@herewegoagain

There wasn't, isn't, and will be not any new high-end hardware on any platfrom ever, that it's computing power is consumed all the way down form day one by the OS or any application running on it.

If something similar materialises, with OS4 installed *and OS4 development continues* then you will see the benefits.

Anyway, I don't understand this:
"What??? PCI, One RAM Slot, What CPU crap is this?"
fast forwarding one week:
"What??? Mulitcore Prossesor? OS4 will never gonna use it's full potential!"

That's crazy!
Are there any specs that can make you (I am not speaking in personal terms but in general) happy?

_________________

@ackcontrls

Quote:

My lack of performance on previously announced products has a lot to do with the information on the situation that has recently become public.

Even making CPU modules??? Don't put the finger on Hyperion-Ainc affair please.

_________________
Philippe 'Elwood' Ferrucci
Sam460 1.10 Ghz
AmigaOS 4 betatester
Amiga Translator Organisation

well, if the high end system will run Ubuntu/sled or something else desktop friendly then I will be hard pressed to resist temptation

I think my wallet would murder me in my sleep if I do that him and buy this system

_________________
Yes I own an Amiga. A non-upgraded A500 that is unpacked once every 3 years.

If you are going to quote me, do so fully or not at all.
/Ins

@Cool_amigaN

Quote:

Cool_amigaN wrote: @herewegoagain *snip* Are there any specs that can make you (I am not speaking in personal terms but in general) happy?

i think you replied to the wrong person, or need to re-read what HWGA said

_________________

@takemehomegrandma

Quote:

I don't agree. OS4 is a non-SMP 32-bit OS, and even if you manage to get it run on *one* of the two cores somehow (thus ignoring the other), you would end up paying 100% of $1500 for a machine that you will use to 50% of its potential. I fail to see how this could be "the best CPU option"...

So, you have a better choice (which I assume does not involve rewriting a big chunk of the OS, breaking compatibility or losing I/O lanes).

What would that be?

--

I'd like to know why this isn't a desktop orientated system, what makes it appropriate or not?

No, it'll not be as fast as a top end Intel system but then neither is a low end Intel system.
The I/O interfaces are for the most part the same as PCs with the exception of some things (SATA, Audio) which look relatively easy to fix.

It'll be expensive yes, but in a small market that's only to be expected. Like it or not that's the price of not using Windows or Linux.

_________________
Whyzzat?

@elwood

Quote:

elwood wrote: @ackcontrls Quote: My lack of performance on previously announced products has a lot to do with the information on the situation that has recently become public. Even making CPU modules??? Don't put the finger on Hyperion-Ainc affair please.

not everything, but alot. dont go jumping to conclusions.

_________________

@ackcontrls

No, I think you've made a terrific choice with PA Semi and that it has great potential to grow with OS4. The press & the entry level critics will certainly (or should) acknowledge that this is a strong system.

Get 'er done & I'll buy her!

_________________
If you're going through hell, keep going. -Winston Churchill

@takemehomegrandma
Quote:

OK, let's try with some pedagogy for two-yearolds

AmigaOS is evolving beyond your little box analogy. The box is changing. The pieces are changing. Some people can't handle change and stay with the same box and pieces and are happy to play with them the way they are until they die. Enjoy your little box while the adults move on to bigger and better things.

_________________
ExecSG Team Lead

@ackcontrls

Quote:

Considerations for versions of the Amiga OS beyond 4.0 have also been taken into consideration.

Well, somehow it was my suspicion that you would already have thought about some of the most obvious questions/doubts regarding your design... If we can get an up-to-date machine with the ability to dual-boot OS4 and Linux for around $1500, that will be just great. I hope it can become a reality.

One thing which puzzled me a little was your comment that you were in negotiations with AI about a licence for a long, long time. Maybe when the dust has settled a little you would be ready to shed some more light on this for us poor die-hard crackpot zealots? Was the OS4 ownership issue really the main obstacle in all this? When/why did things finally start moving? Was there any noticable change in attitude at AI?

@takemehomegrandma

Wow, I'm speechless. What a great combination of a completely inappropriate analogy and an insult to my intelligence. I applaud you, sir.

Now kindly explain to me (text please, not plastic cubes) what, if anything, will go wrong when I run two tasks on a two CPU system, with each task's CPU affinity tweaked so that both CPUs are executing code from different tasks simultaneously. You may assume that the code for these tasks has been written with a multitasking environment in mind, so if they need to share data they already make use of signals, semaphores, etc.

Bonus points: Assume there is no memory protection mechanism in place. Does this affect your answer?

@Kronos

Quote:

Kronos wrote: @CodeSmith Sure if you got an OS designed to give each task it's CPU, and more importantly if you got apps that "know" other task are running at the very same time. Wake me when you worked out those 2.

The first one is a given, OS4 will have to be modified to manage both CPUs. I never said the OS was going to remain unchanged; in fact, the Friedens have said they have already made some changes in that direction in the new memory system. For the second one: Yesterday, I ran Visicalc on my dual core computer, just to see if it worked. It worked rather well actually, and I had some fun playing with this 25 year old program. Did the guy who wrote Visicalc add dual-core support to it? I suspect not...

@CodeSmith

"some changes in that direction in the new memory system" ain't the same as doing a (backward-compatible) SMP-OS.

Dunno what Visical is but if it is from 1982 it's probraly pure MsDOS meaning it doesn't know anything bout multitasking or even how to do proper OS-calls.

The DOS-window in modern Windows is more of an emulator than anything else.


Getting 2 instances of UAE running on 2 cores shouldn't be a problem, but that ain't the goal here ....

