Overall strategy for hardware mods, upgrades, clones

Troubles with your machine? Just want to speak about the latest improvements? This is the place!

Moderators: Mug UK, Zorro 2, Greenious, spiny, Moderator Team

User avatar
Ragstaff
Atari Super Hero
Atari Super Hero
Posts: 572
Joined: Mon Oct 20, 2003 3:39 am
Location: Melbourne Australia
Contact:

Overall strategy for hardware mods, upgrades, clones

Postby Ragstaff » Fri Feb 17, 2017 2:36 pm

This is just meant to be a discussion, a "brain fart" which you can ignore if you want to!
I'm not trying to tell anyone that what they are doing is wrong! We have a great, active hardware development scene, which I think is beyond our expectations for 2017.

I was just thinking the hardware development and mods for the whole Atari / TOS platform, including clones like Firebee, has many OBVIOUS positives, but there is also the one obvious negative. That is, it fractures an already small and varying platform into smaller pieces.
It's a bit like the universe after The Big Bang, with all these nodes drifting further away from the centre.
Some of the nodes have no active software developers or support to keep improve compatibility, or to make special software that takes advantage of that hardware. For example certain obscure video cards, the drivers are long out of development even though there are many improvements that we'd like to see, and there are certainly no special demo's coming along to really give the owners of that hardware some joy.

On the one hand, we go towards "GEM Compatible" software that will run on everything - Firebee's, TT's, Falcons, ST's accelerators, machines with graphics cards and GEM drivers.
That is fine, but has drawbacks too. (I won't go into that in details, but clunkier software, the ST and Falcon are more than "GEM computers", and there is OS fracturing too, lots of different multi-tasking setups, OS choices).
The point I wanted to discuss was at the hardware level rather than just software, and what opportunities there are to bring some convergence back to our hardware platforms, rather than divergence. Whether this could be considered an over-arching strategy that feeds into current decisions being made about hardware

Example 1) VME graphics card for the Mega STE and TT which, instead of being some new chipset with new drivers and new wheels to invent, contains a VIDEL implemented in FPGA (or even Supervidel). This card could also contain a 56k DSP. and a blitter on the FPGA (not for speed - the TT CPU is already faster, but to bring a little ST compatibility to the TT).
This kind of card could converge the Mega STE and TT back towards the centre of gravity, rather than being a mod that takes them further away.

I'm quite a lamer really, so I don't know if you can map those peripherals to the same memory address as original hardware on the Falcon and STE, or if they have to live somewhere else on the VME bus, but even if we can't map them to the same addresses, at least it makes patching software that uses this hardware much easier (it could probably be automated, or on the TT perhaps in realtime), while providing a graphics card that many people want regardless.

Example 2) Something that is already done - the expansion slot exxoss and rpineau are building into their accelerators. To me this can fit into the strategy of convergence of hardware, because there would be two distinct accelerators that can take the same expansions, and if those expansions conform to the strategy then it can bring machines fitted with those accelerators back towards the main "hardware gravitational mass" as well. E.g. the card from Example 1 above could have a version for these accelerators.

User avatar
exxos
Hardware Guru
Hardware Guru
Posts: 4846
Joined: Fri Mar 28, 2003 8:36 pm
Location: England
Contact:

Re: Overall strategy for hardware mods, upgrades, clones

Postby exxos » Fri Feb 17, 2017 3:08 pm

I think the expansion bus is the best idea. As you say, if I do a 68000 booster with the same expansion bus connector as R.pineau's 020 card, then anyone could build a alt-ram card for example and it would work on both boosters. Though this has been talked about I think over a year ago. Not 1 person said they would be interested in developing hardware for the expansion port. This in itself means its down to people like myself and rpineau to develop the stuff. Which is one reason why its taking so long to make progress as we are having to do everything ourselves.

While the expansion has its obvious merits, there is also the issue that from the STFM, STE, TT, MSTE, physical fitting of the boards becomes a problem. rpineau posted about this a few days ago in fact. What will fit in the MSTE likely won't fit in the STFM. So it would still require a different addon card board layout to physically fit in the machines.

It gets to the point that if nobody is going to build any addon cards, then there is little use in having that option. For me, I did develop a alt-ram card to work with my V1 STE booster, but that was a optional addon. Some want the RAM, some don't. So having a expansion port is still useful in some respects. Though for a final board design, it would be easier and cheaper just to have one board and either solder or not solder on the SRAM chips. Does away with the connections and not having to have 2 separate PCBs.

