Homebrew FTA Receiver?

[Techfizzle]

With all the people that are members on this site, at least one of them has to be a eletronics enginer ,and a couple other probly have contacts to have things made oversea's. Why dont satelliteguys ever come up with thier own fta receiver? We could let people think the best options to put on it before it is relased, we could have a team make up new channel maps instead of blind-scanning and getting rid of the no-use channels (test screens and such)

 

[Blindowl1234]

Sounds like a good idea...I wonder if the cost would be prohibitive for making a smaller quanity of boxes?. I imagine that coolsat, fortec and the others make literally thousands of boxes probably for less than half of what we pay now.
I guess anything is possible. If it works I'd buy one. I've never seen a hobby where you have to have so much else to go with something. One thing does 4:2:2, something else does DVBS2 and so on...I've got a lot of cash tied up in this stuff and still don't have what I want. FTA is like old cars not a poor man's hobby Thanks Blind

 

[FTABman0]

That is a great idea! Bet there is someone here that has the right stuff in that department!

Gotta have 4:2:2 like the Quali-TV and make it have OTA ATSC Digital Terrestrial HDTV kinda like the Diamond 9000 HD, and all the other bells and whistles we guys would love to have in a STB! Woo Hoo, Spread it on!

Maybe have a VBox Positioner built in for us C-Banders and have an analog quality meter (Needle Type Meter or could be LCD Bar Graph) in the front so we could see the signal all the time! Little things like that! Hey, I can wish can’t I!

B~Man

 

[Pepper]

Not to be a killjoy (ok, maybe I do enjoy being a killjoy sometimes), but I seriously doubt there's enough demand for something like this to ever bring the cost down enough, especially if it includes all the features we all want. And dare I say, without the funding available from the legions of the eye-patch crowd who want to use it for nefarious purposes.

Now if we as a group decide to figure out the best way to build a PC that has FTA-DVB cards and such in it, and configure the required software and document it so others can build their own more easily, that might be a more attainable goal. Plus you could build it over time and leave out more expensive pieces until you could afford to add on.

I know there are PC cards or USB devices that do 4:2:2 and HD with the right software and enough horsepower. And ATSC tuner cards. Not sure about the other things though.

 

[brentb636]

I like the PC based idea, except that none of the PCI cards can handle a fast, hardware based blindscan. Other than that, the PC idea is a winner ! Ultimate flexibility .

 

[gillham]

I like the PC based idea, except that none of the PCI cards can handle a fast, hardware based blindscan. Other than that, the PC idea is a winner ! Ultimate flexibility .

Are there any PCI attachable DVB-S2 parts that support blindscan? E.g. it is an issue that nobody is building a PCI card yet, or that the ASICes don't exist yet? Getting the right part onto a PCI card might be a lot easier.

But.. a "packaged" solution for DVB/DVB-S2/ATSC with SD/HD/4:2:2 support would be nice. Probably it would be best to contribute to the MythTV Wiki DVB-S sections. Often the information is not as relevant to North America.

Even without a custom box, having prepackaged XML files of Satellites / Transponders / Channels that are kept mostly up to date would be useful if we had tools to convert them to MythTV, or DVBWorld, or XYZ application.

 

[melgarga]

The PC based route is likely the practical one, but a stand alone stb would be the ultimate.
That said, I've heard of a linux based DVB called a Dreambox. There are even HD models I think. I dont know if the DVB HW is integrated or removeable, as a pci card.
There is another PC based, discontinued I believe using the Genpix card I think, Captiveworks? as I recall, dunno, I dont know if it was just for the subculture group or was a valid product for true FTA.
One *might* request HW based blind scan and HW based graphic acceleration from a DVB via pci or usb mfgr, but I wouldnt hold my breath. I dont know the parameters/restrictions of S2 blind scanning, but a sw based routine for standard DVB in a fast enough X86 based box should be doable. Cue resident C++ coders....
I have 2 DVB PCI (not S2) cards and the learning curve is a bitch. They are fickle, the software is problematic, and as with any Windoze based function, likely to go off into lala land on a whim. Raw PC horsepower is an issue especially with features like DVR and HD output.
Any linux users care to comment on PC based dvb, the dreambox in particular? I saw what was likely an 'end of days' blowout of what I suspect was a lowend unit for less than 90 clams shipped on Fleabay...........worthy of a TFTA station????
Linux challenged pc users want to know.......

 

[AcWxRadar]

Why not something along the line of an AB IPBOX 9000HD?

