ESG and Transmeta are designing a fan-less (read: really quiet) set to box with Royal Linux and MythTV built in.
I only hope that since they will be building it on open source, that they will be good GNU citizens, and contribute back to the project.
DVR reference design runs Linux, open source MythTV: “We hope that we will be one of the first to offer a commercial product based on MythTV”
Here’s the original press release from Transmeta
The Australian has a nice peice on the Marketing perspective of ad-zapping. The quote below, to me, is a fundamental reality that over the air networks are going to have to come to grips with in the not too distant future. The question is, will they go down kicking and screaming like the RIAA, or will they learn to somehow embrace technology and turn this into an advantage… My bet’s on the former. My guess is that we are going to be sued and legislated into watching commercials. The big networks are going to turn us all into criminals.
The Australian: Marketing in the age of ad-zappers [January 27, 2005]:
“Though 2005 will not see the death of the 30-second television commercial, it may prove to be the beginning of the end of free-to-air television as the automatic answer to a marketing challenge. The risk-averse marketing industry will be forced to re-evaluate its own conservative conventions and how it spends an estimated $3.5 billion.”
eHomeUpgrade | Buffalo Technology USA Announces LinkTheater Hi-Def Wireless Media Player
Buffalo Technology releases a Wireless Hi-Def DVD/Media player which support DivX, DivX HD, and WMV. A quick look at the press release tells me that it will be available the beginning of February for roughly $350.00 (US). This puts it about $100 more than the IO-Data AvelLink player. It’s hard to see at this point if they have a Linux server or not, so we’ll have to play a bit of wait and see, but it sure would be nice to get a cheap HDTV front end for my High-Def programs captured with MythTV. If the price point drops to the $150 range, then I could see putting one in my family room, and one in the bedroom.
Stand up and take a bow Telewest Broadband (UK Broadband company), the company is investing £20 million in the development of TV-on-demand and personal video recorder (PVR) services in 2005. This appears to be a first step in TV-on-demand. I can only hope that US cable companies get it, and begin to consider long term solutions to a ‘PVR in every home’. PVR’s are great inventions, MythTV has revolutionized the way that I view television, but they could very well become extinct if TV-on-demand ever becomes a reality. Comcast is piloting their Video-on-demand solutions in several regions throughout the U.S. but the content is usually only available for 24 hours after broadcast. It’s a nice start, but hardly ‘on demand’.
PVR UK: PVR (Personal Video Recorder) and DVR (Digital Video Recorder) news, reviews and how-tos for the UK: Video On Demand: CableCos answer to PVRs?
From Brad Templeton’s weblog: Brad Ideas: Changing the nature of TV again
He has written a perl script to create a wishlist for MythTV. It is a great idea, and has far more features and functionality than the ‘WishList’ feature of TiVo. I haven’t tried it out yet, but I’d be interested to use it. My only critique of the concept in general, is that I would now have two places to manage my programming needs (MythTV’s guide, and WishTV).
A few friends and I have taken a leap. We are putting together an order of 8 HD-3000 High Definition recording cards before the famous Broadcast Flag stuff becomes Law (June?).
Right now, each of us are ordering two each, and I plan on eventually building a seperate backend with at least .5 Terabytes of storage. I still plan on transcoding to smaller resolution, but long term, I think the idea is to build enough disk space and tweak my front end to be able to play HDTV natively without the transcoding mess.
A player from I-O data is already available for $250, and they expect DivX-capable players to drop to around $100 within a year (what’s odd about the Times piece is that they neglect to mention that there are actually a lot of players out there that support DivX, the big deal about the I-O Data player is that it supports the high-def version of DivX).
I checked out the player as a possible alternate simple frontend for MythTV and unfortunately, they don’t have a Linux client which serves the content. I think it is only a matter of time before this kind of thing is common place, and can act as a cheap front end for MythTV which you can drop in just about any room that you want.
UPDATE: I got an e-mail back from them saying that they will be releasing a Linux version of their Media File Server. This has once again become a very interesting option for a remote (second/third/etc…) front end for a MythTV backend.
One of the nice features of mythtv is that it can do commercial detection, and from that, build a cutlist. My biggest problem, and what kept me from using this feature was that I thought you had to use the user interface to do this. Alas, I have found that indeed there is a command line way to generate the cutlist. Using mythcommflag -f <filename> you can create the commercial flags, the coming back and using mythcommflag –blanks -f give me the cutlist.
Breaks (computed using only blank frame detection)
13090 : 4 (00:07:16.10) (436)
18678 : 5 (00:10:22.18) (622)
35104 : 4 (00:19:30.04) (1170)
41431 : 5 (00:23:01.01) (1381)
51877 : 4 (00:28:49.07) (1729)
52374 : 5 (00:29:05.24) (1745)
53651 : 4 (00:29:48.11) (1788)
65357 : 5 (00:36:18.17) (2178)
78320 : 4 (00:43:30.20) (2610)
85881 : 5 (00:47:42.21) (2862)
100991 : 4 (00:56:06.11) (3366)
107762 : 5 (00:59:52.02) (3592)
The first column is the frame number (useful for avidemux), the second column is the ‘4:cut out‘ ‘5:cut in‘ column, the third column is the timecode (useful for ProjectX) and the fourth column is seconds (useful for mplayer).
Some simple perl scripting, and we have a cutlist that can be used for avidemux, or as an EDL (edit decision list) for mplayer.
Not much earth shattering to report here. The good news is that I have been recording like a madman, and trying to perfect the process before the new fall season begins. I think I have a pretty good process down at this point.
First, I record the shows (duh), then I use ProjectX to demux them, then I use transcode to multiplex them back together and convert them from MPEG-2 to MPEG-4 (ffmpeg XviD codec) with a lower resolution. This might all seem terribly complex at first, but it’s really quite simple, and the result is a file that comes in at 10% the size of the HD counterpart, and the viewing quality is almost as good.
I’m still waiting to upgrade to Fedora Core 2.0, for two reasons.
- The nvidia driver does not compile against the native kernel
- I haven’t had the time
But as soon as Both of those things change, I’ll definintely upgrade. I have been quite successfully transcoding some shows, but others have the strangest interlacing problems that I haven’t been able to solve.