> It makes me wonder why the hell anyone would mess with PPTP or IPSec stuff, especially since NAT is almost everywhere these days.
I use IPSec because I can buy cheap wireless routers that have hardware accelerated IPSec, and IPSec clients are widely available (built into MacOS X, easily installable in Linux).
IPSec does work with NAT. IPSec AH (authenticate only, not encrypt) mode doesn't, but nobody uses that and many devices don't support it. IPSec ESP works fine through NAT.
>Last I learned in US Politics 101 is that CONGRESS passed the laws, the President signs them, and the > Judiciary reviews them. Yes, his opinion will have large sway with the President, but you gotta ram > crap through Congress first.
You haven't been paying attention the last six years have you? A Republican controlled congress, and a President who has never vetoed anything.. An administration _filled_ with lobbyists appointed to oversee industries they have been lobbying for..
Note that these conflicts of interest were almost unheard of in the Clinton administration, and were much less common in other administrations.
Politics 101 doesn't seem to apply to the Bush Regime. Presidential 'signing statements' which aim to modify congressional legislation? Congress abdicating their power to declare war? Decreasing the separation between church and state? Indefinite detentions, without legal process, evidence, or representation? Preemptive war based on lies? There are examples of each of these in previous administrations, but not all of these and more in from one president.
My first PC was a TI-99/4A, which came with a BASIC interpreter and a bunch of example programs in a book that encouraged me to try them out and modify them. I remember typing in their slots program and playing it with my cousins, and modifying it to get better results for myself.
The gap between what I could make in BASIC and what the professional programs offered was not huge. The same applied to my next system, an Apple ][e. Now, few kids are impressed by a console program. To get to something interesting to them, a GUI app or 3D graphics, requires a huge amount of experience.
The alternative that some mention, high level tools to create GUI apps or graphics ignores the fact that the person gets no experience in the fundamentals and no understanding of what's going on under the covers. Using those tools is more akin to configuring/using a program, rather than actual development.
I haven't finished sync'ing my video iPod yet, but after installing iTunes 7, it brought up a dialog as it went through my library saying it was updating for gapless playback.
This would imply that it's done within the audio files themselves, perhaps making it usable on all iPods.
I'll see shortly.
The new music view/organize modes in iTunes are great. One is divided by albums, showing a small image of the album cover along with all the tracks. The other looks like they licensed "Cover Flow", it looks like flipping through albums in your old physical collection. Definitely better visual feedback then scrolling through a huge text list of songs.
A real product marketing launch would include more information about how to install the product (in this case, it could include information appropriate for each user level ( guru: svn,configure,make,install; linux admin: rpm, apt-get, emerge, etc.; newbie: Upcoming KnoppMyth release)) with links to detailed information for each.
MythTV is pretty much the worst case scenario, because of all the variables involved (hardware receiver cards+drivers, cable TV systems, broadcast TV systems, guide data, etc). Especially when doing HD.. scanning for the channels and setting up guide data can be tricky. The install/setup can be made much easier, it will be very difficult to make it bulletproof.
But, I still think the original point is valid.. doing a real corporate marketing style product launch would be beneficial, and help to promote open source projects and avoid many of the common pitfalls.
Don't forget, the HD-Tivo has two cablecard slots, to enable recording on two HD tuners simultaneously. So, it will cost you somewhere between $10 and $20 per month for the two cable cards (depending on how much your local cable company charges).
With the Tivo service charge on top of that, the cable company DVR begins to look attractive again.
Luckily, Tivo announced that they would allow those of us that spent $300 on a lifetime service subscription to transfer it over to the new Tivo for *only* $200!! Gee, thanks! I spent $1,000 on the HD DirecTV Tivo, and $300 for lifetime service. Now, I can get a new system for only $800 for the box and $200 for the service?!? Sign me up, dude.
Honestly, I do really like the Tivo, and I would like to see the new series 3 (The DirecTV version is painfully slow, and doesn't include all the Tivo Series 2 features). But, I have a really hard time justifying the expense. I certainly won't be an early adopter for $800, and the Tivo service fees may keep me away even after the unit price drops.
It's doubtful that any open source application will be able to use cablecard because of their usage restriction requirements. If it is possible in the future, it would have to also include a bunch of restrictions to MythTV functionality (i.e. the data cannot be unencrypted on the system, so all the editing, transcoding, sharing type stuff is out of the question).
