Amiga.org
Amiga computer related discussion => Amiga Hardware Issues and discussion => Topic started by: Piru on June 21, 2010, 05:25:15 PM
-
(http://image.bayimg.com/bangnaace.jpg)
http://pci-ids.ucw.cz/read/PC/1959
http://en.wikipedia.org/wiki/PWRficient
Thanks to umisef for spotting it (http://amigaworld.net/modules/newbb/viewtopic.php?topic_id=31890&forum=33)
-
Well that's not bad news, if the SPECint2000 / SPECfp2000 scores are to be believed.
According to one source (I posted on one of the other recent thread), the above scores were 1000 and 1500 respectively, those figures being for just one core and at an average power consumption of 7W.
-
"some pretty cool cpu we don't want to tell you about just yet"
...except you did in the output.
FTW, guys!
-
Well spotted.
Two GigE controllers and a 10GigE controller?
I'm sure that they're not all connected (as they'll require off-chip PHYs which might be quite expensive for the 10GigE controller, and they contend for SERDES ports with the PCIe controller), but these are nice things to have :-).
-
Well spotted.
Two GigE controllers and a 10GigE controller?
I'm sure that they're not all connected (as they'll require off-chip PHYs which might be quite expensive for the 10GigE controller, and they contend for SERDES ports with the PCIe controller), but these are nice things to have :-).
The ethernet controllers share the SERDES pins with the PCI-Express lanes. Considering what we're to believe about the rest of the system in terms of PCI-Express slots and onboard devices, I'd guess that nearly all of the SERDES lanes are consumed by PCI-Express. In the AW.net discussion I remember someone saying there was a Realtek ethernet chip on board as well, though I thought that one odd... Go PCI-Express to it, or use that lane for an SGMII to a phy, I'd go straight to a phy...
-
(http://image.bayimg.com/bangnaace.jpg)
http://pci-ids.ucw.cz/read/PC/1959
http://en.wikipedia.org/wiki/PWRficient
Thanks to umisef for spotting it (http://amigaworld.net/modules/newbb/viewtopic.php?topic_id=31890&forum=33)
Yeah, I stated this back on 1/14/2010. thanks
http://www.amiga.org/forums/showthread.php?t=50937&page=39
-
Yeah, I stated this back on 1/14/2010. thanks
http://www.amiga.org/forums/showthread.php?t=50937&page=39
No, actually you GUESSED it.. And with zero to substantiate the guess... Anyone could have guessed, we now have actual confirmation...
So congrats on a great guess, but don't get too carried away Buddy...
-
"some pretty cool cpu we don't want to tell you about just yet"
...except you did in the output.
FTW, guys!
LOL!
:)
Well, this is what happens when "no-techies" tries to do technology!
Ah well, the PA6T was many people's first hand guess since long, including mine.
Didn't think that PA-Semi had had the time to develop it enough to be fit into a consumer product though. Hmm, did Apple continue its development??
-
actually I made an educated assumption based on the information at the time. It seemed pretty aparent to me. I never get carried away! I also never Guess or Gamble.
-
So, how powerful is it?
http://arstechnica.com/old/content/2007/02/8828.ars
Hopefuly not a double post.
-
actually I made an educated assumption based on the information at the time. It seemed pretty aparent to me. I never get carried away! I also never Guess or Gamble.
Yes you do, but you call it "an educated assumption based on the information at the time", which pretty much is the same tag line *any gambler* would use! ;)
-
So, how powerful is it?
http://arstechnica.com/old/content/2007/02/8828.ars
Hopefuly not a double post.
It's actually not as fast as I would have thought:
http://en.wikipedia.org/wiki/Instructions_per_second
8800 MIPS at 2.0 GHz.
Look at the figures of some of the Intel CPUS.
Having said that, it blows away existing and available PowerPC chips that are currently running AmigaOS 4.
-
@HammerD
It does not sound to shabby to me.
With its low wattage, I wonder if overclocking will have a linear or a non-linear graph with the temp.
-
It's actually not as fast as I would have thought:
http://en.wikipedia.org/wiki/Instructions_per_second
8800 MIPS at 2.0 GHz.
Look at the figures of some of the Intel CPUS.
Having said that, it blows away existing and available PowerPC chips that are currently running AmigaOS 4.
How should one read that table? As the instructions / clock cycle rises quite fast with newer cpus, I assume it is mainly because they have more cores?
