Lets talk about A1222 - Tabor

[Iggy]

I see, that is quite disappointing. I was hoping that integrated gpu could be useful  somehow. It would be nice to have itx motherboard in one of those super tiny cases.

If this gpu is "useless" and can't run even AmigaOS in reasonable way I would not bother in including it at all in this motherboard, but as usual not my choice to make.

Of course it can be used.
Hardware acceleration isn't an absolute requirement.
And some of the PCI-E cards that are supported under OS4.1 can be had quite cheaply.

I can find Radeon HD 6450 cards on eBay right now starting at $13.95, which would allow dual OS4/Linux support.
And Radeon HD 7450 cards start at $15.

Personally, in half height cards, I rather like the HD 8490 or the R5 230 (both use the same core).

And R7 240 cards aren't that expensive.

BTW - I had thought the $500 price point was for the board only, if its for the entire system, all the better.

[Iggy]

I wonder what HW acceleration mean? OpenGL?

I'm not concerned about price of GPU that can be used as a someone who is considering $500 motherboard.

If (when) I will get A1220 (or something similar) I don't intend to play games on this computer. MY intent is to use it for AmigaOS4.1 and software (web browser primary + everything that comes with it), instead of emulation.

So, if I could use COMFORTABLY AmigaOS4.1 + browser that will make very nice tiny computer (I have few "towers" in my house already) in one of those tiny ITX cases. Like on PC, I have ancient intel mobo (about 8 year old) and Win 10 run on integrated GPU wonderfully (GMA X4500) (4k don't play well using youtube ).

We will see.

Yeah, when they refer to acceleration they are talking about 2D and 3D operations, some of which can be accessed by OpenGL routines, but also by Warp3D (in its various incarnations).
Actually, all video cards after the R500 series only have 3D primitives, so any 2D library calls are generally rendered using 3D functions (on later cards).

While the integrated graphics from the P1022 will no doubt be more primitive than the Intel mobo you mentioned, they could still be used for basic software.

 But you're right, the price of the addition of a video card is really minor.

[Iggy]

I'm looking at some passively cooled GPU, and I hope I can find nice passive cooler for CPU.

I also hope that fpu issue will be solved.

I know this will sound strange, but are this motherboards REALLY designed from scratch for A-EON?

Is there any indication/rumors A1222 can be delivered as a whole "system" (case, GPU?)?

A passively cooled video card is easy enough to find, and will work as long as your case ventilation is adequate.
A passive cpu cooler is a little trickier.
Its not a particularly hot SOC, and a large heatsink would help, but I'd just add a large, low speed fan to cut noise.

[Iggy]

The more I listen about all of this "projects" (and complete lack of direction) there more I'm convinced I'm going to keep my Amiga Classic and EMULATE AmigaOS4.1. Maybe I will set up separate machine for it based on some cheap i5 itx mobo  and run it under linux/uea thingy... basically "ALICE" as a desktop.

Who selected PPC without fpu, that now it becomes limitation? Why Sam is not produced anymore? Do we really need multiple motherboards so dev. need to make "Ethernet/sound drivers" again and again?

I listened to one talk yesterday night about AmigaOS4.1 and emulation. Conclusion was that it is NOT about hardware it is about AmigaOS4.1 (3.x). Great, but why to even bother with emulation? Make it PC (or old mac) compatible so I don't have to emulate it anymore.

My excitement (at the moment) about A1222 = zero.

I'm not sure I get it either.
I've sent an inquiry to Acube and gotten zero response.
If they'd made another batch of SAM460 boards I would have bought one.

And I 'sort of' understand the need to recoup your development dollars and the money you've invested in cpus (since Tabor was started before the T10XX cpus were available).
But the lag time developing drivers would have been minimized by a T10XX based design, as those cpus share the same core as the X5000.
And they aren't much more expensive than the P1022 (as well as producing better performance per clock cycle, and having a higher maximum frequency).
Then there's the fpu issue...

I don't know for sure (since the developers haven't said), but I suspect this issue was a large part of the reason for the decision not to support this board under MorphOS.

FPU trapping and emulation? Let's face it, what we are more likely to see are specific recompiles intended just for Tabor.
Which means if you've got more than one OS4 system, you're going to need multiple binaries.
Buy an A.L.I.C.E. laptop, run emulation or own an X1000 or an X5000 AND Tabor and you may need two different binaries (or one larger one with code for both types of FPUs).

