Well, if you had provided that information in the first place, I wouldn't continue to question the existance of a low power processor and video H/W from other than Via.
In particular, your mention of (AMD) Athlon processors as low-power left me scratching my head -- I was not aware that AMD had low power processors to support fanless designs until you provided that reference. Also, mention of Intel CPUs (likely mobile ones) that you think would be adequate means nothing to me without some convincing that they are capable of rendering 1080i in real time with or without spcialised MEPG2 accelleration hardware (other formats would be nice to support, but not essential for my purposes, not even MPEG4).
But even that processor list (which I could have googled myself) is not what I was looking for when designing a system. I was not interested in having to design and build a mobo around a processor. I wanted to purchase a fairly well integrated mobo, convincingly capable of rendering HD MPEG2 video in real time, convection cooled, in as small a form factor as possible with a strong preference for open source software.
The nanoITX fit the bill, even as it took forever to become available (heck, mine doesn't even have an integrated coldLuke processor/northbridge -- it has separate Eden CPU and CN400 northbridge) and dissapointingly requires (very modest) active cooling for the 1.0 GHz version.
There was a Commell mini-ITX board that employed the CN400, and X CLE266 driver extention development for it to support HD resolutions, which led me down this path (waiting for the supposedly fanless Eden processor in a nanoITX form factor). The only trouble was that the Commell board required noisy active cooling (I was not about to engineer a heatpipe/enclosure solution from scratch, either).
The alternative was a HushPC or Serener enclosure, using heatpipe technology with a conventional processor, but those would have been even more expensive than what I eneded up building (Eden C3 with CN400, 512 MB RAM in a Silverstone Lascala LC08 with slimline DVD/RW drive and 500 GB hard drive) at a cost of around $1000.
Perhaps I erred in thinking that S/W decoding of HD MPEG2 video required a 2.4 GHx P4 (and the associated cooling and noise), but there was nothing I found that convinced me otherwise. H/W MPEG2 accellerators on common graphics cards (which already started to break the requirement of a small form factor) generally didn't have open source drivers.
The only item that showed promise, after a process of elimination, was the Via Eden C3 with CN400 northbridge in a nice nanoITX form factor.
Frankly, I really like the small form factor, and the existing problems strike me as possible to overcome, particularly since Via has their own xine and mplayer for this platform that leverage the CN400 (albeit without completely open software).
And the $1000 drops way down to around $600 if I eliminate the DVD/RW drive and hard disk, and simply use it as a MythTV frontend (which would fit in an even smaller Silverstone Lascala LC07 case). It's just that I like to occasionally pop in a DVD into the playback device in the media room and not have to go to the server rack, and having local content (not to mention support native development if I care to hack) is nice as well.
My 1.66GHz Athlon stays under 50% load when playing back 1080i MPEG-2
Good lord, that must be loud as heck, with the CPU and PS fan it requres! At least, if it is anything like a similar Athlon tower I have. What silent (and by silent I mean fanless) system do you have that can do 1080i? I *think* an 800 Mhz Via Luke with a CN400 using openchrome should be able to manage, convenction cooled. The 1.0GHz nanoITX board I have is spec'd to use a fansink (albeit putting out 14 dbA noise), so it doesn't quite cut it either.
Still, I'm impressed that your Athlon can do 1080i60 (or 1080i59.97) in real time at all.
Yes, that was pretty much the point of my post. VIA's CPUs are painfully slow
Compared to a power-hungry monster, yes, certainly for the same clock speed. That's exactly why one turns to specialized hardware for MPEG-2 decoding, when using low-power (heat and computational) processors.
I spent an hour or two getting the driver working under the FC4 2.6.15 update, and haven't had an problems so far. Mostly just handling obsoleted functions.
How, if it is binary only?
I can only imagine that perhaps it relies on deprecated or changed kernel or ioctl calls which you've forward ported to 2.6.15.
I might be interested. I guess they bypass xvmc completely then, which would mean there is no dependency on X either. Of course, I haven't looked at the Via versions so I might be guessing just plain wrong.
The opnly thing I seen on the nanoITX are occasional recoverable DMA timeouts on disk transfers. But, I have not ruled out the drive.
While I have not been intentionally stressing hd and eth at the same time, I have been doing builds across the 'net to NFS-mounted directories while playing video.
I think you lost me. With a reasonbly fast CPU, decoding in software (MPlayer, Xine, VLC, ffplay, etc) is FASTER than XVMC, including higher resolutions like 1080.
Hmm.
Last time I checked, one needed around a 2.5 GHz P4 to decode 1080i MPEG2 in software.
