Commit | Line | Data |
---|---|---|
1da177e4 LT |
1 | Frequently Asked Questions: |
2 | =========================== | |
3 | subject: unified zoran driver (zr360x7, zoran, buz, dc10(+), dc30(+), lml33) | |
4 | website: http://mjpeg.sourceforge.net/driver-zoran/ | |
5 | ||
6 | 1. What cards are supported | |
7 | 1.1 What the TV decoder can do an what not | |
8 | 1.2 What the TV encoder can do an what not | |
9 | 2. How do I get this damn thing to work | |
10 | 3. What mainboard should I use (or why doesn't my card work) | |
11 | 4. Programming interface | |
12 | 5. Applications | |
13 | 6. Concerning buffer sizes, quality, output size etc. | |
14 | 7. It hangs/crashes/fails/whatevers! Help! | |
15 | 8. Maintainers/Contacting | |
16 | 9. License | |
17 | ||
18 | =========================== | |
19 | ||
20 | 1. What cards are supported | |
21 | ||
22 | Iomega Buz, Linux Media Labs LML33/LML33R10, Pinnacle/Miro | |
23 | DC10/DC10+/DC30/DC30+ and related boards (available under various names). | |
24 | ||
25 | Iomega Buz: | |
26 | * Zoran zr36067 PCI controller | |
27 | * Zoran zr36060 MJPEG codec | |
28 | * Philips saa7111 TV decoder | |
29 | * Philips saa7185 TV encoder | |
30 | Drivers to use: videodev, i2c-core, i2c-algo-bit, | |
48773e68 | 31 | videocodec, saa7111, saa7185, zr36060, zr36067 |
1da177e4 LT |
32 | Inputs/outputs: Composite and S-video |
33 | Norms: PAL, SECAM (720x576 @ 25 fps), NTSC (720x480 @ 29.97 fps) | |
34 | Card number: 7 | |
35 | ||
fbe60daa MS |
36 | AverMedia 6 Eyes AVS6EYES: |
37 | * Zoran zr36067 PCI controller | |
38 | * Zoran zr36060 MJPEG codec | |
39 | * Samsung ks0127 TV decoder | |
40 | * Conexant bt866 TV encoder | |
41 | Drivers to use: videodev, i2c-core, i2c-algo-bit, | |
42 | videocodec, ks0127, bt866, zr36060, zr36067 | |
43 | Inputs/outputs: Six physical inputs. 1-6 are composite, | |
44 | 1-2, 3-4, 5-6 doubles as S-video, | |
45 | 1-3 triples as component. | |
46 | One composite output. | |
47 | Norms: PAL, SECAM (720x576 @ 25 fps), NTSC (720x480 @ 29.97 fps) | |
48 | Card number: 8 | |
49 | Not autodetected, card=8 is necessary. | |
50 | ||
1da177e4 LT |
51 | Linux Media Labs LML33: |
52 | * Zoran zr36067 PCI controller | |
53 | * Zoran zr36060 MJPEG codec | |
54 | * Brooktree bt819 TV decoder | |
55 | * Brooktree bt856 TV encoder | |
56 | Drivers to use: videodev, i2c-core, i2c-algo-bit, | |
48773e68 | 57 | videocodec, bt819, bt856, zr36060, zr36067 |
1da177e4 LT |
58 | Inputs/outputs: Composite and S-video |
59 | Norms: PAL (720x576 @ 25 fps), NTSC (720x480 @ 29.97 fps) | |
60 | Card number: 5 | |
61 | ||
62 | Linux Media Labs LML33R10: | |
63 | * Zoran zr36067 PCI controller | |
64 | * Zoran zr36060 MJPEG codec | |
65 | * Philips saa7114 TV decoder | |
66 | * Analog Devices adv7170 TV encoder | |
67 | Drivers to use: videodev, i2c-core, i2c-algo-bit, | |
48773e68 | 68 | videocodec, saa7114, adv7170, zr36060, zr36067 |
1da177e4 LT |
69 | Inputs/outputs: Composite and S-video |
70 | Norms: PAL (720x576 @ 25 fps), NTSC (720x480 @ 29.97 fps) | |
71 | Card number: 6 | |
72 | ||
73 | Pinnacle/Miro DC10(new): | |
74 | * Zoran zr36057 PCI controller | |
75 | * Zoran zr36060 MJPEG codec | |
76 | * Philips saa7110a TV decoder | |
77 | * Analog Devices adv7176 TV encoder | |
78 | Drivers to use: videodev, i2c-core, i2c-algo-bit, | |
48773e68 | 79 | videocodec, saa7110, adv7175, zr36060, zr36067 |
1da177e4 LT |
80 | Inputs/outputs: Composite, S-video and Internal |
81 | Norms: PAL, SECAM (768x576 @ 25 fps), NTSC (640x480 @ 29.97 fps) | |
82 | Card number: 1 | |
83 | ||
84 | Pinnacle/Miro DC10+: | |
85 | * Zoran zr36067 PCI controller | |
