gregb wrote:
> The recent discussion about frame size and sync troubleshooting makes me
> wonder about the how other system integrators convince themselves of
> implementation correctness. Are there tools to build a deterministic
> "bit perfect" data stream in WAV format, compare the output of a
lintweaker wrote:
> I just tested the Mytek on the wandboard with csos (cm51). It works
> great!
Did you have to do anything other than plug the DAC in? Didn't need to
blacklist the generic driver? lsmod confirms you are actually using
snd-usb-mytek.ko? logging from dmesg confirms firmware is lo
Triode wrote:
> So if this is causing 100% cpu then there is something more significant
> happening? What were the circumstances?
Code:
[18:10:38.690246] process_newstream:192 resampling from 96000 -> 384000
[18:10:38.690561] process_newstream:239 resampling with
JackOfAll wrote:
> Works for me.
Try it now.
Triode's Profile: http://forums.slimdevices.com/member.php?userid=17
View this thread: http://forums.slimdevices.com/showthread.php?t=98544
___
Thanks Adrian and Clive, I got the resampling working today. (loaded R4
then updated squeezelite and webgui)
It really works and does make a significant improvement over the builtin
interpolation filters. It even works when still using a builtin filter
but still upsampling 2x or 4x. Even with jus
JackOfAll wrote:
> I've been chasing XRUN's on the WB.
>
> >
Code:
> >
> [09:56:38.405777] sendSTAT:151 STAT: STMf
> [09:56:38.405868] codec_open:176 codec open: 'f'
> [09:56:38.405936] stream_file:316 opening local file:
/storage/music/lossless/The_Doors/L_
Triode wrote:
>
> Seems to me we have two options:
> 1) separate resample per track which will result in loss of filter state
> over a gapless transition
> 2) one resample for all tracks, but need to make sure we drain at end of
> a playlist. Downside is that track start marks may not be in the
JackOfAll wrote:
> Triode,
>
> I've spent a couple of hours this afternoon, seemingly chasing rainbows,
> trying to track down why the total number of frames coming out of the
> resample process (per track) did not equal the exact multiple of the
> input frames * oversampling ratio Something
Triode,
I've spent a couple of hours this afternoon, seemingly chasing rainbows,
trying to track down why the total number of samples coming out of the
resample process (per track) did not equal the exact multiple of the
input * oversampling ratio Something cranky is going on after a
period o
KERNEL-WAND-3.0.35-CM50
Changes...
Code:
* Fri Jun 07 2013 - 3.0.35-cm50
- Config changes for SATA support.
Add CONFIG_IMX_HAVE_PLATFORM_AHCI=y
Add CONFIG_ATA=y
Add CONFIG_ATA_VERBOSE_ERROR=y
Add CONFIG_SATA_PMP=y
Add CONFIG_SATA_AHCI=y
Add CONFIG_SATA_A
Triode,
"process.out_frames = odone" in decode._drain() before calling
_write_samples();
Code:
--- /home/clivem/development/git/squeezelite/process.c2013-06-06
19:12:54.896223918 +0100
+++ process.c 2013-06-07 13:24:07.508988610 +0100
@@ -106,10 +111,21
JackOfAll wrote:
> I tried this last night and was having issues with drop-outs at all
> sample rates. I'm trying to remember if I ever used the Amanero with a
> hub or just the WaveIO. It may be a regression in the kernel over the
> last month or so. I'm not sure. It's going to take more time to
JackOfAll wrote:
> KERNEL-WAND-3.0.35-CM49
>
> This update adds the mytekusb2 kernel module. (And fixes the ethernet
> breakage in cm48.) You'll need to install the firmware package (see post
> above) to use the Mytek driver. I don't have a Mytek DAC, so would
> appreciate a thumbs up from anyon
vzs wrote:
> In between did you manage to test Wandboard with Amanero + USB drive
> connected through an USB hub?
I tried this last night and was having issues with drop-outs at all
sample rates. I'm trying to remember if I ever used the Amanero with a
hub or just the WaveIO. It may be a regress
SQUEEZELITE-1.1-2.12.20130606GITFC34DA0
Changes...
