Devialet Chat

(04-Jan-2015, 18:23)Mohmm Wrote: [ -> ]The principle is: "If there is no problem tweak your set-up so long as to produce symptoms."

Hello

Some people like to tweak and try any sort of cables (speaker, USB, Lan….), streamer or software, and change their alimentation and so on… There is no limit to it and as long as it is their passion, whatever their conclusion is, it has to be taken by subjectivity.
I like to stick to a simple system although I have added a few “devices” that I think make a difference but again I am trying to convince anybody about it.
Regarding the white noise problem, the issue is different. I just had a look at your previous mails in details. You are not using Yosemite, you are still on Mountain or Maverick so the problem did not exist at that time….
With Mountain, Maverick the Air Wireless could suffer some interruptions (especially with high definition formats) but the Air Ethernet was clean.
With Yosemite it is bugged with both Air and Ethernet. I have always used the USB/SPDIF (with D Premier) and now I use the USB so I do not really care about it but having the Air (Wireless and Ethernet) not working for such a long time shows that either Devialet has not time to solve it (busy with Phantoms) or they cannot do it so easily…
If you switch to Yosemite you will probably have this issue so think about it before doing it… and better wait for a fix first.
Yosemite does not seem to be easy to “manage”; I have been using JRiver Media Center with it and it makes crash my Mac at some point, or I have to reboot it to use Safari later on. I have no explanation but again people using Jriver with Maverick seems to be fine (I was fine with it too at that time)… I am not so surprised Devialet is very slow to address the issue…
Regards
philoouu

So the problem with these symptoms seems to occur on both Windows and Mac, though there seem to be more reports of it from Windows (& yes, I know this thread is about Windows!). Could this be simply because there are more Windows users reporting the problem here or is it because you're more likely to see the issue with Windows?

Confused definitely has the problem with the beta AIR with screensaver enabled, but it looks much better, if not fixed, with screensaver disabled. This is great as it's a cast iron case where we have evidence that it can be caused by a screensaver. Of course it seems screensavers might not be the only program/cause that produces the problem but it's a start at least.

This is with a simple screen-blanking screensaver right? It can be confusing because the there might be a cpu-hogging screensaver running in the background but not being seen due to the power options just turning the display off. Or it might be simply the display being powered off - no screensaver.

I have no screensaver set, but display turns off after 15 minutes (high performance power settings with a few customizations). I do however suffer dropouts, but no white noise so it's possible my issue is related, but not identical. I can't tie in these dropouts with the display turning off - I might try this today. I notice I have USB set to selectively suspend which could be another possible culprit.

(05-Jan-2015, 09:17)Rufus McDufus Wrote: [ -> ]So the problem with these symptoms seems to occur on both Windows and Mac, though there seem to be more reports of it from Windows (& yes, I know this thread is about Windows!). Could this be simply because there are more Windows users reporting the problem here or is it because you're more likely to see the issue with Windows?

Confused definitely has the problem with the beta AIR with screensaver enabled, but it looks much better, if not fixed, with screensaver disabled. This is great as it's a cast iron case where we have evidence that it can be caused by a screensaver. Of course it seems screensavers might not be the only program/cause that produces the problem but it's a start at least.

This is with a simple screen-blanking screensaver right? It can be confusing because the there might be a cpu-hogging screensaver running in the background but not being seen due to the power options just turning the display off. Or it might be simply the display being powered off - no screensaver.

I have no screensaver set, but display turns off after 15 minutes (high performance power settings with a few customizations). I do however suffer dropouts, but no white noise so it's possible my issue is related, but not identical. I can't tie in these dropouts with the display turning off - I might try this today. I notice I have USB set to selectively suspend which could be another possible culprit.

To clarify, when I have previously referred to a “screensaver” this is simply the default settings of a 2014 Windows 8.1 HP PC, that is the screen powers down / goes blank if no keys are pressed or the mouse not used for 10 minutes. So not a “screensaver” as such, in terms of having some fancy graphic flying around the screen, or similar.

(05-Jan-2015, 06:53)philoouu Wrote: [ -> ]

(04-Jan-2015, 18:23)Mohmm Wrote: [ -> ]The principle is: "If there is no problem tweak your set-up so long as to produce symptoms."

Hello


You are not using Yosemite, you are still on Mountain or Maverick so the problem did not exist at that time….

OK that's true.

Jumping the bandwagon of new OSs is something that I also did in the past. Today it costs me some energy and control not to give in to the temptation (out of curiosity and the urge for trying the new). During the last iterations of the OS I have been learning the problems of incompatible apps the hard way and try not to change a winning horse.

But still I doubt noise on the Mac is a universal phenomenon if you use AIR 2.1.3

Cheers
Monique

@Confused, I forgot to ask you: with the latest beta version of AIR, were you able to recreate the white noise at will by waking the display from the screensaver mode? Or did you just observe that the white noise frequently coincided with the display turning on. The reason I ask is that there may potentially be another Windows process both waking the screensaver and causing the noise.

(05-Jan-2015, 16:27)AlexS Wrote: [ -> ]@Confused, I forgot to ask you: with the latest beta version of AIR, were you able to recreate the white noise at will by waking the display from the screensaver mode? Or did you just observe that the white noise frequently coincided with the display turning on. The reason I ask is that there may potentially be another Windows process both waking the screensaver and causing the noise.

