exxos's Atari, Amiga & retro forum

exxos wrote: 08 Sep 2023 20:14 I don't really class the Falcon as a great machine TBH. It has a lot of cool stuff of course. But as I have said many times in the past, I want to take the remake project to higher speeds which will beats the Falcon on raw CPU power. This opens the door for higher resolutions and such anyway.

For me personally, it is better having a "ST" is backwards compatible with all the existing games and software. Just because the original Atari went in the direction of the Falcon, not mean we have to follow.

Personally I consider the Falcon the best computer ever made, but I would trade it for a reimagined ST compatible if it had graphics that were even better than the Falcon. I would like to see extra graphics abilities similar to console and arcade hardware, such as Sega Genesis, Capcom CPS2, Neo Geo. Hardware sprites, hardware scaling, scrolling playfields, etc. Not sure if anyone else will ever be interested in that kind of thing, but I think it would be amazing. It is disappointing to me that this kind of fantasy system hasn't been attempted with FPGA machines like the MiST yet. If I had the knowledge to make such a thing, that's what I'd do.

stephen_usher wrote: 08 Sep 2023 22:04 Myself, basing it on the TT, with it's full 32 bit architecture and then adding Falcon graphics modes and DSP would be a whole lot better. Double the data bandwidth for a start without even changing the clock.

Not sure about recreating the VME bus, putting a PCI bridge in there might be a good idea.

If you want a TT then go design one. I personally think it would end up a complicated and costly product. Plus nobody came forward to porting the DSP to the H5. It's a lot of work and would need OS patches and someone to write code for it all. Just not going to happen.

The remake project could well go 32bit in the future. Though as the ST remake is costing a lot, I doubt many would pay for a 32bit machine. The aim was to go back to the core ST design to run all legacy software. Start off simple and work up. But I'm doubtful there will be anything past the H5 now. There are just to many things going against making any real progress these days unfortunately.

Xyla wrote: 08 Sep 2023 22:37 Personally I consider the Falcon the best computer ever made, but I would trade it for a reimagined ST compatible if it had graphics that were even better than the Falcon. I would like to see extra graphics abilities similar to console and arcade hardware, such as Sega Genesis, Capcom CPS2, Neo Geo. Hardware sprites, hardware scaling, scrolling playfields, etc. Not sure if anyone else will ever be interested in that kind of thing, but I think it would be amazing. It is disappointing to me that this kind of fantasy system hasn't been attempted with FPGA machines like the MiST yet. If I had the knowledge to make such a thing, that's what I'd do.

Everyone has a ideal / fantasy system. But lack of hardware and software developers kills it all off. It's what kills most projects off. Plus the more complicated something is the more time and cost it takes. There simply isn't anyone anymore to do much of anything . I've been active for 30 years, I'm looking to retire due to health reasons mostly but also the time and cost of everything takes its toll on you.

Sadly there are not many full-time developers left anymore. It's already caused me to abandon many projects. You just have to admit defeat at some point. I don't know how many developers work on MiST etc but I'd assume there's not many active coders anymore.

stephen_usher wrote: 08 Sep 2023 22:04 Myself, basing it on the TT, with it's full 32 bit architecture and then adding Falcon graphics modes and DSP would be a whole lot better. Double the data bandwidth for a start without even changing the clock.

there is such a machine and it is called Falcon Microbox:
https://wizztronics.com/atari-microbox/

exxos wrote: 08 Sep 2023 22:51

stephen_usher wrote: 08 Sep 2023 22:04 Myself, basing it on the TT, with it's full 32 bit architecture and then adding Falcon graphics modes and DSP would be a whole lot better. Double the data bandwidth for a start without even changing the clock.

Not sure about recreating the VME bus, putting a PCI bridge in there might be a good idea.

If you want a TT then go design one. I personally think it would end up a complicated and costly product. Plus nobody came forward to porting the DSP to the H5. It's a lot of work and would need OS patches and someone to write code for it all. Just not going to happen.

The remake project could well go 32bit in the future. Though as the ST remake is costing a lot, I doubt many would pay for a 32bit machine. The aim was to go back to the core ST design to run all legacy software. Start off simple and work up. But I'm doubtful there will be anything past the H5 now. There are just to many things going against making any real progress these days unfortunately.

Well, I never thought that anyone would think about building one, it was purely hypothetical, given the architectural problems with the Falcon.

