Amiga Kit Stocks Apollo Team Accelerators

[kolla]

Yeah, I'm pretty pissed off with the entire apollo core project, wish MikeJ could open up his core sooner rather than later for contributions, it's somewhat silly that there are multiple 68k open or to be opened softcores being developed in parallel.

And to answer my own question of what an ultimate Amiga would look like:
* FPGA based, maybe several for a more modular design
* multiple cores to pick from, providing as good as possible compatibility with existing CPU+FPU combinations, for example 68030+882, 040 or 060.
* OCS/ECS/AGA/AGA+/RTG options, who knows, SAGA may work here.
* Plenty of RAM, many configuration options
* Ethernet, with option of direct access using dedicated driver, or via emulating A2065, "legacy mode"
* SATA, with option of direct access using dedicated driver, or legacy mode via emulating WD33C93A.
* USB, with option of direct access using dedicated driver (and Poseidon), and/or via emulating keyboard, mouse, joysticks.
* serial interface, because it is always so damn useful for debugging and all
* zif connectors for A1200 and A600 keyboards
* pin header for A500 keyboard
* PS2 connector for external Amiga keyboards
* Legacy DB9 joystick ports, and why not 4 of them.
* stuff I may have forgotten.

Fire tests for me is that the system, through legacy options at least, should be able to run AMIX, as well as NetBSD and Linux.

[kolla]

Btw, I communicated briefly with other 68k communities, and the message was the same all over - drop-in replacements for existing chips is what they all want. Implement a fast 68040 in FPGA on a board that plugs into 040 socket, and you have shitloads of customers already. Do one with 030 and one with just 68882, and you suddenly have a huge customer base.

[kolla]

Originally Posted by Akira

And let's not get this one closed too, i saw the shitshow that one was and would not like to see it again.

Sometimes, a shitstorm is exactly what is needed for progress to emerge. Linux would not have had the position it is has today had it not been for a whole series of shitstorms up through the years.

[Signman]

Originally Posted by kolla

Btw, I communicated briefly with other 68k communities, and the message was the same all over - drop-in replacements for existing chips is what they all want. Implement a fast 68040 in FPGA on a board that plugs into 040 socket, and you have shitloads of customers already. Do one with 030 and one with just 68882, and you suddenly have a huge customer base.

Looks like you have a business plan ready to be implemented.

[matthey]

Originally Posted by Akira

And let's not get this one closed too, i saw the shitshow that one was and would not like to see it again.

It was mainly 2 users in that thread who were the problems and acted like they wanted the thread closed so there would be no further discussions. I want discussions and consensus. I wanted A-Eon to work with Gunnar as I thought it would force him to give up some control and listen to reason. I hope there is a tech savvy group or board from the newly formed (A-Eon/Hyperion/iComp/Apollo Team) alliance making the decisions. At least Jens Schönfeld should have some tech and marketing knowledge even if he supported MIPS over the 68k for the Clone-A at one point.

Gunnar's ISA is all wrong. He is obsessed with adding registers even when consensus has said no. The way to overcome consensus is to take control (become the authoritarian) which is what he did. Adding registers is good for performance if they can be used but the integer registers he added and are shared with the SIMD are not orthogonal making them difficult to use especially for compilers, use more energy and use more memory (larger stacks are needed). His focus on high performance is like he is trying to compete with high end desktop processors in some ways but then he only has an old MMX SIMD unit and no virtual MMU. It looks like he is opposed to a compatible 68k FPU because he needs a new vector FPU since his SIMD unit can't have floating point or it would have floating point in the integer register file slowing down a critical path. He is opposed to modes even though a separate 68k_32 and 68k_64 mode may be the best option if/when 64 bit integer unit support is needed (I'm not a big fan of modes either but one for compatibility and code density and one for 64 bit support isn't too bad for a high end CPU). His ISA limits upgrade paths and compatibility. Gunnar may be talented at CPU design but he is terrible at ISA design. Am I the only one who sees this? What are the chances of getting a real development and standardization team?

[Mr.Flibble]

I think some people just need to accept that the Vampire/apollo core isn't for them and move on.

Pissed off about it?

