Best AGA implementation on FPGA?

[grond]

Originally Posted by ReadOnlyCat

With all due respect this is not an "understanding", it is an opinion. Which is fine, but is quite a different animal since users decide what their Amiga is for.

To date nobody has been forced to buy an upgrade they don't want.

Everyone has different needs and use cases and forgetting that only reduces the number of your end users.

You can't please everyone. What's the Amiga need that vampire currently doesn't satisfy? It's too fast and doesn't give you the "I'll go get a coffee while starting this program" feeling of the good old days?

Also if this is about power computing, then why are we even talking about Amigas?

Awwww, replace "is" with "was".

Power computing is far far far ahead of anything an FPGA can do.

Quite to the contrary but this hasn't got anything to do with the Amiga so we shouldn't get into this debate.

[matthey]

Originally Posted by Overflow

How is AeonKit blocking "classic" development?

People are free to develop software and hardware for it.
Is the foundation to the claim based on Trevor deciding to funnel his money into PPC instead of 680xx?

I am fine with Trevor wasting his money on his PPC Amiga hobby. I don't like the way A-EON talks the talk but doesn't walk the walk.

The Talk: They say they support Amiga development, announce a call for Amiga developers (including 68k) and create a new developer web site at www.amigadeveloper.com.
The Walk: The developer web site looks more like an advertisement and has nothing useful. At least some e-mails contacting them have gone unanswered. Meanwhile, Jim Drew who was a bigger Amiga developer than any we have left can't get developer docs to P96 which is used in AmigaOS 4. ThoR and Olsen would like to at least maintain AmigaOS 3 with bug fixes without any kind of support or acknowledgement from up top. FPGA development is completely ignored. It looks like they are just trying to be friendly with 68k users and developers so they can convert them to PPC.

The Talk: They own www.amiga.org, www.amigworld.net, amigakit.com and amistore.net which they use to support the amiga.
The Walk: It is all a big advertisement with censorship of anything they don't want to hear. I have had 2 of my last few amiga.org messages deleted and I was banned for a week with false claims of trolling and libel (I posted the thread content here on EAB). My 68k debugging helped get Chris Young's NetSurf running on the Amiga but now I will not be helping with further recent efforts (I have no plans on returning). Again, it looks like the idea is to lock users into their system where they have to pay for every minor update (worse than Apple).

The Talk: They have announced products for the 68k like the Prisma sound card, MiniMig Plus, "Enhancer software package" and an updated Personal Paint.
The Walk: Prisma and the Minimig Plus should be ready in 2 more weeks (Amiga weeks). The Minimig Plus is kind of like C= trying to sell the A600 when the A1200 was available (outdated when conceived). There is no news in a long time about the software package which is just a compilation of 3rd party software which should be built into the AmigaOS and a 68k AmigaOS update released. Personal Paint did get a 68k release with a few enhancements after pulling the freely available version off aminet (if the update was worth while then this wouldn't have been necessary).

Originally Posted by eXeler0

@MattHey I hear ya, your reasoning makes sense and is somewhat consistent with stuff I heard from other ppl around BigGun. Still, he is putting his talent to good use atm, and for that I'm greatful.
And yea...sometimes adding too many ppl to a project will be the reason for its failure...

For all his flaws, I still like Gunnar. He can't treat his friends like he did meynaf and I and expect them to stick around though. The project is all about Gunnar who likes complete control which he would not give up easily. He does know what he is doing technically and seems to have finally realized that he has to make the core very 68020 ISA compatible.

Originally Posted by clebin

It's a shame you had to bring this topic up here because it doesn't have anything to do with the FPGA or SAGA discussion.

If A-EON/Hyperion supported the 68k and FPGA hardware then it would dramatically help developers and users of these devices. The 68k AmigaOS could be made more API compatible with PPC AmigaOS 4 which would help Amiga development and support in general. Instead we waste limited Amiga developer time and developer resources. Don't try telling them on amiga.org or you will probably be censored and/or banned. It looks like the FPGA development will have to continue the route of reverse engineering and AROS which is wasting time and resources. The uncertainty of support likely affected the Natami project's decision not to push forward. I believe this is a *very* important topic for Amiga FPGA hardware.

[OlafSch]

@matthey

why is investing time in aros wasting resources?

Aros is open, sources are adaptable
Aros is free of charge
Aros means independence

No individuals or small companies can control development

Investing in something closed is wasting resources

[clebin]

Originally Posted by matthey

I am fine with Trevor wasting his money on his PPC Amiga hobby. I don't like the way A-EON talks the talk but doesn't walk the walk.

