[GuitarBizarre]

For about a year now, the sound I've been getting from my Cambridge Audio DacMagic (not the 100, not the Plus, just the normal old version) has been doing strange things.

It started off with the DacMagic simply ceasing to produce sound - I would glance over at the DacMagic, and, invariably, the input Sampling rate light would be on 48KHz - Rather than the 44.1KHz that it was actually supposed to be being given. Sometimes, I would get a situation where this would happen, and I would get short bursts of sound about once a second. Whenever THIS happened, the light would jump back to 44.1KHz each time a burst occured, then go back to 48 and fall silent again.

Turning the unit off and on has always solved both of these problems, and the unit is quite a long way out of warranty, so to save buying a new unit I just kept using it, turning it off and on when needed.


Recently, however, I've been getting more problems - Sometimes, when doing things, PARTICULARLY when doing computer intensive things (I livestream videogames, so frequently tax my computer's power quite a lot), the sound will seem to be filled with glitches - tiny little pops and clicks, at random throughout the track. - Bear in mind, this DOES still happen when I'm not really taxing the computer at all, just less frequently.

The best comparison I can give to explain what this sounds like, is like a scratched record - it doesn't skip or jump, but at random intervals, anywhere from twice a second to once in 5 seconds, it'll sound like a little pop occured and the track will continue. There's no harsh noise or glitch sound, just a tiny break in playback accompanied by a click or pop - to me it sounds as though the click is the result of the gap in playback.

In an effort to try and fix this problem, I changed from using the USB cable to using a TOSLINK cable. Because as a side-effect, this allows me to use 24/96 output from the PC, I changed it to this, and was greeted with a new problem altogether - The sound now had a harsh, grainy "fizz" over the top of everything, and was completely unlistenable.

Convinced the DAC was faulty, I took it elsewhere and tested it with someone elses digital streamer - this was via Coax, (Ghetto coax, in fact - one half of a phono cable) rather than optical or USB, but these played fine, from both internet radio and some 24/96 FLAC files I'd put on a USB stick.

Having brought the unit home again, convinced it was my PC causing the issue, I plugged it in...and miraculously, perfect sound with TOSLINK 24/96 output! This lasted a few hours until I started, once again, getting the same tiny pops and clicks I was getting via USB. Some fiddling around, and a change of samplerate later, the grainy fizz was back as well! Restarted the computer, and sat typing this, all is well - for now. I have no confidence things will stay so idyllic however!


I'm somewhat at a loss - What is going on? Are these pops, clicks, and fizzes, and the random switching to 48KHz the fault of my PC or my DAC? How can I fix them? I can afford to buy a new DAC, but I'm loath to do so if I'm not sure it will solve the problem!

 

[NA Blur]

It sounds like something perhaps the music bit rate is causing the DAC to switch between modes.  I know my m903 will have some USB issues when running USB 2.0 in Windows, but I blame my PC more than anything.
 
Try going back to using USB and install the ASIO driver.  Try a few USB ports and you could even turn off the power management for the various USB HUB devices you have.
 
To do that:
Control Panel
System
Hardware Tab
Device Manager
Find the Universal Serial Bus Controllers section
Select the USB Root Hubs and Right Click
Properties
Power Management
Uncheck Power Management
 
Reboot and try again
 
Be sure to check out their site as it may help you find your fix.  In particular the sections on ASIO setup and Firmware upgrades.
http://www.cambridgeaudio.com/pages/support-centre
 
Let us know if you fix it.

 

[KiruDub]

I'm not a pro with DACs, so someone else might come along with the answer, but it does sound a bit like a buffer size or clocking issue. Here's snippet from a Sound on Sound article from back in '06:
 
"If the buffers are too small and the data runs out before Windows can get back to top them up (playback) or empty them (recording), you'll get a gap in the audio stream — a click or pop."
 
Here's the link: "http://www.soundonsound.com/sos/oct06/articles/pcmusician_1006.htm"
 
Tech has changed since then, but this might at least give you a better understanding what might be happening.

 

[GuitarBizarre]

I'm not tremendously knowledgable about the windows Audio environment, but I'm not convinced that is the case - simply for the fact the windows audio session mixes everything to 24/96 when set to that mode. Even when playing 44.1 audio in the media player, the DAC (When it is playing nice) displays 96KHz on the front, and doesn't stray from it regardless of the audio material being played.
 
If the PC were sending different bitrates (Or if it IS, and I'm misidentifying this), that might make sense, but 24/96 seems to be what windows sends when asked, and everything else gets mixed and/or upsampled.
 
