• 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!

New set of graphs for all speakers reviews

Arc Acoustics

Member
Joined
Mar 17, 2021
Messages
74
Likes
53
Location
Japan
Thank you for another fiddly tracing.
One thing I noticed is, Perlisten S/R4"s" is an on-wall type loudspeaker for THX certification, thereby the measurement is half-space.
(Strictly speaking, I do not see the interference caused by the back wall, so the measurement would be full-space and just ditch the rear hemisphere I guess.)
I'm not sure handling full-space measurement and half-space measurement in the same criterion is the right thing to do because I could not find the post-processing method in ANSI/CTA-2034-A.
(They might use the VituixCAD and if you use the half-space (in the option), VituixCAD will return hor+/-90° as the ER_Rear because they are the only component of the ER_Rear in the frontal hemisphere.)
It should have very similar characteristics (except for the bass) to the S/R4"b", and they have (both measured by Perlisten) so it may be not a big deal.
https://pierreaubert.github.io/spinorama/speakers/Perlisten%20s4s/Perlisten/index_vendor.html
 
OP
P

pierre

Addicted to Fun and Learning
Forum Donor
Joined
Jul 1, 2017
Messages
963
Likes
3,048
Location
Switzerland
Right. It is a windows app and thus doesn’t help me much. There is plenty of great app for data visualization that are open source and run everywhere.
 

krabapple

Major Contributor
Forum Donor
Joined
Apr 15, 2016
Messages
3,169
Likes
3,717
Pierre,

Thanks for this great resource.

Would you consider adding an column for speaker size -- floorstanding vs bookshelf vs center -- to the Spinorama ranking table? It would make it much easier to see e.g. which bookshelf speaker ranks highest under different conditions.

And since I'm asking, also a column for active/passive?

Also, can you add a 'sort by' function to the page?
 
Last edited:
OP
P

pierre

Addicted to Fun and Learning
Forum Donor
Joined
Jul 1, 2017
Messages
963
Likes
3,048
Location
Switzerland
Pierre,

Thanks for this great resource.

Would you consider adding an column for speaker size -- floorstanding vs bookshelf vs center -- to the Spinorama ranking table? It would make it much easier to see e.g. which bookshelf speaker ranks highest under different conditions.

And since I'm asking, also a column for active/passive?

Also, can you add a 'sort by' function to the page?
Hello,

you can already filter and sort but that's not the exactly the same thing. Do you want the extra functionality directly in the table? I also have an export in a Google spreadsheet where you can slice and dice as much as you want if that helps.

Pierre

Here is an example filtering on center speakers and sorting by Score with Sub.

P.S.: I realize it is not super intuituve, if you have a better idea for the UI I am willing to change it.

Screenshot 2022-03-10 at 08.36.35.png
 

kernelpanic

Member
Joined
Mar 19, 2019
Messages
56
Likes
34
@pierre, came across your site today after it was mentioned by the Daily Hifi crew. Amazing job collecting and consolidating all of this data! Any chance you could add price as a search and filter field? It would make it even more powerful to search for speakers based on measurement results as well as budget.
 
OP
P

pierre

Addicted to Fun and Learning
Forum Donor
Joined
Jul 1, 2017
Messages
963
Likes
3,048
Location
Switzerland
@pierre, came across your site today after it was mentioned by the Daily Hifi crew. Amazing job collecting and consolidating all of this data! Any chance you could add price as a search and filter field? It would make it even more powerful to search for speakers based on measurement results as well as budget.
Yes I can do it. That’s easy. Next time I get 2h I will do it.

Done :) You can now filter by price.
You can also sort by F3, F6, Sensitivity and Flatness.
 
Last edited:

Arc Acoustics

Member
Joined
Mar 17, 2021
Messages
74
Likes
53
Location
Japan
Found the slightest mistake.
(Both showing EAC (v1))
As usual, thank you for all these great works!
1654759818681.png
1654760048899.png


Edit: Klipsch RP-600M II should be "Grille off" and "Grille on", and KEF Ref1 Meta should be "Short port" and "Long port".
 
Last edited:

Arc Acoustics

Member
Joined
Mar 17, 2021
Messages
74
Likes
53
Location
Japan

Arc Acoustics

Member
Joined
Mar 17, 2021
Messages
74
Likes
53
Location
Japan
OP
P

pierre

Addicted to Fun and Learning
Forum Donor
Joined
Jul 1, 2017
Messages
963
Likes
3,048
Location
Switzerland
OP
P

pierre

Addicted to Fun and Learning
Forum Donor
Joined
Jul 1, 2017
Messages
963
Likes
3,048
Location
Switzerland
@pierre The data of Neumi BS5P on the chart is "port stuffed".
(actually, the data provided by @hardisj, who might be banned by some small ******* by the way, at the below link looks "port stuffed" only)
I think it should be explained cuz the data "port opened" is pretty much bad.

Hello,

I added back the note section (which disappeared when I revamped the website). It is here but not very visible.

