Capacity differences won't be that big a deal here - both are very large for movie content.
Video codecs are the same.
Copy production is likely to be the same.
But there still are significant differences:
First, you can make a HD-DVD today. Apple's DVD Studio Pro 4, which shipped a few months ago, can make a HD DVD disc. I've made two so far. Heck, here's a torrent to a copyright-free one:
The only player I've tested so far is DVD Player 4.6, running on a Mac G5. I believe Moonlight will be releasing a player for Windows soon. But from a first mover perspective, HD DVD is quite a bit close to having content be released. I wouldn't be surprised if there are 100+ HD-DVD discs on the market before the first Blu-ray.
HD-DVD requires very little capital cost to add to a current DVD production line. Not true of Blu-ray.
Blu-ray has a thinner plastic layer, so all things being equal, it won't be as durable.
And lastly, Sony has lost EVERY consumer format war they've participated in these last few decades. If anything, them owning content has proved to be more disynergistic than synergistic. Other media companies would be more interested in UMD if Sony didn't own a competing company.
Honestly, as others have posted, it's much worse to have two good formats than one, since customer confusion could sink both (leaving HD IPTV a likely winner). Better for the industry to coalesce around one format and declare it an early winner.
I pick HD-DVD, mainly because I can already make them, and I know it's already good enough. If we just had Blu-ray, I'd back it enthusiastically, but I think it's better we geeks declare a winner early (even if it's arbitrary) to avoid a drawn out format war. It's better for one to lose quickly.
Can you give a real-world example where Blu-ray would provide a better experience than HD-DVD for "Hollywood" style content. Sure, as a floppy disc replacement for rewritable files. But for read-only video content?
Bear in mind that once you get a high enough peak data rate, higher data rates don't look any better. So it isn't that the capacity of Blu-ray means all discs will look better - for the vast majority of films, both formats would let you use a maximum legal bitrate throughout the file. It's only titles where the greater capacity of Blu-ray over HD-DVD means you can use a higher average bitrate that you'd see a difference. And I suspect that'll come in somewhere north of 5 hours per side. Not a lot of content out there where you care about watching for more than 5 straight hours without interruption. And given that HD-DVD is cheaper to manufacture (at least at the outset) at a given bitrate, HD-DVD would be cheaper per minute of video. Sure, maybe an entire HD TV show season might take 3 HD-DVD discs instead of 2 Blu-ray discs, but does that difference really have much consumer value?
And how the heck would you know that? The Blu-ray camp has made that assertion, but it simply isn't born out in real-world testing.
Last week, for a test, I put a 123 minute movie on a DVD-9 using MPEG-2, using the HD DVD format (via Apple's DVD Studio Pro 4). Average of around 8.5 Mbps. Looked pretty darn good at 1920x1080.
HD-DVD gives you 30 GB, and the use of H.264 and VC-1 for codecs. No problem AT ALL sticking "Return of the King Extended Edition" on a single side of HD-DVD. So using codecs that are 2x better and 3x more capacity, yeah, HD-DVD is just fine. Single layer HD-DVD will be fine for the vast majority of films, and even offers more minutes per disc at HD than DVD gives us minutes of SD today.
Actually, FireWire is the dominant protocol for compressed video applications, which is a huge market. And it's used widely for external storage as well.
Sure, it isn't used for keyboard and other areas where real-time performance isn't critical, but it doesn't have any advantages there.
Where isosynchronous performance is critical, FireWire does very well.
I do a lot of trade shows, and do a lot of software demos. I work with Mac software and Windows software. Until now, I've lugged a PowerBook, and run VirtualPC or Remote Desktop when I needed to show off Windows stuff. Which kinda worked, and was often better than having to lug two laptops around everywhere.
Now i can have a true dual-boot portable machine?
Or likely better yet, a native-code version of VirtualPC running on MacOS X that can give native full screen performance? So I'm always Apple-Tab away from MacWindows. VirtualPC is going to get a lot more interesting with this - possibly a reason why Microsoft is supporting this effort.
Life is good here for us digital media road warriors.
