+- Devialet Chat (https://devialetchat.com)
+-- Forum: Devialet Chat (https://devialetchat.com/Forum-Devialet-Chat)
+--- Forum: Streaming (https://devialetchat.com/Forum-Streaming)
+--- Thread: Core Infinity listening impressions and comparisons (/Thread-Core-Infinity-listening-impressions-and-comparisons)
RE: Core Infinity listening impressions and comparisons - Wdw - 07-Dec-2017
(07-Dec-2017, 20:22)Mr_Bill Wrote: Wdw,
Thanks for your comments.
If you are listening via AES input - how would the installation of the CI board affect that? Is it the software change?
Truthfully, I'm baffled as to what is effecting this change...I think the Confused One made reference to a new clocking scheme introduced with the Core Infinity which may explain the changes somewhat and then there are firmware/software changes but dunno...
RE: Core Infinity listening impressions and comparisons - Confused - 07-Dec-2017
First a polite request. @patcam and @midi we do have a 'Problems with Core Infinity' thread. Anyone in the future who has a similar issue to yourself will probably look there in hope of finding a solution. Why not post there? Plus, when you do get the new board up and running please come back and tell us your listening impressions here.
@Wdw - I have to say that I was fascinated by your post, and @Mr_Bill's subsequent question 'If you are listening via AES input - how would the installation of the CI board affect that? Is it the software change?' First to say, maybe it is indeed the software change. We have not had a firmware update for a while, maybe Devialet have tweaked something in the firmware that is yielding good results with the Pro. Based on past firmware changes, this is entirely possible. Another thought is maybe it is hardware related? Personally I think this is highly unlikely, in that I would expect the AES/EBU input to work as it should even with the Core Infinity board removed. @ogs has speculated (and he made it clear in his post that he was just speculating) that maybe the alleged 'Super Clock' on the CI board might be somehow re-routed to provide it's goodness elsewhere within the Devialet. Maybe it is, but I suspect this is unlikely. Indeed, if the CI board clock was acting as a 'reclocker' for the AES/EBU input, which is a synchronous input, then it could actually limit performance with the very best sources. Consider that AES/EBU is an AES3 synchronous protocol input, it is therefore extracting the clock signal from whatever is feeding it. This is why devices like the Mutec MC3 can offer improvements. So if you have someone using an absolute top end reclocker, say a Mutec REF10 with MC3+USB and then is the CI board was subsequently reclocking, it would only make things worse. With the best will in the world, the CI clock is unlikely to be as good as a £3K Mutec clock. Another possibility is that it could be good old fashioned expectation bias, quite possible, it can be a very powerful beast, but based on the clarity of Wdw's words, I would tend to think not in this case. Could it be anything else? I do have a final theory here, there was some past speculation that worried that adding a powerful, possibly noisy, computer board into our terribly delicate and sensitive Devialets could cause some issues with electrical noise creeping in where it should not. Lets assume this is indeed the case, users then report SQ improvements, maybe this is because the supposedly negative impact of the CI board actually produces euphonic results. Again, highly unlikely I think, it's just the last possible thing that I can think of!
Much waffle from me, but this really makes me wish that we had a little more technical information from Devialet with respect to exactly what the CI board does. How it otherwise integrates with the rest of the amp, implications for the new USB input, implications for the other inputs, implications of recent firmware / software implementation, I could go on. To be honest, we have never needed one of Devialet's nice white papers more than we do now.
RE: Core Infinity listening impressions and comparisons - Wdw - 07-Dec-2017
(07-Dec-2017, 22:47)Confused Wrote: ........supposedly negative impact of the CI board actually produces euphonic results. Again, highly unlikely I think, it's just the last possible thing that I can think of!
Much waffle from me, but this really makes me wish that we had a little more technical information from Devialet with respect to exactly what the CI board does. How it otherwise integrates with the rest of the amp, implications for the new USB input, implications for the other inputs, implications of recent firmware / software implementation, I could go on. To be honest, we have never needed one of Devialet's nice white papers more than we do now.
The Confused One, a well earned moniker!
I do agree with every one of the many tangents you went on It is odd that the AES would show some improvement and this circuit also served by the excellent clock from our dCS.
Expectation bias could be quite likely as well...I was going somewhat daffy when the D440 was subject to further delays before shipping back, held for a few more days than necessary, awaiting release of the the firmware released last Thursday. I'm certain that I have alienated all of the good natured folks at the Devialet support desk in New York...so having music back in the house was certainly a joy but I can only counter that I had no expectation of an improvement from the board and some concern that induced electrical noise, as you mention, could be likely.
