> you have to optimize the code for a new CPU [...]. [...] if you have a new CPU
> core, you have to optimized the code for it. [...] The problem is, I don't see
> the developer (Morph or AOS) to squeeze the maximum out of this new CPUs.

With high-level programming languages it's rater the compiler which optimizes the scalar code for the respective microarchitecture if you provide the proper compiler switch(es). And indeed, because of the variety of microarchitectures, generation of optimized code is an issue also on MorphOS and OS4 as I described some time back:

https://morph.zone/modules/newbb_plus/viewtopic.php?forum=3&topic_id=7675&start=564

> An e600 core isn't the same as an e5500 or e6500 core.

Exactly. The e6500 and e5500 cores stem from the e500v1 core (via e500v2 and e500mc), which was conceived around 2001, i.e. back when G4/e600 was still going strong. So it was a parallel development for different markets with different demands than G4/e600 and never meant as e600 successor (also indicated by the smaller number).

> Do you really think, FreeScale is that silly? To develop a new core (approx. ten years later)
> and the new one is slower than the old one?

As a general note (i.e. not meant for the e6500/e5500 vs. e600 case specifically) semiconductor companies develop/release new cores that are slower than old cores of theirs all the time, especially if the new core is intended for different markets with different demands.

Quote:

I wouldn't spend 2000 euro neither :) I wonder what final price will be.

Quote:

They (Aeon) have said the AmigaOne X5000 will have an entry price tag of ~€2,220 (for Europeans paying the 20% UK VAT).

Quote:

'was as fast as a Pentium III' - Not in many of the benchmarks, I had Tualtin core products running at 1.6 GHZ that were way more powerful than a P4 1.6 GHZ P4 (except foe memory bandwidth).

'"nearly" as fast as a G5' - but still not as fast, and the G5 scaled to 2.5-2.7 GHz while the systems went dual core and then quad core.

Quote:

So as I worried the X5000 can't even beat the antique G4, and is destroyed by the G5.

Good to know. That glimmering hope of it being a viable replacement for a PowerMac G5 has been put to rest.

Quote:

I think you may be exaggerating a little bit with those remarks, but I won't be surprised if the G5 is faster than an CPU aimed at routers and embedded devices.

What others have written regarding the 4 reasons to consider buying an X5000 seem to be the most reasonable things to consider when making such a choice. I like my Dual 2.7GHz G5 PowerMac in all ways except the amount of electricity it uses and the noise and heat it produces. Okay for cold Winter days and nights, but not something I want to run all the time, specially on hot Summer days.

I now remember one of the X5000 beta testers telling me many months ago that on certain tests, the X5000 was about twice as fast as my X1000. If that is true for many AmigaOS4.1FE and/or MorphOS3.10 applications when it is finally released, it will be something that some users will be willing to pay for.

>> thanks to its faster clock rate, the P5/e5500 comes out as winner in absolute
>> comparison in the majority of tests (except dnetc).

> the X5000 can't even beat the antique G4

As you quoted, the 2.0 GHz X5000 is faster than the 1.5 GHz Mac mini G4 most of the time in the benchmark results compared in this thread (ignoring AltiVec, that is).

Quote:

Yeah, and? The Mac Mini isn't exactly the top end of the G4 line. Clock per clock the G4 is faster. Let's have someone drop benchmarks from a 7448 @ 2.0Ghz for an apples to apples comparison.

My statement stands, obviously.

Quote:

That's where I shake my head. They are taking a router CPU and trying to make it work in a staggeringly expensive desktop computer.

The electricity and heat comparisons are non-issues to me. Electricity is dirt cheap here and the system doesn't run 24/7. Heat is a non factor too. Noise is about the only thing I'd change, but I'm not about to drop $2,500+ for it just to have a much slower computer.

This is boutique manufacturing, and I just don't get it. Anyone who wants one and gets one, more power to them. They are probably much more well off financially than me. I can't justify dropping that kind of dime on a vanity project.

Guys, the only thing I'm trying to say is, if you compare benchmarks, you need a optimized version for a new core. If you use benchmark programs, then this programs has to optimized, too. Otherwise, a comparison is useless.

All these benchmarks and apps are optimized for a G4. If the compiler isn't optimized for a specific core, then the developer has to optimize the code.

