I had a brief audition at the dealer of the Bartok on its own, and then with the Rossini clock. The difference was easily perceptible. To me, the image became more focused, I could pinpoint the location of voices, and could perceive the location of instruments easier. Once the clock was removed from the Bartok, the soundstage seemed to flatten and became less detailed.
Separately, I have been testing adding an alternate external 44.1kHz wordclock signal to the Bartok (not from a dCS Master clock) to see if an even greater improvement in the sound than with the Rossini clock is possible. Still work in progress as it appears the magnitude of any improvement in the sound quality of the Bartok with external clock is dependent on the quality of the external reference clock (I used Cybershaft masterclock).
The reason I am testing this is through my curiosity after reading from another hi-end audio forum that many Vivaldi stack users are finding that adding a very high quality external 10Mhz reference clock to the Vivaldi Clock (which allows for this input) resulted in a significant perceived improvement in sound quality.
Hi @kentyeh, I’d be very interested in hearing any more details you have on your experiments with third party clocks slaved to Cybershaft. I’ve been considering trying the same thing but have found very little information online on whether it’s an option worth pursuing.
Would you mind sharing what clocks you tried between the Cybershaft and the Bartok?
Jeff, I am only just starting on this path of testing non-dCS clocks for the Bartok. So far, I have tried with what equipment I had on hand, a Mutec MC-3+ USB as master clock tied to a Cybershaft OP12 10Mhz Reference clock. The Mutec wordclock was set to 44.1kHz and connected to the Bartok wordclock input 1.
With this setup, there was an apparent improvement in image focus with the Bartok, maybe a 5-10% improvement. I am thinking that if I used a better (lower phase noise) reference clock, such as the OP17 or higher, perhaps the improvement to the Bartok would be significantly better. Unfortunately I am not able to lay my hands on these better units for further testing at the moment.
I must admit I am totally new to this whole topic of clocking etc. so I am only basing on my own logic and from what I have read in the past 3 days on various audio forums. Perhaps others with more experience can help shed more light on this. My ultimate aim is to figure out whether I would do better sound-wise just to get the Rossini Clock for my Bartok, or to rig up my own dual wordclock master clock.
Yes I have heard the 40,000 gbp Abendtrot 10mHz clock with Vivaldi. Yes, there was a perceived improvement but significant for 40 big ones? One thing we know about this hobby is that marginal differences are touted as jaw drooping, wow moments 
In this case I would place the level of improvement akin to replacing a good interconnect with one slightly better. Well it’s a gain. But would I think of buying a cable costing £40,000 if it offered a similar level of improvement?
Thanks for putting this into a real-life perspective!! 
It’s an interesting question. For the price of a Vivaldi clock one could get a Cybershaft OP21A and two quality word clocks to feed their system. The dCS clocks have some advanced features like dither and the temperature calibration, but the Cybershaft certainly has the advantage in phase noise and accuracy. Hard to say which setup would have the advantage!
This is great! I have a Bartok with HP and love it, I’m so tempted to get the clock just to satisfy my Jones for new gear and better the experience. I’m sure it’s super cool. But I don’t want to fall into trading it towards a new Rossini, which is doubtless the temptation.
Jeff, I think what the Vivaldi stack folks were doing was to feed a Cybershaft OP21 or similar reference 10Mhz clock signal to the Vivaldi clock. They reported significant improvement with that setup. They did not try to replace the Vivaldi clock.
Meanwhile, I was curious if for the Bartok, it would be possible to use two quality word clocks and a Cybershaft 10Mhz reference clock instead of the Rossini clock, and which setup would give better results.
Turns out the Cybershaft OP17 (the OP21 would be too expensive) is out of stock so I did not have a chance to go that route. I have now just added a Rossini clock to my Bartok and burning in. Initial impressions has been promising!
Hi kentyeh, congratulations on the new clock!
I’ve read some of those forum postings where they talk about using different 10MHz reference clocks with the Vivaldi clock. Since I don’t yet have a Vivaldi clock the idea of focussing more funds on the 10MHz reference than the word clock is an interesting one.
