View Full Version : About DivX releasing rules (TDX2k2)
CaPPyD
20th July 2002, 17:21
Originally posted by neo_sapien
Of course this has occurred to me. Note that I'm advocating Ogg as a storage container. I don't mean for it to be permanent. It would take something like 5 minutes to demux the component streams and remux them to an MP4 container. Heck, I could do it right now with Graphedit, Virtualdub and mp4tools. The only thing preventing me from making rips directly into MP4 now is the lack of b-frame ISO decoder support. If I didn't mind not having any advanced simple profile options in all my rips, I'd have found it prudent already to switch to making purely MP4 rips.
The thing is, though, in several months when MPEG-4 set tops are released, the Ogg containers will have kept these movies intact and ready for conversion to an MP4 container. With the AVI container, the lack of error detection means that X amount of movies will have gone bad by conversion time.
Now one of these containers will ensure future compliance through conversion to a bonafide MPEG-4 container. The other container will ensure that we have 700 (795) megs of errors.
Man... you are whacked out on something fierce...
I have absolutely NO idea what you are talking about when you say your AVI's go bad....
I've got movies here that I downloaded and burned almost three years ago and I still watch them all the time...
Again, I want some of the drugs yer taking, cuz they seem to be some pretty strong hallucinogens....
:D
neo_sapien
20th July 2002, 17:51
LOL. Don't I wish. I have a copy of the 2cd DivX rip of the LOTR DVD Screener right here with me in an AVI container. About 190 megs of it is totally gone and unrecoverable, and there's maybe 510 megs of useful data. I tried to remux it into something that'll play in vdub, no use. So now I have an unplayable AVI, all because of a lack of error detection. This isn't the only CD that's fallen victim to AVI's suckassitude. Gimme your address, I'll mail you a copy of LOTR on AVI. LOL :p
Christo
21st July 2002, 11:26
Hello everyone,
First of all let's make it clear - no MPEG4/DVD players ona horizon...
High licensing fees had killed MPEG4 hardware support for the near future
Hollywood and other Movie making bonzes opted for MPEG2 Complex profile
It gives 2 times better compression than MPEG2 Simple (current DVD)
This decision isn't final yet but almost certainly will be...
Now, Microsoft pushing their own standard - Corona (WM9)
No support for it from movie making bonzes either
BUT, 80% of DVD players makers, such as Pioneer, Panasonic, Apex, other already
pledged their support regardless of Hollywood decision
I know many of you will say that Windows Media isn't what everyone was waiting for
But this is what we gonna get. In Beta on September 4. In final release by the end
of this year. The players will roll after that. To make it universal standard MS promises
to freeze farther development. That means no WM10,11,etc - Corona (WM9) forever.
Even if the encoding code will be changed abit in a future the decoder always be
backward compatible. Those guys in Seattle really know how to set standards,
don't they ;)
Some may say Windows Media quality plain sucks. I tried WM8, no it doesn't. In fact
I considere it to be more efficient than any DIVX. Of course Corona (WM9) is going
to be even better (20%). Here is something to remember preencoding/premastering (filtering, resizing, etc) very often affects quality more than encoding itself. I have
done a test of converting a DVD/VOB 4 min clip to WM8 and got a file 2 times less in
size with quality equal or better than the original. Preencoding work was done on
Virtual Dub (deinterlacing, filtering, resizing) and encoded with WM8 maximum quality.
ME THINKS CORONA IS OUR FUTURE
bye, now
int 21h
21st July 2002, 11:35
Originally posted by Christo
I have
done a test of converting a DVD/VOB 4 min clip to WM8 and got a file 2 times less in
size with quality equal or better than the original.
This sentence alone reveals it all. If you think you can take input from one lossy format, and convert to another lossy format, and up with something that looks better.. well then.. I have a bridge in Singapore for sale... Seriously though, you're a member of that SmR group right? :p
Hell, while you're at it, you may as well reduce the framerate to something like 20 fps, it will probably look better, less movement of the picture and all.
