Overview Features Instructions Performance Forum Downloads Products OrderV4 Reseller Contact

Welcome to the Apollo Forum

This forum is for people interested in the APOLLO CPU.
Please read the forum usage manual.
VISIT APOLLO IRC CHANNEL



All TopicsNewsPerformanceGamesDemosApolloVampireAROSWorkbenchATARIReleases
Information about the Apollo CPU and FPU.

This Is What I Would Like to See More In Amiga!!page  1 2 

John William

Posts 457
10 Jan 2019 01:23


EXTERNAL LINK 
I would love to see more serious and boring but educational as well and day to use essential and useful and needed applications. I would love to see more applications that I can use for school work purposes, applications like SQL, database, oracle, blah blah blah.

Like...WE WANT MORE SERIOUS professional and non-professional applications!!




Mike Kopack

Posts 261
10 Jan 2019 22:36


Not saying I disagree...
 
  BUT, couple things...
 
  1) Nothing saying you can't grab all sorts of open source stuff that exists for Linux and port it over (such as gnuplot here...) BUT, realize that many of those packages have grown QUITE a bit over the last 20 years, and have a crap ton of dependencies that also have to be ported over.
 
  2) Even with a Vampire, you're looking at a computer that has a fraction of the computational performance of a typical modern Linux desktop machine. When you can emulate an Amiga with a RasPi faster than most Amigas, and the RasPi runs all these open source projects, does it really make sense to try to do this on an Amiga?
 
  Personally, I'd rather see new software for the Amiga that leverages it's capabilities rather than turn into just another machine with a ton of direct ports of stuff I can get on a Linux box... It's why I refuse to run MacPorts or Homebrew on my Mac - why bother? There's nothing about those apps that makes them really "Mac".
 
  But hey, like I said, nothing stopping you from downloading the source, the compiler and going to town making Amiga versions of all the thousands of *nix Open source packages...


Vojin Vidanovic

Posts 1586
11 Jan 2019 10:38


While I do understand the logic, number of Amiga developers is such, I embrace every FLOSS backport - be it game or useful app. As Vamp power grows (V4+) that gap will be a bit lower, enabling better ports.


Andy Hearn

Posts 268
11 Jan 2019 11:16


I hear what you're saying and i'm with you mostly, but wouldn't it be nice to run a vamp, or vamp'd up machine as your daily driver?

I know libre office is a bit pie-in-the-sky, but we're kinda used to that sort of thing round here. Plus given that we have vampire shaped unicorns sitting in various A500/600/2000's, I think we can be forgiven for looking for the pot of gold at the end of the rainbow.

Apparently a port of Libre-office is either being looked at/worked on/ or something, and due real soon now. but initially it looks like it'll be PPC/OS4 based. from there I don't suppose it'll be too much to roll it back to 68k/OS3 with a sprinkling of stardust and Angel tears...

here's hoping anyway :D


Chris Holzapfel

Posts 50
11 Jan 2019 11:30


"I hear what you're saying and i'm with you mostly, but wouldn't it be nice to run a vamp, or vamp'd up machine as your daily driver?"

Yes, that would be in fact something I would do.
With a few key applications most people need in a daily use, good printer connectivity, internet capability and a usable browser. Assuming that everything would run stable, I probably wouldn't need a big Box PC anymore at home. Ok, maybe apart from a PC notebook for everything else.


Andy Hearn

Posts 268
11 Jan 2019 11:57


i'd done some basic mucking about trying to run up my A3k as my daily driver a while back - but there are somethings that having the horsepower of a multicore x86 is good to have hanging around.
  converting audio/video to amiga compatible formats for one thing.
 
  Coffin goes a good long way to having pretty much everything you could need on 68k.
  A solid browser is becoming essential these days - most stuff is going web based. Google docs etc, office365. Netsurf is pretty-dayum good, but I've not tried it with o365 online yet.
  it bit the dust when I tried to log in to teamviewer, but that could just be the capture codes failing


Marlon Beijer

Posts 160
11 Jan 2019 12:00


Andy Hearn wrote:
Apparently a port of Libre-office is either being looked at/worked on/ or something, and due real soon now. but initially it looks like it'll be PPC/OS4 based. from there I don't suppose it'll be too much to roll it back to 68k/OS3 with a sprinkling of stardust and Angel tears...

