The No.1 Website for Pro Audio
Audio interfaces and their AD/DA chips LISTED
Old 16th April 2019
  #421
Gear Maniac
 

The dac chip only works with 24 bits. It is truncated back to the original bit depth. But it does not drop a bit for every -6.1 dB. Digital attenuation would be a real piece of crap if a 16 bit signal dropped 8-10 bits of resolution when attenuated by -40 to -60 dB.
I would swap the AK413 for a AK4414 to test. Have a spare 4Pre to test on. Strong feeling it will not drop in and operate. Data sheet looks compatible.
I have a unit with ESS9018 chips. That gets around the 24 bit issue. But there are other concerns.
Focusrite could upgrade the dac chips. At quantity this is maybe a 1.00 USD cost increase. I think monitor output is not a priority.
Old 16th April 2019
  #422
Gear Nut
 
SPiTFiREgr's Avatar
 

Quote:
Originally Posted by Panelhead View Post
The dac chip only works with 24 bits. It is truncated back to the original bit depth.
Without any dithering applied?
Old 16th April 2019
  #423
Gear Head
 
Sabovic Adis's Avatar
Now I'ma show you a sciolist wizard

Quote:
Originally Posted by SPiTFiREgr View Post
There is no such thing as 8 bits applied to 24 bits. If the final, DAC stage bitdepth is 24-bit , this is your final bitdepth.

If there is a conversion to 32 bits prior the DAC <-> Digital attenuation <-> Conversion to 24 bits again (with dither applied maybe?) <-> into the DAC , that's another story, but I'm not aware of such an implementation.



Utilizing a 32-bit DAC for nearly lossless digital attenuation isn't "cool" , it's a major audio quality advantage. If you are ignorant enough to not be able to understand that (which I wouldn't be surprised, audio industry is full of sciolist wizards), you can keep on feeling confident monitoring with digital attenuation in your 24-bit DAC

A very simple test to understand the audio quality loss of digital attenuation can be done by using the excellent Audio MIDI Setup of macOS. The built-in DAC of Apple machines support 32-bit for this exact reason: digital attenuation. Plug-in your headphones, set the volume really low and listen to some low frequency transients blended with reverb tails. Switch between 24-bit and 32-bit in the format dropdown menu. You won't be able to hear the lower volume details and reverb tails using the 24-bit format.

Another way to test the same thing is by using your audio interface utility app, if it rocks a digital attenuator for the DAC stage. Most probably, your headphone out is rocking an analog attenuator as well, because a headphone amp is present. Depending on your audio interface vendor, you either have a DSP mixer fader for that stereo headphone out and/or a DAC output (digital attenuator) fader (like in Focusrite's Clarett range). Try to evaluate the level of detail in the same low frequency transients blended with reverb tails material, in very low volume. Compare the digital attenuator at 0dB setting with reduced volume at the analog h/w volume knob vs setting the digital attenuator at -30dB and the analog h/w volume knob at 0dB.
Son,

everybody can hear the difference between 256 code values in 8 bit and 65536 cv in 16. Not everybody can hear the difference between 65536 code values in 16 bit and 16777216 cv in 24. And no one can really hear the difference between 16777216 code values in 24 bit and 17179869184 cv 32.
We're talking 1/17179869184 or 5.82076609134674072265625e-11 precision here, hardly a major audio quality advantage. So, suck it up!

- Daddy
Old 17th April 2019
  #424
Gear Nut
 
SPiTFiREgr's Avatar
 

Quote:
Originally Posted by Sabovic Adis View Post
everybody can hear the difference between 256 code values in 8 bit and 65536 cv in 16. Not everybody can hear the difference between 65536 code values in 16 bit and 16777216 cv in 24. And no one can really hear the difference between 16777216 code values in 24 bit and 17179869184 cv 32.
We're talking 1/17179869184 or 5.82076609134674072265625e-11 precision here, hardly a major audio quality advantage. So, suck it up!
Old man,

Can you understand how many of these "16777216 code values" in your 24-bit stream you utilize when you digitally attenuate your stream with -60dBFS?