Not much data with respect to this point! Only once did it make it through the screensaver activating without white noise occurring. On this occasion, “waking” the screen did not cause any issues, not so much as a stutter. Every other time it went into screensaver mode I got the white noise, which simply continued once the screen was reactivated.

Did a little testing.
Windows 7 64-bit, AIR 2.1.2 (yes, I just realised I'm a version behind the current GA release ), JRMC 19 running into the Devialet AIR (WASAPI) device. In the test I was running the JRMC 'Media Server', not the interactive JRMC app. JRMC is accessing my library on a Qnap NAS which is mounted as a shared drive. Everything is hard-wired ethernet. I was using the same 16/44.1 FLAC material throughout. There was no screensaver set.

First test with High Performance power option profile with display set to turn of after 15 minutes caused a 10 second dropout bang on 15 minutes.
I tried again - and got no dropout on 15 minutes. Waited a little. still no dropouts.

I reduced display turn-off period to 5 minutes & restarted AIR. Curiously I got a 10 second (or so) dropout after 10 minutes. I used the PC and restarted AIR, then got a 5 second or so dropout after 9 minutes. I moved the mouse on the PC (no AIR restart) and got a dropout after 8 minutes for about 5 secs.

I then reduced the display turn-off to 2 minutes. Got a 5sec dropout after 3 minutes.

I then reduced the display turn-off to 1 minute and tested multiple times. Sometimes I saw a dropout on the exact 2 minute boundary (I was using a stopwatch by now), other times it could take 2-3 minutes for a dropout. It seemed to always occur within 2-3 minutes. I didn't need to stop/restart AIR to reproduce.


So there seems to be some relation to the display turn-off but strangely that's not the whole story. It seems as if the dropouts only occurred when the display has been turned off by the OS power saver, and then they occur either exactly when that kicks in or some random time after, though that 'random' time could appear to be exactly 1 minute after so it may be occurring at some fixed poll time.
I then started looking at whether something else could be causing it - I tried pinging the ethernet interface on the PC and curiously I got no dropouts whatsoever while I was doing this, though I only tested a couple of times.

I then changed the setup to iTunes into AIR using ALAC files on the local disk. Not a single dropout or white noise

I'm going to upgrade AIR to 2.1.3 and try again. There's nothing useful-looking in the Windows event logs. I might try and attach a debugger to AIR too and see if that gives a clue, even though I don't have the symbol table.

(05-Jan-2015, 17:50)Rufus McDufus Wrote: [ -> ]Did a little testing.
...
So there seems to be some relation to the display turn-off but strangely that's not the whole story.
...

What are your pre-buffering and buffering settings in JRiver? What is your buffer in AIR? The values you have are probably good for debugging, but if you attempt to cure the problem, you may want to increase all of the mentioned buffers.
The problems you describe are bewildering in that the screensaver should not impact the playback to this extent, especially in the high performance mode. They say the reason the screensaver has any effect is that the CPU utilization suddenly drops/increases. But in the high performance mode the CPU should be more immune to these fluctuations. Another cure (I'm not sure how effective it is, but I have it enabled) is to give higher CPU priority to background processes than to the programs.
PS: I actually suspect that your drop-outs have to do with your NAS/JRiver interaction. My gut feeling is that there is no connection between them and the white noise problem we're discussing/debugging here. You may benefit from posting on the JRiver user forum. They've been very responsive to my questions.

(05-Jan-2015, 16:55)Confused Wrote: [ -> ]

(05-Jan-2015, 16:27)AlexS Wrote: [ -> ]@Confused, I forgot to ask you: with the latest beta version of AIR, were you able to recreate the white noise at will by waking the display from the screensaver mode? Or did you just observe that the white noise frequently coincided with the display turning on. The reason I ask is that there may potentially be another Windows process both waking the screensaver and causing the noise.

Not much data with respect to this point! Only once did it make it through the screensaver activating without white noise occurring. On this occasion, “waking” the screen did not cause any issues, not so much as a stutter. Every other time it went into screensaver mode I got the white noise, which simply continued once the screen was reactivated.

But see what matters here is whether you can MANUALLY induce the problem. You're probably tired of testing by now, but if you could manually wake up the monitor from sleep, and that yielded white noise, then there is a direct relationship between the screen waking and the noise. On the hand, if this does not create the problem, it probably means there is a 3rd party process both waking the screen and causing the noise.
I understand the opposite (the monitor going to sleep) did not cause problems. Is that correct?

AlexS, I have to admit my hunch for the dropouts is inclined towards the added complication of accessing music files from a shared drive from the NAS. I'm not actually trying to solve my own problem here but to consistently reproduce it to assist in debugging. I'm not actually a regular AIR user as I tend to prefer an external source.
I'd like to reproduce the actual white noise issue (as opposed to the silent dropouts I hear) and I think my efforts may be better directed at trying to reproduce with a straight iTunes/local music/AIR on the PC as recommended by Devialet.

Oh, I'm using all the defaults for buffer sizes.

I upgraded from AIR 2.1.2 to 2.1.3 yesterday evening and had a lot more trouble even getting dropouts, which was either odd or promising. I've had one or two dropouts but it seems far less consistent & predictable than with 2.1.2