BFEXTU's ProdigySE Project

[BFEXTU]

...and so it begins...

Specs:
MacSE/800k Single Floppy
ProdigySE 4mb board w/ PMMU and 3.0b ROM
ProdigySE Software v1.5 with RAMdisk 1.29 -- maybe the latest
16Mhz 68020 with 68881 FPU and 68851 PMMU (not used unless the 8mb version)
Recoverable/bootable RAMdisk

Status:
Machine boots and works
Board passes self-test (6 iterations)
Prodigy RAMDisk v1.28 and v1.29 both hang/crash the SE
ProdigyControl (control panel) and Prodigy SE Test applications are working
Floppy drive is working, but the eject motor gear could go any time

Research Proejcts:
See if possible to upgrade to 8mb (Same daughter card, different density or fully populate? I forget and haven't cracked the case yet)
Figure out the RAMdisk problem
Software exploration
Might have to rebuild the SE power boards...or not - OK at the moment.
Should probably do the floppy gear repair.
Dump/Archive the ROMs
Review incompatibility with FDHD drives (most likely due to the fact that it uses the old SE ROMs and FDHD drives require the newer SWIM-aware ROM)

 

[retr01]

This will be a fun thread while @Zane Kaminski is busily working on the WarpSE project. @JDW wants the sound distortion to be resolved. I want to have an awesome SE to work alongside my precious SE/30 after I pick a SE up.

So, from the WarpSE thread, I was wondering, why not change the ROMBase, modify some code, and replace the SE ROMs? Then, add two more lines from the 68000 in the Gerber file for a revised SE reloaded board. Finally, some software engineering magic to patch System 6 and 7. Those are three areas to work on to get the SE to the maximum 16 MB of RAM.

 

[retr01]

Of course, with this project here, giving the SE a 68020 with FPU and modifying the ROM and some lines of a cloned board to have 16 MB of RAM?

The 68020 can address 32 lines as it has a 32-bit data path out to the mobo. Did the folks who made the Prodigy board connect all 32 data path lines out of the 68020?

 

[BFEXTU]

As far as I recall from 1988, the '020 on the ProdigySE is using 32-bit addressing, as above.

I created some of the initial dealer support information for the ProdigySE (and the entire SuperMac product line), and I still have notes somewhere. I'll see what's there, but the information is from the era prior to the 8mb upgrade. However, there may also be an additional/related tech note on the 8mb expansion.

And, I think I am remembering that the earliest ProdigySE boards without the PMMU (assuming that is correct) had a 3.0A ROM. So, as above, I think that the max upgrade for the earliest boards was the 8mb RAM module, the 3.0B ROM and the 68851 PMMU. And, the more I write, the more I think this info is correct...but well into the realm of obscure factoids. However, I am very thankful to have found this board!

 

[retr01]

As far as I recall from 1988, the '020 on the ProdigySE is using 32-bit addressing, as above.

Cool.

I created some of the initial dealer support information for the ProdigySE (and the entire product SuperMac product line), and I still have notes somewhere. I'll see what's there, but the information is from the era prior to the 8mb upgrade. However, there may also be an additional/related tech note on the 8mb expansion.

That is awesome. Looking forward to some information.

And, I think I am remembering that the earliest ProdigySE boards without the PMMU (assuming that is correct) had a 3.0A ROM. So, as above, I think that the max upgrade for the earliest boards was the 8mb RAM module, the 3.0B ROM and the 68851 PMMU. And, the more I write, the more I think this info is correct...but well into the realm of obscure factoids. However, I am very thankful to have found this board!

Now that is something that raises eyebrows and starred eyes in my case. The 3.0A and 3.0B ROMs are something would have to locate and study the source code.

 

[BFEXTU]

I added dumping/archiving the ROMs to my task list and made a mental note to remember any ROM I see laying around labeled 3.0A. I have a little pile I recently discovered, but they are 3103's. The 3103 ROM was the old ROM for the stand-alone DataFrame SCSI controller boards that was replaced in the late '80s by the 3105 ROM.

 

[retr01]

@jajan547, can you follow this, too? We need better ROMBase and coding. This is getting started.

Oh, @Crutch brought up a good point that the Macintosh Portable has 9 MB RAM maximum because of the higher ROMBase.

 

[womiller99]

Bravo! Good hacking!

 

[BFEXTU]

I ordered an XGecu T48 Universal Programmer to dump the ROMs. They should be readable. The T56 is 2x the cost and I don't think I need the extra features.

The litmus test will be to read the ROMs and reprogram some replacements and see if they work. I can probably find some UV-erasable 27Cxxx parts.

 

[retr01]

So, I was reading pages 120 to 133 in Apple Guide to the Macintosh Family Hardware, Second Edition by Apple. I realized something.

The Mac SE can address more than 4 MB. I understand that addressable memory does not mean the RAM alone but other memory such as ROM and others. The maximum addressable memory is distributed to all necessary memory for various computer parts. If one part doesn't have memory space, how will it function correctly?