But, as it is today, my MythTV box works well with all my local channels via HD Cable. The one channel I would really like to be able to access is ESPN-HD, but it is encrypted.. so no luck.
Of course, Windows Media Center does not support CableCard either. In fact, WMCE doesn't even support QAM Cable HD. A future version is supposed to support cable/cablecard, but we'll see what kind of requirements/restrictions that brings along.
In any case, I don't think I'll be moving over to WMCE no matter what they offer. I can't see supporting that level of DRM, which effectively locks in Microsoft and the big studios into the positions of power over my hardware.. No thanks, I don't care enough about TV to go for that.
MythTV could really use a marketing guy to help with the new releases (actually, there are many open source projects that could benefit from this). The list of highly technical updates to MythTV don't really do justice to where MythTV is today.
As a MythTV user, here is what I see as important, and having improved in 0.20:
- MythTV is a free / open source PVR application, with support for analog, digital, and HDTV recording in most international standards (i.e. it's usable in the U.S., Europe, Asia, etc.). It includes many features not available in commercial PVR products.
- Automatic commercial detection and removal, or manual skip forward/back.
- Transcode of video to other formats/resolutions -- including DVD export in 0.20.
- Network based structure, allowing 'backend' recording storage on different machine than the 'frontend' display. (i.e. stick the backend with all the cable connections, antennas, loud fans and tons of disk in the basement, put a small/quiet frontend near your TV for output.)
- HDTV support: With supported HD capture card, terrestrial broadcast HD and Cable HD are supported (with the exception of encrypted cable HD channels - which cannot be decrypted on any PC PVR)
- Improved MacOS X support. The 0.20 version has greatly improved the Mac support, especially for the Intel based Macs. Performance optimizations for HD video playback make the Core Duo Mac Minis a great choice for a small/quiet frontend box.
It'll cost you.. but a nice upgrade would be an iMac. The form factor will be similar for him, and it will be a hell of a lot more powserful, while remaining simple to use.
I'm sure a larger screen with better resolution will be a welcome change, and he may even get into some of the new features, like digital picture management or music with iTunes, etc.
The older G4 iMac, which looks a bit like a lamp, would be okay and not too expensive. A newer G5 iMac is a lot like the iOpener (all in one monitor unit), but will cost more.
Or, if form factor is not as important, do like I did with my parents, set him up with a Mac Mini and an LCD monitor. It's still small and quiet, and an older G4 mini should be pretty cheap.
It's similar to ATI's MPEG2 acceleration feature, which has been in their hardware for over a decade. But, that feature is not usable in Linux, because ATI has never released driver source or hardware specs to enable this feature.
The same applies to a lot of the new features that the manufacturers hype.. all the video processing, MPEG4, and WMV accelation. They are all unavailable to Linux users.
If Intel delivers on their open source support claims, I will gladly build my MythTV frontend around an Intel integrated GPU, because nothing else available today offers any meaningful hardware offload.
Define "well". My analysis would be more like "XvMC can be made to work, and greatly decreases the CPU load; but many people have problems using it reliably". Take a look at the MythTV mailing lists, and see how many problem reports and questions there are regarding XvMC.
Also, note that XvMC doesn't address the advanced capabilities of these new video cards, like all the video processing, MPEG4+ acceleration, etc. Add to that the fact that only Nvidia has (closed/binary) driver support for XvMC (ATI cards are capable, but the hardware information was never released to enable support).
Good point.. By licensing the patent, they are legitimizing it. The net effect is that it creates a big barrier to entry for new competitors. This is easy to do when you're the dominant player already seeing tons of revenue. Paying millions in licensing is a non-starter for the small innovators looking to enter that market.
BeOS was sold to Palm a long time ago, and they were supposedly going to use that as their next-gen OS. What ever happened to that plan? How did they now move to Windows?
It's too bad the BeOS technology will just be lost inside Palm.. I'm sure there is little chance of them open sourcing the code instead of just letting it die.
An ethernet port gives all sorts of interesting possibilities.. I wonder what they have planned for this?