If the code is optimized, then:

7448 vs. 7447 = slightly faster; pro: 7448 bigger cache, higher FSB, out of order AltiVec
750GX vs. 7448 = on par (without AltiVec), clock by clock; pro: 750 TDP; con: 750 no AltiVec, scaleability
PA6T vs. 7448 = on par (clock by clock); pro: PA6T TDP, memory bandwidth
970 vs, 7448 = 1.2:1 (clock by clock); pro: 970 scaleability, memory bandwidth; con: 970 TDP
P5 vs. T4 = 1:1.1 (without AltiVec) (clock by clock); pro: P5 scaleability
970 vs. P5 = 1:1 (without AltiVec) (clock by clock); pro: 970 AltiVec; con: 970 TDP

BUT ... all these CPUs are alot slower than Intel best offerings. Because, NXP nor IBM is selling these CPUs for home computer (or they are ten years old). Surprise? I think not.

Is PowerPC or POWER dead? I think not. Otherwise morphOS or AmigaOS is twice as dead.

If you would like to run the cheap train, buy a x86 box for 400$ and start morphOS on an emulator and write with a permanent marker "Amiga" on the case

Everybody who is jelling about the price of the current (new) PowerPC hardware should think about twice. These machines are for a niche market and the word niche is for this market just too big.

Editing typo: P5 vs. T4 = 1:1.1 and 970 vs. P5 = 1:1

[ Editiert durch AltiVeced 26.01.2016 - 16:09 ]

>> http://http.distributed.net/pub/dcti/current-client/dnetc-morphos-ppc.lha
>> http://www.219.dic.at/data/projects/native/AmigaMARK/AmigaMARK-morphos.lha
>> http://amigadev.free.fr/powerpc/nbench_mos.lha
>> http://aminet.net/package/util/moni/stream_mos

> All these benchmarks and apps are optimized for a G4.

How do you know? Which compiler options were used for each of the executables? dnetc uses AltiVec for some algorithms but these were excluded from the comparison for obvious reasons.

> If the compiler isn't optimized for a specific core, then the developer has to
> optimize the code.

In high-level languages you don't have the low-level control regarding machine code the code generator of a compiler has.

> If the code is optimized, then:
> 7448 vs. 7447 = slightly faster [...]
> 750GX vs. 7448 = on par (without AltiVec), clock by clock [...]
> PA6T vs. 7448 = on par (clock by clock) [...]
> 970 vs, 7448 = 1.2:1 (clock by clock) [...]
> P5 vs. T4 = 1:1.1 (without AltiVec) (clock by clock) [...]
> 970 vs. P5 = 1:1 (without AltiVec) (clock by clock) [...]

These specific ratios are only true for a specific benchmark, like Dhrystone, and can't be generalized for everything. An "actual speed of the cpu" does really only exist in terms of a specific task with fixed parameters.

https://morph.zone/modules/newbb_plus/viewtopic.php?topic_id=7553&forum=9&start=1

> NXP nor IBM is selling these CPUs for home computer

They are selling them for whatever purpose you want to use them for, on condition that minimum purchase quantity is met. Of course they had certain target markets in mind during development, but this doesn't mean you can't buy these CPUs if you are not in these markets. A-Eon's AmigaOne X5000/CyrusPlus is proof of that.

> buy a x86 box for 400$ and start morphOS on an emulator

Which emulator would that be than can run current MorphOS?

Andreas,

yes you are right, everything is task specific
If the code is heavy on FP-code or bandwidth use a G5, if it is AltiVec use a T4. If you have pseudo-random access tasks use a CPU with an integrated memory controller and don't use a G5.

I have only tried to put the CPUs in relation to each other ...
How do I know that the code is G4 optimized? Because for a long time it was the only CPU in Amiga land available

And yes, of course you could use every CPU for desktop use. You could also use a T4 as a curler
But that isn't the point. The question is, what's the target market?
IBM could build a really sweet desktop CPU for sure, BUT who will buy such a CPU?
If IBM tomorrow presents a CPU twice as good as everything Intel is able to offer on the desktop, how cares? If Amigaians use this CPU, how cares? Sadly, nobody.

It doesn't matter what outside the WinTel or MacTel world is going on. Nobody cares about "die hard" PowerPC or Amiga fans. Its just irrelevant.

If 2000€ is too much money for a "Amiga" computer, then maybe every morphOS or AmigaOS user should spend 2000€ for a x86 port of the OS.

Edit: It surprises me, that everyone is surprised about the performance of the NXP offerings ... who in hell has really thought, that a T4 is as fast as the latest Intel offerings? And holy ***, the X5000 does costs money ... What would be a good price of the X5000? 1000$? 500$? What production quantity? 5000pcs? 50000pcs?