On my nanoITX mobo, xine -V xvmc runs rings around stock xine.
Re:HD Myth on a Via nano-ITX with CN400
on
MythTV 0.19 Released
·
· Score: 2, Interesting
Sour grapes, perhaps?
Many ATI/NVidia/Intel videocards can do hardware MPEG-1/2 decoding. I have such a card, and I NEVER use it. You can't do any postprocessing, deinterlacing, inverse telecine, noise removal, etc.
Hmm. xine -xvmc and deinterlacing works just fine. The openchrome drivers apparently route the decoded video back for further processing, rather than just feed the chip's display engine, though it wouldn't surprise me if this were possible.
I would STRONGLY recomend staying the hell away from VIA. I made that mistake once, myself. You can expect better performance from an AMD/Intel chip clocked at about half what a VIA chip is, and the AMD/Intel chips will be lower power as well.
I have not seen serious problems (like the infamous DMA problem) with the nano-ITX. It isn't cheap, at around $400 for the 1.0 GHz version mobo, but I was looking for a fanless (or close to it) small form-factor. mini-ITX and ATX boards would not fit the bill.
Find yourself a LV PIII-1.0GHz (12.1W max), an ULV Pentium M-753 1.2GHz (5.5W TDP), an Athlon 4-1.2G (25W max), an Opteron-840EE 1.4GHz (30W TDP), or a mTurion MT-34 1.8GHz (25W TDP) if you're dedicated to ultra-low power, fanless CPUs.
And for H/W MPEG2 decoding to HD resolutions...?
Personally, I'm perfectly happy with a cheaper system, and a few quiet, variable 80mm fans.
Smaller, and fanless, is the goal for me. The 1.0 GHz system doesn't quite get there (it has one fan around 14 dBa which really is whisper quiet), but an 800 Mhz system probably would.
Also, I've read (but not verified) that the via drivers are only supported on older kernels. So, I've been following openchrome with interest.
Finally, I haven't ruled out other issues related to the dropped frames -- remember I'm running a full FC3 right now, with all the servers that are installed by default. Clearly, that needs to change. Until it does, I'm not to harsh on the video drivers for the odd dropped frame (usb traffic, in particular, appears to muck things up).
HD Myth on a Via nano-ITX with CN400
on
MythTV 0.19 Released
·
· Score: 4, Interesting
I've ben putting together a mythfrontend on a Via nano-ITX board in a Silverstone LC08 case.
What attracted me to this platform was the CN400 H/W MPEG2 decoder chip it includes that is capable of deciding HD MPEG2 resolutions (up to 1080i) -- xine plays 1080i on this platform with the 1.0 GHz CPU about 30% idle.
Of course, this is fairly bleeding edge, and there are the occasional dropped frames. Support for the CN400 comes from the openchrome project, which also supports dri/drm, and xine hooks for the resulting xxmc accelleration that takes advantage of the CN400.
It isn't quite fanless -- there is a processor fansink that puts out around 14 dbA. I'm told the 800 Mhz version of the same mobo is fanless, and once I get this stable, will likely spend the $$$ to try one.
The terms of service are basically "it's your bandwidth, do whatever you want with it as long as it's legal". I have an email server, ftp server, etc.
Yeah, "me too".
I also allow remote access via ssh. I do believe that Blarg frowns on excessive upstream bandwidth -- they'd rather host your public sites, but as long as it "isn't a problem", they're cool about any port you open. They gave me the impression that "we'll let you know nicely" if ever there were a problem.
The biggest concern they might have (understandably) would be over port 25 and open email relays. I took care to make sure I wasn't inadvertently running one.
In my dealings with Blarg, I got the impression that this really wasn't a problem: since their customers were paying far more than they had to to access "that there Intarweb", they were generally responsible about what they were doing with the benefits they had.
Recently, Blarg must've struck a deal with Verizon because I now have the option of paying Blarg for my side of the ATM pipe to them instead of Verizon. Obviously it is more convenient for Verizon to deal with Blarg in bulk than people like me for "Advanced Data Services". I took Blarg up on it: it actually saves me about $5 or $10 a month that way! And, supposedly, they make a few pennies on the deal as well that they didn't before. Win, win, win, all around.
FWIW, I selected Blarg on the basis of their sane TOS, most of all.
I would expect that the IRS (in the U.S. anyway) would treat it as transactions in a foreign currency.
Unless you take special actions re foreign currency transactions (and generally individuals can't), you can't just lump them together and be taxed on the net effect. This generally means that in-game profits are taxable to an individual as they occur.
