X32 Discussion

Re: Monitor out noise

Per Søvik said:
Rater pointless, because there is nothing to see, but here is music at -40dBFS, no sound at max monitor levels and finally the noise floor with the X32 turned off. Slope set to 0dB/octave and input level to -20dB, where the input started clipping at -40dBFS from the desk.
View attachment 6263View attachment 6264
Click to expand...

Hmmm, I feel like there's a lot to see there. Not much in the first one, which I removed from the quote, because the question is not what the console is doing with music going through it and masking the noise floor, but what it's doing when there's nothing there.

Your two pictures have lots of 20-30db spikes above the noise floor at 1.somethingkHz and above; why aren't those audible? And they would be more visible with a different scale.

What sampling rate is your console set to? I found that the obnoxious peak at 6.0kHz shifted to 5.somethingkHz when the sampling rate was 44.1kHz.

Also curious why your setup has all those spiky noises when the console is off. What happens when you disconnect from the console, and then when you disconnect the cable from your analyzer?

You are in Norway? You have 240v mains? Wonder if that makes a difference? I can test with 120 volts on hot and 120 volts on "neutral", but not 240v hot-neutral.

See, there was a lot in your pictures, probably more than I'm seeing.

Thanks,
Dan

Edit: For comparison, here is a pic of traces from several of my consoles superimposed upon each other.

IMG_0395.jpg

The 6kHz spike is "only" 9db or so above the rising noise floor and it's clearly audible.
 
Last edited:
Re: Monitor out noise

Dejan Vuckovic said:
OK, so this shows that Behringer at some point in the production became aware of the fault, changed something and fixed the problem. Why not tell us what, so we can fix it as well?
Click to expand...

They did tell us the issue, it's in this thread, start back at post 1 and get back to me in may...lol
As for self repair, it's called voiding the warranty, it's not a new concept either...
 
Re: Monitor out noise

Dejan Vuckovic said:
OK, so this shows that Behringer at some point in the production became aware of the fault, changed something and fixed the problem. Why not tell us what, so we can fix it as well?
Click to expand...

My desk was produced before any desks were delivered to customers, so either the problem has occured at a later date or they picked a good one for me.
 
Re: Monitor out noise

Dan Mortensen said:
Hmmm, I feel like there's a lot to see there. Not much in the first one, which I removed from the quote, because the question is not what the console is doing with music going through it and masking the noise floor, but what it's doing when there's nothing there.
Click to expand...
The first one was for reference, as a calibration so to speak.

Your two pictures have lots of 20-30db spikes above the noise floor at 1.somethingkHz and above; why aren't those audible? And they would be more visible with a different scale..
Click to expand...
What you see is the internal noise of the computer, I can't seem to get rid of it even though I route directly from the asio driver for the interface (Behringer UCA222), and it is audible from the internal sound card output. I guess I could fix it if I disabled everything, but I don't use the computer for this kind of measurement, so it is not normally a problem.
The peaks you see are under -72dB relative to -20dBu, in other words less than -92dBu, and thus not masking very much that is clearly audible at normal listening level gain. The -84dBu spike at 6KHz on your consoles would have been visible though.[/QUOTE]

What sampling rate is your console set to? I found that the obnoxious peak at 6.0kHz shifted to 5.somethingkHz when the sampling rate was 44.1kHz.
Click to expand...
I always run at 44.1, both to be compatible with playback and because I prefer recording at 44.1 unless recording for video. I know there is a small latency price to pay, but I'm hardly ever doing any in-ear monitoring, so that is not really an issue.

Also curious why your setup has all those spiky noises when the console is off. What happens when you disconnect from the console, and then when you disconnect the cable from your analyzer?.
Click to expand...
It is all inside my computer, so no difference.

You are in Norway? You have 240v mains? Wonder if that makes a difference? I can test with 120 volts on hot and 120 volts on "neutral", but not 240v hot-neutral.
Click to expand...
Yup, clean, green and fresh 240V straight from the power plant, going through a UPS just in case.