There's no real point in making a new, suped-up derivative as the market would be vanishingly small and most people can get what they want from an emulator. Better to put the resources into keeping the originals alive and making replacements with "modern" components.

I'm still not 100% sure what to make of all this.

The announcement over on AtariAge seems generally positive and Al seems to think that it could certainly help with bringing more games of a high quality to release for multiple platforms. The prospect of digital downloads for their store and newer games sounds promising too.

It seems like there are some neat projects about, like the XP cartridges and the 2600+. Personally I'll stick to my 7800 since it also runs 2600 games and I have an SD card based cart for it, and though HDMI would be nice, the composite mod I have fitted gives a decent picture.

But I am always a bit wary when a large corporation with no other connection other than a name wants to buy in. I hope it is for genuine reasons and that they have real ideas and it isn't just about an IP grab or something like AmigaKit have been doing.

I guess time will tell, it's one to watch.

By the sounds of it Al wanted to step down anyway, where deleting the forum was even on the cards. So he deciced to pass it onto Atari. A risk yes, but anyone he potentially passed the forum over to would also run the risk of being shut down or messed with.

As he's now being paid by Atari , and they want to seemingly help with various things, overall it was likely the right move.

As to how it all ends , no idea. As long as Atari want to help and not "take over" making unwanted changes etc, then things should be OK. If Atari annoy the community then they will pretty much destroy themselves overnight. Likely why Al wanted to be deeply involved still as the forum is his life's work after all.

stephen_usher wrote: 09 Sep 2023 08:21

exxos wrote: 08 Sep 2023 22:51

If you want a TT then go design one. I personally think it would end up a complicated and costly product. Plus nobody came forward to porting the DSP to the H5. It's a lot of work and would need OS patches and someone to write code for it all. Just not going to happen.

The remake project could well go 32bit in the future. Though as the ST remake is costing a lot, I doubt many would pay for a 32bit machine. The aim was to go back to the core ST design to run all legacy software. Start off simple and work up. But I'm doubtful there will be anything past the H5 now. There are just to many things going against making any real progress these days unfortunately.

Well, I never thought that anyone would think about building one, it was purely hypothetical, given the architectural problems with the Falcon.

There's no real point in making a new, suped-up derivative as the market would be vanishingly small and most people can get what they want from an emulator. Better to put the resources into keeping the originals alive and making replacements with "modern" components.

I think the key to the whole upgrade thing is backward compatibility, which the F030 and the TT has problems with. This is due to several factors; the 68030 CPU being one of them, as the same timings required for some programs cannot be matched, especially if using the caches; a main component in the speed increase.

So whats a better solution?

The ideal machine IMHO is a base 68000 ST system with options for higher speed CPU/bus clock and perhaps some fancy GFX/audio enhancements in parts of the memory map which wont tread on the basic system and are perhaps not 'common hack' memory locations used by demo coders, or metal hitting app devs (if possible to avoid the odd broken demo/app, but by this stage, its probably easier to patch the few remaining problem games/demos/apps).

I know they are not generally appreciated here, but the Vampire080 seems to use this elegant augmentation method to achieve this result.

The apollo system seems to have a - not exploding - but still growing ecosystem based around it, which is now encouraging devs to make use of the (080/AMMX/SAGA) enhancements too (which they are).

I believe at least part of the reason is because of the above stated reasons.

As for the takeover, it may be good news if they are prepared to pay/licence programs/games (at a fair price) and then market them.

This may encourage more people to come out and develop for the ST range.

mrbombermillzy wrote: 09 Sep 2023 09:40 As for the takeover, it may be good news if they are prepared to pay/licence programs/games (at a fair price) and then market them.

This may encourage more people to come out and develop for the ST range.

Yep. If Atari take like 20% profit (after expenses) then it should be good for developers. If the developers get 20%;then I'd assume they would all walk away. Unless Atari promise millions of sales which realistically they can't.

It would be great if we could create a faster ST and have games made for it. It's the CPU being slow which is the main problem. I mean playing a protracker tune maxes the CPU out and we not even at 50khz playback yet.

I doubt any games developers would write in ASM. It would have to be "quick to do" which would involve frameworks / bloatware which will end up needing a ryzen based PC running an emulator to keep up. I don't think that's "in the spirit" of Atari though. But there isn't much option I think.