Example of playing video on Kindle Touch

[knc1]

Originally Posted by geekmaster

I just downloaded the file from the first post again. Viewing the script in a hex editor shows that it still contains 0x0D characters. There are more downloads in the count now, so beware that you need to strip CR chars from it or run it using the command line shown in my previous post.

To make it right before posting:
http://waterlan.home.xs4all.nl/dos2unix.html#DOS2UNIX

[Ehhh]

hmmm.... i re-uploaded 3 hours ago by command dos2unix. and this scripts works fine on my KT.

Originally Posted by knc1

To make it right before posting:
http://waterlan.home.xs4all.nl/dos2unix.html#DOS2UNIX

[geekmaster]

Originally Posted by Ehhh

hmmm.... i re-uploaded 3 hours ago by command dos2unix. and this scripts works fine on my KT.

Great. You should link that one to the first post, where you still have one that contains carriage-returns. I just downloaded it and checked again (5 times now) and it is still wrong. Where did you upload the good one? Can you replace the link in the first post with one for the good file?

Like I said, you need to download it and test the DOWNLOADED version. Are you sure that is the one you said "works fine", or are you testing a DIFFERENT one?

EDIT: Your lack of cooperation and your continuing failure to test your own uploads is giving me a "raised eyebrow" here. Please be more careful. This makes me suspicious of running your binary code when you have refused to publish your source until "it is ready" (whatever that means). Please do not continue to disappoint me. I bumped your karma to 2600 as a sign of respect when you created this thread. Now you need to earn that respect.

[Ehhh]

how to return this karma to you again? i don't need this

Originally Posted by geekmaster

Great. You should link that one to the first post, where you still have one that contains carriage-returns. I just downloaded it and checked again (5 times now) and it is still wrong. Where did you upload the good one? Can you replace the link in the first post with one for the good file?

Like I said, you need to download it and test the DOWNLOADED version. Are you sure that is the one you said "works fine", or are you testing a DIFFERENT one?

EDIT: Your lack of cooperation and your continuing failure to test your own uploads is giving me a "raised eyebrow" here. Please be more careful. This makes me suspicious of running your binary code when you have refused to publish your source until "it is ready" (whatever that means). Please do not continue to disappoint me. I bumped your karma to 2600 as a sign of respect when you created this thread. Now you need to earn that respect.

[geekmaster]

I was only trying to help. The version linked in your first post still has carriage-returns, as you can see in a hex editor. Apparently you do not want my help. Good luck with that...

[knc1]

Originally Posted by Ehhh

hmmm.... i re-uploaded 3 hours ago by command dos2unix. and this scripts works fine on my KT.

Please stop top posting.

[geekmaster]

Originally Posted by knc1

Please stop top posting.

Here is the official ruling on this matter:

RFC1855: http://tools.ietf.org/html/rfc1855

---<snip>---
If you are sending a reply to a message or a posting be sure you
summarize the original at the top of the message, or include just
enough text of the original to give a context. This will make
sure readers understand when they start to read your response.
Since NetNews, especially, is proliferated by distributing the
postings from one host to another, it is possible to see a
response to a message before seeing the original. Giving context
helps everyone. But do not include the entire original!
---<snip>---

And this explains why top posting is so bad: http://www.caliburn.nl/topposting.html

---<snip>---
Top-posting makes posts incomprehensible. Firstly: In normal conversations, one does not answer to something that has not yet been said. So it is unclear to reply to the top, whilst the original message is at the bottom. Secondly: In western society a book is normally read from top to bottom. Top-posting forces one to stray from this convention: Reading some at the top, skipping to the bottom to read the question, and going back to the top to continue. This annoyance increases even more than linear with the number of top-posts in the message. If someone replies to a thread and you forgot what the thread was all about, or that thread was incomplete for some reasons, it will be quite tiresome to rapidly understand what the thread was all about, due to bad posting and irrelevant text which has not been removed.
---<snip>---
We can conclude that there are no good reasons we know of for top-posting. The most top-posts originate from the minimal work people spend on making posts. We think that one should be proud of one's post, that is it contains relevant content, well-formed sentences and no irrelevant 'bullsh*t', before uploading to your newsserver. If the majority of the group will adhere to this convention, the group will be nicer, tidier and easier to read.
---<snip>---

[knc1]

The script file: run_player.sh contained in the KTplayer.tar.gz located in the thread top post - when opened in Geany, Geany claims the line endings are still Windows and offers me the chance to convert them to *nix.

So that is at least two different systems that report the same (mine and GM's).

[NiLuJe]

FWIW, I only see shiny 0x0A EOL (tarball downloaded just now), and the only thing my vimrc is killing is two trailing whitespace .

[geekmaster]

Originally Posted by NiLuJe

FWIW, I only see shiny 0x0A EOL (tarball downloaded just now), and the only thing my vimrc is killing is two trailing whitespace .

That is strange. I will try again....

1) wget from URL in first post.
2) scp to K5.
3) tar zxvf.
4) hexdump. *** only 0x0a line endings. That is good.

Now to see what was wrong before:

1) download with chrome on XP.
2) extract with winrar
3) winhex *** CONTAINS CRs !!! (did winrar insert CRs) ???

Try different tools:

1) download with chrome
2) extract with 7-zip
3) winhex *** CRs again! (both winrar and 7-zip insert CRs)???