Don't hold your breath. The OS4 API is not compatible to OS3, and adds features not present in OS3 API as well. Backporting from OS4 can be a pain.
Besides, LibreOffice uses GTK, so that would have to be ported as well. It's no small task. And working on a port for OS3 would only be possible if they release the sources, which Amigans tend not to do even though the GPL license states so...


Mike Kopack

Posts 261
11 Jan 2019 14:40


But you all are acting like even the V4 will give us suddenly comparable computing horsepower to modern machines... It's not even CLOSE.

I'm not saying the Vampire is a bad board or that it isn't fast or doesn't give amazing capabilities to our old 68K machines, but the simple fact of the matter is that even as a 4x faster 68060, it's still single core, still 32 bit, and (if you assume that it's roughly the performance as the contemporary of the time a Pentium 2 @ 175Mhz) something like 18 generations back in performance.

Think about that for a second... Modern Linux machines running on a 64 bit 4 core/8 thread CPU (or more!) at 4 Ghz vs a 32 bit (roughly) Pentium 2 @ 175 Mhz... night and day.  Yes the Amiga's OS is MUCH less hefty than Linux or Windows, but that only buys you so much in the less overhead. You're still trying to compare a turtle to a cheetah.

IMO PowerPC is a dead end, it's pretty much going nowhere and barely used in modern Unix systems. The new Risc-V open source CPU has some potential, and it would give Amiga a new identity, but not exactly commodity easy to get your hands on hardware, and EVERYTHING would have to be rewritten - given how little uptake there was from 68K->PPC for the Amiga, I don't know how well that would go over either.

So, yeah, I like what Apollo is doing here to extend the 68K line, and hope they have thoughts on how to bring some modern CPU capabilities into next gen cores as well. But don't expect miracles. I don't see anyone that does anything beyond basic office and simple applications being able to really go Amiga 100% these days. Not sure I can even see that on an Amiga X5000 machine...

But by all means, if you want to spend the time backporting the huge Open Source library, go for it! God knows folks have done it to get Linux working on all sorts of CPUs. All I'm saying is, don't expect miracles... Even with a Vampire 4, we're still working with a pretty limited set of hardware by modern computing standards.


Mike Kopack

Posts 261
11 Jan 2019 14:48


I should also add, what I'd REALLY like to see, if we're going to go whole hog on trying to do ports of the Open Source library of apps, is start with a good package management system (just throwing APT as an example out there), so that as stuff gets ported over it's easily findable and installable through that package manager. Updates are easily found and reported, etc.

Just putting it up on Aminet is only so good and eventually Aminet gets flooded with nothing but ports.

Again, just my $0.02.


Greg Thomas

Posts 24
11 Jan 2019 14:51


We want more software, so we need more developers.We can build them or import them.

If there are to be new programmers then there would be new users. This might be possible with the Raspberry Pi port of AROS, but it wont solve any problems inside the next few years.

So we can either attract some new people, or we need to get coding ourselves. With tools like Hollywood, it can become a bit easier for some novices to perhaps work together if there were some infrastructure to organise.

Another temporary solution would be a hardware one. Something like an Android or Linux Bridgeboard. An SBC inside an Amiga would allow running stuff which we shall not be able to run for a few years.

There are some who thinks the Amiga cannot be much more than it is. There are others amongst us who do think it can take its place back amongst the leading OS's where it should be if it weren't for Commodore mismanagement.

Those who see it too hard, hurt too much seeing the Amiga not succeed, don't want to think about it and prefer to enjoy vintage driving their Amiga with white wall tyres.

So if we are to move forward then what are the various ways, and what ways make most logistical sense?


Steve Ferrell

Posts 378
11 Jan 2019 17:42


Greg Thomas wrote:

We want more software, so we need more developers.We can build them or import them.
 
  If there are to be new programmers then there would be new users. This might be possible with the Raspberry Pi port of AROS, but it wont solve any problems inside the next few years.
 
