• 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!

Best spec ADC Chip currently.... ??

nomograf

Member
Joined
Oct 7, 2021
Messages
24
Likes
12
Hey IVX,

I've tried sending an email to order as per your instructions but haven't received a reply. Is there another way?

(Have checked spam etc).

Do the AD you are selling now have the I2S output?

Cheers,
Ruairi
 

phoenixsong

Addicted to Fun and Learning
Joined
Nov 17, 2018
Messages
874
Likes
685
Additional measurements
 

Rja4000

Major Contributor
Forum Donor
Joined
May 31, 2019
Messages
2,682
Likes
4,220
Location
Liège, Belgium
Additional measurements
Linearity is very strange. :-S

Low frequency distortion as well (I wonder how many harmonics @WolfX-700 is allowing in his THD vs frequency measurement)
 

SharpEars

Member
Joined
Oct 3, 2019
Messages
22
Likes
16
Hey IVX,

I've tried sending an email to order as per your instructions but haven't received a reply. Is there another way?

(Have checked spam etc).

Do the AD you are selling now have the I2S output?

Cheers,
Ruairi
I am also interested in I2S output, not just USB from the Cosmos unit. That would really of interest to those of us that want to use the output digitally, either as I2s or convert to S/PDIF, AES/EBU, TOSLINK, etc...
 

staticV3

Master Contributor
Joined
Aug 29, 2019
Messages
7,535
Likes
12,002
Linearity is very strange. :-S

Low frequency distortion as well (I wonder how many harmonics @WolfX-700 is allowing in his THD vs frequency measurement)
Ivan measured linearity as well and got this:
Cosmos__ADC_Linearity.png
Perhaps WolfX-700's measurement was limited to 16bit.

As for the distortion peak at 40Hz, WolfX-700 did another 40Hz FFT and got this (not published for whatever reason):
unknown (1).png
114dB here vs 94 in the sweep that's part of the review.
 

Rja4000

Major Contributor
Forum Donor
Joined
May 31, 2019
Messages
2,682
Likes
4,220
Location
Liège, Belgium
Ivan measured linearity as well and got this:
View attachment 157748
Perhaps WolfX-700's measurement was limited to 16bit.

As for the distortion peak at 40Hz, WolfX-700 did another 40Hz FFT and got this (not published for whatever reason):
View attachment 157749
114dB here vs 94 in the sweep that's part of the review.
Yes, I also think there is probably a mistake in @WolfX-700 measurements here.
This kind of linearity for a device that is said to have 128dBA SNR makes little sense.
 

Rja4000

Major Contributor
Forum Donor
Joined
May 31, 2019
Messages
2,682
Likes
4,220
Location
Liège, Belgium
I am also interested in I2S output, not just USB from the Cosmos unit. That would really of interest to those of us that want to use the output digitally, either as I2s or convert to S/PDIF, AES/EBU, TOSLINK, etc...
Well, even for measurements it's interesting.
It's much easier to get a good measurement when the clocks of ADC and DAC are synched.
I usually use the ADC as the clock source.
 

Hee

Member
Joined
Jan 15, 2020
Messages
32
Likes
23
Cosmos adc can be said to be the best product of 2021.

I saw all the reviews. And I saw that the noise levels of the left and right channels were different in all products.
About 1.3db on the right channel is good.

I wonder if I can modify it with the tweak program provided by Ivan.
 
Last edited:

WolfX-700

Addicted to Fun and Learning
Forum Donor
Joined
Jul 26, 2019
Messages
537
Likes
4,508
Location
China
Yes, I also think there is probably a mistake in @WolfX-700 measurements here.
This kind of linearity for a device that is said to have 128dBA SNR makes little sense.

The linearity measurement difference comes from the filter setting. My website has been using a bandwidth setting of 20-20K, so the result is worse (contains more harmonics and noise). This has always been the traditional setting-I don't think this is a mistake.

Of course, one of the problems is that my previous measurements were mainly audio products, so I insist that the linearity measurement of 20-20K bandwidth is more meaningful. But as this ADC, it is closer to a measurement tool...so I will update the narrowband linearity measurement (800-1200HZ) later today.

