Jump to content
Trash80toHP_Mini

MicroMac Performer Redux: Cloning a 68030 Accelerator for Compact Macs

Recommended Posts

P1010052.thumb.JPG.017664bee6687a5b2b241899fd3f2eb6.JPG

 

P1010050.thumb.JPG.2cb84b8bf15356fc74f4d9c5b0814454.JPG

 

This one is mine, it came as wonderful surprise in my Drexel 128K that had been upgraded to Plus and then to "4MB SE/30" status over the course its remarkably extended useful lifetime. Not one of the original, silk screened student versions we've come to appreciate, but a killer machine from the storied halls of Drexel University nonetheless. But that's another story

 

Share this post


Link to post
Share on other sites

From the part number suffix, we'll assume my PGA 68030 powered Performer from 1994 is the 02 version representing an earlier generation of Performer.

 

PGA-SMT.thumb.JPG.641ab6cd22715074d65d6625fe697416.JPG

 

Sometime in the 1994 timeframe, Micromac switched from the PGA Package to the SMT version of the 68030. The suffix appended to the same part number on the next example is 101A for what it's worth.

 

Can't say for sure, the designations may only represent which of the three connectors were installed for packaging/inventory purposes? If anyone has a Performer, posting pics or just the (c)date, part number, processor package and connector type would be much appreciated.

Share this post


Link to post
Share on other sites

The only apreciable differences between the CPU type versions I've spotted so far was a switch from discrete resistors to resistor packs on the SMT revision. Can anyone spot any other differences?

Share this post


Link to post
Share on other sites

Was going to post that one as well.

At least it made its way onto my desk now - will buzz out the connections once I can find some time.

IMG_3861.JPG

Share this post


Link to post
Share on other sites

Sweet! Are you buzzing that monster above or the simple little card I've made my project? Figured I'd put thruholes for my proposed PCIe 68030 PDS connector on my update of the Performer. Figured you might have something to plug into that someday. [;)]

 

What do you think, will swapping a 50MHz crystal can for the 25MHz can on MicroMac's 16MHz accelerator card run the CPU at 33MHz?

Share this post


Link to post
Share on other sites

I played around with connectors on perf board for a bit and decided my latest notion of using the 68030 PGA socket as a 68030 "PDS" expansion connector will work better than PCIe for adding a memory expansion shield or whatever.

 

Today was really good, my pair of 33MHz SMT 68030s arrived! [:D] Also sourced PLCC-20 thruhole sockets today, dougg3 said the SMT sockets he used were a bear to install. These should be a lot easier unless trace congestion should force use of SMT sockets after all. I may order both.

 

Do SMT PLCC sockets install easily using a hot air station?

Share this post


Link to post
Share on other sites

Realized my verbal explanation of the 68030 PGA Socket interconnect wasn't at all clear and that one of my AI diagrams won't suffice, so:

 

^^$_57.JPG

 

This board handily illustrates what I'll be attempting to fit on my first 10cm x 10cm SEEED board. By dropping support for the Classic, the hole in the middle of the board for the A/B cable goes away.

 

The thruholes you see for its 68 pin socket interface (it's really more, but we'll say 68 for simplicity's sake) will become a three ring PGA array for a socketed 68030. That allows for testing both processor packages on my board.

 

A fifth GAL socket will be added to match the loadout of the Performer.

 

Visualize the SE's 68 pin EuroCard PDS connector (same as used for a NuBus card) added to the above card per the Performer config.

 

Space limitations will likely require dropping the 68000 socket to keep the prototype within the confines of an inexpensive board, but can be added to the final version as could the crazy form factor for Classic compatibility.

 

As a last resort, the FPU socket will be dropped from the prototype form factor to reclaim its PCB acreage. A purist I'm not.  [}:)]

Share this post


Link to post
Share on other sites

Blocking.jpg.d72eee911e404e66dd3444635452b02d.jpg

 

edit: while in merciless KISS editing mode, I decided to replace the standard SE connector on the prototype with the right angled version. Testing with the card mounted vertically in the SE like an SE/30 expansion card in a Late SE or SE/30 chassis simplifies things remarkably well.

Edited by Trash80toHP_Mini
the usual

Share this post


Link to post
Share on other sites

You should leave space for a PDS passthrough :evil:

The PDS in the SE was not spec'd for more than one card but that must be because there are no slot IDs and IRQs for multiple slots.

In the SE/30 at least an accelerator does not use up a seperate slot ID, so possibly we can get away with let's say an accelerator and a network card maybe?

Anyone ever tried that?

Share this post


Link to post
Share on other sites

That would definitely work as that's exactly what Radius did with VidCard connectors on their accelerators. Their Magic Bus Card is a passive PDS adapter for use when no Accelerator is available.

