Some words on senseless attacks on ppc hardware.
From developer point of view.

Software for amiga is made in c/c++.
"Classic" Amiga may run on any cpu as long as it is full 32 bit big endian.
anything may be used ppc, sparc, mipc even itanium.
no reasons to use 68k. 68k gives nothing.
68k is simply slower more expensive more outdated than ppc.


emulators are emulators.
it is stupid to made software for emulator
when simply switch to native code
made everything run many times faster.


there is nothign special in amiga chipset.
commodore never upgrade sprites to 256 colors.
commodore never upgrade playfields to 256 colors.
In 256 colors amiga chipset work as dumb framebuffer.
anything may be used in amiga instead of amiga chipset.


there is no reasons to use aros on x86 or arm.
it is still on win98 level.
no reasons to use something such outdated compared to common os on common hardware.

so no reasons to switch to 68k or x86.

People who want to use x86 should give up with senseless attacks on ppc hardware.
they are should spend their time on improving aros. to get it to at least xp level.
leave ng as it is and work on new ng 2.0 that will be no more than 20 years behind windows
on common hardware.

@ppcamiga1

Quote:

ppcamiga1 wrote: Software for amiga is made in c/c++. "Classic" Amiga may run on any cpu as long as it is full 32 bit big endian.

These are your “rules”. The rest of the Amiga community, both developers and end users, do not care for your biased definitions and rules.

Quote:

emulators are emulators. it is stupid to made software for emulator

It is no way considered objectively stupid to write software for a software CPU or “emulator”. There are many scenarios where it is quite logical and smart, and the wider Amiga user base finds itself in a place and time where writing for software CPUs or emulators is a logical step forward.

Quote:

there is nothign special in amiga chipset.

Correct. Today there is nothing special about Amiga chipsets. But, they are still relevant in terms of running software which was developed to take advantage of those chipsets.

Quote:

there is no reasons to use aros on x86 or arm.

There are many good reasons to do so. One of those reasons is to spur further development of these AROS editions. It’s only with further development that AROS will become more modern and be more like Windows XP or even Windows 7.

Quote:

so no reasons to switch to 68k or x86.

Whatever helps you sleep at night.
The fact remains that:
- Switching to 68k would benefit from the decades of 68k emulation and JIT improvements to yield very high performance of 68k soft CPUs, specifically on x86 hosts.
- Switching to x86 or ARM makes a world of sense given the price/performance and overall availability of x86 and ARM hardware.

Quote:

People who want to use x86 should give up with senseless attacks on ppc hardware. they are should spend their time on improving aros. to get it to at least xp level.

¿Por qué no los dos?

Last edited by agami on 14-Nov-2023 at 11:51 PM.
Last edited by agami on 14-Nov-2023 at 11:45 PM.

_________________
All the way, with 68k

@ppcamiga1

Quote:

Software for amiga is made in c/c++. "Classic" Amiga may run on any cpu as long as it is full 32 bit big endian. anything may be used ppc, sparc, mipc even itanium. no reasons to use 68k. 68k gives nothing. 68k is simply slower more expensive more outdated than ppc.

ARM, PPC, x86, C, C/C++ : designed by female for female (Yin)...

68k and assembler : designed by true male for true male (Yang)...


ppcamiga1 : if you are a fake male, it's YOUR problem !

Why do you want to spread the mediocrity ?

Better to learn and evolve to become a male !

Last edited by CosmosUnivers on 15-Nov-2023 at 11:13 AM.

PopCorn time !!! Lots of popcorn considering where this post will go

_________________
PowerMac G5 2,7Ghz - 2GB - Radeon 9650 - MorphOS 3.14
AmigaONE X1000, 2GB, Sapphire Radeon HD 7700
FPGA Replay + DB 68060 at 85Mhz

:)
I used to think PPC would have a chance on Amiga land if the Power architecture from IBM was adopted and they produced cheaper chips, like what they did for some game consoles some years ago. Today some lower cost versions of their Power10 would be very interesting. The details I don't know but what is true is that those don't exist, I think the cheaper Power10 systems are like 5000$ or something, so it's not viable.

