Dirac > Audyssey XT32 ... Sure, Always?

[Miker 1102]

My only real complaint with Audyssey XT32 is the default reference and flat curves. If you simply run Audyssey in a Next/Next "wizard" like function, it doesn't sound good. Audisssey's belief appears to be flat frequency response is the desired goal. This belief is flawed as people prefer a downward sloping curve. It basically sounds both bright and no base at the same time. If they would add a better downward sloping default curve (like a harmon type curve) it would make things much easier. My favorite instructions for getting Audyssey to produce good results is via this" https://docs.google.com/file/d/1YH_eNHRCxKFCwXMddi28kzXqnUwYHfrD/edit?filetype=msword

Al least I have had good results with it.

Thank you. My first big break was limiting the frequency range. Wow that gave my speakers a chance to work. So should I pick up ratbuddy or hit the new windows app? I mean a lot time audio guy with many years has said that I should be able to adjust my curves on the app..I cannot my fingers are to big?

 

[Miker 1102]

Audyssey XT32 is horrible. It helps with room modes, but in the process screws up the timber of your system. In Harman blind listening test, it was described as "thin", "harsh", "bright" and preferred less than no room correction. This is consistent with my own listening. You can use the app to adjust the timber, but it's a pain in the ass to align properly.

https://www.audiosciencereview.com/...s/target-room-response-and-cinema-x-curve.10/

I lean toward this experience. Recent curve changes and limiting the apps frequency range helped a lot.

 

[Chromatischism]

Best example I have read. I changed my curve, Limited the frequency. Audessy at base line sounds awful..to the point where I question Denon enthusiasts. They also have poor instructions.

Audyssey actually implements a roll-off, but it's canned and doesn't always fit your speakers or room.

 

[peng]

I lean toward this experience. Recent curve changes and limiting the apps frequency range helped a lot.

I know the Audyssey flat curve has been blamed for a long time since Dr. Toole (and Dr. Olive too I think) mentioned that in their studies that most people prefer a tilt/smooth boosts to the low to low frequencies. I have to wonder though, that in their studies, did they allow for the well know Fletcher Munson curve so that they would have conducted their tests at reference level? If not, then I would argue that Audyssey's approach may not be all that wrong, because Audyssey did make it clear that they recommend the use of DEQ along with the reference or flat curve, when listening at below reference level.

Even before I tried putting a tilt on the bass, I never felt the reference curve sounded thin and bright because I always have DEQ on, unless I cranked the volume up to at least -15. Regardless, we have the App now, it isn't hard to implement a tilted target curve, though it needs to be done carefully, not just use the finger to draw the tilt. For it to really work well, I would use REW, do my analysis of the actual FR curve and then apply cuts to the bumps first, then elevate the subs level globally, along with very minor boosts at selected frequencies, and then check with REW again. In doing so, I would sacrifice efficiency but my subs are oversized for my room anyway so its not a problem.

To me that would be the same with Dirac or Anthem, if someone just tweak by shaping the target curve to implement Harman's, it would a hit and miss exercise and all I can say then is, good luck... That would mean people may like the end results because they just luck out, or they "heard" what they expected to hear, or a combination of both..

 

[Sal1950]

Audessy at base line sounds awful.

A gross exaggeration.

 

[Miker 1102]

A gross exaggeration.

Sound is subjective but the Audyssey curve sounds bad to me. Out of the six avrs here, It requires the most work but at least it does work unline YAPO. It also requires a ton of tweaking. I have multiple pdfs for the extra 20 dollar app. I watched almost two hours on the new app for windows which is 200. I don't find it an exaggeration at al.

 

[Chromatischism]

In my experience, it depends on the speakers. I've had speakers that sounded a lot better with it, and I've had speakers that sounded worse with it. That's when I benefitted from the filter frequency cutoff so I get all of the bass EQ that it's so good at, but nothing above that when it doesn't work for my speakers.

 

[Miker 1102]

I know the Audyssey flat curve has been blamed for a long time since Dr. Toole (and Dr. Olive too I think) mentioned that in their studies that most people prefer a tilt/smooth boosts to the low to low frequencies. I have to wonder though, that in their studies, did they allow for the well know Fletcher Munson curve so that they would have conducted their tests at reference level? If not, then I would argue that Audyssey's approach may not be all that wrong, because Audyssey did make it clear that they recommend the use of DEQ along with the reference or flat curve, when listening at below reference level.

