@xeron

Quote:

It's pretty unlikely they'll work.
TV Cards are busmastering cards that DMA directly to the graphics card memory, but you'll still need to call StartDMA() from AmigaOS, which I imagine will just return an error code when it's running on the Mediator.

I don't know enough about Busmastering and DMA devices to say for sure, but.... well, you can always give it a try and see what happens, I suppose. But don't hold your breath. :)

Without overlay support, though, definitely no joy. A TV card without DMA is just filling a PCI slot needlessly.

As a matter of interest for people with TV cards, though, I do fully intend to allow AmiTV to work on classics - my PPC-equipped A4000 has a Prometheus in it - but I need to wait for a fully DMA enabled Prometheus firmware before I can work on it.

@LiveForIt

There's a lot of different cards....

but I'd suggest if anyone does get one for their Miggy that they get one as close to mine as possible! There's so many variations out there!

Mine is of course a WinTV HVR-1300 (Cx2388x/CX22702 combo), but I also have a Cx23881 in front of me and of course an NTSC WinTV Go! (Bt878) and an original PAL Win/TV (Bt848). Anything like them are the best, preferably the Cx2388x unless you like looking at a locked-up Amiga. :)

@sinisrus

What happened, though? Did it display a picture?

@Cap

Yours is a Bt8x8 yes?

Doesn't entirely surprise me if there's something still going on somewhere. I'll try and make it as stable as I can but....

The problem of the Bt8x8 chipset locking the machine is still very much there - if you select a bad input eventually your machine will almost certainly crash.
However, stable inputs seem fine (and a lot better than I ever got it to run in the old version! :) )

The Cx2388x driver won't lock up the system but because it uses Intuiticks instead of interrupts the framerate is slower and you need to keep the window active.
This is a bit of a problem if you want to watch TV while working! When I get interrupts working though.....

In the meantime, go to
www.retroreview.com/iang/AmiTV_0.5.lha
and you can now use the Up and Down arrow keys to switch between inputs.

You may very well find that S-Video inputs are in black and white. I'm investigating that now....

Good luck - let me know how you get on!

@Cap

That's good. That means it's working. :)

Seriously, a blue screen means no input, i.e. the input it's using has nothing connected. That input is the composite video on mine but presumably isn't in yours. Have you tried connecting a video source to all your inputs? One of them will probably display.

This version of AmiTV is a completely different program to the last, BTW. The arrow keys don't do anything in this version... yet. But I think I'll see to that in a moment!

Hi everybody.

http://www.retroreview.com/iang/AmiTV_0.4.lha

is where you want to go if you want to try out my current development version.

It currently looks first for a Cx2388x chipset, then falls back to the Bt8x8. If you only have one TV card installed, like any sane person, this won't make much difference.
If however you have one of each, then use the command line option NOCX2388X or NOBT8X8 to disable one or other of the libraries. Or just delete the relevant library.

Please give generously (feedback, that is!). You'll still need a graphics card capable of overlay support and a 32 bit Workbench (I imagine).

It's possible if you get a blank window try activating it and pressing Enter. I think I fixed that bug but I'm not certain!
It's currently hardcoded to use what on my card is composite video - your input may vary!

Thanks to anyone who tries it and reports their findings!

@Helge

Maybe you're right....

But I definitely think they're aiming to be a large fish in a small pond rather than a small fish in a large pond.
I would imagine the industrial side is more a contingency - hedging their bets, in other words.

Of course, I could be quite wrong, but... we shall see.

@Helge

Obviously, you'd be daft to buy it just in case it ran OS4.

But it's so clearly meant for OS4, that's how everyone's going to treat it. Otherwise why else would they advertise it to Amiga people and even include dumps of UBoot?

There are two options:
1) It's an AOS 4 machine and it'll sell quite nicely (relatively speaking)
2) It's another faceless mobo that runs Linux like any other machine, except slower and more expensive, and it'll hardly sell any at all.

Of course buying on the grounds that it should be OS4 compatible would indeed be a bit daft, but that's clearly what it intends to be, so no harm in looking forward to it!

I want one.

I already have an AOS 4 machine which fine but I still want one!

I'm thinking a fanless PSU and one of those spangly new Solid State hard disks. Utter silence! Mmm.....

Oh yes..... :)

@Rogue

I'll try and write a quick test case when I get chance but it may be a while. I'll email it to you when I've knocked it up.

Thanks!

@agafaster

You want a bigger bitmap if you've got a window on a large screen, which is going to happen more and more as monitors get higher resolutions. I can assemble the output and scale, but it's slower than I'd like, and rules out proper Overlay support.

Interesting what you said about watching the test match. I didn't even think I'd released a version with the tuner activated! It's something I want to add fairly soon to the Cx2388x card.

As for it being a bug that's not in the Cx chipset - I don't know, as it's not raising any interrupts! It's very possible if I work out how to make those interrupts trigger, then it'll still crash!

@agafaster

To be honest, I'm not sure.... but then I'm not sure to which problem you're referring either!

All I know is that on the A1 I can get a good picture using a Cx2388x chipset card, but on my Bt8x8 it raises an interrupt saying it's trying to execute a bad opcode. This sounds to me like it's trying to run from a bad address. However, I've not had time to look into it properly.

Now I've got overlay support I think I don't need to worry about the suspiciously inefficient PCI bus on the A1, as I've got a full-scale PAL picture (768x576) running with no artifacting. Overlay support among other things halves the bandwidth I need as it's using YUV422 (2 bytes) instead of RGB565 (4 bytes).

