Honestly, after using it for a bit it becomes motor memory and there are no problems right-clicking at all. My main complaint with the Mighty Mouse is that the scroll ball gets dirty so easily and stops working â" it was just a week or two ago that I found out how to clean it, before that I'd been page-up and page-downing my way around the internets.
Update: I've been told by the devs that Dirac is optimized for HD live action, wheres my tests have thus far involved SD animated content, so, YMMV. I'll have to try some live action sources next.
Dirac employs wavelet compression, instead of the discrete cosine transforms used in most older codecs (such as H.264/MPEG-4 AVC or SMPTE's VC-1). Dirac is one of several projects attempting to apply wavelets to video compression. Others include Rududu [2], Snow and Tarkin. Wavelet compression has already proven its viability in the JPEG 2000 compression standard for photographic images.
Theora gets a bad rap for being outdated technology, but it does have a few advantages over MPEG-4 ASP: the loop filter, adaptive block sizes, and multiple reference frames, putting it closer to H.264 than MPEG-4 ASP. With these features, it's really a pretty strong showing from Xiph, and things can only get better as the encoder nears 1.0.
As I say below, unfortunately the quality is lacking compared to modern codecs like H.264 and even (dare I say) VC-1. Apparently that's just the nature of using wavelets. While they give a very natural style of compression on still images (JPEG-2000, etc), they do not translate well to moving sequences because, unlike all other current codecs, the image is not broken up into blocks that can then be tracked and diff'd in time.
Still, it'll be interesting to follow Dirac, if only because they're taking a radical new approach with only Michael Niedermayer's Snow as a peer.
While it's very cool what the BBC is doing, and it's good to see wavelet technology being pushed, Dirac 1.0 falls extremely short in my tests (at least on animated material at medium bitrates). In the H.264 era, the quality is unacceptable. Here's hoping they'll be able to keep improving it. On the other hand, I know at least one x264 dev who's convinced that OBMC wavelets will never match the quality of MC block-based approaches without a major breakthrough.
I think they're speaking in contrast to a BSD-style license, where there is no stipulation regarding re-committing modified code. In other words, with the GPL, a company can't absorb the Linux kernel into a proprietary project without having to open up all further modifications. This is unlike the BSD license, which is truly free in the sense that the code can be used in any fashion, proprietary and modified or not.
Exactly. I'll copy/paste my response to this highly innacurate article on Digg:
"Some other major inaccuracies:
This site says that motion compensation was introduced in MPEG-4. What? Motion compensation and motion estimation are at the core of every MPEG and most other codecs.
Also, his understanding of the DCT is way off (and no, you don't need a degree to understand it -- I was building JPEG encoders in 11th grade).
"During an encode, every number in a series is simply halved and the remainders thrown away."... What? The DCT is a complex algorithm that converts an array or matrix of values into frequency space, producing a set of frequency coefficients that represent respective values of cosine waves at different frequencies along the set. Halving the data has nothing to do with it unless you're using a quantizer of two and a quantization matrix of all ones (or a QM of all twos and a quantizer of one...).
This is also wrong: "Think of a quantisation matrix as a palette of values that controls how the pixels in a macroblock are converted from pixels to a formula."
That's the DCT that he's describing. The quantization matrix simply defines what values the frequency coefficients are divided by. I think he has quantization and the DCT mentally swapped.
Oh also, he acts like QPel and GMC are Xvid inventions. In actuality, most MPEG-4 codecs implement these. Granted, DivX's GMC is inferior to Xvid's implementation (one warp point vs three), but it certainly does have it."
Of course, this got dugg down to hell because my comments about the DCT were mistaken for boasting. Hopefully the same won't happen here.
This is one of those things that sounds insightful and seems to make sense at first glance, but falls apart under scrutiny. The reality is that the brain as an external, observable object made up of neurons is completely different from the internal, mental representation made up of mental constructs. There's no reason to believe that we cannot form a complete and accurate map of the brain in human terms. Just because the brain is complex in physical structure, that does not mean that it is impossible to understand semantically. I *think*.
Man, NeXT really was way ahead of the game. It's a shame it was marketed almost exclusively for research and business -- think of how much nicer the personal computing landscape would be today if we had all been running NeXTSTEP instead of Windows throughout the nineties.
Even going up against Compiz, the completely non-accelerated, non-3D Enlightenment DR17 manages to be vastly more beautiful. And it's fast and incredibly light-weight, too. Nothing else delivers eye-candy and performance like Enlightenment.
Nevertheless, here's the original statement that I responded to: "Hurray about those Open Standards huh? Good thing Apple's been a champion of them for years, with Quicktime supporting the book of Open Standards and VLC to support the rest of the nutso formats and encodings."
