MS Proposes JPEG Alternative

automatix writes "Microsoft's new competitor to the omnipresent JPEG format has been shown at WinHEC and is discussed on CNET. The Windows Media Photo format has many promises associated with it. The program manager is claiming 'We can do it in half the size of a JPEG file.'. While 'the philosophy has been that licensing should not be a restriction', it is interesting that the specification requires a click-through agreement to even read it."

Re:Ummmm why?

[blane.bramble]

Well, I clicked the "I do not agree" button, and it still takes you through to the details...

Lossless AND Lossy

[eldavojohn]

I think the most important reason is stated in the opening paragraph of the Wikipedia article:

It can support both lossy as well as lossless compression ...

If you read all of that article, they get to the specifics of how computationally expensive it can be and I must say it offers a high performance.

If you look at JPEG, the Wikipedia article states:

In computing, JPEG (pronounced jay-peg) is a commonly used standard method of lossy compression for photographic images.

And now you see that it only supports lossy. There are other lossless formats out there but I don't think there is another popular MIME file format that is widely used to support both lossy and lossless.

The big selling point here seems to be that you can have one file format to do all regardless of what kind of compression you prefer. I do agree that if it's not released open to the public, it will fail.

Re:Lossless AND Lossy

[harmonica]

If you look at JPEG, the Wikipedia article states:

        In computing, JPEG (pronounced jay-peg) is a commonly used standard method of lossy compression for photographic images.

And now you see that it only supports lossy. There are other lossless formats out there but I don't think there is another popular MIME file format that is widely used to support both lossy and lossless.

Actually, JPEG can do both lossy and lossless, no matter what the Wikipedia article might say (I haven't read it).

Re:JPEG 2000

[jacoplane]

JPEG 2000 [wikipedia.org]:

JPEG 2000 is not widely supported in present software due to the perceived danger of software patents on the mathematics of the compression method, this area of mathematics being heavily patented in general. JPEG 2000 is by itself not license-free, but the contributing companies and organizations agreed that licenses for its first part - the core coding system - can be obtained free of charge from all contributors.

So basically, it's free for the moment, but who knows if it'll stay that way.

It is TIFF hijacked

[Maljin Jolt]

Reading all of 31 pages of the document makes me understand that it is just an attempt to hijack tiff an bend it with MS patented pixel codec to become incompatible with existing tiff technology. Salted with Adobe XMP metadata, ICC metadata and EXIF metadata. All of that registered as a Microsoft trademark. Did I missed something?

Re:first reaction, second reaction

[harmonica]

My first reaction is:

GIF, JPG, and PNG do everything I need -- why a new image format?

There are patents on certain parts of JPEG. Including the ones everyone uses, where the claim is highly disputed, by Microsoft and others are already paying licensing fees.


My second reaction is:

Ok, I'm innovative, so maybe there is a good reason for a new image format. Maybe I'll read more. But then I re-read it's from Microsoft and it's got called Windows in it's name, and I think I've got enough MS and Win in my life -- I really don't want more.

If they came up with a great file format I see no reason why MS would be a problem. Microsoft also helped defining TIFF, so what.

Re:Another Debate

[The Cisco Kid]

Not arguing with you, just further debunking some of the MS spew you quoted.

"Government funding should be for work that is available to everybody."

Even this is MS classic twisting of words.

Open Source work *is* available to everybody. It is so available that no one (commercial company or not) is allowed to take it and lock it up into something that is NOT available to everybody.

"The way the license is written, if you use any open-source software, you have to make the rest of your software open source."

You can *USE* any (Free and/or "Open") software you want, and no requirements are imposed. However, you are not permitted to make that software *part* of *your* software unless *your* software is also 'available to everybody'

Contrast this with proprietary software, where you aren't permitted to use it unless you fork over dough to its 'owners', you arent allowed to inspect it or modify it, or heaven forbid make it part of your software, at all.

Re:Why?