Even before I tried putting a tilt on the bass, I never felt the reference curve sounded thin and bright because I always have DEQ on, unless I cranked the volume up to at least -15. Regardless, we have the App now, it isn't hard to implement a tilted target curve, though it needs to be done carefully, not just use the finger to draw the tilt. For it to really work well, I would use REW, do my analysis of the actual FR curve and then apply cuts to the bumps first, then elevate the subs level globally, along with very minor boosts at selected frequencies, and then check with REW again. In doing so, I would sacrifice efficiency but my subs are oversized for my room anyway so its not a problem.

To me that would be the same with Dirac or Anthem, if someone just tweak by shaping the target curve to implement Harman's, it would a hit and miss exercise and all I can say then is, good luck... That would mean people may like the end results because they just luck out, or they "heard" what they expected to hear, or a combination of both..

I have kept three measurements with app. I read that ratbuddy let's you load them. I don't understand

In my experience, it depends on the speakers. I've had speakers that sounded a lot better with it, and I've had speakers that sounded worse with it. That's when I benefitted from the filter frequency cutoff so I get all of the bass EQ that it's so good at, but nothing above that when it doesn't work for my speakers.

Yes. I found the biggest benefit was the frequency cut off. I cannot adjust the curve right with my fingers to save my life.

 

[peng]

I have kept three measurements with app. I read that ratbuddy let's you load them. I don't understand

Yes. I found the biggest benefit was the frequency cut off. I cannot adjust the curve right with my fingers to save my life.

Not as far as I know unless there is a newer version. Ratbuddyssey (again as far as I know) is just an user interface that you can use to open the Ady file the MultEQ App created and enter the data for editing the target curve into a table; and when you are done you can save the file. Then you can put that file back to the App, may be that's part some users call it "load", export, or copy, I don't know what the right term is. You can see plenty of examples, screenshot etc. on AVSF.

https://www.avsforum.com/threads/announcing-ratbuddyssey-a-tool-for-tweaking-audyssey-multeq-app-files.3006886/

Some people may go overboard and enter numerous data points to try and get a straight flat line 10-20,000 Hz (just kidding..). If you want to get within say +/- 2.5 dB, 1/12 smoothing 20 to 150 Hz, I would say 10 to 20 data points should do it depending on the speakers, subs, and room. After that you obviously should use REW to check not just the mlp, but a few other positions (I did up to 10) as it may improve the mlp but mess up in other positions within you listening bubble, though I have never seen that in my case though I would think that part would be sort of hit and miss by nature, being that it is a manual tweak based on the results of just one position. I would think it will the same if you tweak the target curve with DL or AARC.

 

[Dj7675]

A gross exaggeration.

I could see it sounding awful... someone selects flat, does not run DEQ and/or does not increase their sub level. It can sound very good too with a little know how.

 

[Dj7675]

Thank you. My first big break was limiting the frequency range. Wow that gave my speakers a chance to work. So should I pick up ratbuddy or hit the new windows app? I mean a lot time audio guy with many years has said that I should be able to adjust my curves on the app..I cannot my fingers are to big?

You can certainly give Ratbuddyssey a try. It takes a little bit to get the hang of it but after that, it shouldn’t be an issue. Whether it is worth the time depends on a couple of things..
-Do you feel a need to try to get the bass response smoother?
-Do you want to EQ full range. Like many I have been doing this a while and have used various full range room curves. Lately I have been only using automated EQ under 3-500hz, and doing manual EQ higher to fix any speaker issues. I may be done with full range automated eq. Trying to get the right room curve, with the right amount of slope, for a give set of speakers and room has been difficult to sound good to me. I see that perfectly curve on post measurement (Dirac or Audyssey) and assume it must sound good. I like it for a short while, and then I go fiddle with another full range curve.
The new Audyssey windows app seems to have a lot of functionality and the bonus is it should be much more simple to use than the app.
In the end, I think the app probably does the most important 2 things...
-You can cut off the frequency it corrects and you can do a curve for the bass. You may or may not like correcting higher with the app, or with the new windows version.
I think one of the most interesting threads on ASR is the compilation of comments by @Floyd Toole regarding full range correction via room curves. What I find most ironic is one of the room curves in the AVSForm Ratbuddyssey thread is called the Toole curve... Anyway the compilation of comments from a variety of threads on AVSForum is quite interesting on if we should be using a full range room curve for calibration at all..

Some comments from Floyd Toole about room curve targets, room EQ and more

Room curve targets Every so often it is good to review what we know about room curves, targets, etc. Almost 50 years of double-blind listening tests have shown persuasively that listeners like loudspeakers with flat, smooth, anechoic on-axis and listening-window frequency responses. Those with...

www.audiosciencereview.com

 

[raistlin65]

Best example I have read. I changed my curve, Limited the frequency. Audessy at base line sounds awful..to the point where I question Denon enthusiasts. They also have poor instructions.

