Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Is AIR getting better in the Pro era? Thoughts of little steps.....
#11
Hi, I concur your observation as I did the same test on my system. 

My sms-200 / mutec mc-3 USB through FMC to switch still sounds slightly better to me, but the Devialet Ethernet to my switch is just cheap computer LAN cable, without FMC. 

As the 220 Pro Ethernet port does not accept gigabit connection, where my FMC box only support Gb, can't connect at 100 Mb. So there is still room for improvement.

Btw, I've one more point for the SQ improvement, is I wonder the internal up sampling to 24/192 through Air may be done differently.  Will it be done by FPGA, and other input is done by the BB SRC chips?
Roon HQP > sotm sms-200 > mutec (Cybershaft) > 220 Pro > Cardas Cygnus > Avalon IDEA
Denali 2000T
Reply
#12
(21-Jul-2017, 01:11)Kwan Wrote: Hi, I concur your observation as I did the same test on my system. 

My sms-200 / mutec mc-3 USB through FMC to switch still sounds slightly better to me, but the Devialet Ethernet to my switch is just cheap computer LAN cable, without FMC. 

As the 220 Pro Ethernet port does not accept gigabit connection, where my FMC box only support Gb, can't connect at 100 Mb. So there is still room for improvement.

Btw, I've one more point for the SQ improvement, is I wonder the internal up sampling to 24/192 through Air may be done differently.  Will it be done by FPGA, and other input is done by the BB SRC chips?

Good observation re.: up sampling! To begin with I used Roon's up sampling as this sounds best with the mR/Mutec. With RoonAIR it sounds better when up sampling is performed in the Devialet so for the last two days I've sent tracks with native resolution. Whether all up sampling is done in the BB SRC I do not know. The MAT (DSD to PCM) conversion is done somewhere in the DSP (I think) so it is possible that the stream from AIR is handled here too.
I have not had time to do a new comparison of mR/Mutec vs RoonAIR yet, but your observation matches mine so far.
*
Devialetless!
Roon, ROCK/Audiolense XO/Music on NAS/EtherRegen/RoPieee/USPCB/ISORegen/USPCB/Sound Devices USBPre2/Tannoy GOLD 8
250 Pro CI, MicroRendu(1.4), Mutec MC-3+USB
Reply
#13
Very interesting, in my case I prefer roon totaldac over roon devialet. Both units, devialet and totaldac connected with Ethernet cable. Any thoughts ? Is there away I can make the roon / devialet combo sound better ?
Reply
#14
(21-Jul-2017, 10:13)no32 Wrote: Very interesting, in my case I prefer roon totaldac over roon devialet. Both units, devialet and totaldac connected with Ethernet cable. Any thoughts ? Is there away I can make the roon / devialet combo sound better ?

But your Totaldac is connected via XLR to your Devialet? Makes sense. It's very similar to the situation with the Mutec. External re-clockers in this class sound very good. 
There is only a few things that can be done: improve ethernet transfer from Roon to Devialet as much as possible. This may be cable, switch, FMC's, various settings on the ethernet card and similar.
Let us hope the new Core Infinity board for the Pro amps also improves working conditions for AIR.
*
Devialetless!
Roon, ROCK/Audiolense XO/Music on NAS/EtherRegen/RoPieee/USPCB/ISORegen/USPCB/Sound Devices USBPre2/Tannoy GOLD 8
250 Pro CI, MicroRendu(1.4), Mutec MC-3+USB
Reply
#15
(21-Jul-2017, 10:13)no32 Wrote: Very interesting, in my case I prefer roon totaldac over roon devialet. Both units, devialet and totaldac connected with Ethernet cable. Any thoughts ? Is there away I can make the roon / devialet combo sound better ?

+1

I think the differences are small but overall the Totaldac feeding the Devialet via AES seems a bit calmer. Also I've had the odd dropout and a touch of white noise when running Roon direct to the O d'A, was very brief through and more like crackling fuzzy stuff in one speaker for a few seconds. Totaldac seems way safer.

Guillaume
Industry disclosure: UK distributor for Shunyata Research

220 PRO, totaldac d1 server with additional external power supply, totaldac d1-seven, Echole PSU for Totaldac, Wilson Audio Sasha 2, Shunyata Research cables, Shunyata Hydra Alpha A10 + DPC-6 v3, Various Entreq ground boxes and cables, Entreq Athena level 3 rack, 2 X SOtM sNH-10G with sCLK-EX + 10MHz Master Clock input + sPS-500 PSU, i5 sonicTransporter w/ 1TB SSD

UK
Reply
#16
(21-Jul-2017, 17:18)GuillaumeB Wrote:
(21-Jul-2017, 10:13)no32 Wrote: Very interesting, in my case I prefer roon totaldac over roon devialet. Both units, devialet and totaldac connected with Ethernet cable. Any thoughts ? Is there away I can make the roon / devialet combo sound better ?

+1

I think the differences are small but overall the Totaldac feeding the Devialet via AES seems a bit calmer. Also I've had the odd dropout and a touch of white noise when running Roon direct to the O d'A, was very brief through and more like crackling fuzzy stuff in one speaker for a few seconds. Totaldac seems way safer.