-
well it blows away a 68060 @ 66Mhz - which is about 88 MIPS :) lol
-
Yes you do, but you call it "an educated assumption based on the information at the time", which pretty much is the same tag line *any gambler* would use! ;)
You would know best, as I stated I do not Gamble or Guess. If I did, I would guess you know all about me. But then I would be wrong.
-
Here is a pdf that says it has virtualization.
http://www.systerra.de/documents/PA_Semi_PA6T_1682M_PB_694.pdf
-
It's actually not as fast as I would have thought:
http://en.wikipedia.org/wiki/Instructions_per_second
8800 MIPS at 2.0 GHz.
Look at the figures of some of the Intel CPUS.
Having said that, it blows away existing and available PowerPC chips that are currently running AmigaOS 4.
Divide that value by 2 - only one core gets used (at least initially). The G4 1500 MHz is slower, but optimized software probably fills that gap.
I am pretty curious of a real world usage comparision of OS4 on the X1000 vs. a Mac mini 1500 running MorphOS. Unfortunately we will have to wait until winter for that comparison (and who knows maybe we can compare it against a G4 1667 MHz by then)...
-
Divide that value by 2 - only one core gets used
...and is the CPU even available at 2.0GHz in real life, or only theoretical? The "X1000" is announced at 1.8GHz, and IIRC I remember some comments that "now we have successfully tested it at 1.6GHz" (or whatever)...
(at least initially).
Per design, Amiga can't do what outsiders would call *real* SMP, not even "later".
The G4 1500 MHz is slower, but optimized software probably fills that gap.
As already shown in the MorphOS vs. OS4 speed comparison (not that the G4 1500MHz will be faster, but that optimized software wins *on the same HW*, and thus will reach up beyond the "unoptimized" software).
I am pretty curious of a real world usage comparision of OS4 on the X1000 vs. a Mac mini 1500 running MorphOS. Unfortunately we will have to wait until winter for that comparison (and who knows maybe we can compare it against a G4 1667 MHz by then)...
Bang for the Buck!
-
@tmhg
2.0GHz is frequently quoted in old PASemi notes. I suspect the 1.6/1.8 GHz speeds in the X1000 may be something more to do with the fact it is passively cooled.
-
@tmhg
2.0GHz is frequently quoted in old PASemi notes. I suspect the 1.6/1.8 GHz speeds in the X1000 may be something more to do with the fact it is passively cooled.
Well OK (did never pay that much attention), but let's see *what gets delivered* in the end. The thing isn't even remotely here yet!
:)
-
Well OK (did never pay that much attention), but let's see *what gets delivered* in the end. The thing isn't even remotely here yet!
:)
What interests me, is how can it be a new off the fab line....PA6T.
-
HyperionMP:
"It is unwise to spread this type of unconfirmed rumours.
Whatever people believe they have seen or not seen, I certainly do not want any disappointed users blaming A-EON and/or Hyperion Entertainment for speculation by third parties.
It should be pointed out that measures were taken to throw up "funny answers" when people tried to find out the identity of the CPU."
http://amigaworld.net/modules/newbb/viewtopic.php?topic_id=31896&forum=33#566619
Ah well, we'll see who will be eating the socks in the end...
:)
-
What interests me, is how can it be a new off the fab line....PA6T.
Well, is it?
-
What interests me, is how can it be a new off the fab line....PA6T.
I think there are rumours that Apple allowed another company to take production of the Pa6T for the contracted and warranted customers. Maybe they even were allowed to do some advancments like a die shrink with a transition to another process (45nm). But I don't know if there is any truth in those rumours...
Anyway the PA6T unfortunately (it actually is a nice processor) is a dead end and probably very expensive (well, it gets produced in low quantity and gets primarily sold to mil customers - who have a different approach to calculate costs).
-
What interests me, is how can it be a new off the fab line....PA6T.
Apple bought PASemi, who designed the PWRFicient, but PASemi were a fabless company. I don't recall off the top of my head who was responsible for fabrication, but when buying the company, it would seem reasonable that Apple would have to honour any existing production contracts PASemi / their fabrication partner may have had at the time.
-
The question I have, where do you go from here since there is no follow up chip to the PA6T, so you've got to choose altogether a new chip (even if its ppc based). There is not another ppc chip even in the pipeline for the power/watt combo that PA6T has, so immediately you have painted yourself into a corner regarding a follow up.
-
Apple bought PASemi, who designed the PWRFicient, but PASemi were a fabless company. I don't recall off the top of my head who was responsible for fabrication, but when buying the company, it would seem reasonable that Apple would have to honour any existing production contracts PASemi / their fabrication partner may have had at the time.
