osprey atari coldfire + arm still in development

[shoggoth]

First of all, do try to get the quoting right. I do not like appearing mis-quoted like this. There is a preview feature.

Try to get this right. Again I don't like appearing mis-quoted like this.

for me it is the linux system that is interesting - not X, this is due to drivers available for myriads of devices.

Ok, I get the point of that.

GEM != square box, not necessarily t least. So it won't be GEM-compatible, then?

I stated GEM capabilities - ofcourse GEM compatible is an intent.

Which capabilities - except round buttons and stuff (because adding that to current AES:s isn't a big deal) are we talking about? I've been trying to get a good answer on this one several times now. A clear technical answer, please.

When you say GEM - what exactly are you refering to? Atari GEM? XaAES? Or the concept of GEM itself and its APIs towards the application?

Ok, so then you're not going to add, for example, a new resource format, after all?

yes, here again it seems you are being a bit of a wisecrack. surely, the .rsc format is not that easy to expand - but there needs to be support for that - in addition I will add a new format - hopefully xml format that can be precompiled before runtime.

No, I'm not a wisecrack. I'm trying to get clear answers from you. In this case, if I do use the additional resource format used by your GUI, my application won't run on a system which lacks this functionality. Can we agree on that part?

In this context, I think it's important to think twice about adding such functionality, since we have a current user base to take care of - unless all current users migrate to your system. An alternative to adding this functionality to the OS would be to make a library instead. This library can then be used on other systems to provide the necessary compatibility. That would be a much better solution IMHO.

MiNT is not a distribution, it's a kernel. Which distribution are you referring to?

So there is no mint distribution? Excuse me for being a retard then Well if you could try to have some understanding - usually one installs MiNT or freeMint and it comes with loads of applets for a shell... unless you download just MiNT - but anyway....

What are you talking about? No - FreeMiNT and MiNT are *kernels*. Not distributions.

I'm not calling you a retard. I never did. You did. I do however think you give the impression of someone who isn't very familiar with a modern OS setup on the Atari, yet engaging in technical discussions about the matter.

There are a couple of distros, however. One is called SpareMiNT, and the most common one is probably EasyMiNT (which is based on sparemint). There are also CrippleMiNT, and KGMD (which is outdated). There even used to be an experimental Debian distro based on MiNT (naturally, not official however).

Well, all you are trying to is definately not to get a grip on technical details, several times you state I am saying things Im not..

I can assure you that's not my intention. All I'm trying to do is to sum up the (no offence) short novel you've produced so far and sort out the technical details.

[shoggoth]

Do try to get the quoting right. I don't appreciate getting mis-quoted. Noone does.

Now you are saying I have stated I need no VDI? Im saying I will not use any present VDI sources from the atari scene as they are not efficient enough - here I will create, not port. Just as there is no real reason to use XaAES either, unless as a fallback - well, atleast not until the future brings along more goodies....

I quote myself: "When you say "present VDI accelerators" - what exactly are you referring to?"

- In what way are the current AES:s not optimized for truecolour performance? I asked you this before, but I don't think I got a clear
- answer to that one.

Well, ofcourse here I must admit I have not used the latest versions myself - but...

Ok, then we're getting to the core of the poodle. You're judging stuff you haven't even used.

atari has always been a slow experience - and most people that do have the capabilities of coding speedy routines - they do make demos and avoid system stuff at all cost. Im sure I could outdo present routs, etc.

That's common on machines from that era. It's a result of limited processing power and an urge to squeeze every drop out of the system, not limitations in the AES.

Put XaAES on a fair spec machine, and it's fast. If there is a bottleneck in this context, it's the use of a slow VDI (i.e. not using a VDI replacement), and slow ST-RAM (in the case of the Falcon). My 60Mhz Hades has hardware accellerated graphics (Mach64), and it flies. On my 100Mhz CT60, window redraws are significantly slower due to the ST-RAM bottleneck.

AES gets complex with many windows / interfaces to process - also all parts are not necessary to process if they overlap...

The parts that overlap will not be processed. This is the core of the window handling in the AES. It's how the AES works. You should know that. The AES tells the application which parts that shall be updated, and the application is only permitted to update those parts.

Anyway this is not a bottleneck in the system. If you think it is I beg you to think again, because it's not.

I am very fond of mac os x and they way they handle things, and it is an intent to eventually have vector based gfx - not bitmap - which also means the AES needs to be restructured and still maintain compatability ofcourse. If you look at the sources of AES by digital research that now are public - u may get an idea.

I'm fond of OSX too, but it's a completely different architecture. It doesn't work like GEM at all, since - AFAIK - the application always has access to all parts of it's window. Put that approach in a GEM environment and you'll end up with a very inefficient system. GEM applications just doesn't work that way.
Maybe I'm wrong, but that's my initial impression.

