@wawa

Quote:

funnily enough if this "last chance entry" really ever gets released, all it might have been entry to will likely be already obsolete and discontinued.

Can you try re-writing that so that it makes sense please? Some of the components on the beta testers' Tabor boards have reached obsolescence (they are no longer easily available) if that's what you mean. The board itself will not be obsolete because it's designed to run AmigaOS applications which it will do rather well. Hence, it will do exactly what it says on the tin! If you're buying it to watch 4k video sources and therefore consider 1080p and 576p video standards 'obsolete' then you may consider everything in the Amiga world to be obsolete. What's your point?!

_________________
"Art challenges technology. Technology inspires the art."
John Lasseter, Co-Founder of Pixar Animation Studios

@BigD

I hope you do because you talk about that all the time and recommend it to others

And PPC is dead end anyway and a port to a different platform will not happen because nobody will finance it. AmigaOS will slowly die with the hardware it is based on. Some sales of Tabor will not change that.

Last edited by OlafS25 on 23-Oct-2018 at 12:16 AM.

Quote:

monstercoder wrote: Talk is cheap, bring out your optimized warp3d.library and let's see whats it worth. AFAIK you have produced very little compared to how good you say you are.

I released the Warp3D libraries to the public and offered them to Matthew@A-Eon. They are no longer available from me. A-Eon is the new owner and they are at least not hostile to the 68k like Hyperion. I was hoping they would make a new version of Warp3D available for the 68k by now at least to fix Hyperion bugs I fixed. I can understand the economics though. Matthew asked me about writing a Radeon Warp3D driver but the Radeon card documentation is poor and the cards are much more difficult to program compared to Voodoo cards. I'm not even sure I could do it given the documentation and it really isn't worth it for a few hundred PCI board owners at most. The Mediator boards get more and more expensive all the time which supply and demand tells us will give less new users. None of the FPGA boards or accelerators have used PCI slots so the market is not likely to grow. A Warp3D driver for the Vampire is not likely as the Apollo Core SIMD unit doesn't support floating point.

I recommended Alain Thellier to lead a 68k Warp3D development team. He could finally add a proper software renderer to it at least. I would be willing to help. Wawa is a good tester and can code a little too. It's sad that the best market for Warp3D is probably a renderer for UAE.

Quote:

monstercoder wrote: Yeah exactly. it was "optimized" . He should team up with cosmos and then maybe he could do wonders.

I fixed bugs and shared code and documentation with Cosmos. The last Hyperion Warp3D.library 4.2 was 67648 bytes. My last modified version is 38528 bytes and there are parts I barely touched like stencils (any programs use them?) and some functions only used on startup. You can compare Cosmos versions to that 38528 bytes to see how much more work he has done. I have worked with famous Amiga developers like Frank Wille, Chris Young, NovaCoder, Alain Thellier, Meynaf and ami_stuff who would likely have good things to say about me. I worked with Gunnar too but he now calls me a liar.

Last edited by matthey on 23-Oct-2018 at 12:17 AM.

@BigD

Tabor is more or less Raspberry class hardware but for a multitude of price. It is embedded hardware not comparable to todays desktop hardware even inferior to average smartphones. I assume Wawa wanted to say that the hardware that was developed years ago will be completely obsolete in todays terms until it is ready, But next to the expensive X5000 it is the only amigaos hardware so if anyone wants to buy somthing and not willing to spend the money for X5000 there is no alternative. But that is hardly attracting anyone from outside, more probably people will buy it who have old amigaos hardware already and want something new.

Last edited by OlafS25 on 23-Oct-2018 at 11:59 AM.

Quote:

BigD wrote: There will be no 'new' AmigaOS if that happens. AmigaOS will not survive on x86-64 just as BeOS didn't survive. AmigaOS3.1 is a dead end fork for retro enthusiasts. The Vampire will not lead to a new NG platform on 68k it will simply reinvigorate the Amiga 68k retro scene. That is different. If you want Amiga to continue on 'modern' hardware then buy a Tabor.

I thought "Modern" hardware for personal computing had a real FPU and SIMD unit. I thought a modern OS for a PC had SMP and security. Everything Amiga dies without affordable systems. Betting on Tabor is not a good bet because even if it is successful, it won't do much for the Amiga. Make another bigger run with PPC e500mc cores using the classic FPU and the current customers won't be happy they paid more for less. The current handicapped boards will have to be supported too. At this point, it is probably just trying to get the costs back but Hyperion may be on the edge delaying the OS. The Amiga is a disaster. Only Amiga makes it impossible.