Code:
* Thu Jun 06 2013 - 1.1-2.12.20130606gitfc34da0
- Add squeezelite-process-logging.patch.
- Drop squeezelite-process-cnt.patch.
- Latest git revision fc34da0.
- check free bytes not frames.
---
JackOfAll wrote:
> decode.c .
>
> size_t min_space = decode.direct ? codec->min_space :
> process.max_out_frames;
>
> Shouldn't that be (8 bytes in a frame)
>
> size_t min_space = decode.direct ? codec->min_space :
> (process.max_out_frames * 8);
Oops - fixed in repo - multiply by BY
JJZolx wrote:
> Will it be possible to enable the quad board's SATA port and use it for
> LMS running on the Wandboard?
I expect so. (Definitive answer once I have a quad CPU board in my
hands.)
JackOfAll's Profile: http:
Triode wrote:
> What are you seeing dropped frames on - WB? Is it dropping in this
> place - if so can you confirm if it is mid decode or at the end of the
> decode? We should never get to that code in theory.. Decode should not
> run unless there is space in the output buffer for the size of
Chunkywizard wrote:
> > JJZolx wrote:
> > Will this run on the new Wandboard Quad? Four 1 GHz cores, 2GB RAM,
> > better graphics, plus a SATA header.
> >
> > http://wandboard.org/
> >
> > (and should the community Squeezebox DAC card under development run on
> > it?)> >
>
> Yes and yes. By
JackOfAll wrote:
> Triode,
>
> Just tracking down why I'm dropping frames on the floor from time to
> time. Please can we reset the cnt (from 10 down to zero) back to 10
> after we have (partially) written frames to the out buffer.
>
> >
Code:
> >
> > >
>
> What a
SQUEEZELITE-1.1-2.11.20130606GITDE4F3A1
Changes...
Code:
* Thu Jun 06 2013 - 1.1-2.11.20130606gitde4f3a1
- Add squeezelite-process-cnt.patch (reset cnt to 10 after write some frames)
- Drop squeezelite-process.patch
- Latest git revision de4f3a1.
- add patch
Triode,
Just tracking down why I'm dropping frames on the floor from time to
time. Please can we reset the cnt (from 10 down to zero) back to 10
after we have written some frames to the out buffer.
Code:
--- /home/clivem/development/git/squeezelite/process.c2
John, not had a lot of time today. Started working on a web page to
document the upsample options. 'Squeezelite: Upsample functionality
using the SoX Resampler library'
(http://www.communitysqueeze.org/squeezelite_upsample.jsp). Hopefully
I'll be able to finish it later this evening.
---
JackOfAll wrote:
> Just to add to that, you log in using putty as username "fedora" and
> password "fedora". Then once you have a terminal, you type (or cut and
> paste) the commands, one per line, "enter" key after each one, waiting
> for the previous command to finish and the prompt to reappear
kesey wrote:
> On your Windows PC download putty.exe
> http://www.chiark.greenend.org.uk/~sgtatham/putty/download.html
>
> and off you go:-) You'll need to get the ip address of your Wandboard as
> per the instructions on
> http://www.communitysqueeze.org/quick_start_guide.jsp
Just to add to th
kesey wrote:
> On your Windows PC download putty.exe
> http://www.chiark.greenend.org.uk/~sgtatham/putty/download.html
>
> and off you go:-) You'll need to get the ip address of your Wandboard as
> per the instructions on
> http://www.communitysqueeze.org/quick_start_guide.jsp
Thanks for the he
Squeezed_Rotel wrote:
> I have a CSOS Wand up and running. I currently have the R4 image
> istalled. I want to install the kernel build (kernel-wand-3.0.35-cm49)
> and firmware (mytekusb2-firmware-1.33.10-1) that Clive just built. I
> would like to learn how to open a CMD prompt. Do I use my Win
I have a CSOS Wand up and running. I currently have the R4 image
istalled. I want to install the kernel build (kernel-wand-3.0.35-cm49)
and firmware (mytekusb2-firmware-1.33.10-1) that Clive just built. I
would like to learn how to open a CMD prompt. Do I use my Windows PC
with some linux type ap
LOGITECHMEDIASERVER-7.8.0-0.27.20130605GIT1369983318
Brian, I have just packaged this. Haven't installed or tested it myself
yet.
