Is that a Yiggy only deal, because HQPlayer sounds heavenly with my BIfrost.
This is a mischaracterization. There’s a huge difference between “writing a new OS” and “packaging up a minimalist Linux distribution”.
I’ve gone into much more detail on some of the motivations behind ROCK here. There is way more to it than support.
In general, if you upsample outside any Schiit multibit DAC to their top sampling rates (176.4kHz and 192kHz for the Bifrost), you are bypassing the DAC’s internal upsampler and closed-form filter. Maybe you prefer what HQPlayer does to the Schiit closed-form filter. Or you are not upsampling in HQPlayer, but you are taking advantage of something else it helps with, hard to know without knowing your configuration.
Roon sounds better than Jriber with my DSD ps Audio DAC with the Raat protocol is operating. My DAC is Raat compatible and this is key. Do you use Raat ?
Hi again everyone
Some really interesting topics and facts here. I certainly feel more informed after reading through the very worthwhile posts above
There is a lot of reference to ROCK. I must have been living under one because I have no idea of what that is? Any one care to educate me?
Based on all of this discussion , I’m thinking that I may now build a purpose built PC to act as my Roon core. I’ll leave all of the files on my nas but having the core running from a dedicated pc sounds fun. This will sit in my garage, along side the router and nas and be on 24 / 7.
Any suggestions for components to look out for or avoid? I’ve built many many pc in my time, but never for the sole purpose of hosting Roon (or in fact any single purpose) I’ll spend about 1k if needed.
There’s a whole forum here dedicated to ROCK, but this post is a good starting point.
That makes sense. Is it possible to replicate the closed-form filter within HQPlayer to allow for an easier AB comparison?
It’s a proprietary filter, although based on some old Bell Labs theory. I suspect that a lot of its value is in the particular implementation choices Schiit made for the original theory. Here’s the designer’s Mike Moffat explanation in another forum:
A few comments on filters in general and Mega Combo Burrito in particular. Closed form math refers to coefficient calculation. Granted, input samples are rounded after filter math, except original samples which have zeros added for 18 bit Gumbys and 20 bit Yggys. Agreed, all filters have tradeoffs. I like the way the Mega Combo Burrito sounds. Many agree with me. I only mention Parks McClellan for its ubiquity in unremarkable audio applications. Yes it is linear phase in the sense that what comes in goes out, but it enhances nothing in the time domain. The beauty of a DSP based filter is the convenience of implementing and evaluating the filters discussed above, plus many others. I have been down that road years ago. The earliest time domain algos posited by Bell Labs right before the entry of the US into WW1 (yes, 1) are very interesting in particular. Build yourself a Parks McClellan and hear the soundstage collapse into your speakers. Get out of thinking and into action and build them. This is what we were doing 30 years ago. That is probably why you haven’t heard the names of my collaborators - this happened over a five year period in the 1980’s. We built dozens of them back then and ended up with the unlikely freq time domain combo we have. Allow me to correct a typo. The transition band commences at .986 of nyquist in a Mega Combo Burrito.
Finally a huge underscore on the importance of time domain. We evolved to be able to locate a sound before we could make a decision to either run from or go after it. The Mega Combo Burrito filter uniquely addresses exactly this.
I tried to replicate the “Schiit Sound” with HQP and an IFI Micro DSD Dac. Just for listening comparison.
I didnt come close. If you have success, please let me know.
Mike Moffat’s 30 years in the business since Theta Digital and his many collaborators along the way are a bit of a head start
OK, I CALL UNCLE ON THIS!
I found this too. Trialling both and (from headphones out of laptop) the Audivana had a lot more presence and sparkle. Wondering how i can get roon to sound as good as i much prefer the ux and features of roon???
Yggdrasil’s max input rate is 192kHz and it upsamples internally to a max of 384kHz. This means you can never get Yggdrasil to do NOS. The only Schiit DAC I have seen specifically called out as doing NOS is the Modi Multibit.
I tryed many different software’s on the Mac and I must say that ROON sounds perfect
I use a PS AUDIO DIRECT STREAM DAC on a Macintosh tube preamp and a Macintosh amplifier with B&W 802
I don’t understand why you are telling it should sound worse than other software
It does a bitperfect translation of your music to your DAC
the only thing is your DAC must be good
Everything else is a placebo effect !!!
