• 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.

Audio interface (recording) suggestion

garbulky

Addicted to Fun and Learning
Joined
Feb 14, 2018
Messages
596
Likes
120
#21
I may have skipped over this. You haven't mentioned what is the context are you doing recording in?
For instance for me budget, portability, convenience, and a self powered standalone device and features is very important for recording. Though I have heard better recording using other gear, my need for the above wins out every time.
So I use a Tascam DR70D MK2. It's fantastic for what I do - recording that needs little setup and time to get to pressing the "record" button. I also use an adapter to use the line out to simultaneously record in stereo to an iphone for live stereo recording of music videos. it increases noise slightly but is worth it for the convenience.

I also have a significantly more expensive Roland DAW complete with self powered sliders, 16 channel recording, and all that meant for a real studio. It gets no use due to it not matching my needs.
 

Fone

Member
Joined
Dec 21, 2018
Messages
30
Likes
26
#25
As GARBULKY notes above, you need to nail down exactly what you want to do with the interface. Then narrow down your selection based on "required" features. There are a lot of options but price & lack of certain features will narrow down your choices significantly.

There is an excellent "low latency database" for audio interfaces produced by TAFKAT you might review:

https://www.gearslutz.com/board/mus...erface-low-latency-performance-data-base.html

Low latency can be important to people who are playing live or practicing live at home with software-PC "virtual instruments". It is also an indicator of the quality of drivers and how much effort the company put into making the interface. For absolute low latency performance, the best PCIe interfaces outperform say ThunderBolt or USB. But it is all in the implementation. On the other hand, some audio interfaces have line-outs that allow you to hear analogue sounds without delay; that would be helpful for monitoring "live sound directly from mics or instruments". That is different from low-latency.

To be clear, low latency does not mean the recording path is highest quality; that is not the purpose of the "low latency database".

Traditionally just a couple of third-party software makers produced the drivers for most audio interfaces. RME does them in-house. I think Zoom does also. Can't remember details.

Also, ThunderBolt is a bit sloppy. Recent laptops will generally have TB3 ports (not TB2). I think all interfaces are TB2. So you will need to find a TB2-TB3 conversion cable that works for your interface-laptop (frustratingly, not all will work). This situation might have changed in recent months and I probably am forgetting some important information. On paper, TB should provide better performance than USB but for the homegamer audio interface, I frankly see no benefit today.

I have a RME BabyFace Pro USB. That is a good choice for the homegamer but is expensive. Forum people say the ADC implementation is of higher quality than the DAC implementation is. I don't have a clue. The DAC is fine. The headphone outs are OK but a bit underpowered in my view. The line-outs are servo-balanced which I guess was a compromise given size & price. The software is rock solid and low-latency performance is superb.
 

RayDunzl

Major Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
5,930
Likes
1,876
Location
Riverview, FL
#26
If you'll zip them (yes I know this doesn't reduce the size) it will let you attach them. So zip the two files together.
I think there's still a file size limit. 2 meg?

1546800812536.png


1546800859620.png
 

mansr

Active Member
Joined
Oct 5, 2018
Messages
241
Likes
465
#29
If you'll zip them (yes I know this doesn't reduce the size) it will let you attach them. So zip the two files together.
I tried that, and it exceeded some unspecified maximum size. My patience was insufficient for further mucking about.
 
Joined
Nov 6, 2018
Messages
87
Likes
66
Location
.de
#30
Now the green one is Steinberg UR242 and the red one is Realtek ALC892's line in recording the headphone output of X-Fi Titanium HD.

Zipped flac attached.
View attachment 19970
Boy, that looks terrible. The UR242 clearly has the better ADC but jitter must be off the charts. No wonder given the lousy oscillator spectrum that @mansr posted. At 12.288 MHz, it is possibly hitting up to -30 dBc/Hz @ 10 kHz, at least -50 dBc/Hz. That's a value that even first-generation PLL synthesizers would be utterly ashamed of. (For a 20 meter receiver tuning 14.something MHz, -99 dBm/Hz @ 10 kHz would be considered lousy and prehistoric.) Makes you wonder whether the PLL is even locking properly. If it is, is has to be a really lousy fractional-N job or something. Otherwise one would have to look into what might be keeping it from operating properly. (Dud sample with bad solder joints?)

This is also giving perspective to clock tuning attempts, however. If it takes a plain terrible reference oscillator for ADC jitter to come close to -80 dBc on a 10 kHz tone, how much is there really to be gained with a regular crystal oscillator that's decently clean to begin with?
 
Last edited:
Joined
Nov 6, 2018
Messages
87
Likes
66
Location
.de
#33
@mansr
So to clarify, these are the oscillator spectra in the two different devices you tested? It didn't come out clearly like that, so I'm a bit confused right now.
 

restorer-john

Major Contributor
Joined
Mar 1, 2018
Messages
1,333
Likes
1,649
Location
Gold Coast, Queensland, Australia
#34
Not as spectacularly good as the DAC, but more than good enough for recording purposes.
How good is the DAC in your testing? And @ 1V in, was that in your experience a semi-sweet spot (for the A/D) or does it get better?

