@billt

You're absolutely right. It almost all respects it is worse than an x86 for AMIGA stuff!

By "future of AMIGA" I was thinking of a far off future!

If we're being honest, the AMIGA future will likely be the same as the past 25 years... disappointing.

At least with open source designs we as a community can work on them together.

@ferrels

Come one now. RISC-V is actually getting quite popular and may well displace ARM for low-end and low-power usage. It is unlikely to be desktop scale anytime soon but it is definitely an ISA that exists and is in quite wide, and growing, use. In particular due to the cost-savings of no licence fee.

You have been able to buy RISC-V microcontrollers in single quantities for many years now and they have mature development environments. ESP32 is one quite popular one.

So that "will never reach silicon" is nonsense. They reached silicon and mass production of several different designs many years ago and are widely available, for cheap, even for hobbyists that just want to by single digit number of units.

One of the larger use cases for RISC-V are the microcontrollers on HDDs and SDDs. This is a many-hundred-million-units-per-year market and RISC-V is on its way to completely replace ARM here. When you use several hundred milions of CPUs per year, the cost of licencing becomes a real issue. Other embedded systems and IoT are also going RISC-V in a huge way.

RISC-V, ESP32, has a much more advanced and modern networking stack than AOS4 has. It even has SMB2/3 support, NFS and properly working USB stacks. RISC-V software support for drivers and maturity of toolchains and development frameworks leaves AOS in the dust.

Last edited by V8 on 05-Jul-2022 at 08:17 AM.
Last edited by V8 on 05-Jul-2022 at 06:53 AM.

@V8

Hear hear. And IPv6 of course :)

_________________
B5D6A1D019D5D45BCC56F4782AC220D8B3E2A6CC

@V8

It is the licensing that makes it so appealing and the reason that I thought I would mention it here.

Whilst maybe not relevant today for AMIGA, I see no shame in thinking about how this innovation can benefit our hobby in the future.

After all it has been licensing that has held us back 20 years.

@ferrels

Quote:

You'd have about $23

Quote:

And your math skills are as bad as your reading comprehension. Whether it's RISC-V, some oddball PPC variant, a PPC softcore, ARM variant, etc.....there is some misinformed mental midget showing up at this forum proposing that Amigans should adopt some inferior architecture as the way forward for their beloved OS. It's no wonder that even NG Amigas such as the X1000/X5000 are roughly 20 years behind mainstream PCs in performance the way Amigans have clung onto PPC with a death-grip.

I reckon I'm accurate. Your amazing list of architectures proving your point include : RISC-V and PPC and ARM. You mention PPC twice because you couldn't even think up another one. So that's 3 architectures... I'll add 680x0 and x64 so that's 5... I'd love to hear the other 18.

As for "inferior architecture", I assume you are suggesting that RISC-V is worse that the 68000? In what measurable way?

Quote:

there is some misinformed mental midget showing up at this forum proposing that Amigans should adopt some inferior architecture

At which point have I suggested anything? I posed a question!

@xe54

It could be a nice laptop for UAE. Playing my favourite games on an ultra-niche hardware has its appeal.

@ferrels

Quote:

ferrels wrote: @Karlos Quote: It's nice to see RISC V gaining ground. What ground are you talking about? This architecture has been sitting at the same spot for the past 5 years, which is on some clown's desk on a sheet of paper. Do you realize that this architecture only exists on paper and that it will cost millions in order to produce just the engineering samples needed to design and test a full system? RISC-V is going nowhere and that's where it needs to stay. As for the announcement of a RISC-V laptop, you obviously didn't even follow the links. The laptop doesn't exist and they want cryptocurrency to secure a PRE-ORDER! Oh, and the first 100 folks who are dumb enough to send cryptocurrency get a free NFT! This has scam written all over it. So follow this link and get your free NFT while you still can! LOL! https://xcalibyte.com.cn/en/roma-preorder/

Triggered much? I didn't say anything about the laptop other than the only Amiga like OS with a hope of ever running on one *should such a thing ever materialise* (emphasis added for the hard of comprehension) would be AROS. For obvious reasons.

As for RISCV. Yes, hardware implementing the ISA exists. You can buy evaluation kits based on a number of different manufacturers.

_________________
Doing stupid things for fun...

@pavlor

I hope it runs cool!

@Thread

Aren't we all still waiting for the ACube co-designed Open Source PPC laptop? What use is a RISC V laptop over and above the PPC effort?

