It’s correct that I hide a 10MHz clock that helps the Vivaldi Clock do an even better job, but not correct that I value upsampling.
I heard some very, very interesting comparisons with @PAR and am pretty sure I would choose to run my DAC straight through were it an option with the Rossini.
Why have the Rossini, then? Mixture of box count and funds. Mostly funds, though — I’ll find the space for the boxes if the Vivaldi replacement is also three boxes. (I don’t want an optical drive.) But the funds part really will hurt, and not just for the boxes themselves. The cables too
You’re in a very good place, Ben, and it will only get better. Skynda långsamt as they say.
Found myself reading Serinus’ review of the NHB-108 last night before I turned in. Damn it man, do you have any idea what a bad influence you are? My mummy says I’m not allowed to play with you any more!
And I’m just the hangaround in short pants with the Bartók who’s temporarily allowed to hang with the cool kids! But when I’m bigger I know who I wanna be!
The engineer in me is sceptical to the idea of clocks clocking clocks (and while one of the courses at uni was real time systems, that was a helluva long time ago and I can’t recall with total certainty if I made all the lectures).
But bear with me. Imagine the DAC is a wheel and the clock is a precision motor whose job is to turn the wheel at a very precise speed. And let’s just say that precision motor is very good indeed. How do you improve on it? A master clock is like a second motor (which needs to be even higher precision to add any value) connected to the first one with some kind of transmission. Let’s call this the “active” approach. An alternative solution is to have a bloody great big flywheel (PLL), let’s call this the “passive” approach. While I can see the logic in the active approach I can also see lots of possible problems that the passive approach very elegantly avoids**. I find it hard to convince myself that this will result in a net improvement.
But I assume you have listened and compared fairly carefully? Or was the master clock bought on an article of faith? If you’ve heard the improvement I am not going to make a fool of myself trying to prove that you didn’t with theoretical arguments! But I am curious. And didn’t I read that @anupc works or has worked with something in this area? What do you say?
Maybe this has all been microwaved to a crisp in previous threads in which case please just point me at them. Still playing catch-up here.
** It’s like the story of the space pen. They spent a lot of time and effort inventing a pen that would write in zero gravity only to fail and go back to using pencils. Sometimes elegant is best.
Your wheel with it motor is running, and next to it is a second wheel with the same motor. The chances that both will spin 100% and keep doing that is very slim.
Now add a 3rd wheel or 4th and the difference can get even greater.
But add a device that locks all the wheels together so they now all turn at the same rate, this is what a clock does.
It just gives all the clocks in the system the same clock signal to lock on to.
The vivaldi clock in the clock is the same as whats in the dac, upsampler, and transport. As said it just sends a signal to each, to which each then lock on to and then each work the same.
This is how i understand it.
Not sure whom you were replying to there Duncan. My comment was to Ben’s post where he is using a master clock to clock his Vivaldi word clock which is clocking his Rossini. So this is about improving clock accuracy, not (multi-)clock synchronisation.
I am still interested in your actual observations @all2ofme. While I trust James’s theoretical arguments a tad more than my own they are still no substitute for thorough empirical testing (which in fairness someone at dCS may well also have done).
Yes. What @keiserrg said. What i believe James said if you read it all carefully, is a more accurate clock could theoretically improve the dcs master, but not likely in its own arena ( the short term). I believe what is inferred is a clock reclocking in the long term at least in theory could make improvements. One of the biggest criticisms i notice on vivaldi by users is too many boxes. I dont have that issue… but would enough consumers go for an additional external box making a 5 box system by dcs. I think dcs has made their decision based on gain vs economics but left the door open for the person who wants to find out by giving instructions and a back door to add this extra hi precision clock . I think there are at least 4 or 5 of us who have in fact taken the plunge and not looked back.
Adding the SRS Perf10 was nowhere near the improvement that adding the Vivaldi Clock was. Nor was it as big a leap as swapping a Rossini Clock for the Vivaldi Clock. I accept that I bought it sound unheard and therefore some bias was involved, but gosh golly it MUST have been worth something to the ears because the eyes…well, they protested.
If I ever have the chance I’ll compare the CH Precision T1 to the Perf10. And if they’re close, I’ll cough up and promote my reference clock out of its paper bag/Kallax cupboard
I can put cold, hard financial figures on my evil influence
As @glevethan has said here, I was at least partly responsible for his recent shopping spree. Mind you, he got me into some fun footwear, so we’re even. Cough.
If I were darTZeel I’d market their wares as a Diana Krall/audiophile favourite track escape valve. I sit down to a session with my gold beasties and…just listen. I don’t bounce around things that sound good. I let Roon and my fat forefinger take me where they will. Happy passenger.
Might be fun to give it a spin, but I was just taking a look at the T1’s spec, I doubt it’ll better your SRS considering it has a frequency deviation PPM specification that’s 400 times worse than your Perf10