All the way through this post you seem confused about whether it's A-Eon, Hyperion or even Cloanto that you're writing about.

Meanwhile, Jim Drew who was a bigger Amiga developer than any we have left can't get developer docs to P96 which is used in AmigaOS 4.

Nothing to do with A-Eon. It's not A-Eon's software or even Hyperion's. As I understand it, developers are supposed to pay for a license to write P96 drivers. Why would you expect A-Eon to give away another third-party developer's documentation without their permission just because they're buddies with someone who has a license? That's bonkers.

ThoR and Olsen would like to at least maintain AmigaOS 3 with bug fixes without any kind of support or acknowledgement from up top.

Not A-Eon's job. The only company working on OS 3.X is Cloanto. The only people with the license to do so are Cloanto and Hyperion.

There is no news in a long time about the software package which is just a compilation of 3rd party software which should be built into the AmigaOS and a 68k AmigaOS update released.

Released by who, A-Eon? Why not ask PeterK why he hasn't released a new version of OS 3 with his icon library installed. Sorry for the sarcasm, but that makes about as much sense to me.

If A-EON/Hyperion supported the 68k and FPGA hardware then it would dramatically help developers and users of these devices. The 68k AmigaOS could be made more API compatible with PPC AmigaOS 4 which would help Amiga development and support in general.

Again, conflating the two companies and their roles. The API has nothing to do with A-Eon. As for FPGA development, well you mentioned Minimig+ yourself. I agree that they missed the boat (and I wonder if the project hasn't been shelved in favour of ALICE) but at least it showed an intention. And to me it shows that Hyperion and A-Eon are not joined at the hip. I think PPC and OS4 is a dead end and I've said it for years, but I think A-Eon are also doing a lot of good things and I admire their intentions. I think you expect way too much from them.

[ReadOnlyCat]

Originally Posted by OlafSch

I took part in some of the discussions in the past... power computing does not mean competing with newest i7 multicore system but if you have to make a choice between 100% compatibility and speed then taking speed. Apollo is not designed primarly as a replacement for old hardware but as a real upgrade. Hope this explains it better...

This is how I understood it indeed, thanks for the confirmation!

Originally Posted by grond

To date nobody has been forced to buy an upgrade they don't want.

I'm glad that extortion is still frowned upon.

Originally Posted by grond

You can't please everyone. What's the Amiga need that vampire currently doesn't satisfy? It's too fast and doesn't give you the "I'll go get a coffee while starting this program" feeling of the good old days?

Funny that you say that because this feeling is what caused me to buy my first accelerator for the 1200 back in the day, then to switch to the PC when that was not enough (with much regret, thankfully, OS X reduced that pain a bit later on).

As for your question, I am clearly not the target user for the Vampire since I use my Amigas for their original constraints, even much accelerated ones are too slow for my day to day uses anyway (not to mention that the OS is way outdated) but I like trying to push the envelope of fixed hardware. My interest for expansions lies mainly with storage and I/O which excludes Vampire and Apollo obviously even if these kittens interest me from a technological standpoint.

Originally Posted by grond

Awwww, replace "is" with "was".

Well, you never know the original intent of the poster. But thanks for clarifying!

Originally Posted by grond

Quite to the contrary but this hasn't got anything to do with the Amiga so we shouldn't get into this debate.

FPGAs and ASICs target different use cases and markets indeed, let's leave it at that.

Thanks for the friendly clarifications guys!

[robinsonb5]

Originally Posted by grond

It's too fast and doesn't give you the "I'll go get a coffee while starting this program" feeling of the good old days?

There speaks someone who hasn't recently launched InDesign on Windows!

[wawa]

i must unfortunatelly agree with matthey on the big picture, even if he may err in details.

whatever support aeon has announced for the genuine amiga up till now, (not to talk of what has been actually delivered, since there wasnt much), it looks like either an advertisement attempt, failed attempt to bind the genuine amiga community to their proposals, or attempt to cash on minor upgrades, if its even worth to be called like that.

i actually hate to be blunt about it, but lets take a misleading advertising of prisma megamix as what the users would think of being a regular soundcard, and insisting on it. no wonder the project fails.

given that general tendency, i am most sceptical about any potential involvement of aeon or hyperion with amiga fpga projects. i am sorry to say, if it happened, we might certainly expect insensible design and business decisions, prices driven high, no communication on essential subjects, working behind the scenes forever and so on and so on.

hope the amiga fpga projects will steer clear of such prospects and remain autonomous, as they have been doing up till now, and in any case will seek solutions not to make themselves dependant on others, especially if something like this was attempted by any third party.

