Have tried
I think you may be correct. Tried setting both to channel 1 and both to channel 2. But will play with this.
When I was not getting a signal back due to Apple OS issue must have clicked every box in Focusrite control and REW. There is routing in both.
The FC shows I am banging out some levels. May be seeing crosstalk.
I think you may be correct. Tried setting both to channel 1 and both to channel 2. But will play with this.
When I was not getting a signal back due to Apple OS issue must have clicked every box in Focusrite control and REW. There is routing in both.
The FC shows I am banging out some levels. May be seeing crosstalk.
This is what it looks like from my iPhone screen - which is where I usually write my posts from.
Attachments
Re posts 121, 122 and 125 I agree that this is not a browser property thread and a mod
may easily remove related posts later.
Using my old Apple IPad, the preferred device for spare time:
Post 1 is dominated by line length and so the scale is reduced until characters are less
than about 1 mm in size. Pictures are large, but not too large and actually reduce in size
if I click on them. All other posts on this first thread page have the same small size also.
However if I re-load the page the scale is like it should be. Coming back from a
subsequent page - wrong size again.
Using my main computer with Linux and Firefox browser :
Line length dominates and is longer than the wide screen.
Diagrams are larger than screen size and again reduce to good screen size when I click
on them.
After checking and re-loading a few time when writing this, the representation on this
main computer is correct, but I did not change anything. Wait for some minutes: again
wrong size, re-loading works.
may easily remove related posts later.
Using my old Apple IPad, the preferred device for spare time:
Post 1 is dominated by line length and so the scale is reduced until characters are less
than about 1 mm in size. Pictures are large, but not too large and actually reduce in size
if I click on them. All other posts on this first thread page have the same small size also.
However if I re-load the page the scale is like it should be. Coming back from a
subsequent page - wrong size again.
Using my main computer with Linux and Firefox browser :
Line length dominates and is longer than the wide screen.
Diagrams are larger than screen size and again reduce to good screen size when I click
on them.
After checking and re-loading a few time when writing this, the representation on this
main computer is correct, but I did not change anything. Wait for some minutes: again
wrong size, re-loading works.
Just tried accessing the first page again myself - needed to reload two times to have it displayed correctly.
If I access the first page from the link in the top of this "Reply" page, it shows up correctly the first time.
If I access it from the tread pges navigation panel, need to reload one time.
Weird.
This is on a Win7 computer using the latest Firefox.
This issue is not only happening with this thread, but often when there are large pictures embedded in the text (also in my own postings).
Maybe a marginal compatibility issue of the forum software with some browsers when embedding pictures that are wider than the text window ?
Also, when I type a posting and hit "Return", I often get two line breaks added to the text in my post ... usually editing them out again when previewing the post.
Best regards,
Claas
If I access the first page from the link in the top of this "Reply" page, it shows up correctly the first time.
If I access it from the tread pges navigation panel, need to reload one time.
Weird.
This is on a Win7 computer using the latest Firefox.
This issue is not only happening with this thread, but often when there are large pictures embedded in the text (also in my own postings).
Maybe a marginal compatibility issue of the forum software with some browsers when embedding pictures that are wider than the text window ?
Also, when I type a posting and hit "Return", I often get two line breaks added to the text in my post ... usually editing them out again when previewing the post.
Best regards,
Claas
Can someone check a loopback using a square wave and the o-scope? I'm getting some pretty horrible results, and not sure if it's the audio interface (Behringer UM2) or something else. I had first used it to test my amp and thought it was bad w/ some oscillation and/or other design flaws. Then when I tested just the device, I got similar results. Here are 100hz, 1khz, 10khz, and 20khz square waves.
100hz
1khz
10khz
20khz
100hz
1khz
10khz
20khz
Ok, I had a chance to watch the video and it was very informative. It was nice to learn what that "dither" box I had checked was doing too! This would explain the apparent "ringing" that I was seeing on the lower frequencies, but does not explain the 10khz or 20khz trace that looks like a sine wave. Or does it?