Quote:

I released the Warp3D libraries to the public and offered them to Matthew@A-Eon. They are no longer available from me. A-Eon is the new owner and they are at least not hostile to the 68k like Hyperion. I was hoping they would make a new version of Warp3D available for the 68k by now at least to fix Hyperion bugs I fixed. I can understand the economics though. Matthew asked me about writing a Radeon Warp3D driver but the Radeon card documentation is poor and the cards are much more difficult to program compared to Voodoo cards. I'm not even sure I could do it given the documentation and it really isn't worth it for a few hundred PCI board owners at most. The Mediator boards get more and more expensive all the time which supply and demand tells us will give less new users. None of the FPGA boards or accelerators have used PCI slots so the market is not likely to grow. A Warp3D driver for the Vampire is not likely as the Apollo Core SIMD unit doesn't support floating point.

Elbox block the PCI stuff, our Amiga elites are against any progress for the Amiga Classic 68k. A come back is impossible : no more users, no more coders, no more drivers, no more software...

Hope John Hertell will reverse Mediator and crack the Mach...

Last edited by CosmosUnivers on 23-Oct-2018 at 04:49 AM.

@matthey

Come on now, don't be so shy. Show us how brilliant you are!

@wawa Quote:

wawa wrote: no need to pick on matthey, as if he did nothing, probably just because he takes part on lengthy technical discussions like that on improvements of m68k architecture, even if really leading nowhere.

Maybe because such ISA discussions are more interesting and can have a broader impact?

Manually optimizing binaries isn't fun...
Quote:

I worked with Gunnar too but he now calls me a liar.

Welcome to the club...

@monstercoder

whats your point really?

@cdimauro

Quote:

Manually optimizing binaries isn't fun...

i have not advocated it. i just dont understand why matthey being bashed.

@wawa: because it's a forum, and having a keyboard gives the "right" to people to write whatever they think about, even without any clue about the topic and/or other people.

It's the internet "democracy"...

@OlafS25

Quote:

Tabor is more or less Raspberry class hardware but for a multitude of price. It is embedded hardware not comparable to todays desktop hardware even inferior to average smartphones. I assume Wawa wanted to say that the hardware that was developed years ago will be completely obsolete in todays terms until it is ready, But next to the expensive X5000 it is the only amigaos hardware so if anyone wants to buy somthing and not willing to spend the money for X5000 there is no alternative. But that is hardly attracting anyone from outside, more propably people will buy it who have old amigaos hardware already and want something new.

Exactly what i think and my view ... but many people was thinking here the tabor will be the machine that will make new user. (pls forget this) new users for do what? no smp, no drivers, no security, no browser no change, no gaming, no professional things, no power.

_________________
I love Amiga and new hope by AmigaNG
A 500 + ; CDTV; CD32;
PowerMac G5 Quad 8GB,SSD,SSHD,7800gtx,Radeon R5 230 2GB;
MacBook Pro Retina I7 2.3ghz;
#nomorea-eoninmyhome

@wawa

Whats your point?
just give matthey a chance to show how great he is.
He created a greatly improved warp3d library but does not want to share it ;(

@tlosm

Who cares this late in the day? It's over 24 years after the death of Commodore so a little bit late to course correct and invent the new Linux

This is do or die for A-EON and Hyperion so get behind them or support another niche OS that might survive if the Tabor fails (MorphOS for example). Even if you support MorphOS isn't it nice to have the Tabor to run it on? All choices are good in this small market.

_________________
"Art challenges technology. Technology inspires the art."
John Lasseter, Co-Founder of Pixar Animation Studios

@monstercoder

Quote:

Re: ..Any chance to see OS4 for Vampire boards? by monstercoder on 23-Oct-2018 11:30:23 @wawa Whats your point? just give matthey a chance to show how great he is. He created a greatly improved warp3d library but does not want to share it ;(

as matthey answered you, he actually shared it as far as its been possible. maybe you need to understand that there may be caveats about distributing a binary, which contents and code are actually owned by a third party. improvements to such a binary are usually distributed in form of patches the user needs to apply himself, such as it was a case with patches to kickstart modules.