Or did google chrome insert CRs *INSIDE* the tarball? ??? Let's try chrome to download, then extract it in the kindle:

1) download with chrome
2) scp to kindle
3) tar zxvf
4) hexdump *** only LFs (So, even 7-zip inserts CRs without asking?) Bad!

OK. I need to appologize to Ehhh. I thought he was disregarding my messages about his upload having CRs, when the problem is that winrar and 7-zip insert CRs without warning and without permission. That sucks! Sorry, Ehhh. My tools bit me.

[geekmaster]

The lesson here is to be careful which tools you use to extract a tarball (.tar.gz) file. We cannot trust winrar and we cannot trust 7-zip (at least on the windows versions). They insert windows carriage-return characters into script files when extracting them on a windows host PC, which makes them not run correctly when used on a linux machine like a kindle.

I have extracted linux scripts with winrar and 7-zip in the past (16.3 bazillion times, as I recall) and I do not remember ever seeing this problem before. This is really confusing now, and apparently knc1 has tools that misbehave like this too. It sucks when you cannot trust your tools.

[geekmaster]

Originally Posted by Ehhh

Hi, geekmaster

you can check and compile eink transform plugin with yours algorithms.

best regards, ehhh

Somehow I missed this post. It was the last one on the page and there were posts on the next page. Sorry. Thanks.

The 16x16 matrix in your code should not be needed because the human eye cannot distinguish between more than 64 shades of gray. What is important is the cutoff values used in the 8x8 matrix, which control the display gamma, brightness, and contrast. And because eink behaves with linear gamma like real ink on paper, we do not need gama adjustments, which makes this a lot easier than when dithering for a light-emitting display.

[geekmaster]

It was not that long ago when desktop PCs could play DivX/XviD 480p movies just fine, but you needed a powerful PC to do High Def (720p), especially using the x264 codec (which needs a powerful PC to decode it).

I have been playing various downloaded videos on my K5 and K3. The poor K3 cannot decode x264 fast enough to play, so it needs to be converted to a .gmv.gz file and played later. But on the K5, x264 480p is great, and 720p even plays full speed on some scenes, but slows down when the scenery gets complex. That gives periods of smooth motion alternating with periods a of some dropped frames, and in some cases mostly dropped frames with jerky motion. So, the K5 can ALMOST downconvert 720p in realtime. What this means is that because x264 compresses MUCH better than the old DivX/XviD codecs, we can play pre-scaled 800x600 video in x264 (for maximum compression) at full speed on the K5. The K3 on the other hand probably needs to stay with XviD compression that is much easier to decode.

Anyway, I did not post my "all the options" ffmpeg yet because I was not successful adding alsa sound output to it for kindle sound. That is not a problem for playing video on the K4 though.

I am looking at combining by gmv2vid transcoder and my gmplay player into a plugin for Ehhh's player, now that he provided the source code for his eink plugin. It will be interesting to test them and see whether it is faster to transcode and play raw video output from ffmpeg, or to thanscode and play it from the plugin in Ehhh's player (which uses a portion of the same ffmpeg code that the full player uses). Ffmpeg can play streaming video from the net as well as files, so perhaps we will end up will the best of both when we combine them.

Now that I am doing dithering with no dither table, I am investigating doing eink output with no framebuffer. That looks like it should be faster (or at least more efficient / less battery consumption). Dithering with no dither table, then eink with no framebuffer -- what else can we eliminate to speed this up?

[Ehhh]

The uploaded gstreamer's plugins are the wrappers on ffmeg library )). i have compiled all list of plugins, and it could try to play h264, h262, mjpeg videos.

Originally Posted by geekmaster

It was not that long ago when desktop PCs could play DivX/XviD 480p movies just fine, but you needed a powerful PC to do High Def (720p), especially using the x264 codec (which needs a powerful PC to decode it).

I have been playing various downloaded videos on my K5 and K3. The poor K3 cannot decode x264 fast enough to play, so it needs to be converted to a .gmv.gz file and played later. But on the K5, x264 480p is great, and 720p even plays full speed on some scenes, but slows down when the scenery gets complex. That gives periods of smooth motion alternating with periods a of some dropped frames, and in some cases mostly dropped frames with jerky motion. So, the K5 can ALMOST downconvert 720p in realtime. What this means is that because x264 compresses MUCH better than the old DivX/XviD codecs, we can play pre-scaled 800x600 video in x264 (for maximum compression) at full speed on the K5. The K3 on the other hand probably needs to stay with XviD compression that is much easier to decode.

Anyway, I did not post my "all the options" ffmpeg yet because I was not successful adding alsa sound output to it for kindle sound. That is not a problem for playing video on the K4 though.

I am looking at combining by gmv2vid transcoder and my gmplay player into a plugin for Ehhh's player, now that he provided the source code for his eink plugin. It will be interesting to test them and see whether it is faster to transcode and play raw video output from ffmpeg, or to thanscode and play it from the plugin in Ehhh's player (which uses a portion of the same ffmpeg code that the full player uses). Ffmpeg can play streaming video from the net as well as files, so perhaps we will end up will the best of both when we combine them.

Now that I am doing dithering with no dither table, I am investigating doing eink output with no framebuffer. That looks like it should be faster (or at least more efficient / less battery consumption). Dithering with no dither table, then eink with no framebuffer -- what else can we eliminate to speed this up?

[ixtab]

@Ehhh:

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?