Clearly the square wave performance of this device is poor too (as evidenced by the 100hz trace). Does that explain 10k and 20k? If so, how can it still have a ruler flat frequency response from 20-20k? Do square waves show another aspect of the design like transient response or something like that and are independent of the actual frequency response?
Clearly the square wave performance of this device is poor too (as evidenced by the 100hz trace). Does that explain 10k and 20k? If so, how can it still have a ruler flat frequency response from 20-20k? Do square waves show another aspect of the design like transient response or something like that and are independent of the actual frequency response?
Actually, I guess this would totally explain the 10k appearing as a sine wave. But how about that mess that is the 20k? Does that mean there isn't even a 20k fundamental due to the sample rate? What is the frequency range of 16bit 48k? I could have sworn this was flat out to 20k when I tested it.
I'm not sure what you are asking exactly... What is the sample rate of the scope pics? I was pointing you to the Gibbs Effect (Gibbs phenomenon - Wikipedia and Gibbs Effect) as this explains the square wave shape you observed. The slope on the 100Hz square wave is because of LF attenuation (it's probably AC coupled).
I'm not sure what the 20KHz pic is showing - if that was a square wave then a 20KHz square wave would only have 1 or 2 partials in the pass band before the antialiasing filter if it was sampled at 48KHz. Your interface probably does have a flat frequency response in the audio band. Square waves have harmonics to well over a decade beyond the fundamental - that's what makes them square.
I'm not sure what the 20KHz pic is showing - if that was a square wave then a 20KHz square wave would only have 1 or 2 partials in the pass band before the antialiasing filter if it was sampled at 48KHz. Your interface probably does have a flat frequency response in the audio band. Square waves have harmonics to well over a decade beyond the fundamental - that's what makes them square.
Well...I kind of answered my own question about the 10khz appearing as a sine wave (due to lack of the upper harmonics RE: the Gibbs Effect). But as you point out, the 20khz should only have the fundamental, but it looks super wonky.
As for the 100hz, that is what I don't understand. If the frequency response is flat to 20hz, how can 100hz be attenuated? I thought seeing poor square waves on the low frequency indicated a rolloff. How can you have poor square wave performance *and* a flat FR?
And are you saying that the interface is AC coupled, and the cap isn't large enough for a low enough rolloff?
Oh, and I'm running the interface at 48khz sample rate.
Thanks!!
As for the 100hz, that is what I don't understand. If the frequency response is flat to 20hz, how can 100hz be attenuated? I thought seeing poor square waves on the low frequency indicated a rolloff. How can you have poor square wave performance *and* a flat FR?
And are you saying that the interface is AC coupled, and the cap isn't large enough for a low enough rolloff?
Oh, and I'm running the interface at 48khz sample rate.
Thanks!!
RTA Testing
Mine is sorted out. System used is;
Late 2012 Mac Mini
iPower 12 volt supply for interface
Focusrite Clarett 4Pre interface - one out of box 1 hour, the other well used maybe 1000 hours
Played with REW signal generator and the Akitka 2ppm.
First ran the test around -20 dB, later reran close to the -90.3 dB level used by Stereophile.
Varied the generator and the cable, one copper TB cable and a Corning fiber optic. Hoped to see something from the Corning, but at least with my resolution the two cables seem identical.
Did find one issue. When testing the listening room set up tyhe test show a 60 hz bump around -140 dB. Suspect it is the 500 VA isolation transformer positioned a few feet below the interface. Will eliminate this.
Mine is sorted out. System used is;
Late 2012 Mac Mini
iPower 12 volt supply for interface
Focusrite Clarett 4Pre interface - one out of box 1 hour, the other well used maybe 1000 hours
Played with REW signal generator and the Akitka 2ppm.
First ran the test around -20 dB, later reran close to the -90.3 dB level used by Stereophile.
Varied the generator and the cable, one copper TB cable and a Corning fiber optic. Hoped to see something from the Corning, but at least with my resolution the two cables seem identical.
Did find one issue. When testing the listening room set up tyhe test show a 60 hz bump around -140 dB. Suspect it is the 500 VA isolation transformer positioned a few feet below the interface. Will eliminate this.
