Improved Atari TOS versions available

[ParanoidLittleMan]

http://atari.8bitchip.info/tosimav.html

 

I will answer here on every question about this, what is available now. Even can consider some suggestions what could be added in current releases - so 1.04i, 1.62i, their 'combos' with 2.06 Desktop. There is still some space for it. Please don't come with modding/improving some other TOS versions, especially not to do it with 2.06 . I explained that there is no real need for it.

Even my TOS 1.04i is Mega STE compatible, so no need for low compatible GEMDOS part of 2.06 . This project already took too much time, almost whole year.

[snarkdluG]

In what languages is it available? Only English?

[ParanoidLittleMan]

English and German. I did some work with Spanish versions too. It is possible to make other language versions, but need to understand that it is lot of extra work. If there is some bigger interest, it may happen. I just can't spend 10-20 hours to make it for single user. Not to mention that some understanding of language in question is necessary. So, will need some help too. I can deal beside English with German and languages of my 2 parents - but there are no official TOS versions in Serbian and Hungarian (I translated complete 2.06 to Serbian around 1993) .

The hard part is to deal with different keyboard input - that's where code differs between language versions. Other is mostly just different txt in AES and Desktop RSCs .

In any case, condition is that original TOS version(s) are available for desired language - 1.04 or 1.62 and by need 2.06 .

[jefffulton]

Can I use your software Ramtos TOS switcher to test these out on my machines? I'll test and Buy an image from you if necessary.

I need to swap the UK TOS in my STE with USA versions, so if they work, I'd love to buy a set of chips. I have 1.62 in there now, but any version that will take advantage of the full STE would work.

[pixelmischief]

If you give me a list of terms, I can cover Spanish. I grew up in Miami, which means I speak Spanish. Plus, my wife is Dominican.

[ParanoidLittleMan]

Can I use your software Ramtos TOS switcher to test these out on my machines? I'll test and Buy an image from you if necessary.

 

I need to swap the UK TOS in my STE with USA versions, so if they work, I'd love to buy a set of chips. I have 1.62 in there now, but any version that will take advantage of the full STE would work.

 

That's good question actually. And answer is: no . RamTOS works only with some regular TOS versions. Not with any modded or patched ones.

However, I can easily assemble all it to work from RAM and not ROM, what I already did couple times. So yes, that would be simple way for testing - no need for sending chips, to replace ROMs in machine.

But no thanks. It is well tested, at least English versions. I had some testers for that.

 

The real question is: will some people pay for versions running in RAM ?

Yeah, now some will say : "he just thinks about money, all what wants is money" , Think whatever you want. I spent at least thousand hours in this project. How things are in forums, I really don't see that I should be generous and give it as gift to community who don't care if someone heavily insults me here or in some other forum. I did this in special situation, because I wanted to fix some flaws of TOS in first place. Then some ideas came by self, after suggestions (like here), so there is lot of special, unseen features.

Long talk. Take it as it is, or please leave the shop. I really don't have time for people not trusting that it works. Things are that I don't trust that all who will get it will not try to pirate it.

[ParanoidLittleMan]

If you give me a list of terms, I can cover Spanish. I grew up in Miami, which means I speak Spanish. Plus, my wife is Dominican.

Actually I meant that I will send some questions by need. No need for list of all Spanish txt. in TOS. Only in few cases may be some problems. And contrary to written above, I will need then some tester, who is good with Spanish or some other language. But such testing needs pretty much time, and even investing money - sending chips. Here must to say that despite I said that it can work from RAM, not everything can - for instance not all Virtual floppy functions.

All in all, only if there is some real interest, let's say at least 10 people wants Spanish v. Why I think that it will not happen in next 6 months ?

[jefffulton]

That's good question actually. And answer is: no . RamTOS works only with some regular TOS versions. Not with any modded or patched ones.

However, I can easily assemble all it to work from RAM and not ROM, what I already did couple times. So yes, that would be simple way for testing - no need for sending chips, to replace ROMs in machine.

But no thanks. It is well tested, at least English versions. I had some testers for that.

