@matthey

matthey wrote:
Quote:

There has been talk of AROS SMP for the Raspberry Pi too. SMP with Amiga compatibility is difficult though as Forbid()/Permit() and Disable()/Enable() are kludges without customized support and affordable 68k FPGA only CPUs do not have room for multiple cores.

That's what we call progress. You can't have a good breakfast without cracking open a few eggs.
Anyways...a 64bit OS would run a 32bit one in a sandbox but you know this already.

Also...

PCIe 4.0 debuts on AMD motherboards any day now...
In term of performance, with PCIe 4.0, throughput per lane is 16 GT/s. The link is full duplex, which means the data can be sent and received simultaneously à Total Bandwidth: 32GT/s. No other industry protocol can achieve the bandwidth of the PCIe 4.0 technology (Up to 64 Gbytes/s of total bandwidth for a PCIe 4.0 x16)

@jPV

Quote:

MorphOS OWB haven't solved the endian problems with newer cores, but there are many other things which can still be improved and changed for better compatibility. 3.10 introduced Deadwood's AROS fixes in the MorphOS OWB, as well as other fixes and improvements. 3.11 added many CSS, JavaScript, etc improvements, which I believe bring the current version more compatible than the AROS version. And AFAIK MorphOS team is committed to continue improving the browser with the future releases.

In a clever world there wouldn't be discussions about what specific amiganoid's OWB version is newer or has more fixes, but there would be many people using different amiganoid systems, working together to improve the same OWB codebase, and have it working equally good on all those different systems. Something that Wawa was trying to achieve if I am not wrong.

There is no glory if the MorphOS team continues improving the browser for their few hundreds of users, like there wouldn't be glory if AROS or AmigaOS teams would do the same. We all, summed up together, can be barely call ourselves a community. We should definitely stop with the kiddish arguments and start collaborating for the good of everyone.

@paolone

You know in this limited platform. For most it is a hobby OS.

People bought hardware for it. They just use it for fun and have other computers for serious stuff.

Some developers make executables for all flavours. Others don't. Many reasons for that. At least from a commercial point of view.

At least as long MorphOS and AmigaOS NG is PPC they are doomed to be a hobby OS.

Vampire also has it's problems but the future is brighter. In future we might see more powerfull fpga and maybe a GPU could solve issues with video decoding, 3D and many more. On top of that lot's of classic 68k software and games running native.

AROS is made portable so CPU does not matter much. It can run on anything. If someone would invest in AROS and make it a multimedia, gameing platform I'm quite sure it would succeed.
Not to compete with mothern gameing consoles. That would be just stupid. More of a classic gameing, multimedia platform.

Vampire could very well be the hardware to run it. At least if there was a way to bring the price down. I don't see that many buy this hardware with the price around 400£

Last edited by nikosidis on 09-Jan-2019 at 10:47 PM.
Last edited by nikosidis on 09-Jan-2019 at 10:44 PM.
Last edited by nikosidis on 09-Jan-2019 at 10:41 PM.

@paolone

Quote:

paolone wrote: @jPV In a clever world there wouldn't be discussions about what specific amiganoid's OWB version is newer or has more fixes, but there would be many people using different amiganoid systems, working together to improve the same OWB codebase, and have it working equally good on all those different systems. Something that Wawa was trying to achieve if I am not wrong.

Yes, of course. I naturally agree with that, but I was just correcting the old claim that OlafS25 brought up once again, that AROS version would be the best of the all, because it isn't the absolute truth anymore and it seemed that he hadn't noticed the progress on MorphOS side. Maybe I didn't spend enough time to phrase it polite enough, I'm sorry about it, but because he had some uncertainity in his words ("OWB on Aros X86 is the current most advanced version as far as I know"), I wanted to remind about the progress on the MorphOS side, which in my opinion has added such features that it'd be the most advanced OWB version currently. Of course if someone doesn't agree with it, it can be discussed and I can change my opinion if it'll be proven otherwise.