See, there was a lot in your pictures, probably more than I'm seeing..
Click to expand...
As I stated, rather pointless exercise because my measuring set-up of the day simply wasn't good enough.
However, turning the PRX535 to full gain on mic input should give a gain of 160 dB between dBu in and dB/1m SPL, thus translating a -84dbu spike into 76 dB SPL/1m, which is something I should be able to hear.

Edit: There is a gate on the PRX that might actually not open for such a weak signal even at full mic gain, so probably not the best speakers to verify absence of low level noise. :blush:
 
Last edited:
Re: abusing Cat5 and what to expect....

High speed digital data is what travels over Cat 5 (or 5e or 6).
The difference between the variations are subtle (to the naked eye) changes in the twist patterns of each of the pairs and the relationships between each of the pairs.
These changes were made to permit higher speed data.

If you disrupt these relationships, you may get data errors. In the digital audio world, data errors sound really bad.

Don't abuse digital cables!!

Just because there is still electrical continuity, it doesn't mean it will support the required data rate with an acceptable error rate.

With computers, you may not notice the errors as the protocols are designed to detect the errors and request a resend of the data. This happens fast enough that a person may not notice.

In digital audio it is real time and no resend is possible, so, the data is lost and you hear a bad noise.


Sent from my iPad HD
 
Re: Monitor out noise

Hi Per,

Thanks for the thorough response.

Based on it, though, I think we are talking about two different things, one of which does not prove or disprove the other. I'll comment inline as you did:

Per Søvik said:
The first one was for reference, as a calibration so to speak.
Click to expand...

While the value of calibration is obvious, to me the calibration is the noise of the console into a speaker as it would be in real life. I want to look at the noise, not how far down it is relative to something else. The starting point is that I can hear an obnoxious sound with the monitor attenuator in its normal position and no signal going through the console. The relation of interest is the obnoxious noise to the acceptable noise floor, which I measured as around 9db. Taking that spike down 9 db or so with an outboard equalizer put it right into the noise floor.

Per Søvik said:
What you see is the internal noise of the computer...

The peaks you see are under -72dB relative to -20dBu, in other words less than -92dBu, and thus not masking very much that is clearly audible at normal listening level gain. The -84dBu spike at 6KHz on your consoles would have been visible though.
Click to expand...

I differ with you here. Your random spikes of 20 or so db are clearly visually masking whatever else is there. I don't think anyone can say with certainty what is in your plot and what isn't there, given this. Oh, and "my" spike wasn't 84dbu, it measured that when I shifted the traces up to be more in the middle of the window (I *think* that's what I did; have to fire up the SIM to be sure.) I don't think it matters, as I don't care about the absolute noise floor, post- deciding that it's audibly acceptable. As said earlier, I was more interested in the relationship between the good noise and the bad noise, which stays the same no matter where the traces are on the graph.

To reiterate, I certainly agree that the noise floor of the rest of the console is more than acceptable, but I do hear that bad noise on the monitor out XLR's.

Per Søvik said:
I always run at 44.1, both to be compatible with playback and because I prefer recording at 44.1 unless recording for video. I know there is a small latency price to pay, but I'm hardly ever doing any in-ear monitoring, so that is not really an issue.
Click to expand...

Aha! Another reason your traces will look different than mine. I, too, subjectively felt that the nastiness at 44.1 was less objectionable than the nastiness at 48.0, so maybe you are benefitting from that? Can you try it at 48 and see if you then hear a more objectionable noise?

Per Søvik said:
It is all inside my computer, so no difference.
Click to expand...

And you get exactly the same pattern with no input cables, i.e. antennae, connected to your computer? How about on battery power?

Per Søvik said:
Yup, clean, green and fresh 240V straight from the power plant, going through a UPS just in case.
Click to expand...

It'll be the end of the week before I can set a console up, but I'll see if 240 made up of 2 x 120 makes a difference. It would have the advantages of balanced power...

And my measurements were NOT with a UPS, although show operation is. I'll also confirm that UPS (USV) presence or absence makes no difference in this, as I suspect.

