How do my graphs look?

You seem to have a massive 20 dB boost to low bass...

There's a lot of echo (1/2 second at 400 - 5k Hz).

If you want to see some very good measurements to compare, I suggest downloading the 'gearslutz.mdat' from gearspace.com: https://gearspace.com/board/studio-...unch-gearslutz-rew-results-one-mdat-file.html

Jim1961's mdat in that collection is the best I've seen.

Underdog said:

but I'm not really sure what I'm seeing here.

Click to expand...

Hello, I'm also not quite sure about the measurements :


Sound pressure up to 40 Hz is quite real, but below 40 Hz what sings?
the subwoofer reproduces up to a maximum of 120 Hz. And what plays above 120 on the chart?

skid00 said:

Jim1961's mdat in that collection is the best I've seen.

Click to expand...

Hello, Jim 1961 is one of the best, but..


But the bass is raised below 200 Hz. As if a heavy bulldozer is working nearby.

Anatoliy Gavrilov said:

Здравствуйте, я тоже не совсем уверен в размерах:


Звуковое давление до 40 Гц вполне реально, а ниже 40 Гц что поет?
сабвуфер максимальное поглощение до 120 Гц. А что играет выше 120 на графике?

Click to expand...

Anatoliy Gavrilov said:

Здравствуйте,Джим 1961 один из лучших, А вот поднялся ниже 200 Гц. Как будто рядом работает тяжелый бульдозер.

Click to expand...

Эти измерения лучше:

Anatoliy,​

That bass boost is desirable. Google Harman curve.

And more info about House Curves... https://www.bksv.com/media/doc/17-197.pdf Notice figure 5 on page 4...
And then there is this by Floyd Toole... https://www.aes.org/e-lib/browse.cfm?elib=17839

What speakers are actually playing for "Right", "RS", and "RS2"?

I had assumed that "Right" was the right speaker alone, and that "RS" was the right sub by itself. Not sure what "RS2" might be...perhaps a second sub on the right side?

However, all three of these speakers, "Right", "RS", and "RS2" are +21 dB at 22 Hz, so that does not seem right. Also, "RS" has 5 dB more output at 7 kHz than "Right", which rules out "RS" from being a subwoofer.

Even the center channel has massive output at 22 Hz, so it seems unlikely that it was measured independently of subwoofers.

Another possibility is that "RS" means "Right Surround" and that all speakers were measured with the same subwoofer active. Actually, that's the only reasonable explanation. But I still don't know what "RS2" and "LS2" are. Perhaps rear surround speakers in a 7.1 setup? There's an attached floorplan diagram, but it does not include any useful information about speaker placement or names.

Assuming that my last interpretation is correct, the biggest problem that I see is a large discrepancy between the left and right channels. This will have a negative effect on soundstage and imaging for stereo content, although it may not be that consequential for movies.



Differences are pretty dramatic below 80 Hz (less critical) and above 1.5 kHz (significant).

In the time domain, things are even more crazy.



I see something sort-of resembling a step response from the left channel, but the right channel is not showing a recognizable step response. Ideally, these two curves should match each other closely and look something like this (recent measurement at a friend's house with 2-way Paradigm monitors):



Here's the amplitude response of the left and right channels from the same system:



Note: these measurements were taken with no subwoofer...just a simple 2.0 system. But I hope this gives you an idea of the kind of response you're shooting for. Mainly tighter correlation between left and right channels.

-- David

In an asymmetrical layout I would expect there to be an asymmetrical response... Might be interesting to see what a Maxwell's raytracing room simulator thinks about this...

ddude003 said:

In an asymmetrical layout I would expect there to be an asymmetrical response... Might be interesting to see what a Maxwell's raytracing room simulator thinks about this...

Click to expand...

True, but I suspect there's more going on here than asymmetry. Without knowing more about how these measurements were taken, all bets are off on a rational interpretation of the data.

dsnyder0cnn said:

What speakers are actually playing for "Right", "RS", and "RS2"?

I had assumed that "Right" was the right speaker alone, and that "RS" was the right sub by itself. Not sure what "RS2" might be...perhaps a second sub on the right side?