[damiam]

MNG does all of those, IIRC.

JPEG-LS Vs JPEG

[eldavojohn]

I think you're confusing JPEG with JPEG-LS [jpeg.org]. Yes, they are both "JPEG"s but I don't think many applications natively support JPEG-LS. In fact, I wanted to use it in Photoshop and had to go get a plug-in [datacompression.info]. Whether or not JPEG-LS is as efficient as the new proposed MS format, I do not know. I think that JPEG-LS was slow to catch on because people just didn't care about upgrading their software to use it. I would wager that Microsoft will force third party software to support their new filetype.

I'm not even sure if my browser supports JPEG-LS and I know that programming with JPEG-LS can be a pain if you're looking for libraries to read/write lossless compression image files.

Re:first reaction, second reaction

[Glacial Wanderer]

The fact that you didn't mention raw, dng, or tiff tells me that you don't know enough for your opinion to matter much to me. Without the formats I just mentioned many of your pretty jpg would never have existed. I don't know enough about this new Microsoft format to say whether it is good or not, but I do know that there is a lot of room for improvement to jpg like support for HDR and an alpha channel.

My gut says it doesn't trust Microsoft to properly handle an open image format, but part of me also knows it might take a huge company like Microsoft to move everyone out of the jpg era.

Re:Is JPG open?

[Iphtashu Fitz]

However, JPG isn't an open standard, is it? Isn't it controlled by proprietary licenses as well?

JPEG [wikipedia.org] is a standard, created by the Joint Photographic Experts Group. The problem with it is that it's entirely lossy. It's great for final images but any time you edit an image you'll lose more and more of the detail.

PNG [wikipedia.org] is a newer open standard that was created in part to address the issue of loss. Prior to PNG many people used the GIF format, which is losless as well, but GIF has licensing/patent issues. Most mainstream web browsers, graphics editors, etc. now support PNG graphics natively.

Re:Ummmm why?

[Anonymous Coward]

I clicked on both the agree and disagree buttons. They do in fact go to different pages. Clicking on the I agree button takes you to a very sparse page with a link to download a Word document containing the specifications. When you actually dig around on the page you're directed to when clicking "disagree" to download the specification, you end up back at the same license agreement page.

You must agree to their license to get the specification.

Re:first reaction, second reaction

[BobNET]

I and many many MANY webdesigners would kill, no.. go on a murder rampage for an image format that has alpha transparency to show a background through.

AA! Don't kill me! [ntlworld.com]

Re:JPEG-LS Vs JPEG

[gowen]

Jpeg2000 supports both lossy and lossless compression, and a variety of wavelet-based compression schemes that work better than normal JPEG.

Unfortunately, no-one supports Jpeg2000.

Re:it exists already

[the eric conspiracy]

If you have a DSLR and are not using RAW you are wasting 1/2 or more of the capability of your camera. The cost of a CF card is trivial compared to the cost of even one good lens let alone the entire DLSR kit which is likely to run 2000+.

Re:first reaction, second reaction

[Alan]

The thing is, their browser doesn't even support PNG properly yet (even ie7 I believe), so why would I believe that a) they could support this properly or b) everyone else would.

My other reaction is regarding the photography side of it. Professional photographers aren't going to stop using tiff/raw formats anytime soon, and non-pros are happy enough with jpg because they don't know or care about the format, and really just want something they can get at easily and share/print easily.

Oh, and I don't trust MS not to mess up a potentially good format (if it is that) with licensing issues or other such trickery.

click-through agreement

[rs232]

"You may review these Materials only .. to interface with a Microsoft product"

"MICROSOFT MAKES NO WARRANTY OF ANY KIND"

"If .. ownership .. changes, Your right to use these Materials automatically terminates"

"Microsoft may freely use, reproduce, license, distribute, and otherwise commercialize Your Feedback"

"You will not give Microsoft any Feedback (i) that You have reason to believe is subject to any .. intellectual property claim"

"Microsoft has no obligation to maintain the confidentiality .. of Your Feedback"

"You waive any defenses allowing the dispute to be litigated elsewhere"

"If any part of this Agreement is unenforceable, it will be considered modified to the extent necessary to make it enforceable"

from "Windows Media Photo Specification license agreement [microsoft.com].