Yup.
http://www.theregister.co.uk/2008/05/16/pasemi_apple_support/
-
The question I have, where do you go from here since there is no follow up chip to the PA6T, so you've got to choose altogether a new chip (even if its ppc based). There is not another ppc chip even in the pipeline for the power/watt combo that PA6T has, so immediately you have painted yourself into a corner regarding a follow up.
The answer is very clear. A 1.6 TeraHz Multicore CPU with x86 & PPC embedded cores.
-
I think most users will be underwhelmed with its performance. 8,800 MIPS at 2.0 GHz is about where the leading Pentium processors were back in 2003.
Sounds like there was a surplus of them from 2007 laying around and A-Eon was able to buy them up at a discount and them charge Amiga fanatics a premium price.
-
I think most users will be underwhelmed with its performance. 8,800 MIPS at 2.0 GHz is about where the leading Pentium processors were back in 2003.
It's a question of perspective. MIPS are not the be all and end all. The overall processor and it's memory interface seems a lot better performing than any G4 hardware that currently runs OS4.
A much better indication of performance can be had through running real code. The SPEC CPU2000 (http://www.mrob.com/pub/comp/benchmarks/spec.html#CPU_00) (which are based on several real-world code tests) results for the PA6T appear to rival or outperform figures quoted for PPC970:
PPC970 info here (http://www.realworldtech.com/page.cfm?ArticleID=RWT101502203725&p=2)
(http://www.realworldtech.com/includes/images/articles/powerpc970-2.gif)
The stated performance of a single core in the the PA6T at 2GHz is:
SPECInt2000: 1000
SPECfp2000: 1500
Now let's look at a 1GHz G4 Mac : See here (http://web.archive.org/web/20071201234833/http://www.heise.de/ct/english/02/05/182/)
SPECInt2000: 306
SPECfp2000: varied: 147 - 187
So, even downscaling the results for the PA6T for an estimate at 1.6 GHz, we get:
SPECInt2000: 800
SPECfp2000: 1200
So, that's basically 2.6x the integer performance and 6.4x the (best) floating point performance of a 1GHz G4. In other words, the 1.6GHz PA6T is to the 1GHz G4 what the average 50MHz 060 is the 25MHz 040. Except that you get two of them (if SMP in OS4 ever materialises).
I dunno about anybody else, but the speed up from an 040 to an 060 was pretty conspicuous.
-
Well, only the most fanatical users will pay $2225 USD for technology that performs on par with systems back in 2003. I guess if your hobby is buying obsolete hardware, then knock yourself out.
-
I'm surprised that all of you didn't realize this yesterday when Aeon's design partner was announced. I discussed PA Semi based systems with Varisys months ago (they build a PA6T based CPU card). Once Aeon announced that Varisys helped design the X1000 motherboard, the PA6T was the only CPU that could be on the board on (considering the specs quoted and Varisys' design experience).
Plus, there were clues all over pointing to this for months. What other unique product does Varisys work with? XMOS processors!
Funny, everyone using MorphOS had this one pegged months ago.
-
Well, only the most fanatical users will pay $2225 USD for technology that performs on par with systems back in 2003. I guess if your hobby is buying obsolete hardware, then knock yourself out.
Well, reflecting on your statement I bought (well, built) a nice quad core PC 2 years ago (see sig). It cost lest and is vastly more powerful than any G4 class machine. So, well done me. That was a job well done.
Does it run OS4? No.
Does it run MorphOS? No.
Does it run AROS? Sort of.
So, if my interest was to run OS4 or MorphOS, said system is useless. All the "much faster hardware is available for far less" arguments don't really change that.
-
Funny, everyone using MorphOS had this one pegged months ago.
When they first said low power, dual core 64-bit PPC at 1.6-1.8 GHz, it was my first thought. It's in that "what is X" thread somewhere ;)
-
When they first said low power, dual core 64-bit PPC at 1.6-1.8 GHz, it was my first thought. It's in that "what is X" thread somewhere ;)
Yep, that was too easy a clue. I had an XMOS development board on hand at that point. I was confused by the specs on the PPC component (could have been an AM Titan core or PASemi), but the XMOS reference was clear.
When they announced the XMOS component I felt a little more assured, but the Varisys announcement cinched it. That company offers to help create PASemi based designs. Bingo!
-
... I guess if your hobby is buying obsolete hardware, then knock yourself out.
er... You do realize where you are, right???
:roflmao::roflmao::roflmao:
desiv
-
Couldn't have put that better myself desiv!
