[OT] mini-PCIE with Broadcom Crystal HD Hardware Decoder (BCM970012) for HD playback with free drivers
Dear list,
I just share something I read on the XBMC [1] Web site.
The Broadcom Crystal HD Hardware Decoder (BCM970012) seems to be an alternative for HD playback with free drivers [2]. You can get it at as an mini-PCIE card at eBay for 22 $ [3].
If this is all true in my opinion these are fantastic news.
Thanks,
Paul
[1] http://www.xbmc.org/ [2] http://www.xbmc.org/davilla/2009/12/29/broadcom-crystal-hd-its-magic/ [3] http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=320436937576#ht_2174w...
2010/1/2 Paul Menzel paulepanter@users.sourceforge.net:
I just share something I read on the XBMC [1] Web site.
The Broadcom Crystal HD Hardware Decoder (BCM970012) seems to be an alternative for HD playback with free drivers [2]. You can get it at as an mini-PCIE card at eBay for 22 $ [3].
I'd recommend the BCM70015, which is the better part.
It works as a decoder only, making YUV 4:2:2 frames available, which must then be blitted to whatever output framebuffer setup you have, over the pci express bus. The API seems very straightforward, much simpler that VDPAU, thus one could get a HW accelerated H.264 setup without X11.
I just share something I read on the XBMC [1] Web site. The Broadcom Crystal HD Hardware Decoder (BCM970012) seems to be an alternative for HD playback with free drivers [2]. You can get it at as an mini-PCIE card at eBay for 22 $ [3].
I'd recommend the BCM70015, which is the better part.
where is it possible to buy it ?
It works as a decoder only, making YUV 4:2:2 frames available, which must then be blitted to whatever output framebuffer setup you have, over the pci express bus. The API seems very straightforward, much simpler that VDPAU, thus one could get a HW accelerated H.264 setup without X11.
what about of corrected interlace output ? Most of them videocards can't do it correctly....
Goga
2010/1/2 Goga777 goga777@bk.ru:
what about of corrected interlace output ? Most of them videocards can't do it correctly....
I can't answer that unfortunately.
For those that don't have a mini pci express port, this one might come in handy;
http://www.bplus.com.tw/Adapter/MP1.html
what about of corrected interlace output ? Most of them videocards can't do it correctly....
I can't answer that unfortunately.
let's hope that Crystal HD can do deinterlacing and scaling
in crystalhd/include/7411d.h there's lines
/* scaling on/off */ /* - eCMD_C011_DEC_CHAN_OUTPUT_FORMAT */ typedef enum { eC011_SCALING_OFF = 0x00000000, eC011_SCALING_ON = 0x00000001,
} eC011_SCALING;
/* deinterlacing on/off */ /* - eCMD_C011_DEC_CHAN_OUTPUT_FORMAT */ typedef enum { eC011_DEINTERLACING_OFF = 0x00000000, eC011_DEINTERLACING_ON = 0x00000001,
} eC011_DEINTERLACING;
/* deinterlacing on/off */ /* - eCMD_C011_DEC_CHAN_OUTPUT_FORMAT */ typedef enum { eC011_DEINTERLACING_OFF = 0x00000000, eC011_DEINTERLACING_ON = 0x00000001,
} eC011_DEINTERLACING;
On Sat, Jan 02, 2010 at 08:22:59PM +0300, Goga777 wrote:
what about of corrected interlace output ? Most of them videocards can't do it correctly....
I can't answer that unfortunately.
let's hope that Crystal HD can do deinterlacing and scaling
Yeah, and hopefully it can do "full-fps" deinterlacing, aka 50hz interlaced stream to 50 fps progressive.