Re:Even a better one

[y2dt]

PNG and JPEG are not designed for the same purpose. PNG is lossless and was intended to replace GIF.

http://en.wikipedia.org/wiki/Png [wikipedia.org]

Re:Ummmm why?

[Epistax]

Ok fine no one gets it. A banddand is a range of frequencies you will find in an image. As it turns out, we don't respond to error in an image by the image itself, but by the frequency that the error in the image disrupts. We're worse at seeing disruptions in the high and low range of frequencies, and better in the midrange. Somewhat ironically that means we can take advantage of the high and low and compress more inside those frequency ranges. A DWT or DCT wil give you component pieces for various frequencies which you can simply or delete to form the compression (DCT is JPEG, DWT is JPEG2000). Remember the square blocks in JPEG compression? That's from the DCT. The DWT is more circular so you'll never see square blocking with JPEG2000.

If anyone is interested and wants some not-so-light reading, check out http://foulard.ece.cornell.edu/publications/chandl er_5749_40.pdf [cornell.edu]
It'd be awesome if someone made a compressor for regular images using this technique.

Re:Alternative or Replacement?

[grumbel]

JPEG2000 an alternative? Wake me up in 20 years or so when the patent mess around JPEG2000 is solved...

Re:Ummmm why?

[virtualchoirboy]

Or maybe it's because Forgent Networks filed a lawsuit against MS and Apple and about 40 other companies over the JPEG compression algorithms. If this standard gets adopted and popular, MS can drop JPEG entirely and pay penalties, but no licensing fees while earning licensing fees in return.

I only did one Google search, but easily came up with this [macnn.com] old article from last October. I haven't really followed the case, but it's one reason why MS may have done this.

Re:Big claims indeed!

[tongue]

The patent on LZW compression (the problematic portion of the GIF format) expired on June 30, 2003 [unisys.com], though Unisys claims to have patents and patents pending on a number of improvements to the claims in the original patent.

License to read

[sankyuu]

I was gonna take a look at the specs, since I wanted to see how WMP compares to PNG and JPEG. Did they really get some smart people to come up with a better compression algorithm? The news article was very sparse on detail, so I went to the microsoft site, which had an opening page saying (and I paraphrase):

1.You may read the specs only if you use them to write software which work with Microsoft products, or

2. To provide feedback on the Specs to Microsoft.

Duh... where's the "I wanna know if it's any good" option? I did a google of "Windows Media Photo" to see if there was any more stuff to read, but I didn't get anything more informative than the linked cnet article.

Re:Another Debate

[MasterC]

PNG is only lossless compression so I suppose it's only natural to switch to a file format that can be either lossless or lossy...

You do realize that JPEG can be lossless, right? Just completely skip the quantization step (or, equivalently, use a quantization matrix of ones) and you've got lossless. It's all in the Q matrix.

Re:It is TIFF hijacked

[Holi]

Yes because Wikipedia is always 100% accurate. Try here [digitalpreservation.gov], BTW the AC was right, Microsoft worked with Aldus on the development of Tiff.

Re:Ummmm why?

[hummassa]

I clicked on both the agree and disagree buttons. They do in fact go to different pages. Clicking on the I agree button takes you to a very sparse page with a link to download a Word document containing the specifications. When you actually dig around on the page you're directed to when clicking "disagree" to download the specification, you end up back at the same license agreement page.

You must agree to their license to get the specification.

OR... you download it directly from this place? [microsoft.com]

Re:JPEG-LS Vs JPEG

[gowen]

So does ImageMagick, FWIW. And there's an extension for firefox that supports it, too.