I do understand how much more the other e500 cored cpus cost (quite a bit more than the V2), but all this hassle just so that Aeon/Varisys can incorporate a dirt cheap embedded cpu in a board intended to sell for almost $500?

While Tabor will perform better than emulation (because emulation is intentionally limited to legacy PPC hardware), you have to wonder if an emulator designed to simulate a G4 or G5 level processor on an X86/X64 system wouldn't be just as cheap, perform better, and provide a future bridge to an ISA still actively developed.

And you may question why I'd justify paying four times what Tabor will cost for an X5000/40. But the price actually scales out, since the system provides about four times the potential computing power and is probably at least as powerful as what we'd get from a more modern emulation platform.

One thing I'd like to try, under a future X64 fork of MorphOS is a Qemu based emulation of a G4 processor based system.
I feel fairly confident that today X64 hardware could run our legacy PPC OS/software as a process under a modernized SMP enabled Amiga-like OS.

And we have to consider migration eventually, if not to X64, then to an alternative (and as much as I like Power, as its the only compatible ISA, its too expensive and ARM offers no real advantages over X64, unless you're completely focused on compact portable devices).

So we'll have to see  how this all plays out. I'm not sure that the sale of a few hundred e500V2 based boards is really going to help us long term. They aren't really powerful enough, when compared to other alternatives to really justify the price.
Honestly, if you're going to offer Raspberry Pi performance levels, at several times a Raspberry Pi's price, is that really the 'bargain' you think it is?

I'd rather spend the extra money, on an X5000 or even something as exotic as a TalosII board, and get something I know will be reasonably adequate for my use over a few years.

BUT, that's just me. You all do as you will, and we'll have to see how this goes. After all, I can't predict the future. If I could have in the past, we'd all been using Motorola descended cpus and UNIX-like micro kernel OS right now. :hammer:

[Iggy]

@kamelito

Indeed. If there are anyone on the amigascene I will listen to, its Daniel. Ive never encountered anyone that is to such a degree a "cut the bull%&$#?@!%&$#?@!%&$#?@!%&$#?@!" kinda guy. And its very refreshing.

He did comment that he wasnt overly excited by the CPU choice, but at the end of the day he seemed content with the performance of the Tabor.

Maybe he will chime in with his own comments regarding this.

It would be refreshing, as even I don't appreciate a constant stream of negativity.
And I've repeatedly said I wouldn't continue to post the facts contained in my last post, as it is a bit pointless. This IS happening.
And HEY, I still get what I want, so WTF, eh?

If its adequate for YOUR uses, and it fits your budget, I'd say go for it.
Do NOT make your choices based on mine.
We are very likely to have different goals and preferences.

[Iggy]

The man does have a point about the performance of the e500V2 core outside of floating point.
Its basically the same as any other e500, and actually may outperform the e600 core (aka G4) on some functions.

Freescale cpus continued to progress after the goal of an e700 core was dropped, and many of the features intended for that product ended up being incorporated into later e500s, the e5500, and the e6500.

The e500V2 was never intended for floating point intensive applications, but it can serve in mixed integer/floating point applications.

[Iggy]

@Thellier
:confused: 68k? Interface?? Stack??? "Normal C functions"?
I talk about the PPC ABI used in AOS4.

@Iggy
Yup.  IMHO chances are good that the pretty high performance of that thing in  combination with Thomas Frieden's optimizations save the day

@dirkzwanger
Yes,  and let's not forget that this thing is meant to be an entry level  system. So if real fpu-hungry *non*-native stuff runs "okay", then it  should be fine. And let's also not forget that real critical stuff (e.g.  minigl) can certainly be modified to at least do a high amount of work  in native code.

All in all: let's wait and see and be optimistic (and there's certainly reason to be optimistic)

Well...I didn't say high performance,  but we do have to consider that the majority of all code is not floating point.
And when coded for natively, the e500V2 core should certainly be adequate to the application.
We'll have to see what Thomas pulls out of his...hat? But I wouldn't worry to much about the performance hit in software that can't be recompiled (think about that, how much is there?), and it still just a part of the performance equation.
Factor in stronger integer performance than the SAM460, and things aren't that grim.

[Iggy]

I second that, his signature breaks the browsers view when I read on my phone.

Doesn't affect my low end Samsung.
I say leave it, I know other people that have reinstalled Win98 too.
It's a good platform for DOS games.