No, the statement you responded to was:
What Quicktime files can't you play on Linux?
To which you replied:
Sorenson for one
To which I commented:
libavcodec has had a competent Sorenson decoder for years.
There has been an SVQ < 3 decoder in LAVC for even longer. The OP's point about open standards would be valid if Apple were still promoting Sorenson or some other proprietary codec, but they're not. This is more than Windows is doing: WMV9 may be compatible with VC-1, but newer iterations (WMV9 Advanced Profile) extend past VC-1 spec and are proprietary.
Just for the record (WM is still an awful platform), as far as quality goes, XviD and DivX can't hold a candle to WMV9/VC-1. This is simply because WMV is essentially a Microsoft clone of H.264/MPEG-4 AVC, while DivX and XviD are comitted to the much less complex MPEG-4 ASP.
A lot of the comments here suggest that you buy a several thousand dollar encoder to fix this problem... not really the best solution. In reality, TMPGEnc (already mentioned) has long been on par with CCE quality-wise. Furthermore, it's free.
However, the problem probably is less the encoder than the source. DV is usually incredibly noisy, and thus very very difficult to encode. This can be helped a lot by good filtering. TMPGEnc includes some filtering options that may work for you, but if you're looking for the real deal (and have a lot more time to blow), head on over to doom9.org. Back when I was doing this sort of thing, the Convolution3D AVISynth filter was the best way to increase compressibility, but that was two-three years ago, so things may have changed. That may be a good starting-point, though.
Personal stories aren't worth much on this scale, but yeah, problems started surfacing with mine after three weeks. Most are total non-issues ("staining" which is only visible in specific lighting at a specific angle, the "moo" which happens once in awhile and really isn't that distracting, and the "whine" which I heard once in a very quiet room), however, the one problem that'll be bringing me to the nearest Apple store shortly is a problem with the inverter board (?) which causes the LCD backlight to go out frequently and not come back until I sleep it. It's annoying, but there's enough to love about the 'Book that makes it well worth the minor inconveniences.
Slashdot Mirror
http://bestof.provocateuse.com/images/photos/adam_brody_98.jpg Fucking brilliant!
Honestly, after using it for a bit it becomes motor memory and there are no problems right-clicking at all. My main complaint with the Mighty Mouse is that the scroll ball gets dirty so easily and stops working â" it was just a week or two ago that I found out how to clean it, before that I'd been page-up and page-downing my way around the internets.
Update: I've been told by the devs that Dirac is optimized for HD live action, wheres my tests have thus far involved SD animated content, so, YMMV. I'll have to try some live action sources next.
Dirac employs wavelet compression, instead of the discrete cosine transforms used in most older codecs (such as H.264/MPEG-4 AVC or SMPTE's VC-1). Dirac is one of several projects attempting to apply wavelets to video compression. Others include Rududu [2], Snow and Tarkin. Wavelet compression has already proven its viability in the JPEG 2000 compression standard for photographic images.
Yes it does :|
Theora gets a bad rap for being outdated technology, but it does have a few advantages over MPEG-4 ASP: the loop filter, adaptive block sizes, and multiple reference frames, putting it closer to H.264 than MPEG-4 ASP. With these features, it's really a pretty strong showing from Xiph, and things can only get better as the encoder nears 1.0.
How does it stack up to other codecs?
As I say below, unfortunately the quality is lacking compared to modern codecs like H.264 and even (dare I say) VC-1. Apparently that's just the nature of using wavelets. While they give a very natural style of compression on still images (JPEG-2000, etc), they do not translate well to moving sequences because, unlike all other current codecs, the image is not broken up into blocks that can then be tracked and diff'd in time. Still, it'll be interesting to follow Dirac, if only because they're taking a radical new approach with only Michael Niedermayer's Snow as a peer.
While it's very cool what the BBC is doing, and it's good to see wavelet technology being pushed, Dirac 1.0 falls extremely short in my tests (at least on animated material at medium bitrates). In the H.264 era, the quality is unacceptable. Here's hoping they'll be able to keep improving it. On the other hand, I know at least one x264 dev who's convinced that OBMC wavelets will never match the quality of MC block-based approaches without a major breakthrough.
Gah. I wasn't intending to troll with the "truly free" comment. Please don't mod me down :[
I think they're speaking in contrast to a BSD-style license, where there is no stipulation regarding re-committing modified code. In other words, with the GPL, a company can't absorb the Linux kernel into a proprietary project without having to open up all further modifications. This is unlike the BSD license, which is truly free in the sense that the code can be used in any fashion, proprietary and modified or not.