Yes, Tos apps that dont run well in FreeMint. This is a very interesting subject, incase one could chuck out TOS.

That list would be too long to include here. If you're familiar with the system, you'll know which programs I'm referring to.

Anyway I think you misinterpreted my point regarding TOS. It was not to provide TOS-compatibility for legacy applications. It was about having a ColdFire-aware Atari OS - one that's already here.

[shoggoth]

Folks,

All I wanted to do was to get the guy to answer questions about the technical details of the project, and to get a good explanation as to why he didn't want to use the current code base (XaAES or MyAES). I think my questions were relevant, but somehow the guy simply refuse to give any clear answers at all. I ensure that my intentions are only to ensure that we don't end up with yet another standard and the incompatibilities that follow. When he bash XaAES naturally I'll want to know why (because it's really good nowadays), but he somehow misinterpreted this as love or religion from my behalf. As of yet he hasn't given a clear answer as to why XaAES suck - except that it's coded in a "retarded way" by "slow coders" who "love square buttons". I don't mind him implementing his own AES as long as it's compatible, but I do mind him bashing great community achievements without even having tried them first.

So I PM the guy, and get this (my PM included first). Including this information here is probably a major violation of netiquette, but I'm apparently not allowed to PM the guy, so I'll do it anyway.

-- PeP

hey thx for that - tho i felt there was some hammering just for the sake of it.
I feel better due to this mail - thx - it also changed my view of you.

No hammering just for the sake of it. I'm trying to get some technical details, and things are still a bit unclear. You have not yet answered my questions on the forum properly, and it seems others (judging by Mikro's post) share my view about that.

Keep in mind that some people on this forum are quite skilled on the technical details, and not answering the technical questions properly might cause your project to look bad. It's not about hammering, trust me.

I do have the sources at home for GEM, AES, VDI, GDOS etc. these have the details necessary for me, but I do not know who to contact for sources for TOS and a license - as ATARI games is sold - and I would have to ring my arse off for this - I also cant spread it.

Well, you can most probably forget about obtaining TOS the legal way. You could use the EMUTos sources as a starting point though. On the other hand, for your project I get the impression TOS wouldn't be right.

I just have to try this - as I am not going to wait for noone to do it, also I still expect ACP to get somewhere as there is a different approach - and a skilled person continuing - though, I have a strong wish for something neat and small.

I totally get that approach. I'm not the one who waits for others to do thing either. Kudos for that.

On Atari the current AES:s do not sport modern features as within macosx concerning interfaces and handling.

I've asked you multiple times now. Please examplify, please elaborate. Which modern features are missing, except round buttons and that stuff?

No, Im not impressed of XaAES for this platform its not modern enough, and MiNT devs still say it needs more work - so maybe later it could be good, and that goes for XaAES as well. As of now, they are not versatile enough - but there are plans.

Which MiNT devs are you talking about? I have yet to see you on the developers mailinglist. Or maybe you're using another name there?

XaAES has been ready for mainstream use for more than two years, and the internals extremely structured and modular, just like any modern OS component should be. Have you browsed the source code? In which way is it not versatile enough?

best regards

-- Peter

which mint devs? well I have been chatting on irc to odd skancke , draco, earx... oh and another dude trying to implement bsd aspects into it...

these are names - though u could probably try to drag these into a discussion as well...
you were probably not around the days draco did a lot of work with mint.
I sure as hell never saw you around in 1996 in the atari scene... I have just not been very active lately..

sharing views is one thing.... view = belief - get some knowledge before you get a skyrocket belief concerning software like XaAES that is totally retarded imo when it comes to modern standard.... it is coded by slow coders - that love squares... lol

bye dude - Im not talking to u here.. I just dont like the fuss - if I announce a product under development - I am entitled to do so - without u nagging me about how GOOD these softwares u love are - and that I have to use them - they suck... or maybe you could sell them to steve jobs..

it is very important to be specific I see - as if I never ran sluggish software like oAESis and the likes that didnt even function fully... also if I forget "the" infront of a word that could be attacked too..
U r a newbie dude... Im 31 - u were not even around around 96 - when I started to get to know ppl on irc...

get a life lol...


bye dude - dont pm me anymore.

[jfl]

aRt,
To stay polite I won't say what I think of your attitude but since your software development is closed source, your project is as good as dead anyway. This is 2008, mate. We share code or we're history.

[mikro]

???

This is really funny, I think Peter is not that one who should "get a life" ;-)

This is even hotter candidate for "kill it before any work done" than Phenix/Dolmen was.

