View Full Version : Forget about AVC: Lossless Z-image :)
ia-ia
6th February 2005, 07:39
May be slightly off-topic, but
here is one of the most ridiculous announcements of video compression technology I've seen in years:
http://qbit.com/products.html
For example:
"Using important concepts from Theoretical Physics, Complexity Sciences, and Quantum Mechanics, such as the dual particle-wave nature of matter, and the quantized energy levels of states, Qbit is developing an all-purpose technology for information densification of any file type."
And from press-release:
"Today, Z-Image™ is already achieving a 3 to 5x lossless encoding improvement using interframes, and 10x lossless using intraframes, and we’re making it faster," said Dan Kilbank, founder and CEO of Qbit."
Entire site is very funny (check out "careers" section and read about "critical/desired" skills).
Somehow they managed to get John Sculley (former CEO of Apple and Pepsi) as a co-founder.
Anyway, read press-releases and news, have fun.
hpn
6th February 2005, 15:07
I thought it was some some kind of travesty site, but after reading a bunch of news sites it now seems quite a genuine announcement to me.
Check these:
http://www.telephonyonline.com/ar/telecom_new_compression_transform/
http://www.in-stat.com/promos/05/NA_qbit.htm
http://www.5starsnews.com/digital-video/130597.htm
Now waiting for codec implementations;)
Mug Funky
7th February 2005, 07:36
okay... they trademarked the letter Q.
morsa
7th February 2005, 08:06
Isn't that too much ? :rolleyes:
Now how am I supposed to spell Q-U-E-S-T-I-O-N without infringing copyright law or some patent.I also guess Keyboards will need to start paying royalties to them!!!:eek:
Dali Lama
7th February 2005, 18:38
"The potential impact on [telephone companies] is enormous. If you were to take an HDTV MPEG file, which runs at about 20 Mbps-- if you could reduce that to 2 Mb, now all of a sudden a standard 6 Meg ADSL service can deliver multiple HDTV signals into the home."
This is the funniest quote from one of the articles hpn linked. If someone doesn't get it, just ask :D
-Dali
Leak
7th February 2005, 19:40
Originally posted by Dali Lama
"The potential impact on [telephone companies] is enormous. If you were to take an HDTV MPEG file, which runs at about 20 Mbps-- if you could reduce that to 2 Mb, now all of a sudden a standard 6 Meg ADSL service can deliver multiple HDTV signals into the home."
This is the funniest quote from one of the articles hpn linked. If someone doesn't get it, just ask :D
What, reducing 20 megabits per second to 2 megabits flat?
Yeah, that sure is an achievement... :D
On the other hand, I can do this with 0.1s of video instantly, so what's new? ;)
np: Shnare.sys - Nice Shoes Bitch (Intelligent Toys 1)
CQ
7th February 2005, 22:04
Did you heard about that codec based on MPEG-1 that does dvd quality at 56k and hdtv at 128k? Very good for streaming over dial-up and shitty cable connection. And there are some samples over the net, i think on www.crapsandfullofshitsarticles.com :D :D :D :D :D :D :D :D :D
There are, by now, 2 that wants to be revolutionary on this> this Q thing and the MPEG2@1500k dvd quality MPEG4@500k (or 700?) and AVC@250k and the are enoght days until april to see on the 1st, a list of 20 codecs like MPEG-72284 (brought from the future :D ) with 0,0297k lossless HD and the ADN (dezoxiribo... acid :D) & ARN based video codecs :D :D :D But i think i`d better let asimov to predict the future of video compression.
Short version of upper text: BULLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLSHIT!
DeathTheSheep
7th February 2005, 23:01
And I almost got excited over the SDCS.
Sloot Digital Coding System was something at least REAL and had a ratio of 3,000,000 to 1. Yeah, three friggin million to one. Concept: Rework binary system, keep video NPS structure in 370MB file, and every other conceivable video of nearly any length sucks up only 1024kb each. Strange, huh?
So why is this so unbeleivable? It's a heck of a lot more beleivable than that Sloot crap, which definately sparked a lot of conflicting viewpoints.
Speculation? Partly. Most lossy compression just slams out detail and compresses using either blocks or the standard (QRS=T, FTRTETOYITOTTLFOTOERTT) compression scheme.