I could get crippled internet service from Verizon for some odd $30 a month.
Instead, I pay about $60 a month for much better TOS (static IP, servers within reason, etc.), though Verizon still provides the pipe.
See, my DSL connection to Verizon's CO get's shunted onto the ATM fabric and shows up as a PVC to my ISP. They in-turn pay Verizon big $$$ for the fat pipes to whereever.
Now, Verizon charges for both the ATM PVC I use and the fat pipe my ISP uses. They get paid coming and going.
Verizon is free to charge my ISP (Blarg! in case anyone cares) whatever they want for the fat pipes and the DSL PVCs that get resold to me, at least whatever the market will bear. In fact, they price things so that, compared to their ISP service, they get more money from the likes of me. Used to be, I had to pay Verizon directly for the PVC -- "Advanced Data Services" they called it, but they found out that it was easier to sell those in bulk and let the ISP nickel and dime them out (which put more $$$ in my ISP's hands as well as reducing my overall bill).
So, what's the problem here? Verizon can price their pipes at whatever level the market will bear.
I suspect the real issue is that Verizon does not realize that it is competing against itself: their ISP division has to compete with all the other ISPs for bandwidth on it's own network. So what? They get paid either way. But, from the Verizon ISP perspective, they cry foul that so much more money is made by selling bandwidth to and through third-party ISPs and not them -- one division loses while the other division gains.
Note to Verizon: if it is more profitable to lease bandwidth to ISPs than it is to be one, get out of the ISP business!
I remember when doing speech recognition research, we'd refer to the schwa as the "garbage vowel", the catch-all phoneme for all miscellaneous vocalizations.
Actaully, it's the "hearing something other than French, even when not spoken to them" pissing them off tha that pisses me off. What part of "Je ne parle pas avec vous" (I was not talking to you) don't you understand?
Even that, I can accept: one finds a minority of A**holes everywhere.
But, to try to make it law? And have a court strike that down? And *still* be able to try to make it law? That's just whack.
Watch what happens if I put up a "For Sale" sign in front of my house, instead of "A Vendre".
The same arguments have been applied against spoken speech in public streets, around schools, etc. And, while not having the same Bill 101 force of law, the "offensive" argument is often made.
"Canada has fully public healthcare because there isn't much demand for private alternatives."
Then why have people taken the case for it to be made legal to purchase health care privately all the way to the Supreme Court of Canada? Sounds like pretty stubborn and persistent demand to me.
The Court agreed that preventing this was a violation of rights and freedoms, but stayed it's decision depending on whether Quebec would use the Notwithstanding Clause? Only to have the La Belle Province (Quebec) say it would use the Notwithstanding Clause to overrule the Court?
See, us Canadians, and Quebecois (for those who make a distinction, and there's a valid argument to making one, at several levels), are both a "tolerant" society. We do not take kindly to "offensive" remarks and practices.
In Quebec, for example, it is often considered "offensive" to speak English on a public street, and represents intolerance of the pure laine Quebecois majority.
Similarly, it is "offensive" for Ubisoft to engage in a practice that shows intolerance for EA's desires, EA being the larger, "majority" employer.
Standard Orwellean doublespeak - nothing to see.
Of course, I viewed this attitide as disgustingly fascist (both in Quebec, *and* the rest of Canada), and left. YMMV
Realize that you're talking about a place where the government can overrule the highest court in the land, and routinely does so ("notwithstanding clause"). Democracy unchecked by law is mob rule. Some might argue that Canada sucks less than the U.S. under Bush *in practice*, but it can suck far, far worse, in *principle*. Another example: Canada models it's socialized health care system after Cuba and North Korea. Comparing it *constitutionally* to Nazi Germany doesn't strike me as much of a stretch -- we just haven't found our Hitler yet. (Yeah, yeah, Godwin can shove it).
Again, your insights re. negligence not coming into matters of copyright demonstrates the ignorance of the engineer in the presense of the lawyer.
Now it is the engineer's turn:
Incidentally, you don't have the ability to authorize players, and inserting DVDs into them isn't how one would do so in any event.
Oh! I meant in the technical sense, not the legal sense. Such a player would likely not be authorized in the legal sense. But, the idea is that, technically, possession of the disk enables decryption of contentied previously copied from the disk, for some interval of time.
(especially since an image of the DVD would include any material such as keys that were on it).
There *might be* an area on the DVD that the drive can read, but will not pass to the computer as the drive reads it. This *might be* the disk key. It *might be* used to decrypt "title" keys, which are then used to decrypt "titles".