-- Pasi
in crystalhd/include/7411d.h there's lines
/* scaling on/off */ /* - eCMD_C011_DEC_CHAN_OUTPUT_FORMAT */ typedef enum { eC011_SCALING_OFF = 0x00000000, eC011_SCALING_ON = 0x00000001,
} eC011_SCALING;
/* deinterlacing on/off */ /* - eCMD_C011_DEC_CHAN_OUTPUT_FORMAT */ typedef enum { eC011_DEINTERLACING_OFF = 0x00000000, eC011_DEINTERLACING_ON = 0x00000001,
} eC011_DEINTERLACING;
/* deinterlacing on/off */ /* - eCMD_C011_DEC_CHAN_OUTPUT_FORMAT */ typedef enum { eC011_DEINTERLACING_OFF = 0x00000000, eC011_DEINTERLACING_ON = 0x00000001,
} eC011_DEINTERLACING;
vdr mailing list vdr@linuxtv.org http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr
Problem is, that is just a chip. Any boards out using it? Linux drivers?
Torgeir Veimo wrote:
2010/1/2 Paul Menzel paulepanter@users.sourceforge.net:
I just share something I read on the XBMC [1] Web site.
The Broadcom Crystal HD Hardware Decoder (BCM970012) seems to be an alternative for HD playback with free drivers [2]. You can get it at as an mini-PCIE card at eBay for 22 $ [3].
I'd recommend the BCM70015, which is the better part.
It works as a decoder only, making YUV 4:2:2 frames available, which must then be blitted to whatever output framebuffer setup you have, over the pci express bus. The API seems very straightforward, much simpler that VDPAU, thus one could get a HW accelerated H.264 setup without X11.
Приветствую, Timothy
Problem is, that is just a chip. Any boards out using it? Linux drivers?
no. it's pci-e mini board with chip and Linux open source driver http://www.broadcom.com/docs/support/crystalhd/crystalhd_linux_20091229.zip
Torgeir Veimo wrote:
2010/1/2 Paul Menzel paulepanter@users.sourceforge.net:
I just share something I read on the XBMC [1] Web site.
The Broadcom Crystal HD Hardware Decoder (BCM970012) seems to be an alternative for HD playback with free drivers [2]. You can get it at as an mini-PCIE card at eBay for 22 $ [3].
I'd recommend the BCM70015, which is the better part.
It works as a decoder only, making YUV 4:2:2 frames available, which must then be blitted to whatever output framebuffer setup you have, over the pci express bus. The API seems very straightforward, much simpler that VDPAU, thus one could get a HW accelerated H.264 setup without X11.
I as refering to the BCM70015. No plugin boards for it yet. the board mentioned is for the 0012 version.
I wonder if this stuff will work any better then vdpau/xine. I have yet to get that mess working relablly. using an Asus 8400 silent.
On 1/2/2010 11:28 AM, Goga777 wrote:
Приветствую, Timothy
Problem is, that is just a chip. Any boards out using it? Linux drivers?
no. it's pci-e mini board with chip and Linux open source driver http://www.broadcom.com/docs/support/crystalhd/crystalhd_linux_20091229.zip
Torgeir Veimo wrote:
2010/1/2 Paul Menzelpaulepanter@users.sourceforge.net:
I just share something I read on the XBMC [1] Web site.
The Broadcom Crystal HD Hardware Decoder (BCM970012) seems to be an alternative for HD playback with free drivers [2]. You can get it at as an mini-PCIE card at eBay for 22 $ [3].
I'd recommend the BCM70015, which is the better part.
It works as a decoder only, making YUV 4:2:2 frames available, which must then be blitted to whatever output framebuffer setup you have, over the pci express bus. The API seems very straightforward, much simpler that VDPAU, thus one could get a HW accelerated H.264 setup without X11.
vdr mailing list vdr@linuxtv.org http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr
On Sat, Jan 2, 2010 at 10:06 PM, Timothy D. Lenz tlenz@vorgon.com wrote:
Problem is, that is just a chip. Any boards out using it? Linux drivers?
It is not a single chip. It is a combination of 2 chips. The card will be available soon from Azurewave
Does somebody use it to watch HD channels with VDR? As I understand that solution can give the same result as nVidia Ion?
On Sat, Jan 2, 2010 at 8:29 PM, Manu Abraham abraham.manu@gmail.com wrote:
On Sat, Jan 2, 2010 at 10:06 PM, Timothy D. Lenz tlenz@vorgon.com wrote:
Problem is, that is just a chip. Any boards out using it? Linux drivers?
