ARM or x86 with FPGA emulator

[Kremlar]

Then you just supported my point that you're here to troll and nothing more.

No, I'm here to discuss a topic on a public message board.

Yeah, because everyone should be like Apple. Bunch of garbage they produce since Leopard. No, just no. That will fail just as hard.

So... Apple produces garbage, and Lemote systems are fantastic?  Just trying to set a baseline here.

MIPS and SPARC aren't obsolete.

They are both obsolete and irrelevant on the desktop.

[TeamBlackFox]

Matthey, I'll check it out. Thanks. I've been meaning to get a faster CPU, but that will take a while with my money.

And in regards to that tidbit on 32 vs 64-bit overhead, I  find 32-bit unacceptable for a modern computer. While many Amigans may be happy with that: I will not be, because I primarily use BSD. I have no use for AmigaOS as my daily driver. It is a distraction  and game machine, nothing more.

I agree though that ARM may be a decent target, but I'd have to wait one more ISA generation because I'm not interested in portable computers much, I want workstations and servers. I own a Nexus 7 and a LG Optimus F6 but I'd never want that to be running AmigaOS or AROS as they're already buggy enough. Much of the time they just function as portable media players and remote terminals

[bloodline]

Yeah but both drop compatibility relatively quick. Stuff from the 2000/XP era won't always run due to ABI changes, and since its closed, recompiling against the new ABI isn't possible. Likewise with Apple Rosetta is dead. Apple is a status symbol, it is a symbol of eliteness and that's why stupid people buy it. People with brains like me use Android and cheaper alternatives that work just as well.

Both drop compatibility when developers have caught up with the new architecture. That's simply good business sense.

Apple Rossetta is dead because it doesn't make any sense to keep it going... I don't even have any 32bit x86 Apple apps anymore, let alone any PPC apps.

People buy what they can afford which suits their needs, I'm sure your device suits your needs and fits in your budget, my device does the same. There is no need to be pejorative about anyone's choice.

[TeamBlackFox]

Kremlar, you've worn my patience thin. This is my last response to your post on this thread. You're trolling and I don't appreciate the smart attitude and the high horse approach.

Apple produces expensive rubbish both in hardware and software. Its just an overpriced paperweight running a Mach kernel with proprietary bits and dressed up to look like UNIX. Lemote systems are decent machines, especially from an open hardware perspective. Quality with the first few were lacking but that has somewhat subsided since. I don't care for the main OS that runs on them, but their open nature makes them easily accessible and malleable to various applications.

On the desktop market, SPARC and MIPS are irrelevant, yes, but obsolete? No. There are plenty of chips that would make fine desktop computers. The pieces are out there, it is just up to someone to put them together.

Anyways conversation over. Troll somewhere else, shoo!

[TeamBlackFox]

Both drop compatibility when developers have caught up with the new architecture. That's simply good business sense.

Apple Rossetta is dead because it doesn't make any sense to keep it going... I don't even have any 32bit x86 Apple apps anymore, let alone any PPC apps.

People buy what they can afford which suits their needs, I'm sure your device suits your needs and fits in your budget, my device does the same. There is no need to be pejorative about anyone's choice.

The only reason I give a damn is because their choices influence my range of choices.

I agree that Apple Rosetta has no point, but MS still has people running XP and below! Rather than abruptly cut off the stragglers after a warning or two, they end up weeding them out by hand. It makes no business sense to me, but I don't care for collectivist business approaches anyways.

The fact I can run DOS on a modern x64 laptop is laughably retarded. I still to this day find the addressing modes of x86 making no sense.

[Kremlar]

People buy what they can afford which suits their needs, I'm sure your device suits your needs and fits in your budget, my device does the same. There is no need to be pejorative about anyone's choice.

Right.  Different tools for different jobs.  My phone runs Android, but I'd recommend an iPhone to most non-technical people.

[OlafS3]

Then you just supported my point that you're here to troll and nothing more.