Said *disk key* might only be transferred to the computer in an encrypted fashion, after the computer enters into an exchange whereby it response to a challenge by the drive (an *authentication* phase). This lets two "secure" components (computer and drive) communicate securely over a bus that can be otherwise monitored.
The raw data on the disk that can be returned with a blind "copy" using normal file copy functions would not contain this supposed *disk key*. A special exchange with the drive is required (perhaps via an ioctl mechanism in a *nix device driver).
Of course, CSS has been broken so badly that such an exchange is no longer required, and it is merely a matter of massive amounts of computer power in short order to deduce such a *disk key* from the content itself.
Who actually bothers to rip a DVD with CSS intact, and plays it on an authorized player that handles the decryption? Large-scale pirates, I suppose, but that's not you.
Well, I actually would consider making copies of a DVD with CSS intact on a home file server, to be decrypted by thin clients in the home authorized to do so by way of my having inserted the DVD in them at some point (so the keys could be obtained).
While questionable on legal grounds (the use of an unauthorized player in the client), I do believe that I have a resposibility to protect the copyright holder's property -- to leave copies on a home lan unencrypted might be construed as negligent, rather like not locking up a firearm. This is particularly true if external access to machines on that lan is possible, albeit through an encrypted VPM and firewall.
Of course, ripping a DVD with a consumer DVDROM drive does not obtain all the keys necessary for CSS in the manner in which it is intended to be applied -- a key exchange with the DVDROM drive is necessary for that. Clients having entered into such a key exchange at one time can cache the keys for later decryption of the content served to them.
Now, strictly speaking, CSS can be circumvented without having those keys because they are easy to brute force from the raw content itself.
Anyway, leaving CSS intact and not decrypting CSS with an unauthorized program is not circumvention. Copyright infringement perhaps, but they're not the same thing.
I can't see it being copyright infringement if there is no distribution, but the argument could be made, yes.
FWIW, I do not pirate content. I purchase every CD and DVD I wish to have a copy of. But, the reach of the DMCA in preventing me from managing my use of that content in a convenient manner is utterly maddening.
The notion is that CSS applies to the medium and not the content: it protects the content on that medium.
"And that's your fatal mistake. A CSS encrypted movie on a hard disk is no less protected by a TPM, and eligible for the protections of section 1201 than one on a DVD. The medium is irrelevant.
And please, don't try to make clever arguments. They will not only fail, but they will make you look bad."
Well, then, the MPAA doesn't have a leg to stand on either, do they? That was my whole point: the only defense they appear to have is that they did not defeat CSS on the original, opening up the same defense for the likes of me. Which, as you correctly point out, might seam novel and clever, but no doubt anoying to a judge -- the intent was clearly to defeat CSS, by hook or by crook.
This presumes, of course, that the DVD was encrypted with CSS.
But, even if it wasn't, the DVD was provided on condition that the content not be redistributed, in effect a use license, no? IANAL, but I would think that fair use rights *can* be surrendered by way of such a license.
Let's say they can't. The question then boils down to whether the duplication was a fair use. I'm not sure that it clearly is: it enables parallel examination of the work in a commercial setting by many people, clearly depriving the author of revenue from selling additional copies for that purpose (the benefit to the MPAA is the efficiency obtained by way of parallel, instead of serial, examination).
So, the MPAA derives a financial gain from the copies. The author is deprived of revenue. The copies are not used in an educational setting. The copies are used in whole, and not in part. The copies *might* be used as a criticism of the work, but here, are complete copies necessary?
Like I said, IANAL, but even if the DVD was not protected with CSS, the use does not strike me as fair.
MPAA may have here a good claim to fair use, since their interest in the film is in looking into the privacy of their employees, rather than making copies as a substitute for getting them from the Best Buy
Sure, it looks like the MPAA has a legitimate fair use argument, but the DMCA trumps fair use, no?
Either they ripped the DVD, cracking CSS, and violating the DMCA; or they duplicated the DVD, and used an ordinary player to play those duplicates back. That use of a secondary player to access the work would be a DMCA violation, no?
If the latter case is NOT a DMCA violation, then surely my copying a DVD that I purchased to a hard disk, without defeating CSS, and then playing it back from that hard disk, defeating CSS on the copy wouldn't be either, no? (It may be a patent violation for several reasons, not the least of which might be Dolby Digital decoding in software, but that's a different issue).
The point is that if a straight rip is fair use, then what I do whith that rip as opposed to the original, including defeating CSS on it for personal viewing strikes me as fair use as well. The notion is that CSS applies to the medium and not the content: it protects the content on that medium. This is butressed by the fact that there are hidden areas on the DVD medium that support CSS that can not be copied by an ordinary DVDROM drive -- they are not part of the content ripped.