However, all three of these speakers, "Right", "RS", and "RS2" are +21 dB at 22 Hz, so that does not seem right. Also, "RS" has 5 dB more output at 7 kHz than "Right", which rules out "RS" from being a subwoofer.

Even the center channel has massive output at 22 Hz, so it seems unlikely that it was measured independently of subwoofers.

Another possibility is that "RS" means "Right Surround" and that all speakers were measured with the same subwoofer active. Actually, that's the only reasonable explanation. But I still don't know what "RS2" and "LS2" are. Perhaps rear surround speakers in a 7.1 setup? There's an attached floorplan diagram, but it does not include any useful information about speaker placement or names.

Assuming that my last interpretation is correct, the biggest problem that I see is a large discrepancy between the left and right channels. This will have a negative effect on soundstage and imaging for stereo content, although it may not be that consequential for movies.

View attachment 58049

Differences are pretty dramatic below 80 Hz (less critical) and above 1.5 kHz (significant).

In the time domain, things are even more crazy.

View attachment 58050

I see something sort-of resembling a step response from the left channel, but the right channel is not showing a recognizable step response. Ideally, these two curves should match each other closely and look something like this (recent measurement at a friend's house with 2-way Paradigm monitors):

View attachment 58051

Here's the amplitude response of the left and right channels from the same system:

View attachment 58052

Note: these measurements were taken with no subwoofer...just a simple 2.0 system. But I hope this gives you an idea of the kind of response you're shooting for. Mainly tighter correlation between left and right channels.

-- David

Click to expand...

If that measure is with EQ, on top of that there is symmetry in the speakers vs side walls,
it cannot be extrapolated to what they are asking.
Now, it would be nice if you put the .mdat of your friend's room.
I'm curious to see the acoustics of the room if possible.

written with translator

Greetings

dsnyder0cnn said:

Without knowing more about how these measurements were taken, all bets are off on a rational interpretation of the data.

Click to expand...

Agreed... A clear understanding of what (speaker(s)) is/are being measured and if Audeyssey EQ was enabled for those measurements... Also, what is the (basement?) room made of and are there currently any acoustic treatments? Lots of variables need addressed before much more can be said...

hemiutut said:

If that measure is with EQ, on top of that there is symmetry in the speakers vs side walls,
it cannot be extrapolated to what they are asking.
Now, it would be nice if you put the .mdat of your friend's room.
I'm curious to see the acoustics of the room if possible.

written with translator

Greetings

Click to expand...

I don't have a post-correction .mdat from my friend's room, but I just built new filters for my room today. I don't mind sharing the .mdat for that. This photo shows the measurement conditions pretty well. Fritz Carrera 7 BE monitors about 7.5 ft away from the MLP. Dual SVS 3000 Micro subs. So, a 2.2 system.



Details are in the .mdat file, but here's the actual acoustic amplitude response relative to the target (white) that I designed. Note: this is not a simulation. It's an actual acoustic measurement with the filters in place. I used REW's generator to save a pair of sweeps with timing references to WAV files, one for Left and one for Right. I then played these sweep files back in Roon while recording them in REW.



What we have is -3 dB at 20 Hz with roughly +1/-2 dB deviation from target in the audible band (at least for a 55 year-old like me). The attached .mdat file has what you see here plus the "before" measurements. What's more interesting than amplitude response is improvements to the time domain performance. Here's the step response before and after correction:




Some phase errors still exist due to room asymmetries. I may try to correct some of these, but I'm already happy with the results. Soundstage is much taller with the filter in place (not what I was expecting). Tonality and harmonic density of information is astounding. Looking forward to seeing what else I can do with this relatively modest system.

Other bits include a Gustard X18 DAC and Topping Pre90 analog preamp. Power amps are Orchard Audio monoblocks.

Sorry if this partially derails the OP's discussion, but I put this up as another example of the kind of thing we want to see in REW after applying room correction filters. Not perfect, but clearly an improvement over native response. Emphasizing again the goal of tight correlation between left and right speakers, both in amplitude and phase.