Oh dear. And one could think in 2008, we're small but unite and friendly community.

mikro
(one of the slowest coders ;-)))

[GokMasE]

aRt, if you aren't interested in answering peoples questions and feedback, don't post in a public forum in the first place.
People here have done nothing but politely asking well justified questions about the project you try to promote.

And if you seriously wants us to believe Odd Skancke (who btw is responsible for the marvellous progress of XaAES in recent years) has
given you the idea that FreeMiNT/XaAES is so bad we need to rebuild the OS from scratch.. ..well, that doesn't sound anything like reality.

Finally, I could enter flame mode to tell you what I think about your attitude against especially PeP, but I'll keep it real short:

Apologize

[lp]

I have to agree, closed source = doomed. There was work being done on the MiNT kernel to allow modern drivers to be more easily ported, but I don't know when/if) Odd will get back to kernel work. Not sure what to make of this thread really, only that a lot of specific questions where asked and few answered. That's not how one sells a product. Many were questions I would of asked myself, thank goodness I discovered the thread late.

Anyway I won't drop a dime on anything that won't run MiNT natively. It's proven itself to be solid and reliable, especially on my Hades060. Given enough speed/ram it rocks. There's also no reason the VDI can't be improved, NVDI proves this theory. Same goes for the AES, it's a "component" that can be altered and improved. Mostly the community has lacked some sort of group to oversee such improvements, that's why you have variants in standards.

There's also a certain segment of the user base that likes to remain magic compatible, which in effect stifles forward development.
Those users are even less likely to want some mutant pseudo compatible Atari\debian box, as they think mint is to complex. lol

I applaud the guy for his effort if this thing really exists, however it's not a path I would take.

[shoggoth]

I applaud the guy for his effort if this thing really exists, however it's not a path I would take.

Believe it or not - me too. If it does turn into an actual product, it's a contribution to the platform, naturally.

[earx]

eventhough i sincerely doubt the feasibility of this project i do want to encourage aRt. i will probably buy one of these babies when completed. i just hope it's less of a hassle than the whole ct60 adventure.

[earx]

i also want to add that closed source is fine up until a certain point.. upsides: allows you work comfortably in your own way, 1 person can rule it with an iron fist. downsides: no automatic back-ups, no shadow engineering, no code quality control.

[shoggoth]

While on the subject of ColdFire-based clones, I would like to take the opportunity to mention Didiers work. This stuff runs on off-the-shelf parts, it sports Radeon-accellerated graphics, and it's able to run TOS, FreeMiNT and MyAES at this point.

The parts are cheap, I payed somewhere around 250euro for mine, including graphics card and PSU. I haven't been able to toy around much with it yet, since I haven't got a mouse for it (serial mouse required). Keyboard interface consists of a modified Eiffel connected via CAN.

The source code for this project is GPL, so anyone can get the parts and try it out. Naturally it's not for the newbie at this point, but I strongly believe that with a little more work and "productification" it could very well be the clone we're looking for.

[unseenmenace]

If you're referring to this:-

http://pagesperso-orange.fr/didierm/ct60/ctpci-e.htm

Then yes it does look rather promising. It would be great to see this reach small scale production as a MicroATX board or something.

[shoggoth]

Then yes it does look rather promising. It would be great to see this reach small scale production as a MicroATX board or something.

The M5484LITE is Mini-ITX actually. You don't get a backplate in the package though.

-- PeP

[aRt]

It is here - the osprey f*****g main board + a high-res and low-res display and lcd controller.

wicked ;D

check out the pictures

[aRt]

as a result of me misjudging shoggoth (sorry) a bit gettin a lil irritated, and lots of thinking... Im considering including and porting TOS to arm using a crosscompiler and patching. I have the sources of TOS lacking bios sources and xbios - trying to get the last bits by mailing - and eventually phone calls...

Yet another set of circuitry is necessary - jtag, bdm, and a usb serial interface or so - goin to check out sum nice prices.

[aRt]

Sorry guys if Ive been a bit hard concerning my opinions of MiNT and XaAES, but this baby needs something new its a 35 gram wonder machine imo.

Why not MiNT? well, I do not want to port loads of drivers - so a tint between MiNT and the current strongARM 1110 linux kernel developed by the balloon team at cambridge.

Why not XaAEs, it only sports necessary features, and is a bit off when it comes to GUI possibilities atm, like custom designed GUIs for apps ... I have visited their web-page and I cant say this is what I want for this lil baby. There is as far as I see, no masking capabilities - this comes down to squarishness and rather what points towards windows 3.11 rather than various linux distros , not to forget leopard. Ofcourse, a good GEM could really improve a lot, tho here I would never use XaAES for anything more than a fallback.

