EMM Lab DCC2 & CDSD Better connection?


For the EMM Lab CDSD & DCC2 which BETTER Connection for used maximum PERFORMANCE for 2 Chanal CD.
DCC2 used DST(BNC or ST)? or ANALOG (XLR or RCA)? or PCM (AES or COAX or TOS or ST)?

Which Cables and connectors NEED? only for 2 chanal CD and for SACD 2 chanal?

Thank you for your help.
mehdi
I have found that locking the signal of the EMM DAC 6e w. the Philips to the Philips produced much better sound than slaving the Philips to the EMM DAC 6e. I then thought maybe the ST cables were defective. I then did compare them w. the new cables supplied w. my CDSD and found that the new cables were much better but still the sound was better w. the SACD 1000 as a master to the DAC 6e. Still the CDSD is much better than the SACD 1000 on CD. Not on SACD. In the case of SACD it is only slightly better. The CDSD upsamples SACD to 2 times & outputs it to the DAC 6e. The CDSD upsamples CD to 4 times & outputs it. I didn't ask EMM if this is via the only select inputs or if the SACD 1000 does the same. They mentioned w. the CDSD they were able to design a transport from the ground up to do exactly what they want it to do. I asked them what? They replied in reference to the oversampling. I assume the SACD 1000 does not oversample the outputed digital signal. I also know that they can do a ROM update that will change the sampling algorithm on the CDSD.

I never compared the AES/EBU to the ST cables but they are very emphatic about the glass cables being superior. They also felt using any gels etc. can ruin the sound.

I wonder if the oversampled signal is output via the AES/EBU output. If not then this could explain the difference & your dislike for the oversampled signal.

Also running the AES/EBU on the Philips SACD 1000 elminates the ability to slave the SACD 1000 to the DAC 6e. This might be a similar change to my experience. It might be that the sound w. the SACD 1000 running as the Master is superior. This again might also be the situation w. the CDSD but I haven't experimented here yet. Maybe I will, but I am spinning vinyl these days.
Dgad,
Thank you for your reply. I hope to get a CDSD in very soon. I'll be sure to experiment with the cables.
Play w. the switches & set the SACD 1000 to Master instead of slave & post your responses. I am curious if I am alone in my observations. I had some musician friends over who confirmed my result. I am not sure if this is specific to my unit alone or did everyone experience this.
Elberoth2,

Thank you for the feedback! I knew it will be the case.

Someone said you can't externally sync when AES is used. I don't believe that is true because, even if the glass clock is not working, you can use the BNC clocking interface with the AES and experiment syncing the transport to the DAC or vice versa.

Regards,
Alex
I have only tried CDSD - stock optical cable - DCC2 and slaving to either CDSD or DCC2...at least in these terms, slaving to DCC2 is faaaaar better. Not even close.