86 | * Zoran zr36060 MJPEG codec | |
87 | * Philips saa7110a TV decoder | |
88 | * Analog Devices adv7176 TV encoder | |
89 | Drivers to use: videodev, i2c-core, i2c-algo-bit, | |
90 | videocodec, sa7110, adv7175, zr36060, zr36067 | |
91 | Inputs/outputs: Composite, S-video and Internal | |
92 | Norms: PAL, SECAM (768x576 @ 25 fps), NTSC (640x480 @ 29.97 fps) | |
93 | Card number: 2 | |
94 | ||
95 | Pinnacle/Miro DC10(old): * | |
96 | * Zoran zr36057 PCI controller | |
97 | * Zoran zr36050 MJPEG codec | |
98 | * Zoran zr36016 Video Front End or Fuji md0211 Video Front End (clone?) | |
99 | * Micronas vpx3220a TV decoder | |
100 | * mse3000 TV encoder or Analog Devices adv7176 TV encoder * | |
101 | Drivers to use: videodev, i2c-core, i2c-algo-bit, | |
48773e68 | 102 | videocodec, vpx3220, mse3000/adv7175, zr36050, zr36016, zr36067 |
1da177e4 LT |
103 | Inputs/outputs: Composite, S-video and Internal |
104 | Norms: PAL, SECAM (768x576 @ 25 fps), NTSC (640x480 @ 29.97 fps) | |
105 | Card number: 0 | |
106 | ||
107 | Pinnacle/Miro DC30: * | |
108 | * Zoran zr36057 PCI controller | |
109 | * Zoran zr36050 MJPEG codec | |
110 | * Zoran zr36016 Video Front End | |
111 | * Micronas vpx3225d/vpx3220a/vpx3216b TV decoder | |
112 | * Analog Devices adv7176 TV encoder | |
113 | Drivers to use: videodev, i2c-core, i2c-algo-bit, | |
48773e68 | 114 | videocodec, vpx3220/vpx3224, adv7175, zr36050, zr36016, zr36067 |
1da177e4 LT |
115 | Inputs/outputs: Composite, S-video and Internal |
116 | Norms: PAL, SECAM (768x576 @ 25 fps), NTSC (640x480 @ 29.97 fps) | |
117 | Card number: 3 | |
118 | ||
119 | Pinnacle/Miro DC30+: * | |
120 | * Zoran zr36067 PCI controller | |
121 | * Zoran zr36050 MJPEG codec | |
122 | * Zoran zr36016 Video Front End | |
123 | * Micronas vpx3225d/vpx3220a/vpx3216b TV decoder | |
124 | * Analog Devices adv7176 TV encoder | |
125 | Drivers to use: videodev, i2c-core, i2c-algo-bit, | |
126 | videocodec, vpx3220/vpx3224, adv7175, zr36050, zr36015, zr36067 | |
127 | Inputs/outputs: Composite, S-video and Internal | |
128 | Norms: PAL, SECAM (768x576 @ 25 fps), NTSC (640x480 @ 29.97 fps) | |
129 | Card number: 4 | |
130 | ||
131 | Note: No module for the mse3000 is available yet | |
132 | Note: No module for the vpx3224 is available yet | |
133 | Note: use encoder=X or decoder=X for non-default i2c chips (see i2c-id.h) | |
134 | ||
135 | =========================== | |
136 | ||
137 | 1.1 What the TV decoder can do an what not | |
138 | ||
139 | The best know TV standards are NTSC/PAL/SECAM. but for decoding a frame that | |
140 | information is not enough. There are several formats of the TV standards. | |
48773e68 MCC |
141 | And not every TV decoder is able to handle every format. Also the every |
142 | combination is supported by the driver. There are currently 11 different | |
143 | tv broadcast formats all aver the world. | |
1da177e4 | 144 | |
48773e68 | 145 | The CCIR defines parameters needed for broadcasting the signal. |
1da177e4 | 146 | The CCIR has defined different standards: A,B,D,E,F,G,D,H,I,K,K1,L,M,N,... |
670e9f34 | 147 | The CCIR says not much about the colorsystem used !!! |
1da177e4 LT |
148 | And talking about a colorsystem says not to much about how it is broadcast. |
149 | ||
150 | The CCIR standards A,E,F are not used any more. | |
151 | ||
152 | When you speak about NTSC, you usually mean the standard: CCIR - M using | |
153 | the NTSC colorsystem which is used in the USA, Japan, Mexico, Canada | |
48773e68 | 154 | and a few others. |
1da177e4 LT |
155 | |
156 | When you talk about PAL, you usually mean: CCIR - B/G using the PAL | |
