[UserReview] Vampire V2-128 received and it's just pure p0rn.

[AJCopland]

No, that's completely correct. It is, however, not the driver that patches the graphics.library. It is - for P96 - the rtg.library (core component) or - for CGfx - the cybergraphics.library. The driver goes through a well-documented API of the rtg.library.

I'm confused about the whole P96 situation, as I'm sure 90% of the people are on here.

Is "rtg.library" the P96 dynamic interface library? Like a Windows ".DLL" or Linux ".so"?
If so then what's stop anyone from writing an "rtg.library" with the same function signatures & addresses as the P96 version?

Any 3rd party software will see the "rtg.library", load it, map the functions and run without ever using your code or libraries. Why would a license for P96 be required?

I'm asking as I'm genuinely interested since this is a very common case software that doesn't require a license of any kind.
If they did use P96 code, or libraries then of course that's completely different and would absolutely require a license to do that.

Thanks,

Andy

[AJCopland]

Yes, but...

Surely P96 is something like a windows .dll, and so are the chip and card files. However, the rtg.library has two interfaces: One public, for programs calling into the rtg.library, and one private, for the rtg.library calling into the chip and card interfaces. The latter interface is private, has not been publically available, and requires a license.

In principle yes, if the internal interface would be revealed, or would be reworked. This will/would take some time, though.

But this all aside: How would you feel if you would offer a third party a license to your work you spend three or more years of your life on, providing the work, and its sources, then just receiving an answer saying, "no thanks, I'll just do it all myself and copy your work".

Would you really believe that any follow-up work would really be based on a genuine re-implementation?

Look, it's not as if the work wasn't available for licensing in first place. Authors were there, a seller was there, a potential buyer was there, prototypes were done... Nobody was really trying to harm the Apollo, and everything was done on "good will" basis.

Ok then so the "private" interface is the API for dealing with chipsets (hardware drivers) and various graphics cards.

If I was reimplementing, speculating only here I admit, then I wouldn't bother with that end of it at all. I would just create a very simple system for the one chipset that I was supporting.

I suspect that is what has been done because the Vampire implementation isn't a general purpose system meant to be reused with different chips.
It targets a single design, one done by the same "team" as the author, that supports little more than a framebuffer and some basic operations.

We cannot know more with seeing the source code for it, or at least the team discussing it in more detail, however there's no reason to assume infringment of any license here.
Hooking into "graphics.library" is pretty simple, and writing an "rtg.library" is almost trivial especially if you have access to the hardware designer who's written the underlying chip.
-----

Now as to the second part about how you feel about it, well they took a look at what was offered and the licensing terms. They didn't like them or they decided that they were unnecessary given what they hoped to achieve.

It might be annoying but it's fair on their part to do so, the external interface is public, the private part might simply be far more than they need, so they don't license that part and thus don't use it features. Which is pretty standard stuff.

Sorry that it didn't work out for the P96 API authors but I just can't see what the problem is here. Unless nefarious dealings mean that propriatary code was used in this SAGA P96 implementation it seems that it should be quite legal.

[AJCopland]

The code is up on GitHub by the way here : https://github.com/ezrec/saga-drivers

Looks like a ground up reimplementation.

[AJCopland]

Call me old fashioned, but if the interface was good enough for me to work with, it should also be good enough for me to pay for.

As a gesture yeah sure, but legally? No.

Otherwise every header, function interface, dll, library etc would need to be licensed on a case-by-case or batch basis.
The only ones trying that sort of rubbish are Oracle-vs-Google with the Java library headers.

I'm still not sure what's going on with this stuff. I mean what part of P96 is it actually *using* and therefore what is the fuss about? Looking at the source code I see a lot of calls into AROS and... that's it, no P96 specific stuff all.

So what part of P96 does it use? Is it a library loaded by something to do with P96? Is that what the problem is?

I'm still struggling to understand what the back n' forth between you all is actually over.

Yes if someone is using a library or software that needs a license, and that license has a fee, then I think that someone should pay that fee.
However, I can't figure out what software/library is being used that would require someone to be paid.

[AJCopland]

As long as it is based on the P96 API, it is. There is likely a native CGfx API, though.

I do not know what the conditions are for a third party to offer a CGfx driver. I can only tell you want the conditions for the P96 API are.

Currently, the drivers use the P96 API, and *that* is not valid.

This is only true if it uses the code, compiled libraries which require a license. If it's reimplementing the API based on the interface definitions, i.e. those exported to shared libraries and headers, then it's perfectly legal.

You can copyright a specific implementation, but there is nothing to stop someone else from implementing it themself without using any of the existing code or libraries and then selling or open sourcing that implementation.

[AJCopland]

It's not trivial and it is not what was done here. What was done wasn't difficult (for Jason, that is) because indeed there is very little in the apollo RTG subsystem to set up. The documentation is somewhere in the apollo forum. Basically you set some bits in some registers thereby selecting the colourdepth and resolution you want.

The difficult part is indeed the patching of the graphics.library. And that has not been reimplemented by the apollo team which  means that you need a standard picasso installation and then install the apollo picasso drivers on top of that picasso installation. This is still legal because the picasso files are distributed freely with permission by their authors (e.g. on aminet).

As already mentioned before the work of the original authors is respected and acknowledged. They are invited to talk to the apollo team directly.