How about:
- Content sharing to other DirecTV devices. My primary TV is downstairs, I would like a small satellite HD playback unit that can access content on the main DVR (don't the Moxi DVRs do this?). - PC integration, access music and photos from my PC for display/playback through my HDTV+Stereo. - Basic internet access via unit, WWW browser for display on HDTV.. sometimes this could be handy. - Access guide data, firmware updates, PPV requests, and other DirecTV content via the Internet. This could eliminate the need for a phone connection to the unit. - Additional storage on network (NFS, SMB) drives. Store content encrypted.
Somehow, I doubt they will be as aggressive as I'm hoping.
Also, if you want full/stable support for all the features of a chipset, open source and open specs are needed.
Virtually all modern GPUs have video acceleration support, for MPEG2 and sometimes MPEG4 acceleration. But, in most cases, this is not available in Linux. With HD MPEG2/MPEG4/AVC playback being very CPU intensive, acceleration would be a big benefit for apps like MythTV, VLC, MPlayer, etc.
- ATI was the leader in MPEG2 acceleration, introducing it about 10 years ago. But, they have never released specs or drivers capable of using their hardware capabilities in Linux. - Nvidia has support for XvMC MPEG2 acceleration in their binary drivers. But, it is pretty problematic. Just check out the MythTV mailing lists for all the problem reports. Open source drivers would be much easier to debug. - VIA / S3 made announcements about Linux support, and even released some code. But, their support was incomplete, and developers have gone off on their own with the OpenChrome project to get decent drivers despite VIA. - Most manufacturers have added MPEG4/AVC acceleration in more recent chips. None of these capabilities are usable in Linux.
Intels chips have varying levels of video acceleration support. If they provide access to their video hardware, the Intel chipsets could be an excellent option for MythTV frontends.
I'll believe it when I see it.. Virtually all companies pay lip service to supporting Linux and Open Source. But, not many act on it in a meaningful way. ATI has made statements in the past about their love of Linux, but not much has come from it. We'll see if the AMD ownership makes a difference.
VIA/S3 also made comments about supporting Linux, and even released some source code. But, it turned out to be quite a fiasco.. their code was incomplete and/or had binary components.
This effort from Intel looks to be a lot more complete than the others. I definitely won't be buying any other brands of video cards in hopes of good Linux support..
If it were really 5-20fps, I could see people buying one of these. But, I'll bet the actually difference would be in the 1fps realm (i.e. somewhere in the noise where you really can't tell if it's having an effect).
As you point out, there big selling point is that they do UDP offload, but UDP is a much simpler protocol than TCP, so the offload potential is much less. Also - existing cards already do UDP checksum offload (such as the extremely popular Intel Gig-E NICs).
Then, when you consider that the game companies make the game communications bandwidth efficient, so the games are playable over the WAN, the value becomes even smaller.
I think the product is a hoax, just playing on those gamers who go to ridiculous extremes.
Maybe it's a proof of concept, or thought exercise, but I think there would be many hurdles against actually using this. Such as:
- Discerning keyboard delays vs. user typing delays. - Discerning keyboard delays vs. network latency variability. - Getting the user to connect to a remote host using a direct keyboard interface like telnet. The much more common WWW connections do not expose keyboard input speed, the input is sent as one big request (unless you run some java app, or possibly other active code in the browser). - Compromising the network connection or destination host to expose the keyboard traffic.
I'd say there are a whole lot of more likely exploits that are higher on my list of things to look out for.
Hmm.. Selling a format that I can't play in my car or import into my iPod. Treading over the same ground of DVD-Audio and HDCD, which failed miserably; but offering lower audio quality (standard compressed DVD audio).
I suppose an integrated camera could be useful in some situations.. Adding the GPS coordinates into the picture data would be cool. But, I think I would rather have GPS integrated into my current camera than pictures in my navigation system.
Some more useful integrations would be:
- Digital terrestrial radio: The conversion process is just starting, and almost noone has a digital radio. Adding it into the GPS should be easy, and with the storage capacity in the GPS you could record or timeshift radio. - Satellite Radio: I don't want too many devices mounted in my car.. why not add a Sirius or XM receiver to the single unit. - Wifi access. With all the hotspots available, access should be plentiful.. Updates of software, traffic, and maps could be done. It could also give basic www access, at about the level of a PDA. - Address book, notes, calendar: Much data is useful in the car.
My Garmin Nuvi has MP3 and Picture capabilities, as well as several travel apps. I probably won't upgrade until I can get some of the above features..