Christo
21st July 2002, 11:52
Hey, I was serios
Less noise because of temp smooth 2,
Sharper pictur because of prescise bicubic 0.75 and IVTC
NeVeRLiFt
21st July 2002, 12:03
Me and some friends already have Corona I mean WMP9 and where trying to play around with the codec. I'm gonna test it for Videocapture.
smok3
21st July 2002, 12:48
Originally posted by int 21h
If you think you can take input from one lossy format, and convert to another lossy format, and up with something that looks better.. well then.. I have a bridge in Singapore for sale... if u take ur mp3's and bridge them over a good equalizer then they can sound better (thats subjective), altho some additional (mesurable) noise and distortions will be added, thats just an example.
overall subjective quality generally is not measurable
thats from:
http://www.xiph.org/ogg/vorbis/doc/vorbis-fidelity.html
you may as well reduce the framerate to something like 20 fps, it will probably look better, less movement of the picture and all never tryed that :p , but i did deinterlaced a lot of video stuff just to get that 25fps film like look, it looks cool to me :)
p.s. and i removed that bold tags in quotes, seems like the post looks kinda better without them :devil:
theReal
21st July 2002, 20:33
With the AVI container, the lack of error detection means that X amount of movies will have gone bad by conversion time
I have a good solution for this problem: Mitsui CDRs !! :D
On an SG Ultra CDR, I'm absolutely sure my avi files will still play in ten years (correct storage assumed) without the need for any additional error correction.
Most no-name, and even most brand-name CDRs I would not trust for more than half a year, that's right...
neo_sapien
22nd July 2002, 05:16
I'm using Ogg and Memorex black just to be on the safe side.
theReal
22nd July 2002, 10:11
According to info I read about CDR manufacturers there's only a bunch of actual manufacturers who produce for all the different brands. Memorex, according to this info gets all (or some? or most?) of their CDRs from Ritek Co. - and that is not exactly on the safe side...
I don't trust anything else than Kodak and Mitsui anymore...
neo_sapien
22nd July 2002, 21:12
From what I read, memorex black is japanese, made in a Grade-A fab plant. I asked around, read a few forums, and I kept getting the recommendation of memorex black.
These CD-Rs aren't Ritek, they're from a plant called taijo hinjen (sp?) or something. black is good, because all black cd-rs come from this plant.
SirDavidGuy
22nd July 2002, 22:56
Originally posted by neo_sapien
they're from a plant called taijo hinjen (sp?) or something.
Taiyo Yuden?
Very high quality CD's (Very little C1/C2 errors on reading back), but be wary of burning them at speeds higher than they're rated for.
I know of at least 4 people who have had them explode (One of them being me.)
Mentar
22nd July 2002, 23:22
Wow. doom9 is definitely the best place for information about encoding, and the quality of the boards and most threads are impressive. However, I've got to say that - politely spoken - I have rarely seen that many articles containing opinionated and highly questionable statements on doom9 boards as there are in this thread.
Here's the list of items I disagree with most:
1) The statement that the DivX5 codec "doesn't produce decent 1-CD releases" is (sorry) nonsense. Most of my encodes are anime (generally considered to be on the more difficult end of encoding source material) and I've wasted 6 weeks with tinkering and testing with friends before we unanimously agreed that the quality of DivX5 (no qpels, no GMC, with b-frames for the best quality/size/CPU usage tradeoff) was superior. The theory claims a gain of around 25% in compression, and this is what our testing showed too. We redid many of our old encodes in 3.11SBC with DivX5 (and ogg, see later), and every single feedback we got was "wooo, what an incredible improvement". There may be other reasons to exclude DivX5, but quality is not a valid one.
2) The selection of MP3 strikes me as questionable, too. Vorbis achieves a better quality with a size saving of around 35% compared to MP3. AC3 is valid since it serves a different purpose (5.1 sound), but is also much larger in filesize (hence, a smaller bitrate for the video).
3) The subtitle part made me wonder if the authors really understand the topic at all. Vobsub is no "format" which can be muxed into a video. If you want to mux it, you need to do OCR (for examply subresync from the vobsub package) to for example .srt or .ssa or others, which then can be muxed, either into an .avi carrier or for example into .ogm aswell.
4) The container format discussion. When done correctly, .ogm can be made to run in every single operating system with no problem, you just need to install the right software. Except for the fact that right now Tobias seemed to have stopped development of it (a shame), it offers the most powerful and easy-to use container solution so far. A pity, really.
5) SVCDs "superior" to DivX encoding? I had to chuckle about this one - superior in which way? It forces a lower resolution, it compresses much worse and therefore wastes alot of space. The only "advantage" I can see is that it's easier to download and just burn on CDs to view in SVCD-enabled standalone players. But is that what counts? Diskjockeying during playback kills my fun for sure.
Respectfully, when it comes to both quality AND size-efficiency, I would make a strong case for DivX5 as video codec, Vorbis as audio codec and Ogg as the container. It's no problem to fit a high-quality movie encode with 2 audio tracks and multiple subtitle tracks on a single CD, simply because it uses the available techniques offering the highest quality on the least space, hence enabling higher bitrates in both video and audio with the obvious quality implications.