But, still. That's not going to butter any parsnips.

JPEG alternative

[Anonymous Coward]

For a JPEG alternative look at APT: http://www.intuac.com/userport/john/apt/index.html [intuac.com]. Better compression and mostly patent-free.

Re:Another Debate

[Evil Grinn]

I mean, they used to make only operating systems (which took them a while to perfect)

Microsoft never made only operating systems. Go learn about BASIC.

Re:Ummmm why?

[Whiney Mac Fanboy]

Well, according to NE asia's may edition, the USPTO is going reexamine forgent's patent [nikkeibp.com] (at the behest of The public patent foundation [pubpat.org]. But that's really kind of besides the point 'cause all of MS's products are going to come under patent attacks.

There's just too many software patents out there (and too many broad ones) for MS to work their way around all of them, forgent will just buy up some company with a patent on entropy encoding & turn around & sue MS.

If jpeg patents are MS's fear, a new image format is only going to buy them a little time, but if DRM is their goal, it makes perfect sense.

Licensing should not be a restriction? Try FAT

[massysett]

That from a company that wants to charge license fees [microsoft.com] for FAT? Yeah, right. They might not charge licensing fees now, but if this graphics standard ever gets to be twenty years old, not under active development, and ubiquitous, watch out.

Re:Another Debate

[dominator]

*) Your 2-3MB JPEG photos will end up being not-so-small 20MB PNGs if you decoded them to RGBA and re-compressed them.

*) PNGs are nice and work a lot of places, but things like their alpha channel aren't properly supported on Windows, meaning that they don't work properly eveywhere.

*) All of GIF's patents have expired by now, or were held by IBM, never enforced, and are due to expire in ~2 months. Its lawsuit history is largely irrelevant now, except as historical anectdote. You should feel free to evaluate the recently-unencumbered format on its own merits.

http://en.wikipedia.org/wiki/GIF#Unisys_and_LZW_pa tent_enforcement [wikipedia.org]

Re:Ummmm why?

[Anonymous Coward]

First of all, forget color. A color image is just three channels of black and white.

Imagine you have a back and white image which is pure white noise. Consider what a single horizontal line of that image would look like if you drew it as you would a sound wave, with the bright pixels being high, and the dark pixels being low.

As you step from one pixel to the next, you could have a change of up to 255. There's no predictable pattern. The "frequency" of this noise is high, because the potential difference from one pixel to the next is great.

Now imagine that you apply a smoothing filter to this line of noise, and bring the changes from one pixel down. That is what you get if you blur an image. Now the max differences from one pixel to the next is much lower. The frequencies in a blurry image are low.

There's other ways to consider the frequencies of an image as well. In Wavelets, you would scale the image down to 2x2, and this would be one layer of the image. Then you'd scale it down to 4x4, and scale up the 2x2 image with bilinear filtering and subtract it from the 4x4 image. The 4x4 difference image now represents a different set of frequencies than the 2x2 image did. You store the difference because what you're interested in is the frequency of the 4x4 layer. You want to add that frequency on top of the 2x2 layer when you reconstruct the image, and if you have that "frequency" seperated out, you can compress the data better.

Another way of looking for frequencies in an image is to seperate the image into bitplanes. I think TIFF does this, because it comrpessed the image about the same as seperating the image into bitplanes then compressing with zip. Anyway the idea here is to take all the first bits of each pixel and stick them one after another, and then stick the second bits of all the pixels one after the other... You'll end up with 8 images this way, and you'll find that the image with the highest bits is easily recognizeable and has clear sharp edges, but when you get to the image with the lowest bits, all you have is noise. If you discard that noise when reconstructing the image then you will get banding in the image, but you could in theory interpolate the values of the band above to fill in the noise. You'll lose noise in the image though so stuff will look smoother than it did. Wavelet does somethign similar when it discards the differences and smooths the portions of the image that are in between sharp edges.