Changes...
Code:
* Wed Jun 05 2013 - 7.8.0-0.27.20130605git1369983318
- Latest git revision 5c5e178e3ce33ede6fae559ad2ba37fa844b4a78.
JohnSwenson wrote:
> Are these flags by any chance the flags in the "soxr_quality_spec"
> struct?
Yes.
Code:
#define SOXR_ROLLOFF_SMALL 0u/* <= 0.01 dB */
#define SOXR_ROLLOFF_MEDIUM1u/* <= 0.35 dB */
#define SOXR_ROLLOFF_NONE 2u/* For
Can you please test your Mytek DAC with the Mytek driver and let me know
it is working, please. I'm guessing that when you tested it before it
would have been using the generic usb driver and limited to 96k?
Code:
sudo csos-cleanInstall-testing kernel-wand-3.0.35-cm49
JackOfAll wrote:
> SQUEEZELITE-1.1-2.10.20130605GITA0A1828
>
GREAT! Hopefully I can have some time Thursday night to test this. Just
one question: what is the definition of the "flags" field?
Thanks,
John S.
JohnSwens
Hi Clive
Just notice this morning the time displayed on my radio is about 10
minutes out so my alarm will go off ten minutes early. I guess there is
no time server syncing, can this be added? If you need any more info let
me know.
CW
---
Triode wrote:
> Squeezelite resample branch should now support all codecs and
> dynamically load the library.
>
> John - I think we are at the stage where we can assume resampling is
> available from squeezelite - seems to work well, but interested if you
> have managed to test?
Hi Triode,
I ha
JohnSwenson wrote:
> GREAT! Hopefully I can have some time Thursday night to test this. Just
> one question: what is the definition of the "flags" field?
>
> Thanks,
>
> John S.
Are these flags by any chance the flags in the "soxr_quality_spec"
struct?
When constructing a quality field woul
To Triode, JackOfAll and JohnSwenson.
I consider that amazing progress is being made in the Community Squeeze
Player and OS projects. The audio hardware stuff is necessarily a less
public activity at this stage than the software, but I remain convinced
that great work is going on there. The state
SQUEEZELITE-1.1-2.10.20130605GITA0A1828
Changes...
Code:
* Wed Jun 05 2013 - 1.1-2.10.20130605gita0a1828
- Updated squeezelite-flags.patch.
(flags, precision, passband_end, stopband_start, phase_response)
Update...
Code:
---
Clive - latest squeezelite should take simple options for resampling:
-u vMs|1 should be your preferred option -v -M -s scale -1 dB
-u will select the default recipe with no scaling
Triode's Profile: http://forums.slimde
JohnSwenson wrote:
> That's exactly what I want!
My mind reading skills are improving! :) I've got a couple more loose
ends to deal with, then I'll test and make a new build.
JackOfAll's Profile: http://forums.slimdevices
Triode wrote:
> Clive - latest squeezelite should take simple options for resampling:
> -u vMs|1 should be your preferred option -v -M -s scale -1 dB
> -u will select the default recipe with no scaling
I have another patch (which I need to test) which allows setting all the
quality options dire
JohnSwenson wrote:
> Hi Triode,
> I have been laying out boards every waking moment for the last couple
> weeks, I've been having dreams like TRON only I'm inside multilayer
> boards trying to find processor pins to connect to! Actually kind of a
> cross between TRON and Avatar. Very strange.
>
Chunkywizard wrote:
>
> [edit] whoops. Major faux pas, it was my WHS that was out, just ignore
> me...
>
LOL. Chrony should be keeping the hw clock in sync with network time.
JackOfAll's Profile: http://forums.slimdevic
vzs wrote:
> In between did you manage to test Wandboard with Amanero + USB drive
> connected through an USB hub?
No, sorry! It's on the list for today.
JackOfAll's Profile: http://forums.slimdevices.com/member.php?userid
COMMUNITY-SQUEEZE-WEB-1-56.20130605GITD0D73C4
Changes...