I am a Mac user, and not much of a gamer.. I do play the occasional game for 15 minutes when I'm bored, but that's about it.
I think it would be to Apple's benefit to improve gaming a bit on MacOS, but I don't think that trying to compete with real gaming platforms is a good battle to fight.
What they need to do is:
- Integrate software purchases into iTunes. ITMS is simple and ubiquitous.. expand the scope of the store to include software, and you could guarantee good sales for small developers.
- Concentrate on mini games, which would be fast to download and appeal to the casual gamer. Solitare card games, Tetris, etc.. License old arcade classics, like Pac Man, Galaga, Tempest, etc. Charge a few bucks per game and you'll get plenty of sales.
-- Maybe produce a couple more complex games, like a flight simulator, golf game, racing game, or something like that.
The available bandwidth on digital TV signals is pretty limited.. 19.3Mbps. That may seem like a lot, but when trying to do a 1080i HD broadcast, it becomes pretty precious. Especially when, like PBS, you are splitting the pipe into sub-channels and doing an SD broadcast along with the HD broadcast. The quality of the HD signal suffers, leading to macro blocking and loss of detail.
This already happens quite a bit on PBS signals I have seen, and carving out more of the pipe for data transmission of Disney movies will stress the limitations even more.
It's kind of odd that Disney partnered with PBS, and not ABC to do this.. Since ABC is owned by Disney and all. Maybe ABC cares more about their video quality than PBS, even though ABC uses the more efficient 720p broadcast which leaves more room for other data.
I set up a MythTV server using the "Breezy Badger" release not too long ago. It's still fairly clean, so I can fairly easily just wipe the system and install 6.06 from scratch. But, I'm wondering if this is really necessary.
What are the pros and cons of installing from scratch versus an 'apt-get dist-upgrade' on my existing box?
The WRV54G supports dynamic DNS, so it will automatically update the hostname when its IP address changes. I just use that hostname for VPN connections. Although, it really hasn't been an issue, as I don't think the IP address has ever changed.
Slashdot Mirror
> It makes me wonder why the hell anyone would mess with PPTP or IPSec stuff, especially since NAT is almost everywhere these days.
I use IPSec because I can buy cheap wireless routers that have hardware accelerated IPSec, and IPSec clients are widely available (built into MacOS X, easily installable in Linux).
IPSec does work with NAT. IPSec AH (authenticate only, not encrypt) mode doesn't, but nobody uses that and many devices don't support it. IPSec ESP works fine through NAT.
>Last I learned in US Politics 101 is that CONGRESS passed the laws, the President signs them, and the
> Judiciary reviews them. Yes, his opinion will have large sway with the President, but you gotta ram
> crap through Congress first.
You haven't been paying attention the last six years have you? A Republican controlled congress, and a President who has never vetoed anything.. An administration _filled_ with lobbyists appointed to oversee industries they have been lobbying for..
Note that these conflicts of interest were almost unheard of in the Clinton administration, and were much less common in other administrations.
Politics 101 doesn't seem to apply to the Bush Regime. Presidential 'signing statements' which aim to modify congressional legislation? Congress abdicating their power to declare war? Decreasing the separation between church and state? Indefinite detentions, without legal process, evidence, or representation? Preemptive war based on lies? There are examples of each of these in previous administrations, but not all of these and more in from one president.
I have used it from MacOS several times using FireFox. I'm not sure why it wouldn't work for you.
I don't use it often, the ads are way too annoying, and the interface is slow.
You're dead on..
My first PC was a TI-99/4A, which came with a BASIC interpreter and a bunch of example programs in a book that encouraged me to try them out and modify them. I remember typing in their slots program and playing it with my cousins, and modifying it to get better results for myself.
The gap between what I could make in BASIC and what the professional programs offered was not huge. The same applied to my next system, an Apple ][e. Now, few kids are impressed by a console program. To get to something interesting to them, a GUI app or 3D graphics, requires a huge amount of experience.
The alternative that some mention, high level tools to create GUI apps or graphics ignores the fact that the person gets no experience in the fundamentals and no understanding of what's going on under the covers. Using those tools is more akin to configuring/using a program, rather than actual development.
I haven't finished sync'ing my video iPod yet, but after installing iTunes 7, it brought up a dialog as it went through my library saying it was updating for gapless playback.
