Georg Steger worked a bit more in the backport of reqtools to AmigaOS.
Johan Alfredsson implemented DisplayError() and Stefan Berger improved
his layers.library. DisplayBeep() works now and Henning Kiel committed
a small utility (C:Beep) to use it from the shell. Adam Chodorowski
continued his work on the workbench.library.
Johan Alfredsson rewrote C:List (everything except the ALL option works
now). WiMP can now show some info about windows and screens, thanks
to Henning Kiel. Flavio Stanchina continues in his quest to squish
bugs in header and the build process and Stefan Berger applied some
refactoring to his layers.library. Georg Steger did the same with
the serial, parallel and UnixIO HIDDs. Stefan Berger supported this
by adding a couple of skeletons for the hardware interfaces of the
Flavio Stanchina fixed bugs in the build process and some header files.
Johan Alfredsson finished the C:Info command. Georg Steger improved
the shape window demo (you can now switch the shape on and off) and
fixed a bug in AmigaTerm.
Bugfixes in the new layers.library by Stefan Berger. Georg Steger
ported XaoS, a fast portable real-time interactive fractal zoomer.
Stefan Berger has commited a completely rewritten layers.library which
is much faster and much more simple than the one we had. It's now
based on OO techniques, much easier to extend (we're currently thinking
about child layers and layer groups which will keep their stacking
order). Also, the new layers.library allows to move windows outside
the screen (the first windows have already been lost out there ;-)
Georg Steger committed additional files for ReqTools which should
make it compile on Amiga.
Adam Chodorowski began the work on workbench.library and the
Johan Alfredsson improved the ReqTools file requester. The Volume requester
Michael Schulz committed a PS/2 Mouse.HIDD and removed the respective
code in the keyboard handler.
Adam Chodorowski implemented the Intuition-Functions OpenWorkbench()
and CloseWorkbench(). LockPubScreen() will now open the workbench
if it isn't already open.
Johan Alfredsson added a couple of function to dos.library.
Georg Steger fixed a couple of bugs in the ASL ScreenMode
requester and made the close button in the X11 Window
kill AROS. He also committed a 3D Space Invaders clone.
Michael Schulz is back and has begun to work on an abstract Mouse.HIDD for AROS
x86/Native. Georg Steger finished porting of reqtools.library and ports now all
the reqtools utilities (prefs program) and updated all AROS programs which use
reqtools (for example, MultiView and the text.datatype). He also fixed a few
bugs in diskfont.library and FixFonts, in Intuition and Gadtools. Adam
Chodorowski is working on LoadWB. Johan Alfredsson and Stefan Berger have
finished more and more CLI commands. Henning Kiel is working together
with the author of plex86 to make AROS work with plex86. AROS boots
but it crashes soon after.
Joerg Dietrich committed a ppm.datatype (for PPM pictures).
Tobias Seiler fixed a long standing bug in the keyboard initialization
of AROS Native/i386, so PS/2 keyboards should work now.
Georg Steger ported more and more programs which use Reqtools and
fixed a lot of bugs in the code. The port of Reqtools should be
Johan Alfredsson worked on more DOS commands (Info, Status, Execute)
and fixed a lot of bugs in the dos.library.
Georg Steger finished gadtools.library. Cool Work! He also fixed some
bugs in the graphics.library-to-HIDD driver. On the following day, he
fixed bugs and committed new code for the C library while Johan Alfredsson
cleaned the dos.library. Again one day later, Georg committed the first
version of a port of reqtools.library and fixed some bugs in the
Johan Alfredsson finished his nonvolatile.library and C:Which. Two days
later, he fixed some more Amiga Shell commands and internals.
Georg Steger committed rewritten routines for the DOS functions
MatchFirst(), MatchNext() and MatchEnd(). These should work in all
cases, now. This fixes all those pety bugs like List C:.
Joerg Dietrich committed dtdesc, a tool to create and examine DataType
descriptors. Georg Steger improved the building of the FTXT Datatype with
it. Later that day, Joerg backported our binary.datatype to AmigaOS 3.5.
