• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). Come here to have fun, be ready to be teased and not take online life too seriously. We now measure and review equipment for free! Click here for details.

Measure your C50 in REW and tell us how your bass sounds!

Tim Link

Active Member
Forum Donor
Joined
Apr 10, 2020
Messages
192
Likes
183
Location
Eugene, OR
You are all invited to participate in this experiment to test the idea that a high C50 value as measured by REW equates to good, tight sounding bass. The idea here is that some value of C50 may correlate to subjectively good sounding bass in listening rooms. We're specifically looking at the C50 value in the bass and lower midrange, lets say 800hz and below, approximately. I work at ASC and our general experience is that for small listening environments like we have in our homes a C50 value of at least 10dB throughout the bass range is desirable for good sounding bass, with 15dB or higher being excellent. I have a small sample of client measurements and feedback so far that seem to correlate well.

Any thoughts or suggestions about this idea are of course highly welcomed.


It would be great to get some measurements from people who are hearing muddy bass as well as those who feel the bass in their rooms is very tight and clear. And anything in between.



To participate, here's what to do:

Measure a sweep with REW at your listening position with both (or however many) speakers playing.

Post the C50 plot.

Tell us how clear or just generally good the bass subjectively sounds to you. Comparison to headphone bass clarity is welcome.

Tell us a bit about your setup - how far you are from the speakers, what type they are, how big is the room, etc.

Here's mine.

Bighorn Corner Speakers C50.jpg



The room is 14.5 feet wide. The speaker tweeter centers are 10 feet apart. The microphone is at the listening position 9 feet from the tweeters. The speakers are large corner horn speakers, with bass cabinets stacked to the ceiling. The room is about 22 feet at it's longest, with some staggering of the back wall. It opens up to the kitchen and hallway.
My impression of the bass clarity is that it is overall good. I can follow acoustic bass lines as easily as I can on the headphones and nothing seems to boom, although the headphones do sound tighter and smoother, especially below 100Hz. There is a cancellation at about 80 hz at the listening position, hence the lowered clarity there. That region is an audible problem if I listen for it, although the cancellation zone is actually quite narrow so often it goes unnoticed with music.
 
Last edited:

abdo123

Major Contributor
Forum Donor
Joined
Nov 15, 2020
Messages
4,228
Likes
3,767
Location
Brussels, Belgium
Apparently a good Frequency response curve doesn't mean that you will get a high C50 score

1618642227626.png

1618642245785.png


Speakers are at the front wall, I sit on the back wall so ~3.5 meters distance. the room is 3.75*5 meters.

Speaker's preference score is 6.0, there is no treatment at all in the room.
 

ernestcarl

Major Contributor
Joined
Sep 4, 2019
Messages
1,636
Likes
1,142
Location
Canada
My small listening room dimensions:

4.92m x 3.27m x 2.33m (LxWxH) - basement listening space/desk office. There's single layer of dry wall surrounding the room and open hallway where the right speaker is sitting in the middle of -- but the basement is enclosed in much larger surrounding concrete foundation.

Right speaker's horn waveguide is literally half a foot away from the side-wall, but I've added some acoustic foam treatment directly beside the adjacent wall. This positioning necessitated further EQ -- as well as some extreme off-axis angling at ~22.5 degrees.

*Distance is exactly 2.2m from front of speaker baffle to the microphone.

In the latter two examples I give, the bass line sounds clearer and more palpable. Erin mentioned in his review of the Kali the word "vibrato" and that is exactly the word I would use here. And this vibrato sensation with the latter two FIR and time alignment corrections carries over around much the whole room and not just limited in the couch position. With first two settings, I felt that there was a need to increase the bass gain level. The more aligned the setup, the less need for bass boost as everything sounds clearer even without a boost.

Forgot to mention that I am using loudness compensation in JRiver so whenever levels (per channel) goes below ~85dB SPL, bass gradually increases. But the observation carries over even while disabling loudness compensation DSP.

Same EQ and measurements taken today of my right front main channel (S8) in front of the couch.

1618645370588.png



Now, just ignore the variance artifacts below and above the crossovers -- overlapped where mains use HPF 80Hz 24dB/oct LR and sub use LPF 120Hz 24db externally via JRiver as well as 12dB/oct via fixed internal sub setting.

1618645375235.png

Dip around ~130Hz is from a pervasive room node issue.

I have different headphones so not sure which one to compare with... but I'll post a comparison maybe tomorrow between the more common HD650 and this particular setup.