 

The real question is: will some people pay for versions running in RAM ?

Yeah, now some will say : "he just thinks about money, all what wants is money" , Think whatever you want. I spent at least thousand hours in this project. How things are in forums, I really don't see that I should be generous and give it as gift to community who don't care if someone heavily insults me here or in some other forum. I did this in special situation, because I wanted to fix some flaws of TOS in first place. Then some ideas came by self, after suggestions (like here), so there is lot of special, unseen features.

Long talk. Take it as it is, or please leave the shop. I really don't have time for people not trusting that it works. Things are that I don't trust that all who will get it will not try to pirate it.

 

 

You are 8bitchip, right? I purchased your Hard Disk Driver a few weeks ago and I'll purchase a software RAM version of upgraded TOS 1.62US.

 

You make great stuff, you do awesome HD game conversions with save states, etc

I see no reason why I can't trust your Ram TOS.

 

My question was, will it run in ram or do I need it to be burned onto rom. It looks like there is an electronic ram version that i can run or put into my auto folder and run.

If so, yes, I'll buy it play with it. It sounds cool.

 

If you only want to burn it onto Roms and Sell it, then name the price.

There isn't pricing on your page, but it looks and sounds ver nice.

Edited December 12, 2018 by jefffulton

[ParanoidLittleMan]

jefffulton: I added some new txt on that presentation page. There are some prices and some benchmark too.

http://atari.8bitchip.info/tosimav.html

As said, it can work from RAM, but that is not so good way - some limitations and less SW compatibility.

 

There will be much more about new features, but I need to make that page more attractive and organized. And that needs time, as almost everything.

 

I copy benchmark here:

Done on Mega STE, with same UltraSatan, with 2 SD cards with practically same speed.
Test is just copying SUBDIR with 80 very short files to some other SUBDIR on partition.

Regular TOS 2.06 :

Note: it needs min 32 KB for buffers for 500 MB partition.
Almost empty 500 MB partition:
At 8 MHz: 39 sec , at 16 MHz 27 sec



Regular TOS 1.04, 8 MHz for same as above: 38 sec - marginally faster.

Improved TOS 1.04 (without large sectors), DOS FAT16 partition of 1000 MB, almost empty:
No added buffers, only base 2 KB, what TOS creates after boot.

At 8 MHz: 26 sec , at 16 MHz: 17 sec
With 32 KB added buffers:
8 MHz - 25 sec, 16 MHz - 17 sec - marginal improvement, or just in range of test error.

[jefffulton]

I read your new page. Looks good.

I have a UK STE, but live in the USA. I'll send you an email to discuss paying for and sending you new TOS ROM chips to the USA.

I also have been testing a lot of your HD games via multiple machines and can start giving you some feedback on those when i complete games (so the trainers help) =)

Edited December 13, 2018 by jefffulton

[KLund1]

ParanoidLittleMan You have done some very good work as I read you pages about these improved TOS's. Very impressive!! I'd love to see what you could do for 4.04!!?!!

I still do not see a page to purchase them yet. How could we get versions? Also is there a documant or page that shows specifics on how to use any of the new features you have added?

[ParanoidLittleMan]

This is new for me too, in many things. All started with improved FAT16 of TOS, and I'm sure that it is not easy to understand for most of users.

Then added more and more features, and still have some space in limited 192 KB for TOS 1.04 .

 

What I just realized that is missing is: need to explain in which cases this is simple TOS ROM replacement - and that will be mostly in STE, and 6 ROM chips in ST, Mega ST .

If want 2 chips in ST, then will need 28-32 pin adapters, 2 of them. Because there are no 28 pin EPROMs with capacity of 128 KB (27C1001, 27C010 ... - 32 pins) . Atari used mask programmed, 128KB capacity ROMs for 2 TOS ROM chip releases, and they can fit in 28 pins, since no need for extra pins for programming. Unfortunately, I don't have ROM factory under my bed, so can only ship EPROMs

 

Yeah, there is need for so many explanations. And that's why I need to reorganize whole page to be shorter txt on it. What means that will be links for details.

