Slashdot Mirror
Nancy Goes Head-to-Head With MPEG-4
Justin Rossi writes: "EE Times has an article about Nancy, 'the lightest video codec' which is taking Asia by storm and finally bringing streaming Video to handheld devices. What I wonder is how it shall fare against MPEG-4, Ogg Tarkin, and MC-10."
"Nancy"? Was it named after some coders girlfriend or something?
:P
From a CPU (and therefore an electrical) standpoint the algorithm is better because it uses much simpler mathematics. But I wonder what the video quality would look like. Is it comparable to Mpeg4 based codecs like DivX? This is great for handheld devices, but I doubt it'll make much of a dent on the desktop unless the image quality is a lot better. We already have way more CPU power then we know what to do with
autopr0n is like, down and stuff.
Two formats I can't play in my favorite player (which happens to be just the mediaplayer, but it's the same thing if you are using other players). Will this be the same thing all over again? I don't mind new formats, especially if they are good, but if I can't watch them where I want to, who cares? If the big companies has to buy licenses to get them in their devices, and then force all publishers to use their special software... you know the drill.
I don't care if the software is closed source as long as protocols, codecs, formats, etc are open so anyone can implement and use them.
Well, we never had the bandwidth for real videophones.. they were all choppy as fuck when they came out, but now that we do have the bandwidth people are doing video conferencing with webcams and such all the time. It just isn't exciting anymore.
A cell phone with a cam and enough bandwidth (read 3g networks) might actually be popular since you'll actually be able to get a decent video feed.
autopr0n is like, down and stuff.
Because they can. When everyone says that videophones have already flopped they're assuming the concept just doesn't work or isn't as cool as well all thought it would be.
What you're missing is that once something becomes cheap and convenient, it doesn't need to be supercool. This sounds like one more step towards mass-market feasibility.
Sharp was one of the early adopters of MPEG-4, introducing an MPEG-4 video recorder and a Zaurus with an MPEG-4 player in December 2000.
Interesting, yes, but used where? The article does not say.
They also talk about "block noise" which you can see in DivX quite readily if you have a large piece of video recorded at too low a bitrate.
It is like watching a movie with a 1/4inch chicken wire overlay.
One of the problems with DivX that I have noticed is that it does not handle low light secenes very well...and it seems there are algorithms that compensate, because now some encoders complain about bright/outdoor scenes "going white"...heh.
oh, and this caught my eye...
The company has demonstrated video transmission to a notebook PC at 512 kbits/second, to a PDA at 256 kbits/s and to a cell phone at 28.8 to 32 kbits/s.
...and to charter pipeline (aka charter "sipping straw") at (drum roll please) a max of 12Kbytes a second... Road kill on the information highway.
People are going to ask which Mpeg4 codec is best, and, well that is an issue we will have to treat "Ginger"ly...hehehee
Have you read the moderator guidelines? Well, have you, PUNK? (and I want a Karma: Gnarly option)
I don't seem to find anywhere how well this "nancy" compares in the compression rate arena. How much does it compress with the same amount of lossiness? This is very important for this, because if yu don't care about the rate I could simply use gzip to compress my movies and have no loss at all.
[]'s Victor Bogado da Silva Lins
^[:wq
Debian actually sounds cool. I'd bet anything that if the distro were just called "Deborah" it wouldn't have much marketshare. Names really do affect people. Why do you think no one uses LISP even though it kicks ass?
Actually what I think happened is that the people picked a cool sounding 'foreign' name, like if it had been developed here they might have called it "Ritsko", or "Miho", or "Daikatana", or something, which might sound cool to American ears but retarded to Japanese (at least for a video codec)
autopr0n is like, down and stuff.
It's a low power (power=not much cpu required) designed for mobile devices.
The codec will run "even if CPU power is not high," said Kato. "A 50-Mips CPU can compress and decompress video at 30 frames per second with QCIF [176 x 144-pixel] resolution [using Nancy].
QCIF is a postage stamp, don't get excited... my freakin webcam can do that type of compression right now, this acheives a smaller size I'm guessing. As far as quality is concerned, I don't think thats the main focus.
Their goal is real-time, and low power cpu, and perhaps low bitrate... not highest quality, lowest overall size (MPEG4/DivX, etc)..
I've been waiting for the 1.0 release of Ogg Vorbis for a few years now. Yes, it's a nice CODEC, but the development timeline has been less than ideal for commercial adoption. Ogg Tarkin is still in extremely early development, without even alpha code to show for the effort. While most new audio CODECs have just been proprietary hacks of standard stuff to avoid patent royalties or optimize for streaming, video CODECs are making advances by leaps and bounds. MPEG-4 has the best compression ratio out there, though that may be at the cost of quality. I think that for handhelds and such things, processor requirements may be just as important as compression ratios, and those formats that keep this in mind will flourish.
WARNING: there is a trojan on your
That's a rather glib response, and incorrect. Additions, subtractions, are fairly simple operationsm and bitshifts are blazingly fast (and equivalent to dividing or multiplying by factors of 2) - in contrast, multiplications, divisions, and others are substantially more complex. You can improve performance a LOT if you design your codecs with these guidelines in mind. Check out the research section (fast DCT approximations) of this site - Nancy isn't the only codec to keep this matter in mind.
What I'd really like to know is - how well does nancy scale to higher resolutions? It could be competition for MPEG-4 even in the desktop arena. As someone who uses a 3-year-old laptop that can't really handle the &#($ing huge DivX files (which use pretty outdated technology across the board, whether you realize it or not), I welcome a codec that doesn't stress my system, and will save my battery life to boot.
---If you can't trust a nerd, who can you trust?