All that needs to be done is increase the cost/performance ratio of the hardware, I have lost faith in POWER considering the cost of an IBM POWER server, and the lack of top of the line designs from POWER6+ and up.

Yeah, because everyone should be like Apple. Bunch of garbage they produce since Leopard. No, just no. That will fail just as hard.



MIPS and SPARC aren't obsolete.

I am a 68k fan too, a troll how you call it. My main computers are PCs and I also earn my money with them using Windows.

What Kremlar wanted to explain is that NG (AmigaOS and MorphOS and AROS) are only used by a minority of Amiga users right now. I myself use my own distribution based on Aros 68k in UAE, I do not expect it to replace Windows or Linux, it is just fun to play around with it. A new generation of FPGA devices would have that fun and geek factor for many amigans too, none of these devices will beat a modern PC or your workstations by specification but nevertheless it would be fun to use and I am convinced that even people outside the community will buy it.

Regarding SPARC and MIPS, AROS is the only platform that has proven to be portable, if I remember right only one developer ported it in 3 months to ARM in his spare time. So if you want to get something ported to new platforms set up a bounty and convince people to donate, perhaps even some from outside. I can make a bet with you that neither AmigaOS nor MorphOS will ever be ported to MIPS or SPARC.

I see that you very much dislike X86 (or better hate) but as long as you cannot show benchmarks where SPARC or MIPS outperform the big platforms by magnitudes it will be a hard sell. I also told you that it would be expensive (expecially for Hyperion). Regarding Trevor (a-eon) I do think that investing money to make profit is his motivation, he creates his own personal "toys". Profit, market share and so on is certainly not important for him otherwise he would not act like he does. Besides it is a little funny that you are talking about that he should do this or that to sell more because I do not have the impression that you are thinking in economic terms.

And regarding supporting obsolete technologies, that is the difference between someone only being in contact with machines or with the real market. Not long ago one of my customers still used a MS-DOS application. Microsoft supports obsolete technologies not because they are retro but because customers request it otherwise they will not buy new versions of Windows.

[bloodline]

Apple produces expensive rubbish both in hardware and software. Its just an overpriced paperweight running a Mach kernel with proprietary bits and dressed up to look like UNIX.

My favourite development language is objective-c and I like the power of a 64bit CPU in my mobile devices... Apple gives me both of these things, oh and their OS is a Unix, it's POSIX certified.

[TeamBlackFox]

I didn't call 68k users trolls.

I do realise NG platforms are a minority use. As far as the people using the legacy AmigaOS gp, they're free to do what they want. I've got a 3000, but I don't do much with it.

Anyways, SPARC does pretty well versus Intel, but I don't know enough about hardware to articulate it. I was an electrical engineer in college but quit due to academic issues so I don't know how to exactly compare them properly. The only reason Intel/AMD CPUs are so powerful is the amount of money poured into them. It is like putting a V12 in a Geo Metro. You can do it, but you'd be better off going for system properly designed. A lot more sustainable than modding the hell out of a Geo Metro.

As far as the future of NG Amiga goes I'm unable to participate until that cost/performance ratio goes down, and if I'm not gonna be able to do that on RISC, then I'm going to stick with platforms that show more promise to me on RISC.

CISC just doesn't make sense anymore and there is support for this in the simplicity of RISC. Look at anything else engineering related and you'll see that generally the simpler solution is better.

CISC vs RISC
AR15 vs AK-M
Systemd vs rc init
PlayStation vs Saturn
Gasoline vs Diesel engine

[TeamBlackFox]

My favourite development language is objective-c and I like the power of a 64bit CPU in my mobile devices... Apple gives me both of these things, oh and their OS is a Unix, it's POSIX certified.

I've been over this: Certified UNIX is different from UNIX in the descendant sense.

Kernel is different ( fork of Mach called XNU - X is not UNIX )
Mach is a microkernel developed at CMU, it isn't UNIX which was developed at Bell Labs then forked by Berkeley and AT&T proper as BSD and System V.

NEXTSTEP, Darwin and OSX are POSIX compliant, but not UNIX in the descendant sense. You can dress up Mach all you want but it isn't UNIX.

