With HEXFETs source resistors are absolutely essential for local thermal feedback and degeneration. It will not be stable without.
For laterals source resistor is not essential but still desirable. All devices have very non linear output resistance and a small amount of source or emitter resistance swamps it very effectively. The waste power is very small, for 0.22 ohms it is less than a watt at any point in time.
It does reduce transconductance and damping factor but in a decent implementation both are fixable. Yo can use diodes across them to bypass them at higher currents, for example. There is a suitable example on Rod Eliot's site, and the F5 Turbo uses a similar technique.
On distortion figures, everybody is free to set their own performance requirements and standards. At the same time, it is my experience that you can hear differences in sound between amplifiers which exhibit distortion below 0.1%. Not all of it is because of THD itself, but THD is a poor indicator of an amplifier sound.
Since you are aiming for the lowest possible distortion, you should really read e Self and Cordell books to understand sources of distortion and how they are minimised. Also the First Watt site has some interesting reading on audibility of distortion, and there are a number of threads here, some quite contentious.
The 1058/162 devices are available to us on special order. Renesas is still making these and there are plenty of equivalents from Exicon and a few others. Don't buy off eBay. All of that is fake.
For laterals source resistor is not essential but still desirable. All devices have very non linear output resistance and a small amount of source or emitter resistance swamps it very effectively. The waste power is very small, for 0.22 ohms it is less than a watt at any point in time.
It does reduce transconductance and damping factor but in a decent implementation both are fixable. Yo can use diodes across them to bypass them at higher currents, for example. There is a suitable example on Rod Eliot's site, and the F5 Turbo uses a similar technique.
On distortion figures, everybody is free to set their own performance requirements and standards. At the same time, it is my experience that you can hear differences in sound between amplifiers which exhibit distortion below 0.1%. Not all of it is because of THD itself, but THD is a poor indicator of an amplifier sound.
Since you are aiming for the lowest possible distortion, you should really read e Self and Cordell books to understand sources of distortion and how they are minimised. Also the First Watt site has some interesting reading on audibility of distortion, and there are a number of threads here, some quite contentious.
The 1058/162 devices are available to us on special order. Renesas is still making these and there are plenty of equivalents from Exicon and a few others. Don't buy off eBay. All of that is fake.