48773e68 | 157 | colorsystem which is used in many Countries. |
1da177e4 | 158 | |
48773e68 | 159 | When you talk about SECAM, you mean: CCIR - L using the SECAM Colorsystem |
1da177e4 LT |
160 | which is used in France, and a few others. |
161 | ||
162 | There the other version of SECAM, CCIR - D/K is used in Bulgaria, China, | |
48773e68 | 163 | Slovakai, Hungary, Korea (Rep.), Poland, Rumania and a others. |
1da177e4 | 164 | |
48773e68 | 165 | The CCIR - H uses the PAL colorsystem (sometimes SECAM) and is used in |
1da177e4 LT |
166 | Egypt, Libya, Sri Lanka, Syrain Arab. Rep. |
167 | ||
168 | The CCIR - I uses the PAL colorsystem, and is used in Great Britain, Hong Kong, | |
169 | Ireland, Nigeria, South Africa. | |
170 | ||
171 | The CCIR - N uses the PAL colorsystem and PAL frame size but the NTSC framerate, | |
172 | and is used in Argentinia, Uruguay, an a few others | |
173 | ||
174 | We do not talk about how the audio is broadcast ! | |
175 | ||
48773e68 | 176 | A rather good sites about the TV standards are: |
1da177e4 LT |
177 | http://www.sony.jp/ServiceArea/Voltage_map/ |
178 | http://info.electronicwerkstatt.de/bereiche/fernsehtechnik/frequenzen_und_normen/Fernsehnormen/ | |
179 | and http://www.cabl.com/restaurant/channel.html | |
180 | ||
181 | Other weird things around: NTSC 4.43 is a modificated NTSC, which is mainly | |
182 | used in PAL VCR's that are able to play back NTSC. PAL 60 seems to be the same | |
48773e68 MCC |
183 | as NTSC 4.43 . The Datasheets also talk about NTSC 44, It seems as if it would |
184 | be the same as NTSC 4.43. | |
1da177e4 LT |
185 | NTSC Combs seems to be a decoder mode where the decoder uses a comb filter |
186 | to split coma and luma instead of a Delay line. | |
187 | ||
188 | But I did not defiantly find out what NTSC Comb is. | |
189 | ||
190 | Philips saa7111 TV decoder | |
48773e68 MCC |
191 | was introduced in 1997, is used in the BUZ and |
192 | can handle: PAL B/G/H/I, PAL N, PAL M, NTSC M, NTSC N, NTSC 4.43 and SECAM | |
1da177e4 LT |
193 | |
194 | Philips saa7110a TV decoder | |
195 | was introduced in 1995, is used in the Pinnacle/Miro DC10(new), DC10+ and | |
48773e68 | 196 | can handle: PAL B/G, NTSC M and SECAM |
1da177e4 LT |
197 | |
198 | Philips saa7114 TV decoder | |
48773e68 | 199 | was introduced in 2000, is used in the LML33R10 and |
1da177e4 LT |
200 | can handle: PAL B/G/D/H/I/N, PAL N, PAL M, NTSC M, NTSC 4.43 and SECAM |
201 | ||
202 | Brooktree bt819 TV decoder | |
203 | was introduced in 1996, and is used in the LML33 and | |
204 | can handle: PAL B/D/G/H/I, NTSC M | |
205 | ||
206 | Micronas vpx3220a TV decoder | |
207 | was introduced in 1996, is used in the DC30 and DC30+ and | |
208 | can handle: PAL B/G/H/I, PAL N, PAL M, NTSC M, NTSC 44, PAL 60, SECAM,NTSC Comb | |
209 | ||
fbe60daa MS |
210 | Samsung ks0127 TV decoder |
211 | is used in the AVS6EYES card and | |
212 | can handle: NTSC-M/N/44, PAL-M/N/B/G/H/I/D/K/L and SECAM | |
213 | ||
1da177e4 LT |
214 | =========================== |
215 | ||
216 | 1.2 What the TV encoder can do an what not | |
217 | ||
218 | The TV encoder are doing the "same" as the decoder, but in the oder direction. | |
219 | You feed them digital data and the generate a Composite or SVHS signal. | |
220 | For information about the colorsystems and TV norm take a look in the | |
221 | TV decoder section. | |
222 | ||
223 | Philips saa7185 TV Encoder | |
224 | was introduced in 1996, is used in the BUZ | |
225 | can generate: PAL B/G, NTSC M | |
226 | ||
227 | Brooktree bt856 TV Encoder | |
48773e68 | 228 | was introduced in 1994, is used in the LML33 |
1da177e4 LT |
229 | can generate: PAL B/D/G/H/I/N, PAL M, NTSC M, PAL-N (Argentina) |
230 | ||
231 | Analog Devices adv7170 TV Encoder | |
232 | was introduced in 2000, is used in the LML300R10 | |
233 | can generate: PAL B/D/G/H/I/N, PAL M, NTSC M, PAL 60 | |
234 | ||
235 | Analog Devices adv7175 TV Encoder | |
236 | was introduced in 1996, is used in the DC10, DC10+, DC10 old, DC30, DC30+ | |
237 | can generate: PAL B/D/G/H/I/N, PAL M, NTSC M | |
238 | ||
239 | ITT mse3000 TV encoder | |
240 | was introduced in 1991, is used in the DC10 old | |
241 | can generate: PAL , NTSC , SECAM | |
242 | ||
fbe60daa MS |
243 | Conexant bt866 TV encoder |
244 | is used in AVS6EYES, and | |
be2a608b | 245 | can generate: NTSC/PAL, PALÂM, PALÂN |
fbe60daa | 246 | |
48773e68 | 247 | The adv717x, should be able to produce PAL N. But you find nothing PAL N |
d533f671 | 248 | specific in the registers. Seem that you have to reuse a other standard |
48773e68 | 249 | to generate PAL N, maybe it would work if you use the PAL M settings. |
1da177e4 LT |
250 | |
251 | ========================== | |
252 | ||
253 | 2. How do I get this damn thing to work | |
254 | ||
255 | Load zr36067.o. If it can't autodetect your card, use the card=X insmod | |
256 | option with X being the card number as given in the previous section. | |
257 | To have more than one card, use card=X1[,X2[,X3,[X4[..]]]] | |
258 | ||
259 | To automate this, add the following to your /etc/modprobe.conf: | |
260 | ||
261 | options zr36067 card=X1[,X2[,X3[,X4[..]]]] | |
262 | alias char-major-81-0 zr36067 | |
263 | ||
264 | One thing to keep in mind is that this doesn't load zr36067.o itself yet. It | |
265 | just automates loading. If you start using xawtv, the device won't load on | |
266 | some systems, since you're trying to load modules as a user, which is not | |
267 | allowed ("permission denied"). A quick workaround is to add 'Load "v4l"' to | |
268 | XF86Config-4 when you use X by default, or to run 'v4l-conf -c <device>' in | |
269 | one of your startup scripts (normally rc.local) if you don't use X. Both | |
270 | make sure that the modules are loaded on startup, under the root account. | |
271 | ||
272 | =========================== | |
273 | ||
274 | 3. What mainboard should I use (or why doesn't my card work) | |
275 | ||
276 | <insert lousy disclaimer here>. In short: good=SiS/Intel, bad=VIA. | |
277 | ||
278 | Experience tells us that people with a Buz, on average, have more problems | |
279 | than users with a DC10+/LML33. Also, it tells us that people owning a VIA- | |
280 | based mainboard (ktXXX, MVP3) have more problems than users with a mainboard | |
281 | based on a different chipset. Here's some notes from Andrew Stevens: | |
282 | -- | |
283 | Here's my experience of using LML33 and Buz on various motherboards: | |
284 | ||
285 | VIA MVP3 | |
286 | Forget it. Pointless. Doesn't work. | |
48773e68 | 287 | Intel 430FX (Pentium 200) |
1da177e4 LT |
288 | LML33 perfect, Buz tolerable (3 or 4 frames dropped per movie) |
289 | Intel 440BX (early stepping) | |
290 | LML33 tolerable. Buz starting to get annoying (6-10 frames/hour) | |
291 | Intel 440BX (late stepping) | |
292 | Buz tolerable, LML3 almost perfect (occasional single frame drops) | |
293 | SiS735 | |
294 | LML33 perfect, Buz tolerable. | |
295 | VIA KT133(*) | |
296 | LML33 starting to get annoying, Buz poor enough that I have up. | |
297 | ||
298 | Both 440BX boards were dual CPU versions. | |
299 | -- | |
300 | Bernhard Praschinger later added: | |
301 | -- | |
302 | AMD 751 | |
303 | Buz perfect-tolerable | |
304 | AMD 760 | |
305 | Buz perfect-tolerable | |
306 | -- | |
307 | In general, people on the user mailinglist won't give you much of a chance | |
308 | if you have a VIA-based motherboard. They may be cheap, but sometimes, you'd | |
309 | rather want to spend some more money on better boards. In general, VIA | |
310 | mainboard's IDE/PCI performance will also suck badly compared to others. | |
311 | You'll noticed the DC10+/DC30+ aren't mentioned anywhere in the overview. | |
312 | Basically, you can assume that if the Buz works, the LML33 will work too. If | |
313 | the LML33 works, the DC10+/DC30+ will work too. They're most tolerant to | |
314 | different mainboard chipsets from all of the supported cards. | |
315 | ||
316 | If you experience timeouts during capture, buy a better mainboard or lower | |
317 | the quality/buffersize during capture (see 'Concerning buffer sizes, quality, | |
318 | output size etc.'). If it hangs, there's little we can do as of now. Check | |
319 | your IRQs and make sure the card has its own interrupts. | |
320 | ||
321 | =========================== | |
322 | ||
323 | 4. Programming interface | |
324 | ||
325 | This driver conforms to video4linux and video4linux2, both can be used to | |
326 | use the driver. Since video4linux didn't provide adequate calls to fully | |
327 | use the cards' features, we've introduced several programming extensions, | |
328 | which are currently officially accepted in the 2.4.x branch of the kernel. | |
329 | These extensions are known as the v4l/mjpeg extensions. See zoran.h for | |
330 | details (structs/ioctls). | |
331 | ||
332 | Information - video4linux: | |
333 | http://roadrunner.swansea.linux.org.uk/v4lapi.shtml | |
334 | Documentation/video4linux/API.html | |
335 | /usr/include/linux/videodev.h | |
336 | ||
337 | Information - video4linux/mjpeg extensions: | |
338 | ./zoran.h | |
339 | (also see below) | |
340 | ||
341 | Information - video4linux2: | |
43db48d3 MCC |
342 | http://linuxtv.org |
343 | http://v4l2spec.bytesex.org/ | |
1da177e4 | 344 | /usr/include/linux/videodev2.h |
1da177e4 LT |
345 | |
346 | More information on the video4linux/mjpeg extensions, by Serguei | |
347 | Miridonovi and Rainer Johanni: | |
348 | -- | |
349 | The ioctls for that interface are as follows: | |
350 | ||
351 | BUZIOC_G_PARAMS | |
352 | BUZIOC_S_PARAMS | |
353 | ||
354 | Get and set the parameters of the buz. The user should always do a | |
355 | BUZIOC_G_PARAMS (with a struct buz_params) to obtain the default | |
356 | settings, change what he likes and then make a BUZIOC_S_PARAMS call. | |
357 | ||
358 | BUZIOC_REQBUFS | |
359 | ||
360 | Before being able to capture/playback, the user has to request | |
361 | the buffers he is wanting to use. Fill the structure | |
362 | zoran_requestbuffers with the size (recommended: 256*1024) and | |
363 | the number (recommended 32 up to 256). There are no such restrictions | |
364 | as for the Video for Linux buffers, you should LEAVE SUFFICIENT | |
365 | MEMORY for your system however, else strange things will happen .... | |
366 | On return, the zoran_requestbuffers structure contains number and | |
367 | size of the actually allocated buffers. | |
368 | You should use these numbers for doing a mmap of the buffers | |
369 | into the user space. | |
370 | The BUZIOC_REQBUFS ioctl also makes it happen, that the next mmap | |
371 | maps the MJPEG buffer instead of the V4L buffers. | |
372 | ||
373 | BUZIOC_QBUF_CAPT | |
374 | BUZIOC_QBUF_PLAY | |
375 | ||
376 | Queue a buffer for capture or playback. The first call also starts | |
377 | streaming capture. When streaming capture is going on, you may | |
378 | only queue further buffers or issue syncs until streaming | |
379 | capture is switched off again with a argument of -1 to | |
380 | a BUZIOC_QBUF_CAPT/BUZIOC_QBUF_PLAY ioctl. | |
381 | ||
382 | BUZIOC_SYNC | |
383 | ||
384 | Issue this ioctl when all buffers are queued. This ioctl will | |
385 | block until the first buffer becomes free for saving its | |
386 | data to disk (after BUZIOC_QBUF_CAPT) or for reuse (after BUZIOC_QBUF_PLAY). | |
387 | ||
388 | BUZIOC_G_STATUS | |
389 | ||
390 | Get the status of the input lines (video source connected/norm). | |
391 | ||
392 | For programming example, please, look at lavrec.c and lavplay.c code in | |
393 | lavtools-1.2p2 package (URL: http://www.cicese.mx/~mirsev/DC10plus/) | |
394 | and the 'examples' directory in the original Buz driver distribution. | |
395 | ||
396 | Additional notes for software developers: | |
397 | ||
398 | The driver returns maxwidth and maxheight parameters according to | |
399 | the current TV standard (norm). Therefore, the software which | |
400 | communicates with the driver and "asks" for these parameters should | |
401 | first set the correct norm. Well, it seems logically correct: TV | |
402 | standard is "more constant" for current country than geometry | |
403 | settings of a variety of TV capture cards which may work in ITU or | |
404 | square pixel format. Remember that users now can lock the norm to | |
405 | avoid any ambiguity. | |
406 | -- | |
407 | Please note that lavplay/lavrec are also included in the MJPEG-tools | |
408 | (http://mjpeg.sf.net/). | |
409 | ||
410 | =========================== | |
411 | ||
412 | 5. Applications | |
413 | ||
414 | Applications known to work with this driver: | |
415 | ||
416 | TV viewing: | |
417 | * xawtv | |
418 | * kwintv | |
419 | * probably any TV application that supports video4linux or video4linux2. | |
420 | ||
421 | MJPEG capture/playback: | |
422 | * mjpegtools/lavtools (or Linux Video Studio) | |
423 | * gstreamer | |
424 | * mplayer | |
425 | ||
426 | General raw capture: | |
427 | * xawtv | |
428 | * gstreamer | |
429 | * probably any application that supports video4linux or video4linux2 | |
430 | ||
431 | Video editing: | |
432 | * Cinelerra | |
433 | * MainActor | |
434 | * mjpegtools (or Linux Video Studio) | |
435 | ||
436 | =========================== | |
437 | ||
438 | 6. Concerning buffer sizes, quality, output size etc. | |
439 | ||
440 | The zr36060 can do 1:2 JPEG compression. This is really the theoretical | |
441 | maximum that the chipset can reach. The driver can, however, limit compression | |
442 | to a maximum (size) of 1:4. The reason for this is that some cards (e.g. Buz) | |
443 | can't handle 1:2 compression without stopping capture after only a few minutes. | |
444 | With 1:4, it'll mostly work. If you have a Buz, use 'low_bitrate=1' to go into | |
445 | 1:4 max. compression mode. | |
446 | ||
447 | 100% JPEG quality is thus 1:2 compression in practice. So for a full PAL frame | |
448 | (size 720x576). The JPEG fields are stored in YUY2 format, so the size of the | |
449 | fields are 720x288x16/2 bits/field (2 fields/frame) = 207360 bytes/field x 2 = | |
450 | 414720 bytes/frame (add some more bytes for headers and DHT (huffman)/DQT | |
451 | (quantization) tables, and you'll get to something like 512kB per frame for | |
452 | 1:2 compression. For 1:4 compression, you'd have frames of half this size. | |
453 | ||
454 | Some additional explanation by Martin Samuelsson, which also explains the | |
455 | importance of buffer sizes: | |
456 | -- | |
457 | > Hmm, I do not think it is really that way. With the current (downloaded | |
458 | > at 18:00 Monday) driver I get that output sizes for 10 sec: | |
459 | > -q 50 -b 128 : 24.283.332 Bytes | |
460 | > -q 50 -b 256 : 48.442.368 | |
461 | > -q 25 -b 128 : 24.655.992 | |
462 | > -q 25 -b 256 : 25.859.820 | |
463 | ||
48773e68 | 464 | I woke up, and can't go to sleep again. I'll kill some time explaining why |
1da177e4 LT |
465 | this doesn't look strange to me. |
466 | ||
48773e68 | 467 | Let's do some math using a width of 704 pixels. I'm not sure whether the Buz |
1da177e4 LT |
468 | actually use that number or not, but that's not too important right now. |
469 | ||
48773e68 MCC |
470 | 704x288 pixels, one field, is 202752 pixels. Divided by 64 pixels per block; |
471 | 3168 blocks per field. Each pixel consist of two bytes; 128 bytes per block; | |
472 | 1024 bits per block. 100% in the new driver mean 1:2 compression; the maximum | |
473 | output becomes 512 bits per block. Actually 510, but 512 is simpler to use | |
1da177e4 LT |
474 | for calculations. |
475 | ||
48773e68 MCC |
476 | Let's say that we specify d1q50. We thus want 256 bits per block; times 3168 |
477 | becomes 811008 bits; 101376 bytes per field. We're talking raw bits and bytes | |
478 | here, so we don't need to do any fancy corrections for bits-per-pixel or such | |
1da177e4 LT |
479 | things. 101376 bytes per field. |
480 | ||
48773e68 | 481 | d1 video contains two fields per frame. Those sum up to 202752 bytes per |
1da177e4 LT |
482 | frame, and one of those frames goes into each buffer. |
483 | ||
48773e68 | 484 | But wait a second! -b128 gives 128kB buffers! It's not possible to cram |
1da177e4 LT |
485 | 202752 bytes of JPEG data into 128kB! |
486 | ||
48773e68 | 487 | This is what the driver notice and automatically compensate for in your |
1da177e4 LT |
488 | examples. Let's do some math using this information: |
489 | ||
48773e68 MCC |
490 | 128kB is 131072 bytes. In this buffer, we want to store two fields, which |
491 | leaves 65536 bytes for each field. Using 3168 blocks per field, we get | |
492 | 20.68686868... available bytes per block; 165 bits. We can't allow the | |
493 | request for 256 bits per block when there's only 165 bits available! The -q50 | |
494 | option is silently overridden, and the -b128 option takes precedence, leaving | |
1da177e4 LT |
495 | us with the equivalence of -q32. |
496 | ||
48773e68 MCC |
497 | This gives us a data rate of 165 bits per block, which, times 3168, sums up |
498 | to 65340 bytes per field, out of the allowed 65536. The current driver has | |
499 | another level of rate limiting; it won't accept -q values that fill more than | |
500 | 6/8 of the specified buffers. (I'm not sure why. "Playing it safe" seem to be | |
501 | a safe bet. Personally, I think I would have lowered requested-bits-per-block | |
502 | by one, or something like that.) We can't use 165 bits per block, but have to | |
503 | lower it again, to 6/8 of the available buffer space: We end up with 124 bits | |
504 | per block, the equivalence of -q24. With 128kB buffers, you can't use greater | |
1da177e4 LT |
505 | than -q24 at -d1. (And PAL, and 704 pixels width...) |
506 | ||
48773e68 MCC |
507 | The third example is limited to -q24 through the same process. The second |
508 | example, using very similar calculations, is limited to -q48. The only | |
509 | example that actually grab at the specified -q value is the last one, which | |
1da177e4 LT |
510 | is clearly visible, looking at the file size. |
511 | -- | |
512 | ||
513 | Conclusion: the quality of the resulting movie depends on buffer size, quality, | |
514 | whether or not you use 'low_bitrate=1' as insmod option for the zr36060.c | |
515 | module to do 1:4 instead of 1:2 compression, etc. | |
516 | ||
517 | If you experience timeouts, lowering the quality/buffersize or using | |
518 | 'low_bitrate=1 as insmod option for zr36060.o might actually help, as is | |
519 | proven by the Buz. | |
520 | ||
521 | =========================== | |
522 | ||
523 | 7. It hangs/crashes/fails/whatevers! Help! | |
524 | ||
525 | Make sure that the card has its own interrupts (see /proc/interrupts), check | |
526 | the output of dmesg at high verbosity (load zr36067.o with debug=2, | |
527 | load all other modules with debug=1). Check that your mainboard is favorable | |
528 | (see question 2) and if not, test the card in another computer. Also see the | |
529 | notes given in question 3 and try lowering quality/buffersize/capturesize | |
530 | if recording fails after a period of time. | |
531 | ||
532 | If all this doesn't help, give a clear description of the problem including | |
533 | detailed hardware information (memory+brand, mainboard+chipset+brand, which | |
534 | MJPEG card, processor, other PCI cards that might be of interest), give the | |
535 | system PnP information (/proc/interrupts, /proc/dma, /proc/devices), and give | |
536 | the kernel version, driver version, glibc version, gcc version and any other | |
537 | information that might possibly be of interest. Also provide the dmesg output | |
538 | at high verbosity. See 'Contacting' on how to contact the developers. | |
539 | ||
540 | =========================== | |
541 | ||
542 | 8. Maintainers/Contacting | |
543 | ||
544 | The driver is currently maintained by Laurent Pinchart and Ronald Bultje | |
545 | (<laurent.pinchart@skynet.be> and <rbultje@ronald.bitfreak.net>). For bug | |
546 | reports or questions, please contact the mailinglist instead of the developers | |
547 | individually. For user questions (i.e. bug reports or how-to questions), send | |
548 | an email to <mjpeg-users@lists.sf.net>, for developers (i.e. if you want to | |
549 | help programming), send an email to <mjpeg-developer@lists.sf.net>. See | |
550 | http://www.sf.net/projects/mjpeg/ for subscription information. | |
551 | ||
552 | For bug reports, be sure to include all the information as described in | |
553 | the section 'It hangs/crashes/fails/whatevers! Help!'. Please make sure | |
554 | you're using the latest version (http://mjpeg.sf.net/driver-zoran/). | |
555 | ||
556 | Previous maintainers/developers of this driver include Serguei Miridonov | |
557 | <mirsev@cicese.mx>, Wolfgang Scherr <scherr@net4you.net>, Dave Perks | |
558 | <dperks@ibm.net> and Rainer Johanni <Rainer@Johanni.de>. | |
559 | ||
560 | =========================== | |
561 | ||
562 | 9. License | |
563 | ||
564 | This driver is distributed under the terms of the General Public License. | |
565 | ||
566 | This program is free software; you can redistribute it and/or modify | |
567 | it under the terms of the GNU General Public License as published by | |
568 | the Free Software Foundation; either version 2 of the License, or | |
569 | (at your option) any later version. | |
570 | ||
571 | This program is distributed in the hope that it will be useful, | |
572 | but WITHOUT ANY WARRANTY; without even the implied warranty of | |
573 | MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the | |
574 | GNU General Public License for more details. | |
575 | ||
576 | You should have received a copy of the GNU General Public License | |
577 | along with this program; if not, write to the Free Software | |
578 | Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. | |
579 | ||
580 | See http://www.gnu.org/ for more information. |