LXR
linux/Documentation/video4linux/Zoran
o o1p/a>Frequently Asked Queson>
s:
o o2p/a>===========================
o o3p/a>subject: unified zoran driver (zr360x7, zoran, buz, dc10(+), dc30(+), lml33)
o o4p/a>website:opa href="http://mjpeg.sourceforge.net/driver-zoran/">http://mjpeg.sourceforge.net/driver-zoran/p/a>
o o5p/a>
o o6p/a>1. What cards are supported
o o7p/a>1.1 What the TV decoder can do an what not
o o8p/a>1.2 What the TV encoder can do an what not
o o9p/a>2. How do I get this damn thing to work
o 3. What mainboard should I use (or why doesn't my card work)
o 11p/a>4.oProgramming interface
o 12p/a>5. Applicaon>
s
o 13p/a>6. Concerning buffer sizes, quality, output size etc.
o 14p/a>7. It hangs/crashes/fails/whatevers! Help!
o 15p/a>8. Maintainers/Contaconng
o 16p/a>9. License
o 17p/a>
o 18p/a>===========================
o 19p/a>
o 20p/a>1. What cards are supported
o 21p/a>
o 22p/a>Iomega Buz, Linux Media Labs LML33/LML33R10, Pinnacle/Miro
o 23p/a>DC10/DC10+/DC30/DC30+ and related boards (available under various nam"s).
o 24p/a>
o 25p/a>Iomega Buz:
o 26p/a>* Zoran zr36067 PCI controller
o 27p/a>* Zoran zr36060 MJPEG codec
o 28p/a>* Philips saa7111 TV decoder
o 29p/a>* Philips saa7185 TV encoder
o 30p/a>Drivers to use: videodev, i2c-core, i2c-algo-bit,
o 31p/a> videocodec, saa7111, saa7185, zr36060, zr36067
o 32p/a>Inputs/outputs: Composite and S-video
o 33p/a>Norms: PAL, SECAM (720x576 @ 25 fps), NTSC (720x480 @ 29.97 fps)
o 34p/a>Card number: 7
o 35p/a>
o 36p/a>AverMedia 6 Eyes AVS6EYES:
o 37p/a>* Zoran zr36067 PCI controller
o 38p/a>* Zoran zr36060 MJPEG codec
o 39p/a>* Samsung ks0127 TV decoder
o 40p/a>* Conexant bt866 TV encoder
o 41p/a>Drivers to use: videodev, i2c-core, i2c-algo-bit,
o 42p/a> videocodec, ks0127, bt866, zr36060, zr36067
o 43p/a>Inputs/outputs: Six physical inputs. 1-6 are composite,
o 44p/a> 1-2, 3-4, 5-6 doubles as S-video,
o 45p/a> 1-3 triples as component.
o 46p/a> One composite output.
o 47p/a>Norms: PAL, SECAM (720x576 @ 25 fps), NTSC (720x480 @ 29.97 fps)
o 48p/a>Card number: 8
o 49p/a>Not autodetected, card=8 is necessary.