Code:
* Wed Jun 05 2013 - 1-56.20130605gitd0d73c4
- Latest git revision d0d73c4.
- Add textfield to set squeezelite upsample options.
Update...
Code:
su
JackOfAll wrote:
>
> For Triode...
> squeezelite-flags.patch
>
Thanks - commited to repo.
Triode's Profile: http://forums.slimdevices.com/member.php?userid=17
View this thread: http://forums.slimdevices.com/showthread.p
JackOfAll wrote:
> I have another patch (which I need to test) which allows setting all the
> quality options directly. Sooner or later, (even if John doesn't ask for
> it), someone will want to set precision, passband, stopband, phase
> values directly, rather than using the recipe defines.
>
>
JackOfAll wrote:
> I've had a chance to compare stuff I've statically upsampled, upsampled
> on-the-fly (backend with sox exe), to libsoxr upsampling in squeezelite.
> We're good!
>
> I've also verified that using the ondemand (rather than performance) CPU
> governor is not a problem. (Didn't th
Triode wrote:
> Try now anyway - updated the code for mad and faad with resampling.
> [will go back and clean up once I have all the decodes working if the
> cpu load is ok]
CPU load is OK with everything I have thrown at it: flac, aac, mp3, ogg.
(As an aside, CPU load is less using libsoxr tha
Squeezelite resample branch should now support all codecs and
dynamically load the library.
John - I think we are at the stage where we can assume resampling is
available from squeezelite - seems to work well, but interested if you
have managed to test?
-
JackOfAll wrote:
> >
Code:
> >
> faac -q 500 -w -s -o track.m4a track.wav
>
> >
> results in LC encoded aac in mpeg4 container that
> squeezelite->libfaad will decode.
What were you generating before - was it playable by touch? If so d
JackOfAll wrote:
> Standalone faad bin can decode what I'm creating. squeezelite libfaad
> can't.
>
Code:
faac -q 500 -w -s -o track.m4a track.wav
results in LC encoded aac in mpeg4 container that squeezelite->libfaad
will decode.
---
KERNEL-WAND-3.0.35-CM49
This update adds the mytekusb2 kernel module. (And fixes the ethernet
breakage in cm48.) You'll need to install the firmware package (see post
above) to use the Mytek driver. I don't have a Mytek DAC, so would
appreciate a thumbs up from anyone who does and can confirm it w
MYTEKUSB2-FIRMWARE-1.33.10-1
Changes...
Code:
* Mon Jun 03 2013 - 1.33.10-1
- Initial release.
Install...
Code:
sudo csos-cleanInstall-testing mytekusb2-firmware
---
KERNEL-WAND-3.0.35-CM48
This adds the mytekusb2 kernel module. To use it, you will also need to
install the mytekusb2-firmware package.
Changes...
Code:
* Mon Jun 03 2013 - 3.0.35-cm48
- Add mytekusb2 driver.
Update...
Code:
-
gregb wrote:
> LMS 7.8.0 (on CSOS R5) is not consistently showing album artwork. Any
> hints?
Look back earlier in the thread. Someone else was having problems with a
corrupted cache, that was fixed by deleting the sqlite db and rescan.
-
Perhaps a simple question - I'm running CSOS R5 with LMS 7.8.0
(1369215084 @ Thu May 23 10:14:48 BST 2013).
LMS 7.8.0 (on CSOS R5) is not consistently showing album artwork.
Whereas LMS 7.5.2 (r31632 @ Mon Dec 13 13:07:13 PST 2010) on ReadyNAS
seems to consistently show artwork.
Any hints? Thanks
SQUEEZELITE-1.1-2.7.20130602GIT24C915E
Changes...
Code:
* Sun Jun 02 2013 - 1.1-2.7.20130602git24c915e
- Latest git revision 24c915e.
Update...
Code:
sudo csos-cleanUpdate-testing squeezelite
Standalone faad bin can decode what I'm creating. squeezelite libfaad
can't.