"which outperform the DCT based methods"
Not so fast. None of the current wavelet codecs (Dirac, Snow) outperform H.264 at this time. Probably in the future, but not just yet.
Exactly. I'll copy/paste my response to this highly innacurate article on Digg:
... What? The DCT is a complex algorithm that converts an array or matrix of values into frequency space, producing a set of frequency coefficients that represent respective values of cosine waves at different frequencies along the set. Halving the data has nothing to do with it unless you're using a quantizer of two and a quantization matrix of all ones (or a QM of all twos and a quantizer of one...).
"Some other major inaccuracies:
This site says that motion compensation was introduced in MPEG-4. What? Motion compensation and motion estimation are at the core of every MPEG and most other codecs.
Also, his understanding of the DCT is way off (and no, you don't need a degree to understand it -- I was building JPEG encoders in 11th grade).
"During an encode, every number in a series is simply halved and the remainders thrown away."
This is also wrong:
"Think of a quantisation matrix as a palette of values that controls how the pixels in a macroblock are converted from pixels to a formula."
That's the DCT that he's describing. The quantization matrix simply defines what values the frequency coefficients are divided by. I think he has quantization and the DCT mentally swapped.
Oh also, he acts like QPel and GMC are Xvid inventions. In actuality, most MPEG-4 codecs implement these. Granted, DivX's GMC is inferior to Xvid's implementation (one warp point vs three), but it certainly does have it."
Of course, this got dugg down to hell because my comments about the DCT were mistaken for boasting. Hopefully the same won't happen here.
This is one of those things that sounds insightful and seems to make sense at first glance, but falls apart under scrutiny. The reality is that the brain as an external, observable object made up of neurons is completely different from the internal, mental representation made up of mental constructs. There's no reason to believe that we cannot form a complete and accurate map of the brain in human terms. Just because the brain is complex in physical structure, that does not mean that it is impossible to understand semantically. I *think*.
Hint: as far as most people are concerned, they still aren't. Though it is very easy to argue on both sides of that fence.
Mrs. Longhurst should just kill herself and become AN HERO.
They already have. The new WMV9 Advanced Profile goes beyond VC-1 spec.
Man, NeXT really was way ahead of the game. It's a shame it was marketed almost exclusively for research and business -- think of how much nicer the personal computing landscape would be today if we had all been running NeXTSTEP instead of Windows throughout the nineties.
Even going up against Compiz, the completely non-accelerated, non-3D Enlightenment DR17 manages to be vastly more beautiful. And it's fast and incredibly light-weight, too. Nothing else delivers eye-candy and performance like Enlightenment.
There has been an SVQ < 3 decoder in LAVC for even longer.
The OP's point about open standards would be valid if Apple were still promoting Sorenson or some other proprietary codec, but they're not. This is more than Windows is doing: WMV9 may be compatible with VC-1, but newer iterations (WMV9 Advanced Profile) extend past VC-1 spec and are proprietary.
libavcodec has had a competent Sorenson decoder for years.
Yeah, a system so locked-down and proprietary that the entire system base is open source and based on open standards as old as personal computing. Oh, wait, what?
Just for the record (WM is still an awful platform), as far as quality goes, XviD and DivX can't hold a candle to WMV9/VC-1. This is simply because WMV is essentially a Microsoft clone of H.264/MPEG-4 AVC, while DivX and XviD are comitted to the much less complex MPEG-4 ASP.
Ah, yeah, I guess it's been too long. In that case, QuEnc (a free encoder based on libavcodec from the FFMPEG project) may be the best choice.
A lot of the comments here suggest that you buy a several thousand dollar encoder to fix this problem... not really the best solution. In reality, TMPGEnc (already mentioned) has long been on par with CCE quality-wise. Furthermore, it's free.
However, the problem probably is less the encoder than the source. DV is usually incredibly noisy, and thus very very difficult to encode. This can be helped a lot by good filtering. TMPGEnc includes some filtering options that may work for you, but if you're looking for the real deal (and have a lot more time to blow), head on over to doom9.org. Back when I was doing this sort of thing, the Convolution3D AVISynth filter was the best way to increase compressibility, but that was two-three years ago, so things may have changed. That may be a good starting-point, though.
Personal stories aren't worth much on this scale, but yeah, problems started surfacing with mine after three weeks. Most are total non-issues ("staining" which is only visible in specific lighting at a specific angle, the "moo" which happens once in awhile and really isn't that distracting, and the "whine" which I heard once in a very quiet room), however, the one problem that'll be bringing me to the nearest Apple store shortly is a problem with the inverter board (?) which causes the LCD backlight to go out frequently and not come back until I sleep it. It's annoying, but there's enough to love about the 'Book that makes it well worth the minor inconveniences.