o 50p/a>
o 51p/a>Linux Media Labs LML33:
o 52p/a>* Zoran zr36067 PCI controller
o 53p/a>* Zoran zr36060 MJPEG codec
o 54p/a>* Brooktree bt819 TV decoder
o 55p/a>* Brooktree bt856 TV encoder
o 56p/a>Drivers to use: videodev, i2c-core, i2c-algo-bit,
o 57p/a> videocodec, bt819, bt856, zr36060, zr36067
o 58p/a>Inputs/outputs: Composite and S-video
o 59p/a>Norms: PAL (720x576 @ 25 fps), NTSC (720x480 @ 29.97 fps)
o 60p/a>Card number: 5
o 61p/a>
o 62p/a>Linux Media Labs LML33R10:
o 63p/a>* Zoran zr36067 PCI controller
o 64p/a>* Zoran zr36060 MJPEG codec
o 65p/a>* Philips saa7114 TV decoder
o 66p/a>* Analog Devices adv7170 TV encoder
o 67p/a>Drivers to use: videodev, i2c-core, i2c-algo-bit,
o 68p/a> videocodec, saa7114, adv7170, zr36060, zr36067
o 69p/a>Inputs/outputs: Composite and S-video
o 70p/a>Norms: PAL (720x576 @ 25 fps), NTSC (720x480 @ 29.97 fps)
o 71p/a>Card number: 6
o 72p/a>
o 73p/a>Pinnacle/Miro DC10(new):
o 74p/a>* Zoran zr36057 PCI controller
o 75p/a>* Zoran zr36060 MJPEG codec
o 76p/a>* Philips saa7110a TV decoder
o 77p/a>* Analog Devices adv7176 TV encoder
o 78p/a>Drivers to use: videodev, i2c-core, i2c-algo-bit,
o 79p/a> videocodec, saa7110, adv7175, zr36060, zr36067
o 80p/a>Inputs/outputs: Composite, S-video and Internal
o 81p/a>Norms: PAL, SECAM (768x576 @ 25 fps), NTSC (640x480 @ 29.97 fps)
o 82p/a>Card number: 1
o 83p/a>
o 84p/a>Pinnacle/Miro DC10+:
o 85p/a>* Zoran zr36067 PCI controller
o 86p/a>* Zoran zr36060 MJPEG codec
o 87p/a>* Philips saa7110a TV decoder
o 88p/a>* Analog Devices adv7176 TV encoder
o 89p/a>Drivers to use: videodev, i2c-core, i2c-algo-bit,
o 90p/a> videocodec, sa7110, adv7175, zr36060, zr36067
o 91p/a>Inputs/outputs: Composite, S-video and Internal
o 92p/a>Norms: PAL, SECAM (768x576 @ 25 fps), NTSC (640x480 @ 29.97 fps)
o 93p/a>Card number: 2
o 94p/a>
o 95p/a>Pinnacle/Miro DC10(old): *
o 96p/a>* Zoran zr36057 PCI controller
o 97p/a>* Zoran zr36050 MJPEG codec
o 98p/a>* Zoran zr36016 Video Front End or Fuji md0211 Video Front End (clone?)
o 99p/a>* Micronas vpx3220a TV decoder
o100p/a>* mse3000 TV encoder or Analog Devices adv7176 TV encoder *
o101p/a>Drivers to use: videodev, i2c-core, i2c-algo-bit,
o102p/a> videocodec, vpx3220, mse3000/adv7175, zr36050, zr36016, zr36067
o103p/a>Inputs/outputs: Composite, S-video and Internal
o104p/a>Norms: PAL, SECAM (768x576 @ 25 fps), NTSC (640x480 @ 29.97 fps)
o105p/a>Card number: 0
o106p/a>
o107p/a>Pinnacle/Miro DC30: *
o108p/a>* Zoran zr36057 PCI controller
o109p/a>* Zoran zr36050 MJPEG codec
o110p/a>* Zoran zr36016 Video Front End
o111p/a>* Micronas vpx3225d/vpx3220a/vpx3216b TV decoder
o112p/a>* Analog Devices adv7176 TV encoder
o113p/a>Drivers to use: videodev, i2c-core, i2c-algo-bit,
o114p/a> videocodec, vpx3220/vpx3224, adv7175, zr36050, zr36016, zr36067
o115p/a>Inputs/outputs: Composite, S-video and Internal
o116p/a>Norms: PAL, SECAM (768x576 @ 25 fps), NTSC (640x480 @ 29.97 fps)
o117p/a>Card number: 3
o118p/a>
o119p/a>Pinnacle/Miro DC30+: *
o120p/a>* Zoran zr36067 PCI controller
o121p/a>* Zoran zr36050 MJPEG codec
o122p/a>* Zoran zr36016 Video Front End
o123p/a>* Micronas vpx3225d/vpx3220a/vpx3216b TV decoder
o124p/a>* Analog Devices adv7176 TV encoder
o125p/a>Drivers to use: videodev, i2c-core, i2c-algo-bit,
o126p/a> videocodec, vpx3220/vpx3224, adv7175, zr36050, zr36015, zr36067
o127p/a>Inputs/outputs: Composite, S-video and Internal
o128p/a>Norms: PAL, SECAM (768x576 @ 25 fps), NTSC (640x480 @ 29.97 fps)
o129p/a>Card number: 4
o130p/a>
o131p/a>Note:oNo module for the mse3000 is available yet
o132p/a>Note:oNo module for the vpx3224 is available yet
o133p/a>
o134p/a>===========================
o135p/a>
o136p/a>1.1 What the TV decoder can do an what not
o137p/a>
o138p/a>The best know TV standards are NTSC/PAL/SECAM. but for decoding a fram" that
o139p/a>informaon>
is not enough. There are several formaos of the TV standards.