It seems like a bargain for an almost disposable front end for PC instruments.
 

mansr

Active Member
Joined
Oct 5, 2018
Messages
241
Likes
465
#36
@mansr
So to clarify, these are the oscillator spectra in the two different devices you tested? It didn't come out clearly like that, so I'm a bit confused right now.
The top spectrum (the volcano) is at the clock input on the DAC/ADC chips. The bottom one is at the crystal oscillator. Both are from the same Steinberg UR242 device. This shows that the clock synthesiser in the main DSP chip is what is messing things up badly.
 
Joined
Sep 18, 2018
Messages
43
Likes
11
#37
Thanks to all for all your comments. Sorry for the late response but i have a lot of things to do lately.
I may have skipped over this. You haven't mentioned what is the context are you doing recording in?
True i forgot. I will use primarly for room equalization but i will also use it with a microphone for my desk so my main focus is recording.
Maybe in future i will try some studio monitors but for now i'm good regarding speakers, so balanced output is not that important.

Thanks. A bit messy to read tho, he really should make an excel file.


Overall for now my target is the focusrite scarlett 2i2, for the money seems to be the best performing one around 100€ but i will still get more time to get a definitive decision.
 

Fone

Member
Joined
Dec 21, 2018
Messages
30
Likes
26
#38
Overall for now my target is the focusrite scarlett 2i2, for the money seems to be the best performing one around 100€ but i will still get more time to get a definitive decision.
If you are buying used, make sure to get the second-generation model as there were some improvements. I think the outside box was subtly changed also so there is a way to differentiate this.
 
Joined
Sep 18, 2018
Messages
43
Likes
11
#39
If you are buying used, make sure to get the second-generation model as there were some improvements. I think the outside box was subtly changed also so there is a way to differentiate this.
If i will still go for the scarlet 2i2 i will buy it new from Amazon. It is around 120€ here
Thanks for the tip.
 
Joined
Nov 6, 2018
Messages
87
Likes
66
Location
.de
#40
I am using the combination of an internal soundcard (Asus Xonar D1) with a compact mixer (Mackie 402 VLZ4). Both used, ~70€ total (well, plus some cabling I guess). Here's a loopback test showing why:
rmaa-d1-loopback-hd400-402vlz4.png

The HD400 ground loop eliminator is clearly not making crosstalk or (SMPTE) IMD+N any better when confronted with 2 Vrms from a 100 ohm source impedance (cheap isolation transformer thingies do have their limits in the bass region), but look at that dynamic range. And that's at a -3 dBFS input level no less, as nonlinear input impedance was making itself felt beyond this point. Previously I was getting 112.4 dB on a straight loopback (@ -1.6 dBFS in), so there's barely any loss. That's with a ca. +11.5 dB input gain on the mixer and corresponding about -10 dB main mix setting, for internal levels of ca. +16 dBu. (I can run the mixer maybe 1 dB hotter before the clipping indicator comes on, which nets me another 0.4 dB or so.)
At minimum input gain and input channel level at unity I am getting 105.0 dB @48 kHz, and when cranking up channel level (so same levels at tone control but hotter mix bus), it's 106.6 dB. If I crank the input level back up to ~+12.4 dB but run the mix bus low (main mix @ unity for about +6dBu on the mix bus, reduce channel gain to match), I get 107.4 dB.

The Onyx input amplifier on the 402 VLZ4 mixer is really great, you can run it at maximum gain and it'll still have negligible distortion right up to clipping level (which is more than can be said about a Behringer Xenyx Q1002USB). Not to mention a more than decent EIN of about -131 dBu. The rest of the thing isn't quite up to the same high standards - the tone control stage invariably adds some noise (and a bit of 2nd harmonic at high levels), and while mix bus noise is lower than in some (and running the mix bus low still is more acceptable than running the tone control stage low), output driving abilities are nothing special.

At the end of the day it's a bit of an academic exercise, as I wouldn't actually want to produce music with a more than temperamental ASIO driver. Not sure what kind of 2in/2out interface would be a good match though. Audient iD4 and Mackie Onyx Producer would have converters to match, but EIN spec on the Audient iD4 is not particularly special (though of course it might still be better in practice), and the Mackie appears to be a bit noisier than that. Can anyone think of a 2in/2-out (or 4-in/-4-out) interface with an external power supply? I don't think bus powered would be cutting it. The Focusrite Clarett 2Pre may be a good candidate (EIN still a tad higher but arguably better converters in return)... they're calling it a 10-in / 4-out, but realistically speaking it's no more than a souped-up 4x4 or 6x6. Let's see what it costs... 330€, and to my understanding that's not particularly expensive for what you get. Probably a more practical choice than my little cobbled-together setup for most people. I do still have several analog line level inputs that could be used in parallel to the mic ins, but the optional MIDI bracket is not installed here, and neither is there a Toslink input. In any case, there may be more options than you realize.
 
Top Bottom