-
Are they able to stay in business long enough to make it work?
I do not know but I hope so.
-
I wonder if/when they will come up with a demonstration of the Xena/xcore chipset's capabilities. Otherwise it's going to take a year or two at least before some interesting apps come out that use the chipset after the sale of the X1000 begins (by which time judgement of X1000's success will have already been judged).
It's all very good to say "look! amazing chipset! part of the reason why this is so expensive", but if they want to attract some serious interest, there has to be a demo as well. No, I'm not talking about a 21st century juggler :) But in terms of potential customers saying "wow", it needs to be something that has a similar relative impact.
-
I wonder if/when they will come up with a demonstration of the Xena/xcore chipset's capabilities. Otherwise it's going to take a year or two at least before some interesting apps come out that use the chipset after the sale of the X1000 begins (by which time judgement of X1000's success will have already been judged).
It's all very good to say "look! amazing chipset! part of the reason why this is so expensive", but if they want to attract some serious interest, there has to be a demo as well. No, I'm not talking about a 21st century juggler :) But in terms of potential customers saying "wow", it needs to be something that has a similar relative impact.
The XMOS chip in the "X1000" will probably be as much fun as the FPGA on the Sam.
-
The question I have, where do you go from here since there is no follow up chip to the PA6T, so you've got to choose altogether a new chip (even if its ppc based). There is not another ppc chip even in the pipeline for the power/watt combo that PA6T has, so immediately you have painted yourself into a corner regarding a follow up.
IBM are still actively developing PPC architecture, ALL three games consoles out there use a PPC in some form. We have Xenon, CELL and Broadway. OK Broadway is rubbish and pre G4 speeds but that still leaves two completed CPUs that are PPC compatible. I doubt very much Microsoft would change from PPC back to Intel either, so expect to see a revised Xenon II within a few years (not the rubbish 16 colour s-l-o-w game from the bitmap bros!!)
As others have said it is really an issue of 'will they go bankrupt?' not 'I can build an i7 PS3 beating gaming rig for £500 less!' though. And it is the fastest OS4 box you can buy so if that's all you want and money is no object then have a nice time :)
Working AmigaONEs for sale are pretty rare and for some people SAM isn't fast enough. I totally agree, a 933mhz AmigaONE is probably my limit to how 'simple' I would go if I was to build a machine.
-
"some pretty cool cpu we don't want to tell you about just yet"
...except you did in the output.
FTW, guys!
They didn't "just miss" that, they replaced showconfig with an exe that just printed a message and quit. If thats genuine showconfig output, then someone must have found a copy of the real showconfig somewhere.
-
Maybe someone could put ppc linux on it.
Hopefully it would show what it could really do.
-
The XMOS chip in the "X1000" will probably be as much fun as the FPGA on the Sam.
You know, if you ever get tired of repeating your general opinion, a bot could be written that does it for you :)
-
I wonder if/when they will come up with a demonstration of the Xena/xcore chipset's capabilities. Otherwise it's going to take a year or two at least before some interesting apps come out that use the chipset after the sale of the X1000 begins (by which time judgement of X1000's success will have already been judged).
Never? A-Eon doesn't care about the xs1-l1. It was marketing hype to them, they never even bothered to have the damn chip soldered onto the prototype board. For good reason too, it's completely irrelevant in a desktop system. It blew me away the hype they spun about it.
-
The XMOS chip in the "X1000" will probably be as much fun as the FPGA on the Sam.
It will be worse, a few idiots will probably try to make use of it and all they will achieve will be a wank and a layer of incompatibility with no actual gain for having done so.
-
A-Eon have made it clear several times that the XMOS chip was Varisys' idea, rather than theirs and has been put on as an almost free extra (£10 or so on a £1500+ board).
Yes, they talked about it in their marketing blurb, but they have consistently avoiding saying what it could be used for and instead made it clear that it would be up to the users to find interesting things to do with it.
The hype came from certain over-enthusiastic community members, NOT A-Eon or Hyperion.
-
A-Eon have made it clear several times that the XMOS chip was Varisys' idea, rather than theirs and has been put on as an almost free extra (£10 or so on a £1500+ board).
They have? Where?
The hype came from certain over-enthusiastic community members, NOT A-Eon or Hyperion.
Have you already forgotten about the "What is X?" nonsense?
-
A-Eon have made it clear several times that the XMOS chip was Varisys' idea, rather than theirs and has been put on as an almost free extra (£10 or so on a £1500+ board).