We'd still need a port of AmigaOS to the PPC laptop which Hyperion would struggle with even without a change of ISA IMHO.

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

@BigD

Not all of us no. I'm not waiting for it anyway. If you want an Amiga laptop today, just install MorphOS on a MacBook, AROS on a vanilla laptop (mileage variable) or UAE. Pick your poison.

_________________
Doing stupid things for fun...

@xe54

Quote:

After all it has been licensing that has held us back 20 years.

funding, and market size, I don’t believe the license of PowerPC or 680x0, if Amiga market was profitable, then I’m sure the chipset company’s want have a share of the business, but its not profitable so there no attraction for commercial companies.

Quote:

At least with open-source designs we as a community can work on them together.

Sure secrets are bad, if no one knows the instruction set, you won’t get any software, but most commercial CPU’s you can get the information about instructions and how they work, so software point of view there is no issue. And now most software is written in C/C++/Rust or higher-level language like Javascript, C#, asp .net, Ruby or something like that.

Being able to add or change instructions, is not useful if it’s not part of the standard, because making a unique CPU for a tiny market won’t be profitable, piggybacking on some larger market might a lot more useful way, but who wants a unique CPU’s whit instruction they don’t need. (Turing it into a big-endian or bi-endian CPU, I’m just not sure there is market for that.)

Last edited by NutsAboutAmiga on 05-Jul-2022 at 10:36 PM.
Last edited by NutsAboutAmiga on 05-Jul-2022 at 10:35 PM.
Last edited by NutsAboutAmiga on 05-Jul-2022 at 10:33 PM.

_________________
http://lifeofliveforit.blogspot.no/
Facebook::LiveForIt Software for AmigaOS

@NutsAboutAmiga

Quote:

funding, and market size, I don’t believe the license of PowerPC or 680x0, if Amiga market was profitable, then I’m sure the chipset company’s want have a share of the business, but its not profitable so there no attraction for commercial companies.

Licensing costs is a non-issue for amiga. The amiga market is batches of tens of machines or less. Or when new platforms are introduced it is hundreds of machines.
Licensing costs only matter if you can reach economy of scale which amiga will never do again.
If you build and sell tens of handcrafted boutique motherboards at a time it doesn't really matter if the CPU license is 25cent per unit or not. The board will be really expensive regardless.

As for RISC-V, it is very popular and very successful. But right now it is aimed for the embedded market. I don't think anyone is working on a desktop class CPU so there will not likely be anything remotely useful for building an amiga desktop anytime soon. Decades?

But that is moot anyway. No one is really working on AOS4 since many many years so there is no one around to port it to a new ISA. It has taken them how long to work on Tabor and that is just a PPC with some minor quirks? How long would it take to port to a different ISA?

@V8

When I talked about how licensing had been a problem with AMIGA, i meant in it's entirety, not specifically about CPUS.

All the copyright, patents and lawsuits have taken their toll. The IP, logos, source code, copyrights... it is all a massive mess that can only be solved in court.

If AOS was open sourced in the 90s it would not be so fragmented now.
If the hardware, software and reference designs were all open source we would be in a very different, and far more preferential position.

My personal belief is that the more free a project is for people to work on, the more likely it is to stick around and for people to work on it. RiscOS is a good example here. No in-fighting in the community - just fans who want to help.

If AROS was the "official" fork this entire conversation would be moot but it was the fact that it is considered "unofficial" that people waited for AMIGA Inc in the first place.

Makes me laugh that one of the greatest modern upgrades for AMIGAs is the PI-zero (pistorm project) which uses an ARM processor (but only in the "good" ways) yet those developers aren't arguing about endians or CPUS. They looked at price, performance and accessibility rather than market size and potential CPU partners.

I personally use a MIST as my main AMIGA and the cores on them are really solid and reliable and allow me to be fully compatible with all other kinds of software and hardware. I don't really need more than that.

Really does depend on how you see the future of the AMIGA...
do you see it just as a device for playing your old software?

If we got our faculties together and worked in a coordinated fashion rather than the ridiculous in-fighting and hostile commentary we have instead, we may actually create something special.

Some of the comments on this thread alone are quite vile and ill conceived.

You won't see me editing my words!

As for the availability of desktop level CPUs for RISC-V - I started this thread by announcing that a company has stated that they will are to begin manufacturing a laptop and that per-orders are now being taken!

Whether or not this hardware exists and gets delivered is anyone's guess.