It is not a single chip. It is a combination of 2 chips. The card will be available soon from Azurewave
vdr mailing list vdr@linuxtv.org http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr
Michael Stepanov wrote:
Does somebody use it to watch HD channels with VDR? As I understand that solution can give the same result as nVidia Ion?
yes, ebsi got it working http://www.vdrportal.de/board/thread.php?threadid=91157
"Status update :
Setup :
archlinux, kernel 2.6.28, vdr 1.7.10, vdr-xine 0.9.3, xine-lib-1.2-crystalhd-r634
Hardware :
DVB-S2 TT3200, Motherboard D945GSEJT
Channel :
SKY Cinema HD ORF 1/2 HD
Status :
Working with minor glitches"
Hehe, yes thats me ;)
Lars Bläser schrieb:
Michael Stepanov wrote:
Does somebody use it to watch HD channels with VDR? As I understand that solution can give the same result as nVidia Ion?
yes, ebsi got it working http://www.vdrportal.de/board/thread.php?threadid=91157
"Status update :
Setup :
archlinux, kernel 2.6.28, vdr 1.7.10, vdr-xine 0.9.3, xine-lib-1.2-crystalhd-r634
Hardware :
DVB-S2 TT3200, Motherboard D945GSEJT
Channel :
SKY Cinema HD ORF 1/2 HD
Status :
Working with minor glitches"
vdr mailing list vdr@linuxtv.org http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr
Hello,
i use it with vdr 1.7.10, xine-lib-1.2, vdr-xine, and the xine decoder plugin i wrote at :
http://sourceforge.net/apps/trac/archvdr/browser/branches/libcrystalhd
cu
Edgar (gimli) Hucek
Michael Stepanov schrieb:
Does somebody use it to watch HD channels with VDR? As I understand that solution can give the same result as nVidia Ion?
On Sat, Jan 2, 2010 at 8:29 PM, Manu Abraham abraham.manu@gmail.com wrote:
On Sat, Jan 2, 2010 at 10:06 PM, Timothy D. Lenz tlenz@vorgon.com wrote:
Problem is, that is just a chip. Any boards out using it? Linux drivers?
It is not a single chip. It is a combination of 2 chips. The card will be available soon from Azurewave
vdr mailing list vdr@linuxtv.org http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr
vdr mailing list vdr@linuxtv.org http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr
I wish we could get away from xine. Too many layers. VDR/VDPAU/XINE/XORG. End up with lots of problems, change channels oo fast, somthing crashes, weak signal, something crashes, FF/RW speed changing sometimes crashes, with all the buffering, get slow channel change, slow responce to pause, RW/FF and lots of lip sync problems. Need to reduce some middle ware.
On 1/3/2010 1:08 PM, gimli wrote:
Hello,
i use it with vdr 1.7.10, xine-lib-1.2, vdr-xine, and the xine decoder plugin i wrote at :
http://sourceforge.net/apps/trac/archvdr/browser/branches/libcrystalhd
cu
Edgar (gimli) Hucek
Michael Stepanov schrieb:
Does somebody use it to watch HD channels with VDR? As I understand that solution can give the same result as nVidia Ion?
On Sat, Jan 2, 2010 at 8:29 PM, Manu Abraham abraham.manu@gmail.com wrote:
On Sat, Jan 2, 2010 at 10:06 PM, Timothy D. Lenz tlenz@vorgon.com wrote:
Problem is, that is just a chip. Any boards out using it? Linux drivers?
It is not a single chip. It is a combination of 2 chips. The card will be available soon from Azurewave
vdr mailing list vdr@linuxtv.org http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr
vdr mailing list vdr@linuxtv.org http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr
vdr mailing list vdr@linuxtv.org http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr
On Mon, Jan 4, 2010 at 9:54 AM, Timothy D. Lenz tlenz@vorgon.com wrote:
I wish we could get away from xine. Too many layers. VDR/VDPAU/XINE/XORG. End up with lots of problems, change channels oo fast, somthing crashes, weak signal, something crashes, FF/RW speed changing sometimes crashes, with all the buffering, get slow channel change, slow responce to pause, RW/FF and lots of lip sync problems. Need to reduce some middle ware.
