Future of the project.

Related information and WIP etc
User avatar
JezC
Posts: 2081
Joined: Mon Aug 28, 2017 11:44 pm

Re: Future of the project.

Post by JezC »

I'm also more interested in the faster (possibly less compatible) future board rather than the smaller compatible option...again this is mostly due to having quite a few STF/STFM machines (and a few STEs as well as a Mega 4 & a Falcon) already.

I do understand the wish not to spend too much time on accelerating the H4...but won't the SEC board be a possible option for the H4 (when you get time to finish it off)?
User avatar
exxos
Site Admin
Site Admin
Posts: 23499
Joined: Wed Aug 16, 2017 11:19 pm
Location: UK
Contact:

Re: Future of the project.

Post by exxos »

@JezC The SEC is designed to fit the H4. But it does not speed up RAM. A FPGA system does. In order to make use of a 50MHz CPU, it needs to have faster access to RAM..
https://www.exxosforum.co.uk/atari/ All my hardware guides - mods - games - STOS
https://www.exxosforum.co.uk/atari/store2/ - All my hardware mods for sale - Please help support by making a purchase.
viewtopic.php?f=17&t=1585 Have you done the Mandatory Fixes ?
Just because a lot of people agree on something, doesn't make it a fact. ~exxos ~
People should find solutions to problems, not find problems with solutions.
User avatar
JezC
Posts: 2081
Joined: Mon Aug 28, 2017 11:44 pm

Re: Future of the project.

Post by JezC »

Yes, I get that...and I'm hoping to get to the point where the raw processing power of the H4 derivatives is faster than the Falcon...but I'm probably going to be looking for a faster STF option in the shorter term (as well as the faster FPGA-based one in the longer term).

So, knowing what our possible options for moderate acceleration might be in the interim is still very much of interest (maybe even one of the TF series of accelerators).

I've got one of your earliest 16MHz accelerators to try in one H4 as a first step but I'm sure that won't be the ultimate boost :)
User avatar
exxos
Site Admin
Site Admin
Posts: 23499
Joined: Wed Aug 16, 2017 11:19 pm
Location: UK
Contact:

Re: Future of the project.

Post by exxos »

The SEC will couple with the fast-ram boards for 50MHz fast-ram, that will fly on anything which can run in alt-ram. Once the SEC is finished (albeit slowly) its not much of a jump to a 020 CPU and 32bit. So I am not stopping anything development wise H4 or not.
https://www.exxosforum.co.uk/atari/ All my hardware guides - mods - games - STOS
https://www.exxosforum.co.uk/atari/store2/ - All my hardware mods for sale - Please help support by making a purchase.
viewtopic.php?f=17&t=1585 Have you done the Mandatory Fixes ?
Just because a lot of people agree on something, doesn't make it a fact. ~exxos ~
People should find solutions to problems, not find problems with solutions.
User avatar
derkom
Moderator
Moderator
Posts: 1208
Joined: Sun Jul 29, 2018 6:45 pm

Re: Future of the project.

Post by derkom »

I agree with the faster/enhanced/less compatible crowd. We've already got options when we want a compatible machine, including the current H4. There aren't really many options out there for a truly enhanced ST, and it would be fun to go down that path.

You say you can't work on two boards simultaneously, which I get, but it seems like for those desiring a fully compatible machine, the H4 is nearly all the way there already. Maybe another revision to fix the oopses, implement a couple of small ideas, etc., but when the objective is compatibility, it's already there. Leave the H4, or the slightly tweaked version thereof, as the compatible choice for people who want that, and then head down the path of breaking things to make life exciting.
mikro
Posts: 474
Joined: Mon Aug 28, 2017 11:22 pm
Location: Kosice, Slovakia
Contact:

Re: Future of the project.

Post by mikro »

I, for one, would be really happy if the goal was still to have a cloned ST. I don't really care about acceleration on ST, I love to watch demos, maybe play a game or two, simply - have fun. If I wanted an accelerated ST, I would fire up an emulator.

Not surprisingly, I fully support the idea of replacing ST PCB with Exxos PCB in the original case because again, if I wanted a square totally new PCB, I would go for MiST or Suska.
User avatar
stephen_usher
Posts: 5580
Joined: Mon Nov 13, 2017 7:19 pm
Location: Oxford, UK.
Contact:

Re: Future of the project.

Post by stephen_usher »

I was thinking, if you're going for a full 32 bit machine, why stop at the 020? Why not go for an 030 and build a TT-alike? In that way you can utilise TOS 3.06.

Of course, there's no real need to implement the SCSI but instead add a couple of IDE channels.

The only major issue would be recreating the video sub-system and maybe the DMA sound.
Intro retro computers since before they were retro...
ZX81->Spectrum->Memotech MTX->Sinclair QL->520STM->BBC Micro->TT030->PCs & Sun Workstations.
Added code to the MiNT kernel (still there the last time I checked) + put together MiNTOS.
Collection now with added Macs, Amigas, Suns and Acorns.
User avatar
exxos
Site Admin
Site Admin
Posts: 23499
Joined: Wed Aug 16, 2017 11:19 pm
Location: UK
Contact:

Re: Future of the project.

