My Rossini goes away for Apex next week. Excited like a kid in a sweet shop 
2 Likes
Amazing how quickly an opinion ( and stated only as that) becomes a fact.
3 Likes
Audiophile said: *Yes, the Rossini Apex is better than the pre-Apex Vivaldi. I had the original Vivaldi here many years ago and have heard the Vivaldi countless times over the years…
Personally, I never heard the Vivaldi- but if two reviews indicated same there must be something about it 
1 Like
Honestly, just my opinion, maybe an “A” for effort, but it was a dreadful review. He shouldn’t quit his day job.
I can hear the piano as if it’s an 88-key color palette being worked over by a master musician.
3 Likes
I was impressed when i was at Martin Colloms house for the listening test for HifiCritic on the dCS Apex a few months ago. Definitely better than previous product.
Jon, can you shed any light on how those listening tests were done, or how much time MC typically spends auditioning a product? I ask because his Rossini Apex review includes comparisons, and numerical scores, for the different filtering/upsampling/conversion to DSD options. I found them impossible to make much sense of. I contacted him for clarification, but he was unable even to say which of the options were his reference (ie, what was the 100% score with which all others were compared). I got the impression of a very rushed job.
1 Like
well he spends many days listening. I was there (with one other) for a long day – probably 9 hours overall. But this was a verification process really, against a considerable body of work he had already done.
1 Like
we went through the lots of settings/combinations. I can assure you my hatred of the filter combinations, documentation and app came from trying to make this all work.
Why on earth can one filter number do entirely different things depending on the input sampling rate?
Why is there no “return to default” feature? Or dCS recommended settings? Why are half the filter settings not shown on the unit display?
It is so sad when such excellent hardware (and build quality) is hampered by such mediocrity and awkwardness.
2 Likes
it was not rushed at all, i can assure you.
I doubt there is anyone who disagrees with you about the confusion of settings. Even What Hi-fi were critical. I think dCS must have left it up to the engineers (sorry, guys). The choices need simplifying, and a better presentation in the manual.
Under PCM filters:
“Filter 5. If the sample rate is 176.4, 192, 352.8, or 384 kS/s, the filter has a Gaussian
response, which offers no overshoot on transients and a relaxed roll-off. If the sample
rate is 44.1 kS/s, the filter is an asymmetrical design, which features almost no preringing. Filter 5 is not available if the sample rate is 48, 88.2 or 96 kS/s
Filter 6. If the sample rate is 176.4, 192, 352.8, or 384 kS/s, the filter is an asymmetrical
design, which features almost no pre-ringing. If the sample rate is 44.1 kS/s, the filter is
an alternative sharp filter, which has linear phase and pre-ringing. Filter 6 is not
available if the sample rate is 48, 88.2 or 96 kS/s.”
Now I know what dCS mean by the following, but I have a test lab with Audio Precision and dScope test equipment.
How many users understand this (under DSD filters):
“Filter 1. The default setting, which gives the widest bandwidth and the highest level of
out-of-band noise.
Filter 2. Reduces the out-of-band noise and bandwidth.
Filter 3. Reduces the out-of-band noise and bandwidth further.
Filter 4. Recommended for troubleshooting only. Sharply reduces the out-of-band noise
and also limits the bandwidth to 25 kHz.
Filter 5. Has a relaxed roll-off with a smoother phase response. Removes much of the
out-of-band noise”
So is filter 2 better than 1? Is 3 better than 2? Under what circumstances should you use 4? Is 5 better than 1?
If 1 is the default setting, then “highest level of out of band noise” is a good thing? Or not?
And (if my memory serves me correctly) this filter setting isnt shown on the display.
1 Like
Yes indeed, I listened to the Rossini Apex and Vivaldi non Apex next to eachother in the same system, in no way the Rossini Apex was better than the vivaldi imo.
1 Like
How those various filters noise-shape was measured in Stereophile’s Vivaldi review;
Fig.10 dCS Vivaldi DAC, spectrum of DSD-encoded 1kHz sinewave, DC–100kHz, at –20dBFS into 100k ohms with DSD Filter 1 (left channel blue, right red), DSD Filter 2 (left cyan, right magenta), DSD Filter 3 (left cyan, right blue), DSD Filter 4 (left green, right gray) (linear frequency scale).
It depends almost entirely on your downstream components and how they react to those out-of-band noise. It’s impossible to predict, but discernible via careful trial and listening.
3 Likes
I know that, and you know that. Customers probably dont.
frankly, I never played with the filter options…I used Mapper1 and that’s it.
So- again- a stupid question: let’s say I would prefer Filter 1? for CD and Filter2? for DSD- can I set it as default then permanently?
When you set any filter for any incoming sample rate then this is remembered by the dCS component. So it will be permanently there for that sample rate until you might care to change it. So it can remember a different filter for, say, 16/44.1 or 24/96 or DSD and change between them automatically as you listen to various tracks in a session. BTW, that will be remembered after you switch it off and will be there when you come back.
Just for clarification as the filters used are related to the sample rate and platform ( i.e. PCM or DSD). The numbers do not indicate a particular type (see your user manual) just the sequence in which they appear for that format. As has been mentioned the choice is subjective and that there are no right or wrong choices ( "wrong "choices are not offered which is why not all types are offered for all formats).
After a little while of trying different combinations you will find your favourites and you can then forget about them as it will move between them without you needing to do anything. Aside from one change I have been using the same combinations for 23 years 
.
5 Likes
Ah, thanks! 
I’d like to think they do; dCS users generally have a much better grasp on the platform’s flexibility and how to get the best out of them.
1 Like
In my system, there is a slight veiling with Roon but I still use Roon for its search capabilities.
If people actually perceive a sonic difference with Roon when compared to Mosaic/UPnP, then either they’re not comparing the exact same track, or their Roon Core is misconfigured/misbehaving - DSP Engine, Upsampling, EQ, bit-stuffing, Volume levelling, Normalisation, etc etc. - and should be properly checked and fixed;
For example, here’s an ‘oldish thread on Roon Community that shows how easy it is for Roon Core to be misconfigured and not be bit-perfect;
2 Likes