Struggle replacing Old AVR - tried RZ50, 4800h, and Cinema 70 - what now?

[Steve Dallas]

Here are the new ones I posted yesterday. Thanks for offering to look at them. I'm curious to see what they sound like on yours if there's anyway to use them.

New EQ stands for modified curve
Pure is Pure Direct

The Marantz Pure Direct are weird as the video worked which it's not supposed to - it went away, then came back on. I should have switched to Direct but I didn't think of it at the time.

Marantz 8002 vs Denon 4800h measurements.zip

drive.google.com

Thanks. Before I dedicate any time to this... Are you running the Denon and Marantz standalone, or is the Denon plugged into the Marantz or some such configuration?

One thing that is odd about your pure direct trace is the lack of room-induced nulls. There are typically 2 significant nulls in any given room.

 

[techsamurai]

Thanks. Before I dedicate any time to this... Are you running the Denon and Marantz standalone, or is the Denon plugged into the Marantz or some such configuration?

One thing that is odd about your pure direct trace is the lack of room-induced nulls. There are typically 2 significant nulls in any given room.

The Marantz is running standalone and I have measurements but with a different mic position (close but not the same). But we can't fix the Marantz as you get what you get from its calibration.

The Denon has 6 pre-outs to the Marantz so it's doing no amplification at all - it's in pre-out mode.

My room has a cathedral ceiling with some hallways attached to the main room and the speakers are towards the far side of the cathedral ceiling so they can be seen from the kitchen. Could that account for the lack of nulls or my Psy smoothing?

 

[techsamurai]

The Marantz is running standalone and I have measurements but with a different mic position (close but not the same). But we can't fix the Marantz as you get what you get from its calibration.

The Denon has 6 pre-outs to the Marantz so it's doing no amplification at all - it's in pre-out mode.

My room has a cathedral ceiling with some hallways attached to the main room and the speakers are towards the far side of the cathedral ceiling so they can be seen from the kitchen. Could that account for the lack of nulls or my Psy smoothing?

I can do a fresh set of measurements this evening just to verify things are as expected and slight changes in the mic position don't uncover those nulls.

 

[Steve Dallas]

The Marantz is running standalone and I have measurements but with a different mic position (close but not the same). But we can't fix the Marantz as you get what you get from its calibration.

The Denon has 6 pre-outs to the Marantz so it's doing no amplification at all - it's in pre-out mode.

My room has a cathedral ceiling with some hallways attached to the main room and the speakers are towards the far side of the cathedral ceiling so they can be seen from the kitchen. Could that account for the lack of nulls or my Psy smoothing?

Until you run each on its own, you are comparing apples to oranges. To perform a true evaluation, you need to let each unit use its own amps. Make things as simple as possible and only change 1 thing at a time. You have unknown variables considering the complexity of the systems in these units. Later, after you have the Denon sounding good, if you want to slave the Marantz to the Denon, you can do that and troubleshoot any changes in sound.

I can do a fresh set of measurements this evening just to verify things are as expected and slight changes in the mic position don't uncover those nulls.

As others have said, the mic needs to be stationary between sweep measurements. Even a 1 inch change in any direction makes a meaningful difference in the measurements, which makes comparison difficult. Please re-shoot the measurements with the mic fixed on a stand. Alternatively, learn to take MMM measurements, which averages maybe 50 measurements as you move the mic around the headspace.

It is pretty much impossible to draw any conclusions from the measurements you have taken. Since they are so divergent, (they should not be), it is difficult to even align them. Since most AVRs use 1KHz to set speaker levels, I aligned them there against my preferred target curve. If they are meaningful, which I believe they are not, you can easily see why things sound so different. I don't even know how you could level-match these curves for a meaningful listening comparison.

I strongly suspect you have some double processing happening somewhere. Run the Denon on its own to eliminate that variable.

I now have some suspicions around why you did not like Dirac, which has always produced excellent results for me.

 

[peng]

The ability to play piano like that has always amazed me...

I am amazed the other competitors could not deliver that crescenda as impactful as her despite their seemingly much powerful physique. The peak of that note was probably at least 3 dB higher than the other's.

 

[Steve Dallas]

Here is another take on alignment that makes the Marantz look at little less absurd. (Colors are not the same. Sorry.)

 

[peng]

Until you run each on its own, you are comparing apples to oranges. To perform a true evaluation, you need to let each unit use its own amps. Make things as simple as possible and only change 1 thing at a time. You have unknown variables considering the complexity of the systems in these units. Later, after you have the Denon sounding good, if you want to slave the Marantz to the Denon, you can do that and troubleshoot any changes in sound.