Then, I question Apple's honesty back then. Was Apple honest with the customers? Did Apple intentionally misplace the memory map and only give the end user 4 megabytes?

Most Apple dealers seem to do the dirty work for Apple by lying to customers. I have caught some dealers lying to customers, and they get away with that. "Oh yes, this computer is the tool of the mind." "It will blow you away." "It will be speedy and do anything." I understand how sales work. They tell what customers want to hear. Yet, Apple did things much better in this century than in the former century. Sure, the computers were fun, but I think they are capable of more.

So, let's see how SuperMac did better than Apple with the memory mapping.

 

[BFEXTU]

You can't necessarily judge the early hardware design decisions for a cutting-edge consumer computer in a fledgling industry with a backward-pointing lens based on today's standards. The first Mac was 128K and RAM was extremely expensive. 4mb was a lot at the start and there were many hardware and software design considerations in trying to make systems that would run compatible software, etc. It took a while for them to get the addressing right -- even the Quadra 900 was kind of screwed up. Now, 4gb is a pathetically low amount. A retrograde Moore's Law-driven perspective is your friend.

 

[retr01]

Yeah. What's done is done. Today, we can at least improve it and still have fun with those Apple computers that started a different way of thinking and doing things.

I guess that was a rant post.

 

[BFEXTU]

There are artifacts posted on MacRepo for the ProdigySE, including the User Manual (that had helpful coding tips in it (with the Prodigy's variant of classic programmer hex sig tagging -- 0xDEAD 0xFACE 0xBEEF 0xCAFE), the floppy disk image components/image and a dump of the 3.0B ROM (with the same part # as the one on my board).

When the ROM programmer shows up, I will be able to compare the image with the one already posted. They should match. Also, the contents of the floppy matches the contents on my disk, with the exception that I also have RAMDisk v1.29 vs. 1.28 (disk version). However, neither app works - they both hang (maybe an unimplemented trap error). But, I will look into using newer System versions. The disk has an older system 4.1/5.5 on it, so maybe stepping forward a little will fix the problem. They may have updated the apps without changing the System.

The RAMDisk App crashing/not working was a known/reported problem in 1988...but definitely back-burner.

 

[retr01]

There are artifacts posted on MacRepo for the ProdigySE, including the User Manual (that had helpful coding tips in it (with the Prodigy's variant of classic programmer hex sig tagging

Pages 23 and 27 are cropped off as the person who scanned it goofed up by shifting the manual to the right. It was on a bed scanner, and unfortunately, those physical errors are common. I think that person did not check the scan job before uploading.

Can you find another scan that will have pages 23 and 27?

 

[BFEXTU]

I may have the actual manual around somewhere.

But...page 23 is not too hard to read. Below is the text across the truncated lines.

that demonstrates the techniques
of the Prodigy line of enhancement products

working knowledge of 68000 assembly language
All segments are in MPW format

produce a change in the Ticks global

Page 27 appears similarly easy:
it is fairly
_Pack4 and _Pack5
it would be quite easy
versions of your application
conditionally execute, depending upon

Most important,
real is different between SANE and the 68881
five words (80 bits),

...

The reason for this has to [do] with the 68020

NaN (Not A Number)
If you need some more

what form your program wishes to store

 

[alxlab]

Oh this board is really interesting since it can go to 8MB and has a MMU! Looking forward to seeing what the board looks like.

 

[retr01]

I may have the actual manual around somewhere.

But...page 23 is not too hard to read. Below is the text across the truncated lines.

Cool. Thank you. Hopefully, you will find the actual manual somewhere.

 

[BFEXTU]

Oh this board is really interesting since it can go to 8MB and has a MMU! Looking forward to seeing what the board looks like.

Theoretically, it can go to 32Mb...but might need alternate parts (PAL, Gate Array, etc.)

 

[retr01]

Theoretically, it can go to 32Mb...but might need alternate parts (PAL, Gate Array, etc.)

For the 68020 or better processor that can do a 32-bit data path? The maximum addressability was 2 to the 32nd power, 4,294,967,296 bytes or 4 GB. Imagine being zippy without the BlueSCSI or MacSD.

 

[mg.man]

Looking forward to seeing what the board looks like.

There are a couple of threads over at the other place (with pics)...

Levco / SuperMac Prodigy SE accelerator and utility

About a year ago I bought - on eBay - a Prodigy SE accelerator board, which I intended to plug into the processor expansion slot on my SE which has a SuperDrive. Not knowing much about compact Macs or anything about the board, I put it in a drawer for safekeeping until I was ready. Recently I...

68kmla.org

Prodigy SE by Levco

Hi all, I came across a compact Mac SE last weekend, and noticed one of them had a sticker on the front saying 'Prodigy SE by Levco'. I did some research on Google and came across the details on what it was. It was another motherboard that piggybacked onto the existing motherboard, and had a...

68kmla.org

Cool board, would love to have one!