 

And of course, will need some guides for usage. Virtual floppy is what will need longest guide. And something important: it uses so level disk access code, and because planned gaming support concrete TOS binary must be set for concrete hard disk adapter type. In other words: user must give his adapter type. UltraSatan is most popular in this years, and code for it is good for CosmosEx and GigaFile, + later ICD SCSI adapters. Satandisk (not recommended to buy it now) needs own driver code, IDE needs of course own driver code.

But don't think that it affects usage of regular hard disk driver SW. So, improved TOS variant with low level IDE driver for Virtual Floppy will work fine with any decent hard disk driver SW and adapter - so UltraSatan will work in usual way (with better compatibility, larger partitions) . Only that Virtual Floppy will not work with US, only with IDE adapters.

 

However, there is already written that it can be customized by buyer's wishes. So, must not have all features. I can set in short time new config of features in ASM source, and assemble it. But of course, it is to expect that most will want all it.

 

I think that main problem is that I made some really new solutions, and it is really impossible to explain all it in short.

 

Considering 4.04 - don't mention it now please. As you may see, there is still plenty to do about explanations, WEB pages for what is already programmed.

Indeed I can do it for 4.04 . But this, with 1.04 came as idea for me 10 years ago. I waited so much because knew that will take lot of time (2 months of everyday work on it).

TOS 4.04 will need much more time (and I can not run it in Steem Debugger, so really really much more time). Unless it's sources are available, and is possible to compile it with something better than Alcyon compiler (main culprit probably for not so good FAT16 solution in TOS) . Then me, or with some good C coder we can go on it. Doing it in C should be much easier. Of course, there will be a lot of it what can add as ASM lines in C sources .

[guus.assmann]

Hallo,

 

Improved TOS 1.04 looks interesting. Once I have some more time, I'll have a better look at it.

However I do have a question already: Have you condsidered relocating the OS so it will work on the same addresses as TOS2.06?

As I see it, this would have at least 2 advantages:

1) More space for extra goodies !

2) Can be used in Flash Tos without need for extra logic.

Maybe there's more positive things, but then I'd have to speculate and that's not such good idea.

 

One more question: On your site, you mention 68010. Did you test the OS with this processor?

I've not yet tested if it's really worth replacing the 68000. (Speed increase?)

And some incompattible things even with TOS2.06 are not so good maybe.

Just curiuos for your view on this topic.

 

Anyway, thanks for the efford and information.

 

Another thought just crossed my mind.

You've made many adaptions to games, drivers and this TOS adaption. And hopefully they are documented.

One inevitable thing is that we all have a limited time on earth.

What will happen to all the knowledge once you're time is up?

I know, not a very merry thought, but still one to ponder uppon.

 

BR/

Guus Assmann

 

P.S. Don't be mistaken, I wish you a very long and healthy time among us and hope to learn a lot from you. :-)

[ParanoidLittleMan]

Good questions guus.assmann .

 

I can relocate improved TOS 1.04, 1.62 wherever I want. Because it is fully disassembled, so all I need to do is to change address in first line:

org $E00000 - that would be it - normally, for 1.04 it is org $FC0000 .

 

And I mentioned that improved 1.04 works well in Mega STE - that needs some + code, to proper initialize extra HW in MSTE. And even in case of STE some smaller extra code is needed.

All it is done and tested.

And all it can work from RAM too - but not recommended.

 

I tested with 68010, and works fine. But it is just not wise to go on Mega STE on 68010, unless you have some very special needs - like virtual memory.

And I want to thank to exxos, who sent me for free one 68010 CPU, some 6 Months ago. That test should prove that it will work with 68020-68030 CPUs too, since all changes done are about larger stackframe. There is plus code for emulating move from sr - what is privileged on large stackframe CPUs. Real test would be of course to test it with some accelerator with one of those CPUs in some ST, but as I know it did not happen.

Why no 68010 in Mega STE, or some ST, STE ? Because lot of SW will fail because of larger stackframe, especially games.