Post by exxos »

There are 030 cards out there already . no need to reinvent them. I chose the 020 as its the more compatible than the 030. Evey CPU and OS increase creates its own compatibility issues as well. If people want fastest machine, buy a falcon and a CT60.

The suska has a lot packed into it plus its small if that's what people want. I'm not going to compete with such existing products. Wolfgang has a 030 CPU core now , maybe he will add that to suska. If that's what people want, I have of course no objection if people want to go and support suska and help develop that system.

My goal has always been compatibility as there is nothing out there. Second is speed, but still a ST. Of course we are constantly working on addons .

I might have not been clear about the thought of a basic STM type machine..idea to make it cheaper and easier to finish with the aim of it becoming a complete machine. As people know, machines are failing so much that we are at the door of STs no longer working and fixing them is a lot of work and expense for people. While a new motherboard is expensive, its still going to be cheaper in the long run.

There needs to be more interest in such motherboards and getting boards as complete and cheap as possible is the only way it may encourage more people to get onboard with the project. As I said before many times, the ST has a huge software and games library. We need to make use of that, which means keeping as close to ST as possible.

I know people don't use the floppy drive anymore other than me. But something to consider is the collectors also. Look at the all those 2600 people who spend a fortune on original carts. They could download all the games and have them on a SD card instead. But not everyone wants that. Some like originals and a more retro feel of taking the games out of the box and loading them off floppy. If people don't want that , that's fine I they get a gotek .

I suspect as people here are basically developers, everyone is looking at the H4 for the way forward, I don't disagree . though there are people who are not developers, can't solder, they just want a ST that works which they can play games on. I think there's a lot larger Atari crowed out there than people think. I know going by the amount of stuff which sells around the world in my store.

I think part of the problem with higher prices for Atari's lately is part lack of sellers these days. Years ago evilbay was flood with them. Evilbay drove a lot of sellers away which doesn't help. Though with basically untested / faulty machines being all that is left, people likely will go buy 2 or 3 machines in order to keep 1 working. With supposed serviced machines selling for 200+ people might as well buy a scrap machine and build a H4 instead. It will be more reliable and if people like the thing they can easily plug in upgrades at a later date.

In terms of cases. I'm not a fan of the mega box. While it is smaller, it then needs a second case for the keyboard. Assuming we will 3D print new cases , the STM is small enough but can also hold the keyboard. I'm not saying it will be identical, but just something smaller than a STFM case. Its to keep printing costs down. I'm not saying we won't print STFM cases for the H4 either. Though its keeping the ST alive and generating new blood.

If we could build a complete machine which is affordable, then it makes no sense for people to pay 200+ for a tested machine which will likely soon die anyway.. When people could buy a complete machine which will last for not much more.

Don't get me wrong I appriciate peoples input to the future of the project. Though I think we need to think more towards a doomsday where every ST will be dead and nothing affordable out there to replace it. There is MiST etc and emulators, but nothing like the original hardware other than what we are doing . have to think "new blood" and creating something affordable to attract new people to the ST world. Plus again, we don't want to recreate the MiST or suska either and we have to draw a line somewhere.
https://www.exxosforum.co.uk/atari/ All my hardware guides - mods - games - STOS
https://www.exxosforum.co.uk/atari/store2/ - All my hardware mods for sale - Please help support by making a purchase.
viewtopic.php?f=17&t=1585 Have you done the Mandatory Fixes ?
Just because a lot of people agree on something, doesn't make it a fact. ~exxos ~
People should find solutions to problems, not find problems with solutions.
User avatar
rubber_jonnie
Site Admin
Site Admin
Posts: 10472
Joined: Thu Aug 17, 2017 7:40 pm
Location: Essex
Contact:

Re: Future of the project.

Post by rubber_jonnie »

For me the remake project is more about having a solid, working STFM, with maximum compatibility and the ability to easily add worthwhile add ons like IDE interfaces, 1.44 floppy support or more RAM, but still be able to fit within the original form factor of the ST casework. The H4 and it's successors can do just that.

The problem with STs these days is that they are crumbling. Poor quality PCBs, poor components and an electronically dirty design mean that sometimes even just picking one up is enough for it to die. I just had my test board FDD die. I didn't drop or do anything terrible to it, I picked it up, put it down, then came back to it a few months later.

A board to rejuvenate the many STs out there is a good thing. Unfortunately, the MegaSTE, TT and Falcons are much rarer and very difficult to come by, so unless every owner of these machines starts shouting out that they need replacement boards, the more ubiquitous STFM is where it needs to be at. Well IMHO.

For me, using an FPU or 68020 on an STFM is probably irrelevant (Well 68020 perhaps for speed), as unless you have the software that will use these extra capabilities, then there is no benefit.