MDAT file comparison: RIGHT S8 - same day, time, mic position
 
Last edited:

pozz

Data Ordinator
Forum Donor
Editor
Joined
May 21, 2019
Messages
3,962
Likes
6,330
C50 is for speech clarity. It is the ratio of energy in the first 50ms to the decay tail afterwards.

It doesn't make sense at bass frequencies where the physical waveform periods and integration time for hearing are measured in the hundreds of milliseconds.
 

ernestcarl

Major Contributor
Joined
Sep 4, 2019
Messages
1,636
Likes
1,142
Location
Canada
The idea here is that some value of C50 may correlate to subjectively good sounding bass in listening rooms.

A correlation exists and is related to better GD as well and time alignment around the sub-mains xo. Even if theoretically C50 is a metric supposedly more in line with speech. I already knew that, but my own listening tests agree with the improved measurements making the bass line “better”. It’s not just a difference in SPL or frequency amplitude by any means.
 

ernestcarl

Major Contributor
Joined
Sep 4, 2019
Messages
1,636
Likes
1,142
Location
Canada
C50 is for speech clarity. It is the ratio of energy in the first 50ms to the decay tail afterwards.

It doesn't make sense at bass frequencies where the physical waveform periods and integration time for hearing are measured in the hundreds of milliseconds.

A FDW of 4 cycles also should not make any sense, but it actually correlates to what I hear:

1618664832354.png


In the same vein when I said previously in an older thread that I could after all, surprisingly, hear a (much more subtle) improvement after linearizing the mid and high xo even if this was well after the fact -- and not optimal. The FDW trace also did show a correlation... I know this could be nothing more than pure sighted bias, but that has been (and continues to be my repeatable) experience.

JRiver makes it possible to switch between different DSP preference configs quickly enough. But, of course, it's still not blind.

1618665593685.png


Same EQ and levels, but different time alignments and phase responses.
 

ernestcarl

Major Contributor
Joined
Sep 4, 2019
Messages
1,636
Likes
1,142
Location
Canada
Here's just one more alternate view in REW

Wavelet spectrogram - Linear (% peak)

with correction (trace #3)
1618673439752.png

1/6 resolution, 100% scale, Normalized


no FIR correction (trace #1)
1618672828325.png

1/6 resolution, 100% scale, Normalized


The discontinuity does not appear in the simple frequency response curves without filtering. But when you examine the various time domain information using the different plot views, the relationship between what one hears and what one measures is all there... just hidden.
 
Last edited:
OP
T

Tim Link

Active Member
Forum Donor
Joined
Apr 10, 2020
Messages
192
Likes
183
Location
Eugene, OR
Apparently a good Frequency response curve doesn't mean that you will get a high C50 score

View attachment 124534
View attachment 124535

Speakers are at the front wall, I sit on the back wall so ~3.5 meters distance. the room is 3.75*5 meters.

Speaker's preference score is 6.0, there is no treatment at all in the room.
Yours looks like a more typical untreated room with speakers other than dipoles or large arrays of woofers, or perhaps some of the newer cardioid pattern speakers, which due to their bass dispersion characteristics don't cause bass energy to get stored in the plane of the speakers as much. How would you describe the bass clarity in there compared to headphones?
 
OP
T

Tim Link

Active Member
Forum Donor
Joined
Apr 10, 2020
Messages
192
Likes
183
Location
Eugene, OR
Here's just one more alternate view in REW

Wavelet spectrogram - Linear (% peak)

with correction (trace #3)
View attachment 124578
1/6 resolution, 100% scale, Normalized


no FIR correction (trace #1)
View attachment 124573
1/6 resolution, 100% scale, Normalized


The discontinuity does not appear in the simple frequency response curves without filtering. But when you examine the various time domain information using the different plot views, the relationship between what one hears and what one measures is all there... just hidden.
I've noticed too that when I get the wavelet spectrogram to have a nice contiguous, connected starting pulse it sounds better and the C50 measures better. I don't have any FIR filter devices but with my Behringer digital active crossover I can choose various crossover settings between the woofer and midrange horn. Some of them snap together really nicely and measure well while others don't. I've also seen some of the advanced room correction filters also improve the bass clarity when turned on. So I'd say it's pretty solid proof that they are doing more than just simple frequency response adjustments.
 