 

Surely, everyone should think about what to do with his intellectual work, to save it for time when will not deal with it, or will be not able for it. Steem authors, Hayward brothers published sources some 5 years ago, and gave necessary help for usage. That needs plenty of time. I was thinking about it. I have all files, on 2 computers, + some Flash cards, so it is safe, I think. When I finish with this activity, and it may be very soon, I will do my best to make it all public. At moment there are some free and some commercial SW projects, and even some HW project, on what I work.

 

And there is already some of it published: http://atari.8bitchip.info/playfhd.html

Some asm code too - yeah, big chance for C.J. to find those stolen byes ! Sorry, but it seems that it can't go without pulling out some bad memories. I promise, it will be less and less - unless someone again starts to fill my head with crap.

[+CyranoJ]

Seek. Help. Soon.

[guus.assmann]

Sorry if this is getting a bit of topic, it's about different processors and OS usage.

 

Correct me if I'm wrong, but stackframe size is about the amount of words that the CPU "stores" when a subroutine is called.

In that case, software that's made according to the "rulebooks" should work just fine.

Unfortunately, when the "neodesktop.inf" is used, a 68010 or 68020 system will crash.

Also, I just tested a Mega ST with a 68010 and a CF-card. (PP interface + flash-tos and driver)

It works just fine, until the question is displayed to swap drives. Then it get's into a loop. (Blinking the drive active LED at about 0,5 Hz)

Placing a 68000 in the same system solves this problem.

 

What I'm trying to do is find a good compromise between speed and compatability.

Not by definition for games, but for more serious applications.

Different TOS versions and posibillities to switch between those cannot be avoided.

Switching between processors is not so simple.

I would like a CPLD-based processor that can switch between 68000, 68010 and 68020.

[ParanoidLittleMan]

Stackframe does not differ in case of simple subrutine call - that's even not called stackframe, since there is only return address (long) on stack.

Interrupts, traps have 6 byte stackframe in case of 68000 and 8 byte in case of 68010-30 . Problems appear with SW what assumes that stackframe is 6 bytes.

Like using same code for subrutine call, with added move.w sr,-(sp) - that can be exited with rte on 68000, but with 68010 it will crash, since stackframe holds vector base register too.

 

What you write about 'neodesktop.inf' is with TOS 2.06, I guess, and it's NEWDESK.INF actually - your name is too long for 8.3 format. Or then it is not pure TOS.

Please be more accurate, or give more details. What is Flash TOS actually ?

 

68010 is not really faster than 68000. 68020 can be faster on same clock . But surely will not fit in some CPLD . Only in FPGA.

 

Best compatibility would be 68000 at high clock rate, and TOS 1.04, or 1.62 with Desktop + AES from 2.06 .

All it is already supported and tested by me. Like on Mega STE at 16 MHz. Even better if can draw it at higher clock.

And don't forget that Mega STE has 16 KB cache, while 68020-30 only 512 bytes.

Edited December 29, 2018 by ParanoidLittleMan

[guus.assmann]

Ok,I think I understand Stackframe now. (I only have some 8 bit programming experience)

And I'm sorry for not being accurate. You're right about Newdesk.inf.

 

What I did, was take a Mega ST board with 4MB-ram and Tos 2.06 and Blitter enabled. (No extra software loaded, just "plain vanilla Tos 2.06" )

The CPU is in a socket, so I can swap it easy.

Test program is Gembench 28, V607 build 28. (Only very limited test, I know)

With 68000, 115% Display, 99% CPU and 107% average.

With 68010, 117% Display, 103% CPU and 109% average.

Biggest difference: 68000 gives 99% for integer devision and 68010 is 112%.

All other tests between 1 and 2% faster for 68010.

My conclusion is that mainly some math (and maybe some other instructions) are faster in the 68010. But not a big difference.

 

For the 68020 I cannot do the same. This is a card that fits in the 68000 socket.

But the card has FPU and I'm not 100% sure the clock speed is the same.