Go join up with other like minded souls and build something better, after all, competition is good isn't it?

[Djole501]

Originally Posted by kirk_m

Personally, I think all the development of new Amigas is a wasted effort. I'd rather see Hyperion/A-Eon, et. al. just make some "new, old" boards instead of this X5000 stuff. They'd make a killing selling replacement A4000 boards alone. I'd like to be able to buy replacements for stuff that has died, and, not have to go to places like eBay to be gouged on a 30 year old accelerator. This is also where the Vampire has disappointed me (mine is still in transit), as I was expecting it to be an accelerator in the fashion of the old turbo boards, keeping it retro, versus actually making it an entirely new computer.

The is no reason why you could not run a older version of the core so you would only have a fast CPU, more RAM and RTG if you want it. Nobody is forced to run the latest core which is likely to replace more custom chips.

[clebin]

Originally Posted by kolla

Sometimes, a shitstorm is exactly what is needed for progress to emerge.

Please don't pretend you're helping anyone. Do you see AROS developers getting involved in this nonsense? Do you see Mikej trolling the forums? Say what you like about Gunnar, but do you see him on these pages returning the insults from matthey? Progress occurs quietly, behind the scenes. If shitstorms fuelled progress, Amiga would be the most advanced platform in the world.

[kolla]

Originally Posted by clebin

Do you see AROS developers getting involved in this nonsense?

A few yes.

Do you see Mikej trolling the forums? Say what you like about Gunnar, but do you see him on these pages returning the insults from matthey?

It has happened, and I have seen what he says on IRC. Maybe he understands that exposing his views in public may hurt his project.

Progress occurs quietly, behind the scenes.

Progress happens in all kinds of ways and sometimes, not always, shitstorms are what settles disputes.

If shitstorms fuelled progress, Amiga would be the most advanced platform in the world.

Amigaland shitstorms are nothing, the number of parties involved is simply too low.

[wawa]

Originally Posted by matthey

It was mainly 2 users in that thread who were the problems and acted like they wanted the thread closed so there would be no further discussions.

you are referring to kev who obviously went ballistic about just everything? i have an impression that this outbursts had to hide the actual lack of competence and arguments than other way around. around people who actually do real work none behaves like that. no matter if gunnars every design decision is right or wrong, if i was him i wouldnt tolerate guys like that rageing my forums or ircs.

what about that alternative core proposal? why not post a link to this if it is on internet anyway. is there a possibility to contribute to it? if this is a real thing, why keep it secret?

I want discussions and consensus. I wanted A-Eon to work with Gunnar as I thought it would force him to give up some control and listen to reason.

im not sure it would be a good thing. aeon has apparently no technical expertise themselves. the technical choices they made with every hardware they are proposing look at least suboptimal. as if they were made with some other audience in mind than amiga(ng) scene. wonder what they input might be. money?

I hope there is a tech savvy group or board from the newly formed (A-Eon/Hyperion/iComp/Apollo Team) alliance making the decisions. At least Jens Schönfeld should have some tech and marketing knowledge even if he supported MIPS over the 68k for the Clone-A at one point.

seriously? you propose hyperion as tech savy? after their track of record? you know, that there was contact between hyperion and apollo team and it failed, because they just wanted to cash up without delivering anything in return?

also, there were contacts between gunnar and jens on a1k. jens criticised the project in every aspect, spreeding assumptions that the core is not even 68k comptible, just emulating 68k and such. gunnar openly proposed cooperation. never come to that, which is not surprising, considering differences in conceptions and product offerings.

in conclusion, with your proposed alliance we would simply have more people to the table who want to have their peace of cake. it already starts to take course anyway, with resellers stepping in. i wonder what advantages will arise of it.

[wawa]

Originally Posted by kolla

A few yes.

kalamatee, if you speak of him, may not be courteous at times. but i have not seen him nonsensically rageing in here. on the opposite, his comments are perfectly to the point. calm, competent and constructive. he even called me on my behaviour, righteously.

contributing to an common effort may not always be easy, especiylly emotionally, but useless quarrels dont lead anywhere.