Re:JPEG-LS Vs JPEG

[mikeboone]

Plus, I think that JPEG 2000 is still under the threat of patent litigation [google.com]. Too bad, because lossless JPEG 2000 files are a lot smaller than similar TIFFs.

Re:Another Debate

[jelle]

"Just completely skip the quantization step (or, equivalently, use a quantization matrix of ones) and you've got lossless."

Not exactly. It's close, but it's not lossless. The reason for this is that the DCT is not fully reversible, and even if it were, the JPEG standard does not have a precision standard for the DCT/IDCT in encoders/decoders (not even IEEE-1180 like MPEG1/MPEG2), which means that there can be quite large differences between different encoders/decoders.

And JPEG-LS doesn't even use the DCT...

http://www.jpeg.org/jpeg/jpegls.html [jpeg.org]

Marketing vs. Technical Gore

[abb3w]

If you click on the "I do not accept this agreement." button, it submits the value "I do not accept this agreement.", and you get taken to http://www.microsoft.com/whdc/xps/default.mspx? [microsoft.com], with some generic marketroid babble about how their new spec Whitens teeth, cures BO, and will put a chicken in every pot and pot in every chick.

If you click on the "I accept this agreement and want to download the Windows Media Photo Specification" button, it submits "I accept this agreement and want to download the Windows Media Photo Specification", and should take you to http://www.microsoft.com/whdc/xps/wmphotodwn.mspx? [microsoft.com]. However, I didn't verify that.

Instead, I chose to look at the HTML, and manually submitted my own prefered value via manually entering the URL: http://www.microsoft.com/whdc/xps/wmphotodwn.mspx? I_Reject_The_Agreement_Terms_and_Suspect_Bill_Gate s_Blows_Goats [microsoft.com]. I also got taken to the download page. This page contains the notice "By installing, copying, or otherwise using the software, you agree to be bound by the terms of the license agreement [microsoft.com].", and a download link to the actual specification document at http://download.microsoft.com/download/1/6/a/16acc 601-1b7a-42ad-8d4e-4f0aa156ec3e/WMPhotoSpec_v09.do c [microsoft.com]....

Oops.

Now, while I Am Not A Lawyer, I submitted my rejection of their license terms, so I'd argue in court I shouldn't be bound by them; and since this is a specification, and not itself software, I would also argue that the notice on the page I reached is moot. I suppose the case could be made that since Word macros are a turing-complete programming language, the word document is software, so I thought I'd look through using "less" to be on the safe side. Lo and behold, there is another license embedded:

"READ THIS! THIS IS A LEGAL AGREEMENT BETWEEN MICROSOFT CORPORATION ("MICROSOFT") AND THE RECIPIENT OF THE ABOVE REFERENCED MATERIALS, WHETHER AN INDIVIDUAL OR AN ENTITY ("YOU"). IF YOU HAVE ACCESSED THIS AGREEMENT IN THE PROCESS OF DOWNLOADING THESE MATERIALS ("MATERIALS") FROM A MICROSOFT WEB SITE, BY CLICKING "I ACCEPT", DOWNLOADING, USING OR PROVIDING FEEDBACK ON THE MATERIALS, YOU AGREE TO THESE TERMS. IF THIS AGREEMENT IS ATTACHED TO MATERIALS, BY ACCESSING, USING OR PROVIDING FEEDBACK ON THE ATTACHED MATERIALS, YOU AGREE TO THESE TERMS. IF YOU DO NOT AGREE TO THESE TERMS, YOU ARE NOT AUTHORIZED TO ACCESS, DOWNLOAD, USE OR REVIEW THE MATERIALS."

...followed by a bit more legalese, including that you're not allowed to remove the legalese and redistribute. This "license" strikes me as dangerously like a "license to read", which I'm sure various civil libertarian groups could have lots of fun with. I'd be amused to hear the opinon on a Real Lawyer (TM) as to how binding that would be. Anyone have Larry Lessig's phone number?