[ Editiert durch AltiVeced 26.01.2016 - 20:08 ]

Quote:

It is your choice to buy an X5000 when it is available for purchase, or to continue using cheaper and faster used Mac hardware. Most people are going to agree with you and will not buy an X5000, but I don't berate anyone who decides they want to buy one, and I also don't blame A-Eon for making new PPC computers to run AmigaOS4.1FE (and now MorphOS3.10). I am sure Trevor (and Matthew) are trying to make the best decisions they can given the information available to them, and the advice they receive from their manufacturing partners, as well as the developers at Hyperion. We and many other people may not agree with many of those decisions, or the advice that led to those decisions, but I don't get all worked up, angry, or disrespectful toward the people who are attempting to do something to help the greater Amiga community.

I dislike the amount and/or severity of some of the criticism I see on forum sites, specially when it is coming from people who don't have all the facts and history that have influenced the decisions. When I do criticize any of the decisions that A-Eon managers have made, I try very hard to remain objective, and I have the advantage of knowing the people making some of those decisions a little better than most other users who have no contact directly with Trevor and Matthew. I have had the pleasure of talking with both and feel that I have a good understanding of what they are trying to accomplish, or at the very least, I think I know for certain that their intentions are good for the community, even when some people within the community feel that some or all of the choices management of A-Eon have made over the last several years, in some of the users minds, are actually bad for the community (a view that I don't agree with).

More than anything else, I respect Trevor and Matthew for their commitment and courage to invest so much of their own money and time, trying to achieve their goals, regardless if you agree or disagree with their goals, or the choices and methods used to try to achieve them. Also, I greatly dislike the way in which some of the criticism has been presented, and have found that some times, by some people (not saying this is you), have often spread information and opinions that are blatantly false, usually with great emotion and few or zero facts.

From my perspective, I blame Hyperion for most bad choices, and I firmly believe that all AmigaOS4.x users and programmers would be much better off if AmigaOS4.x had followed the example of MorphOS2.0, by porting AmigaOS4.x to G4 & G5 Mac computers (but I probably shouldn't even mention that any more, as it is ancient history, and we can't go back to change those mistakes).

One advantage the AmigaOS4.x community has over the MorphOS community is that they have far less users who waste time bashing other systems, other users, and other developers, or company choices, and they have the appearance of a more positive and active community of users and 3rd party developers. They are more vocal in a positive way and spread positive news better and more wide spread than the MorphOS users appear to be able to do, and the only explanation most MorphOS users seem to have is that they would rather have less quantity, but better quality, which is okay, but the perspective that many (most) people outside the MorphOS community have of us, is a negative one, which in my opinion has hurt our ability to gain as many new users from the remaining 68k Amiga community as we should have, or could have, but that loss is impossible to gauge or measure, so it is usually dismissed, specially by those who may have done the most harm to the MorphOS community with their hateful, or negative comments on forums like this one.

After all, we can probably count on one hand the number of new users who come to any Amiga inspired OS from outside the former Commodore Amiga 68k community each year, so not offending those former 68k Commodore Amiga users and programmers, or NOT giving them such a negative perspective is very important to our ability to gain more new users and programmers from the community where are most likely to get new users and programmers from.

Quote:

That's where I greatly disagree. If you want to benchmark use real code, not some laboratory ware.
The question for Joe Average is not what does the cpu theoretically, but which computer replays hd video best, which one provides me the most fluent browsing experience. And there will be no subarchitecture optimizes binaries for this and that. Hence, try it with ordinary code and no specially tailored one. That would be misleading.

You are kidding, right?
To not use all features/ capacities of CPU is a wise idea?
To keep it simple ... It's like building a highway with 4 lanes and only use 1 ...

For everyday use a 68k CPU should be enough ...
Give Trevor a call and tell him to use a 750 G3. It's cheap and all you need

Sorry, it's unbelievable ...

Tell the x86 guys to not use the SIMD unit of a i7. It will be fun for sure
I wonder, what do you want in fact?

> If the code is heavy on FP-code or bandwidth use a G5, if it is
> AltiVec use a T4.

AltiVec on a 2.7 GHz PPC970 may be faster than on a 1.8 GHz e6500. And are you sure that bandwidth of PPC970 is better than e6500?

> I have only tried to put the CPUs in relation to each other ...

Already done long ago:
https://morph.zone/modules/newbb_plus/viewtopic.php?forum=3&topic_id=7001&start=743

> How do I know that the code is G4 optimized? Because for a long time
> it was the only CPU in Amiga land available ;-)

For an even longer time, G2 and G3 have been used in "Amiga land". So it may as well be optimized for one of them.