Interesting video about risc v
https://youtu.be/irH5eKzezsE

_________________
Indigo 3D Lounge, my second home.
The Illusion of Choice | Am*ga

@pixie

That was interesting. Thanks for sharing.

_________________
All the way, with 68k

@pixie

Thanks for your positive contribution :)

Video in summary :

* Intel are doing RISC-V now, making it the most popular among manufacturers
* in 5 years they are expecting 15% coverage
* It is great for balancing performance with power consumption
* RISC-V is not going to die out anytime soon due to the open design
* Cost is the lowest out of all of them
* Make electronics without unnecessary parts and unused functions
* Microcontroller chips soon coming from Alibaba : 2.5Ghz 16 core!
* CAD files are available to help you create your own CPUs
* You can "code" your own reference designs using open source software
* They can be directly fabricated at whatever scale without specialist equipment
* Are far more common than people think (she mentions the google chip in my phone that i referenced in my earlier post - the one i was told only, "exists on paper")

RISC-V is a huge deal and will be as ubiquitous as ARM in the coming years, and will certainly power most of the new gadgets. You can laugh at me now but I'll pop back on this thread in 2028, running on my RISC-V powered AmigaOS! It might just be UAE but who really knows!?

@xe54

Interesting topic.

RISC-V's future looks promising. There's a lot of interest in the AI field, partially because they can tailor the CPU for specific tasks, and optimize the power consumption.

The open-source nature has it's good and bad sides. One potential risk, is that the CPU architecture ends up with as many variants as Linux has distros, effectively killing binary compatibility. It already sucks that we can't rely on all new PowerPC CPUs having altivec (or even a standard FPU). JIT recompilation could address this problem at the application level, but it'll likely still be a pain for low-level kernel & driver code.

If we were to switch architecture today, then I'd choose ARM. ARM is making genuine inroads into the desktop market (e.g., Apple M1/M2), and has a real chance of knocking x86/x64 off its desktop perch. It already dominates in mobile, has gigantic momentum behind it, and a huge ecosystem of software and tools. This includes GPUs.**

RISC-V would be a riskier bet at this stage. It's growing, but there are a lot of unknowns. The laptop project you linked to says it'll have a GPU/NPU, but I don't know of a single RISC-V SOC that has a GPU on-board. A lot still needs to be developed, and it's not clear which computing niche RISC-V is going to dominate. There's no guarantee that it will oust ARM or x64 in the mobile or desktop arenas at all, let alone soon.

Hans


P.S., I didn't realize it until recently, but I have a RISC-V microcontroller in my office: an ESP-32.

** We're still waiting for ARM SOC manufacturers to open up their GPU development docs. Actually, they seem pretty cagey about releasing developer documentation in general.

Last edited by Hans on 25-Jul-2022 at 06:44 AM.
Last edited by Hans on 25-Jul-2022 at 06:43 AM.

_________________
http://hdrlab.org.nz/ - Amiga OS 4 projects, programming articles and more. Home of the RadeonHD driver for Amiga OS 4.x project.
https://keasigmadelta.com/ - More of my work.

@Hans

The ISA for RISC-V is well organised into capability levels, so I don't think the binary compatibility fragmentation is going to be as bad. It's not in anybody's interest to make an incompatible part. Extra instructions and features and reliance on those will lock together certain variants and their software and there may be many good reasons for that in some use cases.

_________________
Doing stupid things for fun...

@Hans

Well, the NXP T2080 is a G5 class CPU with AltiVec, why that was not picked for X50x0 is beyond me.

Trever talked about the choice of CPU for A1222 was about cutting costs. Sadly, by picking a CPU that was so different he made it a lot more complicated, and did not help the poor reputation of bringing things to market slowly, while I’m sure other factors was at play as well.

Quote:

JIT recompilation could address this problem at the application level, but it'll likely still be a pain for low-level kernel & driver code.

Well, yes. Sure, with few developers we have, but had few more developers or if drivers where released as open source, some might be able to get things working. an interesting thing about ID Software is that sell games for a few years, and then after some time release there games as open source. This has allowed many of games to reach places where there is little or no commercial value.

Last edited by NutsAboutAmiga on 25-Jul-2022 at 11:16 PM.

_________________
http://lifeofliveforit.blogspot.no/
Facebook::LiveForIt Software for AmigaOS

PADDING

_________________
I HAVE ABS OF STEEL
--
CAN YOU SEE ME? CAN YOU HEAR ME? OK FOR WORK