In relation to FPGA, this has been talked about a lot as well. I was all for a compact FPGA blitter to build onto all my boosters (well aside from the STE one). Though nobody really wants to develop that. Same with FGPA 68000 cores, they can be more efficient but again nobody is interested in it. Same with the MMU, have a faster MMU & CPU, use SRAM, would give the ST line a huge boost in speed, but again nobody is interested in doing that. I get asked about graphics a lot, sure I can add a gfx chip onto the bus, but who is going to write the drivers ? Likely nobody. The list is endless.

The bottom line is to me, all developers of hardware just want to do their own thing and thats it. some users will go to the FPGA route such as firebee or whatever, I myself are only interested in real hardware, so its what I work with. I doubt anyone is going to spend huge amounts of time in developing a FPGA MMU & CPU to use on my boosters. Though even so, while it would be awesome, your half way there to a firebee anyway.

I work with the STFM mostly, its a more common machine IMHO. STE as well I guess. Other machines are around of course, but ST lines was the most popular, its what I work with. If I can give those series of machines a new kick of speed, then thats a good thing. If everyone in the world had a ST booster, maybe more people would start coding again if they had more CPU power to work with. Though AFAIK, nobody else is interested in the STFM or developing for it other than myself.

I really want to stick to real hardware, but that is just me. While FPGA could be used to improve a lot of things, nobody else is interested in it. I don't have the time to figure it all out myself, its just re-inventing stuff again. Mostly people want compatible as possible to use the software, so its what I try and stick to. Of course each hardware change starts to break stuff, going up TOS versions can break stuff.

It's not a easy topic overall. Everyone wants everything, but like I keep saying to people, I am only one person, if loads of people want a gfx card then they need to do some work themselves and develop it and write drivers etc. I'm all for something like a higher resolution ST with 256 colours.. If the ST had more CPU power and faster RAM then it would be awesome. Though then people would have to write games for it to make use of it. Then your back to people can only play them if they have the new hardware. Which really makes it pointless anyone developing any software if less than 10 people around the world will ever be able to use it.

While I do a lot of booster work as you know, I mostly are not spending much time on it these days, I am more for keeping the original hardware running, doing fixes and upgrades such as the new PSU's etc.
4MB STFM 1.44 FD- VELOCE+ 020 STE - Falcon 030 CT60 - Atari 2600 - Atari 7800 - Gigafile - SD Floppy Emulator - PeST - various clutter

http://www.exxoshost.co.uk/atari/ All my hardware guides - mods - games - STOS
http://www.exxoshost.co.uk/atari/last/storenew/ - All my hardware mods for sale - Please help support by making a purchase.
http://ataristeven.exxoshost.co.uk/Steem.htm Latest Steem Emulator

Maeke
Captain Atari
Captain Atari
Posts: 233
Joined: Sun Mar 13, 2016 1:54 pm

Re: Overall strategy for hardware mods, upgrades, clones

Postby Maeke » Fri Feb 17, 2017 3:23 pm

Btw if anyone is willing to help developping a gfx for the expension port, maybe we could use this as a base of work, http://www.harbaum.org/till/atari/index.html, though as he said, exxos can't do it right now (too much work allready in cue for only one person) and i don't have the necessary tools, and training, to do it myself (hadn't do layouts for 20 years except for a small st(f)/e rtc board, but exxos finished this job for me), so using modern layout software for a gfx is actually out of my reach, and i don't have the necessary programmer either.
So if anyone wants to try this project, you're welcome.

Maeke
Captain Atari
Captain Atari
Posts: 233
Joined: Sun Mar 13, 2016 1:54 pm

Re: Overall strategy for hardware mods, upgrades, clones

Postby Maeke » Fri Feb 17, 2017 3:29 pm

forgot to say i'm used to french logic symbols, not US ones, there are enough differences to make US schematics hard to read for me (and again 20 years without working in electronic developpement made me rusty).

