Slashdot Mirror
Build Your Own Linux PVR
linuxwrangler writes "A few weeks ago Russell Pavlicek, Infoworld's 'Open Source' columnist mentioned a personal linux video time-shifter (PVR) he built. In response to reader requests he has now posted a page describing the project." Escaping the monthly fees of TiVo is a good motivation -- and the total cost here isn't bad either.
After all, if one person posts the times of the programs that they want to record, then everyone can have automated recording like Tivo.
Any volunteers for this open source database?
The baby's fine -- please stop sending business cards.
If I build it myself, it won't start to think me gay (not that that's a bad thing...necessarily)
It's curious that there's no reference to Freevo.
So, for $100 more than the cost of a TiVo, which are now at $199 with a $50 rebate, he built a box that has 5 hours less recording time, a worse encoder,a fraction of the features, and how do you even begin to discribe the UI? Non-existant? I will stick to my TiVos.
Step 1: Sell your TV
Step 2: Cancel your cable/satellite/TiVo account
Step 3: Profit!
Oh...crap. I actually have a step 2.
---
Open Source Shirts
MythTV works quite a bit better than this one does. Check it out here.
Yeah, I'm tired of paying monthly fees to a company that treats its customers with respect. Screw these companies supporting Linux too. Man, when TiVo released the 3.0 version of their software with broadband support built-in I nearly puked.
If I hear another person complain about a monthly fee to TiVo, I'm gonna punch them square in the pie hole.
The recording rate of the TV capture card appears to max out at 15 frames per second. If I were intending to archive these programs forever, I'd probably invest in a better card. But for timeshifting shows like the evening news from France (for my wife) and The Red Green Show (for me), 15 fps is adequate.
15 FPS, I'd hardly call that adequate for 30 FPS NTSC television, but to each his own I suppose.
In his first article:
"I remember how the process used to go when I'd scope out solutions in the closed-source world. There would be brochures to peruse. There would be data sheets read. Maybe there would be crippleware demos to run. And then there would be a solution to buy.
Unfortunately, this takes time. Significant time. And in the Internet age, time is critical.
But in the world of open source, I had several options right on my Linux CD. I didn't have to waste time with endless marketing materials."
But you DID have to waste time compiling, testing, setting up, configuring, tweaking, this that and the other thing. I mean c'mon. I've seen (and used) several proprietary solutions that work great with *uncrippled* time-trial demos.
And then you had to waste the time searching for compatible hardware, testing that, taking it back to the store, arguing for a refund, paying 15% restocking, trying something else, etc, etc.
And then you get to the box itself. I'm sorry but a Celeron 400? Sure it might 'work', but not all that well. My p3 600 had trouble capturing tv quality streams without siginificant losses. And it at least had the benefit of UDMA/100 and a 133mhz FSB.
Meh, so some guy made a shitty (functionally and aesthetically) PC that runs linux and plugged it into his TV. Kudo's to you, sir.
I'd really like to do something like this, and use linux to do it. But this guy is full of it. I hate when zealots pretend to be informative.
Gimme a useful article, not a thinly veiled 'MS is TeH SUCK Liniz is tEH GODE!' troll.
I don't need no instructions to know how to rock!!!!
I call it my "tuxvo"
If you combine cron with vcr you can achieve good quality recordings using divx4/5 pretty much off the shelf.
Add an nVidia GeForce 2 MX w/TV-out and nVidia's kernel module you can save yourself the money for the scan converter.
That is, if you don't mind black bars at the edge of the screen. Otherwise go with the scan converter.
But IMHO nothing beats MPlayer for playback. YMMV
Oh and In Soviet Russia tuxvo records you.
I just got Tivo, and while I'm not a huge fan of the fee (I went box lifetime subscription), you have to remember that with the fee you get software updates and fixes and a really good, highly detailed program guide.
Tivos feature set goes well beyond just recording Sopranos on Sunday nights at 8PM.
I'm not saying an HDD-based VCR wouldn't be a good thing, I wouldn't mind having one, but don't compare it to Tivo -- it's not the same.
Bullshit! Apparently he doesn't have a wife that feels the need to carry on a conversation right in the middle of a critical part of the show. ;-)
and likewise, the rewind feature comes in handy so you can reply what you just missed when the "conversation" started, before you could pause
For $300 that this cost, you could have a TiVO. And TiVO works fine w/o a subscription, you just don't get the guide and suggestions that you do with the monthly fee. And it's kludged beyond all belief.
So this guy built a box that 'kinda' works (15fps, vis problems, etc). Check out the AVScience HTPC (home theatre PC) forums at http://www.avsforum.com (Under the HT PC section, they have a dedicated HTPC Linux section).