 

A much longer card would be necessary and we'd need to find the oddball connectors or harvest them from Magic Bus Cards. I've got one that's in the For Parts Only category.

 

The only NIC I've ever had was traded away, but was in the full size SE form factor which wouldn't fit up the SE/30 stovepipe. Any SE compliant PDS card of any sort could be converted for the use with a vertically mounted PerformerRedux in the late SE/SE/30 chassis. Replacing the standard  male "NuBus" type soldertail connector with a female wire wrap connector would do the trick. Soldertails would plug into the PDS and the accelerator would plug vertically into the hacked in EuroDin-96 Female wire wrap socket. Again, SE/30 type chassis required.

 

Haven't even looked at clearances for doing it any other way, baby steps for a change. :mellow:

Share this post


Link to post
Share on other sites

Pretty much the same way we (and I use that term very loosely) got ProtoCache1 up and running. joethezombie did the magical job of brute forcing the basic functions of his IIsi Adapter GAL. Then he and our resident wizard Bolle worked out equations to get that IIsi adaptation up and running. Bolle divined the function of the DiiMO adapter GAL, developed the equations (along with Joe?) and now that's up and running on Bolle's SE/30 DiiMO clone boards.

 

joe's on hiatus for a bit, real life tends to soak up development time. Hoping he'll play with my Performer's GALs upon his return. Bolle has a lot of applicable, open source information from the Amiga side of things and elsewhere for GAL equations needed to interface a 68030 accelerator with a 68000.

 

Me, I'm just playing at electron plumbing again, figured this was as good a time as any to get into the email order PCB game.

 

Did a rough pasteup at work just for you and Bolle. It would have to have a really screwed up PDS pinout though.

 

  PerformerNeu-L.jpg.48769d9a0ccb434f00da18029afd8fa2.jpg

 

Gals look smaller because they're surface mounted rather than socketed, needed the real estate. That's not a PDS passthru, but maybe it's enough pins to support something the likes of the Daughtercard on Bolle's Quesse/Gemini accelerator.

 

Edited by Trash80toHP_Mini

Share this post


Link to post
Share on other sites

I'm going to start pulling together information scattered hither, thither and yon.

 

Bolle: PowerBook 100 Accelerator Insanity thread:

Quote

Those equations usually are pretty basic. There existed a DIY project from german c't magazine that was about building a 030 accelerator for 68000 machines. They had GAL sets for the Amiga as well as the Mac. All the documentation on those is found on the web somewhere. Pretty sure if you compare the GAL connections on their board and on most of the commercial accelerators you will see that they all have a lot in common.

Your Micromac should always work in 030 mode even without the INIT - at least all my 030 accelerators for the SE and Classic do. Disabling the 68000 is done the same way as the onboard 030 is disabled on 030 PDS/Cache upgrades. Pretty sure you won't need a 68000 at all for the upgrades to work.

The logic can be done in a modern PLD. Just look up the TerribleFire. That thing actually should not be too hard to be modified to work in a Mac - minus the onboard DRAM which is software activated on the Amiga I think. SRAM could be implemented with some cache logic instead to speed things up. How cache is hooked up to the 030 bus is also well documented on the net somewhere, including needed GAL code and everything... I have it backed up somewhere in my 030 accelerator folder that is hiding somewhere.

 

Bunsen came up with some great info in a post that seems difficult to quote.

Link to a description of the C't accelerator he provided. THX, B!

 

Edited by Trash80toHP_Mini

Share this post


Link to post
Share on other sites
18 hours ago, Trash80toHP_Mini said:

That would definitely work as that's exactly what Radius did with VidCard connectors on their accelerators.

 

Should I take your word on that? :evil:

Prepared something but are a little bit scared to actually plug it in :p

IMG_3864.JPG

Share this post


Link to post
Share on other sites

edit: UNPLUG THAT MEMORY BOARD FIRST!!!!!!!

hOW THE HECK DID YOU GET THAT CENTER ROW SOLDERED? pASTE/hOT aIR?????

edit3: (*&#$ cAPSlOCK! Tricky tricky: double row then single row!

 

I can buzz the connections on the Magic Bus Card if you'd like. [}:)]

_______________________________________________________________________________

 

Flights of fancy and a 10cm x 10cm PCB to design after I tear my Performer down to the bone. [:D]

 

PerformerNeuRev2.thumb.jpg.6c8b36f1fe038d91450d518e6e7340a9.jpg

 

How many pins are on your Memory Card interface?