I do think a new 68K ASIC with the 68080 core would be very interesting, though they still haven't made a decent compatible MMU from what I remember...

Last edited by Jose on 15-Nov-2023 at 04:04 PM.
Last edited by Jose on 15-Nov-2023 at 04:03 PM.
Last edited by Jose on 15-Nov-2023 at 04:02 PM.
Last edited by Jose on 15-Nov-2023 at 04:00 PM.

_________________

José

@ppcamiga1

"senseless attacks on PPC hardware"......... oh my.
That's a new level of stupid, even for you.
Also, give us a freaking break..... "senseless attack on ppc hardware".
That's one of the most pathetic things Ive ever read.
Did someone hurt your feelings by having a different opinion and different tastes?
Grow a pair Nancy.

In what reality is it up to you to decide what others enjoy?
What sort of brain damage has one suffered to even *think* to try to dictate to others what they should enjoy while concurrently "scalding" them for what they enjoy. It's literal lunacy.

Also, are you even aware of the words "irony" and/or "hypocrisy" ?
Seems you might want to look them up....

/makes thread telling people to stop attacking his machine of choice while attacking pretty much everyone else's machines of choice.


Did it really never dawn on you that most dismissing ppc are in part doing so because of you?
Pretty much *no-one* gets attached to a cpu ISA or byte ordering. Some see ppc as a poor choice for the Amiga and that's their prerogative, but most of the time its mentioned rather than just ignoring it to annoy you.
Most of the ridiculing is at your expense 'cos people despise you.
You MUST be aware of this yes? I fail to believe anyone could be so stupid as to be oblivious to this, but you seem to take the bait nearly every time.
It's hilarious in a child throwing a tantrum to get their own way type of way.

Last but not least though, why the heck do you care what others prefer or enjoy?
Are you really that weak and feeble that you can't just do your own thing and let others do theirs? Or is it narcissism? Or just point blank stupidity coupled with poor mental health?

There's no right or wrong way to enjoy a hobby and only a child or someone with the maturity of a child would even think to suggest otherwise.

Last edited by fishy_fis on 15-Nov-2023 at 06:16 PM.

Quote:

no reasons to use something such outdated compared to common os on common hardware

So you're not going to use PPC or anything Amiga related anymore then?

Quote:

there is no reasons to use aros on x86 or arm. it is still on win98 level

Nonsense. Show me Windows 98 using 16GB of RAM. Show me drivers for Windows 98 for geforce 6xx series cards (Win9x driver support stopped 7 generations of gpus earlier).
Show me Window98 running on a 9th gen i9 (it wont run or even install).....
Even some of the software available to AROS wont run on Win9x because of both hardware reasons and gfx api reasons.
It (AROS) offers cpus ~5x as fast per core than any PPC Amiga hardware.

Yes, its archaic compared to modern OSes, but thats the nature of any AmigaOS or derivative.

And why are you so hung up on byte ordering when you yourself, in this very thread you started said a person should just run native programs?

Anyone whose not delusional could be lead to read everything you write as a clutching of straws in each and every post while disregarding anything youve written prior...
... but no.... that couldnt be the case surely? (wink, wink, nudge, nudge)

Last edited by fishy_fis on 15-Nov-2023 at 06:31 PM.

@ppcamiga1

Quote:

ppcamiga1 wrote: Some words on senseless attacks on ppc hardware. From developer point of view. Software for amiga is made in c/c++.

Not just C/C++. Many of us code in 68k asm. I've been doing it for 35 years. A large part of the OS 3.1 ROM was coded in asm. Many games were too, and some ports also have a significant amount of asm code in them.

Quote:

"Classic" Amiga may run on any cpu as long as it is full 32 bit big endian. anything may be used ppc, sparc, mipc even itanium. no reasons to use 68k. 68k gives nothing.

Not true. Classic Amigas require a CPU that executes 68k code (either directly or through emulation). ppc, sparc, mipc, even itanium are as useless in a classic Amiga as a 68k CPU is in a PC.

Quote:

68k is simply slower more expensive more outdated than ppc.

ppc is a dead end. Modern solutions for classic Amigas are using FPGA and ARM.