[E-Penguin]

It's clear there's a lot of effort being put into the Amiga at the moment, possibly the most I've seen since Commodore went bust. SAGA, Apollo, AROS, Picasso96, TerribleFire, no doubt more besides. It's unfortunate that the various teams involved don't always see eye-to-eye, and if they did it would be a wonderful thing, but isn't it also good to have the competition?

There's clearly a market for the Vampire (I've got one) as there is for modern accelerators for classic machines (got several of those) as there is for classic accelerators (got one of those too).

I'm excited to see where this is going. Fragmentation doesn't have to be a Bad Thing.

That said, if a clear "way forward" were to emerge and the various parties all started pulling together in the same direction, maybe the Amiga has a real future beyond reliving one's youth. An Amiga-ish-powered RaspberryPI-esque board would sell quite nicely, I think.

[OlafSch]

@matthey

I saw your discussions with gunnar on apollo forum until you were removed. You did not ask or giving ideas, you requested him to do what you want. Then you talked about your investors you would bring in if he does what you want. You have nothing to request (the same is true for kolla), you have not given any money, you (and kolla) are not even owning vampire. You do not like Gunnar, you do not even like his project, his design decisions and you do not like Aros but still you pop up in every related thread to inform everyone that you do not like it. I will tell you a well hidden secret.. we know it already. Funny, you ask for a standardization commitee, who will force gunnar later to do what this commitee decides? Simply forget it. The majority of real vampire users seem to be very happy with what they get and for amiga project it is a real big success. Who not wants it not needs to buy it.

@thread

I already explained it a couple of times in the past... apollo/vampire is far more than a accellerator, people who just want that and need something that is designed to integrate with existing hardware like graphic cards should not buy a vampire but a dedicated accellerator. The idea always was to have vampires in existing amigas as the easy solution so people have something to use and devs can port software for it, later add more features and at the end to have standalone devices. To start with standalone devices would have been the more challenging route but it is still the goal. Everybody who is fine with that should buy it, who just needs a simple accellerator should look for something different.

[Toni Wilen]

There is lots of wrong (and right too) in Vampire but I don't (and most likely most other developers that know Amiga hardware and software and how and why they work together the way they do) don't bother to join in useless threads that don't have any real technical discussion.

[Amiga1992]

God damn it IT'S HAPPENING AGAIN


I understand people want to discuss things, but there's so much fragmentation, as I said, and no one is willing to admit it, that it is impossible to agree or have constructive discusisons, and anybody making anything is bound to disappoint a group of "Amiga" users undoubtedly, group which will definitely take it to forums to express their disappointment. Likewise, those who spend so much time making something for "Amiga" will usually be very defensive about it and this will lead to problems.

The biggest issue here is of human interaction.

[matthey]

Originally Posted by Mr.Flibble

I think some people just need to accept that the Vampire/apollo core isn't for them and move on.

Pissed off about it?

Go join up with other like minded souls and build something better, after all, competition is good isn't it?

It is highly beneficial to have one good and generally accepted standard moving forward and toward a goal. A-Eon/AmigaKit is endorsing the Apollo Core way as the standard by their press release statement.

Originally Posted by http://amigakit.com/news/News_Release_Apollo_Accelerators.pdf

The new accelerator is powered by the Apollo/68080 Core which is the natural and modern evolution of latest 680x0 processors. It's 100% code compatible, corrects bugs of 680x0 designs and adds on top most of the cool features which were invented the years after.

The Apollo Core ISA is unlikely to be adopted by any other 68k CPU cores and limits future choices, especially in an ASIC. It was created by one man with little evaluation yet the press release also mentions working together.

Originally Posted by http://amigakit.com/news/News_Release_Apollo_Accelerators.pdf

Gunnar van Boehn of Apollo Team commented: "Lets work together to bring 68K and AMIGA back!”

Gunnar's idea of working together is him making the decisions by himself and everyone supporting him without question. Is this the position of A-Eon/AmigaKit or do they want to put a real development team together like the Amiga used to have?

Originally Posted by clebin

Say what you like about Gunnar, but do you see him on these pages returning the insults from matthey?