No other software can sound “better” !?!
The user interface of Roon is at the moment better than anything else including Audirvana and others
And forget absolutely J river or Amarra…
Hope this helps
In my experience, you shouldn’t underestimate the importance of EQ. I used JRiver for years before I started using Roon. For a long time, I couldn’t get the same SQ from Roon that I was getting from JRiver. I started using Roon anyway, because the integration of Tidal with my library of ripped CDs was too good to pass up. I eventually found that the SQ differences between JRiver and Roon were mostly the result of EQ settings. JRiver has a simple graphic equalizer, which I always used. Once I figured out how to copy those EQ settings to Roon’s parametric equalizer, the difference in SQ disappeared. I didn’t have any experience with parametric EQ, so it took some learning to get it to sound right. But, this has been one of the most important adjustments I have made to my playback system. I know that the audiophile community generally dismisses EQ, but it is, no doubt, a powerful tool.
I am in total agreement with you @Louis_Richardson.
I think one of the reasons the audiophile community dismisses EQ is historic… EQing music used to be done by passive analogue components and resulted in phase issues and a deterioration of the sound due to more circuitry being involved and a longer path-length for the signal.
That’s all gone now we store and eq music digitally, as long as the mathematics is performed at a high enough bit-depth, which it is in Roon.
To my mind, much of the selection and matching of various hi-fi components and the preference of some combinations over other combinations over the years has been due to certain EQ differences between the final combination of components.
As an example… People talk about ‘preferring a certain D-A over another D-A’ because it’s ‘Warmer…Smoother…’ etc.
That’s all achievable in EQ within DSP in my opinion.
Everybody’s preferences are different when it comes to sound, and even the shape and design of peoples heads and ears affect the strength of different frequencies presented to their eardrums (Have a look at Nuraphone headphones, they reckon there’s something in that!)
I personally am more sensitive to mid-range frequencies around 1-2KHz. I really don’t like a ‘forward’ sounding speaker. I have Kef LS50W and they’re great… really open and crystal clear, but… in my room, I think the midrange can get a fraction ‘shouty’. So… I run PEQ and reduce 1K-3K by 4dB. I also boost 80Hz downwards by 3dB as I’ve always liked a ‘warmer’ sound.
A couple of years ago I heard a friend’s many-thousand-pound Naim system, and while it sounded sublime on well-recorded pieces, it ‘shouted’ at me on less well-recorded pieces. He had no way of adjusting the EQ, so, consequently, he said there were lots of recordings he ‘couldn’t listen to anymore’. I thought at the time, that’s a real shame and started researching a system that would give me some control if I needed it, without destroying the clarity and transparency of the better recordings.
I found Roon, upgraded my passive LS50, driven from an AV amp, to LS50W and… haven’t looked back!!
Now… if we can just get Roon to develop a way of specifying a certain EQ preset, to be used in addition to my normal EQ, for a few of my more poorly mixed albums, which can then be auto-recalled when they play, that would be perfection!!
I have to say: You Sir are so right!
This has been my experience as well. If I disable my equalizer settings and put on Steely Dan’s Aja, it sounds perfect. But the vast majority of music I listen to requires some eq tinkering in order to sound “right” to my ears. I actually just use a mild “smiley face” pattern. This works well for me. I generally enjoy a “bright” presentation I guess.
I have also thought that saved eq settings for individual tracks would be a cool feature. Doesn’t iTunes have a feature like that? Honestly though, I don’t know how much I would really use it. For me, the key is finding settings that work well for the majority of recordings, so you can just set it and forget it.
Yep. You’re right. I wouldn’t use it for many recordings. I’ve got mine generally set up how I like it and it works for the majority of albums.
Be nice to have though!
I remember Bob Stuart talking about this when at Meridian. The idea that people put systems together that do one thing at one volume extremely well but don’t deliver in other areas. One of the Meridian goals was to make systems, that culminated in DSP speakers, that would deliver results with any music at all volumes.
I have found this to be the case as I have always been happy with my Meridian systems with any thing I throw at them. Sounds like an add for Meridian but I don’t mean it that way. I was attracted initially by that type of thinking though and stuck with it.