Use vcr
/dev/video
Here are my settings: (works very well)
[defaults]
quality = 100
keyframes = 15
audiobitrate = 128
framerate = 29.97
audiomode = stereo
resolution = 384x288
codec = DivX 4.0
norm = NTSC
source = Television
grabdevice =
freqtab = us-cable
Ahhh, the TiVo faithful... possibly a group actually MORE ravenous about a product than Mac users. I should know, I'm both!
Don't Tivo and Replay have patents on the concept of DVR? It may even be illegal for him to overspend for shoddy substitute.
On the plus side... I have fun playing with my toys too.
It may be nice to have a box and software that *potentially* can work like a Tivo... for the Linux gods. But I have to ask in the name of all the rest of the mortals, where's the GUI? Can we just set this up ONCE and then avoid any kind of hacking? Or is this flawed with the problem of many OSS packages, lack of usability for those who doesn't want/are not able to bother?
IMHO this is the main reason why OSS has not skyrocketed yet, because many people would prefer to pay an extra $100 than having to write batches every time you want to record a TV show or any other kind of menial task.
I know instances of OSS with good interfaces exist. But GUI usually seems to have a very low priority, unfortunately.
The ENIAC Demo Competition
While the main appeal to most people to purchase a TiVo is its timeshifting ability, there is far more to its featureset than just the way it pauses live TV. Anyone who thinks that TiVo is just a digital VCR hasn't really looked at one closely enough to understand the flexibility that it grants you.
"And I don't like the concept of having my destiny linked with the fortunes of the supplying company. I don't need a US$300 doorstop if the TiVo company should someday fail. Hey, if Enron and Worldcom can end up in the toilet, you have to allow for the fact that no one company will be around forever."
I think TiVo has stated that should they go under, they'd supply the current users with some facility to allow the units to continue to function.
"It should also be capable of creating images that can be burned on Video CDs (VCD)."
This is an ability I know someone has hacked into their TiVo. In addition, the new version of WinDVD allows you to view a stream at 1.2x the speed of broadcast, letting you shave down a 60 minute program to 30 minutes or so (after eliminating commercials).
"My God...It's full of ads!" -Fry, about the Internet, Futurama
Plus it's progressive scan, so actually losing 75% of the original 60 frame interlaced NTSC source. This system was definitely designed by a computer guy, not a video guy (as honorable and deep a form of geekhood as any).
:).
He really wants something that can do interlaced capture, like MPEG-2 or some of the higher profile versions of MPEG-4. I believe this is supported in the current CVS of Xvid, at least experimentally. MPEG-4 would give a LOT smaller file sizes than MPEG-2 at the same quality.
Or, since VCD is his goal, he could capture straight to ffmpeg in a VCD compatible profile. Or use SVCD MPEG-2, which would be higher quality, and could still fit a half-hour show on a 800 MB CD-ROM.
Windows Media 9 has great, free, integrated capture that can do interlaced, but that'd be hardly Linux friendly
My video compression blog
Man builds house by pushing large stones into a big pile.
Government of the people, by corporate executives, for corporate profits.
Sure you say you can do a Tivo cheaper. Maybe. But to me, the advantages of doing this outweigh the advantages of Tivo. I personally don't care for the thumbs uping and thumbs downing. I know what I want to watch and I don't want it recording anything else. Couple other folks mentioned the Freevo project and that interests me alot. But a lot of folks forget what you could do. You could hack the program a bit and add a button that lets you Zap a show to a standard DVD-R. You could use NFS and make a similar pc with a WiFi card in it, but minimal hard drive and a TV out card and have the view your recorded shows on any tv. Or start watching it in the living room, pause and go to the bedroom to finish. You could run samba and then mount a disc and setup a doze machine for your kids and they can watch Seasame Street in their room (no TV needed). I could go on, but I would have a rather long post. The advantages far outweigh they nice package Tivo provides.
Gorkman
Ah, but in fact 29.97 is an approximation itself. The actual value is 30000/1001.
:).
As long as we're being geeky
My video compression blog
From my figuring, the author spent $300 building this super-widget. So, he saved $250 over a new TiVo with a lifetime suibscription.
:)
For that, he sacrificed a whole lot!
* 15fps
* poor video quality (vertical lines) on top of the 15fps
* No 'live pause'
* No watching one program while saving another
* No guide, etc.
For me, the tradeoff isn't worth it. Even if TiVo fails as a company, I suspect that the community will come up with a network-hack for getting scheduling information.
And even if you lose that, you still have a better-quality recorder with an easy interface, more features, that's running linux.