I decided to take the other approach and just put an order in for an OP20A and a Japanese word clock that Hasegawa-san recommended has having excellent properties. It will still take some time for them to arrive, but I’ll report back on the results.
A Vivaldi clock may well still be in my future, but until then this setup should provide my Vivaldi with a technically excellent clock signal. Will it sound as good as a Vivaldi clock? Hard to say!
Jeff, I am a bit confused here. If you do not have a Vivaldi ( or other dCS clock) how can you use a 10mHz clock? You cannot use 10mHz directly into a DAC. The DAC requires 44.1 or 48 KHz input. A 10MHz clock does not replace the dCS wordclock. It is a reference source for it.
Perhaps I have misunderstood you?
Hi Jeff, yes, in your case, your plan makes total sense and you have very limited downside financially since the OP20A will definitely be useful even if you add a Vivaldi clock in the future. You can just get one of the word clocks for now for testing purposes since only one frequency is needed at a time.
My wild guess is that this setup may be as good or better than the Vivaldi clock, but the best would be the Vivaldi clock with the OP20A as reference clock.
Look forward to learning about your findings.
Hi Pete, I’ll use the Cybershaft with a word clock that takes the 10MHz reference and converts it to 44.1/48Hz on multiple outputs. The unit I’ll be using is very basic, but according to Hasegawa-san from Cybershaft also very accurate. He stated that it can do justice to the very low phase noise signal generated by the OP20A.
In terms of two key specifications that are important for audio (phase noise and short-term stability), this setup loses nothing to the Vivaldi clock. Of course the Vivaldi clock likely has a better power supply, dither support, and was built specifically for use with the Vivaldi stack, so in practice I wouldn’t be surprised if it would still sound better. The question is how much better 
I hope to find out one day, but until then this setup will do the trick!
Well I can probably guarantee that it will sound different . Whether better is down to your judgement. I remember the week that Rossini was introduced. My dealer called me in to have a listen. However the Rossini clock had not yet arrived. We had a choice of Puccini U clock, Paganini clock and Vivaldi clock . So we tried the Rossini with each in turn. The difference from one to the other was eye ( ear?) opening. It was also surprising how much greater the Rossini resembled the Vivaldi stack with the Vivaldi clock in place ( resembled, no you can’t get a Vivaldi stack equivalent with a Rossini/Vivaldi Clock combination 
).
Interesting. Two questions Jeff:
- Which word clock did Hasegawa-san recommend?
- What was your thinking on choosing the 20A rather than the 21A?
Hi Greg,
Hasegawa-san recommended a unit from a small Japanese company (the non-OCXO version): https://soundwarrior.jp/products/swd-cl10/
As for selecting the 20A, the 21A is sold out at the moment. But honestly for me I don’t think 1dBc/Hz is worth the additional 150k yen. With the 20A I still get the latest and greatest platform with external power supply, etc, so I expect the difference in sound is not that large.
@Gibraltar
Hi Jeff
I followed your link to Soundwarrior website and its all in Japanese (no surprise there). I am wondering if you had any issues ordering from them. Can you advise?
Also, how does your combination work with the Bartok? Can you please give us a short update?
Thank you in advance for your clarification.
Peter
Unfortunately the version with good performance (the non-OCX one) was discontinued recently, so they’re now only available on the used market.
If you want one there is currently one for sale on Yahoo Auctions, which you can buy via buyee.com: https://buyee.jp/item/yahoo/auction/l641290045?conversionType=search_suggest
I’m using the Vivialdi and the Cybershaft + Sound Warrior stands up quite well to the Cybershaft + Vivaldi clock. Both provide a definite improvement over no clock at all. Of course you need two Sound Warriors if you don’t want to switch frequencies manually between 44 and 48.
Thank you Jeff. Just the advice I hoped for… I can use of my existing Cybershaft and at a small expense of the Soundwarrior improve SQ while I save up for the Rossini clock!