Johan Alfredsson commited a first version of a nonvolatile.library
Henning Kiel fixed a couple of bugs in the build process and
Georg Steger improved the ASL file requester code. You can now
delete, create and rename drawers.
AROS runs on Bochs now, if a bit slow. Georg Steger, who fixed the
last bugs, added a README.Bochs how to make it work. Right now,
he's trying to make AROS work with plex86.
Cleanup of the AROS source tree by Bernardo Innocenti, primitive
hardfile support of FFS-formatted floppy disks (ADF) by Georg
Steger and better mouse support for AROS Native/x86 by Tobias Seiler
(mouse buttons work now). Georg threw in some additional code to
change the position of the mouse pointer on the screen and now
you can move the window around. Since VGA.HIDD doesn't support
the XOR drawing mode, yet, there is still some garbage when you
move the window but that should be fixed, soon. :-)
There is now a simple binary.datatype plus the skeleton for the
picture.datatype. The shell supports now color escape sequences
and more language catalogs were added for MultiView and More.
Done by Georg Steger (who else ;-)).
Georg Steger improved the console.device (TAB handling,
stopping output with right mousebutton and keypress).
Unlike on the original AmigaOS, clicking the right
mousebutton in a console window will not
block the whole display but just the output in this window.
There is now a new tool to generate shared libraries,
devices, gadgets, datatypes and HIDDs. And we have
macros to access the PC speaker (which should help
to debug AROS Native/x86). Credits go to Lars Bischoff
and Georg Steger.
The clipboard.device, written by Johan Alfredsson is now
ready (kudos for fixing the last bugs and testing it go
to Georg Steger). Also, we have now ConClip.
We also have a working datatypes.library (for a while now)
but Georg Steger has ported some datatypes (text and ascii
to be precise) from Aminet. The binary datatype will be next
and maybe we'll have a picture.datatype soon, too.
Again, more than one month has passed since the last status
update sigh. It's not that there are no news; it's me. :-)
Ok, again whole six weeks of news:
Michael Schulz had to stop working on AROS Native/i386
because he's moving. Nonetheless, the native version of AROS
boots (you can move the mouse and you will see some windows)
but since there still is no filesystem, you cannot run any
programs. We try to get the sources for SFS to fix this. The
author has been positive to open the sources but he wants to
clean them first.
Work is also being done to make AROS boot with Bochs and
VMware. It already works with VMware but VMware is quite
Also, we try to compile AROS on UAE with the new JIT by
Bernie Meyer. The JIT seems to work, the compile doesn't (yet).
This would be a great step to help to debug AROS.
Some bugs were introduced into BGUI by porting it to AROS
but these were squased. BGUI now compiles on AROS and
AmigaOS. We plan to develop our version of the Workbench
with it. Also, work has begun to move the BGUI site to
the AROS site (which means that it will be much more often
be down. ;-) Honestly, I'm working on fixing this. :-)
Also, the AROS port to NetBSD/x86 has been updated (it
compiles again) plus there is a new port to Solaris/sparc.
The new port compiles but AROS doesn't start, yet.
On a sad note, the AROS project at my company, the
Sowatec AG has not started. On the
positive side, it has not been dropped, either. The problem
is that we simply cannot find enough people to staff it. So
if you're living in Europe, interested in the Amiga and
willing to work with PCs and Unix, please get in contact
with us (the restriction with Europe is just because of the
problems we would have to get a work permission for you plus
the higher risk we would impose in you moving here).
BGUI has been completely ported to AROS, now. This means that the library
can be compiled and that all demos run. Cudos go to Stefan Berger.
Michael Schulz is working on making mice work on AROS Native/i386.
This includes work on a PCI.HIDD, an IRQ.HIDD, the VGA.HIDD and the
Serial.HIDD plus an IDE driver. Trackdisk is there, too, so the
next thing is to find a filesystem to use. We try to get SFS because
it seems one of the best filesystems for the Amiga at this time.