Everyone seems to have their own experience. Not sure why you have so many problems but it doesn't seem very common.
I wish we could get away from xine.
how ? does it possible at all ?
VDR/VDPAU/XINE/XORG. End up with lots of problems, change channels oo fast, somthing crashes, weak signal, something crashes, FF/RW speed changing sometimes crashes, with all the buffering, get slow channel change, slow responce to pause, RW/FF and lots of lip sync problems. Need to reduce some middle ware.
On 1/3/2010 1:08 PM, gimli wrote:
Hello,
i use it with vdr 1.7.10, xine-lib-1.2, vdr-xine, and the xine decoder plugin i wrote at :
http://sourceforge.net/apps/trac/archvdr/browser/branches/libcrystalhd
cu
Edgar (gimli) Hucek
Michael Stepanov schrieb:
Does somebody use it to watch HD channels with VDR? As I understand that solution can give the same result as nVidia Ion?
On Sat, Jan 2, 2010 at 8:29 PM, Manu Abraham abraham.manu@gmail.com wrote:
On Sat, Jan 2, 2010 at 10:06 PM, Timothy D. Lenz tlenz@vorgon.com wrote:
Problem is, that is just a chip. Any boards out using it? Linux drivers?
It is not a single chip. It is a combination of 2 chips. The card will be available soon from Azurewave
Le mardi 05 janvier 2010 à 00:14 +0300, Goga777 a écrit :
I wish we could get away from xine.
how ? does it possible at all ?
softdevice uses ffmpeg with DirectFB, and works wonderfully ("worked", in my case). Lack of X11 brings a lot of little improvements. In a pure STB setup, X11 is really useless, so why not just get rid of it. Xine have also lots of useless features for a STB. The only problem is that display drivers are well maintained in Xorg, but framebuffer drivers used by DirectFB may not be maintained as much.
I switched from softdevice to xineliboutput because I was tired of patching/recompiling the kernel/DirectFB/softdevice/etc with framebuffer fixes. The net result with xineliboutput is that I can use binary packages (e-tobi), which brings visual artifacts and crashes.
Maybe current Intel framebuffer/DirectFB drivers are OK with that Broadcom chip (re. deinterlacing, scaling, etc.), reviving the general interest with softdevice ?
Hi
according of discussion in Official linux kernel mailing list http://www.gossamer-threads.com/lists/linux/kernel/1173758#
in 2.6.34 kernel we will have Crystal HD driver
Goga
On Tue, 05 Jan 2010 10:12:30 +0100 Nicolas Huillard nicolas@huillard.net wrote:
Le mardi 05 janvier 2010 à 00:14 +0300, Goga777 a écrit :
I wish we could get away from xine.
how ? does it possible at all ?
softdevice uses ffmpeg with DirectFB, and works wonderfully ("worked", in my case). Lack of X11 brings a lot of little improvements. In a pure STB setup, X11 is really useless, so why not just get rid of it. Xine have also lots of useless features for a STB. The only problem is that display drivers are well maintained in Xorg, but framebuffer drivers used by DirectFB may not be maintained as much.
I had problems with softdevice, including A/V sync, but I think that was one of the areas the developers were putting a lof of effort into improving. Another thing I like about the xine plugins is that you can leave VDR running in the background for recording and stop and start the player frontend at will. For me xineliboutput and vdr-fbfe worked very well on a Matrox G450 connected to a SD TV, except that it would drop quite a lot of frames to achieve A/V sync.
I agree with Timothy D. Lenz. VDR desperately needs a decent player that works on mainstream graphics cards and allows full use of the OSD along with miscellaneous other features tailored for TV viewing. vdr-sxfe is quite good, but not good enough.