Gunnar should see the truth about his modus operandi as constructive criticism instead of an insult. A group of diverse and knowledgeable people should give a safety check to decisions (I find it invaluable) but he bypassed this when he took complete control and created an authoritarian top down structure instead.

Originally Posted by wawa

you are referring to kev who obviously went ballistic about just everything? i have an impression that this outbursts had to hide the actual lack of competence and arguments than other way around. around people who actually do real work none behaves like that. no matter if gunnars every design decision is right or wrong, if i was him i wouldnt tolerate guys like that rageing my forums or ircs.

I was referring to grond and wysiwyg (the user's first 17 posts were in this locked thread which makes it look like an account created for trolling). These two pro Apollo Core guys were far more insulting than anything said by kev or me even though kev could have remained cooler and more diplomatic. Feel free to review the thread and decide who should have received moderator attention for personal attacks and creating a trolling account.

http://eab.abime.net/showthread.php?t=87295

Originally Posted by wawa

what about that alternative core proposal? why not post a link to this if it is on internet anyway. is there a possibility to contribute to it? if this is a real thing, why keep it secret?

There is no alternative core project at this point. There is only a potential resource to help make a 68060 like core. It's very interesting and I'm still investigating.

Originally Posted by wawa

im not sure it would be a good thing. aeon has apparently no technical expertise themselves. the technical choices they made with every hardware they are proposing look at least suboptimal. as if they were made with some other audience in mind than amiga(ng) scene. wonder what they input might be. money?

Every business goal is to make money and their very existence depends on it. What A-Eon lacks in technical expertise they could make up for by adding directors, hiring employees and partnering with other businesses. They could learn from past mistakes like the Tabor where lack of following standards (standard PPC FPU) has likely cost them (includes Hyperion) hundreds of hours of development expense and lowered the performance, compatibility and value of their product. The advantage of vertical integration with a 68k core is that they could take control of the standards and customize as needed. You lose this advantage with just partnering with an individual who is making poor ISA and design decisions which do not have synergies with your products and goals.

Originally Posted by wawa

seriously? you propose hyperion as tech savy? after their track of record? you know, that there was contact between hyperion and apollo team and it failed, because they just wanted to cash up without delivering anything in return?

I'm not sure Hyperion controls their own destiny or is even under the same management they were before the whole bankruptcy incident. Their financial situation seems to have improved and their relationship with A-Eon seems to have grown since though .

Originally Posted by wawa

also, there were contacts between gunnar and jens on a1k. jens criticised the project in every aspect, spreading assumptions that the core is not even 68k compatible, just emulating 68k and such. gunnar openly proposed cooperation. never come to that, which is not surprising, considering differences in conceptions and product offerings.

I am aware of Jens doubting Gunnar's project. Creating a 68k core is a very complex project so there was good reason to doubt. Credit to Gunnar for persevering (his obsessions are both a blessing and a curse) and getting it working so well. Now it needs changes to fix where his obsessions steered him wrong.

Originally Posted by wawa

in conclusion, with your proposed alliance we would simply have more people to the table who want to have their peace of cake. it already starts to take course anyway, with resellers stepping in. i wonder what advantages will arise of it.

Resellers offer value with support and AmigaKit/A-Eon is charging a reasonable markup. It looks like there is another project where the alliance is much more than as a reseller though. Perhaps the project includes AmigaOS 68k development and an iComp accelerator or Clone-A with FPGA Apollo Core CPU?

Originally Posted by OlafSch

@matthey

I saw your discussions with gunnar on apollo forum until you were removed. You did not ask or giving ideas, you requested him to do what you want. Then you talked about your investors you would bring in if he does what you want. You have nothing to request (the same is true for kolla), you have not given any money, you (and kolla) are not even owning vampire. You do not like Gunnar, you do not even like his project, his design decisions and you do not like Aros but still you pop up in every related thread to inform everyone that you do not like it. I will tell you a well hidden secret.. we know it already. Funny, you ask for a standardization commitee, who will force gunnar later to do what this commitee decides? Simply forget it. The majority of real vampire users seem to be very happy with what they get and for amiga project it is a real big success. Who not wants it not needs to buy it.

