Jump to content

guus.assmann

Members
  • Content Count

    389
  • Joined

  • Last visited

Everything posted by guus.assmann

  1. For 1.44 MB to work, there's also a Gal that needs to be the right revision. If not, it won't work either. If I'm not mistaken, the one you need is in this .zip However I'm not sure which one to get. And I don't have source code. I did replace that one in my computer as well and also placed the FPU and Gal with that. BR/ Guus GAL.zip
  2. If the capacitors are removed and the resistors and possible coils are bridged, the I/O will be faster. However, the computer will not pass the EMC test. That may cause problems in some case. It may cause RF interference. The 68B21 has some higher drive capabilities. That can also cause more RF. Not always a big deal really. BR/ Guus
  3. Hello, The circuits in the Atari are all 5V. So is the RTC. But it will remain working fine at a lower voltage. 4,5V is well within the safe range. Just make sure that the batteries will not be emptied and start leaking. That would make a terrible mess and now you'll get 3 instead of 1. You might consider putting in Eneloop cells. Then you'll get about 3,6V. BR/ Guus
  4. I own at least 2 of those boards, one with original box and documentation. They were designed for the Mega ST and have this size, so the can be securely be mounted in the computer. There's a couple of standoffs at the end of the board, away from the connector. Also there's some metal shielding. There's a 16Mhz oscillator on the board. I've put a socket in and tested with higher frequencies. Above about 24Mhz there's not much improvement anymore. (Gembench test) The speed is determined mainly by the data transfer between CPU and FPU. BR/ Guus
  5. Sorry, I've only used the board a couple of times using Dos. So I'm not sure about any windows version. BR/ Guus
  6. Hello, Yes, you're right that there's a difference. I'm not sure if the "simpler" version will work. Maybe some specs can be found for the board? (Or the readme file with the software) Both the one in my Mega STE and the one that's not mounted do have an FPU socket. And I also have 1 without the FPU socket. That one is also not mounted. BR/ Guus
  7. Hello, I own a Mega STE with an AT-Speed in it. It does contain a 286 and also has a Chrystal on the PCB. (32Mhz is engraved) And the Atari does run at 16Mhz. Also, there does exist an adapter from the PLCC 68000 to a Dip 64 socket that's specially made for an 286 PCB. I don't have the schematics for either card. The Adapter I've just seen is made by GE-Soft. My Mega STE works well, until I move it between rooms. As long as I don't do that and let it sit in it's own place, it's working happily. BR/ Guus
  8. Have a good look at the Gotek documentation. There's a possibillity to use pin 2 for disk-change signal. That may be of use. (Have not tried it myself...) BR/ Guus
  9. There's an adapter board to interface between the PLCC socket and the DIL socket. However that may be hard to get and can cause contact problems. I have to reseat mine every once in a while. So yes, it will function just fine in the MegaSTE, also if that's run at 16Mhz. The 80286 has it's own clock.... BR/ Guus
  10. Hello, The fact that you succeded is proof that you did it right. And the price for the chips is quite good I think. In future, also look at the speed of the chips. 150nS is mostly good enough. (On 8Mhz system 200nS will also do) On a TT you may want to use 120nS. Erasing depends on the light source. Just experiment with shorter times. If you test before programming, it's ok. One very important thing for erasing Eproms. Make sure they're in conducting foam,all pins must be electrically connected. Failing to do so may destroy an Eprom. Especially the older ones like 2716 or 2732. The even older 2708 will definitivly not survive if pins are not connected. I've learned that in the time that an Eprom costs FL 150 (Current equivalent is $ 150). They were not my Eproms, my manager killed them and I found out why. BR/ Guus
  11. What's the speed of the Eproms? 120nS would be ok. Slower may cause problems. BR/ Guus
  12. Hello, That's a matter of taste. Personally, I like GFA Basic better. (Has lots of examples as well). BR/ Guus
  13. Hello Bikerbob, It is very possible that the revision of the test cartridge is the problem and not the hardware itself. I've had this issue, until I used another (multi) testcartridge. It might need a test cable or another fixture to pass this timing test. Does the testcartridge start the menu? Maybe after hitting the space key? You should be able to do all other tests. However, I cannot tell you what revision to use. (I'm not home at the moment) BR/ Guus
  14. Mostly, if you heat up the label, the glue will become soft and the label will come of more easy.
  15. Hello, I have made a PCB, many years ago, that made it possible to use a 65C02 in the XL's or XE's. And the XL7 is already in this forum, look at the attachments I put in. Schematics and source code are there. Just needs some sorting out. There's "logic" in the file names. BR/ Guus
  16. Hello, Technically a bit of topic. And I've made the assumption that Bob has no objections, as he offered to look for some stuff. A while ago, Bob has done some significant speedup of a 1200XL. And I've tested it on the 600XL, 800XL and 130XE. It worked on all boards. ! But I don't recall ecactly what's what. And I also tried to get it to work at 14Mhz and include the OS and Basic-Roms and XE-GS ones as fast Rom. It didn't work and I should have bought a logic analyser to get it to work. Never done that or spend enough time.... BR/ Guus Speedy816_Bob_Wooley.zip PCBS.zip SCH.zip
  17. Politicians discuss about laws and this changes these laws. So they think that discussing the rules and laws of nature can have the same effect. They generally lack the intelligence / knowledge / interest to understand it's not possible.
  18. Hello, This behaviour would happen if the track zero sensor is not working. It can be checked quite easy. One side is an infra-red Led. Just see if there's a voltage drop across it, or accross the resistor that feeds it. And the foto transistor on the other side can be checked as well. Just measure the voltage across it. This should be low when the led is active and shining on the transistor. The voltage should change if you put a piece of paper in the gap. You can also check the voltage on the input of the PIA that processes the signal of the Zero-sensor. Just check the schematic diagrams for this part. BR/ Guus Assmann
  19. A 1,44mb drive in a 1050 is next to impossible. The 1050 has a stepper motor that's driven by it's firmware and also the signals from the heads are processed on the 1050 PCB. In a 3 1/4 floppy unit, these functions are performed by the drive unit. It is possible to put this 1,44mb unit in a XF-drive. Modified firmware is available as well. And you can buy a PCB to (re)build this drive. BR/ Guus
  20. Hello tjlazer, With only 1 Eprom, you have 8 bits data. And the computer needs 16 bits at the same time. Maybe, with some complicated electronics it's possible to multiplex the data.......... ?? This is a bit like buying a Rolls Royce to cross the street and go to work as a toilet cleaner..... :-) BR/ Guus
  21. Doing this work in pure assembler is not crazy, but may be time consuming. If done properly, the resulting code is generally faster than if a "high level language" is used. Having the source in C-code must be a lot of help, provided that some comment is there or the code is not difficult to understand. I wish I had some more knowledge on the 68000 programming, but I don't. My capabilities are more on the hardware side. Replacing SMD, even a 144 pin chip is not a problem for me. Or a PCB design including prototyping. However, this experience is no use for writing software. Coöperation on such a software project would be very helpfull. But it's hard to find others, willing to commit to this. Also, not everyone has the time or the drive to take on such a task. BR/ Guus Assmann
  22. Hello, Maybe it's possible to compromise? Most high level programming languages have the possibility to include code segments in Assembler. (or from other languages) So you might leave the major part as is and only remove the parts that you want to improve. That way, you could leave all library and other things as they are. No need to adapt the source code (and other supporting files) for the other "C-dialect" Maybe this also makes documenting the changes / work easier. BR/ Guus Assmann
  23. 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)
  24. 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.
  25. 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. :-)
×
×
  • Create New...