Path: utzoo!utgpu!news-server.csri.toronto.edu!bonnie.concordia.ca!thunder.mcrcim.mcgill.edu!snorkelwacker.mit.edu!apple!usc!zaphod.mps.ohio-state.edu!unix.cis.pitt.edu!dsinc!ub!uhura.cc.rochester.edu!rochester!pt.cs.cmu.edu!g.gp.cs.cmu.edu!tgl From: tgl@g.gp.cs.cmu.edu (Tom Lane) Newsgroups: comp.graphics Subject: Re: JPEG algorithm results Summary: It doesn't have to be *that* fast Message-ID: <11547@pt.cs.cmu.edu> Date: 11 Jan 91 03:06:54 GMT References: <1991Jan4.025552.810@allgfx.agi.oz> <1051@gistdev.gist.com> Organization: Carnegie-Mellon University, CS/RI Lines: 60 In article <1051@gistdev.gist.com>, flint@gistdev.gist.com (Flint Pellett) writes: > cn@allgfx.agi.oz (Con Neri) writes: > >The JPEG (Joint Photographic Experts Group) draft standard implements a > >DCT based image compression algorithm. ... Typical compression ratios are > >around 30:1 for 24 bit RGB images with reasonably high resolution. > > This doesn't say how much time is required to do the uncompression. > (I could care less if it takes 20 minutes to compress it, as long as it > can be uncompressed rapidly.) Before JPEG can displace something like > GIF, it's going to have to be able to uncompress images in something > less than 5 seconds on a typical PC, (and obviously, if you want it to > handle animation, you're going to have to get down in the fractional > second range.) Gimme a break. Even GIF images usually take longer than 5 seconds to decompress, unless you are talking real small images (or maybe you have a 486?) Software implementations of JPEG *will* be a good bit slower than GIF. However, this is not a good reason to reject it for net posting purposes. Remember that JPEG-compressed files will also be a factor of 5 (more or less) smaller than comparable GIF files. In many cases this translates directly into time savings; for example, I have to download any net-posted image across a 2400bps modem link before I can view it. Assuming a GIF image size of 100Kb (which is on the small side), that takes at least 7 minutes. The same image in JPEG format might be about 20Kb = 1.5 minutes. So if it takes less than 5 minutes to decompress the JPEG image, I have come out ahead on time, and that doesn't even consider the disk space and network bandwidth savings that are accrued all over the net. If I'm going to view the same image over and over, I probably wouldn't want to pay the price of JPEG decompression each time. No problem: convert it to GIF or something else that's cheap to display while I'm working on it. It's still worthwhile to use JPEG format for transmission and archival storage purposes. > I've not seen any concrete data on decompression speed > other than one article that quoted times around 2 minutes (yeech!) > by some package I don't remember the name of. Can anyone who knows > tell us what speed is available right now, both from software and from > hardware solutions? Thanks. The prototype JPEG code that my group is playing around with now takes maybe 3 or 4 minutes to process a typical image (300000 pixels or so) on medium-fast PC class hardware. We haven't made any serious effort to optimize for speed yet, so I think the final result will be significantly better; maybe 1 or 2 minutes. (Incidentally, decompression is about the same speed as compression, and there's no way to make it significantly faster 'cause the algorithm is symmetrical.) Hardware-based solutions will be lots faster, but still not video speed. JPEG is not intended for video applications; there is a separate standards effort (called MPEG) addressing the question of video compression. -- tom lane Internet: tgl@cs.cmu.edu UUCP: !cs.cmu.edu!tgl BITNET: tgl%cs.cmu.edu@cmuccvma CompuServe: >internet:tgl@cs.cmu.edu