Well there was a few other win10 users having the same issue.
This could explain why some people hear differences.
Converting the RME test wavs to flac fixed the issue from memory @ 24bit, but then others complained the 32bit test failed.
1 Like
When took the look on the Motus that time, did you check out Motu 828es too maybe?
No sorry.
The Motu had more than enough output channels, which is all I use it for.
The additional IO would only be helpful if you complete crossover duties outside Roon, and have the ability to route additional analog/digital I/O.
1 Like
Using some Synthies here, etc…And was thinking about Thunderbolt…But anyway thank you 
I know this wasn’t aimed at me, but as the OP I wanted to address it. I definitely try to be aware at all times of potential cognitive biases. That’s why in my original test, I made sure to also test what I perceived as the better quality solution at a lower volume to ensure that my brain wasn’t saying “louder=better”.
But If I carefully examine my biases, if anything I should be subconsciously giving Roon the advantage. I’m really excited about the functionality–just subscribed for a year and strongly considering converting that to lifetime; NUC to run ROCK on just arrived a few days ago; and I’ve had my father-in-law (who lives with us) frantically ripping his CD collection to FLAC so it can be imported. If I’m going to notice a non-existent difference in this case, the better sounding option should be the “new hotness”, because of confirmation bias. But that isn’t what I’m perceiving.
By way of update, I did do a brief A/B test last night based on my new configuration. With the convolution filter for my Clears applied and DAC/Amp EQ disabled, Roon now sounds almost as good to me as native Qobuz with my EQ curve applied. It’s close enough that if I’m not doing a direct comparison I don’t recognize a difference at all. I will test more to be sure, but that’s what I’m hearing so far.
2 Likes
It’s the same device chain (and for that matter the same power supply within a device) from the DAC onward and that’s what counts in the instance to which you are referring.
Or as it’s been commented -
…
Let us know your findings.
Really, the question is whether Roon feeding a different DAC sounds as attenuated(?) as it does feeding the Lumin device? Hard to test since too much would be changed.
I respectfully disagree that it‘s the same device chain, I agree not hearing bits but the analog conversion to it, based on your comment non-DAC transporters devices like Auralic Aries and Lumin U1 do not make any difference then. Bits are Bits.
My chain includes two other DACs - RME & McIntosh DA2, streaming via Roon is attenuated vs streaming via Lumin software (not using the T2 DAC)
Yes, that’s what I mean when I say from the DAC on it’s the same chain.
I couldn’t tell from what I read (I couldn’t find a proper online manual) whether one can use the T2 as only a streamer without using the DAC. Is that how you feed the other DACs you mentioned?
Yes, correct, via USB and/or BNC.
The Lumin (such as many other devices / brands) has config for either using internal DAC with the analogue outputs or using an external DAC via USB or BNC
Hi David,
I noticed you have “WASAPI” mode when using output from your desktop.
I had very bad experience with WASAPI when running Foobar on my desktop.
I am not an expert, but WASAPI is not as good as claimed in terms of audio processing through Windows.
Could it be the source of your problem?
Whatever, installing a Roon server without using its output will prevent making use of WASAPI.
Best regards,
Didier.
Pre Roon, Vortexbox, Squeezelite and many others I used software called MQn+Rewrite for music playback.
https://tirnahifi.org/forum/viewtopic.php?f=15&t=2418&hilit=mqn&start=8000
It was great community feedback developed software. The thread explored the development of a bit perfect software player and how tweaking certain elements of programming, latency, code etc impacted the SQ of the music (bit perfect signal).
The programmer continually used feedback from users over many years to tweak and improve the players sound/code. It was a great sounding player, circa 800 thread pages.
Likewise anyone who has tried different streamers into the same DAC, knows differences in SQ exist between bit perfect streamers. Sometimes these differences are small (and not worth worrying about) while sometimes these differences are quite noticeable. Streamer = software+hardware.
This is before the variation in DAC inputs is even considered.
1 Like
Thanks for your continued indulgence. You’re probaby tired of my pestering. I continue because your experience doesn’t make sense to me.
I think I’ll let sleeping dogs lie.
1 Like
No, not tired at all. But clearly the discussion is turning in circles. Hence I agree on your (edited) feedback - let sleeping dogs lie.
For your reflection:
The Roon forums have various similar user experiences for non-DAC devices as you might have noticed.
This is not a Lumin or T2 discussion. If my experience does not make sense, the fundamental question back to you - do you believe in quality impact of how (i.e. with what hardware) the Bit is fed to the DAC ?, do you believe in the acoustic upside of devices such as the Auralic Aries and Lumin U1 which are doing exactly that. Yes or No ?
1 Like
sure 
still looking for the „best“ way. my bet on streaming DACs
Edit, maybe an interesting (old) discussion in this context about bit perfect tests and SQ What does a (M-DAC) 'Bit Perfect' test tell you about jitter? | pink fish media
1 Like
No. If one needs a streamer sans DAC to improve their system something is wrong elsewhere.
We must agree to disagree. Peace.
3 Likes
My closing thoughts.
There are some things you can only learn through doing and experiencing.
Practical knowledge often provides a deeper understanding of a concept through the very act of doing/experience.
Many many years ago, I was in the bits are bits team.
However, as I listened and experienced various hardware, system config’s, software, sources, DACs, etc I became shockingly aware this theory just didn’t hold.
IMO, reviewers such as GoldenSound have stepped in to fill a needed market void.
Detailed measurements backed by educated subjective listening.
Enjoy the music.
Lets get this thread back onto Qobuz… 
3 Likes
Thank you very much for provided statement and link! Yep, let´s go back to Qobuz.
By way of final update, Qobuz via Roon and Qobuz now sound identically with DSP disabled in Roon and my DAC/Amp’s EQ curve applied for both, and I’ve A/B tested several times over the course of the past several days. The only changes I’m aware of that have occurred in the intervening period are having transitioned to a ROCK installed in my network closet, and Roon updating from 913 to 918.
As an added bonus, playing with several of the DSP options, I now significantly prefer using Roon’s convolution filter and DSD upconversion to my EQ curve on the DAC/Amp. So I think I’m set, and switched my subscription to lifetime yesterday.
9 Likes
Welcome to the “lifers” club…
2 Likes
Eating at the cool kids table. 
4 Likes