 

A good size cache will have much more speed improvement. (Depending on the programm of course)

[leech]

Sorry to go a little off topic, since you mentioned a TOS 4.04 would have to wait until you're satisfied with 1.04/1.62/2.06, but I think it'd be cool to get 4.95 polished up and sold as proper upgrades to the Falcon.

 

Also, any chance for the improvements to 3.06?

 

More on topic, I have one of Exxos' TOS Switchers, that I *think* can hold 4 TOS versions, though I only have a dual switch... any chance on selling chips with that particular PLCC size so I could just swap it out and have your enhanced 1.62/2.06?

[ParanoidLittleMan]

Sorry to go a little off topic, since you mentioned a TOS 4.04 would have to wait until you're satisfied with 1.04/1.62/2.06, but I think it'd be cool to get 4.95 polished up and sold as proper upgrades to the Falcon.

 

Also, any chance for the improvements to 3.06?

 

More on topic, I have one of Exxos' TOS Switchers, that I *think* can hold 4 TOS versions, though I only have a dual switch... any chance on selling chips with that particular PLCC size so I could just swap it out and have your enhanced 1.62/2.06?

I don't think that I said exactly that. It is not only going off topic, but it is really not nice to come out with wish lists when someone promotes his 'ware' . Take what is available or not. Talking about what would be good to have in offer in shop ... not good example. It's not something what I just need to order from manufacturer - I mean improved TOS 3.xx, 2.xx. 4.xx . No, someone or some team needs to spend plenty of time to make it. May be thousands of hours of work. That's what I mentioned - how much time it took already.

I started thread at atari forum without word 'atari' in it's name: https://www.exxoshost.co.uk/forum/viewtopic.php?f=39&t=929

Funny thing is that it is at moment last so-so active thread in SW section there And where they are with it ? Nowhere actually, discussing about diverse C compilers and about is EmuTOS slow or not. That's just pathetic, and yes - off topic. And I add: 0 constructive.

That would be my answer about improvements 'to 3.06' . Since C sources are available, should be done using them, instead much harder 'disassemble it accurately, do all changes in ASM' - that's what I did with 1.04 and 1.62 .

And I would do same with 2.06 if it's object (binary) code would be close to code of 2 mentioned - I checked, and it is not. Surely done with different C compiler (version) . + I explained that no real need for it. We have combo of 1.04/2.06 already over many years.

 

Then, people who have C knowledge, are aware about available TOS sources don't care about TOS improving. They care about EmuTOS and other alternatives. They talk not nice things about TOS, often. They want that people use EmuTOS. So much, that don't care about SW compatibility Pun intented.

 

Please understand that it was lot of work and time spent. And I lost 2 months in Fall because of 2 movings. I'm happy with what I did so far. If this will not be popular, and I really don't expect some big sales, then surely will not deal with other TOS versions. Actually, my motivation was never big considering Falcon and TT TOS versions, or HW upgrades for them. I'm ST man in first place. Maybe Mega STE man in last years

 

Considering shipping of EPROMa or Flash EPROMa in PLCC case - no problem with that. I used them too in some of my HW prototypes, and can program such chips.

[leech]

Main reason I asked about the TT, is since you have already done the work for the 2.06 version, it was my understanding that 3.06 is basically the exact same, with the TT resolution support (and maybe some tweaks to make the 68030 work better).

 

Ha, I always thought that was one of the points of bringing this forward, for if there are other ideas of improvements, and possibilities to get upgrades and much needed fixes to our beloved systems. Clearly you are a very capable person to be able to accomplish what you have done.

 

I thi k one of the problems of the Atari community though is that each of the masterminds that are developing hardware and software for it do it on their own instead of collaborating with others. I understand reasons why that is, and have read far too many flame wars that are mostly pointless over it.

 

Reminds me of comic book villains. If they could ever set aside their conflicts and not being able to share world domination, they could probably defeat the good guys and rule the universe.

 

Ha, I don't mean any offense by calling people comic book villains, I would probably turn into Doctor Doom if I could.