Having checked the DirectSound instructions on that Cambridge Audio link, they seem to imply the same thing: "The sample rate will be locked to a predefined output rate set in the advanced section of the windows output setup"

 
 
Edit: Oh, and just one other thing - I produce music, and as such am somewhat familiar with what happens if a buffer is not large enough or if a VST is using too much CPU power - You're right it does sound like a pop or click, but it doesn't sound like this - Its much more obvious and distorted. This is less dramatic. It also wouldn't explain the weird grainy, metallic "Fizz" being present on 24/96 output, and also wouldn't explain why I only get that aforementioned fizz on 24/96, and only some of the time.
 
In fact, the PC, when I streamed earlier today, (Probably the most intensive thing this PC ever does), had perfect sound, set to 24/96, for about 20 minutes. Then pops and clicks started coming in at random - slight, but there. I didn't get the "Fizz" until after I finished streaming, and had changed the output samplerate to 24/48 then BACK to 24/96!

 

[goodvibes]

The only thing I can think of is if you have a player that is allowing you kernel, Wasapi or ASIO while runnign at the same time as the windows normal output. That could allow both 44 and 48 to reach the DAC. It could than lose the 44k lock when some Windows OS sound (48k) occurs in the background. I have no idea from small amount of playback info described if this fits but it can happen.

 

[GuitarBizarre]

The full audio chain is as follows:
 
Windows Audio settings - DirectSound (No ASIO, no WASAPI, no Kernel - I don't want applications taking exclusive control since I need to be able to listen to many simultaneous software audio sources and I believe the other options all have the potential to disable sound from applications other than the primary playback software.)
 
Windows Samplerates that have given me problems are 16/44.1, 24/44.1, and 24/96. I haven't tried the other sampling rates or bit-depths - 16/44.1 and 24/96 are the most standard, so should be the most compatible, right?
 
Source Material - Mixture of filetypes and compression types, literally 99% of the Library is 16 Bit 44.1KHz though. 
PC Media Player - Foobar2K. No DSP plugins, Audio output is set to "Primary Sound Driver".
Output - Have tried both USB and Toslink. Back panel ports for both. Power suggestion above seems odd to me - the DacMagic isn't USB powered.
 
Into the DacMagic, Steep filter, have tried all 3 inputs (Toslink for the SP/DIF inputs) and gotten the popping on all of them. Grainy Fizz only occurs when 24/96 output is selected, and USB is limited to 16/44.1 so it cannot occur on USB.
 
out of the DacMagic by RCA into NAD C326 amp into Mordaunt Short Mezzo 2 speakers.
 
 
Edit: The Fizz has returned! What was I doing? Nothing more intense than turning off a youtube video... I've just gone through all the supported output modes - ONLY 16/96 and 24/96 exhibit the fizz. Turning the DAC on and off doesn't stop it, changing samplerates multiple times doesn't stop it either. Restarting the computer... well, back in a minute.
 
 
EDIT 2 - Yep, as suspected, the restart has given me back what sounds like a perfectly working 24/96Khz playback mode.

 

[NA Blur]

Try turning the DAC volume / PC volume to full and use the amp to lower the volume.  I read somewhere there is a bug in XP that should be fixed in Vista and above regarding bit sampling.
 
You should at least install and setup the ASIO driver and see if the issue goes away.  You can always uninstall it later.

 

[GuitarBizarre]

The amp already controls the volume - My windows output is always on 100, and my foobar volume is NEARLY always on 100, and has been on 100 during this whole thread. As for the DacMagic - It has no volume control!
 
I am also already using Windows 7 and have been since 2009 when the system worked fine via USB.
 
As for ASIO drivers - I *cannot* use them on this machine for general playback - They forcibly disable playback from other audio sources when in use. I need this machine to be able to play, in some cases, game sounds, messenger sounds, streaming video, AND foobar music, simultaneously. Even if this did solve the problem, I simply wouldn't be able to use the resultant setup the way I have been using the machine in the past.
 
For the first two years or so of owning the product it worked flawlessly using the normal, DirectSound approach, at 16/44. Someone else here in a thread I read said he'd used his using Toslink at 24/96 with directsound, again with no problems at all. I really don't think messing around with ASIO drivers is going to ever get me where I want to go - I want to go back to directsound working, not forward to ASIO turning my daily use into a chore!
 
I'm really just looking to confirm - Which of the two items is causing my problem? Everything I can see settings wise in windows is perfect - exactly the way it should be according to everything I can find from Cambridge Audio.
 
If its the PC, I'll do a format, and hope and pray that the issue isn't hardware related.
 
If its the DAC, I'll buy a new one.