  So we can either attract some new people, or we need to get coding ourselves. With tools like Hollywood, it can become a bit easier for some novices to perhaps work together if there were some infrastructure to organise.
 
  Another temporary solution would be a hardware one. Something like an Android or Linux Bridgeboard. An SBC inside an Amiga would allow running stuff which we shall not be able to run for a few years.
 
  There are some who thinks the Amiga cannot be much more than it is. There are others amongst us who do think it can take its place back amongst the leading OS's where it should be if it weren't for Commodore mismanagement.
 
  Those who see it too hard, hurt too much seeing the Amiga not succeed, don't want to think about it and prefer to enjoy vintage driving their Amiga with white wall tyres.
 
  So if we are to move forward then what are the various ways, and what ways make most logistical sense?

Then I suggest you start coding.  The Vampire targets a small group of dedicated fans who are in love with a dead platform and the days where the Amiga could sustain itself in a marketplace have long since passed.  It takes money or the possibility of making money to draw developers to a platform. And Mike Kopack is correct.  The current Vampire is so far behind contemporary systems that it's simply a toy or a curiosity to non-fans and no serious consumer is going to adopt it as their "daily driver" in 2019.  It's a hobbyist system and that's where it will stay.  Thinking that the Amiga will rise from the ashes like a Phoenix to once again emerge into the marketplace and sustain itself is simply delusional.



Chris Sanz

Posts 25
12 Jan 2019 03:31


Marlon Beijer wrote:

Andy Hearn wrote:
Apparently a port of Libre-office is either being looked at/worked on/ or something, and due real soon now. but initially it looks like it'll be PPC/OS4 based. from there I don't suppose it'll be too much to roll it back to 68k/OS3 with a sprinkling of stardust and Angel tears...

  Don't hold your breath. The OS4 API is not compatible to OS3, and adds features not present in OS3 API as well. Backporting from OS4 can be a pain.
  Besides, LibreOffice uses GTK, so that would have to be ported as well. It's no small task. And working on a port for OS3 would only be possible if they release the sources, which Amigans tend not to do even though the GPL license states so...

Terminills now has source of Final Office and is updating it, been working on it for a while now!



Greg Thomas

Posts 24
12 Jan 2019 07:52


Steve Ferrell wrote:
Then I suggest you start coding.
Done.

Steve Ferrell wrote:
The current Vampire is so far behind contemporary systems that it's simply a toy or a curiosity to non-fans and no serious consumer is going to adopt it as their "daily driver" in 2019.

OK sure, the Vampire is perhaps getting us to the year 2000 and not 2019. Quite a lot more to go. But the Apollo guys are closing the gap. We are a long way behind, but we are catching up, from quite a distance.

Amiga at an architectural level is more advanced than PC. At its core its more efficient. By the time we catch up to the PC, the Windows desktop may be irrelevant as a dominating factor.

The 6 leading platforms are Windows, Mac, Linux, Android, IOS & BSD. Pulling up the rear are MS-DOS, Chrome, OS/2, Amiga, QNX, Symbian, Newton (iPod's) and perhaps things like WebOS (BeOS) on the LG televisions or Firefox OS. Oh and RedFlag used by a billion Chinese of course, though I've never seen it. I held a computer show way back when as the most recent Amiga show in Australia, ACE2K was the everything but Windows show. I've never stopped trying to popularise computer diversity.

There is a desire now for another OS. What is it going to be? Something brand new, or something built on an existing good system. Out of the list what would you get best results building on?
1 QNX, 2 Amiga, 3 BeOS/Haiku and Symbian maybe ahead of BeOS.
QNX is used on Blackberry 10, AMG Mercedes, and various Military Industrial projects such as submarines and the Space Shuttle. Remember when they tried to acquire the Amiga community? Before Microsoft had some quiet word to stay away from Amiga and let it die.

The newest computer platform in the market is Android and IOS both backed with biggest in the world companies. However they are based on Linux and BSD which are both FOSS projects.

Vampire with AROS / ARIX can get us closer. Perhaps in FPGA they can make 150-200 Mhz as a limit. Maybe in 3 years they get ASIC and catapult us to 1.5Ghz multi-core. Then we'd be back in the game.

