• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

REW waterfall vs Amir's speaker measurement waterfall

Pdxwayne

Major Contributor
Joined
Sep 15, 2020
Messages
3,219
Likes
1,172
I like how the bass sounds in my living room when playing Hans Zimmer music.

So I thought I would check the waterfall measurements in my room when playing right speaker with right sub (mic 9.5 ft away, 0 degree toe angle from speaker). Here is how it looks. It seems no obvious big problem for bass?

right_speaker_with_sub_9.5ft_0_degree_waterfall.PNG


I than see Amir's waterfall that only uses 4 msec of decay time. For example:

Triangle Borea BR03 Speaker Measurement CSD waterfall.png




I wonder how do I get similar look with REW for 4 ms? REW seems to default to 300ms and if I use 4 ms, I got no useful information. For example:


right_speaker_with_sub_9.5ft_0_degree_waterfall_4ms.PNG


Any tips?

Thanks!
 

sailor2005

Active Member
Joined
Oct 8, 2020
Messages
121
Likes
168
You won't. Because you are actually measuring your ROOM with REW. Amir measures the SPEAKER. The decay time you see in your measurements includes reflections in the room, Amir measures decay produced by the speaker itself, resonances in cabinet and drivers.
 
OP
Pdxwayne

Pdxwayne

Major Contributor
Joined
Sep 15, 2020
Messages
3,219
Likes
1,172
Also worth noting is that to examine the bass region, restrict the waterfall to 300 Hz.

Thanks!

I was under the impression that sometimes we want to know the balance of the decay for the whole range. For example, I see one site that sells absorption panels using full range measurements?
 
OP
Pdxwayne

Pdxwayne

Major Contributor
Joined
Sep 15, 2020
Messages
3,219
Likes
1,172
You won't. Because you are actually measuring your ROOM with REW. Amir measures the SPEAKER. The decay time you see in your measurements includes reflections in the room, Amir measures decay produced by the speaker itself, resonances in cabinet and drivers.
I see.....So REW is not the right tool. Thanks!
 

Trouble Maker

Addicted to Fun and Learning
Joined
Jan 6, 2020
Messages
676
Likes
709
Location
Columbus, Ohio, US
I see.....So REW is not the right tool. Thanks!

I took it as there are two different questions being asked. Not that your answer to your question was wrong. But I'm also learning so any experts please correct if needed.
Q1: What's the speaker behavior A1:Like this using anechoic measurements e.g. Klippel NFS
Q2: What's the behavior of the speaker in the room A1: Like this using in room measurements e.g. REW & Mic
2 different things.

I'm imagining that reverb/delay times in room are quite long compared to just the speaker. The sound has to bounce around until it looses enough energy to dissipate as it collides with and reflects off things around the room, combines in and out of phase with other like frequencies that are also moving around the room, escape via any openings, looses energy to moving/heating the air molecules (probably not a significant effect in practice but calculable in a sealed environment). Whereas in anechoic it should just be the initial incidence of sound since there should be no reflections.
 
Last edited:
OP
Pdxwayne

Pdxwayne

Major Contributor
Joined
Sep 15, 2020
Messages
3,219
Likes
1,172
If you had a mini amp and a laptop, you could measure outside via ground-plane.
Thank you and @Trouble Maker

I see that Amir's test is to measure speakers without the room. I can still use REW, but measurements should be made outside.

For me, since my room plays a big role in overall decays, using REW for in room waterfall measurements is the correct one to consider in my case.
 
Last edited:

napilopez

Major Contributor
Forum Donor
Joined
Oct 17, 2018
Messages
2,109
Likes
8,420
Location
NYC
I like how the bass sounds in my living room when playing Hans Zimmer music.

So I thought I would check the waterfall measurements in my room when playing right speaker with right sub (mic 9.5 ft away, 0 degree toe angle from speaker). Here is how it looks. It seems no obvious big problem for bass?

View attachment 120285

I than see Amir's waterfall that only uses 4 msec of decay time. For example:

View attachment 120286



I wonder how do I get similar look with REW for 4 ms? REW seems to default to 300ms and if I use 4 ms, I got no useful information. For example:


View attachment 120287

Any tips?

Thanks!

As others have mentioned, from that distance in room, you're not going to get results that match Amir's.

That said, and though I don't really like waterfall plots, you can actually get somewhat similar results with REW to Amir if you take quasi-anechoic measurements. Keep in mind it's not just the time range you need to adjust though; you need to adjust the rise time to be smaller too. I use 0.2ms.

For example, here's Amir's of the JBL HDI-1600:

1616697022792.png


Here's mine of the same speaker. Showing the settings in case anyone wants to try it out when making their own quasi anechoic measurements.

Snag_922096a.png


Not as accurate especially as this was a test measurement, but you see the same issue around the crossover, at least.

The "Window(ms)" setting will obviously depend on whatever window is effective for your own quasi-anechoic data. But to get the 'look' of Amir or stereophile's CSD plots, something roughly akin to these settings should work.

Edit: also note you'll have to fiddle around with the left scroll bar to find the right SPL cut-off.
 
Last edited:
OP
Pdxwayne

Pdxwayne

Major Contributor
Joined
Sep 15, 2020
Messages
3,219
Likes
1,172
As others have mentioned, from that distance in room, you're not going to get results that match Amir's.

That said, and though I don't really like waterfall plots, you can actually get somewhat similar results with REW to Amir if you take quasi-anechoic measurements. Keep in mind it's not just the time range you need to adjust though; you need to adjust the rise time to be smaller too. I use 0.2ms.

For example, here's Amir's of the JBL HDI-1600:

View attachment 120300

Here's mine of the same speaker. Showing the settings in case anyone wants to try it out when making their own quasi anechoic measurements.

View attachment 120302

Not as accurate especially as this was a test measurement, but you see the same issue around the crossover, at least.

The "Window(ms)" setting will obviously depend on whatever window is effective for your own quasi-anechoic data. But to get the 'look' of Amir or stereophile's CSD plots, something roughly akin to these settings should work.
Thanks for the advise about rise time!
 

tuga

Major Contributor
Joined
Feb 5, 2020
Messages
3,984
Likes
4,281
Location
Oxford, England
As others have mentioned, from that distance in room, you're not going to get results that match Amir's.

That said, and though I don't really like waterfall plots, you can actually get somewhat similar results with REW to Amir if you take quasi-anechoic measurements. Keep in mind it's not just the time range you need to adjust though; you need to adjust the rise time to be smaller too. I use 0.2ms.

For example, here's Amir's of the JBL HDI-1600:

View attachment 120300

Here's mine of the same speaker. Showing the settings in case anyone wants to try it out when making their own quasi anechoic measurements.

View attachment 120302

Not as accurate especially as this was a test measurement, but you see the same issue around the crossover, at least.

The "Window(ms)" setting will obviously depend on whatever window is effective for your own quasi-anechoic data. But to get the 'look' of Amir or stereophile's CSD plots, something roughly akin to these settings should work.

That tweeter resonance is way too close for comfort... Thanks for showing us the settings.
 
Top Bottom