Quote:

emulators are emulators. it is stupid to made software for emulator when simply switch to native code made everything run many times faster.

Wrong. Speed isn't the only criteria, or even the main one. If I just wanted speed I wouldn't bother coding on the Amiga, I would just use a PC. I develop for the Amiga largely because I enjoy creating efficient code on a low powered machine. On the PC I don't care, I just code to get the job done. My last project for the PC was written in FreeBASIC, and was plenty fast enough even on a 10 year old machine.

Classic Amigas are retro machines. I want to use them the same way I did 30 years ago, enjoying the architecture of a simpler time when a few lines of asm code could do amazing things. I want to realize some of the dreams I had for the Amiga back in the day - but didn't have the time, knowledge, or tools to make them a reality. And I want to play all those classic games and try out those classic apps I didn't have the time or money for back then.

Quote:

there is nothign special in amiga chipset. commodore never upgrade sprites to 256 colors. commodore never upgrade playfields to 256 colors. In 256 colors amiga chipset work as dumb framebuffer. anything may be used in amiga instead of amiga chipset.

Wrong. The Amiga chipset is very special, unlike anything on the PC or other computers of the time - or even today. I always wanted to get into hitting the hardware to create amazing games and demos, but didn't have sufficient reason to. Now I want to explore that part of the Amiga (and other retro computers too like the C64 and Atari - but Amiga takes precedence). This isn't about getting results, it's about the joy of solving puzzles and the satisfaction of getting the most out of the least.

You can do anything with a dumb fame buffer and enough processing power, but that's no challenge and isn't satisfying. To me it's cheating - cheating me out of the enjoyment I get from utilizing the Amiga chipset's special features.

Your problem is that you are thinking of development as a means to an end. For many of us the development itself is the end. Once I have finished a project I lose interest in it, much like an artist loses interest in a work when it's completed. And like an artist I enjoy working with a particular medium - even an ancient one that cannot achieve the results of modern technologies.

Quote:

there is no reasons to use aros on x86 or arm. it is still on win98 level.

And what's wrong with that? AROS x86 is Amiga OS on PC hardware. The whole idea of it was to preserve the OS that we knew and loved when the Amiga was no longer being produced. But most of us accepted that Windows had become a better OS. AROS was just for retro computing. Now PCs are powerful enough that we don't need AROS for that anymore - emulation does the job.

Quote:

People who want to use x86 should give up with senseless attacks on ppc hardware. they are should spend their time on improving aros. to get it to at least xp level. leave ng as it is and work on new ng 2.0 that will be no more than 20 years behind windows on common hardware.

Back in 90's I argued that ppc was the wrong way to go. If we were going to switch CPUs it should be to x86 on the PC, as it was cheap, ubiquitous and would become more powerful than other architectures. And I was right. ppc was an expensive dead end.

Quote:

ppcamiga1 wrote: Some words on senseless attacks on ppc hardware.

No attack needed, this architecture is practically dead.

Jose Quote:

I used to think PPC would have a chance on Amiga land if the Power architecture from IBM was adopted and they produced cheaper chips, like what they did for some game consoles some years ago. Today some lower cost versions of their Power10 would be very interesting. The details I don't know but what is true is that those don't exist, I think the cheaper Power10 systems are like 5000$ or something, so it's not viable.

IBM was unlikely to produce "cheaper" POWER chips. They need high end server and workstation features and POWER backward compatibility making their POWER cores huge. They also demand high margins.

https://en.wikipedia.org/wiki/Gekko_(processor)#Development Quote:

The project was announced in 1999 when IBM and Nintendo agreed to a $1 billion dollar contract (IBM's largest ever single order) for a CPU running at approximately 400 MHz.