This would imply that it's done within the audio files themselves, perhaps making it usable on all iPods.
I'll see shortly.
The new music view/organize modes in iTunes are great. One is divided by albums, showing a small image of the album cover along with all the tracks. The other looks like they licensed "Cover Flow", it looks like flipping through albums in your old physical collection. Definitely better visual feedback then scrolling through a huge text list of songs.
A real product marketing launch would include more information about how to install the product (in this case, it could include information appropriate for each user level ( guru: svn,configure,make,install; linux admin: rpm, apt-get, emerge, etc.; newbie: Upcoming KnoppMyth release)) with links to detailed information for each.
MythTV is pretty much the worst case scenario, because of all the variables involved (hardware receiver cards+drivers, cable TV systems, broadcast TV systems, guide data, etc). Especially when doing HD.. scanning for the channels and setting up guide data can be tricky. The install/setup can be made much easier, it will be very difficult to make it bulletproof.
But, I still think the original point is valid.. doing a real corporate marketing style product launch would be beneficial, and help to promote open source projects and avoid many of the common pitfalls.
Don't forget, the HD-Tivo has two cablecard slots, to enable recording on two HD tuners simultaneously. So, it will cost you somewhere between $10 and $20 per month for the two cable cards (depending on how much your local cable company charges).
With the Tivo service charge on top of that, the cable company DVR begins to look attractive again.
Luckily, Tivo announced that they would allow those of us that spent $300 on a lifetime service subscription to transfer it over to the new Tivo for *only* $200!! Gee, thanks! I spent $1,000 on the HD DirecTV Tivo, and $300 for lifetime service. Now, I can get a new system for only $800 for the box and $200 for the service?!? Sign me up, dude.
Honestly, I do really like the Tivo, and I would like to see the new series 3 (The DirecTV version is painfully slow, and doesn't include all the Tivo Series 2 features). But, I have a really hard time justifying the expense. I certainly won't be an early adopter for $800, and the Tivo service fees may keep me away even after the unit price drops.
It's doubtful that any open source application will be able to use cablecard because of their usage restriction requirements. If it is possible in the future, it would have to also include a bunch of restrictions to MythTV functionality (i.e. the data cannot be unencrypted on the system, so all the editing, transcoding, sharing type stuff is out of the question).
But, as it is today, my MythTV box works well with all my local channels via HD Cable. The one channel I would really like to be able to access is ESPN-HD, but it is encrypted.. so no luck.
Of course, Windows Media Center does not support CableCard either. In fact, WMCE doesn't even support QAM Cable HD. A future version is supposed to support cable/cablecard, but we'll see what kind of requirements/restrictions that brings along.
In any case, I don't think I'll be moving over to WMCE no matter what they offer. I can't see supporting that level of DRM, which effectively locks in Microsoft and the big studios into the positions of power over my hardware.. No thanks, I don't care enough about TV to go for that.
MythTV could really use a marketing guy to help with the new releases (actually, there are many open source projects that could benefit from this). The list of highly technical updates to MythTV don't really do justice to where MythTV is today.
As a MythTV user, here is what I see as important, and having improved in 0.20:
- MythTV is a free / open source PVR application, with support for analog, digital, and HDTV recording in most international standards (i.e. it's usable in the U.S., Europe, Asia, etc.). It includes many features not available in commercial PVR products.
- Automatic commercial detection and removal, or manual skip forward/back.
- Transcode of video to other formats/resolutions -- including DVD export in 0.20.
- Network based structure, allowing 'backend' recording storage on different machine than the 'frontend' display. (i.e. stick the backend with all the cable connections, antennas, loud fans and tons of disk in the basement, put a small/quiet frontend near your TV for output.)
- HDTV support: With supported HD capture card, terrestrial broadcast HD and Cable HD are supported (with the exception of encrypted cable HD channels - which cannot be decrypted on any PC PVR)
- Improved MacOS X support. The 0.20 version has greatly improved the Mac support, especially for the Intel based Macs. Performance optimizations for HD video playback make the Core Duo Mac Minis a great choice for a small/quiet frontend box.
It'll cost you.. but a nice upgrade would be an iMac. The form factor will be similar for him, and it will be a hell of a lot more powserful, while remaining simple to use.
I'm sure a larger screen with better resolution will be a welcome change, and he may even get into some of the new features, like digital picture management or music with iTunes, etc.