We need software. Particularly to make the Amiga a developers fun-park.

Perhaps as I learn to code, I should make a blog for others to do the same so we can 'build' new developers from the people which we already have. Hmm...

What assets have we got, and how can we maximise them? If we can leverage our skills (like sharing with others) then we maybe be able to show a plan for rehabilitation of the Amiga is more than a mere delusion.

The OP had some really good points to ponder.


Gunnar von Boehn
(Apollo Team Member)
Posts 4440
12 Jan 2019 08:16


Hello Greg,

Regarding APOLLO features:

Please have a look here: CLICK HERE 
APOLLO has actually a 64bit support already.
Some of the 64bit features are even already used with legacy Amiga software.




Greg Thomas

Posts 24
12 Jan 2019 08:35


Its great that 68080 can use some of the new features with older software. In 3 years time PC's will probably be 10 cores @ 6Ghz as normal. We have some time to go.

So Gunnar, am I right in thinking that the early works which you are doing right now are actually the most important? Like you are working now on the perfect seed which can later grow into a giant Oak tree?

Adding extra cores, and higher frequencies being easier than building well tuned features which you do at the moment?

I've learnt to stop pissing you off for specifics of your secret plan for the future which you haven't got to work just yet.
So now I just ask for generalities ;-)


Gunnar von Boehn
(Apollo Team Member)
Posts 4440
12 Jan 2019 09:03


Greg Thomas wrote:

So Gunnar, am I right in thinking that the early works which you are doing right now are actually the most important? Like you are working now on the perfect seed which can later grow into a giant Oak tree?

We have done all the major works.
In fact, the APOLLO 68080 is a very good CPU core.

Lets try to collect some facts:

* The 68000 ISA is a very good architecture.
-68k ISA has a good and powerful instruction set.
-It has a wealth of flexible and very powerful EA modes.
-The instruction set very readable for humans.
-Its encoding is dense - which makes programs shorter than competitors

* 68080 adds some improvements on to this
- more registers
- 64bit operations
- very powerful AMMX instructions, greatly improving Video and Game performance.

APOLLO 68080 is a very strong CPU.

As you correctly pointed out the current 68080 could be put in an ASIC - this would allow a much higher clockrate. This is mostly a matter of dollars only.

Regarding features like 8 cores etc.
Yes APOLLO support from architecture SMP.
 


Mike Kopack

Posts 261
12 Jan 2019 16:08


Gunnar - PLEASE don't take my statements as an attack against what you and your team are doing! I think Vampire is an amazing product and I can't wait to get my V2 next week for my 500 and a V4 standalone whenever they become available.



Steve Ferrell

Posts 378
12 Jan 2019 21:18


Greg Thomas wrote:

Steve Ferrell wrote:
Then I suggest you start coding.
Done.
 
 
Steve Ferrell wrote:
The current Vampire is so far behind contemporary systems that it's simply a toy or a curiosity to non-fans and no serious consumer is going to adopt it as their "daily driver" in 2019.

  OK sure, the Vampire is perhaps getting us to the year 2000 and not 2019. Quite a lot more to go. But the Apollo guys are closing the gap. We are a long way behind, but we are catching up, from quite a distance.
 
  Amiga at an architectural level is more advanced than PC. At its core its more efficient. By the time we catch up to the PC, the Windows desktop may be irrelevant as a dominating factor.
 
  The 6 leading platforms are Windows, Mac, Linux, Android, IOS & BSD. Pulling up the rear are MS-DOS, Chrome, OS/2, Amiga, QNX, Symbian, Newton (iPod's) and perhaps things like WebOS (BeOS) on the LG televisions or Firefox OS. Oh and RedFlag used by a billion Chinese of course, though I've never seen it. I held a computer show way back when as the most recent Amiga show in Australia, ACE2K was the everything but Windows show. I've never stopped trying to popularise computer diversity.
 
  There is a desire now for another OS. What is it going to be? Something brand new, or something built on an existing good system. Out of the list what would you get best results building on?
  1 QNX, 2 Amiga, 3 BeOS/Haiku and Symbian maybe ahead of BeOS.
  QNX is used on Blackberry 10, AMG Mercedes, and various Military Industrial projects such as submarines and the Space Shuttle. Remember when they tried to acquire the Amiga community? Before Microsoft had some quiet word to stay away from Amiga and let it die.
 