Look here for a correspondance in both symbols https://fr.wikipedia.org/wiki/Fonction_logique (it's in french but it doesn't need translation for the part i want to show).

rpineau
Captain Atari
Captain Atari
Posts: 470
Joined: Wed Jun 29, 2011 6:39 am
Location: California / USA
Contact:

Re: Overall strategy for hardware mods, upgrades, clones

Postby rpineau » Fri Feb 17, 2017 5:17 pm

So, interesting indeed.
For the expansion port, this is highly dependent on the CPU. On the 68000 booster that Chris does, we put all the 68000 signals. On the 68020 booster me and Juliusz are doing, we put all the 68020 signals and even added the 68030 ones for the future. But if you develop a card for the 68020 expansion it'll work on the 68020 and 68030 but not the 68000 (32 bit bus vs 16 bit bus, DSACKx vs DTACK, ...) unless you add a lot of logic for it on the 68000 card (aka doing the same adaptation as what we don on the 68020 card for the 68000 bus). Also a card that takes advantages of the extra 68030 signals will not work on the 68020 (STERM , cache burst signals ,...).
So even with a common form factor expansion with a set pinout we would run into some issues (even though less so than with totally different expansion busses).
Regarding the blitter and what not on the VME bus, the blitter need to be able to be bus master, the Atari VME slot doesn't allow bus master.. so at least that part won't work.
As for video card I started working on one for our 68020 booster that will use the Epson S1D13506. it has its own DRAM (one external chip) and can do 800x600 in 16 bits. The component is 16 bit so can be adapted to VME bus, 68000 expansion or 68020/030 expansion bus ... but we need drivers (fVDI ?).
So it's not that easy to get convergence.
Rodolphe
Falcon + AB040 + Eclipse PCI + ATI Rage VGA card + NE2000 Ethernec + HxC Floppy Emulator
MegaSTE 4MB + CosmosEx / 1040 STF for hardware dev
http://www.rti-zone.org/atari.php

joska
Fuji Shaped Bastard
Fuji Shaped Bastard
Posts: 3239
Joined: Tue Oct 30, 2007 2:55 pm
Location: Florø, Norway
Contact:

Re: Overall strategy for hardware mods, upgrades, clones

Postby joska » Fri Feb 17, 2017 5:21 pm

Ragstaff wrote:Something that is already done - the expansion slot exxoss and rpineau are building into their accelerators.


There already is such a port, the MegaBus. It accepts everything that's piggybacked to the CPU, as well as the existing graphics cards, alt-RAM, network cards, IDE-cards etc that uses this port. It can also solve the "68000 in different locations"-problem in the STFM by creating variants of the 68000->MegaBus adapter that locates the MegaBus connector to the same place in all STFM's.
Jo Even

Firebee - Falcon060 - Milan060 - Falcon040 - MIST - Mega ST - STM - STE - Amiga 600 - Sharp MZ700 - MSX - Amstrad CPC - C64

rpineau
Captain Atari
Captain Atari
Posts: 470
Joined: Wed Jun 29, 2011 6:39 am
Location: California / USA
Contact:

Re: Overall strategy for hardware mods, upgrades, clones

Postby rpineau » Fri Feb 17, 2017 5:37 pm

@Joska,I agree that the megabus is great for 68000 machine. But it's fairly limiting for machine that have faster CPU and can use 32 bits data bus and faster clock. This is why I "create" the current expansion bus on our 68020 card so that new card can be 16 or 32 bits (the video controller I named above is a 16 bit component, the DRAM controller we will be working on will do 32 bit data transfer, the network card I plan on making will use a SMSC LAN 9218i which is also 32 bit, ... all these card are already at various stage of development, some are even fully routed and can be produced for our current dev board version, but we first want to fully finish the 68020 card).
We even added 8 /IRQx signals on the bus and plan to use a trim down version of the VHDL code for the 68901 from Suska as an interrupt controller.
So with the Mega bus, some of these card would be limited and might not be able to all run together.
Rodolphe
Falcon + AB040 + Eclipse PCI + ATI Rage VGA card + NE2000 Ethernec + HxC Floppy Emulator
MegaSTE 4MB + CosmosEx / 1040 STF for hardware dev
http://www.rti-zone.org/atari.php

User avatar
Eero Tamminen
Atari God
Atari God
Posts: 1457
Joined: Sun Jul 31, 2011 1:11 pm

Re: Overall strategy for hardware mods, upgrades, clones

Postby Eero Tamminen » Fri Feb 17, 2017 7:34 pm

rpineau wrote:As for video card I started working on one for our 68020 booster that will use the Epson S1D13506. it has its own DRAM (one external chip) and can do 800x600 in 16 bits. The component is 16 bit so can be adapted to VME bus, 68000 expansion or 68020/030 expansion bus ... but we need drivers (fVDI ?).