No, PowerBooks dropped ADB in favor of USB back with the Bronze keyboard G3 - 1999? That generation still had SCSI, but the last pre-Titanium had FireWire and AirPort. All the G4 PowerBooks have been USB/1394/802.11.
iBooks never had ADB - they were USB from their introduction, as was the iMac.
Apple machines were using today's standard Windows interfaces ports well before most Windows machines had them:).
Far from factious. Cell would be enormously good at video processing kinds of applications, and if it's cheap enough could cetainly compete with the ASIC, DSP, and (increasingly) FPGA systems that are used in the field today.
I'm sure a set top box vendor loves the idea of a chip that can decode compressed video, process it, and run a Java applet in the same piece of silicon.
Actually, the I-frame only is quite a bit easier to decode, since it doesn't require motion estimation, doesn't reference multiple frames, etcetera. It's almost identical to a series of JPEGs at 29.97 frames per second.
In normal IBBP MPEG-2, a given B-frame will reference frames either direction of it, requiring two frames be buffered in memory, and a lot of moving around of data from those two frames.
QuickTime 7 is required for H.264 playback in order to support B-frames. QuickTime's original architecture back in 1991 required that frames only reference previous frames, not future frames - a feature that is a basic requirement of H.264. Apple has been working on a major rewrite for years to get this working.
The good news is that out of order playback means better MPEG-1, MPEG-2, and MPEG-4 Part 2 implementations are also now technically possible within QuickTime.
Great testament to the improvement in compression
on
Apple Easter Egg
·
· Score: 1
Wow,
That file was from right around when I started doing compression, and it's an amazing testament to how much things have gotten better.
Video is Apple Video, codenamed "Road Pizza" pre-Cinepak. No data rate control. 5-bit per channel RGB.
Audio is uncompressed 8-bit. Pre IMA, and well pre MP3 for computer playback. Computers of that era weren't fast enough play back MP3 in real-time, let alone with video along with it!
File totals 300 KBytes/Sec, or 2400 Kbps. One could easily provide better quality at 100 Kbps today. 24x improvement in a little over a decade - not Moore's Law, but not bad!
Video professionals aren't using Betamax
on
Apple Backs Blu-ray
·
· Score: 2, Insightful
Well, video professionals are still using the analog Betacam SP and Digital Betcam. They're based on the Betamax tape shell, but run at higher speeds and have much better image quality than Betamax did.
Sony Professional has certainly made enough profit on those formats to make up for the Betamax losses by now.
Whomever wrote the original article is pretty fuzzy on a lot of details.
Just to clear things up a bit:
VC-9 was based on Windows Media Video 9, which is the commercial release version of the WMV codec, plus the Advanced Profile extensions. It was later renamed VC-1. No difference between the two.
H.264 and VC-1 do have significant technical differences (I go worried when he described his research on this point as "Another source told us recently that they had had the codec explained to them, and confirmed that it did "pretty much" the same as H.264." Well, pretty much in the same way that MPEG-1 and RealVideo do pretty much the same thing. They're both codecs, but have significant differences with real-world differences. For example, VC-1 uses larger blocks than H.264, which helps with some content and hurts with others. H.264 supports multiple reference frames, which can improve compression, but slows encode and decode.
Lastly, these issues aren't that unusual - I doubt it would even be possible to build a competitive codec without stepping on a whole lot of patents. Microsoft has IP in H.264, after all. It's still not possible to build a patent free MPEG-2 decoder.
Well, there may be room for heroic optimization yet, but there aren't any G4 machines that can decode 1080i MPEG-2 in real-time. Not a problem on my dual G5, certainly.
Maybe they're limited by the limited FSB on the G4?
Oh, modern codecs can do a LOT better than the MPEG-2 subset in DVD.
H.264, which will be supported on that box in QuickTime 7, is roughly 3x more efficient than MPEG-2. So DVD quality at around 2 Mbps. You could do great 1080 24p HD at around 8 Mbps with it. WMV9 Advanced Profile from Microsoft offers equivalent performance, although likely won't be on the miniMac in a high performance version for a while.
Not sure if the G4 will be able to do HD decode of H.264, though. Apple has only demoed full HD decode on dual G5. Maybe if they're able to do some video card offloading or something.