Yes, they talked about it in their marketing blurb, but they have consistently avoiding saying what it could be used for and instead made it clear that it would be up to the users to find interesting things to do with it.
The hype came from certain over-enthusiastic community members, NOT A-Eon or Hyperion.
It's the main selling point on the A-eon website. Since they won't talk about the CPU, "Xena" and "Xorro" are the only way the X1000 has to distinguish itself. The whole color scheme of the original "X" logo came from the XMOS website.
Just because nobody has yet managed to come up with a purpose for the XMOS chip to be on the mobo doesn't mean it isn't A-eon's main marketing ploy.
-
They have? Where?
In at least 2 or 3 posts on AW.net and iirc in an AF interview.
I might fish out the references if work is really slow tomorrow...
Have you already forgotten about the "What is X?" nonsense?
Nope - but I read that as generating interest in something a little unusual (which the XMOS is) rather than promising world-changing features and I don't recall them saying what the XMOS might be actually used for - the earlier versions of the website specifically said it would be up to the users to find uses for it.
-
In at least 2 or 3 posts on AW.net and iirc in an AF interview.
Also at the A-Eon site: Our hardware designers had a brilliant idea: "Why not add an XMOS chip?"
And the rest of their text about xena:
"Once there were custom chips; for the AmigaOne X generation, we have customisable chips. XMOS calls it "Software Defined Silicon", we call it 'Xena', a nod to the old custom chip names. It's the inheritor of the 'transputer' concept, and it's something we're quite excited about.
Capable of eight concurrent real-time threads with shared memory space, at up to 500 MIPS, Xena gives the X1000 a very flexible, very expandable co-processor. The uses are endless; control hardware, DSP functions, robotics, display - even SID chip and console emulators.
Xena is not simply strap-on extra adding an extra half GHz of processing power, it's a different kind of thing to a general purpose CPU altogether. It's an event-driven processor, which means it can respond immediately to events such as external signals, rather than having to wait on an interrupt. This makes it appropriate to true real-time functions. It has many input/output lines which are software configurable, making it ideal for ultra-low latency data sampling applications and extremely easy to turn into control hardware for... well, virtually anything. The I/O can also be configured to communicate with extra XMOS chips that can run the processor's code in a highly parallel fashion, and for serious power applications you can just keep on adding processors.
The Amiga has seen some truly ingenious hacks and add-ons; Xena can take this to a whole new level. It will take a while for the full possibilities to be realised, but we urge you to visit XMOS and discover more for yourselves."
Seems pretty ok description of it's capabilities. In the teaser phase they were perhaps more "enthusiastic".
Also I recommend to read their xorro description as well. It has about all the public details of the connections of/to xena.
-
Well, only the most fanatical users will pay $2225 USD for technology that performs on par with systems back in 2003. I guess if your hobby is buying obsolete hardware, then knock yourself out.
Agree it's comparable to 2003 Intel Pentium speeds but for a box that can run Amiga OS that's impressive.
BTW does the estimated $2222 US include OS4 ?
-
The hype came from certain over-enthusiastic community members, NOT A-Eon or Hyperion.
Then why are they calling it a X1000 then? They even hyped it in the system's name.
-
IBM are still actively developing PPC architecture, ALL three games consoles out there use a PPC in some form. We have Xenon, CELL and Broadway. OK Broadway is rubbish and pre G4 speeds but that still leaves two completed CPUs that are PPC compatible. I doubt very much Microsoft would change from PPC back to Intel either, so expect to see a revised Xenon II within a few years (not the rubbish 16 colour s-l-o-w game from the bitmap bros!!).
IBM/LSI, AMCC and Freescale have all announced new chips recently, it isn't dead by any means. PPC isn't targeting the desktop anymore but they are getting faster. There is an upgrade path but with Amiga volumes don't expect them to be cheap.
They wont be competitive with most modern PCs but I don't think they'll be quite so bad as some are suggesting. The expect X1000 is going to be a lot better than a 2003 vintage PC. I don't know how the core will compare but it has a dual channel on-die memory system - similar to that found in today's mainstream PCs.
Unfortunately SPEC has not had a good reputation for a long time and it's got a lot worse recently. The only way to find out real performance is to try it and see.
-
Then why are they calling it a X1000 then? They even hyped it in the system's name.
"monkey bazooka" sounded more awkward
-
Lets get some points out the way.....
* Xena is likely a method to "lock in" to the hardware, custom chip custom encryption, custom security, and hard to copy.