Edit: I should add that the MLP is not the couch behind the mic. That's just there for guests and because I have no other place to put it. MLP is a chair that I roll out in front of the couch for serious listening. The chair has height adjustment to get each listener's ears to the right level for best sound. Just in case you were puzzled by the mic position relative to the couch.

skid00 said:

That bass boost is desirable. Google Harman curve

Click to expand...

Yes, I liked the Harman curve and it is quite real in our settings. In it, the high-frequency roll-off and low-frequency rise are 3 dB relative to a frequency of 2 kHz, The Jim 1961 has a bass boost of +7dB - that's relative to 2kHz.

dsnyder0cnn​

Are these measurements from the same microphone position? Very high quality sound settings.
And if you put a microphone for measurements at 5-6 points in front of the chair?
That is, to make a composite graph of the frequency response in REW from 5-6 measurements.

Anatoliy Gavrilov said:

dsnyder0cnn​

Are these measurements from the same microphone position? Very high quality sound settings.
And if you put a microphone for measurements at 5-6 points in front of the chair?
That is, to make a composite graph of the frequency response in REW from 5-6 measurements.

Click to expand...

Yes. Measurements from my room were taken from the same microphone position, before and after.

It's been a while since I've taking the time to measure sweeps from multiple locations…not just the MLP, but last time I did using Acourate filters, the correlation was quite good. The tonality is decent from seats on the couch behind the listening chair. However, soundstage and imaging magic mostly happens only at the MLP.

dsnyder0cnn said:

Details are in the .mdat file, but here's the actual acoustic amplitude response relative to the target (white) that I designed. Note: this is not a simulation. It's an actual acoustic measurement with the filters in place.

Click to expand...

Impressive result. But your RT60 metrics seem rather high, and your Clarity metrics rather low. Just curious as to your take on that?

thothsong said:

Impressive result. But your RT60 metrics seem rather high, and your Clarity metrics rather low. Just curious as to your take on that?

Click to expand...

Yeah. I know. Current listening environment is a loft space in a rental home. The back half of the right wall opens to the floor below. If we were staying here, I'd cover the walls of that staircase with acoustic panels to kill the echo. IIRC, RT60 shoots up to over 4 seconds at high frequencies because of that area.

I would want to add diffusion to the ceiling also. I have the panels, but I don't want to have to repair the ceiling to get our deposit back. Hoping I can do more with the room once we find the house we want to buy.

We have not heard from the OP, @Underdog in a while. I'm still curious to understand how they took their measurements and why they think the left and right channels are so radically different. It has to be more than acoustics.

Yes. And based on his other forum posting history, he knows exactly what he is doing, and he prefers that amount of bass boost.

Anatoliy Gavrilov said:

Yes, I liked the Harman curve and it is quite real in our settings. In it, the high-frequency roll-off and low-frequency rise are 3 dB relative to a frequency of 2 kHz, The Jim 1961 has a bass boost of +7dB - that's relative to 2kHz.

Click to expand...

skid00 said:

Yes. And based on his other forum posting history, he knows exactly what he is doing, and he prefers that amount of bass boost.

Click to expand...

I like subbass as much as the next guy, but there's nearly 30 dB of difference between the left and right channels at 25 Hz. That's not normal. If he knows what he's doing, he should be able to figure that out.

He asked for feedback…my feedback is to change things in his room/system or measurement protocol until the left and right channels match…ideally within +/- 1 dB.

Kind-of strange that he created this thread a week ago and then vanished. I guess it is the holidays, but why ask a question when you won't have time to come back and read the answers? Otherwise, it's been a good discussion.

Hmm... It looks me like @skid00 and @dsnyder0cnn are talking about two different people... @skid00 is talking about Jim 1961's preference of bass and boost and @dsnyder0cnn is talking about the OPs difference between what appears to be a left and right channel... Did I miss something here? No worries...

@dsnyder0cnn your kit looks very nice and your filters look like they are doing a great job getting to your target curve... And I am wondering if you have looked into or tried Serkan Gur's inversion method?

dsnyder0cnn,​

Anatoliy and I were talking about jim1961's REW graph.​

And I should have pointed out that Jim's other data (impulse response, distortion, etc) were excellent, too. But the OP seemed to be an REW amateur, so I left that out.

Forum postings leave a lot to be desired!