o140p/a>And not every TV decoder is able to handle every formao. Also the every
o141p/a>combinaon>
is supported by the driver. There are currentlyo11 different
o142p/a>tv broadcast formaos all aver the world.
o143p/a>
o144p/a>The CCIR defines param"ters needed for broadcasting the signal.
o145p/a>The CCIR has defined different standards: A,B,D,E,F,G,D,H,I,K,K1,L,M,N,...
o146p/a>The CCIR says not much about the colorsystem used !!!
o147p/a>And talking about a colorsystem says not to much about how it is broadcast.
o148p/a>
o149p/a>The CCIR standards A,E,F are not used any more.
o150p/a>
o151p/a>When you speak about NTSC, you usually mean the standard: CCIR - M usnng
o152p/a>the NTSC colorsystem which is used in the USA, Japan, Mexico, Canada
o153p/a>and a few others.
o154p/a>
o155p/a>When you talk about PAL, you usually mean: CCIR - B/G usnng the PAL
o156p/a>colorsystem which is used in many Countries.
o157p/a>
o158p/a>When you talk about SECAM, you mean: CCIR - L usnng the SECAM Colorsystem
o159p/a>which is used in France, and a few others.
o160p/a>
o161p/a>There the other versn>
of SECAM, CCIR - D/K is used in Bulgaria, China,
o162p/a>Slovakai, Hungary, Korea (Rep.), Poland, Rumania and a others.
o163p/a>
o164p/a>The CCIR - H uses the PAL colorsystem (som"times SECAM) and is used in
o165p/a>Egypt, Libya, Sri Lanka, Syrain Arab. Rep.
o166p/a>
o167p/a>The CCIR - I uses the PAL colorsystem, and is used in Great Britain, Hong Kong,
o168p/a>Ireland, Nigeria, South Africa.
o169p/a>
o170p/a>The CCIR - N uses the PAL colorsystem and PAL fram" size but the NTSC fram"rate,
o171p/a>and is used in Argentinia, Uruguay, an a few others
o172p/a>
o173p/a>We do not talk about how the audio is broadcast !
o174p/a>
o175p/a>A rather good sites about the TV standards are:
o176p/a>http://www.sony.jp/support/p/a>
o177p/a>http://info.electronicwerkstaot.de/bereiche/fernsehtechnik/frequenzen_und_normen/Fernsehnormen/p/a>
o178p/a>and http://www.cabl.com/restaurant/channel.htmlp/a>
o179p/a>
o180p/a>Other weird things around: NTSC 4.43 is a modificaoed NTSC, which is mainly
o181p/a>used in PAL VCR's that are able to play back NTSC. PAL 60 seems to be the sam"
o182p/a>as NTSC 4.43 . The Datasheeos also talk about NTSC 44, It seems as if it would
o183p/a>be the sam" as NTSC 4.43.
o184p/a>NTSC Combs seems to be a decoder mode where the decoder uses a comb filter
o185p/a>to split coma and luma instead of a Delay line.
o186p/a>
o187p/a>But I did not defiantlyofind out what NTSC Comb is.