Baffling...but my wife commented immediately when it was just warming up and playing in the background....I think my initial post may have been a tad too over the top but it certainly does sound better.
Best
RE: Core Infinity listening impressions and comparisons - ogs - 07-Dec-2017
@Confused, the only thing we know about the clocking on the CI is what @Celts88 posted earlier in this thread (post 16) where a Devialet employee states that
"The Core Infinity board will effectively be the one to generate and distribute the purest clock signals, including the word clock. Those signals are then rerouted on the motherboard"
So it seems the whole amp benefits from the improved clocks on the CI. If all this is correct it explains the reports we have seen that SPDIF and AES sounds better than before. This could also explain what Thierry Massoubre meant when he said to me in an email "- We will take this opportunity to do a few modifications on the motherboard to improve the performance of your unit"
We may know more about this in the future, but we will certainly know if our amps sound better when we get them back with the CI installed!
RE: Core Infinity listening impressions and comparisons - SwissBear - 08-Dec-2017
A short comment here: it is not easy IMHO to make comparisons between two setups when days not to say weeks have passed between the removal of the 'old' system and the arrival of the 'upgraded' one. As @f1eng said in another thread, the volatility of the audition memory is high and the validity of perceptions can probably not exceed the order of magnitude of minutes/hours.
Having said that, I have received my D900 upgraded to D1000 with the CI board Tuesday this week. After having let it burn-in for some time, I have made a first 'serious' listening session tonight. At this stage, given the level of maturity of the firmware as we know it, there is still a very audible difference between the Roon Ethernet input of the Devialet and the AES/EBU one, when fed by a rather 'sophisticated' source like @Antoine has initiated in this forum (Paul Hynes LPSU, MacMini, SOtM sMS-200 Ultra, tx-USB Ultra, Mutec MC-3+USB, all these three devices being powered by the PH and clocked by a Mutec Ref-10).
Differences are clearly in favor of the latter, especially in terms of general dynamics of the sound, as well as micro-dynamics.
Just my 2c.
RE: Core Infinity listening impressions and comparisons - sam1000 - 08-Dec-2017
I can’t wait to get my 250 pro so I can do head to head comparison with and without CI board. I’ll send my other unit for upgrade after Xmas.
Having said that, I did run SMS-200 with upgraded power supply and MuteC-3. After listening to Roon AIR , I sold that combination. Ref-10 might be in another league though.
RE: Core Infinity listening impressions and comparisons - baddog - 08-Dec-2017
I just wanted to thank everyone for posting their impressions and experiences. I got a late jump on the upgrade as I was in South Korea for the initial two weeks of the program, so my units just arrived in NY for servicing this Monday. It does sound like Devialet exceeded expectations on this upgrade, good for them!
RE: Core Infinity listening impressions and comparisons - Wdw - 08-Dec-2017
(08-Dec-2017, 02:30)sam1000 Wrote: I can’t wait to get my 250 pro so I can do head to head comparison with and without CI board. I’ll send my other unit for upgrade after Xmas.
Having said that, I did run SMS-200 with upgraded power supply and MuteC-3. After listening to Roon AIR , I sold that combination. Ref-10 might be in another league though.
Look forward to your comparisons...keep us posted!
RE: Core Infinity listening impressions and comparisons - Damon - 08-Dec-2017
(07-Dec-2017, 23:26)Wdw Wrote: Baffling...but my wife commented immediately when it was just warming up and playing in the background..... but it certainly does sound better.
In all seriousness, the same behaviour occurred in our house. My wife also commented on the change in sound; this was within two minutes of the CI unit being put back on the shelf and fired up. I think this type of reaction from unbiased (and uninterested) listeners, strongly indicates that expectation bias is not a factor in her awareness of the sonic changes.
Although I understand (sort of) the concept that simply by observing and experiment or event, that we change the outcome of that event, I am confident in dismissing, or diminishing, the influence of bias, as I did not expect any change sound quality. Nor was I wanting a change. If anything, I was simply glad to have the amp back, and if anything, I was apprehensive about potential problems as a result of the CI board install.
Amici, this is a really good thing. To the listening chair, Batman!
RE: Core Infinity listening impressions and comparisons - ssfas - 08-Dec-2017
The uPnP sounds superb, but is unstable - moreso than with the old streaming card.
Was using a dedicated 5G network from a Kindle Fire HD running bubbleupnp. The router is the ASUS RT-66N with a MCRU PSU.
Back streaming by usb from Aries.
UPDATE: Put the Dev on a guest 5G network and 30 minutes of 24/96 and not a hiccup.