Probably not.

Do you care? As a sciolist audio wizard, probably not either.

If you are so frustrated with that 32-bit DAC "nonsense", why don't you send a message to the mailing list for IC designers at Asahi Kasei, ESS, Apogee, Apple, Google? Let them have a laugh as well
Old 18th April 2019
  #425
Gear Head
 
Sabovic Adis's Avatar
Son,

Quote:
Originally Posted by SPiTFiREgr View Post
Old man,

Can you understand how many of these "16777216 code values" in your 24-bit stream you utilize when you digitally attenuate your stream with -60dBFS?
you do that in my studio and I'll fire you instantly! Like, literarlly:





Quote:
Probably not.
Do you care? As a sciolist audio wizard, probably not either.

If you are so frustrated with that 32-bit DAC "nonsense", why don't you send a message to the mailing list for IC designers at Asahi Kasei, ESS, Apogee, Apple, Google? Let them have a laugh as well
There you go, Son, for when you can't think for yourself...


- Daddy
Old 18th April 2019
  #426
Gear Nut
 
SPiTFiREgr's Avatar
 

Quote:
Originally Posted by Sabovic Adis View Post
Son,



you do that in my studio and I'll fire you instantly! Like, literarlly
So you are pretty much confirming the same thing:



This is the Clarett monitor pot position for the "instant fire" digital attenuation value
Old 18th April 2019
  #427
Gear Maniac
 

Can you adjust the gain elsewhere? I try to keep my 4Pre monitor pot in 12 - 2 o’clock range. It has a barely audible output when set at 9:30 - 10:00.
Old 18th April 2019
  #428
Lives for gear
Pad the signal comming in. If you have to run the output that low, your headroom is not going to be good.
Old 19th April 2019
  #429
Gear Nut
 
SPiTFiREgr's Avatar
 

Quote:
Originally Posted by Panelhead View Post
Can you adjust the gain elsewhere? I try to keep my 4Pre monitor pot in 12 - 2 o’clock range. It has a barely audible output when set at 9:30 - 10:00.
I can and I'm doing it, that's exactly what I'm saying. If I didn't, I would have to use the "digital volume control" of the interface/dac, thus reducing the bitdepth.

Quote:
Originally Posted by elegentdrum View Post
Pad the signal comming in. If you have to run the output that low, your headroom is not going to be good.
I actually completely disable the digital volume control, using the "None" setting on Focusrite Control and adjust the output gain using a passive volume attenuator

That was the whole argument:

Digital volume control on 24-bit DACs vs Digital volume control on 32-bit DACs
Old 27th May 2019
  #430
Here for the gear
 

I have both MK I and MK II Apogee Symphony 8x8 modules and here is what I found out about their AD/DA converters:
MK I: AD CS5381 x4, DA Apogee’s own chip x1

MK II: AD AK5388AEQ x4, DA ES9018S x1
Old 27th May 2019
  #431
Gear Maniac
 

Are you sure about the D/A conversion chip in the MK 1? Seems if Apogee built a chip they would use it across the product line to increase volume.
To justify building a proprietary chip requires high volumes and high prices to cover the R&D.
Old 27th May 2019
  #432
Lives for gear
 
s wave's Avatar
here is a good older list of AD DA - https://taylorcopaq.jimdo.com/2018/0...-converter-ic/
Old 30th May 2019
  #433
Here for the gear
 

Quote:
Originally Posted by Panelhead View Post
Are you sure about the D/A conversion chip in the MK 1? Seems if Apogee built a chip they would use it across the product line to increase volume.
To justify building a proprietary chip requires high volumes and high prices to cover the R&D.
Yes, it might just be a rebranded DAC. Here is a picture of it:
Attached Thumbnails
Audio interfaces and their AD/DA chips LISTED-apogee-dac.jpg  
Old 31st May 2019
  #434
Lives for gear
 
s wave's Avatar
CS5381 specs https://statics.cirrus.com/pubs/proD.../CS5381_F2.pdf

