Howto - Distortion Measurements with REW

I don't have a Focusrite, but I'd try another ASIO driver, e.g. ASIO4ALL or other although I'd expect the Focusrite driver to be a good ASIO implementation.

For Win, some people reported they also need to disable the other non-soundcard interfaces in their settings, e.g. in my case, the internal Realtek in my laptop.
 

LKA

Member
2014-12-28 5:31 pm
As I see my old modified usb card SB1240 is better than many newer ones.
 

Attachments

  • X-Fi-USB-HD-modded-Loopback-1k.png
    X-Fi-USB-HD-modded-Loopback-1k.png
    32.5 KB · Views: 33
  • X-Fi-USB-HD-modded-Loopback-10k.png
    X-Fi-USB-HD-modded-Loopback-10k.png
    59.1 KB · Views: 33
  • X-Fi-USB-HD-modded-Loopback-CCIF.png
    X-Fi-USB-HD-modded-Loopback-CCIF.png
    31.6 KB · Views: 28
  • X-Fi-USB-HD-modded-Viktor1k-9-6d-BFS.png
    X-Fi-USB-HD-modded-Viktor1k-9-6d-BFS.png
    33.4 KB · Views: 27
  • X-Fi-USB-HD-modded-Viktor10k-9-6d-BFS.png
    X-Fi-USB-HD-modded-Viktor10k-9-6d-BFS.png
    31.1 KB · Views: 33
This is an external USB card, X-Fi USB HD (SB1240)

1st improvement
2nd improvement
I replaced 5 opamps
Thanks for sharing your mods. That was quite a bit of work! Having a $99 sound interface that is almost that good is fine for me. I wonder if opamp upgrades inside my 2i4 will improve performance?
 
I'd say check it with a loopback connection or re-check the loopback connection. Also try maxing out the Buffer samples in the ASIO Control Panel. Third thing you can check is your Volume Mixer in Win, then re-check levels in R.E.W.

Without R.E.W., does the audio interface both play back audio and record audio?

I just checked for interference with mains by putting the UMC202HD on top of a DC PSU, and sure enough, bumps at 60, 180, 300Hz.

Tested self-noise measurements with Browser open several tabs + Wi-Fi, then Browser closed + Wi-Fi, then Browser closed and Wi-Fi Off. Measurements seem close to me.
 
To check whether measurements with a smaller frequency range can provide less THD in calculations because my card has some spurious high-freq spikes, I did two measurements: first is 20-20kHz in the measurements window, and second is 10-48kHz. Both are with the card set at 96kHz using the ASIO driver. The respective measurements are below.

Does anyone know why I don't get any calculated figures in the Distortion window for the smaller range of frequencies?

Dist20-20k.jpg


You can see above that the table below the graph, with the Fundamental, Harmonics and THD has no numbers at all.

Now here is the larger frequency range:

Dist10-48k.jpg


Here, we do see the measured and calculated numbers for each entry.

Do you sometimes get these missing numbers? Is it user error?
 
Your graph cursor is placed before the data starts - it is at 19.4 Hz and your measurement starts at 20...
Thanks John, indeed user error and I didn't notice until later last night. I was puzzled for a while because another two set of measurements did show both tables with figures as I hadn't move any cursor that time and that initial set of measurements on reloading didn't show any figures anymore...

Very nifty to compare several measurements in a row like that.

Thanks for this fantastic software, I used it a few years ago for room correction and now for something different and learning new possibilities regularly.