I tried a comparison of the Qobuz Android app and Roon ARC with a Dragonfly Cobalt attached and can say that Roon ARC definitely sounds richer with more frequency range to my ears. The Qobuz app was way too muddy and punchy. I have to try it with UAPP next…
1 Like
If you use filters in Roon, the soundlevel is often lower then the original and your brain thinks that loud is better, so did you compensate for that?
Great question.
I am not sure why but the crossfade did “seem” to be causing what I heard - degradation.
I had an audiophile friend over yesterday and as we all do, did some serious intentional listening. We didn’t notice a difference with crossfade disabled.
With crossfade on, there was a difference - small? Yes.
Imagined? Maybe. I am so curious - is there a real difference?
My dealer, who sold me the Roon, insists the degradation is real, as does his staff who were here when we installed the new amps.
We also did A/B testing (crossfade was active) and we agreed that the LINN KLIMAX DSM sounded better.
After this discussion, which I appreciate the many’s POVs, I am questioning what we heard.
In any case, the difference isn’t very significant and the Roon interface is remarkably better than LINN.
I need more time trying to understand why.
It’s funny that I look for differences with all my “upgrades”. logically, when I looked at my rack purchase - HRS - the idea that a rack would change system sound made zero sense, but you know, it did. My wife noticed it immediately.
The whole soundstage opened up.
I can say the same thing about the footers on my Wilsons.
The subtle changes in sound caused by tiny things, which frankly, I don’t understand, is really fun for me to explore and try to understand.
Everyone has an opinion. In the end, it is what our perceptions are, how the music feels, making tweaks in search of that “sound”, that “feeling” that moves you, resonates with your body.
It’s a crazy and rewarding hobby (lifestyle) for me. I am in my room about 2 hours a day, every day when not traveling.
This makes me happy. And after any upgrade, I always say, this is it. It’s perfect. No more upgrades.
And then ……
8 Likes
@Simon_Arnold3 - lost track slightly - did you do a bitperfect/null test with crossfade on/off?
Crossfade is obviously not bitperfect during a transition - and shouldn’t do anything elsewhere - but is there any effect if you have only one track in the play queue?
Maybe Roon Devs need to answer that…
1 Like
I guess it upsamples anyway at all times even if there is no fading going on
It would be easy to check if crossfade is bit-perfect in the middle of tracks. Unfortunately I don’t have a subscription anymore so I can’t do it. But even if it isn’t, just going from 16bit to 64bit then back to 24bit should not be audible.
It shouldn’t of course; but it maybe also depends on what code might remain ‘in circuit’ to manage combined levels during the crossfade.
If the code that remains in the signal path did anything besides bit depth conversions when outside of the crossfade window, it would be a bug. Roon can chime in, or someone can just analyze the output and put it to rest.
It would… albeit a pretty benign bug if all that it did was drop the level slightly as a consequence of not clipping during transitions. I am, of course, guessing. But at least it’s a hypothesis that can be tested.
Based on my understanding of crossover, the signals being crossed over are each scaled with factors between 0 and 1, and the sum of the two factors is always 1. Thus, if none of the signals clipped individually, the crossover would not clip either.
1 Like
If I’d read the manual properly when doing crossfades in Reaper, I’d be able to explain better… but it’s potentially more complicated than that - depending on the effect you are trying to achieve.
We are, however, on a tangent. If it can be demonstrated to have no effect outside of transitions, then it - errr - has no effect.
How can a cross fade be ever bit perfect it’s mixing two tracks together.
It can’t. But outside of the transition region, it should have no effect.
The mixing is done only during transitions from a track to the next, in a short window. If that window is say 1 second, then the last second of the current track will be mixed with the first second of the next track. Outside of that window, nothing happens (or should happen) to the signal. Ideally, that’s where it should still be bit perfect.
Not sure it works like that but in theory it’s not being altered before or after the cross points.
That’s the theory that needs testing… hence the original question… 
If it’s the same bit test file it seems to pass but it also seems to crash the test mechanism of the DAC and it then hangs so hard to tell.
Thanks for trying it… 
I love this thread! It makes me wish I had ears that could tell a difference. I’m lucky when I can tell a difference from Spotify to Tidal MQA! (Just kidding, threw that in to rile you up!)
4 Likes
Yup. I love this community where music makes a difference. I would say that of the folks who visit us and want to hear “the system” only 10% actually listen.
When they do, some of them have tears when they listen. I have a lady friend with hearing aids who was crying when I put on Rumors - Fleetwood Mac on vinyl. She kept saying thank you. She said she hasn’t really heard music in over 10 years.
The other 90% of the folks that listen are looking at their phone, talking over the music and not being present. Sort of pisses me off a little then I usually shut it down and they don’t even notice.
So yes, I am fortunate that I really hear and feel the music. And my special friends who come over do as well. With a little scotch, some weed and the lighting just right…. Mmmmmm, Mmmmmmm Good.
9 Likes