Crackling/distortion issue with new D100

[ecsrun]

I have the D100 hooked up and I am having nothing but problems with it... It does not freeze up when switching tracks like the C50/C48 units I tried, but I get a bit of crackle/pop periodically when using the USB port in ASIO mode. Perhaps every 20-30 seconds, I will hear a brief, although extremely annoying, crackling sound. It is very low - not like a loud pop.

I already tried replacing the player software, the computer itself, every single cable... Added a furman power conditioner, etc etc. Every imaginable combination of buffer settings has been tried.. including completely different music players.. even tried comparing windows 7 to windows 8 to see if that made a difference.

The DAC is attached to a dedicated USB 2.0 highspeed controller that is not shared with any other devices in every computer that I test with. All computers I am testing with are very new, and extremely overkill for music playback. For example, core i7 + 16GB ram + SSD. I have the NAD M51 DAC sitting on my media cabinet right next to the D100. If I take the exact same belken gold 2.0 USB cable and attach it to the NAD, the playback is perfect... No crackle or pop.

The D100 has no problems when using optical or coaxial..

Very noticeable on the 192Khz Norah Jones track 'Carnival Town' when the volume is turned up to moderate listening levels, i.e. 70-80db.

The only different that I can detect between the NAD M51 and the D100 is that when I go into the device manager in windows... I can select the USB root hub and see the power usage for all devices.. with the McIntosh USB Audio device, it requires and draws '0 mA' from the bus. With the M51, there is a 10 mA power requirement and usage. Additionally, if you go into the properties for the EHCI host controller, the 'bandwidth used' by the McIntosh during playback never goes above 6%. The 'bandwidth used' by the NAD M51 hovers around 35%.

If anyone takes an interest, I am happy to post pictures or video.

Anyone else receive this unit and have any crackling in ASIO USB mode?

[redm]

Interesting problem. I suspect there are bound to be more of these little issues as the bugs get worked out. If your using it async and the NAD works fine, then its likely a problem with the D100, regardless of if it's specific to your config or not. Did you make a report to McIntosh yet?

[ecsrun]

I grabbed a USB analyzer tool.. appears that the D100 chipset is intended to be self powered. My guess is that this was done to try to isolate the D100 from the noise coming in via the USB power. I can confirm the device is reporting as being in high speed mode.

I still have no idea why it uses so much less bandwidth than the other DACs I have here at my house to compare it with...

[ecsrun]

That is a great question. My experience with contacting McIntosh directly via phone led me to Chuck. I am not entirely sure he is up to speed on digital technology - he asked me why I would want to use ASIO mode. I felt like saying "how about because I paid for it and it is written on the specs as a supported feature??!"

Ron appears to be the focal point for McIntosh digital, but I have not received a response from him dating back to my last email over the C50 problems that I had.

[benmoree]

If you are running a windows CPU have you downloaded and installed the USB driver for the d100?

[tweet]

Sounds like a ASIO driver/latency issue is causing buffer underruns. Try a google search for "PCI Latency Settings to Fix Clicking and Popping Problems"

[ecsrun]

Yes I am using the McIntosh D100 driver.

These are brand new PCs with literally nothing installed on them. Even windows was freshly installed for these tests. The behavior happens across 5 different machines... but all work with the NAD M51 flawlessly. However, the M51 is a completely different DAC chipset.....

I have been through a myriad of DACs over the last 2 months and this is the first time I have encountered this.

Regarding PCI latency.. or IRQ latency.. I already downloaded the DPC Latency check tool from Thesycon. All of my machines get flawless results.. usually between 50-100 microseconds.

[junker]

Originally Posted by ecsrun

he asked me why I would want to use ASIO mode. I felt like saying "how about because I paid for it and it is written on the specs as a supported feature??!

Not familiar with ASIO since I'm using a MacMini, but where on the D100 spec sheet or brochure does it say anything about supporting ASIO?

No problems in the heavy usage I've put mine through thus far. Takes a second to display the correct bit depth and sampling frequency on the display after starting playback, but other than that it's been perfectly reliable for me.

[ecsrun]

Two things I can try..

1) Different USB cable - perhaps the Kimber you are using? So far, I have only tried the audioquest carbon and a belken gold 2.0

2) Mac Mini - can you tell me which model of the mac mini you have? Since it is an i7 I am assuming it is the latest generation with no optical drive.. can you tell me specifically which USB port you are using on the back of the mini?

Thanks,

[junker]

Sure. Yeah it's a late 2012 model 2.3 GHz Quad-Core i7 with SSD and 8 GB. There are 4 USB ports on the back, and after checking "system information" and USB Prober from the dev site, I decided to to use #3 (looking from the front) or #2 when looking from the rear. I did see that it didn't draw any USB current as well. Both of my previous DACs (DragonFly and MicroMega MyDAC) both showed a draw of 50mA I believe.

Integer mode works perfectly with Audirvana+ v1.4x and directly from iTunes all the way up through 24/192KHz files. Also, use VLC, etc with no issues. Never ever had and cutting, popping, or anything of the sort.

On Windows I would start by testing without ASIO and only use the McIntosh driver (made by Thesycon?) Maybe there is a race condition or they are competing for a shared resource or something.

About the USB Cable... yeah, I like the basic Kimber Kable quite a bit. The conductors are 6% silver, the connectors are both soldered and crimped, and the sheath and connectors are well made. Nate Mansfield from Kimber suggested taking off the ferrite RF cores (they just pop off with a small screwdriver) unless you have RF issues. Another reason to keep the run short as possible. Their USB AG cable does't have them, except by special request. Some guys have been very excited about the MapleShade stuff too. Pretty wild designs for sure with almost no dielectric effect. The WireWorld Starlight and Silver StarLight cables also look nice IMO, but I tend to stay pretty loyal to Kimber. Ray picked-up the phone when I was calling to leave a message one Saturday in 2005 and literally hung out on the phone and helped me out for an hour or two. Great guy, great customer service, and it was clear that he loves what he does.

I'd be okay trying one of the Gold Belkins as well, but I wanted to keep the USB cable as short as possible and most of those are 6 foot. In fact, I almost wish I would have bought the 0.5m ($50) instead of the 1.0m ($60) - it would have been plenty enough for the way I have my media rack set-up currently. But when I get an MC452 and probably place the MacMini->D100->MC452 on individual wood stands the length might be needed. Now that I think about it... if I eventually find that I only need 0.5m I may just go ahead sell the the 1m Cu and buy the 0.5m Kimber USB Ag for $100 - it's the longer lengths that start to get crazy expensive. Thing to keep in mind about USB digital audio is that it is streamed and not a packet-based protocol, so the digital signal is essentially a high frequency analog signal where above a certain threshold it's a 1, and below it's a zero. Not technically 100% accurate but close enough. So, even though it is digital the specified length limit is 5m, and longer lengths could pick-up RF interference etc which would affect the signal. How much of it is audible? I'm sure at some level it is. I just want short, reliable, well-made, and ideally from an American company that I can stand behind.