Code:
[22:23:10.989336] codec_open:169 codec open: 'a'
[22:23:10.989398] codec_open:182 closing codec
[22:23:10.989473] faad_open:484 opening mp4 stream
[22:23:10.989541] stream_file:31
JackOfAll wrote:
> Yep, I realise there wouldn't be upsampling. Was just trying to get aac
> to play. I don't have any native material myself so was trying to create
> some from wav files with ffmpeg / faac.
You need to create mp4 files. [see the faad.c file to see the pain that
that causes!]
Triode wrote:
> There's no code for faad with resampling yet - have it working here but
> the test branch will be broken other than flac and wav.
Yep, I realise there wouldn't be upsampling. Was just trying to get aac
to play. I don't have any native material myself so was trying to create
some
JackOfAll wrote:
> H. libfaad doesn't want to decode anything created with ffmpeg or
> faac...
>
> >
Code:
> >
> faad_decode:345 error reading stream header
>
> >
There's no code for faad with resampling yet - have it working here bu
Triode wrote:
> Do you have some 320k aac material? I think the decode is the most cpu
> intensive. So that's probably the one to add support for next to see if
> lossy codec should be supported?
H. libfaad doesn't want to decode anything created with ffmpeg or
faac...
Code:
COMMUNITY-SQUEEZE-WEB-1-55.20130602GIT741BCB6
Changes...
Code:
* Thu May 30 2013 - 1-55.20130602git741bcb6
- Latest git revision 741bcb6.
- Add UPSAMPLE option to squeezelite config.
- Update from Ghandi. (Swedish lang properties.)
Upda
SQUEEZELITE-1.1-2.6.20130602GITDF9BDAE
Changes...
Code:
* Sun Jun 02 2013 - 1.1-2.6.20130602gitdf9bdae
- Add UPSAMPLE option to systemd squeezelite.service.
Update...
Code:
sudo csos-cleanUpdate-testing squeezel
SQUEEZELITE-1.1-2.5.20130602GITDF9BDAE
Changes...
Code:
* Sun Jun 02 2013 - 1.1-2.5.20130602gitdf9bdae
- squeezelite-process.patch
- Latest git revision df9bdae.
Update...
Code:
sudo csos-cleanUpdate-testing s
Triode wrote:
> I've made a change along similar lines to yours which I think will
> work. Could you try the latest googlecode version, or the patch which
> it includes:
>
Thanks, Triode, for such a swift and positive response. I did a git
pull to get your latest
Code:
Triode wrote:
> Sounds promsing... I've updated googlecode with anther version which
> should be gapless. Need to look at the cmd line options.
>
I'm going to be out for a few hours. Will look later this evening.
Triode wrote:
> Do you have some 320k aac material? I think the decode is the
JackOfAll wrote:
> Adrian, not that my opinion matters more or less than anyone else, but
> for the record I'm signing-off on this, doing the upsampling on the
> front-end, 352k8 / 384k, not a problem.
>
> I've had a chance to compare stuff I've statically upsampled, upsampled
> on-the-fly (back
Adrian, not that my opinion matters more or less than anyone else, but
for the record I'm signing-off on this, doing the upsampling on the
front-end, 352k8 / 384k, not a problem.
I've had a chance to compare stuff I've statically upsampled, upsampled
on-the-fly (backend with sox exe), to libsoxr
SQUEEZELITE-1.1-2.4.20130601GIT7FE386F
Changes...
Code:
* Sat Jun 01 2013 - 1.1-2.4.20130601git7fe386f
- Update upsample patch. (unsigned int -> size_t).
Update...
Code:
sudo csos-cleanUpdate-testing squeezelite
Adrian,
Not pretty I know. I'll leave it to you for an elegant solution.
If the track rate already equals the max we can resample to, don't
resample! (If you try to create the soxr process with input_rate ==
output_rate, it's SEGV's. ;))
The values that you pass into sox_process really do need
dsdreamer wrote:
> @Triode
>
> I see you are busy with upsampling schemes, but I wanted to report my
> issues in configuring different DACs using Jivelite.
>
> I had good success with a Musical Fidelity UAC2 VLink 192 interface, but
> utterly failed with an AudioEngine D2 UAC1 and also failed H
@Triode
I see you are busy with upsampling schemes, but I wanted to report my
issues in configuring different DACs using Jivelite.
