Quote:
Originally Posted by 2fastdriving
Ivan knows best.
I left on 10mhz for a few days and didn't listen to it.
Tonight... Wow! Sounds incredible. I couldn't really hear a difference on the first few days. Now I can switch the clock in and out and there's a clear difference.
I'm hearing a bigger soundstage. More detail. More dynamics.
Everyone says to leave this thing on 24/7 but it worries me... I live in an area prone to power outages. I had a quick chat with Dan and he helped me pick out an inexpensive UPS for just my clock, nothing else. I'll leave it running all the time and not need to worry.
For the record, I tried a bunch of different clock settings and even changed my normal k01x filters... But after Ivan said what he did above, I set it back to my favorite filter and upsample options and 10mhz clock. I'm keeping it this way for a while. Sounds incredible, and I finally think I'm in the SOTA digital realm!
|
Congratulations again on the new clock! Another vote for 10 mHz being 'the' best approach to clocking based upon my testing and listening over many years and quite a few clocks...
Any clock with an oven-stabilized oscillator of any kind, either Quartz (OCXO) or Rubidium must (my opinion) be left on 24/7. As it is, these clocks take at least 10-30 minutes to truly reach thermal equilibrium and initial good performance levels. Based upon my listening and owning the Esoteric G-0s (Rb type), G-03x (OCXO type) and now on my 2nd owned Cybershaft (all OCXO) with testing of 3 additional Cybershaft units (1 PROD and 2 pre-PROD prototypes, all OCXO), it's even more than that. Once the clock's OCXO reaches initial thermal equilibrium, it continues to mature for a few hours to where it will definitely sound better later that day and the next until it hits its true steady-state best behavior. When I power on my system for some time now, I will always turn on power conditioners and then a few minutes later, the clock and then walk away until well after the OCXO oven monitor's LED on the front lights up and the clock can get moving towards it equilibrium. When I am a bit more patient, I leave the clock run for 1-2 hours before I power up anything else and let the D-02 and P-02 achieve master-sync. When I do, I notice that the components achieve sync in 1-3 seconds versus a longer period (10-15 seconds) when I am more impatient and only wait for the oven monitor to light up.
When its all said and done though, it's all about the sound; my system will sound better many hours after the clock is on and everything is giving time to recharge and stabilize.
In the studio realm or in high-tech video or microwave/other communications infrastructures where master clocking is used extensively, they virtually never turn the clocks off. Please bear that in mind...
My 2 cents...
There was another question here or on another thread regarding using of a GPS-disciplined oscillator instead of another master clock technology. GPSDO is useful when you have a much lower precision OCXO or other mechanism; then GPSDO or other 1pps master-sync regulators are of use. For my money, I would not strap a GPSDO implementation to an Esoteric G-02X or G-01 or other word clock; I'd go for a full-blown 10 mHz OCXO-based approach with the lowest phase noise possible (and therefore lowest jitter) to provide a true 10 mHz master. There are several great choices in the industry for this,....
With respect to the Grandioso G1 which is all 10 mHz to begin with, I don't think it is necessary to have an external 10 mHz master clock. It would be useful to see the actual phase noise and Allen Variance data for the G1 (and G-01) because PPM/PPB is the red herring in the clock industry and it matters little if the phase noise is high at 1, 10 and 100, but that notwithstanding, you would probably need one hell of a 10 mHz master clock to truly make an argument for strapping a master clock onto what is already itself a perfectly good master implementation (G1).
Again my 2 cents....