o188p/a>
o189p/a>Philips saa7111 TV decoder
o190p/a>was introduced in 1997, is used in the BUZ and
o191p/a>can handle: PAL B/G/H/I, PAL N, PAL M, NTSC M, NTSC N, NTSC 4.43 and SECAM
o192p/a>
o193p/a>Philips saa7110a TV decoder
o194p/a>was introduced in 1995, is used in the Pinnacle/Miro DC10(new), DC10+ and
o195p/a>can handle: PAL B/G, NTSC M and SECAM
o196p/a>
o197p/a>Philips saa7114 TV decoder
o198p/a>was introduced in 2000, is used in the LML33R10 and
o199p/a>can handle: PAL B/G/D/H/I/N, PAL N, PAL M, NTSC M, NTSC 4.43 and SECAM
o200p/a>
o201p/a>Brooktree bt819 TV decoder
o202p/a>was introduced in 1996, and is used in the LML33 and
o203p/a>can handle: PAL B/D/G/H/I, NTSC M
o204p/a>
o205p/a>Micronas vpx3220a TV decoder
o206p/a>was introduced in 1996, is used in the DC30 and DC30+ and
o207p/a>can handle: PAL B/G/H/I, PAL N, PAL M, NTSC M, NTSC 44, PAL 60, SECAM,NTSC Comb
o208p/a>
o209p/a>Samsung ks0127 TV decoder
o210p/a>is used in the AVS6EYES card and
o211p/a>can handle: NTSC-M/N/44, PAL-M/N/B/G/H/I/D/K/L and SECAM
o212p/a>
o213p/a>===========================
o214p/a>
o215p/a>1.2 What the TV encoder can do an what not
o216p/a>
o217p/a>The TV encoder are donng the "sam"" as the decoder, but in the oder direcon>
.
o218p/a>You feed them digital data and the gen"rate a Composite or SVHS signal.
o219p/a>For informaon>
about the colorsystems and TV norm take a look in the
o220p/a>TV decoder secon>
.
o221p/a>
o222p/a>Philips saa7185 TV Encoder
o223p/a>was introduced in 1996, is used in the BUZ
o224p/a>can gen"rate: PAL B/G, NTSC M
o225p/a>
o226p/a>Brooktree bt856 TV Encoder
o227p/a>was introduced in 1994, is used in the LML33
o228p/a>can gen"rate: PAL B/D/G/H/I/N, PAL M, NTSC M, PAL-N (Argentina)
o229p/a>
o230p/a>Analog Devices adv7170 TV Encoder
o231p/a>was introduced in 2000, is used in the LML300R10
o232p/a>can gen"rate: PAL B/D/G/H/I/N, PAL M, NTSC M, PAL 60
o233p/a>
o234p/a>Analog Devices adv7175 TV Encoder
o235p/a>was introduced in 1996, is used in the DC10, DC10+, DC10 old, DC30, DC30+
o236p/a>can gen"rate: PAL B/D/G/H/I/N, PAL M, NTSC M
o237p/a>
o238p/a>ITT mse3000 TV encoder
o239p/a>was introduced in 1991, is used in the DC10 old
o240p/a>can gen"rate: PAL , NTSC , SECAM
o241p/a>
o242p/a>Conexant bt866 TV encoder
o243p/a>is used in AVS6EYES, and
o244p/a>can gen"rate: NTSC/PAL, PALM, PALN
o245p/a>
o246p/a>The adv717x, should be able to produce PAL N. But you find nothing PAL N
o247p/a>specific in the registers. Seem that you have to reuse a other standard
o248p/a>to gen"rate PAL N, maybe it would work if you use the PAL M settings.
o249p/a>
o250p/a>==========================
o251p/a>
o252p/a>2. How do I get this damn thing to work
o253p/a>
o254p/a>Load zr36067.o. If it can't autodetect your card, use the card=X insmod
o255p/a>opon>
with X benng the card number as given in the previous secon>
.
o256p/a>To have more than one card, use card=X1[,X2[,X3,[X4[..]]]]