At the other end of the spectrum, I FINALLY got a compatible LC08 case from Silverstone (it was a show demo unit I begged and pleaded to purchase -- lead time for "real" ones is about 4 weeks) for my nanoITX mobo, and put it together last night. Took lots of pictures (including showing the diffs between the old and new LC08 case), and plan to write something up and post it online.
FC3 and Lindows run fine on it, but I have to get the accelerated CN400 (H/W HD MPEG2 decoding) drivers, and build MythTV for it.
I sink my own email, thankyouverymuch ( *points to TCP port 25 on his mail server* ).
There's something wrong with a world where the flesh-and-blood mailman delivers mail to a wholely insecure metal box outside my home, but my computer has to "get" mail from the electonic equivalent of a P.O. Box.
No, I won't relay your spam.
IMNSHO, when you buy a house it should come with a routable/28 subnet, at least.
Slashdot Mirror
Well, if you had provided that information in the first place, I wouldn't continue to question the existance of a low power processor and video H/W from other than Via.
In particular, your mention of (AMD) Athlon processors as low-power left me scratching my head -- I was not aware that AMD had low power processors to support fanless designs until you provided that reference. Also, mention of Intel CPUs (likely mobile ones) that you think would be adequate means nothing to me without some convincing that they are capable of rendering 1080i in real time with or without spcialised MEPG2 accelleration hardware (other formats would be nice to support, but not essential for my purposes, not even MPEG4).
But even that processor list (which I could have googled myself) is not what I was looking for when designing a system. I was not interested in having to design and build a mobo around a processor. I wanted to purchase a fairly well integrated mobo, convincingly capable of rendering HD MPEG2 video in real time, convection cooled, in as small a form factor as possible with a strong preference for open source software.
The nanoITX fit the bill, even as it took forever to become available (heck, mine doesn't even have an integrated coldLuke processor/northbridge -- it has separate Eden CPU and CN400 northbridge) and dissapointingly requires (very modest) active cooling for the 1.0 GHz version.
There was a Commell mini-ITX board that employed the CN400, and X CLE266 driver extention development for it to support HD resolutions, which led me down this path (waiting for the supposedly fanless Eden processor in a nanoITX form factor). The only trouble was that the Commell board required noisy active cooling (I was not about to engineer a heatpipe/enclosure solution from scratch, either).
The alternative was a HushPC or Serener enclosure, using heatpipe technology with a conventional processor, but those would have been even more expensive than what I eneded up building (Eden C3 with CN400, 512 MB RAM in a Silverstone Lascala LC08 with slimline DVD/RW drive and 500 GB hard drive) at a cost of around $1000.
Perhaps I erred in thinking that S/W decoding of HD MPEG2 video required a 2.4 GHx P4 (and the associated cooling and noise), but there was nothing I found that convinced me otherwise. H/W MPEG2 accellerators on common graphics cards (which already started to break the requirement of a small form factor) generally didn't have open source drivers.
The only item that showed promise, after a process of elimination, was the Via Eden C3 with CN400 northbridge in a nice nanoITX form factor.
Frankly, I really like the small form factor, and the existing problems strike me as possible to overcome, particularly since Via has their own xine and mplayer for this platform that leverage the CN400 (albeit without completely open software).
And the $1000 drops way down to around $600 if I eliminate the DVD/RW drive and hard disk, and simply use it as a MythTV frontend (which would fit in an even smaller Silverstone Lascala LC07 case). It's just that I like to occasionally pop in a DVD into the playback device in the media room and not have to go to the server rack, and having local content (not to mention support native development if I care to hack) is nice as well.
Good lord, that must be loud as heck, with the CPU and PS fan it requres! At least, if it is anything like a similar Athlon tower I have. What silent (and by silent I mean fanless) system do you have that can do 1080i? I *think* an 800 Mhz Via Luke with a CN400 using openchrome should be able to manage, convenction cooled. The 1.0GHz nanoITX board I have is spec'd to use a fansink (albeit putting out 14 dbA noise), so it doesn't quite cut it either.
Still, I'm impressed that your Athlon can do 1080i60 (or 1080i59.97) in real time at all.
Yes, that was pretty much the point of my post. VIA's CPUs are painfully slow
Compared to a power-hungry monster, yes, certainly for the same clock speed. That's exactly why one turns to specialized hardware for MPEG-2 decoding, when using low-power (heat and computational) processors.
How, if it is binary only?
I can only imagine that perhaps it relies on deprecated or changed kernel or ioctl calls which you've forward ported to 2.6.15.