Guillaume

There is a separate thread on the Roon forum regarding this crackling issue. Would you like to join in that discussion as well? I think it could be beneficial now when the Roon guys are trying to figure out what is causing this Smile

https://community.roonlabs.com/t/crackle...rnet/28559
Bluesound Node > Matrix Audio X-SPDIF 2 > Genelec 8351B & 7360A
Devialet 1000 Pro
Bluesound Node 2i > Genelec 8330
Tampere, Finland
Reply
#17
I think the old adage that all threads that mention AIR, not matter how positively, seam to drift into comments of crackling, drop-outs, etc. A shame, but it does indicate that AIR perhaps has it's intrinsic issues that may never be solved. Let's hope the new OS Board is stable.

Anyway, my original observations in this thread have got me thinking about the USB input on the Pro. Most of my Devialet AES vs USB experience comes from the 'old' Expert models, where I have tried Melco, Aurender, dCS, CAD CAT and of course the microRendu.  Pretty consistently I have preferred the results via AES, perhaps most notably with an Aurender N10, which can output USB and AES, making a direct comparison possible.  However, that was then, now we have the new 'Pro' model.  I have been told that the USB board in the Pro is identical to the one used in the former 'Expert' model, but when I did briefly try the mR direct via USB on the Pro, it did seam to perform a lot better than it did with the 'old' Expert.  Specifically bass accuracy, so maybe something to do with galvanic isolation, leakage loops, or perhaps quite simply that the asynchronous performance in the Pro is better.  Anyway, much speculation.

My question is has anyone tried any USB to AES or S/PDIF comparisons with the Pro? Or indeed, has anyone found good results or a preference for USB with the Pro?

This question is kind of 'semi' off topic here, but it is related to my original comments, with both AIR and USB being asynchronous protocols. Maybe this deserves it's own thread if it gets some interest, although I suspect most of us have not tried any serious A/B USB versus AES tests? If I ever get the time and am in the right mood, I might try revisiting the mR direct to USB some time, it might be interesting. I did try it a while back, but only very briefly.
1000 Pro - KEF Blade - iFi Zen Stream - Mutec REF10 - MC3+USB - Pro-Ject Signature 12
Reply
#18
As long as UDP is used for AIR we'll always have talk of drop-outs and crackling unfortunately. When (if) the OS board runs RAAT we will see the more robust TCP in use and I think there will be few that has problems with packet-loss.

Very few units have USB and AES outs from the same internal source. A USB to AES DDC can be used, but then you're not comparing equal sizes. Still, this is an interesting comparison to make. I wish Devialet would provide us with a block-diagram showing where inputs are terminated. AIR and USB may have less processing of the data flow than AES and SPDIF. If your thinking is correct, a microRendu (maybe with a ISORegen) or the new ultraRendu should be comparable to AIR. AES/SPDIF would suffer because of the internal signal path or some other factor. I am working on a comparison of RoonAIR and mR/Mutec, but this takes more time than I first thought!
*
Devialetless!
Roon, ROCK/Audiolense XO/Music on NAS/EtherRegen/RoPieee/USPCB/ISORegen/USPCB/Sound Devices USBPre2/Tannoy GOLD 8
250 Pro CI, MicroRendu(1.4), Mutec MC-3+USB
Reply
#19
(24-Jul-2017, 15:22)ogs Wrote: ...
Very few units have USB and AES outs from the same internal source. A USB to AES DDC can be used, but then you're not comparing equal sizes. Still, this is an interesting comparison to make.
...

Some food for thought there and in Confused's post above.  Although the Mutec is usually talked of as a re-clocker and its benefits are presumed to derive from that, I wonder how much of the improved sound quality is "simply" due to use of AES rather than direct USB.  Maybe interesting in principle, but impossible to disentangle in practice I suppose.

(24-Jul-2017, 15:22)ogs Wrote: ...
I wish Devialet would provide us with a block-diagram showing where inputs are terminated. AIR and USB may have less processing of the data flow than AES and SPDIF.
...

I agree, it would be really helpful to have a complete block diagram of the amplifier showing the data and control flows.

What exactly do you have in mind by "less processing of the data flow"?  Do you mean there might be some DSP done on what comes in via AES/SPDIF but not on the AIR/USB data?  Given that it's all digital anyway, why would you be concerned about that?  (I am genuinely interested in your thinking, not being argumentative.)
Roon (Mac Mini), Wilson Benesch Full Circle, Expert 1000 Pro CI, Kaiser Chiara
Warwickshire, UK
Reply
#20
(24-Jul-2017, 15:37)thumb5 Wrote: What exactly do you have in mind by "less processing of the data flow"?  Do you mean there might be some DSP done on what comes in via AES/SPDIF but not on the AIR/USB data?  Given that it's all digital anyway, why would you be concerned about that?  (I am genuinely interested in your thinking, not being argumentative.)

I do not know really... we'd need the block-diagram to know exactly, but I am thinking that AES/SPDIF has a separate receiver chip while AIR is "directly" connected and USB with the XMOS processor may be connected similarly to AIR. This is all guesswork anyway, but from listening I have a feeling AES is somehow less transparent than AIR which leads me to this.
*
Devialetless!
Roon, ROCK/Audiolense XO/Music on NAS/EtherRegen/RoPieee/USPCB/ISORegen/USPCB/Sound Devices USBPre2/Tannoy GOLD 8
250 Pro CI, MicroRendu(1.4), Mutec MC-3+USB
Reply


Forum Jump:


Users browsing this thread: 1 Guest(s)