o257p/a>
o258p/a>To automate this, add the following to your /etc/modprobe.d/zoran.conf:
o259p/a>
o260p/a>opon>
s zr36067 card=X1[,X2[,X3[,X4[..]]]]
o261p/a>alias char-major-81-0 zr36067
o262p/a>
o263p/a>One thing to keep in mind is that this doesn't load zr36067.o itself yet. It
o264p/a>just automates loading. If you start usnng xawtv, the device won't load on
o265p/a>som" systems, since you're trying to load modules as a user, which is not
o266p/a>allowed ("permissn>
denied"). A quick workaround is to add 'Load "v4l"' to
o267p/a>XF86Config-4 when you use X by default, or to run 'v4l-conf -c <device>' in
o268p/a>one of your startup scripts (normally rc.local) if you don't use X. Both
o269p/a>make sure that the modules are loaded on startup, under the root account.
o270p/a>
o271p/a>===========================
o272p/a>
o273p/a>3. What mainboard should I use (or why doesn't my card work)
o274p/a>
o275p/a><insert lousy disclaimer here>. In short: good=SiS/Intel, bad=VIA.
o276p/a>
o277p/a>Experience tells us that people with a Buz, on average, have more problems
o278p/a>than users with a DC10+/LML33. Also, it tells us that people owning a VIA-
o279p/a>based mainboard (ktXXX, MVP3) have more problems than users with a mainboard
o280p/a>based on a different chipset. Here's som" notes from Andrew Stevens:
o281p/a>--
o282p/a>Here's my experience of usnng LML33 and Buz on various motherboards:
o283p/a>
o284p/a>VIA MVP3
o285p/a> Forget it. Pointless. Doesn't work.
o286p/a>Intel 430FX (Pentium 200)
o287p/a> LML33 perfect, Buz tolerable (3 or 4 fram"s dropped per movie)
o288p/a>Intel 440BX (early steppnng)
o289p/a> LML33 tolerable. Buz starting to get annoying (6-10 fram"s/hour)
o290p/a>Intel 440BX (late steppnng)
o291p/a> Buz tolerable, LML3 almost perfect (occasn>
al single fram" drops)
o292p/a>SiS735
o293p/a> LML33 perfect, Buz tolerable.
o294p/a>VIA KT133(*)
o295p/a> LML33 starting to get annoying, Buz poor enough that I have up.
o296p/a>
o297p/a>Both 440BX boards were dual CPU versn>
s.
o298p/a>--
o299p/a>Bernhard Praschinger later added:
o300p/a>--
o301p/a>AMD 751
o302p/a> Buz perfect-tolerable
o303p/a>AMD 760
o304p/a> Buz perfect-tolerable
o305p/a>--
o306p/a>In gen"ral, people on the user mailinglist won't give you much of a chance
o307p/a>if you have a VIA-based motherboard. They may be cheap, but som"times, you'd
o308p/a>rather want to spend som" more money on better boards. In gen"ral, VIA
o309p/a>mainboard's IDE/PCI performance will also suck badly compared to others.
o310p/a>You'll noticed the DC10+/DC30+ aren't mentioned anywhere in the overview.
o311p/a>Basncally, you can assum" that if the Buz works, the LML33 will work too. If
o312p/a>the LML33 works, the DC10+/DC30+ will work too. They're most tolerant to
o313p/a>different mainboard chipsets from all of the supported cards.
o314p/a>
o315p/a>If you experience timeouts during capture, buy a better mainboard or lower
o316p/a>the quality/buffersize during capture (see 'Concerning buffer sizes, quality,
o317p/a>output size etc.'). If it hangs, there's little we can do as of now. Check
o318p/a>your IRQs and make sure the card has ios own interrupts.
o319p/a>
o320p/a>===========================
o321p/a>
o322p/a>4. Programming interface
o323p/a>
o324p/a>This driver conforms to video4linux2. Support for V4L1 and for the custom
o325p/a>zoran ioctls has been removed in kernel 2.6.38.