DMAing isn't too much of a problem at the moment because I've split the bitmap into different chunks, one per line, but it would still fail on large lines (1600x1200 resolution would probably fail, for instance).

The main problems:
1) Cx2388x chipset raises interrupts which never trigger the associated routine (works on Bt8x8)
2) Bt8x8 executes the wrong area of memory. I may know how to get round this but if so it'll only work up to resolutions of about 576 lines until I get the new SDK.
3) Bt8x8 raises too many interrupts when the signal is poor and freezes the hardware solid. Don't know if this is AOS or the A1 at fault.

Bug 1 is a mystery to me.
Bug 2 I think I may be able to get round.
Bug 3 may mean I just can't use interrupts on the Bt8x8, which may slow things down immediately. The next rev of AOS however may allow me to DMA straight into the Overlay, so it may cancel it out.

From what I've seen so far, though, I'd have to recommend the Cx2388x chipset over the Bt8x8. It just seems much more solid. Unfortunately, of course, most people have Bt8x8 chipsets!

@Rogue

Interesting.

Maybe it's a side-effect of the old SDK I'm using with the new memory system then, because that's what's happening, even in main memory.

If I do AllocMem(size, MEMF_PUBLIC | MEMF_HWALIGNED) and size is greater than about 2K-3K then StartDMA() returns 0 or -1 (not sure which). If it's less then it returns 1, as you'd expect, and I can use it. This is the limitation I'm working with....

@Rogue

No, I quite understand what you're saying. It's not as though you're overstaffed working on OS 4 or anything!

What I don't understand, though...

Why does IExec->StartDMA() fail with a return code of -1 when the block isn't contiguous? I'd expect it to
return the correct string of DMAStructs for Scatter/Gather, rather than just failing.... I'm sure there's a good reason for it though. :)

I am indeed writing support for the Bt8x8 chipset but....

a) Using it correctly (with interrupts) is liable to hang an AmigaOne solid. I'm hoping this isn't a problem with future OS4 devices.
b) I can't get it to work until the new SDK comes out.

So in other words, don't hold your breath as even when it is 'working' it's liable to crash your machine!

The Cx2388x chipset is indeed the successor to the Bt878 (which was the successor to the Bt848), but it seems a lot more stable so far.... though that might just be chance!

@Hans

I'm afraid I can't actually do any Bt8x8 work until I get the new SDK... again!

I've been messing around with it all morning and my code looks fine, but I've just noticed something - the RISC program is held in main memory on the Bt8x8!
Of course this means the card will probably need a nice contiguous lump of memory, so we're back
to needing AllocVecTags() again.
The Conexant chipset has 2K or something of SRAM onboard, so the program is loaded directly onto the card.

So I'm afraid that once again you're back on the waiting game until the new SDK. Sorry!

In the meantime, anyone who's got a Cx2388x card please try it!

Hi everybody,

At long last, people who have a Radeon with Overlay support (I use a Radeon 7000) and a Conexant CX2388x chipset TV card
can try and get a picture.

This is just a preliminary version of my new API which uses libraries and stuff, so nothing does anything really, except the window.

Can people with a Radeon, a 32-bit Workbench, and a CX2388x-based TV card try it please?

On my WinTV HVR1600 or whatever it is the input is the standard composite one, but it'll vary according to which card you use.

All feedback gratefully received!

Oh, it's at http://www.retroreview.com/iang/AmiTV_0.02.lha

I can answer my own question!

Q: When does DMA not DMA?
A: When you're DMAing to the virtual address instead of the physical address!

I was using the same value as I was passing into StartDMA, rather than the DMAStruct block which was passed out of it!

I now have in RGB24 mode a C64 screen (my test machine, a C64C!).
In Overlay mode, I've also got part of a C64! But only in the wrong places. The colours are right though!

At last... progress....

Oh, @Hans:

That was a microsnippet of code. The rest of it wasn't important!

And as for the Bt8x8, I've not forgotten it, but the Cx23882 seems much more reliable and less prone to oddities by far; plus the Bt8x8 crashes AmigaOnes solid when they drop the signal if you enable interrupts (mind you, I can't get the Cx23882 to trigger an interrupt at all!). My main target is now the Cx2388x chipset as it's the current one, but I'll try and keep the Bt8x8 working.

@Chip

That is a possibility, yes, but only for the MPEG stream I think. I think DVPlayer needs a stream of data rather than a raw bitmap which is why the TV card is offering.
The DVB part of the card does output a valid MPEG stream though I think so I may well do that. Of course I'll need to buy DvPlayer for that though!

No, it's not the first line of a very bad joke.

I have finally made my TV card mostly working again, BUT I can't see anything.... it's just not writing to memory.

It thinks it's writing to memory, but it's not. This is what I'm doing (from memory, there will be inaccuracies):



This all succeeds, no problemo. DMA 'starts' up ok on each line otherwise it yells loudly.

Then I program the Connexant chip:



That's more or less it, anyway.

This used to work.
It doesn't now!

I can check the Cx2388x chip's RISC Program Counter, and it's going properly.
I can even check where it thinks it's DMAing to, and it's the correct value (about 0x6d69e000 or something). It's always in one of the mFrameBufferLine regions.

However, my data is never being changed. It's DMAing to it, but the target address never changes from 0xDEDEDEDE (my default value).

Remember I'm using the latest update, but the latest SDK also, which of course is 50.22 so doesn't have any AllocVecTags() or anything.

Any help much appreciated, it's very frustrating when you can see things working but there's no evidence of it!

Thanks!