There should already be couple of 16-bit drivers for fVDI on which to base a new one (latest one's Vincent's Amiga one), so that seems fairly feasible.

joska
Fuji Shaped Bastard
Fuji Shaped Bastard
Posts: 3239
Joined: Tue Oct 30, 2007 2:55 pm
Location: Florø, Norway
Contact:

Re: Overall strategy for hardware mods, upgrades, clones

Postby joska » Fri Feb 17, 2017 8:32 pm

rpineau wrote:But it's fairly limiting for machine that have faster CPU and can use 32 bits data bus and faster clock.


That is a good point. Maybe one could use the MegaBus, and add a second socket for the extensions? That would increase it's usability a lot.
Jo Even

Firebee - Falcon060 - Milan060 - Falcon040 - MIST - Mega ST - STM - STE - Amiga 600 - Sharp MZ700 - MSX - Amstrad CPC - C64

rpineau
Captain Atari
Captain Atari
Posts: 470
Joined: Wed Jun 29, 2011 6:39 am
Location: California / USA
Contact:

Re: Overall strategy for hardware mods, upgrades, clones

Postby rpineau » Sat Feb 18, 2017 3:53 am

Latest fvdi source update was r483 from Johan in svn.
The github page is also not showing much update (https://github.com/jklockars/fVDI_from_ ... aster/fvdi ). latest log is from "Thu Jan 24 11:46:17 2013"
Vicent Riviere has a fork (https://github.com/vinriviere/fvdi ) which seem to have more recent changes he did for the Amiga stuff you mentioned.
So I guess Vincent's fork is the one we should look at for implementing new drivers.
Rodolphe
Falcon + AB040 + Eclipse PCI + ATI Rage VGA card + NE2000 Ethernec + HxC Floppy Emulator
MegaSTE 4MB + CosmosEx / 1040 STF for hardware dev
http://www.rti-zone.org/atari.php

Faucon2001
Atari Super Hero
Atari Super Hero
Posts: 531
Joined: Sat Oct 26, 2013 11:19 pm
Location: Brasil
Contact:

Re: Overall strategy for hardware mods, upgrades, clones

Postby Faucon2001 » Sat Feb 18, 2017 11:53 am

I don't have any hardware knowledge, and I am really impressed by the ingenious work you are all doing, but summing to Ragstaff comments, I wonder if there is another way to think hardware development. Let me explain what I mean based on the following:
1- we can't separate hardware and software.
2- most of the good Atari softwares are no longer developed
3- even though there are still developments, ressources are limited.
4- compatibility with old software is an issue difficult to solve.
So taking for example the video card expansion, why putting a new video chip which needs a specific driver to be developed and FVDI which is still incomplete and in standby, instead of an ET4000 which is managed directly by NVDI and compatible with many legacy softwares?

Wouldn't it be an better path? is it feasible or a stupid idea? Can't we reuse/improve solutions done in the past and which are known to work with legacy softwares?
Philippe

Firebee, Falcon, STE, Aranym Box, Hatari Pi Box.
My music http://www.philippeworld.net/
My photography http://phil-67.deviantart.com/
EasyAraMint, BeeKey and BeePi https://sites.google.com/site/emaappsarch/home

BlankVector
Captain Atari
Captain Atari
Posts: 377
Joined: Wed Oct 24, 2007 7:52 pm
Location: Paris, France
Contact:

Re: Overall strategy for hardware mods, upgrades, clones

Postby BlankVector » Sat Feb 18, 2017 1:11 pm

rpineau wrote:Latest fvdi source update was r483 from Johan in svn.
The github page is also not showing much update (https://github.com/jklockars/fVDI_from_ ... aster/fvdi ). latest log is from "Thu Jan 24 11:46:17 2013"
Vicent Riviere has a fork (https://github.com/vinriviere/fvdi ) which seem to have more recent changes he did for the Amiga stuff you mentioned.
So I guess Vincent's fork is the one we should look at for implementing new drivers.
Rodolphe


Johan is currently moving official fVDI sources to GitHub. It is work in progress, and the repository you mentioned is part of the process. This will take time, as Johan is gathering all his old backups to recreate the full fVDI history as completely as possible.

This is why I (Vincent) have setup a private temporary fork on GitHub. I will contribute to Johan's official sources when he is ready.

And indeed, considering the crazy bugs I needed to fix in fVDI recently in order to get simple drivers working, I strongly recommend to use my repository as base for new driver development. Until my fixes are merged into official sources.