The SFS developers seem to be very positive to this, we'll see what
comes from it.
As for the Sowatec AG hiring
developers for AROS, all I can say at this time is that we had several
job interviews and things look quite good. The only thing that might
break it at this time (as I see it) is that we get not enough people.
We figured that we need four developers and last I looked, we had three
and one interview still open. Don't hold your breath but press your
Nils Henrik Lorentzen has begun to write a framebuffer.HIDD which
uses Linux' framebuffer for rendering. Right now, it's mostly for
debugging (it's a synchronous interface instead of X11s' asynchronous
one) but maybe, there will be a AROS Native/linux-i386, ie. a
version of AROS that will boot with the help of a Linux kernel
but without the need to install a complete Linux system. We'll
see what can be done here.
And I'm diving into the depths of Ami, the SDK for the next Amiga
by Amiga Inc.</A>. As it looks, the
SDK is a bit disappointing because it's very rough and has lots of
holes but on the positive side, Amiga's crew is very helpful and
the developer support is just great. sigh Had we just had such
a support in the old C= days. Many thanks for their effort go to
Ray A. Akey and Gary Peake and all the members of the Tao and Amiga
support staff that I'm bombing with EMails. ;-) If they can keep this
attitude, then I have no doubt that in five or ten years, people
will name Ami in the same breath as Windows, EPOC and Linux.
Something really to look forward to.
As for stats, I'm proud that we made 40k hits on our web page and
this year has also been very productive from the EMail side: Last
year, there were about 300KB mails/month in the developer list.
This year, we have 1MB mails/month! Thanks go to all those
members of the development team that have not been mentioned here
but who do important stuff, nonetheless (in no order): Henning
Kiel and Lars Bischoff for reporting bugs, Georg Steger for
cleaning up after the other developers have raged through the
code and calling me back to earth when necessary, Jaime Dias for
asking "stupid" questions that make us think what we're doing,
Manuel Lemos for trusting us with BGUI, Requin Frederic for
helping Michael Schulz, Tobias Seiler for continuing the NetBSD
port, Bernardo Innocenti for bugging people to allow us to port
their software to AROS and last but not least all those people
who I forgot. :-) Thanks to you all; it's your
work that makes this possible.
And at the end, a short note to all those people who started own
ports of the AmigaOS: Good luck to you all. I think you will find
out soon enough why AROS took soo long. The AROS team is still
willing to share their knowledge and code with everyone out
there. And we can do this because of the MPL, the only true Open
Source license. :-)
Some new software is ported to AROS: The freetype library (www.freetype.org),
BGUI and SnoopDOS. Also DOpus becomes more and more usable.
Many bugs in the Gfx.HIDD were fixed and there is now a
quite usable cybergrafx.library. Also the first version
of an IDE driver for x86/Native was committed. Also
a trackdisk.device is there, so we can start porting
a filesystem to AROS.
As for BGUI, we would like someone who will keep it alive on
Amiga (that is, make sure that our changes don't make BGUI
unusable on the Amiga). If you want to help, please contact
Also please note that my PGP-key has changed (2048bit instead
of only 1024bit for better security :-). You find my new version
here (Shift-Click on the link to download).
Lastly, my Amiga SDK has arrived. You can expect an AROS Amie/Native
but probably not soon.
This is an old announcement which I fogot to copy on the
WWW site. Sorry for that. :-(
There was no status update for more than 6 weeks (which is too much).
I'll do better in the future. :-) In the last few days, our WWW counter
silently skipped 33.333 hits.
The Sowatec AG has decided to give AROS a chance. Several people who
applied for the job to work on AROS have been contacted to arrange for
job interviews. That means for me that I will finally be able to meet
some of the people I worked with for the last five years ;-) It doesn't
mean that there will be an AROS development team in Switzerland, though.
