While the technical achievement itself is impressive, what has been ignored in this review is the total amount of disk storage that is required to store JUST ONE HOUR of high-def 1080i video uncompressed DTV signal (you got it, uncompressed, they cant mpeg encode on the fly with these things yet, its too processor intensive and the specialized hardware is expensive) is 35 GIGABYTES.
So in other words, you need basically an ENTIRE STANDARD TIVO's worth of storage allocated for just one hour. So if you want to store any decent amount of programs, youre gonna need 700 gig just to do 20 hours of 1080i programming. Thats 4 200 gig ATA drives if you want 800 gig. Thats a helluva lot of heat to be generated in such a small chassis not to mention a huge expense.
Whats needed for mass adoption of this technology is terabyte sized hard disks, because having to create a contiguous file system using a RAID of 200 gig drives is rediuclously impractical and expensive to do. Not to mention prone to failure.
Can we get the guy who built this to build, oh, a dozen more of these and like a hundred bowling balls and set them up in front of a certain exec's office at a certain corporate headquarters in Orem?
This is Jason Perlow, formerly Software Manager and Developer Liason on the Zaurus here.
The 5600's memory architecture is different from the SL 5500's, but effectively the end user has the same about of avalaible storage and program execution memory as they did on the 5500. The reason is that all program storage now occurs on a JFFS2 partition on the FLASH, as opposed to running in a RAMDISK in the RAM. In the 5500, the RAM is partitioned into a 32MB ramdisk (MTDRAM) for program and 32MB of RAM for program execution. On the 5600, 32MB of the 64MB of flash is allocated to user storage on a JFFS2 partition (which is a compressed file system, so technically you can get more than the 32MB) and the 32MB RAM is used -exclusively- for program execution.
Why did they move user storage to Flash instead of running it in RAM? Simple. Battery dies, you don't lose your data.
On a related note, Sharp Electronics launched ZAURUS.COM today, which includes a Sourceforge 2.5 based custom implementation designed by Tony Guntharp, Tim Purdue's colleague in the original SourceForge project.
Mr. Iwanyk, 32 years old, first suspected that his TiVo thought he was gay, since it inexplicably kept recording programs with gay themes. A film studio executive in Los Angeles and the self-described "straightest guy on earth," he tried to tame TiVo's gay fixation by recording war movies and other "guy stuff."
"The problem was, I overcompensated," he says. "It started giving me documentaries on Joseph Goebbels and Adolf Eichmann. It stopped thinking I was gay and decided I was a crazy guy reminiscing about the Third Reich."
_______
What, doesn't he know anything about the Nazis? A whole bunch of them loved to dress up in women's clothes. That TIVO is brilliant.
Back in 1993-1994 when IBM was still working on the OS/2 for PowerPC and WorkPlace OS, and Taligent and "Pink" were still on the drawing board, IBM was planning to release the PowerPC 620 Series, a 64-bit version of the PowerPC 604. They intended to use it to run a 64-bit version of OS/2 that ran on the Mach kernel.
The design was scrapped because back then the manufacturing process was way too expensive to be cost effective in mass producing the chip. And we all know what happened to PowerPC OS/2.
Sure, nobody ever thought jetliners filled with aviation fuel would hit the towers. However, if anyone saw the recent NOVA "Why The Towers Fell" you would see that various shortcomings in the design of the building aided in the collapse.
http://www.pbs.org/wgbh/nova/wtc/
Engineer's report here:
http://www.pbs.org/wgbh/nova/wtc/collapse2.html
1) The walls made of sheet rock in the building's core completely shattered and exposed the metal beams when the planes hit, and thus the fires melted the core. if they were made of concrete, chances are the buildings might have been still standing.
2) use of spray on fire insulation on the metal beams. When the planes hit it shook the insulation off and exposed the bare beams, losing their fire protection.
3) Angle clips used in the floor trusses werent strong enough to hold the trusses up when the floors started to buckle due to the tremendous heat, causing the pancaking effect which caused the floors to collapse.