Maeke
Captain Atari
Captain Atari
Posts: 233
Joined: Sun Mar 13, 2016 1:54 pm

Re: Overall strategy for hardware mods, upgrades, clones

Postby Maeke » Sat Feb 18, 2017 2:02 pm

Faucon2001 wrote:I don't have any hardware knowledge, and I am really impressed by the ingenious work you are all doing, but summing to Ragstaff comments, I wonder if there is another way to think hardware development. Let me explain what I mean based on the following:
1- we can't separate hardware and software.
2- most of the good Atari softwares are no longer developed
3- even though there are still developments, ressources are limited.
4- compatibility with old software is an issue difficult to solve.
So taking for example the video card expansion, why putting a new video chip which needs a specific driver to be developed and FVDI which is still incomplete and in standby, instead of an ET4000 which is managed directly by NVDI and compatible with many legacy softwares?

Wouldn't it be an better path? is it feasible or a stupid idea? Can't we reuse/improve solutions done in the past and which are known to work with legacy softwares?


The main reason to search for new video chips is that the et4000 will sooner or later become very hard to find AND not all et4000 are compatible.

joska
Fuji Shaped Bastard
Fuji Shaped Bastard
Posts: 3239
Joined: Tue Oct 30, 2007 2:55 pm
Location: Florø, Norway
Contact:

Re: Overall strategy for hardware mods, upgrades, clones

Postby joska » Sat Feb 18, 2017 2:54 pm

Ideally one would create a shifter AND videl compatible graphics card using an FPGA. Then you would use NVDI (or just TOS) VDI simply by setting the video base address and fiddle with the LineA variables to get a larger screen resolution.

There is no advantage in using ET4000 except that the drivers already exists.
Jo Even

Firebee - Falcon060 - Milan060 - Falcon040 - MIST - Mega ST - STM - STE - Amiga 600 - Sharp MZ700 - MSX - Amstrad CPC - C64

Maeke
Captain Atari
Captain Atari
Posts: 233
Joined: Sun Mar 13, 2016 1:54 pm

Re: Overall strategy for hardware mods, upgrades, clones

Postby Maeke » Sat Feb 18, 2017 3:35 pm

joska wrote:Ideally one would create a shifter AND videl compatible graphics card using an FPGA. Then you would use NVDI (or just TOS) VDI simply by setting the video base address and fiddle with the LineA variables to get a larger screen resolution.

There is no advantage in using ET4000 except that the drivers already exists.


And it's a lottery to find one compatible if you don't know how to tell the difference.

rpineau
Captain Atari
Captain Atari
Posts: 470
Joined: Wed Jun 29, 2011 6:39 am
Location: California / USA
Contact:

Re: Overall strategy for hardware mods, upgrades, clones

Postby rpineau » Sat Feb 18, 2017 5:47 pm

Thank for the update Vincent.

As for the ET4000, it's always hard to find one and they are going to become impossible to find, Add to this that not all are compatible depending on the DAC they use.
The new chip I'm looking at is not the most powerful thing in the video world, but has a direct 68000 and 68030 mode. So this makes a few things easier to do on the hardware side.
If we make a fVDI driver and make this card for our 68020 bus as well as VME, Mega Bus and internal Falcon bus, we can get more people to use it.
Unless someone offers to make a video card for all Atari (different version for whatever port is available) using a FPGA that would be great. So far, no one is making new video hardware for the Ataris except if you have a CT60/63 ...

Rodolphe
Falcon + AB040 + Eclipse PCI + ATI Rage VGA card + NE2000 Ethernec + HxC Floppy Emulator
MegaSTE 4MB + CosmosEx / 1040 STF for hardware dev
http://www.rti-zone.org/atari.php

rpineau
Captain Atari
Captain Atari
Posts: 470
Joined: Wed Jun 29, 2011 6:39 am
Location: California / USA
Contact:

Re: Overall strategy for hardware mods, upgrades, clones

Postby rpineau » Sat Feb 18, 2017 5:47 pm

double post.. deleted.
Falcon + AB040 + Eclipse PCI + ATI Rage VGA card + NE2000 Ethernec + HxC Floppy Emulator
MegaSTE 4MB + CosmosEx / 1040 STF for hardware dev
http://www.rti-zone.org/atari.php

Maeke
Captain Atari
Captain Atari
Posts: 233
Joined: Sun Mar 13, 2016 1:54 pm

Re: Overall strategy for hardware mods, upgrades, clones

Postby Maeke » Sat Feb 18, 2017 7:02 pm

rpineau wrote:Thank for the update Vincent.