[ParanoidLittleMan]

I made only combination of TOS 1.04 or 1.62 GEMDOS part with Desktop/AES part of 2.06 . Did not mod or improve TOS 2.06 self. It is possible because design of TOS is that GEMDOS part starts AES and Desktop like some APP (after floppy, hard disk boot, AUTO run) . So, can run some other AES version, not only what goes with that TOS version. But GEMDOS part must be modded, if nothing else it's location(s) in ROM space.

3.06 of course can not be used with some TOS 1.xx, because TT has lot of different HW in compare to ST(E) . Btw. 2.06 and 3.06 practically differ only in TT HW support in later.

 

There are teams in Atari community - like Hatari, EmuTOS team. Or some others. I don't see myself as team worker. And I can certainly do many things faster alone. Communication can take lot of time and even energy.

I'm really shocked about what trivial things they exchange posts at exxos forum.

 

Surely there is competition in all it. And that's as everything - has it's good and bad sides. Healthy competition is good. But when some start to spit other's work, even to spread lies, + overglorifying own stuff, that's not good.

And recent experience with newbie game programmer, who just refused some simple advice. It's easier to get insulted than learn some things. And we all can always to learn something.

I don't see actually some real difference in how people acting in real life and on Internet, forums. Same bad things from noisy people, while good ones are usually silent. Some want cheap fame, some work really hard, and many of them left this waters because low feedback, or some really bad actions - like Steven Seagal some months ago, after some really not nice posts performed by now new SW section moderator at Atari forum without Atari in it's name. Comic books are everything but not realistic. We don't need superheroes. Just less lawyers and better politicians

[leech]

Ah, I see, so you basically just took the feature set of 2.06 and backported, so to speak, into 1.04/1.62, plus some other modifications. It had initially sounded to me like you had improved the Mega STe (2.06). And if I am getting that wrong, sorry it is 2am here.

 

Ha, the comparison to comic book villains was just a jab at how so many in the community just result to insulting each other instead of helping each other out.

 

Seems at least with FreeMiNT they are doing some great work to add stability and features to it, was just hoping some people could get together with deep knowledge of TOS and upgrade it like some did for AmigaOS, where they released 3.1.4 recently after decades, we have a new AmigaOS release with better CPU support, and a ton of improvements.

[+Sauron]

Ah, I see, so you basically just took the feature set of 2.06 and backported, so to speak, into 1.04/1.62, plus some other modifications. It had initially sounded to me like you had improved the Mega STe (2.06). And if I am getting that wrong, sorry it is 2am here.

 

Ha, the comparison to comic book villains was just a jab at how so many in the community just result to insulting each other instead of helping each other out.

 

Seems at least with FreeMiNT they are doing some great work to add stability and features to it, was just hoping some people could get together with deep knowledge of TOS and upgrade it like some did for AmigaOS, where they released 3.1.4 recently after decades, we have a new AmigaOS release with better CPU support, and a ton of improvements.

 

The AmigaOS has professional developers working on such things because they still have a large enough active base of users to support it, albeit barely. Unfortunately the same can't be said of the ST, so it's all now reliant on open source projects. It would be nice to see EmuTOS grow enough to a point where it can truly be a drop-in replacement for TOS, but it needs much better compatibility than it currently has.

[ParanoidLittleMan]

It would be nice to not go off topic.

And what I did is not that kind of TOS replacing . Even comparing this with EmuTOS is bad idea. Complete different approach, very different aims and ideas how to make it more user friendly and support for new ways of storage.

Does user base size really matter ?

So, I will say clearly: This is not TOS replacement, this is TOS improving. Keeping it's original design, fashion, and fixing some not good solutions (FAT16), adding new features, while maintaining original RAM layout, way of booting from disks and so on - yes, mostly in purpose of good SW compatibility. EmuTOS will be never much SW compatible with their current "we do all according to TOS DOCs" approach. + despite all bad what I said about TOS and it's flaws, it is done on high level - those programmers at DR knew something. Disk code is where it is still much better than in EmuTOS , and VDI ...