The question remains then, after Eazel dies as company, what happens to the code. The obvious answer is "well, the community, you dolt" but that doesnt really answer developmental issues.
Immediately of course, the GNOME project can host it, but as most of the developers worked at Eazel and are presumably moving on (perhaps to other Linux companies to contribute to Nautilus, hopefully) somebody should be fostering new development, as well as the hosting of the Eazel services.
It would seem that Ximian should be the heir apparent, since they already package Nautilus with their desktop, and they have a lot of GNOME developers, but bringing this project over may be putting too much on their plate.
My vote goes to Sun. Why Sun? So far, they are the one 3rd party UNIX company with the most interest in GNOME for desktop use with the release of OpenOffice and porting of GNOME to Solaris, and have people that know how to write for the desktop. Theres no danger of Sun going under either, and they have the cash to hire a good amount of the original developers. They also have the network infrastructure to host Eazel services. This would open up some interesting possibilities, such as the integration of Cobalts way-cool BlueLinQ technology with Nautilus.
A third possibility is to share the original Eazel developers between 4 of the top Unix companies, perhaps a split among Sun, IBM, and HP and Redhat, and have them throw a certain amount of money into a pile they've already presumably reserved for GNOME Foundation. This would ensure not one single company "takes control" over active development.
Slashdot Mirror
Here are the specs on the unit:
http://www.vssll.com/pvr921.html
It uses a 250 gig drive to store up to 25 hours of HDTV. However I beleive this is 25 hours of 720p or 480p, not 1080i.
The 35 gig per 1 hour of uncompressed 1080i figure I got from somone at SONY who's been working on similar technology.
While the technical achievement itself is impressive, what has been ignored in this review is the total amount of disk storage that is required to store JUST ONE HOUR of high-def 1080i video uncompressed DTV signal (you got it, uncompressed, they cant mpeg encode on the fly with these things yet, its too processor intensive and the specialized hardware is expensive) is 35 GIGABYTES.
So in other words, you need basically an ENTIRE STANDARD TIVO's worth of storage allocated for just one hour. So if you want to store any decent amount of programs, youre gonna need 700 gig just to do 20 hours of 1080i programming. Thats 4 200 gig ATA drives if you want 800 gig. Thats a helluva lot of heat to be generated in such a small chassis not to mention a huge expense.
Whats needed for mass adoption of this technology is terabyte sized hard disks, because having to create a contiguous file system using a RAID of 200 gig drives is rediuclously impractical and expensive to do. Not to mention prone to failure.
Can we get the guy who built this to build, oh, a dozen more of these and like a hundred bowling balls and set them up in front of a certain exec's office at a certain corporate headquarters in Orem?
We should all wear this T-shirt with pride: Smite Thy Enemies!
This is Jason Perlow, formerly Software Manager and Developer Liason on the Zaurus here.
The 5600's memory architecture is different from the SL 5500's, but effectively the end user has the same about of avalaible storage and program execution memory as they did on the 5500. The reason is that all program storage now occurs on a JFFS2 partition on the FLASH, as opposed to running in a RAMDISK in the RAM. In the 5500, the RAM is partitioned into a 32MB ramdisk (MTDRAM) for program and 32MB of RAM for program execution. On the 5600, 32MB of the 64MB of flash is allocated to user storage on a JFFS2 partition (which is a compressed file system, so technically you can get more than the 32MB) and the 32MB RAM is used -exclusively- for program execution.
Why did they move user storage to Flash instead of running it in RAM? Simple. Battery dies, you don't lose your data.
Yes. Its linux.
On a related note, Sharp Electronics launched ZAURUS.COM today, which includes a Sourceforge 2.5 based custom implementation designed by Tony Guntharp, Tim Purdue's colleague in the original SourceForge project.
Mr. Iwanyk, 32 years old, first suspected that his TiVo thought he was gay, since it inexplicably kept recording programs with gay themes. A film studio executive in Los Angeles and the self-described "straightest guy on earth," he tried to tame TiVo's gay fixation by recording war movies and other "guy stuff."