On Tuesday 05 Jan 2010, Tony Houghton wrote:
I had problems with softdevice, including A/V sync, but I think that was one of the areas the developers were putting a lof of effort into improving. Another thing I like about the xine plugins is that you can leave VDR running in the background for recording and stop and start the player frontend at will. For me xineliboutput and vdr-fbfe worked very well on a Matrox G450 connected to a SD TV, except that it would drop quite a lot of frames to achieve A/V sync.
I'm currently using a Matrox G450 with softdevice and it works pretty much flawlessly for me at the moment with a bog-standard CRT tele'.
I agree with Timothy D. Lenz. VDR desperately needs a decent player that works on mainstream graphics cards and allows full use of the OSD along with miscellaneous other features tailored for TV viewing. vdr-sxfe is quite good, but not good enough.
I'm watching this sort of thread with interest...no free HD content in the UK yet (at least over DVB-T) but it's on the way in the next few years. No immediate need for getting an HD tele' or beefing up my vdr box for HD playback but putting in a small, purpose-built decoder card definitely seems to be the way forward.
It would be nice to see support for the Broadcom HD decoders in FFMPEG so they can be used by softdevice but I'm not holding my breath! Once I have HD transmissions at my disposal, I'll definitely be buying such a decoder card or whatever they've been superseded by! Maybe someone will also have made some DVB-T2 devices by then!
Cheers,
Laz
On Tue, 5 Jan 2010 15:01:41 +0000 Laz laz@club-burniston.co.uk wrote:
I'm watching this sort of thread with interest...no free HD content in the UK yet (at least over DVB-T) but it's on the way in the next few years. No immediate need for getting an HD tele' or beefing up my vdr box for HD playback but putting in a small, purpose-built decoder card definitely seems to be the way forward.
It would be nice to see support for the Broadcom HD decoders in FFMPEG so they can be used by softdevice but I'm not holding my breath! Once I have HD transmissions at my disposal, I'll definitely be buying such a decoder card or whatever they've been superseded by! Maybe someone will also have made some DVB-T2 devices by then!
The trouble with a purpose-built decoder is that it takes up a valuable PCI(-E) slot when there are plenty of motherboards with onboard graphics which should be able to do hardware decoding, even if it does currently limit the choice to NVidia. In the short term (providing software as well as drivers supports these cards) a separate decoder has the advantage that you could pair it with ATI graphics and benefit from FRC too.
I think that the future may ultimately lie with OpenCL.
On Tue, 2010-01-05 at 17:29 +0000, Tony Houghton wrote:
On Tue, 5 Jan 2010 15:01:41 +0000 Laz laz@club-burniston.co.uk wrote:
The trouble with a purpose-built decoder is that it takes up a valuable PCI(-E) slot when there are plenty of motherboards with onboard graphics which should be able to do hardware decoding, even if it does currently limit the choice to NVidia. In the short term (providing software as well as drivers supports these cards) a separate decoder has the advantage that you could pair it with ATI graphics and benefit from FRC too.
I think that the future may ultimately lie with OpenCL.
The Crystal HD decoder will doubtless appear in netbooks very soon, and I fully expect thin Intel Atom PCs to also feature it on-board. When those boards/machines appear, we will be on the way to a *reliable* open source STB which supports modern HD codecs + playback.
As the driver is being taken into the linux mainline kernel, it'll be one less piece of the puzzle to have to checkout nightly SVNs for or rely on a binary blob from NVidia for. That additional choice is surely of great benefit to us all.
Maybe then I'll be able to replace my aging FF technotrend card + full-height case :) ... and best of all I'll then be able to justify spending a small fortune on an LCD or Plasma :D
gdh
On Tue, 05 Jan 2010 18:32:03 +0000 Gavin Hamill gdh@acentral.co.uk wrote:
The Crystal HD decoder will doubtless appear in netbooks very soon, and I fully expect thin Intel Atom PCs to also feature it on-board. When those boards/machines appear, we will be on the way to a *reliable* open source STB which supports modern HD codecs + playback.