Ah, thanks Grond I think that answers it for me

[AJCopland]

No-one is saying that the Vampire will only run AROS, or use AROS ROMs!

It's running Workbench 3.1 right now and has a licensed 3.1 ROM included.

The fact that they're investigating AROS and it's replacement ROM is just that an investigation, a test to see if it's a viable idea.

I would prefer it if the damn thing just map-rommed whatever ROM you've got and work without having to patch the ROM but that's the current situation not necessarily always going to be the case.

ThomasRichter you are spreading so much FUD about all of this. They (Apollo) do NOT depend on any OpenSource software or hardware in the current Vampire 2 boards for A600.
They do not have to OpenSource the core or board designs unless they desire doing so, they might be making money from these Vampire boards but you don't seem to mind when companies do that so what is your problem with them doing that? (if they are)

It's weird you argue about paying licenses for things that don't need a license because someone "deserves" money, but after 7 years of hard work by a team developing these accelerators and their hardware/FPGA design you then say that they should have to OpenSource it all for no benefit to them.

You are annoyingly inconsistent in this thread depending on who you're discussing.

So, no AROS is not the only way, and Gunnar is not forcing it on anyone, he's impressed by it but nothing more.
OS 3.1 does run on it, people have shown plenty of videos of it doing just that.
It's not 100% final, even their "Silver" core is still in testing and it's the "Gold" core (all of which will be free updates) that will have most of the bugs fixed. The current release is more like an "alpha" quality build with all features implemented but still with bugs.

All this arguing is doing is wasting a lot of time and typing on the internet.

[AJCopland]

Yes, it does. P96 support for a commercial platform requires a license. Is the P96 driver licensed? No, it's not.

Oh, wait, and P96 is "just there" and "nobody developed it", and "nobody deserves money for it"? I'm not demanding money for my pocket, remember? I'm asking for money for the original developers that also spend a lot of time - probably more than 7 years - to develop it.

You're saying this is worth nothing, and the work can just be taken?

The important part is that P96 is optional, the driver is a separate download, it does not *come-with* the Vampire so it does not require a license.

You can argue that someone who does download it, and does use it with AOS3.1, and does use all of the free (from the developers) tools might want to give them something but as Nicholas pointed out, and as we discussed earlier there's no actual license required here.

It's reverse engineered interface, you dislike that, but that doesn't change it's legality, it is legal.

Apparently, you don't understand a thing here. It's about honesty and ethnics. If my product depends on somebody else's product to enable its full functionality, and this other product requires licensing, I need to get this license. And no, I don't attempt to simply work around that, I tackle the problem honestly.

If I don't want to pay for the license for P96, I cannot simply provide the P96 driver for AmigaOs. For AROS, the situation might be different because AROS does not use P96, but for AmigaOs, it does.

Try to understand the situation for just a moment from the perspective of the P96 developers, please.

I am a software developer, I've seen every game that I've ever worked on pirated.
I understand that frustration, but here we're talking about very different things.
All computer hardware depends on other peoples products to "enable its full functionality" to varying degrees. In this case you could use something other than P96 but that doesn't seem necessary.

Your argument here is still spurious, reverse engineering is legal, if they had used the Picasso SDK to develop it then they would have required a license. However they used a reverse engineered interface and so they do not.

If it was easy (or even possible) to contact the original P96 developers, to get that SDK at all, to negotiate with them then maybe that would be a viable route but the case here is that it isn't necessary or legally required.

Why we're having to dispute it is bizarre, we're talking about long established legal facts with every precedent imaginable going for it that is widely done on a daily basis.

The driver can be written without violating the license, and it has been. The tools themselves are freely available from AmiNet. That's both end of the puzzle, the legal tools, the legal driver for it.

[AJCopland]

thanx... as i understand it you have to read it out and upload it. But better to ask kipper there directly about the details

Yep, you have to prove that you have a legal ROM image. It's so they can give you "back" a modified version.

I don't understand the reasoning or why they didn't just make it work with any existing 3.1 ROM but then I'm not the one developing the hardware

[AJCopland]

So in a sense, you need to decide: Vampire with AROS, or AmigaOs without the Vampire.

But this is false, even if (and most ppl argue it isn't) the P96 driver is a problem you can still use Vampire with AmigaOS just fine, just in that case without P96. It might need something else or maybe the AROS version needs porting to AmigaOS.

Either way everyone's points still stand that you seem to be arguing against something that the developers (P96) themselves don't have a case for.

[AJCopland]

Thomas I thought I should make clear that I'm not against you here, just arguing against you conflating unrelated things namely AmigaOS+Vampire+P96.

These are 3 separate items.

You can buy the Vampie2 for your A600 and use it with AmigaOS or AROS perfectly legally - the totally separate, optional, download of the P96 driver has no bearing on that.

If you want too then you can download the shareware (http://aminet.net/package/driver/video/Picasso96) Picasso96 from AmiNet and use that with the driver.

The driver is an open source, clean room, implementation using reversed engineered headers and so is legal in much the way that AMD cloned x86, or Apollo/TG86k/etc replicates the 680x0 ISA.

The only thing that would need paying for is Picasso96 from AmiNet as it's Shareware and the requested amount is $20... if anyone knows how to pay, or indeed how to contact someone in order too pay them, then that's the only part that needs any money to exchange hands.

I'd be happy to pay that, if there's a way to do so. Paypal? To whom?