Propeller help for the old !
elder1943
Posts: 8
I am a new user of the Prop chip.· I am truly impressed with the versitility an possibilities.· I have a 3 older pieces of lab grade test equipment (such as Fluke 8860A) which are well built and still would be useful except that they have antique microprocessors.· In the case of the Fluke 8860A the "in bound" slave processor is a Mostek (young people, pls google "Mostek") MK 14574P-5.· Not only is the chip·impossible to get·but Mostek vanished·years·ago.
My question (before I start) has anyone "looked"/published anything using the Prop... together with the proper interface chips to emulate older microprocessors.· I understand this is a small "nitch" application which would be mostly a labor of love rather then profit.
My goal is a pcb "header" which would plug into the origional socket.
Thanks·
Jim
elder1943
·
My question (before I start) has anyone "looked"/published anything using the Prop... together with the proper interface chips to emulate older microprocessors.· I understand this is a small "nitch" application which would be mostly a labor of love rather then profit.
My goal is a pcb "header" which would plug into the origional socket.
Thanks·
Jim
elder1943
·
Comments
This is a work in progress but it does nun CP/M quite well already. These are becoming quite long threads and the latest versions of ZiCog are near the ends.
Whether this can be modified to drive any real peripheral devices I don't know. Depends on what they are and what circuit they sit in.
Certainly a labour of love rather than profit but at least one person is looking into using this to replace an entire Z80 CPU card in some instrumentation device and run the code extracted from it's PROMS.
potatohead and co. are working on a 6502 emulation in the Hydra forum.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
For me, the past is not over yet.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
For me, the past is not over yet.
The place to start would be to get a data sheet for the chip. Analyzing that, along with the propeller's capabilities, would show if an exact hardware level emulation is possible.
If it is not possible with a Propeller, it certainly would be with an FPGA.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
www.mikronauts.com - a new blog about microcontrollers
The difficulty in emulating a processor is threefold:
1. You need detailed documentation for the ISA, hardware specs.· Only recently did the Atari 2600 emulators figure out some of the borderline 6502 BCD ADC/SBC cases.· (It turns out the docs the C64 crowd had developed weren't completely correct.)· Sometimes undocumented instructions get used - stuff which isn't in the normal manuals.
2. Complete emulation is more difficult than partial emulation.· If you know what features get used, then your task gets much easier.· But if you have to emulate every feature (because you may not know which features are used) then your task is much larger.
3. Requiring 100% accurate timing complicates the task a hundred-fold.· It's one thing to execute the instruction stream, it's quite another to execute the instruction stream and interface with external hardware which expects·B to follow A exactly 1.117usec later.
The Propeller also brings with it it's own set of challenges and limitations.· Each cog only has 498 words of memory for code and fast storage, so emulator tricks like jump tables for instruction decoding aren't as useful.· HUB RAM is only 32K and that has to include your emulator as well, so emulating a processor with 32K or 64K of RAM means you need external memory - and you can't just drop a SRAM onto a bus.· Eight cogs is a lot of processing power, but only if they are working on loosely coupled tasks.
That's not to say it can't be done, but doing so takes a lot of dedication.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Composite NTSC sprite driver: http://forums.parallax.com/showthread.php?p=800114
NTSC color bars (template): http://forums.parallax.com/showthread.php?p=803904
shop.ebay.com/?_from=R40&_trksid=p3907.m38.l1313&_nkw=Fluke+8860A&_sacat=See-All-Categories
Just a thought,
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Watching the world pass me by, one photon at a time.
Jerry
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
For me, the past is not over yet.
··· I think it is easy to buy a new USB multimeter in place of your Fluke 8860A. Parallax has one of those for around hundred bucks. I think this is simpler than·emulating·an obscure·or "not completely" documented processor? My $0.02
Sincerely,
Venkata.
I am old enough to have used Mostek parts, and I think I may still have some of their chips in my parts box. Unfortunately not the one in the Fluke. If the code for the processor is recoverable (external eprom/rom/etc.) it might be possible to emulate it on a prop. If not you would have to interface the prop to the measuring circuitry directly and write equivalent software. Not an easy task, but probably easier or at least not much more difficult than emulating the Mostek chip.
I have to agree with VG that it may be better to replace it. Unless you are really fond of that Fluke and want to restore it as a labor of love it is just not worth it. I speak from experience. I just resurrected an old Fluke scopemeter and although it works well the screen is very difficult to read in all but the best of lighting.
Actually what I had in mind was creating plug-in replacements for the tubes, probably using MOSFETS. I did this fairly sucessfully a long time back on a Gonset mobile shortwave converter.
I would probably use the filament winding on the power transformer to create the required voltages. My "round tuit" list is so long now that the proposed conversions probably won't ever happen. BTW, the HP 201C and the ancient Ferris RF generator still work great. I may even have a spare acorn tube for the oscillator in the Ferris.
I did have a monstrous Tektronix 545 oscilloscope but I gave it to a retired electronics technician (in his 80s). He uses it mostly to amaze and entertain the children in his neighborhood. He and I both used those things when they were brand new!
Jerry
I am an "old fart". I just "recapped" my EIP 545A. The Fluke 8860A, Tek 465/75/85 and its brothers/sisters are from the golden age of test equipment. Mercedes autos today are junk. Very high priced Chryslers with an undeserved symbol on the hood. The MB golden age was (for me) 1979/1989. They built some truly increditable cars. Much of the test equipment is worth preserving from say from 1985/2005 because of the quality and the simple fact it is repairable/maintainable by you and I We have already have had a major scope company abandon a 2xxx line of scopes filled with unobtainable VSLI. Shoddy controls/case/power supply throughout. This "improvement in design" made maintaining/repairing the 'boxes" too expensive/impossible for the individual with a small budget. (see Tek2 forum).
The robots I build often have SMD parts on the PCB. I "work" the PCB under a video microscope with a 15 inch VGA display which I built rather then use the simpler less costly USB stuff. Yes, it is a labor of love and with the 8860A and I may have hit the wall. It appears that the "Mostek" part "may have been a custom part made for the Fluke company, as I have been unable to find a trace or mention of the MP outside of the Fluke manual.
The Tek2 guys are already building many of the small parts needed to keep the Tek quality test equipment running. I was hoping that with modern fast processors creating a "clone" Mostek might be doable. Maybe not......
And no, the hundred dollar "just as good" new USB based subs are not as good, as useable, as operator friendly, etc, etc, etc............
My most profound thanks for all who took the time to respond.
Jim Ward
elder1943
Jim Ward
elder1943