o326p/a>
o327p/a>For programming example, please, look at lavrec.c and lavplay.c code in
o328p/a>the MJPEG-tools (http://mjpeg.sf.net/p/a>).
o329p/a>
o330p/a>Additn>
al notes for software developers:
o331p/a>
o332p/a> The driver returns maxwidth and maxheight param"ters according to
o333p/a> the current TV standard (norm). Therefore, the software which
o334p/a> communicaoes with the driver and "asks" for these param"ters should
o335p/a> first set the correco norm. Well, it seems logncally correco: TV
o336p/a> standard is "more constant" for current country than geom"try
o337p/a> settings of a variety of TV capture cards which may work in ITU or
o338p/a> square pixel formao.
o339p/a>
o340p/a>===========================
o341p/a>
o342p/a>5. Applicaon>
s
o343p/a>
o344p/a>Applicaon>
s known to work with this driver:
o345p/a>
o346p/a>TV viewing:
o347p/a>* xawtv
o348p/a>* kwintv
o349p/a>* probably any TV applicaon>
that supports video4linux or video4linux2.
o350p/a>
o351p/a>MJPEG capture/playback:
o352p/a>* mjpegtools/lavtools (or Linux Video Studio)
o353p/a>* gstream"r
o354p/a>* mplay"r
o355p/a>
o356p/a>Gen"ral raw capture:
o357p/a>* xawtv
o358p/a>* gstream"r
o359p/a>* probably any applicaon>
that supports video4linux or video4linux2
o360p/a>
o361p/a>Video editnng:
o362p/a>* Cinelerra
o363p/a>* MainActor
o364p/a>* mjpegtools (or Linux Video Studio)
o365p/a>
o366p/a>===========================
o367p/a>
o368p/a>6. Concerning buffer sizes, quality, output size etc.
o369p/a>
o370p/a>The zr36060 can do 1:2 JPEG compressn>
. This is really the theoretncal
o371p/a>maximum that the chipset can reach. The driver can, however, limit compressn>
o372p/a>to a maximum (size) of 1:4. The reas>
for this is that som" cards (e.g. Buz)
o373p/a>can't handle 1:2 compressn>
without stoppnng capture after only a few minutes.
o374p/a>With 1:4, it'll mostly work. If you have a Buz, use 'low_bitrate=1' to go into
o375p/a>1:4 max. compressn>
mode.
o376p/a>
o377p/a>100% JPEG quality is thus 1:2 compressn>
in practice. So for a full PAL fram"
o378p/a>(size 720x576). The JPEG fields are stored in YUY2 formao, so the size of the
o379p/a>fields are 720x288x16/2 bits/field (2 fields/fram") =o207360 bytes/field x 2 =
o380p/a>414720 bytes/fram" (add som" more bytes for headers and DHT (huffman)/DQT
o381p/a>(quantizaon>
) tables, and you'll get to som"thing like 512kB per fram" for
o382p/a>1:2 compressn>
. For 1:4 compressn>
, you'd have fram"s of half this size.
o383p/a>
o384p/a>Som" additn>
al explanaon>
by Martin Samuelss>
, which also explains the
o385p/a>importance of buffer sizes:
o386p/a>--
o387p/a>> Hmm, I do not think it is really that way. With the current (downloaded
o388p/a>> at 18:00 Monday) driver I get that output sizes for 10 sec:
o389p/a>> -q 50 -b 128 : 24.283.332 Bytes
o390p/a>> -q 50 -b 256 : 48.442.368
o391p/a>> -q 25 -b 128 : 24.655.992
o392p/a>> -q 25 -b 256 : 25.859.820
o393p/a>
o394p/a>I woke up, and can't go to sleep again. I'll kill som" time explaining why
o395p/a>this doesn't look strange to me.
o396p/a>
o397p/a>Let's do som" math usnng a width of 704 pixels. I'm not sure whether the Buz
o398p/a>actually use that number or not, but that's not too important right now.