I might be interested. I guess they bypass xvmc completely then, which would mean there is no dependency on X either. Of course, I haven't looked at the Via versions so I might be guessing just plain wrong.
While I have not been intentionally stressing hd and eth at the same time, I have been doing builds across the 'net to NFS-mounted directories while playing video.
Hmm.
Last time I checked, one needed around a 2.5 GHz P4 to decode 1080i MPEG2 in software.
On my nanoITX mobo, xine -V xvmc runs rings around stock xine.
Many ATI/NVidia/Intel videocards can do hardware MPEG-1/2 decoding. I have such a card, and I NEVER use it. You can't do any postprocessing, deinterlacing, inverse telecine, noise removal, etc.
Hmm. xine -xvmc and deinterlacing works just fine. The openchrome drivers apparently route the decoded video back for further processing, rather than just feed the chip's display engine, though it wouldn't surprise me if this were possible.
I would STRONGLY recomend staying the hell away from VIA. I made that mistake once, myself. You can expect better performance from an AMD/Intel chip clocked at about half what a VIA chip is, and the AMD/Intel chips will be lower power as well.
I have not seen serious problems (like the infamous DMA problem) with the nano-ITX. It isn't cheap, at around $400 for the 1.0 GHz version mobo, but I was looking for a fanless (or close to it) small form-factor. mini-ITX and ATX boards would not fit the bill.
Find yourself a LV PIII-1.0GHz (12.1W max), an ULV Pentium M-753 1.2GHz (5.5W TDP), an Athlon 4-1.2G (25W max), an Opteron-840EE 1.4GHz (30W TDP), or a mTurion MT-34 1.8GHz (25W TDP) if you're dedicated to ultra-low power, fanless CPUs.
And for H/W MPEG2 decoding to HD resolutions...?
Personally, I'm perfectly happy with a cheaper system, and a few quiet, variable 80mm fans.
Smaller, and fanless, is the goal for me. The 1.0 GHz system doesn't quite get there (it has one fan around 14 dBa which really is whisper quiet), but an 800 Mhz system probably would.
To each his or her own, I suppose.
Also, I've read (but not verified) that the via drivers are only supported on older kernels. So, I've been following openchrome with interest.
Finally, I haven't ruled out other issues related to the dropped frames -- remember I'm running a full FC3 right now, with all the servers that are installed by default. Clearly, that needs to change. Until it does, I'm not to harsh on the video drivers for the odd dropped frame (usb traffic, in particular, appears to muck things up).
What attracted me to this platform was the CN400 H/W MPEG2 decoder chip it includes that is capable of deciding HD MPEG2 resolutions (up to 1080i) -- xine plays 1080i on this platform with the 1.0 GHz CPU about 30% idle.
Of course, this is fairly bleeding edge, and there are the occasional dropped frames. Support for the CN400 comes from the openchrome project, which also supports dri/drm, and xine hooks for the resulting xxmc accelleration that takes advantage of the CN400.
It isn't quite fanless -- there is a processor fansink that puts out around 14 dbA. I'm told the 800 Mhz version of the same mobo is fanless, and once I get this stable, will likely spend the $$$ to try one.
Where's a +1 Satire when you need it? (Somehow, +1 Funny does not really apply here).
Yeah, "me too".
I also allow remote access via ssh. I do believe that Blarg frowns on excessive upstream bandwidth -- they'd rather host your public sites, but as long as it "isn't a problem", they're cool about any port you open. They gave me the impression that "we'll let you know nicely" if ever there were a problem.
The biggest concern they might have (understandably) would be over port 25 and open email relays. I took care to make sure I wasn't inadvertently running one.
In my dealings with Blarg, I got the impression that this really wasn't a problem: since their customers were paying far more than they had to to access "that there Intarweb", they were generally responsible about what they were doing with the benefits they had.
Recently, Blarg must've struck a deal with Verizon because I now have the option of paying Blarg for my side of the ATM pipe to them instead of Verizon. Obviously it is more convenient for Verizon to deal with Blarg in bulk than people like me for "Advanced Data Services". I took Blarg up on it: it actually saves me about $5 or $10 a month that way! And, supposedly, they make a few pennies on the deal as well that they didn't before. Win, win, win, all around.
FWIW, I selected Blarg on the basis of their sane TOS, most of all.
I would expect that the IRS (in the U.S. anyway) would treat it as transactions in a foreign currency.
Unless you take special actions re foreign currency transactions (and generally individuals can't), you can't just lump them together and be taxed on the net effect. This generally means that in-game profits are taxable to an individual as they occur.