As others have said, the mic needs to be stationary between sweep measurements. Even a 1 inch change in any direction makes a meaningful difference in the measurements, which makes comparison difficult. Please re-shoot the measurements with the mic fixed on a stand. Alternatively, learn to take MMM measurements, which averages maybe 50 measurements as you move the mic around the headspace.

It is pretty much impossible to draw any conclusions from the measurements you have taken. Since they are so divergent, (they should not be), it is difficult to even align them. Since most AVRs use 1KHz to set speaker levels, I aligned them there against my preferred target curve. If they are meaningful, which I believe they are not, you can easily see why things sound so different. I don't even know how you could level-match these curves for a meaningful listening comparison.

I strongly suspect you have some double processing happening somewhere. Run the Denon on its own to eliminate that variable.

View attachment 322722


I now have some suspicions around why you did not like Dirac, which has always produced excellent results for me.

View attachment 322725

I think it could be because he's using room correction to eq, sort of tone control to get results that please him. You and I are using RC to achieve the best inroom response first, and then we may try to tweak the target curve for more, or less exaggerrated bass.

 

[techsamurai]

Until you run each on its own, you are comparing apples to oranges. To perform a true evaluation, you need to let each unit use its own amps. Make things as simple as possible and only change 1 thing at a time. You have unknown variables considering the complexity of the systems in these units. Later, after you have the Denon sounding good, if you want to slave the Marantz to the Denon, you can do that and troubleshoot any changes in sound.



As others have said, the mic needs to be stationary between sweep measurements. Even a 1 inch change in any direction makes a meaningful difference in the measurements, which makes comparison difficult. Please re-shoot the measurements with the mic fixed on a stand. Alternatively, learn to take MMM measurements, which averages maybe 50 measurements as you move the mic around the headspace.

It is pretty much impossible to draw any conclusions from the measurements you have taken. Since they are so divergent, (they should not be), it is difficult to even align them. Since most AVRs use 1KHz to set speaker levels, I aligned them there against my preferred target curve. If they are meaningful, which I believe they are not, you can easily see why things sound so different. I don't even know how you could level-match these curves for a meaningful listening comparison.

I strongly suspect you have some double processing happening somewhere. Run the Denon on its own to eliminate that variable.

View attachment 322722


I now have some suspicions around why you did not like Dirac, which has always produced excellent results for me.

View attachment 322725

Thanks, I might try the amps and see what they sound like. Do I need to run another calibration as I take them out of pre-amp mode or are volumes that only thing that will be affected?

 

[Steve Dallas]

Thanks, I might try the amps and see what they sound like. Do I need to run another calibration as I take them out of pre-amp mode or are volumes that only thing that will be affected?

I would run a calibration to eliminate all unknown variables.

 

[techsamurai]

Here's my Audyssey app curve that gives me the blue response. Let me post the Denon L+R. Looks like the Harman curve a bit doesn't it?

I can't control the 30-50hz well on the iPad and I boosted the 10khz+ because of the room's attenuation

 

[peng]

Here's my Audyssey app curve that gives me the blue response. Let me post the Denon L+R. Looks like the Harman curve a bit doesn't it?

I can't control the 30-50hz well on the iPad and I boosted the 10khz+ because of the room's attenuation

View attachment 322754

Not really, Harman curve tilts the deep bass up, yours down, so more like you are doing the opposite.

Take a look of the Amir's in his Audyssey review thread. His looks like the Harman curve.

 

[techsamurai]

Not really, Harman curve tilts the deep bass up, yours down, so more like you are doing the opposite.

Take a look of the Amir's in his Audyssey review thread. His looks like the Harman curve.

Yeah, like I said I have no control over the deeper bass on the app - I clicked 100 times to manage to mark a higher curve than it wanted. It's a horrible bug... The Pro should cost $20 especially since Audyssey is flawed by design and the Pro version is necessary to fix it. I can't believe it doesn't take post calibration measurements and re-update itself.

 

[EWL5]

Yeah, like I said I have no control over the deeper bass on the app - I clicked 100 times to manage to mark a higher curve than it wanted. It's a horrible bug... The Pro should cost $20 especially since Audyssey is flawed by design and the Pro version is necessary to fix it. I can't believe it doesn't take post calibration measurements and re-update itself.

Do you still have your LFE crossover well below 80Hz?

 

[techsamurai]

Do you still have your LFE crossover well below 80Hz?

40hz for Mains
60hz for center
80hz for surrounds

40hz for 2ch playback (music)

 

[peng]

Yeah, like I said I have no control over the deeper bass on the app - I clicked 100 times to manage to mark a higher curve than it wanted. It's a horrible bug... The Pro should cost $20 especially since Audyssey is flawed by design and the Pro version is necessary to fix it. I can't believe it doesn't take post calibration measurements and re-update itself.