Just too bad that the ogg development seems to have stopped by now. I hope that the next carrier format to make the limelight will not fall back behind where ogg made it so far.
neo_sapien
23rd July 2002, 04:46
Yeah, that's the one. I burn them at 12x, and the box says "Write from 1x up to 16x Multi Speed".
Originally posted by SirDavidGuy
Taiyo Yuden?
Very high quality CD's (Very little C1/C2 errors on reading back), but be wary of burning them at speeds higher than they're rated for.
I know of at least 4 people who have had them explode (One of them being me.)
Tom Servo
31st August 2002, 14:35
I've been testing the various builds of the WMV9 beta codec.
Generally the picture is fine, at least as good as Divx5 (postprocessing disabled, or else it'd be unfair). While 2601 was getting terrible blocky on highspeed motion, prolly due alpha nature, 2773 and 2798 improved a lot, and 2848 shows almost no blocks at all anymore, only on large uniform surfaces or very highspeed motion (and that only if bitrate congestion happens, for both cases. Talking with MS, I found out that deblocking will go into the decoder). Also WME7.1 doesnt allow VBR even less 2pass, so I couldn't test that one.
I'm still awaiting about any specs how a WM9 DVD layout should look like. If I got those infos, I'll make WM9 MiniDVDs from then. :)
If the majority of the DVD standalone players will support WM9 in future, I think it's worthy to consider doing WM9 MiniDVDs.
And about TDX rules, dunno what to think about it, but this packaging rule is kinda stupid:
Must be packed with RAR, compression is NOT ALLOWED and broken into 15 or 20 MB volumes
Even if it results only into avg. 5% compression, better than nothing.
oddball
2nd September 2002, 02:55
Have you ever tried to decompress 700 megs of compressed rar files? It takes AGES compared to uncompressed. The 5% is usually actually more like 1% or less from what I've seen. The movie is already compressed pretty much as much as it can be. The gains aren't great enough for the length of time it takes to download those few extra megs on broadband and more of a pain for decompressing them. Uncompressed is how I like them archived thanks.
SirDavidGuy
2nd September 2002, 15:05
It's because releases are usually split into parts, since they're often released on IRC or newsgroups.
((( atom )))
2nd September 2002, 21:26
another (the main, i guess) reason is that warez-groups pirate public servers and don't want the admin to easily find these huge files. with these small files he just doen't have a chance to simply search for and delete everything about 100mb or so.
/jc
theReal
3rd September 2002, 00:00
you can upload small parts to a lot of different free (or inexpensive) web-accounts but you won't find a free account where you can upload 700MB.
Then, if there's only a 20MB part of a rar archive on one server, the admin doesn't have any good arguments to delete the file because he probably can't prove the content is illegal.
However I think downloading movies like this is a pain in the ass and I rather use the time to rent a DVD and make a good Divx backup myself. It's more fun :)
((( atom )))
3rd September 2002, 00:12
no no, people simply hack vulnerable ftp-servers (windows)with scripts, kids can use and start their own server-process. i've seen some doing that. rather funny, lol. good bandwidth they can achive..
but as you said: go and make it on your own is teh eral fun..
mikemikez2
3rd September 2002, 12:53
The most important reason for using rar archives is to protect the data in it. I know some groups don't turn on the rar option called "protect archive from damage" but they are just stupid.
Next to that when you got a small part of your upload/download that's messed up you can make things right by just re-upping/downloading/fxp'ing only one 15MB chunk.
(some use 20MB chunks now and DVD-R's are mostly 50MB chunks)
I know some parts of these so called "rules" are questionable but the major advantage is UNIFORMITY. When you are in the scene you will learn to apreciate this a lot and it's way more important and contructive to the scene than just using some tiny bit better codec or so...
Some might just bash this "scene" nomatter what but I think it's pretty amazing how it operates within an extremely distributed "architecture" (which is what keeps the scene alive) and especially how well and organized stuff is being done.
Little mistake in a rip? Same or next day you'll find a nice FIX or rerelease....
Next to that, "the scene" shows us that nowadays the distribution network media corporation have can very well be replaced by people working for free on the internet. (Even under current political pressure) This distribution network is what media companies say gives them the right to charge insane amounts of money for cheap copies that contain maybe a few cents of actual efforts made by the artists.
Don't let them shut down the internet! Don't let them get more controll over over media in general by legally closing down distribution at the level of encoding!
So much for my speech..:)
Mike
vBulletin® v3.8.11, Copyright ©2000-2026, vBulletin Solutions Inc.