Of course, if someone at a unix command prompt incanted something clever (say, curl -o Bill_Blows_Goats.txt -C 8261 http://download.microsoft.com/download/1/6/a/16acc 601-1b7a-42ad-8d4e-4f0aa156ec3e/WMPhotoSpec_v09.do c [microsoft.com] — and don't forget to remove the Slashdot inserted spaces) the Microsoft server would only give them the meaty parts (albeit in a form even OpenOffice would probably gag on), and omit the license. I'd be amused to hear the opinion of a Real Lawyer as to how binding the agreement co

Re:JPEG-LS Vs JPEG

[gowen]

It's not [fedcir.gov]. Lizardtech finally ran out of routes for appeal earlier this year.

Half right

[SuperKendall]

GIF was good enough. PNG exists because the world is full of geeks who think that they are lawyers.

Or they wanted alpha channels and greater bit depths and better compression options.

ZIP is good enough. RAR and 7z (and bz2 and ...) are stupid.
WAV is not a compressed format. Apples and oranges.

Damn straight

MP3 is good enough. AAC is evil. (hint - the difference isn't quality, it is control)

AAC is MP4, a patented but open standard - just like MP3. Apples use of AAC in ITMS wraps the Fairplay DRM around that you are so coyly referring to, but really has nothing to do with AAC itself.

The original comparison stands in two cases.

Re:Marketing vs. Technical Gore

[Heisenbug]

Hey, I'm not a lawyer, but I am in law school. As a rule of thumb, you should remember that laws aren't evaluated by computers -- they're evaluated by humans. Especially in contract law, which is all about equity, the judge is going to do whatever he thinks is common sense, with an eye towards making the market work.

So, you know it's a loophole, Microsoft knows it's a loophole -- I bet the judge will know too, and shoot it down. I don't really know the answer, but I suspect that you would stand up and say, "your honor, I read the html, and found a way to respond that would trick the machine into giving me the benefit of the contract without sending back the exact text I was supposed to." Microsoft would stand up and say, "your honor, this is like figuring out a way to get a vending machine to give you candy without putting in any money. What would happen if that was legal?"

Re:What advantages does it offer?

[Anonymous Coward]

Um, JPEG already offer totally lossless rotation.

I do it all the time with jpegtran.

You can also crop and a number of other operations, without ever decoding and reencoding the file.

The MS shill clearly has not heard of it.

I'm really not sure why it's not more used, but somehow it's not.

Re:Ummmm why?

[ajs]

To quote MS in answer to your question:

Objectives for Introducing a New Still Image Format

Today's file formats for continuous tone images present many limitations in maintaining the highest image quality or delivering the most optimal system performance. Windows Media(TM) Photo was designed to remove these limitations. The design objectives include:

• High performance, embedded system friendly compression
  
  • Small memory footprint
    
  • Simple, integer-only operations (no divides)
  
  
• Industry-leading compression quality
  
• Lossless or lossy compression using the same algorithm
  
• Support a very wide range of pixel formats:
  
  • Monochrome, RGB, CMYK or n-Channel image representation
    
  • 8 or 16-bit unsigned integer
    
  • 16 or 32-bit signed integer
    
  • 16 or 32-bit floating point
    
  • Several packed bit formats
    
    • 1bpc monochrome
      
    • 5 or 10bpc RGB
      
    • RGBE Radiance
  
  
• Simple, extensible TIFF-like container structure
  
• Planar or interleaved alpha channel
  
• Embedded ICC Profile
  
• EXIF and XMP metadata

Windows Media(TM) Photo is the only format that offers high dynamic range image encoding, lossless or lossy compression, multiple color formats, and performance that enables practical in-device implementation.

Re:Yeah, Mk...

[ceeam]

Sure I can:

http://www.microsoft.com/mscorp/ip/tech/fat.asp [microsoft.com]

Re:Ummmm why?