This new shitaki utilizes equations to de/compress data losslessly. It is first analyzed, a blueprint is formed, a pattern is calculated, and the data is conformed to that pattern using intense physics-based formulas to compress it (giving you 10X), and even adjusting the original pattern as it goes (therefore the more robust 20-30X).
Quite an acheivement over the privious compression schemes, which only replaced sequences of common data with markers, or rearranged data to fit into a pre-set pattern--especially over the "lossy" solutions that simply discard data the viewer doesn't notice (supposedly).
Speculation? Well,
99% Speculation
1% pure, unrivaled, undisputed fact.
This is equation compression. Very hard to program, harder to implement, but not quite as hard as you might think.
That's life, that's the new compression. Speculation? Absolutely. Reality? Absolutely.
Cheers,
Sirber
8th February 2005, 02:34
Just saying BS is shorter to write :p
CQ
8th February 2005, 06:36
yeah, but its not that expressive :D
Mug Funky
8th February 2005, 15:33
big problem with "equation compression"... it's bollocks.
consider that for most real world mathematical problems no analytical solution exists. hell, we can't even track the motion of more than a couple of stars at once before we need to fall back numerical solutions (think 300,000,000 calculations instead of 1 simple 4 term equation with substituted values).
determinism is one thing, but practicality is another - what's the use of using numerical methods to decompress a movie when it takes 2 days to decompress a frame?
true there's a VERY small subset of situations out there that can be expressed analytically and neatly, but certainly not enough that you can throw a video at some magical Equationator (TM) and get out a tiny file.
though if quantum computing is cracked any time soon, all hell will break loose. imagine brute-force cracking AES encryption in less than a second? video compression might increase in efficiency by several orders of magnitude as well.
only problem with quantum computing is that you can't fit it in your head without needing a very stiff drink.
Sir_Lagsalot
10th February 2005, 07:25
From one of their news stories:
"This radical new approach to compression will first be used to more efficiently move two-dimensional images, such as video, medical images, satellite photos, etc. by reducing files sizes by a factor of 10. Later this year, Qbit will introduce a more robust form of its technology that will reduce any data file by a factor of 32."
This is great! I'll be able to compress my zip/rar/ace files by a factor of 32!!! Then, to save even more space, I'll compress the resulting files by a factor of 32! I'll keep doing that till all my files are losslessly compressed to 1 bit.
iapir
10th February 2005, 13:09
I read once that the "bus speed" of each of our ear is about 10 kbps. That makes 20 kbps for what we can hear. So I assume that a perfect compression of sound could go as low as this ? (maybe a bit more if different people hear different things from the same source)
Maybe it's the same for the eyes.
CQ
10th February 2005, 13:14
@Sir_lagsalot - Why don't you use rar to compress, in small steps, the file to 1 bit? :D :D :D :D :D :D :D :D :D :D
Man,see the compression FAQ (faqs.org i guess).
Sir_Lagsalot
11th February 2005, 01:45
Originally posted by CQ
@Sir_lagsalot - Why don't you use rar to compress, in small steps, the file to 1 bit? :D :D :D :D :D :D :D :D :D :D
Man,see the compression FAQ (faqs.org i guess).
Unfortunally, Rar is unable to decompress my 1bit compressed copy of the FAQ, I'll have to wait for Qbit to finish their "like magic" quantum codec to read it I quess...
E-Male
11th February 2005, 02:36
Originally posted by iapir
I read once that the "bus speed" of each of our ear is about 10 kbps. That makes 20 kbps for what we can hear. So I assume that a perfect compression of sound could go as low as this ? (maybe a bit more if different people hear different things from the same source)
Maybe it's the same for the eyes.
it's not that easy
our eyes have a limited resolution
but we have a brain, the best interpolating device known to mankind
ever wondered how much of what you see and hear really comes from your eyes and ears? and how much your brain adds from experience
to top that we have the ability of abstract thinking, which to my knowledge can't be electronically duplicated (at leats for now)
so talkign about bandwith of ears or eyes is very much simplifying
also: how did they measure that numbers?
CQ
11th February 2005, 06:39
Sir_Lagsalot... I was joking :D .
Sir_Lagsalot
11th February 2005, 07:24
Originally posted by CQ
Sir_Lagsalot... I was joking :D .
That makes 2 of us.
vBulletin® v3.8.11, Copyright ©2000-2026, vBulletin Solutions Inc.