This is a European model and not available for US or N.A. markets yet.

With this receiver, you can simply purchase different tuner modules and "pop" em into the receiver to get different formats. It only holds two modules at a time, but you can buy like four unique modules. Basically, it is a dual tuner unit, but you can select which format of video tuners you want.

A group of engineers could manage to figure out how to convert an existing FTA receiver to have plug-and-play tuner modules that would process all types of formats and have one box that is the main control unit with one tuner and then a separate module with up to six unique plug-and-play tuner modules.

If you wanted the additional tuner module, you could purchase it separately (at a much lower expense) and connect it to a tap off the front end of the main tuner (the I.F. section) in the main box and then its output would feed back into the main receiver at the RF level.

If you didn't want anything but the basic receiver, you are not out anything extra. But, if you wanted more you wouldn't have to purchase an entirely different and whole other receiver box, just the tuner sections and a small chassis to contain them.

Of course, there would have to be some additional controls built into the primary receiver box to control which tuner you wanted to select at any given time with some automatic fashion. That would make the main box more expensive anyway, but likely not as expensive as three or five or whatever entire receivers.

Just an idea.

Radar

 

[Dishman Dan]

I agree that a PC based solution would be and is more viable. The bad thing about the PC that is that it is not "idiot proof". A stand alone receiver would not require an engineers degree to set up and tweak everytime something is changed or added or crashed!

 

[wizard101]

All the STB chip manufactures have application notes, test boards and a vast amount of supporting documentation. Obviously their goal is to sell as many devices as possible.
These hardware systems are not tremendously complex. Receive-demodulate-digitize-process-display. Many of the functions/optimizations are simply teaks to the software.
As an RF engineer, I can make this comment. On the software side I have no clue. While I completely believe this is possible it would take the expertise of people from more than one specialty, and an organizer.

 

[Dishman Dan]

I might of not been too clear on my last comment - sorry. I meant to say that a stand alone receiver would be more "idiot proof". Unlike a computer that anytime you add something, there is always something else that need upgraded or a setting changed and the computers do crash. We had a bunch of games on our computer(s) over the years. Whenever something was fixed or upgraded on one end, it messed up something else up. We now have 2 stand alone gaming systems which is a lot less headaches than fighting the computer. At last - "idiot proof" and no more crashes! Now we spend more time enjoying instead of trying to get the computer "tweaked". My satellite receivers give me the same piece of mind!

 

[gillham]

All the STB chip manufactures have application notes, test boards and a vast amount of supporting documentation. Obviously their goal is to sell as many devices as possible.
These hardware systems are not tremendously complex. Receive-demodulate-digitize-process-display. Many of the functions/optimizations are simply teaks to the software.

I think the issue is one of cheap silicon to decode 4:2:2P@ML and 4:2:2P@HL. The formerly Amphion, now Conexant CS6652 or CS6654 is about all I could find with hardware decode for 4:2:2. I couldn't find any pricing information.

I don't think it is too practical to build a 4:2:2 STB from scratch, but if some reference design could be built using a 4:2:2 capable chipset instead, that might be a winner.

The PC is still going to beat it in flexibility, but usually not in ease of use.

 

[asmodee]

These are just the initial comments from a nobody on these forums, so take it for what it's worth (near nothing).

My degree is in electronics engineering, but it's the digital side. I haven't a clue on the analog/RF side of things. Also, I'm a software developer. I could do a custom receiver hardware (digital side) + software if someone could handle the analog and RF stuff.

That being said, I doubt there'd be enough demand to make such a project worthwhile at the price it'd have to be sold at...

You have to consider for something with a small run... I have no idea how big Sat Guys is, but even if you could sell 100 units. I'd be surprised if you could get a final design done for $10k and that's not accounting for people's time investment.

So you take 100 units and divide that by $10k and you're going to have to add $100 to the cost of each receiver just to cover prototyping expenses. Then add in however much the time of the people working on the project is worth, then the cost of the parts, etc. and you're better off with a semi-custom receiver.

Say take a single board computer or even an existing dev kit, add in the RF stuff and maybe an mpeg decoder hardware (depending on whether that's the best approach cost wise), add in IR receiver, AV connectors, a decent looking box and you're done.

Less prototyping costs, less time from the designers, but a higher production cost. In such small quantity you're probably further ahead with a higher per unit cost and lower prototyping cost.

 

[techno935]

I'm planning on making a linux PC to handle 4:2:2 decoding for myself. May as well slave it to the 8100 or the almost outdated 4000.