AWG or sound sampling device?

Jump to Latest
NickKUK

NickKUK

Member
Joined 2019
#1
I'm looking for next suggestions on my audio journey so I thought I'd pick some experienced and brilliant minds (is that enough sucking up? 😀). I'm trying to decide which would be more useful and what options I have in those fields.

For an ARB I want to use it for audio (including measuring noise/THD), RF (AM/FM) tuning, and also DSD (preferably up to 512, ie 24MHz). Budget ~£300-400. Bonus for USB/LAN connection. The natural selection would be the Silent SDG1000X which would work with the SDS1104-E for bode plots where currently I have a python program allowing me to use the Mac mini as a bode plot sine generator between ~4-20KHz using the headphone output.

For a sound sampling input device it would need to work with a Mac mini (USB/ethernet/thunderwhatsitsface) or in a Linux VM with a preference for open source for working in VMs and to counter the Apple 3 generation support. I used to have an Audiophile firewire but that's dead as a doornail.
There's not really a large range of USB sampling devices for the Mac and what there is tends to be overpriced for what they are.
 
#2
So reading though the the marketing blurb, the focusrite scarlett 2i2gen3 seems to tick the boxes:
  • 24bit 192khz
  • 2x balanced TRS inputs
  • 2x balanced TRS outputs
  • connects to mac via usb without drivers - making it great for linuxVMs running REW
  • £149.

I did look at the spdif in/out clarett but at £249 it’s starting to get too much.
 
#3
I have the focusrite solo. Unsure of its architecture; I know it does stereo out, but is the mic xlr input "left" and the trs input "right"? Or can they be assigned that way...

Its important to have stereo I/O in REW so it can do the gated measurements, intended to capture speaker FR before room effects muddle the measurement.

I realize it cant do the RF bands...
 
#5
jjasniew said:
I have the focusrite solo. Unsure of its architecture; I know it does stereo out, but is the mic xlr input "left" and the trs input "right"? Or can they be assigned that way...

Its important to have stereo I/O in REW so it can do the gated measurements, intended to capture speaker FR before room effects muddle the measurement.

I realize it cant do the RF bands...
Click to expand...

I had a look at the solo but the mic input seemed hard coded when i read the guide. It supposedly has balanced input on music and the balanced outputs L&R.

I was thinking that cross talk measurements may warrant 2x input and 2x outputs. So the reason looking at the 2i2 was that second input was a balanced combo.
 
#6
mbrennwa said:
If I'd need a sound interface for audio testing, I'd look at an RME DA/AD unit. For example this:
https://www.rme-audio.de/de_adi-2-pro-fs-be.html

These things are a bit more expensive than the usual suspects, but I guess they are also a bit better.
Click to expand...

I tend to be a little touchy given the speed of OSX driver models going out support, hence a main usb profile works well. I’m aware of the custom usb profile used to control settings etc so the more standard it is the better.
I’ll have a look into the RME.
 
#9
oops, does SDS1104-E really Ok for your THD+N/SNR tests? Usually, if the scope resolution is 8bits manufacturers don't mention that. So, SDS1104-E will have about 40db SNR, is that what are you looking for? Scarlett, whatever the particular model, suppose to be 16bits at least, just add there some notch and measure THD/THD+N as low as you like, but an 8bits scope is no way 😵
 
#10
IVX said:
oops, does SDS1104-E really Ok for your THD+N/SNR tests? Usually, if the scope resolution is 8bits manufacturers don't mention that. So, SDS1104-E will have about 40db SNR, is that what are you looking for? Scarlett, whatever the particular model, suppose to be 16bits at least, just add there some notch and measure THD/THD+N as low as you like, but an 8bits scope is no way 😵
Click to expand...

Precisely why i’m looking for more bits. If it was static i could use a 32bit 1MSPS dac but the best SNR comes in at 62SPS..
Ideally as much bits as possible barring the horrendious EMI i have in my man cave!
 
#12
IVX said:
So you don't need any new interface if already have Scarlett, right? When I had need measure THD+N -115-120db I prepared a trivial notch filter($10 + my time) and used that with my old $80 Asus Sonar U7.
Click to expand...
So that’s my thinking - the scarlett would provide 24bit ~100dB balanced input and output. The RME would provide ~124dB IIRC with all the bells and whistles.

Theres lots of PC options but USB/mac balanced i/o less so. Plus there the marketing BS hype with any Mac targeted gear.
 
#17
NickKUK said:
Theres lots of PC options but USB/mac balanced i/o less so. Plus there the marketing BS hype with any Mac targeted gear.
Click to expand...
Not sure what you mean by this. MacOS has built-in support for standard USB audio. You don't need any fancy drivers and stuff, except for the most exotic audio interfaces that don't work according the USB Audio standards. The same can be said for Linux.

Maybe the reason why you'll see more "Windows supported" tags is because that seems necessary to mention explicitly for Windows, but it is not for MacOS and Linux.
 
#18
mbrennwa said:
Not sure what you mean by this. MacOS has built-in support for standard USB audio. You don't need any fancy drivers and stuff, except for the most exotic audio interfaces that don't work according the USB Audio standards. The same can be said for Linux.

Maybe the reason why you'll see more "Windows supported" tags is because that seems necessary to mention explicitly for Windows, but it is not for MacOS and Linux.
Click to expand...

Usb sound vs pci. I’ve written usb drivers on osx. I agree the usb cards that support the usb audio profile “just work”. Nowadays there’s also more support for direct USB without going through the kernel in the audio framework.
 
#20
The CP2114 supports only 48 kHz, 16 bit digital audio. This is typically not an issue for source USB audio as the device capabilities are reported to the host and any sample rate conversion (for say 44.1 kHz audio) is done automatically by the host. However, some audio devices may require 24-bit digital audio data on the I2S data stream. In this case, the CP2114 will send the useful 16-bit audio to the connected audio device on SDOUT in the most significant 16 bits and pad the remaining 8 bits of data with 0s. Likewise the CP2114 will read the MSB 16 bits of data on DIN and throw out the LSB 8 bits from SDIN. The CP2114 can be configured in 16 bit or 24 bit mode via a configuration option in the CP2114 one-time programmable ROM.
Click to expand...

And the award for the most useless audio usb bridge goes to…

24bit i/o then only exports 16bit 48khz via USB… top 16bits then zeros the remaining bits of the 24 sent.. speachless
 
You must log in or register to reply here.
Top Bottom