The older G4 iMac, which looks a bit like a lamp, would be okay and not too expensive. A newer G5 iMac is a lot like the iOpener (all in one monitor unit), but will cost more.
Or, if form factor is not as important, do like I did with my parents, set him up with a Mac Mini and an LCD monitor. It's still small and quiet, and an older G4 mini should be pretty cheap.
No, it's not.
It's similar to ATI's MPEG2 acceleration feature, which has been in their hardware for over a decade. But, that feature is not usable in Linux, because ATI has never released driver source or hardware specs to enable this feature.
The same applies to a lot of the new features that the manufacturers hype.. all the video processing, MPEG4, and WMV accelation. They are all unavailable to Linux users.
If Intel delivers on their open source support claims, I will gladly build my MythTV frontend around an Intel integrated GPU, because nothing else available today offers any meaningful hardware offload.
Define "well". My analysis would be more like "XvMC can be made to work, and greatly decreases the CPU load; but many people have problems using it reliably". Take a look at the MythTV mailing lists, and see how many problem reports and questions there are regarding XvMC.
Also, note that XvMC doesn't address the advanced capabilities of these new video cards, like all the video processing, MPEG4+ acceleration, etc. Add to that the fact that only Nvidia has (closed/binary) driver support for XvMC (ATI cards are capable, but the hardware information was never released to enable support).
Good point.. By licensing the patent, they are legitimizing it. The net effect is that it creates a big barrier to entry for new competitors. This is easy to do when you're the dominant player already seeing tons of revenue. Paying millions in licensing is a non-starter for the small innovators looking to enter that market.
BeOS was sold to Palm a long time ago, and they were supposedly going to use that as their next-gen OS. What ever happened to that plan? How did they now move to Windows?
It's too bad the BeOS technology will just be lost inside Palm.. I'm sure there is little chance of them open sourcing the code instead of just letting it die.
An ethernet port gives all sorts of interesting possibilities.. I wonder what they have planned for this?
How about:
- Content sharing to other DirecTV devices. My primary TV is downstairs, I would like a small satellite HD playback unit that can access content on the main DVR (don't the Moxi DVRs do this?).
- PC integration, access music and photos from my PC for display/playback through my HDTV+Stereo.
- Basic internet access via unit, WWW browser for display on HDTV.. sometimes this could be handy.
- Access guide data, firmware updates, PPV requests, and other DirecTV content via the Internet. This could eliminate the need for a phone connection to the unit.
- Additional storage on network (NFS, SMB) drives. Store content encrypted.
Somehow, I doubt they will be as aggressive as I'm hoping.
Also, if you want full/stable support for all the features of a chipset, open source and open specs are needed.
Virtually all modern GPUs have video acceleration support, for MPEG2 and sometimes MPEG4 acceleration. But, in most cases, this is not available in Linux. With HD MPEG2/MPEG4/AVC playback being very CPU intensive, acceleration would be a big benefit for apps like MythTV, VLC, MPlayer, etc.
- ATI was the leader in MPEG2 acceleration, introducing it about 10 years ago. But, they have never released specs or drivers capable of using their hardware capabilities in Linux.
- Nvidia has support for XvMC MPEG2 acceleration in their binary drivers. But, it is pretty problematic. Just check out the MythTV mailing lists for all the problem reports. Open source drivers would be much easier to debug.
- VIA / S3 made announcements about Linux support, and even released some code. But, their support was incomplete, and developers have gone off on their own with the OpenChrome project to get decent drivers despite VIA.
- Most manufacturers have added MPEG4/AVC acceleration in more recent chips. None of these capabilities are usable in Linux.
Intels chips have varying levels of video acceleration support. If they provide access to their video hardware, the Intel chipsets could be an excellent option for MythTV frontends.
I'll believe it when I see it.. Virtually all companies pay lip service to supporting Linux and Open Source. But, not many act on it in a meaningful way. ATI has made statements in the past about their love of Linux, but not much has come from it. We'll see if the AMD ownership makes a difference.
VIA/S3 also made comments about supporting Linux, and even released some source code. But, it turned out to be quite a fiasco.. their code was incomplete and/or had binary components.
This effort from Intel looks to be a lot more complete than the others. I definitely won't be buying any other brands of video cards in hopes of good Linux support..