IANAA. Do not construe this as tax advice.
They have never been available wherever I lived.
Up in 98272 Verizon Land, Verizon charged less for their own crippled internet service than they did to lease the pipes to other ISPs.
Instead, I pay about $60 a month for much better TOS (static IP, servers within reason, etc.), though Verizon still provides the pipe.
See, my DSL connection to Verizon's CO get's shunted onto the ATM fabric and shows up as a PVC to my ISP. They in-turn pay Verizon big $$$ for the fat pipes to whereever.
Now, Verizon charges for both the ATM PVC I use and the fat pipe my ISP uses. They get paid coming and going.
Verizon is free to charge my ISP (Blarg! in case anyone cares) whatever they want for the fat pipes and the DSL PVCs that get resold to me, at least whatever the market will bear. In fact, they price things so that, compared to their ISP service, they get more money from the likes of me. Used to be, I had to pay Verizon directly for the PVC -- "Advanced Data Services" they called it, but they found out that it was easier to sell those in bulk and let the ISP nickel and dime them out (which put more $$$ in my ISP's hands as well as reducing my overall bill).
So, what's the problem here? Verizon can price their pipes at whatever level the market will bear.
I suspect the real issue is that Verizon does not realize that it is competing against itself: their ISP division has to compete with all the other ISPs for bandwidth on it's own network. So what? They get paid either way. But, from the Verizon ISP perspective, they cry foul that so much more money is made by selling bandwidth to and through third-party ISPs and not them -- one division loses while the other division gains.
Note to Verizon: if it is more profitable to lease bandwidth to ISPs than it is to be one, get out of the ISP business!
Heh.
I remember when doing speech recognition research, we'd refer to the schwa as the "garbage vowel", the catch-all phoneme for all miscellaneous vocalizations.
Even that, I can accept: one finds a minority of A**holes everywhere.
But, to try to make it law? And have a court strike that down? And *still* be able to try to make it law? That's just whack.
The same arguments have been applied against spoken speech in public streets, around schools, etc. And, while not having the same Bill 101 force of law, the "offensive" argument is often made.
Then why have people taken the case for it to be made legal to purchase health care privately all the way to the Supreme Court of Canada? Sounds like pretty stubborn and persistent demand to me.
The Court agreed that preventing this was a violation of rights and freedoms, but stayed it's decision depending on whether Quebec would use the Notwithstanding Clause? Only to have the La Belle Province (Quebec) say it would use the Notwithstanding Clause to overrule the Court?
In Quebec, for example, it is often considered "offensive" to speak English on a public street, and represents intolerance of the pure laine Quebecois majority.
Similarly, it is "offensive" for Ubisoft to engage in a practice that shows intolerance for EA's desires, EA being the larger, "majority" employer.
Standard Orwellean doublespeak - nothing to see.
Of course, I viewed this attitide as disgustingly fascist (both in Quebec, *and* the rest of Canada), and left. YMMV
Realize that you're talking about a place where the government can overrule the highest court in the land, and routinely does so ("notwithstanding clause"). Democracy unchecked by law is mob rule. Some might argue that Canada sucks less than the U.S. under Bush *in practice*, but it can suck far, far worse, in *principle*. Another example: Canada models it's socialized health care system after Cuba and North Korea. Comparing it *constitutionally* to Nazi Germany doesn't strike me as much of a stretch -- we just haven't found our Hitler yet. (Yeah, yeah, Godwin can shove it).
Now it is the engineer's turn:
Incidentally, you don't have the ability to authorize players, and inserting DVDs into them isn't how one would do so in any event.
Oh! I meant in the technical sense, not the legal sense. Such a player would likely not be authorized in the legal sense. But, the idea is that, technically, possession of the disk enables decryption of contentied previously copied from the disk, for some interval of time.
(especially since an image of the DVD would include any material such as keys that were on it).
There *might be* an area on the DVD that the drive can read, but will not pass to the computer as the drive reads it. This *might be* the disk key. It *might be* used to decrypt "title" keys, which are then used to decrypt "titles".
Said *disk key* might only be transferred to the computer in an encrypted fashion, after the computer enters into an exchange whereby it response to a challenge by the drive (an *authentication* phase). This lets two "secure" components (computer and drive) communicate securely over a bus that can be otherwise monitored.
The raw data on the disk that can be returned with a blind "copy" using normal file copy functions would not contain this supposed *disk key*. A special exchange with the drive is required (perhaps via an ioctl mechanism in a *nix device driver).
