I'm playing Apple Lossless files.

My setup: WD My Book Live, Raspyfi, WIFI adapter, Schiit Modi DAC

I've been playing the same songs for testing and been getting consistent skips playing songs.

ERROR: handle_hc_chhltd_intr_dma:2537: handle_hc_chhltd_intr_dma: Channnel 3, DMA Mode -- ChHltd set, but reason for halting is unknown, hcint 0x00000002, insts 0x04600001

9:1:1: cannot set freq 44100 to ep 0x5

Can you explain what's going on here?

I have returned to RC2 for the time being.

Cheers
Andy

I'd love to see this working so if I can help with testing of any sort let me know. I can see Pi is on its limits here. By the way is NFS an option to reduce CPU/LAN load for NAS shares?

The flies I have been testing with have all been 44.1/16 FLAC so there shouldn't be any need to resample with the DAC accepting 32, 44.1 & 48.

I'm not linux capable at code level but spend quite a bit of time setting up and maintaining Linux servers and VPS's. I'm also GMT +12 if this helps answer questions over this side of the globe.

Cheers
Andy

first of all, thanks for all the feedbacks.

We already found out the problem that causes the Database Update take so long, we're gonna release a fix soon. And also this causes the CPU to be extra loaded.

The values are so important because on the Raspberry PI the USB bus and Ethernet are shared, so it takes a careful otpimization to let its tiny resources handle this. And on RC2 you hadn't this problem because there was a resampling to 44/16 always activated. So there was no problem to handle higher resolutions (especially on Async DACS) .

So, having a bitperfect playback with the PI (I point out, its not a problem of raw computing power, but the way that USB and Ethernet are managed) requires all to be working perfect. We thought we were there, but as it points out, there are lot of combinations possible and thanks to your feedbacks I feel confident we will achieve better performances .

Mike - OK after an overnight database update the CPU load is down to 5% for MPD and CIFS is 1% so it was the database update causing the extra load.
I am checking this remotely so will check the popping when in the office.

You said try changing the buffer values up and down. Can you explain what is going on here and why the settings are so critical. If seems like the setup is tettering on not working yet RC2 appeared to be solid and worked with no tweaks after installing.

Cheers
Andy

Second: if you heard popping or noise, just go to Sources Settings -> advanced and try to gradually increase the values. If this worsens the situation, try with decreasing. This depends on your specific configuration, so there's not a definitve answer there...

As said before in another thread I also have glitches when playing music from NAS (through Ethernet) with RaspyFi 1.0 and my Lavry DA11.
Please acknowledge us of a fix for this issue.

Pinpon.

Decided to move my comments from the front page which was getting a bit messy.

This is the popping and skipping on VS1.0 but not returning to RC2. I mentioned that the SSH seems slower on 1.0. I have 2 SD cards now, one with 1.0 and one with RC2.

Looking at TOP I get the following approx CPU load figures whilst playing 44.1/16 bit FLAC from Ubuntu Samba over wired LAN:
RC2 - mpd ~2% / CIFS 1%
VS1.0 - mpd ~ 35% / CIFS `20%

To be fair the RC2 updated its database quite fast whilt the 1.0 was still going when looking at the TOP figures. However 1.0 seems smuch slower than RC2 which I am guessing is causing the problems.

Can you advise where to go now with what is causing the higer load.

As mentioned previously I have tried all the Kernet options.

For the moment I am sticking with RC2 as it works well for me but happy to test any ideas you have to get 1.0 up and running.