> It surprises me, that everyone is surprised about the performance
> of the NXP offerings ... who in hell has really thought, that a T4
> is as fast as the latest Intel offerings?

The comparison was with a decade-old Motorola/Freescale offering, not with any latest Intel offering. That's worlds apart.

Quote:

No, I am not kidding. I am pretty serious.
Of course an optimized code for every processor would speed things up. But it messes things up, too. Look, today we have quite some differerent cpus for MorphOS already (603e, 604, 5200B, 750CXe, 7447ies, some other G4s, 970ies). But binaries don't care about that issue. They are identical for all these processors. The only exception is Altivec. Nobody wants special binaries for a particular processor, what an ugly mess that would be... If that were the case arguments against the Tabor board would be void, too. The thing is about usability in real life: One architecture/ISA - one binary. But no sub ISA binaries.

All in all it's definitely not a G3 that is good for all, but the processor that executes generic code fastest is choice #1. In MorphOS ppc land that is ibm 970. Unfortunately the 970 bysed systems have their downsides (bulky, noisy, leaky, energy-hungry).

The other thing is to realize when a horse that gets ridden is actually dead. And the ppc horse is a dead for desktop usage since a while. Put you saddle on the back of a new horse if you want to ride on.

[ Editiert durch Zylesea 27.01.2016 - 00:12 ]

> today we have quite some differerent cpus for MorphOS already
> (603e, 604, 5200B, 750CXe, 7447ies, some other G4s, 970ies).

...and 460 with 440 core :-)

> binaries [...] are identical for all these processors. The only exception
> is Altivec. Nobody wants special binaries for a particular processor,
> what an ugly mess that would be...

We already have that with special E-UAE binary for 970, and we had it on m68k :-)

Quote:

And with four times the threads, the e6500 is likely to have an advantage anyway.

BTW - Andreas, did Freescale get around to enhancing AltiVec on the e6500?

>>> If the code is heavy on FP-code or bandwidth use a G5, if it is
>>> AltiVec use a T4.

>> AltiVec on a 2.7 GHz PPC970 may be faster than on a 1.8 GHz e6500.
>> And are you sure that bandwidth of PPC970 is better than e6500?

> with four times the threads, the e6500 is likely to have an advantage anyway.

It'd be even more with the T4240 instead of the T4160 (four times the threads of 2x PPC970MP) or T4080 (four times the threads of 1x PPC970MP or 2x PPC970FX).

> Andreas, did Freescale get around to enhancing AltiVec on the e6500?

Yes, as reported 3½ years ago:

https://morph.zone/modules/newbb_plus/viewtopic.php?forum=3&topic_id=7183&start=805

"AltiVec technology for the e6500 core is essentially the same as AltiVec technology from the e600 core, except for the following:
- Adds new instructions for computing absolute differences [...] These speed up in the inner loop of motion estimation video processing
- Adds new instructions for dealing with misaligned vectors more easily [...]
- Adds new instructions for dealing with elements of vectors [...] These allow loading/storing of arbitrary elements to arbitrary addresses
- Instructions for moving data from GPRn to vector register [...]"

Doesn't look like too much.
It is hard to tell how much of a performance increase this will provide, but then, it beats not having AltiVec at all.

> It is hard to tell how much of a performance increase this will provide

None, unless the new instructions are being used. And using the new instructions means the code won't run on CPUs with "old" AltiVec, I guess. I wonder what ramifications the fact that data stream (dst) instructions are no-oped on the new AltiVec will entail, if any.

The OS4 Kommunity seems to run on a different clock than the rest of the world. Where is the X5000? Where is everything else promised? The X5000 was announced on Jan 8, 2014. I hope this boat anchor isn't holding back further proper MorphOS development.

1523804_741436842533370_1201160379_o.jpg

> The X5000 was announced on Jan 8, 2014.

Yes, this was when the "AmigaOne X5000" name was decided upon. The Cyrus mainboard itself was announced even half a year earlier:

https://morph.zone/modules/newbb_plus/viewtopic.php?forum=3&topic_id=7001&start=780
http://www.amiga-news.de/en/news/AN-2013-07-00015-EN.html

> I hope this boat anchor isn't holding back further proper MorphOS development.

At least the delay of the X5000 won't hold back MorphOS development ;-)

Quote:

Good point, this is really snail pace.

Quote:

Maybe not if the machine actually gets released to end users. If A-Eonkit goes belly up and the X5000 never gets released, then it seems to me an awful lot of development time has been wasted that could have gone into the ISA switch.