That is strange, Amir did it. I have done it, others have done it so maybe the new models (3800/4800 behaves differently). Steve Dallas had the 4700 before but has the 4800 now, maybe he can't control it either, can't wait to get his feedback.

Amir's:

 

[techsamurai]

That is strange, Amir did it. I have done it, others have done it so maybe the new models (3800/4800 behaves differently). Steve Dallas had the 4700 before but has the 4800 now, maybe he can't control it either, can't wait to get his feedback.

Amir's:

Hmm, it could be my "locks" right next to it - let me start with a fresh one and play with it. It also unexpectedly affects the other side of the range

 

[Steve Dallas]

That is strange, Amir did it. I have done it, others have done it so maybe the new models (3800/4800 behaves differently). Steve Dallas had the 4700 before but has the 4800 now, maybe he can't control it either, can't wait to get his feedback.

Amir's:

If I drop my finger all the way to left, I can easily control down to 20Hz. I use an Android tablet, though. There is no difference in usability of the app between the 4700 and 4800. Only the output file is different.

Yeah, like I said I have no control over the deeper bass on the app - I clicked 100 times to manage to mark a higher curve than it wanted. It's a horrible bug... The Pro should cost $20 especially since Audyssey is flawed by design and the Pro version is necessary to fix it. I can't believe it doesn't take post calibration measurements and re-update itself.

Audyssey is not flawed by design. Its default implementation works perfectly fine for 90%+ of its buyers. Your use case is special--probably the most special I have seen yet.

 

[peng]

Hmm, it could be my "locks" right next to it - let me start with a fresh one and play with it. It also unexpectedly affects the other side of the range

Also keep in mind the 703's -3 dB point seems to be around 45 Hz, that's anechoic, in-room it may go quite a bit lower, may be down to 35 or 32 Hz, but 20 Hz may be its reach, so that could be a limiting factor for the Audyssey's predicted/wishful thinking curves for the FL, FR C etc., but for the customized target curve, you should be able to drag it all the to the left as your sub(s) should have no trouble extending to 20 Hz or lower.

 

[Steve Dallas]

I alluded to this earlier, but I want to take the time to explain it.

Equalizing a speaker in a room with directivity considerations...

Many B&W speakers have considerable directivity errors. That means the frequency response of the off axis sound varies from the direct sound. The off axis sound is what is reflected from walls, ceilings, floors. While the direct sound may be fairly (or even very) even, the reflected sound may not be. How does EQing that uneven sound work in a room? The answer is: unpredictably.

Consider the measurement from your speakers posted earlier in the thread:

The direct sound is purple, and I assume red is 30 degrees off axis and blue is 60 degrees (?) off axis. You can see how much the frequency response changes from 0 to 30 to 60 degrees. Sixty degrees and greater is typically what is reflected back to you from the side walls. As reflected from the side walls, you are not hearing the purple line; you are hearing the blue line and worse. Those reflections mix with the purple line at a slight delay (phase difference) before reaching your ears.

This is a standard beamwidth plot, which illustrates the radiation pattern of the speaker according to its directivity properties:

This is not entirely accurate, but imagine the area inside the red lines is the direct sound. Within +/- 30 degrees, you hear all frequencies fairly evenly in a balanced speaker, or in your case, something like the purple direct sound line to the blue 30 degree line in your speaker's FR graph above. As you move off axis, the FR changes considerably. Imagine the magenta, green, and gray lines in the beamwidth graph are reflecting off the side walls with very uneven FR and mixing with your direct sound.

To make it easier to visualize, I will stand it up like the speaker would stand:

Imagine applying EQ at 8KHz, for example. If I am pushing the DRC curve around trying to affect parts of the FR I measured in my room, does the EQ change affect the magenta and gray lines in the same way? In the same amount? What happens when we introduce phase?

Here is a better B&W speaker. The area within the red lines is smoother, but things still get very ragged outside the red zone.

This speaker is also difficult to EQ due to its ragged reflections. (I owned the 805N for 20 years.)

Hopefully, this illustrates the difficulty of what you are trying to accomplish above about 1000Hz. Very much measurement, re-measurement, and trial and error will be required to get great results.

BTW, a typical room asserts its influence up to around 1000Hz. Below that, DRC is very effective at smoothing FR. Above that, I usually find it best to let the speakers be themselves after I have optimized placement, which is the 1st thing I do. Scratch that. The 1st thing I do is choose a speaker with even directivity an smooth FR, such as a Revel or KEF, then I optimize placement and continue on from there.

 

[morpheusX]

Sorry if this was already answered, but for those with crackling noises on REW, you need to go to the ASIO driver, and increase the buffer size.
More details here: https://www.avnirvana.com/threads/distortion-noise-coming-from-my-speakers-measuring-with-rew.8623/