Is it intended for the netbook market then? I didn't realise that. Hopefully Intel will realise that VGA is obsolete (and that they need to stop using inefficient chipsets with cheap & noisy coolers) otherwise Ion will still be the only viable option for an Atom HTPC, making Crystal HD theoretically redundant.
As far as I understand Intel presented their new chips on CES show called Arrandale and Clarkdale. Both has integrated graphics accelerator and other things. Those will be based on i7, i5 and i3 cores. In addition to Atom based "pineview" chips that also include graphics accelerator.
So now we have new options in case Atom will not provide enough juice.
Lets hope they will also release Linux drivers for those chips.
On Tue, Jan 5, 2010 at 9:53 PM, Tony Houghton h@realh.co.uk wrote:
On Tue, 05 Jan 2010 18:32:03 +0000 Gavin Hamill gdh@acentral.co.uk wrote:
The Crystal HD decoder will doubtless appear in netbooks very soon, and I fully expect thin Intel Atom PCs to also feature it on-board. When those boards/machines appear, we will be on the way to a *reliable* open source STB which supports modern HD codecs + playback.
Is it intended for the netbook market then? I didn't realise that. Hopefully Intel will realise that VGA is obsolete (and that they need to stop using inefficient chipsets with cheap & noisy coolers) otherwise Ion will still be the only viable option for an Atom HTPC, making Crystal HD theoretically redundant.
-- TH * http://www.realh.co.uk
vdr mailing list vdr@linuxtv.org http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr
On Wed, 6 Jan 2010 20:28:14 +0200 Alex Betis alex.betis@gmail.com wrote:
As far as I understand Intel presented their new chips on CES show called Arrandale and Clarkdale. Both has integrated graphics accelerator and other things. Those will be based on i7, i5 and i3 cores. In addition to Atom based "pineview" chips that also include graphics accelerator.
So now we have new options in case Atom will not provide enough juice.
Lets hope they will also release Linux drivers for those chips.
And that they're better than the Poulsbo drivers.
Have anyone else seen this problem?
When I watch an ongoing recording vdr does not see it grow.
Say I record a 60min episode and start to watch it 10min in. Pressing 'Ok' will show me that the recording is 10min in size. After 5min it still shows 10min size. If I stop watching and immidiately resume watching it, vdr will show 15min in size.
If I let it continue till 15min it will stop my watching the recording as if it was ended. I can the of course resume watching again, as the size will always be the size when I start watching the recording.
/Johan
On Tue, Jan 5, 2010 at 1:10 PM, Johan Andersson jna@jna.pp.se wrote:
Have anyone else seen this problem?
When I watch an ongoing recording vdr does not see it grow.
Say I record a 60min episode and start to watch it 10min in. Pressing 'Ok' will show me that the recording is 10min in size. After 5min it still shows 10min size. If I stop watching and immidiately resume watching it, vdr will show 15min in size.
If I let it continue till 15min it will stop my watching the recording as if it was ended. I can the of course resume watching again, as the size will always be the size when I start watching the recording.
This sounds like a similar problem I've already reported, in which case Klaus is aware of it and has been working to fix it. Maybe he'll see this thread and elaborate further. It's nice to hear someone else reporting this bug as well so we know it's affecting other users.
On 05.01.2010 22:32, VDR User wrote:
On Tue, Jan 5, 2010 at 1:10 PM, Johan Andersson jna@jna.pp.se wrote:
Have anyone else seen this problem?
When I watch an ongoing recording vdr does not see it grow.
Say I record a 60min episode and start to watch it 10min in. Pressing 'Ok' will show me that the recording is 10min in size. After 5min it still shows 10min size. If I stop watching and immidiately resume watching it, vdr will show 15min in size.
If I let it continue till 15min it will stop my watching the recording as if it was ended. I can the of course resume watching again, as the size will always be the size when I start watching the recording.
This sounds like a similar problem I've already reported, in which case Klaus is aware of it and has been working to fix it. Maybe he'll see this thread and elaborate further. It's nice to hear someone else reporting this bug as well so we know it's affecting other users.
Since I can't reproduce this behavior here, I'm afraid I can't help you. Somebody who can reproduce this will have to debug it.