If it were really 5-20fps, I could see people buying one of these. But, I'll bet the actually difference would be in the 1fps realm (i.e. somewhere in the noise where you really can't tell if it's having an effect).
As you point out, there big selling point is that they do UDP offload, but UDP is a much simpler protocol than TCP, so the offload potential is much less. Also - existing cards already do UDP checksum offload (such as the extremely popular Intel Gig-E NICs).
Then, when you consider that the game companies make the game communications bandwidth efficient, so the games are playable over the WAN, the value becomes even smaller.
I think the product is a hoax, just playing on those gamers who go to ridiculous extremes.
Maybe it's a proof of concept, or thought exercise, but I think there would be many hurdles against actually using this. Such as:
- Discerning keyboard delays vs. user typing delays.
- Discerning keyboard delays vs. network latency variability.
- Getting the user to connect to a remote host using a direct keyboard interface like telnet. The much more common WWW connections do not expose keyboard input speed, the input is sent as one big request (unless you run some java app, or possibly other active code in the browser).
- Compromising the network connection or destination host to expose the keyboard traffic.
I'd say there are a whole lot of more likely exploits that are higher on my list of things to look out for.
Hmm.. Selling a format that I can't play in my car or import into my iPod. Treading over the same ground of DVD-Audio and HDCD, which failed miserably; but offering lower audio quality (standard compressed DVD audio).
Should be a smashing success.
I suppose an integrated camera could be useful in some situations.. Adding the GPS coordinates into the picture data would be cool. But, I think I would rather have GPS integrated into my current camera than pictures in my navigation system.
Some more useful integrations would be:
- Digital terrestrial radio: The conversion process is just starting, and almost noone has a digital radio. Adding it into the GPS should be easy, and with the storage capacity in the GPS you could record or timeshift radio.
- Satellite Radio: I don't want too many devices mounted in my car.. why not add a Sirius or XM receiver to the single unit.
- Wifi access. With all the hotspots available, access should be plentiful.. Updates of software, traffic, and maps could be done. It could also give basic www access, at about the level of a PDA.
- Address book, notes, calendar: Much data is useful in the car.
My Garmin Nuvi has MP3 and Picture capabilities, as well as several travel apps. I probably won't upgrade until I can get some of the above features..
I am a Mac user, and not much of a gamer.. I do play the occasional game for 15 minutes when I'm bored, but that's about it.
I think it would be to Apple's benefit to improve gaming a bit on MacOS, but I don't think that trying to compete with real gaming platforms is a good battle to fight.
What they need to do is:
- Integrate software purchases into iTunes. ITMS is simple and ubiquitous.. expand the scope of the store to include software, and you could guarantee good sales for small developers.
- Concentrate on mini games, which would be fast to download and appeal to the casual gamer. Solitare card games, Tetris, etc.. License old arcade classics, like Pac Man, Galaga, Tempest, etc. Charge a few bucks per game and you'll get plenty of sales.
-- Maybe produce a couple more complex games, like a flight simulator, golf game, racing game, or something like that.
The available bandwidth on digital TV signals is pretty limited.. 19.3Mbps. That may seem like a lot, but when trying to do a 1080i HD broadcast, it becomes pretty precious. Especially when, like PBS, you are splitting the pipe into sub-channels and doing an SD broadcast along with the HD broadcast. The quality of the HD signal suffers, leading to macro blocking and loss of detail.
This already happens quite a bit on PBS signals I have seen, and carving out more of the pipe for data transmission of Disney movies will stress the limitations even more.
It's kind of odd that Disney partnered with PBS, and not ABC to do this.. Since ABC is owned by Disney and all. Maybe ABC cares more about their video quality than PBS, even though ABC uses the more efficient 720p broadcast which leaves more room for other data.
I set up a MythTV server using the "Breezy Badger" release not too long ago. It's still fairly clean, so I can fairly easily just wipe the system and install 6.06 from scratch. But, I'm wondering if this is really necessary.
What are the pros and cons of installing from scratch versus an 'apt-get dist-upgrade' on my existing box?
The WRV54G supports dynamic DNS, so it will automatically update the hostname when its IP address changes. I just use that hostname for VPN connections. Although, it really hasn't been an issue, as I don't think the IP address has ever changed.
The Check Point devices also support Dynamic DNS.