Onkyo TX-RZ810 in the house

[Tony V.]

Today being "Boxing Day" up in Canada (this tends to be the best day to go shopping for the best deals in A/V here) I came across an Onkyo RZ810 for $900 off the regular price at Bestbuy. My wife by my side says to me "seeing as you have worked so hard on renovating the basement storage/laundry room, you can buy this for the Theater room" on top of that she says "let's buy some Home theater seating while we are at it" Of course I asked if she was sure but long story short we have a new receiver and seats for 7 (3 in the front row and 4 in the back) pictures to follow once they are delivered.

NOTE: Update, I ended up returning this unit and upgrading to a RZ 920 as the 810 only had the option of processing either 7.2 channels or 5.2 with 2 Atmos channels. Where the 920 has processing for 9.2 channels.

The 810 has everything I need and clames to have 130watts per ch although I'm scepticle of that but it's not small and has a nice size tordodal power supply up front and has compleatly separate analog amplification and digital processing blocks as well as solid copper bus bars to limit interference. Non-phase-shift amp circuits shift the frequency roll-off point higher than the listening level, and combined with Onkyo new AccuReflex technology that works alongside its proprietary auto EQ room correction.

While on the topic of auto room EQ I've been thinking a lot and there is some aknowlagment in the industry that EQ above 400hz if applied unevenly between left and right channels can cause the image to collapse and I've exsperimented with it and found this to be true. Onkyo has gone this route and may be turning heads away but I'm in the club that it may be a good thing to leave the front soundstage alone or at the very least do it externally using REW as a tool and outboard EQ and amps as I already do.

Anyhow Atmos here I come...now just to redo the ceiling of the theater room and install some speakers.

 

[Sonnie]

Awesome deal Tony... congrats!

I would not do without EQ above 400Hz for anything in the world. Read our recent speaker evaluation. It's actually the opposite... if you don't equalize it, there is a much better chance of the frequency responses varying between speakers, which is when your image will be skewed. Frequency matching speakers is critical to imaging, sound stage and depth acuity.

 

[Tony V.]

Totally agree that EQ above is needed but care and attention needs to be taken and doing it with REW and manually inputting the results should achieve the same results if not better don't you think?

 

[Sonnie]

It just depends on the platform... Dirac Live does a wonderful job, and it can be verified with REW. Audyssey is okay, and much better than the Onkyo Auto-EQ, YPAO, MCACC, etc... but we've found it still isn't as good as Dirac Live. DL corrects frequency response and impulse response... so it's more than just equalization.

 

[Tony V.]

Sadly a unit with DL was out of my price range and I'm going to give this a chance. My room and speakers are quite easy to work with so hopefully I can get near perfect results.

 

[Sonnie]

Yeah... I think you'll be fine. That's a really nice unit and a super good deal.

 

[Tony V.]

So after some digging (I should have looked into this farther before buying) I ended up returning this unit and upgrading to a RZ 920 as the 810 only had the option of processing either 7.2 channels or 5.2 with 2 Atmos channels. Where the 920 has processing for 9.2 channels built in. so now the upgrade begins. Home theater seats arrive on Friday Jan 5th and Ive already removed the old sofa and love seat that was in the theater.

The RZ920 also boasts full 9 channels of class D amplification so reaching the true 135watts per channel at 8ohms should be no issue. I will still run my mains off an external amp just because I can

 

[JBrax]

Congrats on the new gear Tony! You’re about to head down that upgrade rabbit hole. Atmos is a really nice upgrade and you’ll enjoy it.

 

[Matthew J Poes]

Sounds like a fun time Tony!

Tony I hope to have an article that goes over EQ above FS soon, but I'm taking too long to write it. I keep wanting to say things that aren't really true and needing to correct myself.

In any case, there is an argument to be made that EQ above FS should never be done with minimum phase EQ filters (i.e. PEQ). John talks about doing this only vary carefully, but unfortunately REW doesn't let you do this the way I think of being careful.

The right way to EQ above FS is to measure the speaker free space (outdoors or using gating) and applying a low Q (broad) EQ to correct general shape problems. The measurements need to be polar, and you want to correct things that are in all the axis's and not something that only appears in certain measurements.

Dirac is different and you can't recreate what Dirac does, so trying to compare is really difficult. Dirac uses FIR filters which are not minimum phase. As such it can correct for both amplitude and phase anomalies separately. It also uses all of the measurements averaged in such a way as to minimize the variance across measurements. I'm actually working on doing these calculations for my paper as we speak. It is possible to do this, but it isn't common. You want to EQ the response of the speakers such that across the measurements the normalized mean equals 0 at every frequency. Dirac effectively does this with its approach, but doing this manually is difficult, and doing so would likely cause huge phase problems. If not done symmetrically between the speakers the phase response of each speaker may also become mismatched which is audible (this isn't absolute phase but rather speaker relative phase).

If you do apply EQ above FS I would do so carefully. FS is actually probably below 150hz in your room. Between 150hz and about 1khz I find you really need to take a good number of measurements and EQ so the average variance is 0. Then above 1khz I normally would only apply general shaping of the response based on the free space measurements or general house curve.

 

[Tony V.]

So do you recommend using a wide Q for frequency adjustment rather than a narrow one? My Yamaha YDP2006 EQs are very flexible for not only adjustment for narrow to wide Q but also I can pick any frequency with a maximum of 6 filters a channel and one adjustable highpass and one lowpass filter. I can link both channels together so that any adjustment I do to one happens evenly to the other.

 

[Matthew J Poes]

So do you recommend using a wide Q for frequency adjustment rather than a narrow one? My Yamaha YDP2006 EQs are very flexible for not only adjustment for narrow to wide Q but also I can pick any frequency with a maximum of 6 filters a channel and one adjustable highpass and one lowpass filter. I can link both channels together so that any adjustment I do to one happens evenly to the other.

Yes wide is generally better. We don’t hear narrow band anomalies anyway, so even though it might not measure as well, it should be fine.

Earl Geddes wrote a paper in JAES on this topic and I believe it’s covered in one or two of his books. If I find the chapters I’ll share. His view (which is shared by Dr Toole as well) is that you shouldn’t try to correct the response of the room above Fs which is the Schroeder frequency. That the speaker itself should be flat on axis and have a smooth and decaying off axis response. If the speaker doesn’t have a smooth on and off axis response they both consider them flawed speakers. That’s where people like me need to come up with options for people who don’t own JBL M2’s or Geddes Sumas. John seems to share my view that correcting a speakers uneven response is ok as long as you correct the speaker and not the room. That means focus on the major anomalies and not on the little reflection related cancelations. That means using fairly broad eq to correct the general shape of the response to be more flat and not try to correct room or speaker deficiencies caused by diffraction or reflections.

Since I know Flavio sometimes reads my posts I’ll say for his sake, this is the rule that I apply to minimum phase eq. This is not the rule I think is universal. It is my understanding that both Floyd Toole and Earl Geddes would disagree with me and argue that even FIR based eq systems applied to minimize position to position variance is bad. However my own experience is that Dirac is different than other correction systems I have experience with and it’s very fine grained correction does generally sound better in an obvious manner. That it’s algorithm and use of FIr filters above Fs do no harm.

 

[Tony V.]

Thank you Matt, very insightful.
Thankfully my room measures quite well and my EVs are really good sounding speakers to begin with. (Remember, these are studio monitors) so they are designed with a flat response. I know when they were bought new in 1985 they cost $12k