As for the ET4000, it's always hard to find one and they are going to become impossible to find, Add to this that not all are compatible depending on the DAC they use.
The new chip I'm looking at is not the most powerful thing in the video world, but has a direct 68000 and 68030 mode. So this makes a few things easier to do on the hardware side.
If we make a fVDI driver and make this card for our 68020 bus as well as VME, Mega Bus and internal Falcon bus, we can get more people to use it.
Unless someone offers to make a video card for all Atari (different version for whatever port is available) using a FPGA that would be great. So far, no one is making new video hardware for the Ataris except if you have a CT60/63 ...

Rodolphe


CT60/63 meaning of course a falcon, am i wrong?

User avatar
Dark Willow
Atari Super Hero
Atari Super Hero
Posts: 668
Joined: Fri Mar 12, 2004 10:59 am
Location: UK
Contact:

Re: Overall strategy for hardware mods, upgrades, clones

Postby Dark Willow » Mon Feb 20, 2017 12:44 am

I recall putting out a concept for a new VME gfx card in the past, going with existing supported chipsets to avoid the driver headaches. The ET4K does have that advantage. Supply is surprisingly not an issue. Clones are still produced in China for some reason (embedded systems, I guess). However my first thought in this area would be the Radeon 9xxxx - drivers are out there, and it's a far more capable (and abundant) chipset. I'm not sure of the technical possibilities for a Megabus to PCI bridge, but given the potential flexibility it offers, if feasible it could dramatically expand the number of users of PCI on the Atari platform beyond CT-PCI and clone owners - making developing drivers for off-the-shelf cards more attractive and worthwhile - as well making hardware upgrades far more affordable
TT030, 10mb/16mb, 3.5" SCSI multi-card reader (2gb CF/512mb SD), Matrix TC1006 gfx card, EtherNEC, Epson GT scanner, VanillaMint | 1040STe, 4mb, TOS 1.62, MegaFile 60, SM125 | Lynx II | 800XL, 64k, 1050 disk drive, XC12 tape deck, MicroLink parallel adapter

PuppySoft:Projects & More

rpineau
Captain Atari
Captain Atari
Posts: 470
Joined: Wed Jun 29, 2011 6:39 am
Location: California / USA
Contact:

Re: Overall strategy for hardware mods, upgrades, clones

Postby rpineau » Mon Feb 20, 2017 4:26 am

PCI "bridges" like the one on the CT-PCI are still available (PLX PCI 9054 is about $75 on mouser and probably around the same price in other online reseller). Most of these require a full 32 bit address and data (like the expansion port on the CT60/63/60e or the one we're putting on our 68020 card).
We know some subset of the PCI bus were done on Falcon (Eclipse card) and I think on Mega ST using a CPLD but these only works with 1 video card (I tried a 2 slot PCI adapter on mine with an ethernet PCI card in the 2nd slot and when the pcibios prg starts it bombs so it might only be a software issue..).
As for the Radeon card, I think only a few are supported and not easy to find.
If the ET4000 if as you say is still produce could be a good idea if we can find a reliable source.
But I think this discussion goes beyond just video card and indeed PCI would open the options for more cards (we're swiping the issue regarding space in the machines under the rug for now ;) ).

I don't know of any other PCI I/O Accelerator besides the PLX PXI 9054 (that's what it's called.. a PCI I/O Accelerator) but I didn't spent too much time looking for other ones as we know this one is somewhat supported. We would still need some 680x0 bus to PLX 9054 glue (a CPLD should be able to do that :) ) for each "platform"... direct 68000 CPU to PLX, Mega ST Bus to PLX, 68020 bus to PLX, 68030 bus to PLX (if we want synchronous transfer using STERM instead of the DSACKx common to 020 and 030) ..

So interesting but still not straightforward and most costly than , if we take the video card example, using the Epson S1D13506 with a CPLD on our 68020 bus as the Epson S1D13506 is around $20, the 2MB DRAM for it is about $*, add the CPLD and the various component, the PCB and connectors and you can get a video card for about $50 to $60 that does 800x600 in 16 bit "True color".

None of this is set in stone and if a good and affordable solution comes along I'll be more than happy to use it.