7. REVISION HISTORYReleaseDateChangesA1December 2002Initial ReleaseA2October 2003Changed front page description of digital filterImproved distortion specification from -105 dB to -110 dBModified serial port timing specifications for slave mode operationAdded pull-down resistors to recommended input bufferA3May 2004Changed full-scale voltage specification to reflect VA supply voltageAdded Applications section about capacitor value on FILT+ pinChanged input impedance specification from 37 to 2.5 kΩChanged impedance specification on FILT+ from 35 to 4.5 kΩA4August 2004Add Lead free part numberF1July 2005Replaced diagrams showing OVFL functionality (see Figures 20 and 21)Replaced Figures 13, 15, 18 and 19 to demonstrate pre-emption of the MSB. Increased maximum digital current (ID) specification at 5 V from 43 mA to 46 mA. .
Old 20th June 2019
  #435
RTG
Gear Maniac
 

Curious if anyone knows the chips used in the Arturia Audiofuse.
Old 25th September 2019
  #436
Here for the gear
 

Roland Quad-Capture UA-55 -> AK4620B(VF-E2)

https://www.akm.com/akm/en/file/datasheet/AK4620BVF.pdf
Old 4th October 2019
  #437
Quote:
Originally Posted by cinstance View Post
Yes, it might just be a rebranded DAC. Here is a picture of it:
It is an Apogee branded ES9018S, the same 8ch DCA used in the MK II as well.
Old 4th October 2019
  #438
Lives for gear
 
s wave's Avatar
Quote:
Originally Posted by Category 5 View Post
It is an Apogee branded ES9018S, the same 8ch DCA used in the MK II as well.
is that the MKll of apollo / steinberg or tascam? thx
Old 5th October 2019
  #439
Quote:
Originally Posted by s wave View Post
is that the MKll of apollo / steinberg or tascam? thx
I was referring to Symphony Mk II. It uses the same IC, but they no longer brand it as an Apogee chip.

Apogee is largely responsible for the overwhelming popularity of the Sabre DACs. AKMs latest chips are as good, or better, but not nearly as universally praised by end users for some reason.
Old 11th October 2019
  #440
Gear Maniac
Quote:
Originally Posted by ardis View Post
Do you think maybe a future proofing, where 32 bits may be something the industry moves into?
Very late reply here, but why not... The I2S interface that's usually used to interface to converter IC is inherently arbitrary bit depth since it's most significant bit first. Whichever side is the master just picks the number of bits and any extra bits at the tail end (if the interface bitdepth is higher than the actual bitdepth) are just set to zero / ignored.

Properly dithered 24 bits is always going to be enough since that's pretty much at the limit of what's physically possible due to the inherent thermal noise of any real world resistances (whether they be actual resistors or other components with resistive behavior). "32 bits" is purely marketing speak. The internal processing path inside a converter is varying bitdepth in any case since different parts inside an oversampling delta-sigma-converter require different bitdepth vs samplerate.
Old 11th October 2019
  #441
Lives for gear
 
Lady Gaia's Avatar
32-bit resolution at I/O is either meaningless, or essentially the same as 24-bit resolution behind a pre-amp with a 48dB range (or equivalent attenuation range on output), depending on how you look at it. You could make a reasonable argument that both are true. As an intermediate representation in a purely digital environment, on the other hand, it’s a completely different story. A 32-bit floating point representation effectively eliminates any concerns about gain staging in a digital processing chain.
Old 4 weeks ago
  #442
Gear Nut
 

What about Audient Sumo?
Old 5 days ago
  #443
byz
Here for the gear
 

What about Audient iD4 converters and opamps?

Last edited by byz; 4 days ago at 08:38 AM..
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Forum Jump
Forum Jump