Former Sennheiser engineer Axel Grell made a $200 pair of ANC true wireless earbuds - the TWS/1
Nov 10, 2021 at 9:59 AM Post #93 of 454
Joined
Mar 29, 2014
Posts
6
Likes
44
When is the official frequency graph going to release?
Now!

TWS1 FRQ.jpg

These are the official raw graphs (no smoothing). Measured under these conditions:
grell TWS/1
27.09.2021
21°C 55% RH 1009 hPa
Sine sweep 200 steps
0dB FS / Volume max
aptX Classic
IEC 318-4 (AKA IEC 711) coupler
Calibrated 427nV = 0 dB SPL

The green one is measured in ANC Off mode, the red one in ANC On mode. The differences between the modes are intentional.

Measurements only show graphs in charts. The only way to find out how headphones, especially in-ear ones, interact with your ear and how they sound is by ... listening to the headphones!

Enjoy the music!


Axel
 
grell audio Simplicity in all aspects but sound. Stay updated on grell audio at their sponsor profile on Head-Fi.
 
https://www.instagram.com/grellaudio/ https://grellaudio.com/ info@grellaudio.com
Nov 10, 2021 at 10:24 AM Post #94 of 454

angelom

1000+ Head-Fier
Joined
Apr 19, 2015
Posts
1,021
Likes
684
Now!


These are the official raw graphs (no smoothing). Measured under these conditions:
grell TWS/1
27.09.2021
21°C 55% RH 1009 hPa
Sine sweep 200 steps
0dB FS / Volume max
aptX Classic
IEC 318-4 (AKA IEC 711) coupler
Calibrated 427nV = 0 dB SPL

The green one is measured in ANC Off mode, the red one in ANC On mode. The differences between the modes are intentional.

Measurements only show graphs in charts. The only way to find out how headphones, especially in-ear ones, interact with your ear and how they sound is by ... listening to the headphones!

Enjoy the music!


Axel

Thank you for your post Axel, including the raw graph.

Is possible for you posting the smoothed graph too? I know that specially with earphones the personal fit is making a difference in sound, but the smoothed graph is giving a better general idea of the real sound of this (or other) earphone/headphone than the raw graph.
 
Last edited:
Nov 10, 2021 at 12:58 PM Post #95 of 454

crinacle

Member of the Trade: In-Ear Fidelity
Joined
Feb 3, 2011
Posts
2,399
Likes
3,287
Location
SG
Now!

TWS1 FRQ.jpg
These are the official raw graphs (no smoothing). Measured under these conditions:
grell TWS/1
27.09.2021
21°C 55% RH 1009 hPa
Sine sweep 200 steps
0dB FS / Volume max
aptX Classic
IEC 318-4 (AKA IEC 711) coupler
Calibrated 427nV = 0 dB SPL

The green one is measured in ANC Off mode, the red one in ANC On mode. The differences between the modes are intentional.

Measurements only show graphs in charts. The only way to find out how headphones, especially in-ear ones, interact with your ear and how they sound is by ... listening to the headphones!

Enjoy the music!


Axel

Hey Axel,

The unit I received for review seem to be significantly out of spec relative to the graph you just posted, also confirmed on the GRAS RA0402. Are there any software updates I need to install or do I need to get an exchange set up?
 
Nov 10, 2021 at 2:45 PM Post #96 of 454
Joined
Mar 29, 2014
Posts
6
Likes
44
Thank you for your post Axel, including the raw graph.

Is possible for you posting the smoothed graph too? I know that specially with earphones the personal fit is making a difference in sound, but the smoothed graph is giving a better general idea of the real sound of this (or other) earphone/headphone than the raw graph
I don't think that this gives a better picture as peaks are much better perceived than dips. The smoothing makes it all equal.
Hey Axel,

The unit I received for review seem to be significantly out of spec relative to the graph you just posted, also confirmed on the GRAS RA0402. Are there any software updates I need to install or do I need to get an exchange set up?
Hi Crinacle
This is strange! I've personally measured all samples that where sent out to reviewers. They were all good!
On the bottom of the battery case is a laser etched code (hard to see). Would you please be so kind and send me that code? I#d like to exchange your sample, but there is only one left here. And I need that!

Thank you!

Axel
 
grell audio Simplicity in all aspects but sound. Stay updated on grell audio at their sponsor profile on Head-Fi.
 
https://www.instagram.com/grellaudio/ https://grellaudio.com/ info@grellaudio.com
Nov 10, 2021 at 2:47 PM Post #97 of 454
Joined
Mar 29, 2014
Posts
6
Likes
44
I don't think that this gives a better picture as peaks are much better perceived than dips. The smoothing makes it all equal.

Hi Crinacle
This is strange! I've personally measured all samples that where sent out to reviewers. They were all good!
On the bottom of the battery case is a laser etched code (hard to see). Would you please be so kind and send me that code? I#d like to exchange your sample, but there is only one left here. And I need that!

Thank you!

Axel
I've never measured or heard a TWS/1 with such a frequency response.
 
grell audio Simplicity in all aspects but sound. Stay updated on grell audio at their sponsor profile on Head-Fi.
 
https://www.instagram.com/grellaudio/ https://grellaudio.com/ info@grellaudio.com
Nov 11, 2021 at 11:42 AM Post #98 of 454

angelom

1000+ Head-Fier
Joined
Apr 19, 2015
Posts
1,021
Likes
684
I don't think that this gives a better picture as peaks are much better perceived than dips. The smoothing makes it all equal.