Per Søvik said:
As I stated, rather pointless exercise because my measuring set-up of the day simply wasn't good enough.
However, turning the PRX535 to full gain on mic input should give a gain of 160 dB between dBu in and dB/1m SPL, thus translating a -84dbu spike into 76 dB SPL/1m, which is something I should be able to hear.

Edit: There is a gate on the PRX that might actually not open for such a weak signal even at full mic gain, so probably not the best speakers to verify absence of low level noise. :blush:
Click to expand...

Emphasis added.

Not pointless because it's not showing anything, but probably pointless for the reason you state that I politely emphasized.

Very useful, though, in that it helps us narrow down variables that could contribute to the problem. I haven't thought about 44.1 vs. 48 hardly at all.

Also useful because you have not conclusively shown that you DON'T have the problem AFAIC, which keeps my theory alive, at least temporarily, that all consoles have the problem. Not that I like it, but it's nice to be right once in a while, and it's rare enough. :?~:-?~:???:

Thanks,
Dan
 
Last edited:
Re: X32 Discussion

Thanks for this, Rob, but it begs the question:

Is it abusing a snake to stand on it or straighten out a kink? Or mildly kink it?

If so, that is a scary concept for live sound, one which has not been obvious with other digital snakes.

Is there something special about AES50 that is not special about other digital formats (if that is the right word)?



Rob Spence said:
Re: abusing Cat5 and what to expect....

High speed digital data is what travels over Cat 5 (or 5e or 6).
The difference between the variations are subtle (to the naked eye) changes in the twist patterns of each of the pairs and the relationships between each of the pairs.
These changes were made to permit higher speed data.

If you disrupt these relationships, you may get data errors. In the digital audio world, data errors sound really bad.

Don't abuse digital cables!!

Just because there is still electrical continuity, it doesn't mean it will support the required data rate with an acceptable error rate.

With computers, you may not notice the errors as the protocols are designed to detect the errors and request a resend of the data. This happens fast enough that a person may not notice.

In digital audio it is real time and no resend is possible, so, the data is lost and you hear a bad noise.


Sent from my iPad HD
Click to expand...
 
Last edited:
Re: Monitor out noise

Hello

pictures two and three have spikes at different frequencies - what happens if you short the input of your measuring system ?


For whatever it is worth - I received my console mid-august as soon they became available - and only problem I have is ch16/32 "helicopter" - mainly on ch 32

Monitor output and headphone output are quiet besides normal even hiss, when cranked up.

I run my system @ 44,1
I switched it to 48 just to try, but there was no difference.

While playing around with "helicopter" I noticed, that if ch 32 is all full gain / fader full up ( which naturally is no real life situation...) switching channel select buttons becom VERY audible. So the cable going vertically inside console next to ch 16/32 must carry some serious amount of data, which radiates some and the preamps are capable of picking it up.


Terminating the input with 600 ohms resistor makes "chopper" go away leaving low-level whining. I pumped channel 32 compressor gain up 24dB to get some reading on X32 master meter. Without input termination ch 32 noise is @ -8dB and with termination -33dB. So proper termination takes care of 25dB - same goes for channel switching noise.


Some of these days - perhaps - I will open the console and reroute that one cable causing noise.



Nuuska
 
Re: Monitor out noise

Dan Mortensen said:
<snip>Also useful because you have not conclusively shown that you DON'T have the problem AFAIC, which keeps my theory alive, at least temporarily, that all consoles have the problem. Not that I like it, but it's nice to be right once in a while, and it's rare enough.
Click to expand...