[TheRaven64]

AAC also has no reason to exist, your [sic] right about the control

No. MP3 is MPEG-1 audio layer 3. It was part of the initial MPEG specification. It was about as good as could be done with the processing power available at the time, but used a fairly primitive psycho-acoustic model and had noticeable artefacts. The MPEG-2 specification introduced an additional way of encoding audio, the Advanced Audio CODEC (AAC), which gave significantly better compression. This was refined (new profiles were added) in MPEG-4. All of these provided significant improvement over the original.

bzip2 serves a different purpose to zip and is more of a pointless replacement for gzip

No. Gzip is a stream compressor. Bzip2 is a block compressor. You can add gzip to a stream with minimal latency. Bzip2 requires blocks of 100-900KB to work with. If you sent an IM session through bzip2, then it would add huge delays. Gzip would not. This is why gzip is used for things like HTTP - you can just add it into the output stream and decompress it at the browser's end. Bzip2, however, gives significantly better compression ratios on files, for precisely the same reason. They do not serve the same purpose (although some people do seem to persist in using gzip as if it were a block compressor).

Hello new feature, same as the old.

[jandrese]

According to the C|Net article:

Additionally, Microsoft's new image format allows such things as rotating the image without the need to decode it and subsequently encode it again, he said.

Uh, JPEG already does this. Has nobody at C|Net ever heard of jpegtran? Does the MS format allow non 90 degree turns or something?

Re:first reaction, second reaction

[jonadab]

> doesn't even support PNG properly yet (even ie7 I believe)

As near as I can tell (from what testing I have done), the PNG support in IE7 is excellent. There are still some issues with CSS2, but PNG support seems to be good to go.

IE Support of PNG as good as GIF

[Anonymous Coward]

It seems I say this time and time again, but the idea that GIF is better supported by IE than PNG is completely and utterly incorrect. Every feature of GIF that IE supports, IE also supports with PNG. There is no reason whatsoever to prefer GIF over PNG, and in fact there are lots of reasons not to (PNG is nearly always smaller, for example, and is W3C standard, if you care).

"But wait," I hear you say! "The slashbots are always saying that IE support of PNG is bad! Why are you disagreeing?" And the answer, my friend, is that I'm not disagreeing. GIF is very primitive format compared to PNG. There are many advanced features of PNG that GIF does not support that IE does not always support.

For example, GIF is limited to 8-bit color and optional 1-bit transparency. This means that a GIF image can only display 256 colors, with one color optionally labled as "transparent". GIF cannot do true color, it cannot do alpha transparency. If you are willing to use GIF instead of PNG, you are by definition not using these features of PNG, because GIF does not support them in the first place.

Guess what? 8-bit PNGs with 1-bit transparency work just fine in IE, and are smaller than GIF! Exactly the same features, which work just as well! I'll take a moment to note that some versions of the GIMP that I've used will not produce 8-bit PNGs with 1-bit transparency and require some command-line post processing to produce them, but that's a problem with the GIMP, not a limitation of the PNG specification or a flaw in IE's rendering of 8-bit PNGs with 1-bit transparency.

PNG can also do 16-bit, 24-bit and I believe 32-bit color, and guess what? IE supports it! So you can produce richer, more colorful graphics with PNGs than you can with GIFs.

Then, there's alpha transparency, the holy grail of PNG. With alpha transparency, you can produce anti-aliased text that will display properly on any background, along with a host of other effects that anyone who has tried web-design for more than about two hours will tell you is sorely needed. This feature -- which GIF doesn't support either -- is not supported properly by IE (although there is a hack that can make it work, apparently).

So let's recap: everything you can do with GIF (with the exception of animation) can be done with PNG with better compression to boot, and without creating any problems with IE. There are some things that PNG can do that GIF can't do anyway (like alpha transparency) that IE chokes on. If you're using GIF in your websites because IE doesn't handle PNG's alpha transparency features properly, then you're stupid, because GIF doesn't either.

Capiche?