[wawa]

@matthey

i understand your disappointment. however even if what gunnar and his current team currently proposes may not be perfect, there is hardly a concurrent proposal in this direction, nor is there any to be expected any soon. so, while this would be better to make more coordinated effort, the calculation on part of apollo team may be, that it would take longer, and they want to deliver. and said that, even gunnar doesnt seem as stubborn as not to reconsider critical voices and finally act accordingly when he discovers the points were valid. just to mention bitfield implementation in this context.

so from this perspective there are certainly drawbacks what concerns apollo/vampire project, as it isnt completely open source for instance, even if the hardware part itself appears to be, but in general its worth support, for this reason alone that it ignites some current in the scene, content among the users and opens some prospcts for the future, especially that none of use knows exactly where it leads.

and finally just a short comment on your aros remark. of course if not the legal hell around amiga system aros might be considered unnecessary effort. i might err, but from where i stand it looks like the genuine system alas will never be secure. as in secure to use and to upgrade, except aome basement efforts. as soon as grown up and in the open field such efforts would be at mercy of true ip holders or pretenders, not easily to distinguish from each other. secondly aros gives opportunita to involve people of different goals and differnet platforms of interest in one project, the code being tested on a wider base and teoretically better structures because of this. i wouldnt miss this opportunity. so while i support apollo/vampire project i am rather investing my time with aros68k for the time being. and i doubt this investment will be entirely lost.

[matthey]

Originally Posted by clebin

All the way through this post you seem confused about whether it's A-Eon, Hyperion or even Cloanto that you're writing about.

I do make some business assumptions which have a small possibility of being incorrect. That is that A-EON has considerable control at Hyperion after bailing them out of bankruptcy. The bailout would have taken a considerable sum of money to not only pay off existing debts but put them in a position where they will not end up back in the same situation. A business like this can usually be bought for nearly the same cost as a loan. If someone is willing to make such a risky loan, the business assets are usually used as collateral for a secured loan. A software business with outdated technology and intellectual property like Hyperion would have little value on the auction block as they own few tangible assets. Either through majority control (probably private stock) or a secured loan to Hyperion, A-EON likely has significant influence in, if not complete control of Hyperion. It is possible Trevor is into philanthropy for his Amiga hobby but then why nickel and dime Amiga users for software that should be included in the AmigaOS. It is also possible that Trevor is not very business savvy at all and asked for too little from the nearly worthless bankrupt Hyperion. I would have required majority ownership in Hyperion for the loan and kept it alive as a software house for A-EON in order to fulfill the contract with Amiga Inc. and keep the lawyers away. Cloanto is a separate entity unless A-EON bought more than the rights to Personal Paint.

Originally Posted by clebin

Nothing to do with A-Eon. It's not A-Eon's software or even Hyperion's. As I understand it, developers are supposed to pay for a license to write P96 drivers. Why would you expect A-Eon to give away another third-party developer's documentation without their permission just because they're buddies with someone who has a license? That's bonkers.

If A-EON/Hyperion really wanted to help, they would check their existing P96 license to see if it would apply to the 68k AmigaOS as well. They may already be licensed to give the driver development SDK to all AmigaOS developers like they did for AmigaOS 4 before the graphics.library was integrated in AmigaOS 4 (AmigaOS 4 was 68k in beta so it is likely they distributed the P96 SDK to 68k AmigaOS owners). They may be able to sell an AmigaOS 4 software upgrade with a new graphics.library from AmigaOS 4 for AmigaOS 3. It depends on if the license says "AmigaOS", "AmigaOS 4", "AmigaOne" and/or "PPC only". At the very least, they should have contacts or be able to provide contacts for the P96 owners if they wanted to help. Instead, ThoR is trying to contact the owners through his contacts with the programmers which he talks about on A-EON's own web site. One of the biggest Amiga developers ever, Jim Drew, gets no help when asking for P96 documentation. The AmigaOS 4 censors at amiga.org also remove anything they decide doesn't support AmigaOS 4 and PPC. A-EON has talked about encouraging developer support but they don't walk the walk. I was not always against A-EON or even Hyperion but I look at their actions and I have had enough. I originally wanted to update the 68k AmigaOS 3 to have an API more similar to AmigaOS 4 making Amiga development easier but no more. If they want a clean break and more division then so be it.

Originally Posted by clebin

Again, conflating the two companies and their roles. The API has nothing to do with A-Eon. As for FPGA development, well you mentioned Minimig+ yourself. I agree that they missed the boat (and I wonder if the project hasn't been shelved in favour of ALICE) but at least it showed an intention. And to me it shows that Hyperion and A-Eon are not joined at the hip. I think PPC and OS4 is a dead end and I've said it for years, but I think A-Eon are also doing a lot of good things and I admire their intentions. I think you expect way too much from them.