It only compares if you wanted to play around building a PVR for personal humor. It doesn't make sense as a solution, though
Steve
DirecTivo units, as well as MS's UlitmateTV, can do this now. The disadvantage with the former is that DirecTV is required. I'm not sure about UltimateTV, because, really, I'd rather not...
"My God...It's full of ads!" -Fry, about the Internet, Futurama
the company with the worst pop-up/under ads ever
Just as MP3.com is only one vendor of MPEG audio files, X10.com is only one vendor of devices that follow the X-10 power-line data communication protocol.
Will I retire or break 10K?
I've been working on a PC-based Tivo-like system for a couple of months now and have published some (hopefully) useful information here.
One critical factor is the choice of a "smart" or a "dumb" capture card and deciding whether you want to be able to export your recordings to DVDR/CDR disks in DVD/SVCD/VCD format.
If you just want plain Tivo-like functionality then you can use DivX as your compression method and get reasonable results with a software-based realtime encoder.
I've compared the two options and reviewed the Pinnacle PCTV card (dumb BT8x8 capture) and the Hauppauge PVR-250 (smart -- it has onbard hardware MPEG encoding).
Is this stuff any good? Here's a clue -- hardware companies should stick to making hardware and leave the software writing to software professionals. This clearly hasn't happened in the tuner/capture-card industry.
Most of the work to date has been done under Windows but I'm currently working on using this hardware config under Linux and will update the project site accordingly.
However, if you want to then export your DivX files to DVD/SVCD/VCD you're going to get sub-optimal quality because you're transcoding between two lossy formats. Since the stuff I like to keep for posterity on CDR/DVDR is more than likely going to be material like good movies or music concerts, I have opted to use an MPEG1/MPEG2 encoder and avoid re-encoding.
There are also a couple of video samples demonstrating the differences between the three most popular options:
1. Realtime MPEG capture using a dumb card
2. Non-realtime encoding using TMPGenc
3. Realtime MPEG capture using a hardware encoder.
There's a heap more to do on this project but it's coming along quite nicely.
Instant TiVO.... Oh wait, for Linux .....
Try, http://gatos.sourceforge.net/ or ftp://ftp.xig.com/pub/3Ddemos/extras/README.xvamp
... Governments are instituted among Men, deriving their just Powers from the Consent of the Governed...
What's wrong with you people? For years, everyone whined that Linux was being ignored in the server market, in the embedded market, etc.
Suddenly, a few years ago, TiVo produced a Linux-based PVR. They even (*GASP!*) encouraged hacking it!
Now, you want to "avoid the service cost?"
Brilliant. Take one of the few companies that stands to actually make a successful business based on Linux, and attempt to put them out of business because you're too cheap to pay $12 a month for the service, to support a company that has supported the Linux community over the years.
What, exactly, is the reason for not wanting to support TiVo? Besides being complete skinflints, that is...if you can't afford $12/mo., you shouldn't be spending $300-500 for a luxury item like a PVR in the first place.
.@.
So if you can't really use a TiVO without a subscription
You can if it's an old enough TiVo. You just need to: A) find one that originally came with 1.3, B) find a 1.3 image you can put onto the drive.
It makes you wonder if these comments were actually posted BY Tivo
No, they were posted by people with a clue.
I would love to purchase or build a PVR. But I have absolutely no interest in adding to my monthly bills
Then you have several choices.
1) Build your own/buy a HTPC solution. Understand that, currently, it's simply not as good - the UI is poor, the scheeduling isn't as good, and it's not as easy to integrate or be used by others. On the upside, it'll be a lot easier to rip digital video for archive purposes than a TiVo (but not a Replay).
2) Buy an old TiVo from eBay, get an image, deal with the annoy screens.
3) Buy a new TiVo, purchase a lifetime subscription. For $250 it's done. You're looking at a sum-total cost of $450 at this moment for a basic TiVo ($200 for the box, $250 for the sub) and that's it.
4) Buy a new DirecTiVo (if you have DirecTV), and if you have the TotalChoice Premier package then the $5/mo fee is waived.
5) Buy a ReplayTV -- the cost of the subscription is included.
In any of the above situations your monthly bills have not changed.
All these comments saying it isnt as good as TiVO are totally off-topic and are just clogging up the discussion
No they're not. If you believe that build-your-own is as good a solution as a TiVo, you're wrong. It's not. The example given in this article isn't even a good implementation -- you can build one that does a helluva lot better job, but it's still not as good.
If you want to talk about paranoia regarding TiVo/Replay "watching you" then that's another thing (not that I buy into the level of paranoia that concerns people). If you want to talk about worrying about the company folding, that's a valid issue as well. But whining about increased monthly fees is just bullshit - it means you haven't done your homework and are talking out of your ass.