I see this project, and it sub projects as having the possibility to keep these machines going fo many more years to come, just like putting a newer more up to date chassis in a car.
Collector of many retro things!
800XL and 65XE both with Ultimate1MB,VBXL/XE & PokeyMax, SIDE3, SDrive Max, 2x 1010 cassette, 2x 1050 one with Happy mod, 3x 2600 Jr, 7800 and Lynx II
Approx 20 STs, including a 520 STM, 520 STFMs, 3x Mega ST, MSTE & 2x 32 Mhz boosted STEs
Plus the rest, totalling around 50 machines including a QL, 3x BBC Model B, Electron, Spectrums, ZX81 etc...
User avatar
exxos
Site Admin
Site Admin
Posts: 23499
Joined: Wed Aug 16, 2017 11:19 pm
Location: UK
Contact:

Re: Future of the project.

Post by exxos »

Things like the Falcon, I know Wolfgang is working on chip clones and has been for some time for it. But that work could still be ongoing for some years. But it will end up being the same problem that the chips will not be totally accurate. Just look at the problems recreating the ST chipset... There is just no support there or interesting recreating the original chips. For one it is not a simple endeavour particularly where not many of us even can code FPGA stuff. I can see nothing but headaches trying to go in that direction.

From a software point of view, again "ST" line is where all the software is at. Some stuff will run on a Falcon, but like you say, there is a limited number of titles specifically done for the Falcon and indeed the CT60.

As some know I have a Veloce 020 STE here, whether it be the CPU, machine, or TOS version, I basically gave up because I could hardly get anything to run on it... Move on to the next step which is the Falcon, TOS404, then the CT60 etc etc.. I just want to avoid that direction. There are already a lot of things out there like MiST, Emulators, all great of course, but currently are not a fully compatible ST system.

Like others have said also, each machine after the ST as its own features and upgrades like MSTE, Falcon etc. while there are titles which may well use STE features such as DAC playback.. we are not really building a STE... As I've mentioned several times when all this began, why do we even have to go to the many routes that Atari did anyway ? If we are going to break something, let's break something which we actually designed to start with.

Accelerators such as 020, or 030 with TF's stuff, or PAK or whatever, can still be added into the H4 if people want it. While the 030 is "faster".. Its data cache breaks self modifying code, but also I have seen as others have, that the data cache can actually hinder performance slightly as well. , In terms of ST software, the 030 I think is actually worse than the 020.

In terms of my SEC booster, as said before, the CPU basically wires up like a 020... It technically isn't a 68000 CPU.. It has missing pins just like the 020 CPU does and have to be emulated... So theoretically should not be much of a jump to change the SEC CPU for a 020 CPU.

Aside the CPU aspect, the Falcon has the DSP sound system, but basically while this is really good, nobody is really doing software for it, it would be incredibly complicated and time-consuming to add this into a ST design, when no software would use it anyway. Would basically fall back on myself to do the hardware and software for any future upgrades.

Now I have spoken to some demo coders last year who would be willing to adapt tracker code to work on a new sound system and one guy even suggested some features for improvement as well. I have mentioned this before, but if we had a sound system with its own RAM, and able to do pitch change and volume control all in hardware.. Then while currently the tracker code maxes out the CPU, we could have several channels of CD audio quality with very little CPU overhead.

For example, four channels in tracker modules have to be mixed into a single channel. This mixing takes up a lot of CPU time whereas this could easily be done in hardware if we had four separate playback channels. So right now we have just saved like 50% CPU time. because such a sound system would have its own RAM, it would not eat up the main system RAM, and the CPU would not have to fetch data from ram to send it to the audio playback circuit, this again frees up a lot of CPU time.

Such a circuit will be relatively simple to construct I think, and as I would end up doing the initial playback routines I can give the demo coders some examples of how the system works and then coders can take over using such a system. It would give them something new to play about with it is what we need to do as well, is encourage coders to actually code for such machines. Such programming will be relatively simple to get started with, whereas I suspect the DSP on the Falcon etc is a lot more complicated to get to grips with and so people will just not bother.

At heart it is still a "ST" system, which will be able to use all legacy software, but we are also adding in new features which people /coders can play around with. Such a sound system is not dependent on the TOS version. and it could be used in any programming language from STOS, GFA,ASM etc. We need to encourage more people to code for such machines not just play games on them.

So my overall aim is to build a solid ST legacy system, then just simply go in our own direction for upgrades and add-ons. At least we can document all the hardware and software properly this time, rather than just guessing at how things work with buggy TOS routines etc.
https://www.exxosforum.co.uk/atari/ All my hardware guides - mods - games - STOS
https://www.exxosforum.co.uk/atari/store2/ - All my hardware mods for sale - Please help support by making a purchase.
viewtopic.php?f=17&t=1585 Have you done the Mandatory Fixes ?
Just because a lot of people agree on something, doesn't make it a fact. ~exxos ~
People should find solutions to problems, not find problems with solutions.
Locked

Return to “H4 MONGREL EDITION DEVELOPMENT & INFO”