Quote:

There is no glory if the MorphOS team continues improving the browser for their few hundreds of users, like there wouldn't be glory if AROS or AmigaOS teams would do the same. We all, summed up together, can be barely call ourselves a community. We should definitely stop with the kiddish arguments and start collaborating for the good of everyone.

I think MorphOS team hasn't got resources and time for setting up a bigger cooperation with this so far, they're just spending the time they can from other tasks to do the things they see necessary to get things at least a little better shape. And they just started experiment with this recently, after waiting long enough if Fab would make a comeback, so there really haven't been chance for larger cooperation yet. I don't know if there would be competent developers on other camps who really would have time and interest to join this effort. AROS version went ahead on its own too at first, but now it seems to be quiet on that front too. Is the develpoer who took it forward around anymore?

I'm all in for joining efforts of all NG camps together, and I really think we should do more cooperation, because NG platforms are all in the same boat with everything. They are so similar that it's just waste if code written for one platform wouldn't be shared for others. It would just help everyone if we get anyone interested on any of the NG platforms. I've been thinking it more and more lately that it's so useless to fight with each other and we should share our programs, but unfortunately it's too rare still. It's the same goal we're trying to achieve and same dreams to get better Amiga systems, but unfortunately there are people who are a bit too thin-skinned for everything and jump against each other for the smallest reason with no tolerance, while we just should get word spreaded and let the world know about offerings of each platform.

_________________
- The wiki based MorphOS Library - Your starting point for MorphOS
- Software made by jPV^RNO

@wawa

Quote:

wawa wrote: lets wait a bit and see how aros owb behaves on vampire. condsidering that odyssey 1.25 already compiles for m68k it might become an option one day. if valid, remains to be seen...

It'll be a great enrichment if/when Odyssey will work on 68k/Apollo. But for the heavy tasks Vampire it just lacks raw cpu power and RAM.

_________________
My programs: via.bckrs.de
MorphOS user since V0.4 (2001)

@jPV

I think so too

All camps should work together on the common base API and a common set of applications (here I think of browser and something like Office) but I fear the egos are too big for that. Also there are different decisions regarding technology in the different platforms (Examples are Poseidon in Aros and MorphOS, something different in AmigaOS, MUI (Zune) in Aros and MorphOS, different in AmigaOS, AmiTCP compatible in all three

Let's put it other way.

If all software where up to date on all Amiga platforms who would with their mind right go with expensive, inferior PPC hardware when you could have i386 hardware that would do the tasks faster and much cheaper.

It you make one great game that is only for one platform. Advertise and make it popular, people would buy exactly that hardware if the price where right.

You can have a Pi for less than what one PS4 game costs.

Last edited by nikosidis on 10-Jan-2019 at 10:32 AM.
Last edited by nikosidis on 10-Jan-2019 at 10:32 AM.
Last edited by nikosidis on 10-Jan-2019 at 10:31 AM.

@nikosidis

I doubt aros is so unpopular because the software sucks, i have seen aros videos on youtube where people play quake 3 and play with some emulators on aros so it can't be that bad.

@jPV

Quote:

They are so similar that it's just waste if code written for one platform wouldn't be shared for others.

unfortunately it is only your personal opinion.

@Zylesea

Quote:

It'll be a great enrichment if/when Odyssey will work on 68k/Apollo. But for the heavy tasks Vampire it just lacks raw cpu power and RAM.

i think so too. and i was getteing at the project to compile aros odyssey 1.25 for both ppc and m68k being aware that it probably doesnt make much sense. but vampire v4 standalone having 256mb as far as i recollect might arrive at the edge, when odyssey can at least be started. this is so far as i recall the memory requirement on i386. and remember m68k binaries hav usually smaller footprint and possibly a bit lower demands.

unfortunattely i have not enough own abilities, debugging tools and any further support from the community to debug the resulting binaries. knowing that even skilled morphos developers have considered it impossible or too time/effort consuming to port this version of the webkit engine to big endian i have put it on backburner. our sources are still open though, naturally.