Of course, CSS has been broken so badly that such an exchange is no longer required, and it is merely a matter of massive amounts of computer power in short order to deduce such a *disk key* from the content itself.
Who actually bothers to rip a DVD with CSS intact, and plays it on an authorized player that handles the decryption? Large-scale pirates, I suppose, but that's not you.
Well, I actually would consider making copies of a DVD with CSS intact on a home file server, to be decrypted by thin clients in the home authorized to do so by way of my having inserted the DVD in them at some point (so the keys could be obtained).
While questionable on legal grounds (the use of an unauthorized player in the client), I do believe that I have a resposibility to protect the copyright holder's property -- to leave copies on a home lan unencrypted might be construed as negligent, rather like not locking up a firearm. This is particularly true if external access to machines on that lan is possible, albeit through an encrypted VPM and firewall.
Of course, ripping a DVD with a consumer DVDROM drive does not obtain all the keys necessary for CSS in the manner in which it is intended to be applied -- a key exchange with the DVDROM drive is necessary for that. Clients having entered into such a key exchange at one time can cache the keys for later decryption of the content served to them.
Now, strictly speaking, CSS can be circumvented without having those keys because they are easy to brute force from the raw content itself.
Anyway, leaving CSS intact and not decrypting CSS with an unauthorized program is not circumvention. Copyright infringement perhaps, but they're not the same thing.
I can't see it being copyright infringement if there is no distribution, but the argument could be made, yes.
FWIW, I do not pirate content. I purchase every CD and DVD I wish to have a copy of. But, the reach of the DMCA in preventing me from managing my use of that content in a convenient manner is utterly maddening.
Again, thanks for the insights.
"And that's your fatal mistake. A CSS encrypted movie on a hard disk is no less protected by a TPM, and eligible for the protections of section 1201 than one on a DVD. The medium is irrelevant.
And please, don't try to make clever arguments. They will not only fail, but they will make you look bad."
Well, then, the MPAA doesn't have a leg to stand on either, do they? That was my whole point: the only defense they appear to have is that they did not defeat CSS on the original, opening up the same defense for the likes of me. Which, as you correctly point out, might seam novel and clever, but no doubt anoying to a judge -- the intent was clearly to defeat CSS, by hook or by crook.
This presumes, of course, that the DVD was encrypted with CSS.
But, even if it wasn't, the DVD was provided on condition that the content not be redistributed, in effect a use license, no? IANAL, but I would think that fair use rights *can* be surrendered by way of such a license.
Let's say they can't. The question then boils down to whether the duplication was a fair use. I'm not sure that it clearly is: it enables parallel examination of the work in a commercial setting by many people, clearly depriving the author of revenue from selling additional copies for that purpose (the benefit to the MPAA is the efficiency obtained by way of parallel, instead of serial, examination).
So, the MPAA derives a financial gain from the copies. The author is deprived of revenue. The copies are not used in an educational setting. The copies are used in whole, and not in part. The copies *might* be used as a criticism of the work, but here, are complete copies necessary?
Like I said, IANAL, but even if the DVD was not protected with CSS, the use does not strike me as fair.
Sure, it looks like the MPAA has a legitimate fair use argument, but the DMCA trumps fair use, no?
Either they ripped the DVD, cracking CSS, and violating the DMCA; or they duplicated the DVD, and used an ordinary player to play those duplicates back. That use of a secondary player to access the work would be a DMCA violation, no?
If the latter case is NOT a DMCA violation, then surely my copying a DVD that I purchased to a hard disk, without defeating CSS, and then playing it back from that hard disk, defeating CSS on the copy wouldn't be either, no? (It may be a patent violation for several reasons, not the least of which might be Dolby Digital decoding in software, but that's a different issue).
The point is that if a straight rip is fair use, then what I do whith that rip as opposed to the original, including defeating CSS on it for personal viewing strikes me as fair use as well. The notion is that CSS applies to the medium and not the content: it protects the content on that medium. This is butressed by the fact that there are hidden areas on the DVD medium that support CSS that can not be copied by an ordinary DVDROM drive -- they are not part of the content ripped.
FC3 and Lindows run fine on it, but I have to get the accelerated CN400 (H/W HD MPEG2 decoding) drivers, and build MythTV for it.
I sink my own email, thankyouverymuch ( *points to TCP port 25 on his mail server* ).
There's something wrong with a world where the flesh-and-blood mailman delivers mail to a wholely insecure metal box outside my home, but my computer has to "get" mail from the electonic equivalent of a P.O. Box.
No, I won't relay your spam.
IMNSHO, when you buy a house it should come with a routable /28 subnet, at least.