The Nintendo console PPC CPU core was the best general purpose PPC CPU console design although it was basically a limited OoO G3 at 2.3 DMIPS/MHz with some custom SIMD instructions. That was better than any ARM cores back then but at a very high cost. The IBM Cell based PPC CPU core in the PS3 was not a general purpose design but a stripped down in-order design that was clocked up for higher performance from the SIMD units. I recently made a table from 7-zip compression and decompression results where the Cell PPC CPU core had roughly the single core performance of an in-order CPU core at 1/3 of the clock speed (3GHz Cell CPU core has 1GHz like performance). The expensive OoO IBM PPC G5 CPU core shows disappointing results with very cheap in-order RISC-V and ARM cores producing better results. Even a very expensive POWER9 CPU core is outperformed in decompression by all the in-order cores. The PPC G5 and POWER9 cores have significantly higher peak performance than the in-order cores but compilers can't exploit it even with OoO cores which should give them a huge advantage over in-order RISC cores which are notoriously difficult to extract performance from and which was the point of my exploration. The SiFive U74 in-order core was deliberately designed to reduce stalls and provide easy to extract consistent performance which is in contrast to the in-order Cell PPC CPU core which is very stall prone. I am more impressed by the small and cheap SiFive U74 core, even with the relatively weak RISC-V ISA, than the expensive IBM designs.

https://amigaworld.net/modules/newbb/viewtopic.php?topic_id=45079&forum=17#864827 Quote:

There is no mention of the branch prediction penalty but it gives 2.5 DMIPS/MHz for the U74 core. This is competitive with the also in-order 8 stage superscalar Cortex-A53 and the 7-zip results back it up. The SiFive U74 core outperforms the Cortex-A53 and competes with the later generation Cortex-A55. single core | compression/MHz | decompression/MHz SiFive_U74 0.70 0.92 Cortex-A53 0.56 0.92 Cortex-A55 0.63 1.03 IBM_Cell_PPE 0.23 0.33 IBM_PPC_G5 0.49 0.82 POWER9 1.08 0.83 The first 4 cores above are in-order cores while the last 2 are OoO PPC/Power cores since how poor the PPC G5 and Cell performed came up in another thread and some Amiga users want AmigaOS 4 for the POWER9 Blackbird. The in-order SiFive U74 destroys the in-order Cell PPE and even outperforms the OoO PPC G5 in this benchmark while decompression is better than the OoO POWER9. The SiFive U74 uses a 28nm process which is better than the older PPC cores except the POWER9 but all of these PPC cores have a huge advantage in peak processing power and transistor counts (Cell has 256 GFlops of single precision fp performance and uses 234 million transistors for example). The PPC cores need perfectly scheduled code or they stall a lot, especially the Cell. The little SiFive U74 in-order core is a little miser that uses a 68060 like pipeline to reduce stalls even though the weak RISC-V ISA rarely if ever is able to use both ALUs "incorporated into the pipeline in order to allow a zero cycle load-to-use latency where the first stage is used for the address generation and the last stage can be used to operate on the data". A CISC ISA is required to take full advantage of both ALUs like the 68060 pipeline was capable of and which provides incredible performance potential, even for an in-order core.

Jose Quote:

I do think a new 68K ASIC with the 68080 core would be very interesting, though they still haven't made a decent compatible MMU from what I remember...

A MMU is not a problem in FPGA. The problem is Gunnar optimizing for a FPGA to maximize performance in a FPGA and a MMU is counter to his goal. The 68k "MMU" changed significantly and it is likely that a modern MMU would need some changes. ThoR tried to work with Gunnar on a "minimal" MMU design and compatibility layer but Gunnar is unreasonable. The AC68080 has a MPU but no virtual addresses.

bhabbott Quote:

Not just C/C++. Many of us code in 68k asm. I've been doing it for 35 years. A large part of the OS 3.1 ROM was coded in asm. Many games were too, and some ports also have a significant amount of asm code in them.

Most of the AmigaOS was written in C with only a few modules in 68k assembler like FFS. Some assembler was used. Libraries were more likely to use some assembler than be 100% C code. Some of the C code is written for 68k features and not just big endian.

base->lh_Library.lib_OpenCnt++;

Simple code like this to bump the open count of a library is common on the 68k but is not safe on the PPC. The 68k uses a single instruction atomic RMW operation while RISC uses separate load, add and store instructions where another program can change the value before it is updated. Most Amiga code is written for a CISC big endian CPU and the 68k remains the best choice for compatibility.