By this token, AIX, Tru64 and other UNIXes which use fully custom kernels aren't UNIX either in my sense and that is OK in my book.

Objective C is better than C++ and C# but there are some reservations I have about its design.

[OlafS3]

I didn't call 68k users trolls.

I do realise NG platforms are a minority use. As far as the people using the legacy AmigaOS gp, they're free to do what they want. I've got a 3000, but I don't do much with it.

Anyways, SPARC does pretty well versus Intel, but I don't know enough about hardware to articulate it. I was an electrical engineer in college but quit due to academic issues so I don't know how to exactly compare them properly. The only reason Intel/AMD CPUs are so powerful is the amount of money poured into them. It is like putting a V12 in a Geo Metro. You can do it, but you'd be better off going for system properly designed. A lot more sustainable than modding the hell out of a Geo Metro.

As far as the future of NG Amiga goes I'm unable to participate until that cost/performance ratio goes down, and if I'm not gonna be able to do that on RISC, then I'm going to stick with platforms that show more promise to me on RISC.

CISC just doesn't make sense anymore and there is support for this in the simplicity of RISC. Look at anything else engineering related and you'll see that generally the simpler solution is better.

CISC vs RISC
AR15 vs AK-M
Systemd vs rc init
PlayStation vs Saturn
Gasoline vs Diesel engine

As I said Aros is very portable, if you want something on SPARC or MIPS set up a bounty for it and convince people to donate to get it ported. That is the only realistic advice I can give. Or directly contact Trevor D., he is a kind person, perhaps you can convince him to finance the port of AmigaOS (but I have high doubts there). But of course you must offer more than emotions but hard facts and advantages that outweigh the disadvantages of such a change.

And if you do not like AROS and do not want to buy a AmigaOS system you could still buy MorphOS and a used Mac. If none is interesting for you but only AmigaOS ported to SPARC/MIPS it becomes difficult.

And why does CISC make no sense anymore? They are very living and except ARM there is nothing that plays a role at all.

[matthey]

Matthey, I'll check it out. Thanks. I've been meaning to get a faster CPU, but that will take a while with my money.

Maybe there will be a cheaper big box 68k fpga accelerator sometime next year with 128MB or more of memory, ethernet, etc.

And in regards to that tidbit on 32 vs 64-bit overhead, I  find 32-bit unacceptable for a modern computer. While many Amigans may be happy with that: I will not be, because I primarily use BSD. I have no use for AmigaOS as my daily driver. It is a distraction and game machine, nothing more.

There is a place for both small 32 bit "fun" devices which can still do a lot of work and 64 bit power machines for large compiles, big number crunching and servers. Frank Wille programmed the Solid Gold and SQRXZ games for a 68k with less than 1 MB of memory but he also writes code for NetBSD which he uses on the SUN server that is used for the vbcc links above:

http://sun.hasenbraten.de/
http://sun.hasenbraten.de/~frank/projects/

Hmm, another Amiga user with connections to BSD. He does support PPC Amigas also although he doesn't like the attitude of the AmigaOS 4 developers. The AmigaOS 4 developer attitude turns me off even more than the high prices. I love the 68k but PPC would be acceptable for a high end AmigaOS if that was a good choice that was going somewhere.

I agree though that ARM may be a decent target, but I'd have to wait one more ISA generation because I'm not interested in portable computers much, I want workstations and servers. I own a Nexus 7 and a LG Optimus F6 but I'd never want that to be running AmigaOS or AROS as they're already buggy enough. Much of the time they just function as portable media players and remote terminals

The AmigaOS 3.1 on is not so buggy although some parts are kludgy. The AmigaOS never got much server software and even the networking support is lacking. It doesn't have the security, memory protection or resource tracking that would be highly desirable for a server. Maybe AROS could be improved to be an acceptable server but BSD will likely always be a better choice for what you are interested in.