Klaus
Not a problem for me. I was just interested to hear if this was a known issue with a simple fix. I'll do some debugging on my own.
The most likely suspect is the client side NFS. I'll post my findings.
/Johan
Klaus Schmidinger skrev:
On 05.01.2010 22:32, VDR User wrote:
On Tue, Jan 5, 2010 at 1:10 PM, Johan Andersson jna@jna.pp.se wrote:
Since I can't reproduce this behavior here, I'm afraid I can't help you. Somebody who can reproduce this will have to debug it.
Klaus
vdr mailing list vdr@linuxtv.org http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr
.
The problem does not occur if storage is on local disk or if vdr is verion 1.6.0. So this is seemingly .ts files on an NFS server.
I tried NFS v2,v3 and noac mount option (no client attribute cache), the error still occurs. I will continue debugging...
/Johan
Johan Andersson skrev:
Not a problem for me. I was just interested to hear if this was a known issue with a simple fix. I'll do some debugging on my own.
The most likely suspect is the client side NFS. I'll post my findings.
/Johan
Klaus Schmidinger skrev:
On 05.01.2010 22:32, VDR User wrote:
On Tue, Jan 5, 2010 at 1:10 PM, Johan Andersson jna@jna.pp.se wrote:
Since I can't reproduce this behavior here, I'm afraid I can't help you. Somebody who can reproduce this will have to debug it.
On 09.01.2010 13:36, Johan Andersson wrote:
The problem does not occur if storage is on local disk or if vdr is verion 1.6.0. So this is seemingly .ts files on an NFS server.
I tried NFS v2,v3 and noac mount option (no client attribute cache), the error still occurs. I will continue debugging...
You might want to try commenting out the line
#define USE_FADVISE
in tools.c.
Just an idea...
Klaus
Johan Andersson skrev:
Not a problem for me. I was just interested to hear if this was a known issue with a simple fix. I'll do some debugging on my own.
The most likely suspect is the client side NFS. I'll post my findings.
/Johan
Klaus Schmidinger skrev:
On 05.01.2010 22:32, VDR User wrote:
On Tue, Jan 5, 2010 at 1:10 PM, Johan Andersson jna@jna.pp.se wrote:
Since I can't reproduce this behavior here, I'm afraid I can't help you. Somebody who can reproduce this will have to debug it.
En/na Paul Menzel ha escrit:
Dear list,
I just share something I read on the XBMC [1] Web site.
The Broadcom Crystal HD Hardware Decoder (BCM970012) seems to be an alternative for HD playback with free drivers [2]. You can get it at as an mini-PCIE card at eBay for 22 $ [3].
If this is all true in my opinion these are fantastic news.
This is great news indeed. In the comments someone mentions this motherboard:
http://www.intel.com/products/desktop/motherboards/D510MO/D510MO-overview.ht...
It can be passively cooled, has a pci and a mini pci-e slot, but it lacks hdmi.
Bye
I just share something I read on the XBMC [1] Web site.
The Broadcom Crystal HD Hardware Decoder (BCM970012) seems to be an alternative for HD playback with free drivers [2]. You can get it at as an mini-PCIE card at eBay for 22 $ [3].
If this is all true in my opinion these are fantastic news.
This is great news indeed. In the comments someone mentions this motherboard:
http://www.intel.com/products/desktop/motherboards/D510MO/D510MO-overview.ht...
It can be passively cooled, has a pci and a mini pci-e slot, but it lacks hdmi.
have a look at Intel® Desktop Board D945GSEJT http://www.intel.com/products/desktop/motherboards/D945GSEJT/D945GSEJT-overv...
-
Alex Betis -
Gavin Hamill -
gimli -
Goga777 -
Johan Andersson -
Klaus Schmidinger -
Lars Bläser -
Laz -
Luca Olivetti -
Manu Abraham -
Michael Stepanov -
Nicolas Huillard -
Pasi Kärkkäinen -
Paul Menzel -
Timothy D. Lenz -
Tony Houghton -
Torgeir Veimo -
VDR User