Last edited by matthey on 15-Nov-2023 at 09:34 PM.

@Thread

Not enough popcorn (it was pretty obvious it was going to take this way) with the usual suspects making ultra long answsers that nearly no one will be willing to read.

_________________
PowerMac G5 2,7Ghz - 2GB - Radeon 9650 - MorphOS 3.14
AmigaONE X1000, 2GB, Sapphire Radeon HD 7700
FPGA Replay + DB 68060 at 85Mhz

"Senseless attacks"... Lol.

The only senseless thing is to keep paying a fortune for hardware that's simultaneously both underpowered and underutilized compared to commodity hardware that's much cheaper and easier to find.

PowerPC had it's day. It's time to let it rest in peace. There are 8-bit processors with more market share.

_________________
Doing stupid things for fun...

K-L Quote:

Not enough popcorn (it was pretty obvious it was going to take this way) with the usual suspects making ultra long answers that nearly no one will be willing to read.

The people who like PPC aren't the developers and don't understand the technical reasons why PPC failed. This is in contrast to 68k supporters who love the 68k because of the technical reasons and is at least partially the reason why many of us stayed on the Amiga. On most other platforms, users don't even know or care what is under the hood. This reality can be ignored but expect the same kind of success as PPC AmigaNOne.

Last edited by matthey on 15-Nov-2023 at 09:49 PM.

@matthey

I don't care at all about all this nonsense PPC vs 68k or vs anithyng else BTW.

AW has become such a joke with all this "CPU wars" with lots of technical stuff to prove who has the biggest.

But don't pay attention to me, go on, I still have som popcorn remaining.

_________________
PowerMac G5 2,7Ghz - 2GB - Radeon 9650 - MorphOS 3.14
AmigaONE X1000, 2GB, Sapphire Radeon HD 7700
FPGA Replay + DB 68060 at 85Mhz

K-L Quote:

I don't care at all about all this nonsense PPC vs 68k or vs anything else BTW. AW has become such a joke with all this "CPU wars" with lots of technical stuff to prove who has the biggest. But don't pay attention to me, go on, I still have some popcorn remaining.

The PPC fans had the "biggest" mentality where 68k Amiga fans like elegance. The 68k fans are grateful for crumbs that fall from the table as long as there isn't a 3 cycle load-to-use delay between the crumbs falling. We are easy to please as far as performance as even a simple in-order superscalar 68060 rarely reaching 100MHz is a prized possession in our community and can't be found on shelves anywhere. Fall too far outside of nostalgia and compatibility though and that is not 68k Amiga anymore.

Last edited by matthey on 15-Nov-2023 at 11:06 PM.

@ppcamiga1

At this stage it's just honest criticism.

The only way out of expensive hardware that lags massively behind the mainstream is to switch CPU architecture.

Really, this hard decision should have been made in 2001. the same month of November that the OS4 contract was signed, AMD released an AthlonXP clocked at 1.6Ghz, while the top of the range single CPU Mac G4 was only hitting 833Mhz. I didn't pay attention to the computer market outside of Amiga at that time but it should have been obvious to the people in charge, which way the wind was blowing.

No OS4 development had taken place at that point and there was hardly a legacy of PPC software to preserve compatibility with. Amithlon was perfectly positioned as the transisition product bringing unimagined levels of performance to system friendly 68k software and would have kept the userbase happy for the 2 to 3 years of development of the new OS4 for the same hardware you were running Amithlon on.

The new OS would have been free of almost all of the problems that have plagued OS4 since the beggining. It really would have payed dividends in the long run and we wouldn't find ourselves in the position we're in today.

@Rob

I don't think this has led to a problem. OS3.2 suits retro 68k fans and OS4.x PowerPC a rabid bunch of NG fans. Emulation on x86-64 including Amiga Forever and Win-UAE, AROS, MorphOS, Arm/FPGA implementation with 68k OS distros covers everyone else!

Not many people yearn for PPC native apps IMHO!

Last edited by BigD on 15-Nov-2023 at 11:45 PM.
Last edited by BigD on 15-Nov-2023 at 11:44 PM.