OP
T

Tim Link

Active Member
Forum Donor
Joined
Apr 10, 2020
Messages
192
Likes
183
Location
Eugene, OR
C50 is for speech clarity. It is the ratio of energy in the first 50ms to the decay tail afterwards.

It doesn't make sense at bass frequencies where the physical waveform periods and integration time for hearing are measured in the hundreds of milliseconds.
At ASC we have been using the MATT test to measure bass clarity in rooms. We know it works because if you listen to it on headphones you can clearly here where all the bass tones stop and start. In most real listening rooms if you play it there are frequency bands where it is hard to hear the stops and starts because the bass continues for too long. We've been trying to see if other more widely accepted clarity measurements correlate to the MATT test well and C50 for various reasons seems to. We are definitely not using it for it's intended purpose here but it seems to work because of how it's calculated. The MATT test is tone bursts that last 1/8th of a second with 1/8th second silences between them. A room needs to settle down reasonably well within about a 16th of a second for the silent gaps to be clearly heard. I think it's very interesting that correcting group delay and using advanced room correction software also seems to increase measured clarity, both C50 and the MATT test.
 
Last edited:

Hipper

Addicted to Fun and Learning
Joined
Jun 16, 2019
Messages
637
Likes
512
Location
Herts., England
My room is 13' x 14' x 8' high. Speakers are VMPS RM30Ms. I listen 5' from the drivers and positioning is using approximately 'The Thirds' arrangement. Full treatment is 21 GIK Soffit Traps in wall-wall and wall-ceiling corners and a variety of panels. I didn't listen to bass without treatment but past experience told me it would be poor and boomy. With treatment it's much better, the key being the lower decay times. It was not perfect and I added EQ to make it better still. My whole strategy was to try and replicate the detail of headphone sound. The biggest problem I found to achieve this was not in the bass region but in the treble - percussion etc.. I think I've got more or less there by preventing side wall reflections and increasing the dB in that region.

In order to see how I was doing I used REW but only the SPL, waterfall and spectrograms. The other stuff like RT60 and Group Delay I never understood (or took much time to try to). I never noticed the 'Clarity' tab and don't understand it.

Below are SPL and C50 graphs showing no and full treatment but with no EQ. I've used the same parameters of the OP with no smoothing for the SPL graph.

No Treatment-Full Treatment SPL.jpg

No Treatment C50.jpg

Full Treatment C50.jpg


Adding EQ (only from 0-200Hz) gives this.

No EQ-EQ 9.18.jpg

With EQ.jpg

#
What it all means I've no idea!
 
Last edited:

Absolute

Addicted to Fun and Learning
Forum Donor
Joined
Feb 5, 2017
Messages
775
Likes
1,441
This is not my set-up, but I thought it would be interesting to see what a full-on double bass array frequency corrected with Audiolense looks like in a quite large dedicated room. Best bass quality I've ever experienced.

OMF-c50.jpg


OMF-waterfall.jpg
 

30 Ounce

Active Member
Joined
Apr 9, 2019
Messages
116
Likes
152
My speakers are B&W Signature 805 with Velodyne HGS15 sub that is eq'd with Behringer parametric EQ. Room is small 15X12 feet with floor to ceiling bass traps in the front corners. Sonex on side wall and ceiling reflections. Hardwood floors. I would term bass quality as excellent with acoustic vs electric bass easily discernible. The pat of a bass drum as well as tone development and decay are both excellent. Equal to or better than my headphones.
C50.jpg
 

ernestcarl

Major Contributor
Joined
Sep 4, 2019
Messages
1,636
Likes
1,142
Location
Canada
Forgot to mention that I am using loudness compensation in JRiver so whenever levels (per channel) goes below ~85dB SPL, bass gradually increases. But the observation carries over even while disabling loudness compensation DSP.

Sorry, I meant ~75dB which is about the average listening volume.

---

With regards to how it sounds in comparison to the HD650 -- equalized using Amir's PEQ settings:

I switched around between three different presets for the sofa setup:
1. Tested with stereo fronts playing only
2. Upmixed stereo to 4.1 ch with -5dB surrounds (default) for stronger front stage focus
3. Upmixed stereo to 4.1 ch with no reduction of upmixed surround volume for a more "live" or stronger envelopement

For the headphones, I switched around between Redline monitor (crossfeed plugin) being turned on and off.

1618776371347.png


Parallel playback into two DAC outputs (zone link feature in JRiver) was used -- switching bet. pausing the "sofa" zone and putting on the headphones, then taking it off and unpausing the sofa zone. Volume was balanced between the two zone outputs, of course.