What is A-EON's role then? Amiga free loan philanthropy? Amiga propaganda organizers and censors in charge of converting Amiga 68k users to PPC?

[clebin]

Originally Posted by matthey

I do make some business assumptions which have a small possibility of being incorrect. That is that A-EON has considerable control at Hyperion after bailing them out of bankruptcy.

Everything you say depends on those assumptions being correct. You might be convinced enough to trash them for it, but I'm not so sure. If what you say is true, I think we'd see more evidence. I think A-Eon would want recognition and they'd want to give OS4 a publicity boost. The news of Trevor getting on board would probably buy OS4 another year without a 4.2 release. Do you also think it's possible that someone like Ben Hermans could employ tricks to keep Hyperion going without a large cash injection?

What is A-EON's role then? Amiga free loan philanthropy? Amiga propaganda organizers and censors in charge of converting Amiga 68k users to PPC?

On hardware, same as ACube. I can't explain why ACube want to continue making new PPC machines, but they do.

On software, I think A-Eon recognise that while the hardware market remains fairly vibrant, the software scene is stone cold. That must be a concern for A-Eon as a hardware vendor and AmigaKit as a reseller. A coordinated effort is sorely needed to revive things. Hyperion aren't going to do it and nobody would trust them if they did. So A-Eon are trying to step into that role. Again, these are just my assumptions and I could also be wrong.

(Last off-topic post from me now)

[matthey]

Originally Posted by clebin

Everything you say depends on those assumptions being correct. You might be convinced enough to trash them for it, but I'm not so sure. If what you say is true, I think we'd see more evidence. I think A-Eon would want recognition and they'd want to give OS4 a publicity boost. Even the most fervent fanboys are losing faith right now but Trevor getting on board would probably buy them another year without a 4.2 release. Do you also think it's possible that someone like Ben Hermans could employ tricks to keep Hyperion going without a large cash injection?

We can only assume since the businesses are private. Logic points toward A-EON acquiring more than some antiquated software to sell. In any case, Hyperion should be indebted to A-EON more than financially. This is why I believe A-EON could help if they wanted to. By the way, businesses near bankruptcy usually have already employed all their tricks to stay out of bankruptcy.

Originally Posted by clebin

On hardware, same as ACube. I can't explain why ACube want to continue making new PPC machines, but they do.

It is still pretty easy to throw together off the shelf SoC PPC hardware even if it is not modern. There is a niche market which is not saturated like the general PC market. As long as there are a handful of wealthy Amiga PPC users willing to pay up for outdated hardware and software then it may be profitable. It just sucks for the Amiga masses who would like something affordable. Better retro compatibility with 68k and AGA hardware compatibility should help enlarge the market also. This is where the FPGA technology is entering the picture and will likely be outselling PPC Amiga hardware soon. We already have 3 AGA implementations for FPGA and they keep getting better and faster. There is huge development talent available in this area but they could achieve so much more if they were working together. A-EON could have been the ones to unite and energize the Amiga platform with a new AmigaOS for the 68k (with more AmigaOS 4 like API for easier development) and standardization but it appears that they chose the niche failed Hyperion business model.

Originally Posted by clebin

On software, I think A-Eon recognise that while the hardware market remains fairly vibrant, the software scene is stone cold. That must be a concern for A-Eon as a hardware vendor and AmigaKit as a reseller. A coordinated effort is sorely needed to revive things. Hyperion aren't going to do it and nobody would trust them if they did. So A-Eon are trying to step into that role. Again, these are just my assumptions and I could also be wrong.

I believe you are correct about the cold Amiga software scene and A-EON realizing it. Niche hardware only has niche software. The Amiga will never be relevant again without affordable hardware and standardization. I certainly am demotivated to do any developing on the Amiga where I have to do more work for less. I love the Amiga but there is no future for it without major changes.

[Overflow]

The reason why I reacted a bit was cause I dont see Trevors actions as a hindrance to the "community". He has created a positive vibe with his positive drive/attitude. That positive atmosphere seems to be somewhat fading now, as the OS development is lagging faar behind for utilization of the "new" PPC hardware.

But I concede to your point that he could have done MORE if he had focused ALL this energy into a FPGA group like Apollo Team, or his own project in that direction.

The Vampire seems to generate alot of buzz, one can only hope it translates into some nice updates of older software to work well with higher monitor resolution, and take advantage of "everyone" having 68060+++ level performance with 128 megabyte of ram. Something that was rather scarse just a month or two back.