THD+N frequency sweep, there was a discussion last night. First of all, the 40HZ of the chart I published is -10Xdb. As we all know, when AP measures this chart, the results are often worse than Dashboard. It is verified on the Dashboard that the distortion and noise of 40HZ are indeed the worst (SINAD114).
 

Rja4000

Major Contributor
Forum Donor
Joined
May 31, 2019
Messages
2,682
Likes
4,220
Location
Liège, Belgium
I will update the narrowband linearity measurement (800-1200HZ)
I think @amirm is using a very narrow band around 200Hz.
I wonder if that's not just one FFT window bin.

@WolfX-700 And how many harmonics do you allow in THD vs Frequency ?
You can chose in the AP settings for THD measurement.
I also wonder if and where @amirm sets a limit for that test.
 

Blumlein 88

Grand Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
20,523
Likes
37,053
I think @amirm is using a very narrow band around 200Hz.
I wonder if that's not just one FFT window bin.

@WolfX-700 And how many harmonics do you allow in THD vs Frequency ?
You can chose in the AP settings for THD measurement.
I also wonder if and where @amirm sets a limit for that test.
I don't do any standard linearity measurement. I use a quarter sample rate tone (11,025 hz for 44.1). Any harmonics are out of band. I use a signal that only has one bit level turned on at a given time. Pretty much any sigma delta DAC has clean linearity to 20 or more bits. Those with low enough noise I can confirm as having the proper level down to 23 bits and pretty much 24 bits. I believe even the noisy ones are getting linearity correct all the way down, they just get swamped by noise at the lowest 3 or 4 bits.
 

WolfX-700

Addicted to Fun and Learning
Forum Donor
Joined
Jul 26, 2019
Messages
537
Likes
4,508
Location
China
I think @amirm is using a very narrow band around 200Hz.
I wonder if that's not just one FFT window bin.

@WolfX-700 And how many harmonics do you allow in THD vs Frequency ?
You can chose in the AP settings for THD measurement.
I also wonder if and where @amirm sets a limit for that test.
My chart is THD+N frequency sweep and the bandwidth range is 20-20K. Not THD frequency sweep
 

Rja4000

Major Contributor
Forum Donor
Joined
May 31, 2019
Messages
2,682
Likes
4,220
Location
Liège, Belgium
that is simply impossible even theoretically if the linearity is 16bit but the distortions harmonics level -140db.
The linearity test to me is redundant because it duplicates THD and is less precise. Any nonlinearities will be detected by the THD 100.0% unavoidable.
Well, if you add some compression, I'm not sure you would see it in THD.
I posted something about that here.
I may be wrong, though.
 

WolfX-700

Addicted to Fun and Learning
Forum Donor
Joined
Jul 26, 2019
Messages
537
Likes
4,508
Location
China
I have a dream.

I dream that one day HiFi measurement will be popularized by everyone, and everyone who is interested in it can understand the real situation of their equipment at a very low cost.

And E1DA Cosmos ADC is our HOPE.

Thanks @IVX for developing this product
 

Rja4000

Major Contributor
Forum Donor
Joined
May 31, 2019
Messages
2,682
Likes
4,220
Location
Liège, Belgium
Rja4000, THD is the sum of harmonics powers. Any nonlinearity will add harmonics to the sin(x) function stimulus. Why do we need to test the linearity one more time after the THD test?
If the sine remains the same but only its level changes "dynamically" (with input level), a static sine signal will still be undistorted.
But, still, on music, you'll hear a difference.
 
Last edited:

JohnYang1997

Master Contributor
Technical Expert
Audio Company
Joined
Dec 28, 2018
Messages
7,175
Likes
18,292
Location
China
Rja4000, THD is the sum of harmonics powers. Any nonlinearity will add harmonics to the sin(x) function stimulus. Why do we need to test the linearity one more time after the THD test?
That's not what linearity test is.

Look here:
 
Top Bottom