 

[GuitarBizarre]

Update - I have borrowed a DAC to use for a few days and will see if it has the same issues. Here's hoping not - If this DAC is fine I'll be confident enough in my own DAC's being faulty, that I'll buy a new one and avoid hours more troubleshooting!

 

[GuitarBizarre]

OK, do, using DPC Latency checker I've narrowed down the problems.
 
1 - The pops and clicks are DPC latency, caused by my streaming program, and there's really nothing I can do about it. Luckily it only happens when streaming.
 
2 - The other DAC worked eprfectly fine playing 96KHz material, but my personal DAC will still bug out and start giving a strange fizzy, clipping-style distortion. This ONLY happens on 96KHz settings in windows. Other sampling rates are perfectly fine...

 

[KiruDub]

Don't you love computers? Nothing beats an intermittent problem for hours of fun! 

 

[GuitarBizarre]

Well, it only does this on 96KHz material.
 
I suspect it is dying, slowly but surely, but I CAN still use it. I'll wait until it dies forever, unless someone has a great idea on how to fix it.
 

 

[anwaypasible]

it sounds like the circuit is aged.
what exactly is unclear.. maybe it is the DAC, maybe it is some capacitors (but could be something else i guess).
 
when you raise the sample rate to 96khz .. there should be more electrical need .. and if that need isnt met, you get the distortion.
but, if the DAC is the problem.. the only real way to know is to test the circuit leading up to the DAC (or simply change out the DAC for a different one)
 
reason why i think it is the circuit itself and not the DAC .. because when you used the coax input, the DAC worked fine you said.
and if the DAC worked fine, then chances are it really isnt the DAC
it's kinda like there are some portions of the circuit dedicated to the different input types, and they've aged to the point where problems occur.. but choosing the one input you havent been using is still 'fresh' or 'within the operating limit of the aged|weakened circuit'
and if that holds true, then switching to coax input only might allow the thing to work for another few years .. but i would expect the other inputs to slowly grow worse.
 
see.. when you said you tried the coax input and it worked fine, then you brought it home and everything worked for an hour .. that kinda tells me there is a slight problem with the circuit, and the whole thing shifted as if a gelatin.. and then when you went back to the other input, the gelatin slowly shifted back.
 
 
the only other thing i can really think of,
the coax cord places a different resistance onto the circuit, and the weakened circuit doesnt feel the resistance is a bother.
i dont know if the coax resistance is specific to the cord, or maybe the cord AND the soundcard it was connected to.
i also dont know if the resistance is identical from one piece of hardware to the next.
 
see.. the optical cable probably doesnt put resistance onto the circuit (and if it does.. the size of resistance is in a totally different league)
the usb input might be isolated from the rest of the circuit, but the isolation being used could also add to the resistance the circuit sees (as if it doesnt really 'turn on' until some electricity goes through and warms it up enough to trigger it on)
 
i am not one to think any capacitor will be a direct replacement for the old one, but that is because i think of the electron worm itself.. as well as the timing of the gaps between each worm being presented or accepted (as well as any phase or size alterations of the worm).
but
maybe you open the thing up, find some capacitors.. note the value, then find some replacements to solder in .. but as i said, maybe it isnt a bad capacitor but some other piece.
 
 
maybe you could do what i cant afford to do..
get a digital multimeter that can tell you if there is a small voltage difference.
because 12.0 vs 12.1 might not be enough when the difference is 12.002 vs 12.004
i would get some plastic to cover the tips of the multimeter probes to prevent the metal from touching anything else.. they are like training wheels to keep shaky hands from coming into contact with more than one thing causing more problems.
just be sure the plastic is soft enough (or has a hidden hole already put there) to allow you to simply press the probe onto the metal and then press down a little bit firm to remove the gap between probe and solder.
that way maybe you could test the voltage of the capacitor with the thing running and maybe come to realize the voltage is lower than it should be .. thus causing the DAC to work less than it should.
 
 
i figure a DAC working with less voltage than required is like a soldier running with more weight compared to another person that isnt carrying anything at all.
it is going to change how fast the person runs.. and it is going to change how well the person balances themselves.
but truth be told.. the real answer probably lies with how well the brain|personality of the runner enjoys the extra weight.
 
one more thing to be said.. maybe your motherboard is adjusting some electrical characteristics on the usb connection as if a premium service offered by the board manufacturer (or maybe an FCC requirement).. and that is why you are able to go from broke to working again, because the gelatin in the circuit is changing, and the usb connection is adjusting.. and when the two start to seperate as the gelatin slides, then the problem comes back.
you got to think, maybe when you reboot the computer the gelatin has some time to slide back or possibly 'recharge' itself and slowly drain again like a battery.