I think you're missing the point. HE AAC is more than twice as efficient as today's leading class of codecs (AAC, WMA, Ogg). Twice is a big deal! Think of the difference between, say, MPEG-2 and H.264 or WMV9 Advanced Profile. It took video codecs a full DECADE to get the kind of improvement jump we're getting with HE AAC. That 20 Kbps stream can be a great sounding 44.1 with HE AAC - better than that 64 Kbps VBR stream you cite.
The technology has been around for a while in enterprise systems, but is only now trickling down to desktop use.
And AAC PS (parametric stereo) is just around the corner, which is more efficient yet.
Yes, HE AAC and AAC+ are the same thing. HE AAC is the name that MPEG gives it, and AAC+ is Coding Technologies name for their implementation.
Next up is AAC PS, for parametric stereo, which applies the SBR techniques to synthesizing stereo. Gives another big leap yet for music listening - 24 Kbps is good enough for people who can live with MP3 @ 160 or so.
Heck, Apple doesn't even do ASP (Advanced Simple Profile). They're simple profile only, for both encode and playback. And their SP encoder is quite infamously weak.
Squeeze is a great Mac MPEG-4 encoder. The other major commercial tool for that on Mac is Compression Master, which is a much better encoder for SP, and does good ASP as well. And there's ffmpegX and DivX for Mac as well.
Actually, VCD uses only Layer II audio, not MP3. There aren't any controlling patents or licensing fees for MPEG-1.
Your general point is very apt, of course. Except for VCD, virtually all media technologies require various patent licensing, and in practice these haven't resulted in any company gaining undo control over the technology. It just means that makes of encoders, players, and/or content have to pay a fee to make the stuff. But the licensing contracts don't let a company revoke or re-negotiate the license after it has been launched.
Slashdot Mirror
Capacity differences won't be that big a deal here - both are very large for movie content.
m g.torrent?1C6B407CD6671B2BB03F55C49D67CEB584A74D90
Video codecs are the same.
Copy production is likely to be the same.
But there still are significant differences:
First, you can make a HD-DVD today. Apple's DVD Studio Pro 4, which shipped a few months ago, can make a HD DVD disc. I've made two so far. Heck, here's a torrent to a copyright-free one:
http://216.99.212.233:6969/torrents/HD_DVD_TEST.d
The only player I've tested so far is DVD Player 4.6, running on a Mac G5. I believe Moonlight will be releasing a player for Windows soon. But from a first mover perspective, HD DVD is quite a bit close to having content be released. I wouldn't be surprised if there are 100+ HD-DVD discs on the market before the first Blu-ray.
HD-DVD requires very little capital cost to add to a current DVD production line. Not true of Blu-ray.
Blu-ray has a thinner plastic layer, so all things being equal, it won't be as durable.
And lastly, Sony has lost EVERY consumer format war they've participated in these last few decades. If anything, them owning content has proved to be more disynergistic than synergistic. Other media companies would be more interested in UMD if Sony didn't own a competing company.
Honestly, as others have posted, it's much worse to have two good formats than one, since customer confusion could sink both (leaving HD IPTV a likely winner). Better for the industry to coalesce around one format and declare it an early winner.
I pick HD-DVD, mainly because I can already make them, and I know it's already good enough. If we just had Blu-ray, I'd back it enthusiastically, but I think it's better we geeks declare a winner early (even if it's arbitrary) to avoid a drawn out format war. It's better for one to lose quickly.
Yeah, and note that they didn't even SHOW ANY CONTENT. It was basically a comparison of marketing sheets.
Can you give a real-world example where Blu-ray would provide a better experience than HD-DVD for "Hollywood" style content. Sure, as a floppy disc replacement for rewritable files. But for read-only video content?
Bear in mind that once you get a high enough peak data rate, higher data rates don't look any better. So it isn't that the capacity of Blu-ray means all discs will look better - for the vast majority of films, both formats would let you use a maximum legal bitrate throughout the file. It's only titles where the greater capacity of Blu-ray over HD-DVD means you can use a higher average bitrate that you'd see a difference. And I suspect that'll come in somewhere north of 5 hours per side. Not a lot of content out there where you care about watching for more than 5 straight hours without interruption. And given that HD-DVD is cheaper to manufacture (at least at the outset) at a given bitrate, HD-DVD would be cheaper per minute of video. Sure, maybe an entire HD TV show season might take 3 HD-DVD discs instead of 2 Blu-ray discs, but does that difference really have much consumer value?