 

edit: forgot to switch back to sockets for the GALs in this layout. Also might drop the FPU for a 68030 PGA Socket for the prototyping stage. SMT GAL sockets free up a lot of acreage on the solder side and should fit on the component side with little problem. Dunno, images are still spinning around in the noggin. Condensing the prototype down to 10cm x 10cm Accelerator/VMem inclusive board for the SE PDS sans FPU is probably the sane thing to do. Implementing the FPU subset of the PGA CPU socket down to an FPU socket after the SMT CPU config is finalized will be child's play.

 

Edited by Trash80toHP_Mini
the usual

Share this post


Link to post
Share on other sites

This would have been too easy...

At least it boots but as soon as you try to access the NIC it just freezes. Doesn't matter if I have the RAM plugged in or not.

Also all combinations of accelerator driver, CompactVirtual and Asante drivers do not make a difference.

 

The accelerator has another port which can be used by a graphics card I think. This should be useable as a PDS passthrough as well.

Time to buzz the connections at that connector out as well. Maybe some slight bending of some signals is needed to pull off the additional PDS card thing.

 

CompactVirtual will let you set an address range for a custom PDS card, so it can do its magic and move around device addresses while not moving things into the specified address range of the PDS card.

Share this post


Link to post
Share on other sites

That Accelerator's way too complicated! KISS! I'll bet an SE PDS expansion card will work without a hiccup on the Performer. We'll see when I get to the point of testing.

 

1 - confirm my board's functional in the Drexel Plus/Killy Klip

2 - install Killy Klip/Performer on SE CPU - PDS is unblocked in an SE/Killy Klip Config

------- if that works I can test an SE PDS card if I can find one. :/

3 - desolder SE's CPU

4 - install machine pin socket strips in CPU's thruholes

5 - test SE with 68000 in machine pin socket
6 - remove 68000

7 - install tall machine pin header strips into CPU socket

8 - install Performer on machine pins sans 68000

9 - sacrifice lamb on the alter of the digital gods - then flip power switch

 

Hopefully celebrate over lambchops at dinner. [}:)]

 

Edited by Trash80toHP_Mini

Share this post


Link to post
Share on other sites

Just checked SE vs. SE/30 PDS positioning. Looks like I'll need to scootch the 96 pin connector back from the front edge of the card a bit for the PCB to fit up through the SE/30 chassis opening. But it'll fit, no biggie!

Share this post


Link to post
Share on other sites
20 hours ago, Trash80toHP_Mini said:

Pretty much the same way we (and I use that term very loosely) got ProtoCache1 up and running. joethezombie did the magical job of brute forcing the basic functions of his IIsi Adapter GAL. Then he and our resident wizard Bolle worked out equations to get that IIsi adaptation up and running. Bolle divined the function of the DiiMO adapter GAL, developed the equations (along with Joe?) and now that's up and running on Bolle's SE/30 DiiMO clone boards.

 

joe's on hiatus for a bit, real life tends to soak up development time. Hoping he'll play with my Performer's GALs upon his return. Bolle has a lot of applicable, open source information from the Amiga side of things and elsewhere for GAL equations needed to interface a 68030 accelerator with a 68000.

 

 

 

 

Ohh, I didn't know any of that ever took place. I never bother looking at alot of those threads because 99% of the time it turns out to be a bunch of discussion and nothing ever happens or nobody ever does anything (me included alot of times). 

 

Thats the one big wammy. another big wammy is to me "protocache" is deceiving because I dont know what it is, or own anything related to it. But eh..

 

I digress...

Share this post


Link to post
Share on other sites

ProtoCache1 was just a code name I picked out of a hat for the wire wrap prototype. joe liked the Rogue One tie in so we stuck with it. Unfortunately, most everything we discussed was in a series of massive email threads, not out in the open forums the way I would prefer it. Bolle jumped into the project with both feet and despite my tangential meanderings, I think we made a good team effort of it.

 

I'd like to keep this discussion out in the open, lots of folks have something to contribute. I've been leaving a trail Portable crumbs along the thread to see if you'd jump into the game. [:D]

 

Edited by Trash80toHP_Mini

Share this post


Link to post
Share on other sites

@Bolle

Have you begun buzzing that board?

Does it identify itself as the Gemini Accelerator? Probably as they named the VMem board Mercury.

The 76(?) pin board interconnect would be what I'm eager to explore.

Were you already thinking about using a 72pin SIMM?

 

My next AI session will be to break the VMem board and Performer out of the unified version onto two 10x10 PCB representations with that interconnect. A block diagram showing the components as is on Performer and Mercury would be one version.

 

A large scale visual thinker's conceptual layout will place the tandem pairs of Mercury SIMMs in a standard stacked memory bank configuration. I'll append a stretched version of my SIMMspender 30pin SIMM bank to 72pin SIMM converter to that so that three pages aligned will make visualization a snap. I know you like to see things arranged in much the same manner, do you have any suggestions?

 

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×