now matthey has apparently delivered his work to current owners (or licensees?) of warp3d in order (as he obviously hoped) that it may be legally distributed. if it doesnt happen, he is not the person to blame. personally i see here only the best intentions.

you need to understand that the whole pool of amiga software is a legal minefield with often unclear relationships, which you can barely safely work with, trying to improve it. this is what may be the cause that prevents better progress as much in this as in other cases. personally this was the reason for me to switch to aros. im very pleased that thor and olsen finally got their update to kickstart 3.1 out of the doors under such circumstances. if it had happened ten years ago id probably reconsider. but even today as you see thei work may occur to be endangered by the uncertainty of legal status and thus the subject of legal actions, preventing them to proceed.

sigh, nuff said..

Last edited by wawa on 23-Oct-2018 at 12:43 PM.

@BigD

Quote:

Even if you support MorphOS isn't it nice to have the Tabor to run it on?

MorphOS will never ever run on it.

Quote:

All choices are good in this small market.

Actually, no. Not if it adds even more burden on the few remaining developers who still bother to release applications and develop operating systems. For performance reasons alone, Tabor requires software developers to add specific support for it.

Do not take my word for it. This website is filled with discussion threads regarding the P1022's incompatible CPU design. The poor choice of CPU has caused and will cause countless hours of additional (and otherwise unnecessary) development time, which is an extremely rare resource in this community.

Is it too late to change anything about it? Obviously. The damage is done. Still, no need to sugarcoat it.

@ASiegel

The P2040/2041 (which have complete FPU implementation) were released around 2 to 3 YEARS after the P1022 (2011/2012 versus 2009). That delay would have meant a delay in designing the board until 2013 and would have likely meant we'd still be waiting until 2022 at least until we have had a shippable product to buy. By that time Hyperion would have been bankrupt and the Vampire V4 would have stolen ALL the excitement and good will (and spare change) left in the small fragmented Amiga market. A-EON were forced to compromise with the CPU but it means the AmigaOS market can grow however slowly. Tower57 shows what is possible. Amiga coding has never been easy but it is a lot easier to code if you have relatively low cost machines available to purchase to allow people to buy and use the software that you create! Without the Tabor, AmigaOS 4.x was for those that had £1,700+ of spare change

P2040/2041 based machines could and should come in the future.

Last edited by BigD on 23-Oct-2018 at 01:05 PM.
Last edited by BigD on 23-Oct-2018 at 01:04 PM.

_________________
"Art challenges technology. Technology inspires the art."
John Lasseter, Co-Founder of Pixar Animation Studios

@tonyw

Quote:

tonyw wrote: @Fl@sh There isn't enough memory space for it (you need at least 512 MB just to run the basic OS) and code compiled for 68k would crawl.

Isn't this a massive exaggeration? I don't think it's too different to MorphOS and on my Efika the OS requires about 30-40 MB RAM. Of course some applications are very demanding most importantly Qdyssey. But with 128 MB RAM quite something is possible.

But I fully agree and also think a port of OS4 to the Apollocore is not a wise idea. Enhancing OS 3.x or AROS 68k is the way better approach for Apollo.

_________________
My programs: via.bckrs.de
MorphOS user since V0.4 (2001)

@K-L



Quote:

K-L wrote: @tlosm NO Amiga computer is cheap. Try to buy anything Amiga related that doesn't cost a huge amount of money whether it is for Classic or NG (and whether it is used or new). The Amiga market is NOT cheap nowadays. Period. (And it canoot be compared with anything named "actual market").

My Mac mini and powerbook were both cheap and the only OS on their harddisk is MorphOS. And before ppl say bla bla used stuff bla bla Applew is evil bla bla. Right: it's used and I also dislike Apple and think they are kind of evil. But on the other hand I very much like realism: There is cheap and available hardware, glad the the MorphOS team took opportunity and supports it. Custom hardware makes no sense in a tiny niche.

_________________
My programs: via.bckrs.de
MorphOS user since V0.4 (2001)

@BigD

Quote:

BigD wrote: @tlosm Even if you support MorphOS isn't it nice to have the Tabor to run it on? All choices are good in this small market.

morphOS team announced several times they will not support Tabor (because of the non standard cpu).
I guess we will rather see MorphOS supporting the coming Raptor µATX board. About the same price league as Tabor, but state of the art technology.

_________________
My programs: via.bckrs.de
MorphOS user since V0.4 (2001)