On Wed, Sep 17, 2008 at 8:57 PM, mhampton <[EMAIL PROTECTED]> wrote: > > I thought I'd mention that using ffmpeg and Sage/singular/gfan/tachyon > I was able to finish the little pilot project I had in mind: taking a > 5d Groebner fan, intersecting it with a hyperplane, rotating it in 4d, > and animating the projection into three dimensions. I am still > struggling with the options to ffmpeg; I eventually just used "-qmax > 2" and nothing else but I am sure that is far from optimal (I did try > Vincent's options and it didn't work on OS X, I think I am missing the > codex). I think the results are enjoyable: > > http://www.d.umn.edu/~mhampton/gf5.mp4
Cool! > > Cheers, > Marshall Hampton > > On Sep 3, 2:52 am, Vincent Beffara <[EMAIL PROTECTED]> wrote: >> Hi, >> >> > Thanks very much for this response. ffmpeg looks very useful to me, I >> > am checking it out right now. It is unclear to me what the overlap is >> > with mplayer/mencoder. It seems that ffmpeg is somewhat leaner and >> > more portable, so I am thinking of using it as the encoder for your >> > first suggestion (create a sequence of jpegs >> >> Rather use a non-destructive compression like tiff or tga, the artifacts >> of jpg might become too visible after encoding especially if they vary >> from frame to frame. Disk is cheap these days. >> >> > with the tachyon >> > raytracer, then convert to mp4). ffmpeg seems quite fast to me >> > compared to using imagemagick with GIFs. >> >> It should be slower than animated GIFs, but I agree that it is quite >> efficient. And quite portable, I even managed to compile it on Solaris >> after some tweaking. There are two things to be careful about, and for >> which it is quite better than mencoder in my experience: >> >> - Colorspace encoding : for some reason mencoder wants YCbCr, and >> standard packages provide RGB ... Besides, ffmpeg doesn't like 16 bits >> per channel, and figuring that one out was tricky (the error message >> is a bit cryptic and sounds like "syntax error"); >> >> - Quality settings for the h264 codec. That is actually the big one. The >> main parameter is _not_ the bitrate ! Well, increasing the bitrate >> helps, but by default the "hi quality" mode is off and you need some >> tweaking. At the time I was experimenting to put bits of Dimensions >> online, and it turned out that the following parameters gave a good >> result: >> >> -vcodec libx264 -b $(VBR) -level 30 -mbd 2 -cmp 2 -subcmp 2 -g 300 >> -qmin 1 -qmax 31 -flags +loop -chroma 1 -partitions partp8x8+partb8x8 >> -flags2 +mixed_refs -me_method 8 -subq 7 -trellis 2 -refs 1 -coder 1 >> -me umh -me_range 16 -bf 0 -sc_threshold 40 -keyint_min 25 >> >> (Yes, that's all a long string of command-line arguments.) For 384x288 >> resolution, a VBR of 200 kbps is enough. And I believe you will not >> find it so fast :-) >> >> One last thing: Apple's implementation of h264 is not quite as complete >> as the one of x264 (which I find quite ironic), so if you are not >> careful the file you produce will not be readable on a mac ... >> >> I am attaching a Makefile that I used to encode video, in case you find >> it useful. Be sure to check out mp4creator after ffmpeg. >> >> Have fun, >> >> /vincent >> >> -- >> Vincent Beffara >> UMPA - ENS-Lyon >> 46 allée d'Italie >> 69364 Lyon Cedex 07 >> Tél : 04 72 72 85 25 >> >> Makefile >> 1KViewDownload > > > --~--~---------~--~----~------------~-------~--~----~ To post to this group, send email to sage-devel@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/sage-devel URLs: http://www.sagemath.org -~----------~----~----~----~------~----~------~--~---