Last edited by wawa on 10-Jan-2019 at 11:06 AM.

@monstercoder

Quote:

I doubt aros is so unpopular because the software sucks, i have seen aros videos on youtube where people play quake 3 and play with some emulators on aros so it can't be that bad.

i have been withessing two main arguments against aros:
1. it is not amiga(os) (tm)
- one cant do anything about this..
2. it doesnt run genuine m68k binaries.
- well it does. at least the amiga-m68k target. and perhaps big endian arm (arspi) target michal is working on will do so some day.

other than that, im not sure how big the community is, but following the forums the user base might actually be similar on all camps. there aint enough drama though around aros and thereofore it gets less attention.

@monstercoder

I did not say that at all. I doubt there is any that have used or use AROS as much as I do. The OS is very stable. That said it depends on the hardware and drivers. As you probably know the DELL D520 is fantastic hardware running AROS. I can play games native like you say "quake III" or Amiga 68k games and other emulated games for hours without any problems.

What I said is that there are software, games on other alternatives that is not on AROS and also the other way. There are also software that is on all platforms, but some have more recent versions or better developed.

With AROS most recent version of Odyssey with up to date SSL, I doubt there is versions on other platforms that are more compatible, faster or more stable. That said, streaming audio, video does not work well.
My biggest hope for 2019 is to get that fixed as it will make AROS much more usefull.

On MorphOS version I heard that streaming video, audio with Odyssey work great and they have overlay, something AROS don't have.
This is what I'm talking about regarding choosing platform.

Last edited by nikosidis on 10-Jan-2019 at 11:25 AM.

for my point of wiew aros on vampire will be good and faster.
why i think this ? ... just because this is aros running on qemu on a powermac g5, where qemu x86 on G5 OsX host have a performance like the vampire v2 68080 in computing.

gave a look.

https://youtu.be/sqZJ0WL-cLI

Last edited by tlosm on 10-Jan-2019 at 11:56 AM.

_________________
I love Amiga and new hope by AmigaNG
A 500 + ; CDTV; CD32;
PowerMac G5 Quad 8GB,SSD,SSHD,7800gtx,Radeon R5 230 2GB;
MacBook Pro Retina I7 2.3ghz;
#nomorea-eoninmyhome

@wawa

Quote:

wawa wrote: @monstercoder Quote: I doubt aros is so unpopular because the software sucks, i have seen aros videos on youtube where people play quake 3 and play with some emulators on aros so it can't be that bad. i have been withessing two main arguments against aros: 1. it is not amiga(os) (tm) - one cant do anything about this.. 2. it doesnt run genuine m68k binaries. - well it does. at least the amiga-m68k target. and perhaps big endian arm (arspi) target michal is working on will do so some day.

I would add that for an outsider like me, it looks to lack management, coherent design and vision. Both MorphOS and OS4 look way more polished and complete packages that work out of the box and have consistent feeling both visually and functionally. But on AROS you get somehow unfinished and incoherent feeling of everything.

It feels that AROS is developed more without control, programmers come and go, and do what they find fun to implement, but nobody is taking responsibility that basic things get designed, done, and tested throughout. I might be wrong, but this is how it appears after using other alternatives, and I don't think I'm alone with this kind of impression.

You'll have to have a complete package nowadays to get casual users to come back and stay. It's not like people would be willing to finetune, hack and patch, and search everything they'd like to have, like we used to do in 90s with classic Amigas. Times have changed from that. AROS distributions seem to be one man projects with not enough resources or ability, or underlying system doesn't allow them, to make that finished products.