Hi Dan, I totally get your point of view, although in my world it is the difference between max level/nominal level and any noise that is the important relationship. I will however try to make a measurement that actually shows the true noise floor of the monitor outs, and thus any noise in relation to that noise floor.
I don't disagree with you in your statement that all consoles might have this problem if you define the problem as being any discernible noise that rises above the white noise of the circuitry, but to me the problem is purely academical if that noise can't be heard. Since I'm able to turn the headphone outs to max without hearing a noise, and know without a doubt that at that setting my hearing will be severely suffering after a short time even at nominal level, and that 0dBFS at that setting will ruin both my headphones and my ears, I have to conclude beyond any doubt that my monitor outs are noise free for all intents and purposes.
As far as measuring methodology goes, you haven't stated (today and to my recollection) what were the conditions of your measurements and what kind of time they are averaged over. What I was testing was the outputs without anything prior to the main fader turned up, and without making scene changes and other adjustments that will run the faders or provoke any noise from effects etc. If there are particular conditions or operations that aggrevates the problem, I'll try to replicate those, and certainly run the test at both sampling frequencies.
 
Re: Monitor out noise

Timo Ulkuniemi said:
Hello

pictures two and three have spikes at different frequencies - what happens if you short the input of your measuring system ?
Click to expand...

The pictures show instant measurements (you can see the average at the lower frequencies if you look closely, and I might just turn off the instant trace next time), so some of it is random and some of it is drift that I see when I let it run for a while. Shorting the inpus have no effect, it is all internal to the computer and is caused by my inability to turn off the internal source of the noise, I'm sure I could fix the computer for measuring purposes if I took the time and didn't mind about rendering it temporarily useless for anything but routing signals from the asio to the spectrum analyzer.
 
Re: X32 Discussion

Dan Mortensen said:
Thanks for this, Rob, but it begs the question:

Is it abusing a snake to stand on it or straighten out a kink? Or mildly kink it?

If so, that is a scary concept for live sound, one which has not been obvious with other digital snakes.

Is there something special about AES50 that is not special about other digital formats (if that is the right word)?
Click to expand...

I don't think it is the aes50 format that is a problem or special, as the Ramcat cable I use works fine on a Midas aes50 system.
When I have had the problem the cable was not moving or kinked in any way. I can't reproduce the problem by wiggling the cable or ethercons either.
It would appear to be more of an issue in the console/stage racks. Not for all of us, only some of us!

Darren
 
Last edited:
Re: X32 Discussion

Dan Mortensen said:
Hi Pat,

Thanks for the description of a possible problem, but the question at hand is something happening anywhere in the middle of a cable, not at any connector.

The specific questions, quoted from the earlier post but reworded a little to maybe be clearer:

Do you or anyone else at Behringer or a company owned by Behringer have any knowledge of why an AES50 signal stream would be disrupted by some physical intrusion (bumping, squeezing, kinking, stepping on, etc.) occurring in the middle of a Cat 5 cable which does not break continuity on any conductors?

And I guess the question to be asked before that one is "Can this disruption happen?". I felt like it did in the experience I shared earlier, but maybe I'm wrong and it was something else?

Are there any other physical/environmental/electrical/RF interference factors we need to worry about when connecting our S16's to the console or each other?

Thanks,
Dan

Edit #2: Wait, maybe you were anwering the third question and not the first two? If so, my apologies for not guessing which one you were addressing. If not, the above stands. And your tag line "if this doesn't solve the issue..." made me think you thought it answered all questions.
Click to expand...

Hello Dan,

AES50 is a rock-solid connection and used by Midas with their Klark Teknik SuperMac technology. There are no known issues with the protocol and the technology. If ever this can only be related to cable or connection problems. The issue you are reporting Dan is an interesting one therefore I would like to learn more about your application to try and resolve this issue for you.
I will PM you straight away and we can discuss the matter in more detail.

Kind Regards
Jim Knowles
CARE EMEA / Tech Support
MUSIC Group / BEHRINGER
 
Suggested Modification to Producer Console

Hi.

Not sure if this is the right place for this, but I don't think there is a "Producer" thread yet.

I think the Producer would be much more usable if input channels 9-16 could appear on the right-hand set of faders. It would eliminate lots of layer changing while in regular mixing mode. And since there are no scribble strips you presently are having to double-label the eight input channels.

I realise this would need a hardware redesign, to acccomodate the extra layer button.

Does anyone else think this would be a good idea?

Thanks,
Mick Berg.
 
Last edited:
Top Bottom