  The newest computer platform in the market is Android and IOS both backed with biggest in the world companies. However they are based on Linux and BSD which are both FOSS projects.
 
  Vampire with AROS / ARIX can get us closer. Perhaps in FPGA they can make 150-200 Mhz as a limit. Maybe in 3 years they get ASIC and catapult us to 1.5Ghz multi-core. Then we'd be back in the game.
 
  We need software. Particularly to make the Amiga a developers fun-park.
 
  Perhaps as I learn to code, I should make a blog for others to do the same so we can 'build' new developers from the people which we already have. Hmm...
 
  What assets have we got, and how can we maximise them? If we can leverage our skills (like sharing with others) then we maybe be able to show a plan for rehabilitation of the Amiga is more than a mere delusion.
 
  The OP had some really good points to ponder.

If you're serious about learning to code for any system, I'd suggest starting with C or C++ first.  Many people (mostly folks who have never programmed in anything other than BASIC) will vehemently disagree with me, but speaking from experience as someone who learned to program first in BASIC and later in C/C++, BASIC only taught me to program poorly and with some bad habits as well.  C/C++ is also standardized so that once you learn it, you can port your code to almost any system.  With BASIC, there are as many varieties as there are years in a millennium and most of these variants are not compatible across systems nor standardized, so porting code from one system to another is difficult if not nearly impossible.  BASIC has too many drawbacks to be used for any serious development....again, all the folks who use AMOS/Blitz/HiSoft/PowerBASIC, etc, will lose their minds over my comments, but just ask them what major applications out there were developed in any variant of BASIC that are currently available in the market or used in the scientific communities....they'll be hard pressed to even give an answer.

The only thing that I found difficult to understand with regard to C/C++ was pointers.  But once I figured out that pointers are simply a way to dynamically allocate resources, the rest was easy.  Another thing to remember is that no serious cross-platform development occurs in BASIC.  C/C++ make up well over 90% of the code base being used today.

Here are a couple books that will help immensely.  The first is: C for Dummies. It's great for learning the fundamentals of C. EXTERNAL LINK 
The other book is a bit more advanced.  It is: Programming in C by Kochan.  EXTERNAL LINK 
And don't get scared away from C by people who say you should learn C++ first instead of C.  C++ is a super-set of C, so if you learn to program in C first, you'll  have no problems with C++.  C++ just adds more concepts on top of C such as standard templates, inheritance, and function overloading.




Chris Sanz

Posts 25
12 Jan 2019 23:31


Vojin Vidanovic wrote:

While I do understand the logic, number of Amiga developers is such, I embrace every FLOSS backport - be it game or useful app. As Vamp power grows (V4+) that gap will be a bit lower, enabling better ports.

+1 Good hardware enables good software!


Greg Thomas

Posts 24
12 Jan 2019 23:39


Steve Ferrell wrote:
I'd suggest starting with C or C++ first.

Thanks for your tips Steve,that sounds very helpful. In my case I did C and Pascal at uni over 20 years ago on the PC. And I did simulated Assembly language as well at a rudimentary level. Ironically Assembly is what appealed and made sense to me, though I only did a tiny bit. As soon as I saw COBOL I fled in terror, and I hated Java too.

But I've never done any programming on Amiga except for scripting. I wonder if there is anything out there explaining that kind of thing. if there is not, then I guess me writing it might be more useful than any code I'm able to come up with this year. So in keeping with the original posters points, do you know any Amiga centric resources? or do you suggest other areas to get up to speed, regroup and then head back to Amiga camp later?

Like many of us here, I'm very devoted to the Amiga and strongly want to see it survive the level now that its on its last life. Apollo have worked hard and got us an extra life to last another level or two. It looks like AROS will just make it right before before they run out of serious coders. I think we have 3 - 5 years to make good use of the Retro computing fad, 68080, AROS, and an anti Silicon-Valley market desire for another platform.

Bravo Apollo team!

posts 32page  1 2