And how the heck would you know that? The Blu-ray camp has made that assertion, but it simply isn't born out in real-world testing.
Last week, for a test, I put a 123 minute movie on a DVD-9 using MPEG-2, using the HD DVD format (via Apple's DVD Studio Pro 4). Average of around 8.5 Mbps. Looked pretty darn good at 1920x1080.
HD-DVD gives you 30 GB, and the use of H.264 and VC-1 for codecs. No problem AT ALL sticking "Return of the King Extended Edition" on a single side of HD-DVD. So using codecs that are 2x better and 3x more capacity, yeah, HD-DVD is just fine. Single layer HD-DVD will be fine for the vast majority of films, and even offers more minutes per disc at HD than DVD gives us minutes of SD today.
Actually, FireWire is the dominant protocol for compressed video applications, which is a huge market. And it's used widely for external storage as well.
Sure, it isn't used for keyboard and other areas where real-time performance isn't critical, but it doesn't have any advantages there.
Where isosynchronous performance is critical, FireWire does very well.
I do a lot of trade shows, and do a lot of software demos. I work with Mac software and Windows software. Until now, I've lugged a PowerBook, and run VirtualPC or Remote Desktop when I needed to show off Windows stuff. Which kinda worked, and was often better than having to lug two laptops around everywhere.
Now i can have a true dual-boot portable machine?
Or likely better yet, a native-code version of VirtualPC running on MacOS X that can give native full screen performance? So I'm always Apple-Tab away from MacWindows. VirtualPC is going to get a lot more interesting with this - possibly a reason why Microsoft is supporting this effort.
Life is good here for us digital media road warriors.
The point was that PowerBooks have been externally modern for some time. There's no practical difference in how the internal bits talk to each other.
No, PowerBooks dropped ADB in favor of USB back with the Bronze keyboard G3 - 1999? That generation still had SCSI, but the last pre-Titanium had FireWire and AirPort. All the G4 PowerBooks have been USB/1394/802.11.
:).
iBooks never had ADB - they were USB from their introduction, as was the iMac.
Apple machines were using today's standard Windows interfaces ports well before most Windows machines had them
I just launched iTunes 4.7, and was prompted to download 4.8.
Not via software update, but it's something.
Far from factious. Cell would be enormously good at video processing kinds of applications, and if it's cheap enough could cetainly compete with the ASIC, DSP, and (increasingly) FPGA systems that are used in the field today.
I'm sure a set top box vendor loves the idea of a chip that can decode compressed video, process it, and run a Java applet in the same piece of silicon.
Actually, the I-frame only is quite a bit easier to decode, since it doesn't require motion estimation, doesn't reference multiple frames, etcetera. It's almost identical to a series of JPEGs at 29.97 frames per second.
In normal IBBP MPEG-2, a given B-frame will reference frames either direction of it, requiring two frames be buffered in memory, and a lot of moving around of data from those two frames.
QuickTime 7 is required for H.264 playback in order to support B-frames. QuickTime's original architecture back in 1991 required that frames only reference previous frames, not future frames - a feature that is a basic requirement of H.264. Apple has been working on a major rewrite for years to get this working.
The good news is that out of order playback means better MPEG-1, MPEG-2, and MPEG-4 Part 2 implementations are also now technically possible within QuickTime.
Wow,
That file was from right around when I started doing compression, and it's an amazing testament to how much things have gotten better.
Video is Apple Video, codenamed "Road Pizza" pre-Cinepak. No data rate control. 5-bit per channel RGB.
Audio is uncompressed 8-bit. Pre IMA, and well pre MP3 for computer playback. Computers of that era weren't fast enough play back MP3 in real-time, let alone with video along with it!
File totals 300 KBytes/Sec, or 2400 Kbps. One could easily provide better quality at 100 Kbps today. 24x improvement in a little over a decade - not Moore's Law, but not bad!
Well, video professionals are still using the analog Betacam SP and Digital Betcam. They're based on the Betamax tape shell, but run at higher speeds and have much better image quality than Betamax did.