I was asked to join the Apollo Team by Gunnar and I was never removed from the Apollo Team or forum. I left after Meynaf when it became pointless for me to be there any longer. I would be happy to send you the private e-mails where Gunnar asked me to join originally and return if you are willing to keep them private. I do like Gunnar but I recognize his flaws and when he sabotages his own project (Does a parent who lets their child do anything they want love their child the most?). We actually worked together very well until he decided to take a left turn with the project and head in another direction by himself. I miss the Natami days when we had constructive forum discussions where Gunnar was not in complete control. Contacting the potential embedded/ASIC partner and mentioning investors and creating a business didn't get much enthusiasm, maybe because it threatened his control?

I have allowed some of my code to be included in AROS which is kind of odd for something I don't like. I support the concept of AROS but the implementation is a little harder to love, at least on the 68k. Give me the choice of AROS or a developed 68k AmigaOS and I will choose the latter every time but I don't think I'm alone.

[wawa]

Originally Posted by matthey

Gunnar van Boehn of Apollo Team commented: "Lets work together to bring 68K and AMIGA back!”

this is usual marketing template as used by amigakit usually. it always sounds the same, no matter who is quoted. it doesnt mean anything so far other than they have an agreement to sell the bundle of the core and the accelerator. matt, really, you fall for this?

Gunnar's idea of working together is him making the decisions by himself and everyone supporting him without question. Is this the position of A-Eon/AmigaKit or do they want to put a real development team together like the Amiga used to have?

i trust gunnar decisions much more than the combined autorities aeon might put together. gunnar has designed 68k highend fpga core. aeon simply let design few motherboards around standard components by a professional company for a premium price. i see investments capacities. im not sure i see competence. and i fear i see conflict of interests in consequence of already made investments.

I was referring to grond and wysiwyg (the user's first 17 posts were in this locked thread which makes it look like an account created for trolling). These two pro Apollo Core guys were far more insulting than anything said by kev or me even though kev could have remained cooler and more diplomatic. Feel free to review the thread and decide who should have received moderator attention for personal attacks and creating a trolling account.

ive been witness of this. nothing can match kevs spewing insults all around him in hardly comprehensive stream of consciousness kinda bursts. i really wondered if he is sane. grond can be offensive at times and wysiwyg joined in to comment on kev, obviously. still no comparison.

There is no alternative core project at this point. There is only a potential resource to help make a 68060 like core. It's very interesting and I'm still investigating.

sorry, but as i thought. hot air versus delivered work.

look. if someone delivers an alternative to apollo core it wont become standard. if not, it might, no matter what aeon or whoever else does. its simply a matter of social dynamics.

Perhaps the project includes AmigaOS 68k development and an iComp accelerator or Clone-A with FPGA Apollo Core CPU?

we will see. i dont go for assumptions.

I do like Gunnar but I recognize his flaws and when he sabotages his own project (Does a parent who lets their child do anything they want love their child the most?). We actually worked together very well until he decided to take a left turn with the project and head in another direction by himself.

i would be very much more suspicious if gunnar was really perfect!

I miss the Natami days when we had constructive forum discussions where Gunnar was not in complete control.

yet neither natami nor apollo core became what you wanted. what does that tell us?

Contacting the potential embedded/ASIC partner and mentioning investors and creating a business didn't get much enthusiasm, maybe because it threatened his control?

or maybe because it was too early and too unrealistic at that point? gunnar is still referring to asic as remote option. they would have to sell enough fpga bords though to cover the costs.

I have allowed some of my code to be included in AROS which is kind of odd for something I don't like. I support the concept of AROS but the implementation is a little harder to love, at least on the 68k. Give me the choice of AROS or a developed 68k AmigaOS and I will choose the latter every time but I don't think I'm alone.

i dont question your attitude towards aros. even though the copymem patches have been replaced lately by kalamatee with apparently more appropriate cpu specific asm inlines. it isnt in setpatch anymore. we had a massive speedup thank these and other late commits to amiga-m68k target.

as for implementation it is a matter of work and engagement. i let you wait till you are allowed to work on amigaos, while i can help with aros today.