[TeamBlackFox]

BSD and IRIX for servers definitely! I'm just saying I like workstations and servers a lot more than portables. I know AmigaOS isn't a server suitable OS. The attitude of the OS 4 devs isn't as bad as some GNU/Linux people. Now you see why I dislike GNU/Linux, in the Arch forums they censor alternative init systems and prevent people from denouncing systemd. That and the Gentoo craze are two reasons I have no trust from GNU and co.

Speaking of ARM though, AMD did announce an Opteron that is ARM based, if it is fast enough I'd support a move to ARM by either NG solution. That would be more my style as AMD already have NUMA clone bus systems and a powerful RISC on that would be badass!

[biggun]

CISC just doesn't make sense anymore and there is support for this in the simplicity of RISC. Look at anything else engineering related and you'll see that generally the simpler solution is better.

You repeat marketing slogans of the 90th.
These slogans are technically outdated today.

Some facts:
* CISC has stronger instructions than RISC

In the late 80th and early 90th making a simpler RISC decoders allowed
to invest this into other parts of the chip to improve performance.
As this time RISC chips had a performance advantage.

But technology moved on...
The decode is NOT the bottleneck anymore.
CISC chips can today decode the same amount of instructions as RISC chips.

Today the most complex part and biggest parts of the CPU are the Caches.

Today the main challanges when designing a CPU are:


1) Resolving Instructions dependancies.
CISC has here some advantage - as a CISC CPU an do more with less.

2) Cache access is the other big bottleneck.
CISC has here two big advantages.
* CISC instructions are much more compact. Therefore they can do more with the same ICache Bandwidth.
* The number of possible parallel DCache read and DCache writes
is today the real limiting factor today in CPU design.

As RISC chips lack by design real immediate handling - all RISC chips needs to do more DCache reads to do the same amount of work as CISC chips.


Think about the above technical facts and rethink your opinion about the RISC vs CISC discussion,

[psxphill]

BeOS failed permanently when it switched from PowerPC, and dropped it for x86

Sounds an awful lot like history repeats itself.

BeOS switched to x86 because it had failed, it didn't fail because it switched to x86. The original BeBox was a PC with a PowerPC instead of an x86, there was no real advantage to them fitting a PowerPC except you had to buy their computer.

History could therefore easily repeat itself. Avoiding x86 will give you a long drawn out death, switching to x86 may or may not pay off but it will play out much quicker.

The only sane decision is between ARM or x86, which is why they own so much of the market between them. The ARM is considerably slower than x86 though, I'd prefer an Atom because it competes pretty well against the ARM but you can also go to the top end with Core I*.

Sorry, I phrased that poorly:

To read/write 64 bits of data from or to the main memory:

AMD Phenom: 33-40 cycles on average
MIPS R16000A: 11-15 cycles on average.

I apologise for the confusion.

That is pretty vague, the R16000A is 200mhz while the Phenom starts at 1.8ghz.
I'd take 40 cycles at 1.8ghz over 15 cycles at 200mhz.

The cycles increase because ram speed lags behind cpu speed, but that is what large caches are for.

If you underclock that Phenom then the average number of cycles would decrease, but it wouldn't get faster.
 
 Your metric is irrelevant.

OlafS3,

Tekmote.NL sells some for €200-400, mostly due to import tariffs. If there was more demand I'm sure the cost would drop.

Where is the demand going to come from? How many people want such a slow computer? The netbooks have appalling battery life.

[TeamBlackFox]

Biggun,

You're not correct at all, because the majority of designs that are getting notice besides RISC are VLIW and EDGE. In VLIW's case it is like super-RISC and even adopts some CISCy advantages.  VLIW is register-register/load-store architecture and basically just adds instruction level parallelism.

EDGE is another way to add instruction parallism by adding one advantage CISC genuinely has: variable length instruction words.

Anyways no, RISC is going to always be simpler and more efficient for most forms of computing. Even x86 cores nowadays break instructions down into simpler ones before processing them.

Of course, biggun you can always try proving me wrong by building this super orthogonal CPU and trying to benchmark it against a processor of the same application that is RISC. I'll be waiting.