I switched over to attenuators a while ago, although I have no experience with Gold Point I can say they are a distinct improvement over pots - if you're not interested in the techy speccy stuff then probably best stop reading now ;-).
I recently ran some tests on a brand new "decent" 10k log pot out of curiosity and found that at -18dB the left channel was down 2dB on the right (that difference was worst at -48dB registering a massive 25dB difference). The stepped attenuator matches left/right better than 0.05dB right through the range down to -68dB. I also calculated that THD+N was around 5dB worse going through just one pot than it was going through my whole preamp. I use Vishay enpw 0.1% resistors but they're SMD the equivalent through hole version would be Vishay RN series - bulk foil look attractive but you're way down the route of diminishing returns if you go there.
As for the capacitors it's important to know their function in the circuit, if you replace the RIAA stage caps you need to either get close tolerance or be prepared to over order and measure their actual value - I wouldn't go for any more than 1% deviation from spec.
I recently ran some tests on a brand new "decent" 10k log pot out of curiosity and found that at -18dB the left channel was down 2dB on the right (that difference was worst at -48dB registering a massive 25dB difference). The stepped attenuator matches left/right better than 0.05dB right through the range down to -68dB. I also calculated that THD+N was around 5dB worse going through just one pot than it was going through my whole preamp. I use Vishay enpw 0.1% resistors but they're SMD the equivalent through hole version would be Vishay RN series - bulk foil look attractive but you're way down the route of diminishing returns if you go there.
As for the capacitors it's important to know their function in the circuit, if you replace the RIAA stage caps you need to either get close tolerance or be prepared to over order and measure their actual value - I wouldn't go for any more than 1% deviation from spec.