* 99% of all X1000's will remain in their original packaging and retained by collectors, its obsolescence is irrelevant, it's a collectors piece, for vintage uber geeks, thats the target market and they may not even been fans of the Amiga......
* Its compatibility, performance, price and usability are all irrelevant as its not going to exit the packaging. It will be resold at increasingly escalated prices and a short production run is an absolute certainty.
X1000 = Fiat 500 Abarth (Gutless piece of **** but colour coded and stylish), If Clarkson reviewed it for the cool wall it would a jewel around Kristin Scott Thomas's neck, polished and kept in a safe and brought out on occasion to polish its packaging.
I can't see how anyone can disagree with this.......isn't it obvious ? ;)
-
IBM/LSI, AMCC and Freescale have all announced new chips recently, it isn't dead by any means. PPC isn't targeting the desktop anymore but they are getting faster. There is an upgrade path but with Amiga volumes don't expect them to be cheap.
They wont be competitive with most modern PCs but I don't think they'll be quite so bad as some are suggesting. The expect X1000 is going to be a lot better than a 2003 vintage PC. I don't know how the core will compare but it has a dual channel on-die memory system - similar to that found in today's mainstream PCs.
Unfortunately SPEC has not had a good reputation for a long time and it's got a lot worse recently. The only way to find out real performance is to try it and see.
I doubt very much that a machine based on the tri core sextet thread capable IBM Xenon CPU as in the £125 xbox 360 @ 3.2ghz would cost more than £400/$500 etc. It is also significantly faster than the existing G5 based x1000 by a huge margin...ie current quad core PC speeds in reality.
And Sony/Toshiba/IBM are not going to let CELL die, again with over 45 million CPUs produced for various applications price is not a sensitive issue as with old G5 PPC architecture stuff.
Anyway I did a bit of research on the XMOS co-pro, it can execute 8 threads on each of the 4 cores yes but only two threads in reality (ie two every 1/4 of the total CPU speed so 100mhz processing per pair for a 400mhz XMOS running 8 threads on a core).
The problem is the OS has no support for such esoteric hardware. Maybe the XMOS would make a good geometry setup engine like a baby version of the CELL SPUs on PS3 for your GPU but who knows. And a quick and dirty kludge like WarpOS or PowerUP is not going to work well.
The real issue is the people making the OS and the hardware need to be the same damned company in step with each other 100% of the design phase, look how many revisions it took MS from Windows v 1.0 to achieve effective and useful multitasking. Using Dr Tim King's TriPos for Amiga in 1985 was nowhere near as complicated as integrating a dual core cpu and quad core co-pro with the ability to run 32 threads integrated into a single thread single core OS like OS4 to be sure my friends. I don't envy the programmers tasked with that and given the tiny sales of x1000 it's not going to be Hyperion I bet.
-
Lets get some points out the way.....
* Xena is likely a method to "lock in" to the hardware, custom chip custom encryption, custom security, and hard to copy.
* 99% of all X1000's will remain in their original packaging and retained by collectors, its obsolescence is irrelevant, it's a collectors piece, for vintage uber geeks, thats the target market and they may not even been fans of the Amiga......
* Its compatibility, performance, price and usability are all irrelevant as its not going to exit the packaging. It will be resold at increasingly escalated prices and a short production run is an absolute certainty.
X1000 = Fiat 500 Abarth (Gutless piece of sh*t but colour coded and stylish), If Clarkson reviewed it for the cool wall it would a jewel around Kristin Scott Thomas's neck, polished and kept in a safe and brought out on occasion to polish its packaging.
I can't see how anyone can disagree with this.......isn't it obvious ? ;)
I don't disagree with any of that except that the Abarth Fiat 500 is the best in its class....if you want a euro shoebox of a car or have a tiny garage ;)
Anyway the issue is A-EON are marketing it as the second coming, and that's why you get such a huge disparity in the reception received by Amiga users who lived in the Reagan years of A1000 and know for a fact how it absolutely slam dunked EVERY desktop computer available in 1985 in every department at any price. Interesting yes, expensive yes, FA to do with Amiga 1000 or any kind of spiritual successor to A1000 vs Mac vs ST vs 286 EGA PC. Sure the ST was cheaper but it was like comparing an M3 and a rolling chassis of an M3 with no engine or drivetrain lol it was the bare minimum for a computer even in 1985 with zero custom hardware and an off the shelf non multitasking GUI.
-
@Piru
So, you'll be lining up to buy one as it comes out ?
-
"8,800 MIPS at 2.0 GHz is about where the leading Pentium processors were back in 2003."