So don't hope for too much, yet. :-)
Unfortunately, the date for the World of Alternatives is very unlucky
for me; I'll return from my holidays on Saturday and therefore, I'll
not be able to attend :-( Since the World of Amiga in Cologne seems
to be a dead event this year (Who's left to attend ? Who is willing
to attend??) I'll try to make it the London Amiga show or whatever
is held this or early next year. Sorry if you did expect me to show
On a more happy news, BGUI has been handed over to us and integrated
into AROS (you can download the contrib sources for your copy). Manuel
Lemus promised me that it compiles on Amiga but, unfortunately, it
doesn't compile on AROS, yet (the old lack-of-manpower problem). There
should be no major problems, though and porting has started. I just
can't say when it will be finished. Anyone wanting to work on BGUI
only (and not on AROS) should still apply for a CVS account to
synchronise his work with ours.
Also, Steve Crietzman of the Open Amiga Foundation seems to be pretty
far with finding out which legal obstacles could hurt AROS/Lite. His
effort is currently blocked by the effort of the Sowatec AG to found
an AROS development team (we'll have to sort our legal obstacles first).
As usual, the legal dept. drains all the energies ;-)
On the AROS side, work is still progressing, maybe even at a faster
pace than usual ;-) The AROS Native/i386 flavour is currently broken
(the HIDD system has been greatly enhanced and Michael Schulz currently
cannot take care of it) but AROS Linux/i386 works quite fine. Nils
has committed a couple of patches which greatly enhance the speed of
the graphics output (it's now much faster then on the WoA in Cologne
last year in November). Right now, bugs in the build system and exec
are hunted by various AROS developers and I hope to see lots of progress
in that area when I return from my holidays.
As for the survey which I started months ago, I hope I can summarize
them after the holidays. :-)
First of all, there are no news regarding the commercial venture
which I spoke about last time. Several ideas are currently weighted
against each other and, as usual when money comes into play, things
progress very slowly. In the mean time, AROS itself progresses at
its usual slow pace. :-)
As for job appliances, we need five people and only three persons
which we can use, applied, yet. So there are still two, high payed
Currently, we use DOpus 4, kindly made available to the Amiga community
by GP Software, to find and fix bugs in AROS.
More than two weeks since the last update. No, we're not dead :-) Work
is still progressing; for example there is now a port of DOpus 4 (see our
screenshots page). Many things
still don't work but it has been ported in just a couple of hours of
work (mostly disabling things in DOpus that AROS doesn't support, yet).
Promising, I'd say.
But the more important news is that we finally got the attention of Amiga
(mostly because of the efforts of Steve Crietzman of OAF and Gary Peake
of Amiga; thanks a lot guys). I'd like to say that's good news but that's
not completely clear, yet. The point is that Amiga would like to help us
but their lawyers object. :-) Welcome to the wonderful world of business.
Currently, we're trying to find a way thought his legal mess which a)
doesn't hurt Amie, b) gives AROS room to breathe, c) makes Amigas lawyers
happy and d) helps the community. c) seems to be the most problematic
point here but we're working on it. Some very interesting ideas have been
proposed and now we and Amiga are checking what can be done and what is
futile. At this time, just a big thank at Amiga for their openmindedness
and the time they spent on this during the last three weeks when they
were also really busy with preparing everything for the Amiga 2K show.
Just in case you wonder why nothing was announced after fleecys' comment
in the IRC conference: The lawyers covered Bill with Red Tape when they
wanted to do the announcement. So you have to take my word for it. :-)
Hopefully, you can read something in the Executive Summary, soon. No
promises, though; this is really ugly and every lawyer asked until now
strongly suggested to proceed with extreme care. Oh well...
Something on the more positive front, whatever happenes, the AROS sources
will stay available at least under MPL. Also, 4 people already applied for
working on AROS. Unfortunately, only one of them can move to Switzerland
and that especially unfortunate as my boss has come up with some plans
to get this thing going. So please, if you can imagine working on this
full time, send your curriculum, etc. to email@example.com. As a
teaser: You will earn at least Sfr 72.000/year (that's about 45.000 Euros)
because otherwise, the Swiss Foreignpolice will not allow you to move
to Switzerland (basically, the idea is that if we don't need highly
trained specialists for that job, we should be able to find someone
native for that; good for you in this case). Additional bonus: If this
project fails, we can easily find something else for you to do. So no
matter what happens, you can rely on having a job. :-)
There is now a summary of votes on this
server which I will update once per day. Please continue voting.