I had good success with a Musical Fidelity UAC2 VLink 192 interface, but
utterly failed with an AudioEngine D2 UAC1 and also failed HiFimeDIY
Sabre DAC UAC1 device.
JackOfAll wrote:
> I tried to run a quick test on the WB with "-u". Got no output, so tried
> running it on my desktop from Eclipse debugger. Looks like a deadlock/
> resource problem on the output buffer mutex.
>
> Thread 1: slimproto_run() at slimproto.c:531 LOCK_O (wants and is
> waiting for
SQUEEZELITE-1.1-2.3.20130601GIT7FE386F
Changes...
Code:
* Sat Jun 01 2013 - 1.1-2.3.20130601git7fe386f
- Quick hack patch to drop/re-aquire output buffer mutex in process.c.
Update...
Code:
sudo csos-cleanUpdate
I tried to run a quick test on the WB with "-u". Got no output, so tried
running it on my desktop from Eclipse debugger. Looks like a deadlock on
the output buffer mutex.
Thread 1: slimproto_run() at slimproto.c:531 LOCK_O
Thread 3: output_thread() at output.c:662 LOCK
JackOfAll wrote:
> I've packaged it, but haven't actually tested myself. Have had enough
> for one day.
>
> SQUEEZELITE-1.1-2.2.20130601GIT7FE386F
>
> Changes...
> >
Code:
> >
> * Sat Jun 01 2013 - 1.1-2.2.20130601git7fe386f
> - Latest git revision 7fe386f.
> -
Triode wrote:
> Oops - try it now...
I've packaged it, but haven't actually tested myself. Have had enough
for one day.
SQUEEZELITE-1.1-2.2.20130601GIT7FE386F
Changes...
Code:
* Sat Jun 01 2013 - 1.1-2.2.20130601git7fe386f
- Latest git revision 7fe386f.
- Switc
JackOfAll wrote:
> Sure.
>
> Need to check-in process.c?
Oops - try it now...
Triode's Profile: http://forums.slimdevices.com/member.php?userid=17
View this thread: http://forums.slimdevices.com/showthread.php?t=98544
_
Triode wrote:
> I think my initial branch will do this, but its just upsampling with a
> power of 2 multipler to the highest rate below max sample rate. Is this
> logic ok?
Sure.
Triode wrote:
> Interested in what your tests are saying... Resample branch on
> googlecode should work for pcm a
JackOfAll wrote:
> I'll be interested if I can work out why. ;) I've now gone off at a bit
> of a tangent again with an exe from Example 3 that I have put a Java
> wrapper around, so I can switch between libsoxr re-sampling and sox exe
> re-sampling on the back-end.
>
> One other thought before
surfek wrote:
> Is it just the configuration or just CSOS LMS cannot support video? (if
> just config - then how to change it?)
Edit /etc/sysconfig/squeezeboxserver. Clue: the "SQUEEZEBOX_OPTIONS"
line contains the --no options. You probably just want to comment
out the line, restart the ser
I know I am not the first one, but I thought I should report on my
experience with the $42 UAE23 DAC from here:
http://hifimediy.com/index.php?route=product/product&product_id=83
First of all it came up working without a hitch as long as I configured
it via the Web Interface (it is not possible t
I think LMS is configured to not support video (it supports only 'music'
type of libraries, video and pictures are disabled)
Is it just the configuration or just CSOS LMS cannot support video? (if
just config - then how to change it?)
thanks for your help,
Michal
--
JackOfAll wrote:
> 'Changes'
> (https://github.com/CommunitySqueeze/Web-GUI/commit/d4946d7264557fc713286676f752fd5948a060e1)
> will be in next release. Thank you.
>
> PS. Please don't take this as a criticism. I really appreciate the diff.
> Any chance that in future you could use '-u' (unified)
Triode wrote:
> Hum - thats interesting - only just started to look at it, will be
> intested if you can work out why...