"The problem was, I overcompensated," he says. "It started giving me documentaries on Joseph Goebbels and Adolf Eichmann. It stopped thinking I was gay and decided I was a crazy guy reminiscing about the Third Reich."
_______
What, doesn't he know anything about the Nazis? A whole bunch of them loved to dress up in women's clothes. That TIVO is brilliant.
Its the C700, not the C300 for the japanese mini-laptop.
http://www.zauruszone.com/files/sl5600pics.zip
correction guys, its 64MB flash, 32MB RAM.
Get em while they last! Make sure to get to the booth EARLY, cause at JavaOne this year we had 3-4 hour lines going around the BLOCK at Moscone.
Back in 1993-1994 when IBM was still working on the OS/2 for PowerPC and WorkPlace OS, and Taligent and "Pink" were still on the drawing board, IBM was planning to release the PowerPC 620 Series, a 64-bit version of the PowerPC 604. They intended to use it to run a 64-bit version of OS/2 that ran on the Mach kernel.
The design was scrapped because back then the manufacturing process was way too expensive to be cost effective in mass producing the chip. And we all know what happened to PowerPC OS/2.
http://www.byte.com/art/9411/sec8/art5.htm
You should check out eGullet.com. Its sort of a Slashdot, but for food freaks. Its a free site and currently there is no advertising.
We frequesntly get food celebs like Tony Bourdain participating, as well as others who use pseudonyms.
We have boards about regional restaurants, wine, cooking, food news, and our user base is very international.
Sure, nobody ever thought jetliners filled with aviation fuel would hit the towers. However, if anyone saw the recent NOVA "Why The Towers Fell" you would see that various shortcomings in the design of the building aided in the collapse.
http://www.pbs.org/wgbh/nova/wtc/
Engineer's report here:
http://www.pbs.org/wgbh/nova/wtc/collapse2.html
1) The walls made of sheet rock in the building's core completely shattered and exposed the metal beams when the planes hit, and thus the fires melted the core. if they were made of concrete, chances are the buildings might have been still standing.
2) use of spray on fire insulation on the metal beams. When the planes hit it shook the insulation off and exposed the bare beams, losing their fire protection.
3) Angle clips used in the floor trusses werent strong enough to hold the trusses up when the floors started to buckle due to the tremendous heat, causing the pancaking effect which caused the floors to collapse.
2)
Representatives of the Familiar distribution have already stated that it WILL be ported to the Zaurus.
Jason Perlow
Moderator,
http://groups.yahoo.com/group/sharp-linux
The question remains then, after Eazel dies as company, what happens to the code. The obvious answer is "well, the community, you dolt" but that doesnt really answer developmental issues.
Immediately of course, the GNOME project can host it, but as most of the developers worked at Eazel and are presumably moving on (perhaps to other Linux companies to contribute to Nautilus, hopefully) somebody should be fostering new development, as well as the hosting of the Eazel services.
It would seem that Ximian should be the heir apparent, since they already package Nautilus with their desktop, and they have a lot of GNOME developers, but bringing this project over may be putting too much on their plate.
My vote goes to Sun. Why Sun? So far, they are the one 3rd party UNIX company with the most interest in GNOME for desktop use with the release of OpenOffice and porting of GNOME to Solaris, and have people that know how to write for the desktop. Theres no danger of Sun going under either, and they have the cash to hire a good amount of the original developers. They also have the network infrastructure to host Eazel services. This would open up some interesting possibilities, such as the integration of Cobalts way-cool BlueLinQ technology with Nautilus.
A third possibility is to share the original Eazel developers between 4 of the top Unix companies, perhaps a split among Sun, IBM, and HP and Redhat, and have them throw a certain amount of money into a pile they've already presumably reserved for GNOME Foundation. This would ensure not one single company "takes control" over active development.
Jason Perlow
Sr. Technology Editor
Linux Magazine