_________________
- We don't need good ideas, we haven't run out on bad ones yet
- blame Canada

@takemehomegrandma

Quote:

By concept, Amiga can't be SMP. Forcing SMP into the picture will break Amiga.

What makes you think that? The only limitation is Fobid()/Permit(). But that does not mean, it will not work at all. It only means, it will stall all cores instead of only one. This is a requirement anyway if you do e.g. memtransfers (incl. DMA) with multiple sources/destination (say you write io buffers with DMA but have multiple threads doing a line decoding - all threads need to stall when you fill the buffer).
This is no different to say with one core. In that case the only core available will stall. The difference is, because of the assumption, AmigaOS had one core only, Forbid()/Permit() has been used in various places where semaphores or mutexes would have been a better solution.

The conclusion ? It would still run, but not as efficient as it could. Would it be slower then on a single core ? I doubt this, except stalling all cpus introduces an additional overhead compared to stalling one core only (but usually, that's not the case, is it?).

The fact that AmigaOS uses shared memory for its signaling and message passing OTHO makes it a perfect candidate for multiple cores. Would be interesting to see what happens if a thread on one core crashes and how this would affect the other cores (e.g.illage op code exception on multiple cores).

_________________
AmigaOne XE G3/933/VIA/FM801/R200 (fixed),
G3SE/600/Voodoo3/Sil680/RTL8139/SBLive! (noiseless!)
µA1-MK2/G3/933/R200/CMI8738
XE/G4 (broken 7450/800)

@Kronos

OK, I'm starting to get irritated by these obtuse replies.

First, I used Visicalc as an example. How about this: I can run VB6 as well on my dual core box too. That doesn't run in a sandbox and I'm pretty certain Microsoft didn't code that with multiproc in mind. Yet somehow it works.

Second, I don't understand why you two are having so much trouble understanding SMP. THE OS IS SUPPOSED TO TAKE CARE OF IT, NOT THE APPS. The only effect that SMP has on a program that was already written to be multitasking is to magnify the effect of thread safety bugs. That's it. The OS needs changes to the syncronization functions, and to optimize certain things differently.

Jeez, I'm getting flashbacks to the times I was explaining to my friends that this computer called the amiga could really run many programs at the same time. The same sort of "I call BS" like "I call BS because you'd need one keyboard and screen per program" and "I call BS because all your timing loops would get out of whack"

Just accept it, OK? if Windows can manage SMP without breaking their legendary backwards compatibility, I'm sure we can too.

@CodeSmith

Quote:

The first one is a given, OS4 will have to be modified to manage both CPUs.

It has to be modified to run on that cpu/board anyway.

_________________
AmigaOne XE G3/933/VIA/FM801/R200 (fixed),
G3SE/600/Voodoo3/Sil680/RTL8139/SBLive! (noiseless!)
µA1-MK2/G3/933/R200/CMI8738
XE/G4 (broken 7450/800)

@takemehomegrandma
Quote:

OK, let's try with some pedagogy for two-yearolds:

thats really pathetic. if at any time you feel like actually providing some arguments instead of just behaving like an idiot please feel free.

/stone

@CodeSmith

MMmmmm SMP on x86 started with the PentiumPro (maybe earlier) which means something like mid nineties ...... Windows was sandboxed atleast one time during the Win3.x to XP transition, so it's hard to say wether old SW really runs on the metal or mor in an emulation.

The structures of AmigaOS on the other hand haven't been really updated eversince 1985.

Rememeber how the PPC-port was supposed to be "just a matter of a compiler switch" and how hard it really turned out ?

Adapting Forbid/Permit (and more important what happens in between those 2) may sound easy, but getting all details, sideefects and quirks right to allow reasonable compability might turn out a nightmare.

Reasonable approach would be sandboxing the whole shebang and only allow new SW to run on the 2nd core, but that thought is pure evil as we all know

_________________
- We don't need good ideas, we haven't run out on bad ones yet
- blame Canada

@stone

He's throwing a hissy fit because this is one thread he can't bring the Efika into and feel all smug about.

@dirigent

Quote:

If we can get an up-to-date machine with the ability to dual-boot OS4 and Linux for around $1500, that will be just great. I hope it can become a reality.

Yes, that's very interesting. I still would like to see some NVidea support, though. I'd like to look more into Cg, and am thinking of buying an x86 box or even a mac to play around with it. Combined with some stuff I'm doing on the PS3, that system might be a better alternative then a MacIntel. I could at least test vmx and multihreaded code on a nice dev box, and probably port or code with both, Linux and AOS in mind...(doesn't mean I shouldn't add a winbox, too, but that is not that important any more). So, yeah, I am looking forward to this system. I still have time. I won't buy anything before 2008 anyway. Hopefully until then, all this issues (technically and legally) will be resolved.

_________________
AmigaOne XE G3/933/VIA/FM801/R200 (fixed),
G3SE/600/Voodoo3/Sil680/RTL8139/SBLive! (noiseless!)
µA1-MK2/G3/933/R200/CMI8738
XE/G4 (broken 7450/800)

@Jorge

Quote:

Jorge wrote: @CodeSmith Quote: The first one is a given, OS4 will have to be modified to manage both CPUs. It has to be modified to run on that cpu/board anyway.

Indeed. The question is *how* that is gonna happen, since it would require Amiga Inc and ACK to legally get the full rights to *all* OS4 sources, including ExecSG from the Friedens.
Even if the lawsuit is ruled in the plaintiffs’ favor, I remain very sceptical about that.

Oh and what's the timeline for the PA Semi to reach volume production? Somewhere I read "end of 2007".