Also it has a subwoofer out and I connected it to my YAMAHA NS SW300 with the pure input and the soundbar outputs a whole lot of midrange as well as bass from the subwoofer out. So I connected it to the inputs that I can adjust the crossover frequency. Is there any sort of mini audio equaliser I can buy to adjust the crossover frequency going to my subwoofer?

[@joentell]

MiniDSP 2x4 or the OSD iDSP I reviewed uses Parametric EQ which is what you want. You can also apply crossovers. kzbin.info/www/bejne/ZoTbk6SMpc6apcU

[@Steve30x]

@@joentell thank you. I'll check it out.

[@Steve30x]

@@joentell I just looked for it and it's neither on Amazon or the OSD website anymore

[@joentell]

@@Steve30x bit.ly/OSD-iDSP (affiliate link)

[@Steve30x]

@@joentell I'll have to try this on my computer because I keep getting redirected to Amazon UK with the link. I'm not sure I can buy it here in Ireland. I searched for it on amazon.com and found it. Its not on Amazon UK or Amazon Germany. I would have to import it to ireland. that would add 33% to the price because I would have to add 23% VAT and 10% Import duties. But it might be worth it

[@H2OFSU]

Joe, could you show differences it makes with the off, light, medium, heavy settings for Dynamic Volume?

[@joentell]

Those are compressors, so they wouldn't show much on a frequency response chart. Those would be more obvious if you looked at the waveform of the sound coming out. You would see less variation between loud and soft volume. Essentially, it's a dynamic range compressor.

[@H2OFSU]

@@joentell Do you know how “cinema eq” effects the treble? It says softens, guessing it rolls off faster?

[@joentell]

@@H2OFSU I maybe can measure that in a future video. In stereo mode, I wasn't able to get the option to show up.

[@pb24dagrk]

@H2OFSU CinemaEQ applies a about a -4dB treble at 20khz, starting from 1khz acting as the hinge. Think of a straight line from 1khz to -4dB at 20khz.

[@rv112xy]

Great video!

[@ddcrocky3831]

When running Audyssey Deos that change the warmth of marantz sound to like denon lets say ect

[@glengatt3941]

I don't understand this should I set it with odyssey?

[@Sky1]

I ran Odyssey today it cleaned up things quite a bit from the last time I ran it and I ran it twice the last time and both times it came out kind of harsh sounding sounding to my ear. Go figure so I was really interested in this video to see if Odyssey really works. Sorry about the spelling I'm doing voice to text.

[@joentell]

I think it's best to turn down the treble a bit. I will probably make a video soon a out how to get Audyssey to sound more similar to Dirac.

[@erod9088]

Most measurements and room correction above 300 Hz isn't audible to the human ear in the same fashion as the measurement adjustments. If you have decent speakers, you don't need any correction above the transition. You won't hear it anyway, other than a potentially deader room. The brain filters the bad stuff out at higher frequencies.

[@joentell]

Changes in frequency response above 300hz are definitely audible. Test it easily yourself. Change the treble response on your AVR which starts at 1khz. You can hear the changes.

[@CalJennings]

The real test is "How does it sound to you?"

[@bostonquad2068]

I was thinking the same thing with the computer on top of aVR

[@joelopez7459]

Subs were a must, at least towers

[@tomcavanaugh5237]

It sets the speakers to large, virtually turns off the subwoofer, and sets all the other speakers in strange lopsided ways. Or, at least that's what it did for me.

[@wric01]

No need for that 35 bucks mic. It end with same measurement yards in settings with measurement tape as audyessy.

[@blueflame0003]

I have x8500h and I didn't use audessey

[@FURognar]

Why not? With the multeq app you could get some great results!

[@danielmiller469]

Yeah I’ve learned something today Odyssey sucks

[@FURognar]

It only sucks if you dont know what to do with it. Compared to Dirac or Trinnov optimizer, yeah it sucks compared to those. But compared to the other calibration like YPAO, MCACC, AccuEQ and whatever Sony uses, Audyssey is better than all those.

[@wa2368]

@@FURognar YPAO is better

[@FURognar]

@@wa2368 you might like the sound of YPAO better. That doesnt mean its better at correcting for the problems introduced by room modes better than Audyssey can. From what I understand, XT32 is a better profile than YPAO. But YPAO's default curve is more musical.

[@wa2368]

@@FURognar Fix room issues with proper placement and physical room treatments first. No software offers any magic fix. Do that before you even start this discussion.

[@FURognar]

@@wa2368 of course. For a dedicated space, room treatments are a must. But not always possible for shared spaces like a living room

[@mladia]

So can you summarize it for us in the cheap seats Joe? I only have MultiEQ (s740h). I ran Audyssey.... Used ratbuddyssey . Now I know I should use Reference NOT Flat. I should probably not turn on Dynamic EQ.... and limit MultiEQ Filter Frequency to Schroeder. I Eq'd the subs with REW and MiniDsp 2x4 HD. Did I miss anything?

[@joentell]

I think you got it! Maybe just add a target curve you like. Check out the Harman target or NAD target curve. Both use similar philosophies. Don't use tone control if you use a sub since it only seems to affect the main speakers and not the sub which will throw off the balance.