_________________
"Art challenges technology. Technology inspires the art."
John Lasseter, Co-Founder of Pixar Animation Studios

To me it is all about the software.

Back in the 80/90s people I know bought an Amiga because it DID something. Most members of the club had no idea of the hardware that drove it Many ran business on it because there was some really good software for the Amiga back in the day. Some as a hobby to help a group of people with DTP and similar. The Amiga scene was about an end goal or solution until the dark days. Sure some in the group bought the Amiga because it had the hardware it did, but that was a minority. Regardless they all used it as their daily driver to get a goal done.

The software in Amiga is now really old, incompatible, buggy, and more trouble than it is worth to do a simple task.

So sure, it would be nice if the ideal hardware platform was chosen at some point, but to what end? The magic coding bunnies aren't going to magically create useful and proper software after a OS lands on a new CPU. AOS4 has chased away most talent who never will come back to it and they barely can get multiprocessor working (after 3 years) on a system you can hardly buy used at this point, sure there may be a gem with an audio program or finance software but they are very isolated as the OS is pretty stale and dead (unless someone can prove me wrong).

That is why I use what I do, because it does the job easily and effectively, yes the hardware is old. But, it is cheap and the performance for most things I use it for major overkill. So I can use it as a daily driver for most things, and I get great support from the MorphOS Team and the extended set of programmers that wrote kick ass code. Most of the software has had constant updating over many years at this point and It is a total solution almost, so now it makes perfect sense to get it on something else.

Last I ran OS4 or OS3 to try and be a daily driver as much as possible they failed pretty miserably. I still love my classics and OS 3.2 but besides the OS I'm running software from the 90's and earlier and it can be painful.

@matthey

Quote:

IBM was unlikely to produce "cheaper" POWER chips. They need high end server and workstation features and POWER backward compatibility making their POWER cores huge. They also demand high margins.

It also looks the recent open sourcing of POWER so people can freely fabricate their own designs hasn't encouraged any new lower cost chips to be produced either.

Quote:

Simple code like this to bump the open count of a library is common on the 68k but is not safe on the PPC. The 68k uses a single instruction atomic RMW operation while RISC uses separate load, add and store instructions where another program can change the value before it is updated. Most Amiga code is written for a CISC big endian CPU and the 68k remains the best choice for compatibility.

The most sensitive would be Forbid/Permit. Actually more sensitive would be Disable/Enable. And these need to modify a single byte. PPC doesn't like to work with scalars that aren't words. Especially with alignment and Amiga structures are not long aligned.

However PPC does include instructions for atomic access. The lwarx and stwcx pair. In typical PPC style it must be loaded, modified, then stored back. This looks like a logical way to perform an atomic lock. Or atomic increment. However it must be accessed as a whole word. Easy enough to long align it and mask it off. But that looks somewhat sloppy and code should be able to directly modify the intended scalar. Of course the 68K had no problem doing a simple atomic add. But I do wonder, if 256 tasks in a row Forbid, then relinquished it temporary with a Wait, that would surely break it.

Amusingly enough, there is a Microsoft blog that talks about it. It's fairly recent from 2018, but the original must surely be 2 decades old going by the context, since the only time I recall Microsoft touching PPC was for the short lived WindowsNT PPC port.

https://devblogs.microsoft.com/oldnewthing/20180814-00/?p=99485

@CosmosUnivers

Quote:

CosmosUnivers wrote: @ppcamiga1 Quote: Software for amiga is made in c/c++. "Classic" Amiga may run on any cpu as long as it is full 32 bit big endian. anything may be used ppc, sparc, mipc even itanium. no reasons to use 68k. 68k gives nothing. 68k is simply slower more expensive more outdated than ppc. ARM, PPC, x86, C, C/C++ : designed by female for female (Yin)... 68k and assembler : designed by true male for true male (Yang)... ppcamiga1 : if you are a fake male, it's YOUR problem ! Why do you want to spread the mediocrity ? Better to learn and evolve to become a male !

So, you're replying to THE parrot/troll of the forum with... this?!?

Do you SERIOUSLY believe on what you're written or is it another joke?