Is really a pity you think a smoothed graph isn't useful. It is really for me and for a long time (specially from Tyll's very excellent work in the past in this aspect) because the smoothed graph is usually giving me personally a very good close representation of how I'm hearing a headphone or earphone.

Anyway, thank you for your answer.
 
Last edited:
Nov 11, 2021 at 12:55 PM Post #100 of 454

tinyman392

Be nice to noobs, we were all noobs at one point in our life.
Joined
Apr 27, 2011
Posts
8,707
Likes
1,679
Thank you for your post Axel, including the raw graph.

Is possible for you posting the smoothed graph too? I know that specially with earphones the personal fit is making a difference in sound, but the smoothed graph is giving a better general idea of the real sound of this (or other) earphone/headphone than the raw graph.
That's a pretty smooth graph on its own IMO. Smoothening it out would only get rid of the odd bump in the 250 Hz range and possibly reduce the amplitude of the spikes > 5kHz. I feel like a compensated graph (DF, Harman, etc.) would be more useful than just simply smoothing.
 
Nov 11, 2021 at 1:06 PM Post #101 of 454

angelom

1000+ Head-Fier
Joined
Apr 19, 2015
Posts
1,021
Likes
684
That's a pretty smooth graph on its own IMO
If you want saying that the raw graph above that Axel posting and the smoothed graph are actually very similar in this case, then I don't want this earphone.
I feel like a compensated graph (DF, Harman, etc.) would be more useful than just simply smoothing

Is common that the person or website giving you a compensated/smoothed graph will telling you that their reference point is more like harman or diffuse field or something similar that they thinking is correct.
 
Last edited:
Nov 11, 2021 at 1:13 PM Post #102 of 454

tinyman392

Be nice to noobs, we were all noobs at one point in our life.
Joined
Apr 27, 2011
Posts
8,707
Likes
1,679
If you want saying that the raw graph above that Axel posting and the smoothed graph are actually very similar in this case, then I don't want this earphone.


Is common that the person or website giving you a compensated/smoothed graph will telling you that their reference point is more like harman or diffuse field or something similar that they thinking is correct.
I kind of told you what the smoothed version of the graph would end up looking like, it's not really going to change much as all it might do is cut the amplitude of the spikes > 5kHz (depending on how much smoothing is applied). If you don't want the earphone, don't get it (Edit: I'm on the fence about it too, but mainly due to the amount of bass boost it has).

The main reason I prefer a compensated graph (any) is that it does cut down on the pinna gain and makes that a lot easier to compare to. Historically I'm accustomed to a DF compensation due to the way Tyll did things so I always compared to that, but nowadays everyone and their mother has their own compensation curve. Also, many different measurement rigs will also have different compensation that needs to be done to account for the stuff above 1k.
 
Last edited:
Nov 11, 2021 at 1:17 PM Post #103 of 454

angelom

1000+ Head-Fier
Joined
Apr 19, 2015
Posts
1,021
Likes
684
I kind of told you what the smoothed version of the graph would end up looking like, it's not really going to change much as all it might do is cut the amplitude of the spikes > 5kHz (depending on how much smoothing is applied). If you don't want the earphone, don't get it.

I prefer seeing the real compensated graph than the interpretation that you or other person is giving of the original raw graph.
 
Nov 11, 2021 at 1:56 PM Post #104 of 454
Joined
Mar 23, 2009
Posts
6,220
Likes
2,757
Location
Pittsburgh, Pennsylvania USA
I kind of told you what the smoothed version of the graph would end up looking like, it's not really going to change much as all it might do is cut the amplitude of the spikes > 5kHz (depending on how much smoothing is applied). If you don't want the earphone, don't get it (Edit: I'm on the fence about it too, but mainly due to the amount of bass boost it has).

The main reason I prefer a compensated graph (any) is that it does cut down on the pinna gain and makes that a lot easier to compare to. Historically I'm accustomed to a DF compensation due to the way Tyll did things so I always compared to that, but nowadays everyone and their mother has their own compensation curve. Also, many different measurement rigs will also have different compensation that needs to be done to account for the stuff above 1k.
& @angelom
Add to that, everyone (and their mother) has a unique compensation curve / frequency response due to the shape of their ear, head width, oral and nasal cavity, etc.

Although there are fewer factors that “change” the sound between the in-ear headphone and your cochlea than with over ear headphones or speakers, that fact actually makes it harder/impossible to tune an in-ear that will match how most people would perceive a “neutral studio monitor speaker setup” with all of your physical FR traits. To put it in another way, the tuning of an in-ear that matches what I would hear from a linear studio monitor would not sound linear to you… it would sound funky!

It looks like the TWS/1 worked with Sonarworks to develop an app called SoundID that tailors the sound for your ears based on a quiz of A/B comparison tests. You basically build your own bespoke compensation curve/EQ.
 
grell audio Simplicity in all aspects but sound. Stay updated on grell audio at their sponsor profile on Head-Fi.
 
https://www.instagram.com/grellaudio/ https://grellaudio.com/ info@grellaudio.com
Nov 11, 2021 at 3:34 PM Post #105 of 454

tinyman392

Be nice to noobs, we were all noobs at one point in our life.
Joined
Apr 27, 2011
Posts
8,707
Likes
1,679
I prefer seeing the real compensated graph than the interpretation that you or other person is giving of the original raw graph.
Wait, are you asking for a smoothed graph or a compensated graph (these are two very different things)? I have no clue what a real compensated graph is, could you elaborate?
 

Users who are viewing this thread

Top