Screenshot 2022-07-25 at 12.52.26.png
 

Arc Acoustics

Member
Joined
Mar 17, 2021
Messages
74
Likes
53
Location
Japan

sweetchaos

Major Contributor
The Curator
Joined
Nov 29, 2019
Messages
3,872
Likes
11,557
Location
BC, Canada
Is anyone checking the veracity of the results on this site https://pierreaubert.github.io/spinorama/? Is that a completely false result appeared in relation to Elac Debut 2.0 b5.2 where it says "Bass extension: 200Hz at -6dB" this is completely false how can I believe the other results?
I would recommend that you look at the graph first
it's clearly a gated frequency response, so data below 200hz is missing.
CEA2034.jpg

The algorithm therefore assumes 200hz as the -6db point, which of course it isn't.

this is completely false how can I believe the other results?
that's a negative way of looking at things. good luck with that.
 

HelloHell

Member
Joined
Sep 3, 2022
Messages
32
Likes
30
I would recommend that you look at the graph first
it's clearly a gated frequency response, so data below 200hz is missing.
CEA2034.jpg

The algorithm therefore assumes 200hz as the -6db point, which of course it isn't.


that's a negative way of looking at things. good luck with that.
Thanks for clarifying, but people who only see the result in the shortcut or don't understand measurements see all other speakers with bass extension at around 60hz 38hz and then look at the Elac measurement and think that this is 3.25" mid/woofer...
 

beaRA

Active Member
Joined
Apr 16, 2021
Messages
223
Likes
315
Thanks for clarifying, but people who only see the result in the shortcut or don't understand measurements see all other speakers with bass extension at around 60hz 38hz and then look at the Elac measurement and think that this is 3.25" mid/woofer...
I think I heard on a live stream that Erin would be measuring the Elac B5.2, so this gated measurement may be superseded for the data summary.
 

sweetchaos

Major Contributor
The Curator
Joined
Nov 29, 2019
Messages
3,872
Likes
11,557
Location
BC, Canada
I think I heard on a live stream that Erin would be measuring the Elac B5.2, so this gated measurement may be superseded for the data summary.
Correct.
Erin has measured these and he said that they're not that good.
The data is not available yet on his website yet, but I expect that shortly.
He talked about it on *daily hifi podcast aug 29*, in case you're wondering.
 
OP
P

pierre

Addicted to Fun and Learning
Forum Donor
Joined
Jul 1, 2017
Messages
963
Likes
3,048
Location
Switzerland
I would recommend that you look at the graph first
it's clearly a gated frequency response, so data below 200hz is missing.
CEA2034.jpg

The algorithm therefore assumes 200hz as the -6db point, which of course it isn't.
that's a negative way of looking at things. good luck with that.
hello,

Data is from Nuyes and that’s the first time I get data from him. we are trying to find a format that works well. His data is gated at 5ms so valid above 200Hz. Below that he
does a ground plane method. Fine. Except that merging the on axis and the other spl curves is not super pretty. Current result is:

Code:
Reading cache ... (loaded 1 speakers)
2022-09-03 15:51:57,767 INFO services.py:1470 -- View the Ray dashboard at http://127.0.0.1:8265
Queing 1 speakers for EQ computations
(optim_save_peq pid=816956)          SPK auEQ
(optim_save_peq pid=816956) -----------------
(optim_save_peq pid=816956) NBD  ON 0.31 0.20
(optim_save_peq pid=816956) NBD  LW 0.29 0.17
(optim_save_peq pid=816956) NBD PIR 0.30 0.24
(optim_save_peq pid=816956) SM  PIR 0.12 0.91
(optim_save_peq pid=816956) SM   SP 0.51 0.93
(optim_save_peq pid=816956) LFX       48   47
(optim_save_peq pid=816956) LFQ     0.41 0.58
(optim_save_peq pid=816956) -----------------
(optim_save_peq pid=816956) Score    4.1  6.4
(optim_save_peq pid=816956) w/sub    6.3  8.6
(optim_save_peq pid=816956) -----------------
(optim_save_peq pid=816956) +4.05 +6.38 Elac Debut 2.0 B5.2

you can also filter the speaker’s measurement by quality. High will only show klippel like measurements, medium will show good quasi-anechoic or anechoic data, low is everything else.
filters_eq.png
 

Attachments

  • 1662267466932.png
    1662267466932.png
    252.9 KB · Views: 61
Last edited:

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
15,891
Likes
35,912
Location
The Neitherlands
@pierre First... great website. Never knew of its existence (totally missed it somehow)

What is the reasoning behind the blue bar for bass extension. Granted the numbers are displayed but I would consider 100Hz not 'extended' and then, by logic, I would the blue bar to be all on the left for 100Hz and for 20Hz on the right.

The same comment on smoothness. Why is the yellow bar not 1.0 on the right ?

Flattness a similar case why not have +/-10dB (which is terrible) all the way on the left ?

Just want to understand the reasoning. Maybe the scales are auto generated but see no reason not to fix the scales.
 
Last edited:
Top Bottom