[Iggy]

I know, but I waited for my BlizzardPPC 15 years so, 2.5 years is nothing .

Yes, and how long did it take OS4 to be considered out of beta?

Look guys, blame Mincea, he signed the contract to develop the drivers AND claimed to be responsible for the poorly functioning beta driver the developers have for the X1000.
Tabor HAS to ship with a network driver.
You might be able to resort to a USB doggle, but that is far from ideal.

Right now we need functioning network drivers for the X5000 and Tabor for both Linux and OS4.

[Iggy]

...Is this about money or lack of time or both? Last few months there is only... SILENCE about Tabor.

Well, they were paid in full according to Aeon, so its not about money.
And porting to a e500v2 cored cpu can't be that different than porting to an e5500 cored cpu.

So...my guess is it really is the network driver that is stumping them.

[Iggy]

Honestly, 2.5 years is way too long to wait for a NIC driver. The actual amount of code that has to be written isn't that great, and open source Linux drivers could probably be used for reference.

In all fairness, having a Linux model doesn't always offer as much utility as people think.
Of all the NG OS', AROS seems to be the operating system that benefits most from Linux examples.
The PPC OS' usually require tighter more streamlined solutions.

But this difference is why I favor NG OS', as Linux uses an ever growing brute force model, while NG OS' descend from a 68K OS where resources were much tighter and coding efficiency had to be maximized.

And my own coding is suited to this as I always wrote for modularity, breaking down tasks into separate sub parts.
What worked well once to limit resource demands on 68K OS' now serves well in SMP OS' to maximize threading.

Also, I think we've had some really creative coders in the Amiga community, and some of them are still here turning out improved solutions.

Further, I just exchanged a message with Ralph Schmidt today, and I made a point of thanking him for his use of a micro kernel in MorphOS. With no apologies to Torvalds, I still think this is a more efficient and stable way to base an OS (unlike the massive monolithic kernel Linux is based on - where the kernel alone is larger than AmigaOS in its entirety).

We have a history of thinking outside of the box, basically descended from our heritage with the first commonly available multimedia computer.
And even some of our dead end were pretty cool (like I would have loved to see the QNX based fork of AmigaOS continue to be developed).

SO...what's the point of this post? Well, Linux models reflect their inefficient UNIX heritage. And as someone who moved from 68K Xenix based systems to more efficient 68K OS' that were being promoted for process control applications, I KNOW there are better ways to do things than the mainstream standards.

Basically, just because Windows, OSX/MacOS and Linux (at a very distant third) dominate the market, it DOESN'T mean we have to model our approaches on these OS'.
Sometimes different can be better.

[Iggy]

Sorry, a crash led to a duplicate post.

Jim

[Iggy]

Some of us have been saying this since 2001 but were drowned out by emotionally challenged halfwits for the most part.

Now, now, Nik,
That argument will only get you called a blue team troll, even if it's true.

[Iggy]

@thread

Info about Tabor and ethernet (and more):

Trevor's blog - We are Amiga!

#6

I found the part about RX video card support more interesting myself.
But as most of you know, Tabor doesn't really do it for me.

Hans' work though...always impressive.

[Iggy]

I fail to see how any os4 capable computer would not be a botleneck . And what Amiga application would take advantage of a RX based card?  Most advanced games we have Are from the late 90s . Not to mention our dogslow chipsets and lack og multithread support

Current hardware doesn't really take full advantage of any of the high end video cards from the 6750 up.
So what?
If I had the choice of a Radeon HD 7750, and R7 250, or an RX580, I'd prefer the RX580. For one thing, it's a current card.
And if we aim lower, say an RX550, it shouldn't be too cpu limited.

That said, since I'm a MorphOS user, and I haven't bought an X5000/20 because I'm hoping for an X5000/40, if I had one  I'd probably drop down to something like an HD 4770. I'd get a driver with overlay support and and video card that's reasonably suited the cpu's capabilities.

I've used a 7750, it's not that much more impressive. And the R7 250 kind if a dog, slower than my R7 260X, which in turn can't touch an HD7850.

What I'm hoping for is R600 (or possibly R700) AGP support for our legacy systems, and 3D acceleration for all the new Radeon HD cards that have been added to the support list.

All that being said, I still think what Hans de Ruiter has accomplished under OS4 in terms of video card support is remarkable.

Pity there isn't someone with that level of talent writing network and sound drivers. You might have those dreadful P1022 based boards in your hands if those issues were resolved.