ParanoidLittleMan

Not sure that it really conflicts. Someone has concrete experience and explanation ? Even IDE port on Monster should work well, because AdSpeed slows down to 8 MHz when accessing regular memory, ROM and bus (and IDE port is on it). Of course, there may be conflict about how to install it physically.

It's sold. Yes, much faster than Satandisk - about 6 times.

It is not at all sure that Satandisks are the problem at you. Possible reasons: crap SD cards - and there is lot of it on market, especially online. Weak PSU for Satandisk. Failure in ACSI port, DMA section in your Mega ST. Note that I don't talk about 'bad DMA chip' - that's case with early STEs, and that's true, with published bad versions of chip. Overall unreliable work of your Mega ST. Should at least check it's power supply, and recap by need. And yes, Satandisk can be problem too, but I really did not hear about much problems with them. Satandisk works fine with all ST, Mega ST, STE (with good DMA chip), Mega STE, TT .

VME bus has own address space. I know that IDE adapter via VME worked not without adding extra address logic for IDE port. So, most likely enhanced joystick port is not accessible there too. Whole thing seems too complicated. And will need to add more than just connectors for the port self - there are 6 chips and lot of resistors in STE at those connectors . I guess that getting STE will be better solution, especially for someone inexperienced with electronic.

That's pretty bad formulation. Since MonsTer self has TOS 2.06 support, you don't need this solution described here.

What average user can do is to clean up it. Ensure good contacts - in case of Atari ST it means mostly reseating socketed chips (those square ones in first place), removing rust from pins by need. That must be done carefully, not to damage sockets, which are from stiff plastic. If that helps not, it will need someone with experience and equipment. And may need replacement of some component(s) . Btw. the reason why nothing on screen is most likely that it fails at some stage of early HW initialisation.

New home computer in years 2019-2020 should met some criteria: 1st and most important is computing power. 2nd is OS - and that's at least 100x, if not 1000x more complex now than in 80-es. Internet support, 3D graphic is elementary now. Multimedia, of course. In Atari there was some early level multimedia, + GUI, and that's all what was there basically. Vampire is just not enough powerful for today. And as I see, nothing from 1 GHz 68K compatible CPU - all is still in FPGA. And even 1 GHz would be too slow for today, when multi core, 3-4 GHz CPU is affordable. There are reasons why Apple and Motorola abandoned 68K line, and went on RISC based Power PC. 68000 was great for it's time, but it was designed 40 years ago. Sometimes you just need to change basic architecture to go forward. EmuTOS as main OS in 2019 ? Actually, then better old good TOS 1.xx-4.xx - at least they will run old Atari SW well. As said, today need some very complex OS in compare to old times, and not ROM based, for sure. I think that this thread leads nowhere. Today, launching new computer, what is not PC based, compatible is extremely demanding, and will cost hundreds of millions. And years of development. It was not easy even in 1985, demands were much higher than in 1982 - year of 8-bit machines without real OS, just some simple basic and tape loader. New designs happen only in portable area - tablets, smart phones, and even there, OS is not really new - Linux or Windows based.

That was so-so possible in early years of home/personal computers, when machines were with OS in ROM, of small size. And HW was not so complex too. Now, modern OS consists of some million lines of program code. That's something what needs bigger team. Even Atari TOS, what is only some 200 KB took plenty of time, and they even could not finish V 1.00 at time, so first machines were supplied with TOS on floppy (was bad advertisement) . Other thing is that making some up to date and to current user needs version of TOS, GEM - for whatever CPU would be practically making it from scratch (zero) . That would need years of work of many people. Even big company like Google did not made Android as complete new OS, it is Linux based.

The real question here is status of Atari corporation today. I don't think that much from old Atari left at all. Name is sold, probably multiple times. And name is what is good advertisement. Even younger people know what name Atari means, at least know that was pioneer in video gaming, that made some home computers. I personally don't expect that they - current owners, leadership even thinking about some new 'home computer' - what is now little obsolete term. As I see, some don't consider PC as home computer, but it is it actually. And even can connect it to new TVs, unlike it was in 80-es. I see idea about some new, keyboard computer as bad. Or better said, we have plenty of it - called notebook 😀

According to what I read about AdSpeed it is 16 MHz CPU with some cache, so basically very similar to what is in Mega STE . What means that should work with any TOS version. Couldn't find some photo of installed ST version, so can not say how will be good to connect to CPU lines (mostly address) with GAL for address decoding. In any case, those lines - A17-A23, /AS, /RW, /DTACK are accessible at other locations too.

Older v. - black case, 2 SD slots vertically ordered. Needs about 10 V power with DC connector (internal regulator and Greatz bridge). With ACSI cable. Works flawless. Price is 40 Euros + shipping.

It is known problem that with IMP MMU chip 2.5 MB (2 MB + 0.5 MB) works not properly, and only 2 MB works. I had that problem in my Mega ST, and others too.

Well, calling something hack is a bit matter of taste. Usually it means doing something not supported by basic design. Or some simple solution what works for purpose, but may be problematic in some situations. TOS 2.06 is done with intention to work on ST machines too - there is code what detects HW, and will not try to use HW available only in Mega STE or STE. TOS 2.06 can not fit in ST TOS ROM space. That was actually not future proof design error - they set TOS 1.00 ROM address space on very high location, right under HW register address space, and it can be max 192 KB long - $FC0000-$FEFFFF . At $FF0000 starts HW address space. Because 2.06 is longer (actually even 1.06 did not fit in 192 KB, so they moved TOS ROM address space to $E00000 in STE because that). In other words, it was simply not possible to make TOS 2.06 as simple put ROM chips in machine solution. Using TOS from RAM is not really good solution if you want to use it regularly. That's something what can be called rather hack, but again, tastes ...

I made it in my Mega ST. Here is how looks: EPROMs go in original sockets, so not higher. I soldered GAL on top of 1 20 pin TTL chip - that's where flat cable from CPU goes. But it can be placed anywhere (near) . Need to solder some 15 lines, most is between CPU and GAL . That's only way without some intermediate board, sockets. I'm ready to ship this - 4 programmed EPROMs (US 2.06 TOS is what I used here) + programmed GAL for decoding new address space of 2.06 for ST, Mega ST.

SM124 in monochrome (B/W) monitor. Composite is for color monitors .