If I agreed with all of you, I would not attempt this project, Id sit and wait for a fat block on my desk as a result of ACP... Deviance is also the cause of innovation.

aRt

PS! Sorry if I offended you guys, but I became irritated due to negativity between the lines - as shoggoth said - there has been many promises and none kept, making negativity understandable, though I did not know of this. Ofcourse the atari software of today is good for an atari of today, tho - here I wish to innovate still keeping compatibility if possible.

[christos]

First of all let me disagree with you on MiNT and XaAES/MyAES. I think they look fantastic and behave great as well but that is a matter of taste or difference of opinion or whatever. I think the main issue at hand is that what you are trying to make breaks compatibility with Atari software unless you insert an emulation layer. Therefore it is more of Atari inspired than Atari derived. This is not at all bad, it's just different. Basically what it means is that if I am a software author, in order to support your platform I need to make two versions of my program one for current Atari and clones and one for yours unless I run it through emulation. And it will be much more than a matter of recompile.

[aRt]

First of all let me disagree with you on MiNT and XaAES/MyAES. I think they look fantastic and behave great as well but that is a matter of taste or difference of opinion or whatever. I think the main issue at hand is that what you are trying to make breaks compatibility with Atari software unless you insert an emulation layer. Therefore it is more of Atari inspired than Atari derived. This is not at all bad, it's just different. Basically what it means is that if I am a software author, in order to support your platform I need to make two versions of my program one for current Atari and clones and one for yours unless I run it through emulation. And it will be much more than a matter of recompile.

I do not need to break compatibility - some extra features will not hurt - as adding a masking layer forexample - this enables buttons of custom shapes, and the click works within the mask - like in osx.

You do not need to make an extra version of your software unless you address hardware directly, which is infact a no no - when it comes to programming applications though not demos.

a recompile is all that should be necessary - and if youd like a nicer interface - it _could_ be added. If the coldfire addon goes well, there will probably be a great chance of no need to recompile.

[aRt]

oh, whoever added a title of retro freak - thanks a lot - cause I freakin luv it!!!! thx mods / ops!!!!!!!

[christos]

OK, I think I understand now. You plan to add the MiNT api and the AES api to the linux kernel so that you won't need emulation. Sorry but I am not very experienced on this, just trying to understand. That is an interesting idea. Remains to be seen how well it will play out.

[aRt]

OK, I think I understand now. You plan to add the MiNT api and the AES api to the linux kernel so that you won't need emulation. Sorry but I am not very experienced on this, just trying to understand. That is an interesting idea. Remains to be seen how well it will play out.

oh no prob, its actually quite hard for me to answer clearly at times, as I have so many notes, and most of it is still in my head, I could have been clearer on the subject.

Yup, remains to see - atm. Im surfing the net for mcu´s to accelerate the lcd display - to releave the bus yielding more cpu overhead pr sec for instance.

[aRt]

aRt, if you aren't interested in answering peoples questions and feedback, don't post in a public forum in the first place.
People here have done nothing but politely asking well justified questions about the project you try to promote.

And if you seriously wants us to believe Odd Skancke (who btw is responsible for the marvellous progress of XaAES in recent years) has
given you the idea that FreeMiNT/XaAES is so bad we need to rebuild the OS from scratch.. ..well, that doesn't sound anything like reality.

Finally, I could enter flame mode to tell you what I think about your attitude against especially PeP, but I'll keep it real short:

Apologize

hey, not interested is not the word - though, as I stated to christos - there was a lot of negativity inbetween words and lines - seemingly polite- yet again sorry.

MiNT does not support the necessary for this project due to lack of drivers - and I do not want to port them, when I can follow the otosis example - and just change a linux kernel that has drivers. XaAES does not yield the necessary for a good application environment - as forexample custom apps etc. I have answered to best ability and been misjudged so I had to answer even twice.

4 Ataris of today - MiNT and XaAES are good enough for, for sure - atleats for those that dont need the extra look and feel of forexample mac os x. Compatibility will be kept - I did not say I wouldnt - still ppl went on about this and that - when I stated I would keep GEM, AES, VDI capabilities..

Sorry if I offended u - btw - there are now some pics of the board back in this thread.

[shoggoth]

Art,

Instead of throwing the monkey back and forth in this forum - how about putting a detailed specification on your web page instead? And I mean like a schematic overview of the internal software design, not only a feature list? Which part of the system does what - comparisons between your system and current solutions, flaws found in the current code base and how you plan to remedy etc.

Personally this is what spawned all the questions in my case - we've seen pictures of the hardware, and a flash-based web page (why?) only stating features, not solutions.

Just a suggestion.

-- PeP

[alexh]

The pictures of the Osprey board have an ARM chip (Intel StrongARM SA-1110) that hasnt been made in a LONG LONG time... why use an obsolete part?

[alexh]

In fact the entire thing (board, lcd and touch screen) looks like the Official Intel StrongARM development Kit circa 2000