I've been trying to use AROS lately when I have ported some of my programs for it, and as I wanted to take an easy way, I just installed a hosted version on a Linux PC. I bet many others would like to take that route to have a peek at the OS. I'm not sure if it's because of that, but many things seem to be very unstable or just don't work correctly. I haven't been able to test Odyssey properly, because it always crashes if not in start, but after one or two pages, or even if I start and quit without going anywhere. Also OpenURL crashed if not configured, the Which command doesn't find commands from SYS:C/ even though it finds them from other locations in path, Zune isn't fully compatible with MUI4/5, and lots of these kinds of things. Some things work just fine, so I don't know what's causing these things.. hosted version, something in my setup, AROS generally?

_________________
- The wiki based MorphOS Library - Your starting point for MorphOS
- Software made by jPV^RNO

@jPV

Quote:

It feels that AROS is developed more without control,

consider this a quality as much as a handicap. aros may be more anarchistic in approach, but you already have "controlled" opportunities with morphos and os4. there is a choice for every one. and even so there a re certain naturally developed hierarchies of competence one needs to observe to contribute successfully.

Quote:

programmers come and go

same as on os4 and morphos, certainly. its hard to force anyone to work on his hobby in his spare time and take responsibilities as if he was compensated enough.

Quote:

You'll have to have a complete package nowadays to get casual users to come back and stay.

there are distributions and there are users apparently, even if im not very interested in i386 platform. other than that the effort we are pulling is to provide amiga users with similar opportunity. how long i will be motivated myself, and if the users accept it, i cant say. im just doing what i can so far. without much expectations.

Quote:

I haven't been able to test...

i cant answer that in detail here. i dont even know what platform you refer to. i386 abi v0? it would be best to discuss it in appropriate context (like on aros-exec, dev-ml or on our slack channel), because this is not the subject here. all i can say right away, i have been only testing my own compile of odyssey for i386 abi v1 and it was rather stable. there are known problems with video playback. but exactly in this area i have not been able to gain assistance form morphos team members.

Last edited by wawa on 10-Jan-2019 at 01:50 PM.

@jPV

this is not true if using Aros 68k. You can replace even Zune with MUI 3.8 when using a desktop that not depends on Zune (I use Magellan)

I have just tested which command... it perfectly works ;)

In Aros Vision you can even change between Zune, MUI 3.8 and MUI 5 (or however you call it)

Last edited by OlafS25 on 10-Jan-2019 at 02:00 PM.

@OlafS25

when targeting amiga you can simply compile generic amiga binaries and they should work. if not there is a problem that needs to be taken care of. but jpv is talking of i386 i sense.

@wawa

I know ;)

I only wanted to correct him because people are always talking about Aros has this limitation or that but that is not true, at least not true for all platforms

@jPV

There are some coders that just do what they like and others listen to users and fix whatever is wrong. One of the major problems is that it is way to few that report bugs. Reporting bugs is also a matter of a detailed description, else it is hopeless.

You are right about MUI not being fully up to date with OS4 or MorphOS.

The AROS night builds look ugly. The distributions are fine but as you say MorphOS and OS4 does look more polished. Also like you say. Distributions are one man job. It is realy hard to do everything.

Every time I tried AROS hosted I had some kind of problems and it did not feel stable as running native.

I have no idea why and the lack of gfx acceleration makes me totaly uninterested in that version.

Running hosted is still great for developers. Maybe some day it can be great for users too.

It for sure solves a lot regarding hardware.

Last edited by nikosidis on 10-Jan-2019 at 02:21 PM.
Last edited by nikosidis on 10-Jan-2019 at 02:19 PM.
Last edited by nikosidis on 10-Jan-2019 at 02:18 PM.

Something AROS/68k lacks, is everything related to native screenmode settings, overscan etc.

As for web browsers... it's like trying to train an elephant to balance on a house of cards. IMO time is better spent exploring what one can use Internet for without a browser, implementing native lightweight clients for common protocols, specialized Amiga frontends to relevant services etc.

Last edited by kolla on 10-Jan-2019 at 03:18 PM.

_________________
B5D6A1D019D5D45BCC56F4782AC220D8B3E2A6CC