I'll be interested if I can work out why. ;) I've now gone off at a bit
of a tangent again with an exe from Example 3 that I have put a Java
wrapper around, so I can switch be
JackOfAll wrote:
> Triode,
>
> I'm trying to replicate the "rate -v -s -M" I'm using for the back-end
> resampling with the sox binary, using squeezelite and libsoxr. (For the
> moment, my quick hack is in the flac decoder, in the callback, running
> the decoded flac->PCM data through libsoxr be
Gandhi wrote:
>
> Please find attached the changes to ApplicationResources_sv.properties
> for updating it to the latest version (1-53).
>
'Changes'
(https://github.com/CommunitySqueeze/Web-GUI/commit/d4946d7264557fc713286676f752fd5948a060e1)
will be in next release. Thank you.
PS. Please do
Triode,
I'm trying to replicate the "rate -v -s -M" I'm using for the back-end
resampling with the sox binary, using squeezelite and libsoxr. (For the
moment, my quick hack is in the flac decoder, in the callback, running
the decoded flac->PCM data through libsoxr before writing it to the
output
JackOf All,
Please find attached the changes to ApplicationResources_sv.properties
for updating it to the latest version (1-53).
Thanks!
+---+
|Filename: ApplicationResource.properties_20130531_diff.txt |
|Download: http:
JackOfAll wrote:
> I'm back using an nfs share for media at the moment. Trying to remember
> if I run the Amanero on the hub when I was testing with a USB drive
> Don't think I did. It would have been (192k max) with a WaveIO and the
> USB drive, both connected via the hub. I wouldn't expect
vzs wrote:
>
> - Isn't 92.5% CPU usage a bit on the limit? I'm using a similar minimum
> phase filter on my Intel G620 PC + LMS7.8 "rate -v -M -b 87.5 352800"
> and on track start SoX uses even 45% of one of the cores.
>
Depends on how you look at it, I guess. On paper 10% usage of a single
co
JackOfAll wrote:
> COMMUNITY-SQUEEZE-WEB-1-53.20130530GIT4870C6E
>
> Primarily for Pascal...
>
> Changes...
> >
Code:
> >
> * Thu May 30 2013 - 1-53.20130530git4870c6e
> - Latest git revision 4870c6e.
> - Do not populate Squeezelite config MAC address (using wir
JackOfAll wrote:
> One word answer, yes. You'll be using one proc core exclusively for the
> up/over sampling 44k1->352k8 or 48k -> 384k.
> Example below with my server side sox re-sampling on the WB, 44k1 ->
> 352k8, very high quality, minimum phase.
That's wonderful news! I asked about this on
COMMUNITY-SQUEEZE-WEB-1-53.20130530GIT4870C6E
Primarily for Pascal...
Changes...
Code:
* Thu May 30 2013 - 1-53.20130530git4870c6e
- Latest git revision 4870c6e.
- Do not populate Squeezelite config MAC address (using wired interface
MAC) by default unless the
gregb wrote:
> I am suspicious the "newer" LMS mechanism of rotating files has fooled
> me to believe log file access via LMS web page was not working, I
> suspect LMS scan completed just before the file rotation, and LMS showed
> a blank page representing zero length log files. It's clumsy to ac
JJZolx wrote:
> Will this run on the new Wandboard Quad? Four 1 GHz cores, 2GB RAM,
> better graphics, plus a SATA header.
>
> http://wandboard.org/
>
> (and should the community Squeezebox DAC card under development run on
> it?)
Yes and yes. By the way the Quad runs fast DDR also (533MHz vs
I am suspicious the "newer" LMS mechanism of rotating files has fooled
me to believe log file access via LMS web page was not working, I
suspect LMS scan completed just before the file rotation, and LMS showed
a blank page representing zero length log files. It's clumsy to access
log files via ssh
gregb wrote:
> LMS log and scanner access via web page is working for me today (via
> Firefox 21; same browser as last week). Odd.
I am suspicious the "newer" LMS mechanism of rotating files has fooled
me to believe log file access via LMS web page was not working, I
suspect LMS scan completed ju
JohnSwenson wrote:
> Yes, I would consider a windowed SINC to be a simple filter. That's not
> to say that is actually going to be the exact function used. The ones
> that seem to have the most detrimental effect sonically are more complex
> functions designed to achieve higher stop band rejectio
701 - 800 of 1560 matches
Mail list logo