So its as fast as a computer I picked out of the trash recently, a dell pentium 2ghz machine...
Amd has new 64 bit 3ghz chips with 6 and 8 cores that will cost 300$ for 6 cores or 400$ for 8 cores, and remember thats brand new bleading edge tech price. 18 months after release, they will be half that price.
I sure don't see an x1000 in my future, I'd rather for 900$ buy a new linux box with 8 processing cores,12 gigs of ram and 2 terrabytes of hard drives.
For the same 900$ I can buy a sam board and run amiga os on hardware thats slower than 10 year old pc hardware. I just don't see the whole lets stick with ppc, its just retarded at this point, in my view anyhow.
Even people like me who are huge amiga fans can not justify the expense based on how outdated the hardware is. They are killing any chances it had by offering substandard slow ass hugely inflated priced hardware.
-
"
"8,800 MIPS at 2.0 GHz is about where the leading Pentium processors were back in 2003."
So its as fast as a computer I picked out of the trash recently, a dell pentium 2ghz machine...
"
At what wattage, 95 watts ? Even the lower power ones are about 45 Watts, no comparison.
-
Anyway the issue is A-EON are marketing it as the second coming,
I think that's because they have now the control of Amiga & OS developments.
First time since 199x AOS devs have control of AOS future.
And the HW is a big jump forward in terms of performance (IMO, it's 5...10 fears forward from SAM) and up-to-date expansion slots (15 years forward from SAM) + GPGPU etc.
(xena is there mainly/just to bring some I/O flexibility over standard HW, like the video slot of original Amigas, waiting for people to utilize it, or like the geekport of bebox, just like they seem to say now, after the tease is over, since january)
about CPU performance:
It should be like p3@2Ghz. A lot better than P4@2Ghz. + it has integrated RAM controllers + integrated PCIe bridge + more cache , unlike those 2ghz pentiums, unlike any other PPC on desktop. It should significantly outperform systems like original xbox360. And whole system (HW) is more modern than the last G5 mac. Only time will tell how many years it takes for the SW to support it fully.
@everybody
A-Eon does not seem to plan to sell many hundreds of x1000. Surely they "just" pave the way for next gen multicore HW & ASMP AOS4.
-
And yet a far more apt description somehow.
"monkey bazooka" sounded more awkward
-
GPGPU
Isn't that a bit premature considering there isn't even 3D support for the included Radeon X700 card yet?
As far as I can tell both ATI and nVidia are keeping their GPU stream processing drivers closed source (and binaries are only available for x86 and amd64). So where is this GPGPU stuff coming form then?
-
Per design, Amiga can't do what outsiders would call *real* SMP, not even "later".
By the wikipedia definition of SMP, even classic AmigaOS can be extendet to do real SMP. Are you perhaps thinking about NUMA?
-
By the wikipedia definition of SMP, even classic AmigaOS can be extendet to do real SMP. Are you perhaps thinking about NUMA?
Oh dear gods no...
AmigaOS as it is today, either in classic or NG cannot and likely will never support SMP. The APIs simply do not allow for SMP and would need major reworking before SMP is possible.
-
Using a gang scheduler for SMP would allow Forbid/Permit-style disabling of the multitasking to be implemented. Nasty code that uses Forbid to speed things up will have the dubious distinction of being slower though because it will still run on one core only.
Personally, I think the FPGA approach to expanding the Amiga (eg. making slave processors to the master single-threaded CPU) is the best way but SMP is certainly possible.
-
Right up until something running in one core wanted to talk to something running in another.
-
Using a gang scheduler for SMP would allow Forbid/Permit-style disabling of the multitasking to be implemented. Nasty code that uses Forbid to speed things up will have the dubious distinction of being slower though because it will still run on one core only.
Personally, I think the FPGA approach to expanding the Amiga (eg. making slave processors to the master single-threaded CPU) is the best way but SMP is certainly possible.
There's no reason why AmigaOS couldn't do something entirely different. More than a few folks are working on system designs that don't rely on cache coherency and related paradigms to do local parallel processing. And I'm not talking about co-processors or other types of dedicated out-of-band hardware.
-
Using a gang scheduler for SMP would allow Forbid/Permit-style disabling of the multitasking to be implemented. Nasty code that uses Forbid to speed things up will have the dubious distinction of being slower though because it will still run on one core only.
The problem is that there is a large proportion of the Amiga's software catalog falls into the "nasty code" category - mostly for optimization reasons. If you implement SMP, you have to break API compatibility. To attempt to have both would at best leave you with a system with (application and probably system) stability somewhat south of Charles Manson.