Your speakers, your room, and your speakers placement within your room all impact Audyssey's ability to provide room correction.

This is why people are objecting to your sweeping generalizations.

 

[Dj7675]

I have kept three measurements with app. I read that ratbuddy let's you load them. I don't understand

Yes. I found the biggest benefit was the frequency cut off. I cannot adjust the curve right with my fingers to save my life.

My best sounding results with Audyssey were very simple:
1-Limit correction to 3-500hz
2-On all speakers, for the room curve put a point at 200 on the 0 line, and a second point at 20hz (something between +3 and +6)
3-On your receiver, post calibration after sending the curve back to the receiver, add the same amount of boost to the sub(s) you put in your room curve

 

[Chromatischism]

My best sounding results with Audyssey were very simple:
1-Limit correction to 3-500hz
2-On all speakers, for the room curve put a point at 200 on the 0 line, and a second point at 20hz (something between +3 and +6)
3-On your receiver, post calibration after sending the curve back to the receiver, add the same amount of boost to the sub(s) you put in your room curve

Question: why is the sub level increase needed if you adjusted the curve?

 

[Dj7675]

Question: why is the sub level increase needed if you adjusted the curve?

Because of how Audyssey correct the sub. See attached. See the curve created vs the corrected response. If you don't compensate for this, the sub correctly slope will be correct, but the level won't. The following was taken from the link I posted in post 5 of this thread. It is a good read. It is not mine but a nice compilation of things that helped me understand Audyssey a bit better to make it sound good. Member @Jon AA I believe may have originally posted it on avsforum or audioholics (can't recall). Also it is probably best to confirm the total amount of boost needed by measuring with REW.
https://docs.google.com/file/d/1YH_eNHRCxKFCwXMddi28kzXqnUwYHfrD/edit?filetype=msword

 

[GalZohar]

I don't understand how you get better results by editing the curve to account for peaks that weren't originally handled.
If a peak wasn't handled due to spatial averaging - Then you should probably just adjust your measurement positions.
If a peak wasn't handled because Audyssey can't flatten it or thinks the filter to flatten it would cause more harm than good, how does introducing a dip to the target curve helps in any way? Supposedly it should still avoid reducing that dip for the same reason and not make anything useful from it. That is, unless your edit works around some unexplained Audyssey flaw which I've never seen shown anywhere.

 

[peng]

I don't understand how you get better results by editing the curve to account for peaks that weren't originally handled.
If a peak wasn't handled due to spatial averaging - Then you should probably just adjust your measurement positions.
If a peak wasn't handled because Audyssey can't flatten it or thinks the filter to flatten it would cause more harm than good, how does introducing a dip to the target curve helps in any way? Supposedly it should still avoid reducing that dip for the same reason and not make anything useful from it. That is, unless your edit works around some unexplained Audyssey flaw which I've never seen shown anywhere.

You can call it "flaw", but can also call it imperfections or something else. As I mentioned before an amplifier may be design to have 0 % distortions, but if you measure even the AHB2 you will see that it has distortions at any output level. The distortions measured were low because of the feedback applied, if not, distortions would have been much higher

I think the same apply to REQ, even the best ones such as Dirac's latest greatest version will not get you a flat line even if the target curve is. The target, flat or not is just the target. When you measure the actual results you will still see dips and bumps, but some can do a better job than others for sure, and that would also depend on other factors.

 

[GalZohar]

That doesn't explain how adding a dip to the target curve helps Audyssey better correct the peak it failed to correct earlier. Unless that dip isn't fixed due to spatial averaging (that is, it doesn't consider the dip as it's only in 1 place). If we just measure in the same spot, can you still get any flatter response by setting a non-flat curve? How does that actually work?

 

[NirreFirre]

I was looking at Dirac (in the new Onkyo RZ50 review thread) and my old university's department where they started the research 10+ years ago and one thing led to another.. I don't know if this thread or another has discussed the upcoming features of "multi speaker active correction" but I found some papers from 2014-15 (and references back to 2012!) and also 1-2 Dirac patents from those years that are pretty detailed and very interesting reads so I thought if sharing them. It's mind blowing how long time it takes for someone to implement, design, manufacture and sell these technologies into AV.. but I guess Dirac's partners in other the car industries have had it for years

Personal multichannel audio precompensation controller design​

Audio precompensation controller design using a variable set of support loudspeakers

http://www.signal.uu.se/Publications/abstracts/a141.html (Adrian's PhD thesis)

 

[Sal1950]

I cannot adjust the curve right with my fingers to save my life.

Use a android emulator like Bluestacks to run the software on your PC.
I can use either my 32in computer monitor or my 75in Sony TV.
Tweaking the curves then is a cakewalk.