Since there seems to be a confusion how the polling works: My
script looks for the questions and then for a digit and counts
Unnoticed, our WWW site has passed 20'000 hits this weekend. :-)
AmiWorld On-Line has translated versions of their AROS article. There is
now a german, an english, an italian and even a japanese version! Follow
this link to the index of all articles.
We are currently doing a poll on options. Please read
Over 2000 games have been ported to AROS, yesterday: MAME now runs on AROS.
:-) Have a look yourself at our screenshots page.
There is now a mailing list for announcements only (basically the same you
can read here :-) Follow this link to subscribe.
Also, the sources for Scalos V2 have been handed over to AROS. Jörg
is lacking the time to give Scalos the attention it deserves and
therefore allowed us to include his work into our contrib area.
No official representative for this work has been elected, yet,
so this is your chance to contribute.
There are still problems to build AROS Native/i386 with EGCS (GCC 2.95.2) which
should be resolved, soon. Some more screenshots have been added. It's DOOM, this
time. Unlike other Amiga ports of DOOM, DOOM can be run several times on AROS
and you can even play multiplayer games on one machine (not very useful,
though: The keyboard is just too small for four players :-).
There is a new support page for AROS for PPC Amiga. It will be the main
contact page for people interested in AROS on PPC cards on Amiga hardware and
especially those people who are interested in supporting this by
working on it.
A longstanding bug with the dates in the ToC has been fixed (it's not
perfect but better than before).
T>he autodocs are now here.
We have begun to move contributed software (for example, examples from
Fish disks and Aminet, the small and big games) outside of AROS.
These things have now their own graph (that's why the size of AROS
became smaller last week) and the code is in AROS/contrib.
There is progress on the AROS i386/native front. A very basic VGA.HIDD
has been written plus some additional stuff (timer, trackdisk, etc).
Code for Menus has been comitted. Unlike on the Amiga, Menus are handled
by a normal task which gets all input events as long as the user presses
the right mousebutton. Therefore, menus don't block, they are very easily
enhanced and the input handler is much more simple. And if blocking menus
ever become an issue, this can added on a per-program basis by looking for
AROS va devenir l'aile de développement de OAF (Open Amiga Foundation).
OAF va maintenir les relations publiques et le contact avec Amiga,
ils seront chargés de collecter et de compiler les informations et de nous
tenir au courant sur les principaux évênements afin que nous puissions nous
concentrer sur le développement. Ils s'occuperont également de la trésorerie
ainsi que des problèmes légaux ce qui nous soulage véritablement.
Nous sommes en train de travailler sur Intuition (bugfix, meilleure
intégration, menus fonctionnels), les pilotes matériels pour AROS
natif/x86 (vga.hidd et scsi.device/hidd), commodities, datatypes et
Surtout pour le dernier, nous avons besoin de personnes connaissant
WarpUp et PowerUp. Mais aussi des personnes sachant écrire des pilotes
pour un PC (principalement comment effectuer des Exec handle PCs' IRQs
un pilote pour le IDE chipset).
Des tests de vitesse ont montré que nous avons atteint 80% de la
vitesse native du matériel (Quake atteint 60FPS en 320x200 avec
Linux/x86 et 50FPS avec AROS sous Linux/x86). Ainsi, nos problèmes
graphiques semblent résolu. Les liens et la section 'statut' ont
été mis à jour.
Il y a deux bons articles à propos d'AROS sur le net. Le premier est
ici (c'est une interview avec Aaron digulla) et l'autre est ici
(background et quelques détails sur AROS). Cet article est un peu
vieux mais il est bien écrit (et est disponible en Italien).