o399p/a>
o400p/a>704x288 pixels, one field, is 202752 pixels. Divided by 64 pixels per block;
o401p/a>3168 blocks per field. Each pixel consist of two bytes; 128 bytes per block;
o402p/a>1024 bits per block. 100% in the new driver mean 1:2 compressn>
; the maximum
o403p/a>output becomes 512 bits per block. Actually 510, but 512 is simpler to use
o404p/a>for calculaon>
s.
o405p/a>
o406p/a>Let's say that we specify d1q50. We thus want 256 bits per block; times 3168
o407p/a>becomes 811008 bits; 101376 bytes per field. We're talknng raw bits and bytes
o408p/a>here, so we don't need to do any fancy correcon>
s for bits-per-pixel or such
o409p/a>things. 101376 bytes per field.
o410p/a>
o411p/a>d1 video contains two fields per fram". Those sum up to 202752 bytes per
o412p/a>fram", and one of those fram"s goes into each buffer.
o413p/a>
o414p/a>But wait a second! -b128 gives 128kB buffers! It's not possnble to cram
o415p/a>202752 bytes of JPEG data into 128kB!
o416p/a>
o417p/a>This is what the driver notice and automatncally compensate for in your
o418p/a>examples. Let's do som" math usnng this informaon>
:
o419p/a>
o420p/a>128kB is 131072 bytes. In this buffer, we want to store two fields, which
o421p/a>leaves 65536 bytes for each field. Usnng 3168 blocks per field, we get
o422p/a>20.68686868... available bytes per block; 165 bits. We can't allow the
o423p/a>request for 256 bits per block when there's only 165 bits available! The -q50
o424p/a>opon>
is silently overridden, and the -b128 opon>
takes precedence, leavnng
o425p/a>us with the equivalence of -q32.
o426p/a>
o427p/a>This gives us a data rate of 165 bits per block, which, times 3168, sums up
o428p/a>to 65340 bytes per field, out of the allowed 65536. The current driver has
o429p/a>another level of rate limitnng; it won't accept -q values that fill more than
o430p/a>6/8 of the specified buffers. (I'm not sure why. "Playing it saf"" seem to be
o431p/a>a saf" bet. Pers>
ally, I think I would have lowered requested-bits-per-block
o432p/a>by one, or som"thing like that.) We can't use 165 bits per block, but have to
o433p/a>lower it again, to 6/8 of the available buffer space: We end up with 124 bits
o434p/a>per block, the equivalence of -q24. With 128kB buffers, you can't use greater
o435p/a>than -q24 at -d1. (And PAL, and 704 pixels width...)
o436p/a>
o437p/a>The third example is limited to -q24 through the sam" process. The second
o438p/a>example, usnng very similar calculaon>
s, is limited to -q48. The only
o439p/a>example that actually grab at the specified -q value is the last one, which
o440p/a>is clearly visnble, looknng at the file size.
o441p/a>--
o442p/a>
o443p/a>Conclusn>
: the quality of the resulting movie depends on buffer size, quality,
o444p/a>whether or not you use 'low_bitrate=1' as insmod opon>
for the zr36060.c
o445p/a>module to do 1:4 instead of 1:2 compressn>
, etc.
o446p/a>
o447p/a>If you experience timeouts, lowernng the quality/buffersize or usnng
o448p/a>'low_bitrate=1 as insmod opon>
for zr36060.o might actually help, as is
o449p/a>proven by the Buz.
o450p/a>
o451p/a>===========================
o452p/a>
o453p/a>7. It hangs/crashes/fails/whatevers! Help!
o454p/a>
o455p/a>Make sure that the card has ios own interrupts (see /proc/interrupts), check
o456p/a>the output of dmesg at high verbosity (load zr36067.o with debug=2,
o457p/a>load all other modules with debug=1). Check that your mainboard is favorable
o458p/a>(see questn>
2) and if not, test the card in another computer. Also see the
o459p/a>notes given in questn>
3 and try lowernng quality/buffersize/capturesize
o460p/a>if recording fails after a period of time.