[matthey]

Originally Posted by wawa

this is usual marketing template as used by amigakit usually. it always sounds the same, no matter who is quoted. it doesnt mean anything so far other than they have an agreement to sell the bundle of the core and the accelerator. matt, really, you fall for this?

The press release looks like typical thoughtless fanfare alright. These kinds of statements make it difficult to know where AmigaKit/A-Eon really stands which defeats the purpose of notifying the public.

Originally Posted by wawa

i trust gunnar decisions much more than the combined autorities aeon might put together. gunnar has designed 68k highend fpga core. aeon simply let design few motherboards around standard components by a professional company for a premium price. i see investments capacities. im not sure i see competence. and i fear i see conflict of interests in consequence of already made investments.

Gunnar has received a certain aura of competence due to his accomplishment but relying on this for trusting his decisions is a fallacy. Without checks or reviews by peers, he is prone to make mistakes and miss opportunities by himself. Abandoning team decision making and consensus is arrogant and unproductive.

Originally Posted by wawa

or maybe because it was too early and too unrealistic at that point? gunnar is still referring to asic as remote option. they would have to sell enough fpga boards though to cover the costs.

I don't think it was too early anymore than my ISA creation was too early. We would have been better off finding out the needs and wants of potential embedded customers sooner than later. We would have been better off finding out changes to the VHDL that would make creation and verification easier early rather than later. We would have been better off figuring out the financial situation earlier rather than later. If InnovASIC had partnered, we might already have been able to afford an ASIC although a few more investors would have reduced the outlays and risk. Gunnar failed to assess his resources and continued to hyper optimize his CPU design and ISA for an FPGA. He became as much of a loose end as the lack of an Amiga OS which was ready for a product. Uncertainty kills investment and there has been enough to kill most Amiga projects over the years. My foresight has been better than most Amiga people's hindsight but if it was perfect then I wouldn't have bothered wasting my time.

Originally Posted by wawa

as for implementation it is a matter of work and engagement. i let you wait till you are allowed to work on amigaos, while i can help with aros today.

The 68k and AmigaOS are running out of opportunities. I see the 68k and Amiga potential still and talent in the existing community but it is beyond me to put Humpty Dumpty back together again. I'm not sure if current attempts are even serious.

[wawa]

Originally Posted by matthey

arrogant and unproductive.

and from the other side likely the community sometimes makes impression of arrogant and unproductive. see tonis comment above. especially except of the usual lack of expertise in the discussions there is also no general consensus.

If InnovASIC had partnered, we might already have been able to afford an ASIC although a few more investors would have reduced the outlays and risk. Gunnar failed to assess his resources and continued to hyper optimize his CPU design and ISA for an FPGA. He became as much of a loose end as the lack of an Amiga OS which was ready for a product.

so you dragged a deal along. but maybe he wasnt ready? maybe he thinks its too early to set things in stone and maybe he is right? beyond that the os isnt ready either, so there wouldnt be any complete product, such as could be delivered to general public. this is a construction site, lets face it.

The 68k and AmigaOS are running out of opportunities. I see the 68k and Amiga potential still and talent in the existing community but it is beyond me to put Humpty Dumpty back together again. I'm not sure if current attempts are even serious.

i can very well understand this feeling. but shepperding this herd is simply impossible.

[Amiga1992]

Originally Posted by matthey

The 68k and AmigaOS are running out of opportunities. I see the 68k and Amiga potential still and talent in the existing community but it is beyond me to put Humpty Dumpty back together again. I'm not sure if current attempts are even serious.

You guys really, really don't know when to just leave things be.

I think there's plenty of opportunity to do stuff, but the more sour this community turns the less people want to do stuff for it, the more disgruntled hardware devs quitting you will see.
This kind of ranting discussions are what every time I am thinking about updates for my Amiga software, make me wonder "is it even worth it to release for this kind of scene anymore?". And I just make one niche program with a friend, can't imagine how people doing things that are more popular, like a Vampire, AROS, or whatever, must feel with all this. If I was any of them I'd quit for much less bother.

Maybe the thing is that actually nobody should put "Humpty Dumpty" back together.