Left and Right I/O Are Reversed In v1.100

I just upgraded to the v1.100 software and the left and right inputs/outputs are swapped. Loop the left channel to the left input and it shows when the right display button is set and vice versa.
Also, the right (left?) gain readout is behaving strangely, indicating -78dB or so when it should be 0dB.
:unamused:

Funny you mention that. I am using v.9999 and noticed the same thing about the Left and Right inputs being reversed about a 1/2hr ago. I thought there was something wrong with my preamp at first, but then connected a signal from another signal generator and verified it. I did not experience the gain readouts being off. UPDATE- I double checked my measurement since I had been doing an Automated AMPFreq Chirp measurement using the right channel as a reference so I thought I would try it using the Left channel as the reference (cable wise), and it gave me poor results. I then exited out of the program and came back in and re-measured the FR using the Expo Chirp from the main menu and the channels are correct now…

Now that’s weird - v0.9999 worked fine for me.

I did just now launch the application again and now the acquision is very slow (64k FFT), seems slower than with v0.9999.

see my edited post above…

I did the automated FR chirp test, re-booted and the channels were still reversed.

Hi @Rammis, there’s a bug where the channels can get flipped at higher sample rate. It very rarely happens at 48K, and is more likely to happen at 192K. I think the issue is that once the data streams get put into a hardware fifo (internal to the chip) the ordering is lost and when sample rates are changed, it’s unknown how many samples are left in the fifo. If it’s an even number, then there’s no flip, but if it’s an odd number, then there’s a flip. So, when you change sample rates from 48K, double check things aren’t flipped. Hopefully this is understood soon.

Is that the same for ver .9999 ? I was indeed playing with sample rates yesterday and I have know doubt that my inputs were reversed for some period of my testing. I noticed it 1st when adjusting the balance pot of the preamp and things were changing opposite of what I expected. Then a little later I was getting noise on one channel after switch to a different preamp input- the kind caused by a poor cable connection. When I moved the cable for the right channel by, the noisy left channel changed. Then I removed the left cable input and the Right channel disappeared. Then I hooked my sig gen to the each input and verified they were reversed. It was not too late at night for my mind to be playing tricks :thinking:

I never had that happen with v0.9999, and I use the highest sampling rate. It happened immediately and constantly with v1.100. I went back to v0.9999 and the problem dissipared.

HI @Rammis, you are indeed correct. This is a regression with 1.1. You have to hit the sample rate button twice to make it stick. We’ll get a release out on Monday or sooner.

Hi @Rammis, fixed in 1.110

The swap between left and right channels is still there at 192k sampling rate, but seems OK at lower rates. Also, the Right ‘Gain dB’ readout is broken at all sample rates (the left is OK).

Good find, thanks very much! The gain for the right channel depended on the left channel peak, so if the left channel peak goes away, the gain is wrong. For 1.12 that will be changed, along with 192K swap.

Hi @Rammis, I just verified it appears fixed in 1.12: