Log in

View Full Version : ffdshow tryouts project: Discussion & Development


Pages : 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 [77] 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308

Leak
19th June 2008, 08:51
It is not in any way important to me, but I am curious why it is impossible? libavcodec itself does offer the functionality, at least for some codecs.
Well, since all non-I-frames in MPEG2 fully depend on neighbouring decoded frames you can't just "decode it to half the size" and expect something watchable to come out of it.

And since you have to decode it at full resolution scaling it down afterwards only increases CPU load - and you can already do that using ffdshow's Resize & Aspect filter, anyway.

Kurosu
19th June 2008, 11:00
Well, since all non-I-frames in MPEG2 fully depend on neighbouring decoded frames you can't just "decode it to half the size" and expect something watchable to come out of it.

And since you have to decode it at full resolution scaling it down afterwards only increases CPU load - and you can already do that using ffdshow's Resize & Aspect filter, anyway.

This is somewhat funny: would you also explain to David Huffman, were he alive, what variable length coding is?

Distortion is indeed a side effect, but it does improve decoding speed. The mplayer man page for instance, and google, have explanations about the lowres decoding option found in lavc.

Leak
19th June 2008, 14:18
This is somewhat funny: would you also explain to David Huffman, were he alive, what variable length coding is?
Errr... so you're implying that David Huffman is to Variable Length Coding as Reimar is to MPEG2? :confused: (Sorry, I really genuinely don't understand that comment of yours...)

LoRd_MuldeR
19th June 2008, 14:32
Errr... so you're implying that David Huffman is to Variable Length Coding as Reimar is to MPEG2? :confused: (Sorry, I really genuinely don't understand that comment of yours...)

I think what he actually meant is:
You don't need to explain how MPEG-2 decoding works to Reimar, because as an MPlayer/Codec developer he should be very familiar with it...

Leak
19th June 2008, 15:08
I think what he actually meant is:
You don't need to explain how MPEG-2 decoding works to Reimar, because as an MPlayer/Codec developer he should be very familiar with it...
Ehehe... ummm, since he asked that question above I didn't consider that he could be an mplayer dev... :)

Thanks for clearing that up...

Mr VacBob
20th June 2008, 03:06
For libavcodec's mpeg-2 decoder there already is a "use speedup tricks" option in ffdshow. I don't know what that does precisely.

It (CODEC_FLAG2_FAST) uses a simpler dequantization step which results in more zero DCT coefficients. It's not that much faster, really.

What he asked for is controlled by AVCodecContext.lowres, which certainly works for MPEG2. I think it might already be exposed in the UI too.

nxmehta
20th June 2008, 11:09
Hi all,

There's a bug in ffdshow (reported at http://sourceforge.net/tracker/index.php?func=detail&aid=1924063&group_id=173941&atid=867360 ) that makes some of the software I'm running unusable. I was just wondering if this bug was fixed in any of the nightly builds? If not, is it something that might be fixed in the future?

Thanks.

leeperry
20th June 2008, 13:29
is it me, or is it only possible to resize horizontally in ffdshow ?!

jmartinr
20th June 2008, 19:44
is it me, or is it only possible to resize horizontally in ffdshow ?!

It's you :p. Horizontal and vertical resize works fine here. Maybe the problem is in your renderer.

leeperry
20th June 2008, 20:32
well you can keep the exact same width and only resize vertically ?

now how do you do that ?

I've got a profile to resize to 1024x768, yet sometimes it resizes to 1152x768 with SD content.....I want freaking 1024*xxx :(

cyberbeing
20th June 2008, 21:20
It works here as well. Use "Specify horizontal and vertical size" > only change the vertical size > make sure Aspect ratio is set to "no aspect ratio correction".

I'm assuming you're using YV12 and YUY2 with VMR/EVR? With those color spaces VMR/EVR will sometimes use odd internal resolutions (which seems to be a built-in restriction) yet it will still display at your desired resolution.

clsid
20th June 2008, 21:21
Maybe those are anamorphic encodes? Check your PAR settings. Because it should work.

JarrettH
20th June 2008, 22:25
think we'll have a new beta soon? :devil:

clsid
20th June 2008, 22:37
At the end of the month if no regressions are reported.

LoRd_MuldeR
20th June 2008, 23:51
At the end of the month if no regressions are reported.
Apparently the "Cubic Blending" deinterlacer doesn't deinterlace at all or it fails horrible to remove combing artifacts (which I think isn't possible with that method)

Yong
21st June 2008, 05:19
Guys, is it possible to make "custom" OSD display as list?
if the video resolution is small, OSD will got truncated if display all in one line like this
http://xs128.xs.to/xs128/08255/clipboard01263.png

i hope this feature request isn't too hard for devs :p

mike7
21st June 2008, 08:41
Guys, is it possible to make "custom" OSD display as list?

use \n for new line, ex.: %idimar\nRemaining time: %rtime

Reimar
21st June 2008, 09:26
Ehehe... ummm, since he asked that question above I didn't consider that he could be an mplayer dev... :)

Thanks for clearing that up...

Well, you started from the assumption that decoding at half resolution is not possible in a useful way. But the reality is, that even quarter resolution decoding is possible with (IMO) acceptable quality losses simply by setting a libavcodec flag. Half resolution in my tests reduces decoding CPU usage by 60%, quarter resolution by another 20% (compared to half res) for high bitrate HD material.
So my question was about the part I did not know: why it would not be possible to export this functionality via ffdshow (seems I was mistaken to assume you knew about this libavcodec feature).

leeperry
21st June 2008, 10:42
Apparently the "Cubic Blending" deinterlacer doesn't deinterlace at all or it fails horrible to remove combing artifacts (which I think isn't possible with that method)

yeah the deinterlacing section of ffdshow is very poor.

the "cubic blending" doesn't do anything whatsoever, and the "linear blending" is very blurry.

the best deinterlace I've found so far is the PS "blending" script in MPC...but it only works w/ EVR or VMR9

Seb.26 told me that it's using a very simple script, but because noone is willing to add new features to ffdshow......it's not gonna happen anytime soon.

ffdshow is so great, but it's got a few annoying bugs that noone will ever fix I guess :(

for instance....when using Avisynth scripts, very often Reclock says "no video stream found", I stop/press play again, and it's fine...

clsid
21st June 2008, 11:02
but because noone is willing to add new features to ffdshow......it's not gonna happen anytime soon.Comments like this ain't gonna help in upping the motivation of people. Mumbles something about free, open-source, spare time. Currently we only have the manpower to maintain ffdshow. For actual development we simply need a truckload of programmers with no personal life.


I assume that those deinterlacing problems exist for a long time already, and they are no recent regressions. So they won't hold back a beta. But anyway, I am no C++ programmer, and I don't really care about deinterlacing functionality.

leeperry
21st June 2008, 11:10
For actual development we simply need a truckload of programmers with no personal life.
Yeah, where are those guys ?! :D

I'm willing to betatest until the end of days.

well ffdshow is a fantastic piece of software, but it would need some polish on several points(buggy winamp2 plugin support, buggy AVISYNTH script support, buggy deinterlacing, etc etc)

but well there's no alternative to ffdshow anyway.....so I'll just STFU and enjoy what I get(for free at that, as you stated above) ;)

:thanks: all for your fantastic work so far!

Yong
21st June 2008, 15:05
use \n for new line, ex.: %idimar\nRemaining time: %rtime
i did tried with \n but i was using blackslash instead :eek:
what a stupid me :D
thanks btw.

This should make the OSD look more tidy:)

LoRd_MuldeR
21st June 2008, 15:24
yeah the deinterlacing section of ffdshow is very poor.

That's not true. Try "KernelBob" or "TomsMoComp" for example. :rolleyes:

Maybe there are some "better" Deinterlacers out there, but ffdshow includes some very good deinterlacers that work at realtime.
Having deinterlacers in ffdshow that don't work in realtime are kind of pointless...

My tip: If none of the "built-in" deinterlacers makes you happy:
1. Use ffdshow's Avisynth option to load your favorite deinterlacer, e.g. Yadif or TDeint. That's what I do for DVB-S playback.
2. ffdshow can use DScaler plugins, which offers various additional deinterlacers

the "cubic blending" doesn't do anything whatsoever, and the "linear blending" is very blurry.

Yes, "Cubic Blending" is apparently broken. But that's all.

"Linear Blending" is working as it's expected to work. If you don't like that (very simple) method, don't use it :p

ffdshow is so great, but it's got a few annoying bugs that noone will ever fix I guess :(

namely?

for instance....when using Avisynth scripts, very often Reclock says "no video stream found", I stop/press play again, and it's fine...

Why do you think it's the job of the ffdshow developers to workaround bugs in Reclock?


But anyway, I am no C++ programmer, and I don't really care about deinterlacing functionality.
Well, for me it's very important, because I use ffdshow for DVB-S playback. So deinterlace is essential!
And in my opinion ffdshow offers a great number of good deinterlacing methods :)

All I wanted to report is: One single deinterlacer filter appears broken, while all the others work fine.
If you can't fix "Cubic Interpolation", simply disable it for the time being. I won't miss it...

clsid
21st June 2008, 15:38
I'll add it to the bug list.

leeperry
21st June 2008, 18:04
Yes, "Cubic Blending" is apparently broken. But that's all.

"Linear Blending" is working as it's expected to work. If you don't like that (very simple) method, don't use it :p

well I only use deinterlacers for poorly encoded videos with combing artefacts, or interlaced WMV's.

and for this purpose, blending is my only option I think ?! maybe you have some better suggestion ?

in ffdshow, bicubic blending doesn't do jack and linear blending is blurry to hell.

the blending PS script of MPC is not blurry at all, yet it does the job very nicely.....it's in fact a very simple blending script....but it doesn't work with HR, only VMR9 and EVR :(

Why do you think it's the job of the ffdshow developers to workaround bugs in Reclock?

if I don't use the Avisynth capabilities of ffdshow, it NEVER happens.
if I enable it, half of the time it happens....BUT if I press stop/play, everything's fine again :confused:

several friends of mine have the exact same problem...

leeperry
21st June 2008, 18:10
I'll add it to the bug list.

any chance that you add the terrible winamp2 plugin support too, please ?

using the Izotope OzoneMP plugin is freezing to death each time you open a new file :(

there's some deadlock, mostly ffdshow kills the DLL before it closed down :(

AnnaFan777
22nd June 2008, 14:48
Hi, thanks for the great ffdshow,

I'd like to suggest on " resize if "

resize if certain dimension or aspect ratio (or codec like DV) is met:

for example:
resize if -> 720x480 (or the codec is DV) then resize to 640x480
rezise if -> 320x240 (VCD) then resize to 640x480

thanks for the consideration..

AnnaFan777
22nd June 2008, 15:27
Hi

I use ffdshow to watch DV and divx, some are progressive, some interlaced..

I was thinking "Deinterlace IF" option:

activate deinterlace if interlace pattern (comb) is dectected for ?? sec
or disable deinterlace if no comb is dectected for ?? sec

PS: I know it's just a few clicks away, but some are lazy,
and many users have no idea about interlace, they have de-int on,
so they always watch blurry progressive videos.


thanks for the consideration..

clsid
22nd June 2008, 16:09
The resize stuff can mostly be accomplished by using the Profiles/Presets functionality of ffdshow.

You should post your deinterlace feature request on the SourceForge project page. Because otherwise tomorrow everyone will have forgotten about it.

Mercury_22
22nd June 2008, 22:54
In ffdshow_rev2020_20080622_clsid_x64 everything it's OK Finally a working x64 version :thanks:

P.S. NO XviD encoding but who needs it ? :p

Inventive Software
23rd June 2008, 02:28
I have a feature request for the remote API (external key support, basically). Is there a way (or future possibility) within that menu to add another shortcut key or combo for a Bob deinterlacing filter, as well as the Deinterlace filter? To visualise this, imagine there being 2 (or more) sections in the Deinterlacing tab: one for bob filters (LeakKernelDeint, DGBob, Kernel Bob etc), the other for the rest of them (and you could add one for the interpolation filters, for example). Then in the remote API have a key that switches between chosen filters for each kind to accomodate different video content (for example Bob for sports like football and F1, Blend for TV shows, etc). Simpler suggestion: just truncate the Deinterlacing filters and have Bob filters on their own tab, but of course don't allow them both to operate at the same time.

LoRd_MuldeR
23rd June 2008, 16:20
Hi

I use ffdshow to watch DV and divx, some are progressive, some interlaced..

I was thinking "Deinterlace IF" option:

activate deinterlace if interlace pattern (comb) is dectected for ?? sec
or disable deinterlace if no comb is dectected for ?? sec

PS: I know it's just a few clicks away, but some are lazy,
and many users have no idea about interlace, they have de-int on,
so they always watch blurry progressive videos.


thanks for the consideration..

I second that request :)

It would be really useful for DVB playback too, because it switches between film and interlaced all the time :rolleyes:

BTW: I know that TDeint (full = false) as well as DScaler already have combing detection code...

rickardk
24th June 2008, 17:07
Anything changed with input stage off ffdshow audio recently?
I used to be able to have ffdshow audio connected after madFlac by setting format: uncompressed to "all supported".

But this does not work now I think.

clsid
24th June 2008, 20:34
Works fine for me. Tested with latest revision.

Inventive Software
25th June 2008, 02:00
libavcodec MPEG-1 decoder's broken with MPG files, it displays the wrong aspect ratio in MPC. Setting to libmpeg2 fixes the issue.

clsid
25th June 2008, 17:06
Works fine for me. Are those files perhaps anamorphic?

Inventive Software
25th June 2008, 17:13
I think so, the SAR reported is 5 digits / 5 digits, so I guess it's Anamorphic.

clsid
25th June 2008, 17:37
Added to bug list.

thuan
26th June 2008, 15:22
I'm using ffdshow_rev2026_20080625_clsid and keyboard hotkeys don't work. Can anyone confirm?

Inventive Software
26th June 2008, 16:35
You're using Ctrl+Alt when you press them, right?

thuan
26th June 2008, 18:03
Yes, of course. I have used these hot keys for a long time.

Inventive Software
26th June 2008, 19:32
Confirmed, it's broken with 2026. It was working with an older revision, but I think it was a couple hundred versions old.

thuan
27th June 2008, 03:25
Not really, it still worked fine for a couple of build back, or even the 2020 build, but I have no time to check on it now.

LoRd_MuldeR
27th June 2008, 03:28
I'm using ffdshow_rev2026_20080625_clsid and keyboard hotkeys don't work. Can anyone confirm?

Works fine, using the rev2020 ICL build...

clsid
27th June 2008, 11:53
It probably got broken by 2024. I'll revert that later.

iron2000
27th June 2008, 12:18
Not critical but-
Is it possible to have the system time in the OSD without the "System time:" heading?
And also make it 12hr instead of 24hr?

clsid
27th June 2008, 12:36
Is it possible to have the system time in the OSD without the "System time:" heading?Yes, if you create a user defined OSD string.
And also make it 12hr instead of 24hr?Nope.

Thunderbolt8
28th June 2008, 15:46
I noticed there seems to be again a problem with .ass subs (inside a mkv file). for a long time switching to a 2nd subtitle inside a muxed file resulted in some sort of strange characters and .srt like fonttype & white colour. then for a shorter time it worked fine with multiple (.ass) subtitle inside one .mkv file, and now theres no difference which of the multiple subs I choose inside that .mkv, only the first default one is displayed all the time.

clsid
28th June 2008, 16:14
Subtitle related stuff hasn't been changed recently. So any inconsistent behavior you are getting is probably splitter/file related. But anyhow, the subtitle functionality of ffdshow is still incomplete and there is nobody available to fix it. Haruhiko seems to be busy with other things.

rack04
28th June 2008, 17:25
Is anyone else experiencing broken AAC playback with ffdshow_rev2026_20080625_clsid.exe? Even if I set ffdshow audio decoder to prefer in MPC HC if won't attach to the filter.

Edit. I uninstalled ArcSoft and all is well. What was weird is that it even effected the DXVA operation of MPC HC.