Sony Professional has certainly made enough profit on those formats to make up for the Betamax losses by now.
By your definition, MPEG-2 sucks since the first Highlander DVD was so bad.
Nope.
The only video codecs for which patent license free decoding is available that I know of are H.263, MPEG-1, and VP3/Theora.
Whomever wrote the original article is pretty fuzzy on a lot of details.
Just to clear things up a bit:
VC-9 was based on Windows Media Video 9, which is the commercial release version of the WMV codec, plus the Advanced Profile extensions. It was later renamed VC-1. No difference between the two.
H.264 and VC-1 do have significant technical differences (I go worried when he described his research on this point as "Another source told us recently that they had had the codec explained to them, and confirmed that it did "pretty much" the same as H.264." Well, pretty much in the same way that MPEG-1 and RealVideo do pretty much the same thing. They're both codecs, but have significant differences with real-world differences. For example, VC-1 uses larger blocks than H.264, which helps with some content and hurts with others. H.264 supports multiple reference frames, which can improve compression, but slows encode and decode.
Lastly, these issues aren't that unusual - I doubt it would even be possible to build a competitive codec without stepping on a whole lot of patents. Microsoft has IP in H.264, after all. It's still not possible to build a patent free MPEG-2 decoder.
Sure. That's what MovieLink and CinemaNow are all about.
And DVD quality is more than enough for most folks. If you can sustain 1.5 Mbps, you can get a great looking 480p.
Well, there may be room for heroic optimization yet, but there aren't any G4 machines that can decode 1080i MPEG-2 in real-time. Not a problem on my dual G5, certainly.
Maybe they're limited by the limited FSB on the G4?
Oh, modern codecs can do a LOT better than the MPEG-2 subset in DVD.
H.264, which will be supported on that box in QuickTime 7, is roughly 3x more efficient than MPEG-2. So DVD quality at around 2 Mbps. You could do great 1080 24p HD at around 8 Mbps with it. WMV9 Advanced Profile from Microsoft offers equivalent performance, although likely won't be on the miniMac in a high performance version for a while.
Not sure if the G4 will be able to do HD decode of H.264, though. Apple has only demoed full HD decode on dual G5. Maybe if they're able to do some video card offloading or something.
I think you're missing the point. HE AAC is more than twice as efficient as today's leading class of codecs (AAC, WMA, Ogg). Twice is a big deal! Think of the difference between, say, MPEG-2 and H.264 or WMV9 Advanced Profile. It took video codecs a full DECADE to get the kind of improvement jump we're getting with HE AAC. That 20 Kbps stream can be a great sounding 44.1 with HE AAC - better than that 64 Kbps VBR stream you cite.
The technology has been around for a while in enterprise systems, but is only now trickling down to desktop use.
And AAC PS (parametric stereo) is just around the corner, which is more efficient yet.
Yes, HE AAC and AAC+ are the same thing. HE AAC is the name that MPEG gives it, and AAC+ is Coding Technologies name for their implementation.
Next up is AAC PS, for parametric stereo, which applies the SBR techniques to synthesizing stereo. Gives another big leap yet for music listening - 24 Kbps is good enough for people who can live with MP3 @ 160 or so.
Heck, Apple doesn't even do ASP (Advanced Simple Profile). They're simple profile only, for both encode and playback. And their SP encoder is quite infamously weak.
Squeeze is a great Mac MPEG-4 encoder. The other major commercial tool for that on Mac is Compression Master, which is a much better encoder for SP, and does good ASP as well. And there's ffmpegX and DivX for Mac as well.
The formats are more recent, but there aren't any known patents with non-waived license fees covering Layer I and Layer II.
This isn't true for Layer III, which is why Layer III wasn't used in VCD, or anywhere really as a MPEG audio codec.
Actually, VCD uses only Layer II audio, not MP3. There aren't any controlling patents or licensing fees for MPEG-1.
Your general point is very apt, of course. Except for VCD, virtually all media technologies require various patent licensing, and in practice these haven't resulted in any company gaining undo control over the technology. It just means that makes of encoders, players, and/or content have to pay a fee to make the stuff. But the licensing contracts don't let a company revoke or re-negotiate the license after it has been launched.