At which point you may as well just go over to Haiku or some other OS that has these features built into it from the beginning and run UAE.
--edit--
Tbh I think the Dragonfly's experimental internal clustering model would probably be easier to apply to th
-
Using a gang scheduler for SMP would allow Forbid/Permit-style disabling of the multitasking to be implemented. Nasty code that uses Forbid to speed things up will have the dubious distinction of being slower though because it will still run on one core only.
Personally, I think the FPGA approach to expanding the Amiga (eg. making slave processors to the master single-threaded CPU) is the best way but SMP is certainly possible.
That´s what I was thinking. Note though, that even the nasty code will have the benefit of the fastest single core Amiga ever. So quit your legacy apps and you have full multicore speed or run them and expect twice the performance impact they would normally have. This is certainly faster than any emulation.
-
So quit your legacy apps and you have full multicore speed or run them and expect twice the performance impact they would normally have. This is certainly faster than any emulation.
Except that your legacy apps will always need emulation to some degree. They're 68k remember.
And if you're going down that road then I'm sorry to say that it's been done before and better courtesy of Amithlon. Which had the added benefit of not running on a 10 times overpriced under performing hardware platform.
-
Using a gang scheduler for SMP would allow Forbid/Permit-style disabling of the multitasking to be implemented. Nasty code that uses Forbid to speed things up will have the dubious distinction of being slower though because it will still run on one core only.
I think gang scheduler would work if Forbid()/Permit() would only be used by user programs. Problems is that Forbid()/Permit() pairs are used throughout drivers, system libs and user programs. Therefor I think a gang scheduler will either severely limit heavy IO task by CPU intensive tasks or the other way around.
In the end you will need to replace all the Forbid()/Permit() around IOs by specific locking mechanism in order to combine heavy IO with efficient SMP.
greets,
Staf.
-
What if you had enough *native* apps to get rid of legacy ones ?
You will always be able to run them with UAE anyway...
Why are we still talking about *legacy* in 2010 ? We need to move forward...
-
What if you had enough *native* apps to get rid of legacy ones ?
You won't. The potential market is way too small for serious commercial apps. If your software library consists of ports of X or linux apps why not run linux directly then?
-
You won't. The potential market is way too small for serious commercial apps. If your software library consists of ports of X or linux apps why not run linux directly then?
Really ? As I see it most important 68k apps have already been ported...
A PPC-only MorphOS (plus UAE for older apps/games) will be enough already.
What if someday MorphOS gets ported to some new architecture ? Does this mean we'll have:
1. 68k emulator
2. ppc emulator
3. new xx processor native apps
?
-
Really ? As I see it most important 68k apps have already been ported...
?
Huh ?
Name one paint/gfx-app that has been ported over to PPC (OS4 or MorphOS) and is still actively maintained.
Name one word-processor .....
There is some movement in sound&music, but still lots of important apps left behind.
And no running badly ported Linux-apps by installing half-a-linux on OS4 doesn't cut the beef .....
-
I doubt very much Microsoft would change from PPC back to Intel either,
It's a difficult call. Microsoft went to IBM because they benefited from Sony's R&D budget. Sony might not want to get screwed by IBM again.
-
Isn't that a bit premature considering there isn't even 3D support for the included Radeon X700 card yet?
As far as I can tell both ATI and nVidia are keeping their GPU stream processing drivers closed source (and binaries are only available for x86 and amd64). So where is this GPGPU stuff coming form then?
No idea.
-
You won't. The potential market is way too small for serious commercial apps. If your software library consists of ports of X or linux apps why not run linux directly then?
Well, I'm not a coder but simply a user, and as I'd say not a novice user, every Linux distro I've used *feels*, and *looks* well, yuck.
GNOME looks like a homemade amateur GUI and just doesn't feel responsive. KDE 3 was OK, but KDE 4 is rubbish.
-
No idea.
Why did you bring it up up then (GPGPU)? Did you read somewhere that Hyperion plans to support it?
-
Name one paint/gfx-app that has been ported over to PPC (OS4 or MorphOS) and is still actively maintained.
Name one word-processor .....
There is some movement in sound&music, but still lots of important apps left behind.
There's none. But name me one 68k word processor or gfx/paint app that's been updated in the last 5 years ?
I guess there's none... I'd rather see some serious/rather up to date app like Blender (for 3D) ported to native, rather than being able to use a useless legacy app...
And for that we don't need legacy apps...
Who's using Lightwave on a Pegasos ?