Rodolphe
Falcon + AB040 + Eclipse PCI + ATI Rage VGA card + NE2000 Ethernec + HxC Floppy Emulator
MegaSTE 4MB + CosmosEx / 1040 STF for hardware dev
http://www.rti-zone.org/atari.php

Maeke
Captain Atari
Captain Atari
Posts: 233
Joined: Sun Mar 13, 2016 1:54 pm

Re: Overall strategy for hardware mods, upgrades, clones

Postby Maeke » Mon Feb 20, 2017 8:39 am

besides designing a gfx specific for the atari line of computers would have another advantage, the needed space could be reduced. Making them easyer to fit in a st(m)/f(m)/e or a falcon.

joska
Fuji Shaped Bastard
Fuji Shaped Bastard
Posts: 3239
Joined: Tue Oct 30, 2007 2:55 pm
Location: Florø, Norway
Contact:

Re: Overall strategy for hardware mods, upgrades, clones

Postby joska » Mon Feb 20, 2017 9:03 am

Maeke wrote:besides designing a gfx specific for the atari line of computers would have another advantage, the needed space could be reduced. Making them easyer to fit in a st(m)/f(m)/e or a falcon.


Exactly. And it would be easy to route the Shifter/Videl output through them, making it a lot more userfriendly than the traditional graphics card. They would also use less power than obsolete PCI/ISA graphics card, which isn't a bad thing considering that the ST/Falcon PSU isn't very powerful.

I've had various graphics cards for my Falcon, TT and ST's for 20 years now. I also have a Milan that use PCI graphics card. Yes, colours are nice. Large resolutions are even nicer. But there is no denying that there's a lot of issues too. There is so much software out there that just does not work with anything else than the shifter and videl. And not only that, software that appears to be clean GEM applications has problems in colour modes on graphics cards because the AES and VDI does not really handle anything above 256 colours at all. Basically, the application is on it's own if it wants to create and display truecolour bitmaps. So there's not really that much software out there that supports this. Try finding a good pixel painter that works in 16 or 32 bit mode on an ATI Rage... There's loads of them that works on the Videl though.

And that's where I think the SuperVidel is spot on. I think the SV concept is the only one that offers both large resolutions and colours and still keeps compatibility.
Jo Even

Firebee - Falcon060 - Milan060 - Falcon040 - MIST - Mega ST - STM - STE - Amiga 600 - Sharp MZ700 - MSX - Amstrad CPC - C64

Maeke
Captain Atari
Captain Atari
Posts: 233
Joined: Sun Mar 13, 2016 1:54 pm

Re: Overall strategy for hardware mods, upgrades, clones

Postby Maeke » Mon Feb 20, 2017 9:37 am

I won't talk about the compatibility issue, since i never had a gfx for my ste or my (recently acquired and restored) stf, i didn't know about this, but i can relate for the psu problem, indeed unliike the pc's psu, the genuine atari psu were made for the componants of the computer they were meant to power on, so they didn't need to accept much power loads, and thus weren't very powerfull, after all the stf(m)/e psu just had to support the motherboard, the floppy drive, the mouse (nod a big stress) and joysticks most of the time, even the cartridges didn't need much power i think, but a gfx can give a very heavy load, especially a pc one.

User avatar
Ragstaff
Atari Super Hero
Atari Super Hero
Posts: 572
Joined: Mon Oct 20, 2003 3:39 am
Location: Melbourne Australia
Contact:

Re: Overall strategy for hardware mods, upgrades, clones

Postby Ragstaff » Tue Feb 21, 2017 3:50 am

What do you think is better? Video/Supervidel in FPGA, or VIDEL emulation on a small, low-power chip like a single-core ARM CPU which costs a few dollars? These go in mobile phones and use hardly any power, you could probably make a tiny PCB that would fit inside an ST case

User avatar
leech
Captain Atari
Captain Atari
Posts: 355
Joined: Tue Dec 01, 2015 3:26 pm

Re: Overall strategy for hardware mods, upgrades, clones

Postby leech » Wed Feb 22, 2017 6:09 am

I would love for various SuperVidel upgrades to the older ST/TT computers. A VME with SuperVidel on it would cover MegaSTe and TT0030, maybe even toss on the NetUSBee stuff? That would be the ultimate upgrade for those two computers at least.


Social Media

     

Return to “Hardware”

Who is online

Users browsing this forum: No registered users and 5 guests