The reason I am asking is since Pass and Wayne selected to remove degeneration resistors and use CFM at the source and experienced better dynamics together with better slew rate*, that this would be a good indication of what to do 
But like you have mentioned several times, one have to look at the whole circuit and I understand that. It would be fun to learn the difference of VFB and CFB on the same amplifier, so I guess the only way to find out is to do.
* 100V/mS for INT-25 and XA25 compared to 50V/mS for X and XA.8 series
But like you have mentioned several times, one have to look at the whole circuit and I understand that. It would be fun to learn the difference of VFB and CFB on the same amplifier, so I guess the only way to find out is to do.
* 100V/mS for INT-25 and XA25 compared to 50V/mS for X and XA.8 series
I just want to interject with something.
The Onsemi NSVJ3910 (Vds 25V) is $0.43 per pcs if you order 100 pcs via Mouser. With those prices, you can match to 1% tightness.
That was it, that was all I had to say. Now back to the regular schedule
Read the full JFet noise test report: https://www.mvaudiolabs.com/diy/modern-jfet-noise-measurements/#setup
The Onsemi NSVJ3910 (Vds 25V) is $0.43 per pcs if you order 100 pcs via Mouser. With those prices, you can match to 1% tightness.
That was it, that was all I had to say. Now back to the regular schedule
A tiny follow up on post #529. The NSVJ6904DSB6 is a dual package of the NSVJ3910. The NSVJ6904 is a CASE 318BD, this happen to be the same size as the 2SK2145BL in the 2-3L1C package. The difference is the source connection, which is flipped on the NSVJ6904, which is sad since it would have been drop in. Such is life. The fix is easy. Use a 6-pin footprint and create an extra track such that Pin-2 of the Toshiba/LSK part is connected to Pin-5 of the Onsemi part, now both packages have a source trace available.