Short conclusion (Sennheiser HD650):

Bass is quite good, BUT
  • less SPL & articulation
  • somewhat bloated/wooly
  • mids and highs have that boxy closed-in sound
  • Redline Monitor significantly reduced that boxy, closed-in sound (not completely)
  • surprisingly harsher at LOUD volumes
  • sadly lacks the physical sensation and "punch" of a real sub (obviously)
The open-back headphones sounded nice, but nowhere near as "open" and tactile as real speakers.
 
OP
T

Tim Link

Active Member
Forum Donor
Joined
Apr 10, 2020
Messages
192
Likes
183
Location
Eugene, OR
My room is 13' x 14' x 8' high. Speakers are VMPS RM30Ms. I listen 5' from the drivers and positioning is using approximately 'The Thirds' arrangement. Full treatment is 21 GIK Soffit Traps in wall-wall and wall-ceiling corners and a variety of panels. I didn't listen to bass without treatment but past experience told me it would be poor and boomy. With treatment it's much better, the key being the lower decay times. It was not perfect and I added EQ to make it better still. My whole strategy was to try and replicate the detail of headphone sound. The biggest problem I found to achieve this was not in the bass region but in the treble - percussion etc.. I think I've got more or less there by preventing side wall reflections and increasing the dB in that region.

In order to see how I was doing I used REW but only the SPL, waterfall and spectrograms. The other stuff like RT60 and Group Delay I never understood (or took much time to try to). I never noticed the 'Clarity' tab and don't understand it.

Below are SPL and C50 graphs showing no and full treatment but with no EQ. I've used the same parameters of the OP with no smoothing for the SPL graph.

View attachment 124726
View attachment 124727
View attachment 124728

Adding EQ (only from 0-200Hz) gives this.

View attachment 124730
View attachment 124729
#
What it all means I've no idea!
Your bass clarity went up a lot. I imagine it sounds a lot better after treatment than before, no?
 
OP
T

Tim Link

Active Member
Forum Donor
Joined
Apr 10, 2020
Messages
192
Likes
183
Location
Eugene, OR
This is not my set-up, but I thought it would be interesting to see what a full-on double bass array frequency corrected with Audiolense looks like in a quite large dedicated room. Best bass quality I've ever experienced.

View attachment 124752

View attachment 124755
That's impressive. I take it there are no acoustic treatments in that room?
 
OP
T

Tim Link

Active Member
Forum Donor
Joined
Apr 10, 2020
Messages
192
Likes
183
Location
Eugene, OR
Sorry, I meant ~75dB which is about the average listening volume.

---

With regards to how it sounds in comparison to the HD650 -- equalized using Amir's PEQ settings:

I switched around between three different presets for the sofa setup:
1. Tested with stereo fronts playing only
2. Upmixed stereo to 4.1 ch with -5dB surrounds (default) for stronger front stage focus
3. Upmixed stereo to 4.1 ch with no reduction of upmixed surround volume for a more "live" or stronger envelopement

For the headphones, I switched around between Redline monitor (crossfeed plugin) being turned on and off.

View attachment 124842

Parallel playback into two DAC outputs (zone link feature in JRiver) was used -- switching bet. pausing the "sofa" zone and putting on the headphones, then taking it off and unpausing the sofa zone. Volume was balanced between the two zone outputs, of course.

Short conclusion (Sennheiser HD650):

Bass is quite good, BUT
  • less SPL & articulation
  • somewhat bloated/wooly
  • mids and highs have that boxy closed-in sound
  • Redline Monitor significantly reduced that boxy, closed-in sound (not completely)
  • surprisingly harsher at LOUD volumes
  • sadly lacks the physical sensation and "punch" of a real sub (obviously)
The open-back headphones sounded nice, but nowhere near as "open" and tactile as real speakers.
You've beat the headphones across the board! With your in room measurements I would expect practically no advantage in clarity between headphones and in-room. I expect that a C50 of 20 or higher in the bass is pretty much audibly perfect.
 

ernestcarl

Major Contributor
Joined
Sep 4, 2019
Messages
1,636
Likes
1,142
Location
Canada
You've beat the headphones across the board! With your in room measurements I would expect practically no advantage in clarity between headphones and in-room. I expect that a C50 of 20 or higher in the bass is pretty much audibly perfect.

The bass in those senns do sound much better than what I have in another space...


So I've made some casual measurements of my open plan living room (single big square space of the kitchen, dining, and great room) today, and the results are not nearly as good. There aren't really any true listening positions as I rarely sit on the couch, so I'll include the clarity traces in a couple of areas.

Can't recall the exact area, but it's a very common mid-sized townhouse with a square, open plan living space configuration. Half of the square contains the kitchen and dining areas, and the other half is the 'great room' -- yeah, not really that grand or huge -- speakers are only about an inch from the front wall of the great room with the couch at the other end (maybe 10 inches from the rear wall), while to the left side of the great room are the other two mentioned dining and kitchen areas. Distance of the speakers to the couch is around 4.3-4.5 meters.

Acoustic treatment products are banned from this space.

Speakers are the (now discontinued) Mackie MR8 mk3 active studio monitors with their rears ports bunged (-2dB HF shelving engaged).

I've simplified the audio chain -- believe me, it was way more complicated before (whole house audio network):

TV female 3.5mm out > JDS labs subjective3 equalizer (-7 dB low shelf filter) > ROLLS Promatch (RCA to XLR) > TC Electronic Level Pilot > Active Monitors


*Notably, a large oblong shaped glass coffee table is directly right in front of the speakers (temporarily semi-permanently -- to keep kids away from the electronics).


Frequency Response (psy smoothing)

1618879580229.png


Bass is not excellent -- but acceptably good to me -- and, very much so all-over the place depending on where you sit/stand -- so not for critical listening or up to ASR's audiophile "hi-fi" standards.

Looks like there's quite a wide dip in the mid bass on our couch listening area.

The bass around the kitchen area is anemic, but mids and treble are quite clear for when listening to the news or following youtube cooking channel videos.

Substantial low frequency shelving EQ has to be applied as one doesn't want too much low bass in the dining table since that's just annoying as hell.


Clarity (C50)

1618879625995.png


I would say siting in the left seat makes me want to increase the bass tone control, but if I moved closer to the right seat afterwards... it makes me want to dial it back down -- C50 is not enough to explain why this is the case without looking at the SPL traces as well. In the end, it's better to have it somewhere in the middle.


1618879630508.png


As expected: the bass, mids, and highs are worst at the left-most corner of dining room which is more than 90 degrees off-axis the plane of the speakers (red trace).

Also, I'm not really sure if I should care much about the bass while toasting bread or frying some eggs on the kitchen stove/counter area (blue trace).


-----


Eventually, maybe I'll get a miniDSP 2x4 to supplement or fully replace the JDS subjective3 tone control equalizer that's currently in use.


Manual correction

1618882662735.png



Predicted EQ curves


1618882705869.gif


Nothing too complicated like invididual channel correction here. I'm also not planning on getting a sub to supplement the bass in this room. To me, it's good enough for general purpose listening or TV watching.
 
Last edited:

ernestcarl

Major Contributor
Joined
Sep 4, 2019
Messages
1,636
Likes
1,142
Location
Canada
Addendum to my first reply in this thread.

1618917714086.png


We can see how the sub/LFE channels (actually that's just a single sub with a breakdown of the output signals) here are substantially supporting and improving the clarity of the bass around the front and surround satellites' weakest null point.
 

mikewxyz

Member
Joined
Jun 19, 2019
Messages
32
Likes
32
Aside from what correlation Tim is hoping to establish, I’m interested in what clarity values mean for the home audio person in general. Most of what you find on the web deals with concert halls. Example: This is a great site for listening to different Clarity 80 values for the big spaces - https://www.concerthalls.org/?page_id=151

Googling optimum Clarity values I found this table –
OPTIMAL CLARITY in dB
-1.6 < C80 < 1.6 Instrumental music
-2 < C80 < +2 Symphonic music
0 < C80 < +4 Opera
C50 > +3 Speech
https://www.acoustex.de/fileadmin/user_upload/industry_acoustex_12dodicifacce_R02.pdf

Data for my Left and Right speakers shown. The red line in my REW graph represents C50. The green line is C80. My listening room is a large open floor plan. The “room” is ~ 6K ft^3 with openings to adjacent rooms. The mic was ~ 8 ft away from the speakers. My clarity values are higher than what is recommended in the above table.

I would be interested to hear what others consider a preferred C50 or C80 value and if these clarity values are correlated to other room attributes.

C50_C80 R Speaker.jpg
C50_C80 L speaker.jpg
 
Top Bottom