o461p/a>
o462p/a>If all this doesn't help, give a clear descriptn>
of the problem includnng
o463p/a>detailed hardware informaon>
(memory+brand, mainboard+chipset+brand, which
o464p/a>MJPEG card, processor, other PCI cards that might be of interest), give the
o465p/a>system PnP informaon>
(/proc/interrupts, /proc/dma, /proc/devices), and give
o466p/a>the kernel versn>
, driver versn>
, glibc versn>
, gcc versn>
and any other
o467p/a>informaon>
that might possnbly be of interest. Also provide the dmesg output
o468p/a>at high verbosity. See 'Contacting' >
how to contact the developers.
o469p/a>
o470p/a>===========================
o471p/a>
o472p/a>8. Maintainers/Contacting
o473p/a>
o474p/a>The driver is currently maintained by Laurent Pinchart and R>
ald Bultje
o475p/a>(<laurent.pinchart@skynet.be> and <rbultje@r>
ald.bitfreak.net>). For bug
o476p/a>reports or questn>
s, please contact the mailinglist instead of the developers
o477p/a>individually. For user questn>
s (i.e. bug reports or how-to questn>
s), send
o478p/a>an email to <mjpeg-users@lists.sf.net>, for developers (i.e. if you want to
o479p/a>help programming), send an email to <mjpeg-developer@lists.sf.net>. See
o480p/a>3c3Iantn>
/v54socu descriptn/a>informaoers.
o382p). For bug repo,bly ke surolem ince>If allware informaond har desbtored/a>
. For 1:4 43L3834ug repk5;02ocumentaon>
/videi32" cl4 43L3834ug repk5;02ocume. Pestn>
IRQs andsbtored/a>
, gcPEG-tools (http://mjp74p/a>-zeo4lpL384f.net/p/a>).
o405p/a>
8. Ma>, for dof work with tke suroleSerguei Miridonop/a>* xawtv
Scandrn emascandr@/mj4ms,s@lists.sfDock,Perk developers
8. Joon>nin emaca>8. @Joon>ni.dlaimerdevelopers
o469p/a>
o470p/a>===========================
o393p/a>
o324p/a>Tho324eved hus, i startup, y bmsst inste">o356p/Public L /pnmprdevelopers
o405p/a>
ompress9p/a>he mainrerefore, th;Basncallyre4eved hus,3p/a>nd/od allif8. The only
o2p/a startup, y bmsst inste"NUe">o356p/Public L /pnmpnforpublii32 mai. The only
oinsteFrereSore, theFick ="Doc; eiocesssn>
, gc2st insteL /pnmp,ork in ITU or
o2(. Check as ins)ersn>aschinsn>
, grdevelopers
o450p/a>
ompress9p/a>he mai4eved hus, i 100h focan asstC10+/Dbople fulth debug=2,
o3er bWITHOUT ANY WARRANTY;ompressn> Andnce>Ifmplthe , to4l the h debug=2,
o2MERCHANTABILITYug rFITNESS FOR A PARTICULAR PURPOSE. verbost), give the
o3"NUe">o356p/Public L /pnmpnfod alfor so463 developers.
o405p/a>
omYsncmainboalock,>
ti beea copyst inste"NUe">o356p/Public L /pnmp">o405p/a>
o2aloe ex to work 9p/a>he;n>
2) anwritbly knsteFrereSore, th">o405p/a>
oiFick ="Doc, Inc., 675 Mv3L Avp,oCambrid, onMA 02139, USel, bad=VIA.
o469p/a>
o480p/a>
/video4lin5x/Zoran#L511" idv3L411" class="lin5" nam5v3L411

to -rigim" aLXRefore, the/a>prov">o480p/a>LXRe334p/a>tyL411, work /a>If href=>the kerne/a>">o480p/a sento:lxi@n>
/v.no">lxi@n>
/v.noL411.

lxi.n>
/v.no kinso shoerede/a>">o480p/ao4linux/Zorre4p0+/-n>
" i.no">Re4p0+/gtoo" i ASL411, est. Alrst iL>
/vEach of the /a>fr>Iflar ca s in /proccace 1995.