This is wonderful news @trag !! I know @ClassicHasClass and I are standing by. Please let us know if here is any other way we could help 
Apple Network Server MacOS based ROMs found
- Thread starter eric
- Start date
-
Board Nominations
Nominations have now closed and the results are available here. -
Hey Guest, MARCHintosh 2026 is upon us. Check out community projects, join GlobalTalk, and have fun!
I used kansas_deinterleave to deinterleave all the ROMs. I think most people just want ANS 2.0.
This four chip arrangement is not flashable by Apple's flash utility which expects different chips.
Also, your ROM DIMMs won't have the write enable pin wired so we can't make our own flash utility?
This four chip arrangement is not flashable by Apple's flash utility which expects different chips.
Also, your ROM DIMMs won't have the write enable pin wired so we can't make our own flash utility?
Last edited:
ANS 2.0 means the first one above, "ANS Deinterleaved.zip"?
That is correct. Actually, I made a provision to enable the WE pin to all of the chips from the WE pin(s) on the DIMM. It's been 20 years since I laid the thing out. IIRC, connecting them requires installing a couple of 0 ohm resisters and maybe a wire jumper. But certainly, it would require your own programming utility, which follows the programming requirements for the HY29F800 from Hyundai or Hynix (same procedure for both). There's some theoretical discussion about programming in the large PEx thread I linked above.
My thought was to build an external programmer similar to Doug's mechanism for the Mac II family.
This is probably the first posting in that speculation: https://68kmla.org/bb/threads/pex-rom-project.23568/post-417941
And here is a later post laying out some of the solid options. Never got any further as far as building/testing hardware: https://68kmla.org/bb/threads/pex-rom-project.23568/post-418272
Last edited:
Yes. That's the only one with 1 MiB files.
Any reason to not connect them?
Code for flashing various chips exists in New World Mac firmware updaters (Open Firmware code), and in open source utilities such as flashrom. I have a fork of flashrom for Mac OS X.
I'll read those later. I was thinking, could the ROM DIMM have a switch to disable itself so that I can boot my Power Mac 8600 into its built-in version of Open Firmware, load firmware updating code from disk or serial port, then use the switch to enable the DIMM (bypass the built-in ROM), then run the firmware updating code? That way, I could use the Power Mac 8600 as a programmer for itself or other Macs. It would be useful for ANS since ANS doesn't have built-in ROM.
Apple flashing schematics were linked at
https://68kmla.org/bb/threads/mac-os-on-ans-modding-a-rom-dimm.51120/post-578579
Apple's flashing utility (for classic Mac OS) could be updated to add support for other chips. However, I don't think you can boot Mac OS, then disable the built-in ROM while Mac OS is running unless you disable interrupts and exceptions. So in this case, the ROM DIMM must already have a ROM that works with the Mac.
Enabling the use of WE was almost an after thought in the lay out. The _WE pins are all hard wired high. I think I left clear traces between _WE and Vss, but they would have to be physically cut and then the line to the DIMM _WE pins connected with resistors and I think in one case a jumper. It's doable but inelegant.
Also, any machine in which they're use would need to hold the _WE pins in the DIMM socket high or the ROM would not function. It seemed safer to just wire them high.
With more time to massage the layout I could have arranged resistors to tie _WE lightly high, so the status of the DIMM socket wouldn't matter, unless it dragged _WE low, but that wasn't my goal at the time and that could be problematical, if the DIMM sockets are tying those pins high in some way. Without knowing what Apple was doing with the DIMM socket WE pins in general, connecting them represented too many unknowns.
I don't remember what the status of Vpp is on the DIMM, but I don't think the chips I used need Vpp at all, being Flash, rather than EEPROM.
A new module could be designed in that fashion. Just add some jumpers or DIP switch for the _CE pins to the chips and for the DIMM socket pin which disables the on-board ROM. If you're asking if there is a software way to do so, there are _CE pins on the DIMM module, but I'm not sure what they do on the DIMM socket, but there is no software method to control the pin that disables the on-board ROM. That would need to be a physical switch.
There's a pin in the DIMM module which leads back to the _CE pins of the on-board ROMs. Connect that DIMM module pin to Vss on the DIMM, and it supplies Vss into the DIMM socket, which connects back to the _CE pins of the on-board ROM, disabling them.
Apple almost certainly had some kind of stand alone flashing utility they could use. But once the modules were in development machines it was more convenient to not need to pull the module and replace it repeatedly.
Cameron,
I just found this. I still ready and willing to do whatever I can to help just let me know m e know.
I also have a rough plan to get the Deep Dish to you. I am waiting on a restoration shop in southern California
to get around to working on my vantage VW Karmann Ghia. I will have to trailer it so I can bring the ANS to you
during that trip which I hope will be in May or June.
adios,
jt
I don't know how I never noticed this thread until now. Visiting here too seldom.
I can make copies of those ROMs, but will need some help processing the dumps to get them ready. I still have thirteen blank ROM circuit boards and hundreds of flash chips.
Process the dumps. Program them onto chips, solder the chips to boards. Done. Probably only take me 8 or 10 years based on past experience.
Seriously, I don't have a lot of attention to spare, but can probably get it done, if someone else will take the lead on getting the files to be programmed massaged as needed.
The ROM module in the X100, the X500 and X600 Macs, the ANS, the Beige G3 and the PowerExpress are all the same physically, although the Beige G3 is wired for 3V supply, instead of 5V supply.
Here's the saga of making the ROMs for the PowerExpress:
PowerExpress (9700) ROM Thread
Which also covers in some detail the process of going from 4 MB dump to four files suitable for programming on X16 flash chips.
Not sure if I referenced any files in my old webspace in there. My ISP changed since then, so if I did, referenced files would be here:
https://sphinxgroup.org/Firmware/
If more than thirteen modules are needed, I still have the board design and JLCPCB was pretty cheap to produce the previous 20 I made. In fact, I think the gerber files are in the folder referenced above.
I've also used this design to make Rev. B and Rev. C ROM modules for the Beige G3 (adds 2 drive support on ATA busses), and Kansas ROM modules for the X500/X600, which doesn't really do much, but in theory should make speculative processing work properly with G3 upgrades on those machines.
Is there a story behind how John found the ROMs? I didn't read the thread in as much detail as I should, but I gathered he found two versions and that Classichasclass found another?
In case anyone doesn't ahve it I've attached the Apple Hardware Developer Note for the ANS. I'm much more a hardware guy than a software guy.
The main differences between the ANS and the PowerMac 9500/9600 is that the ANS has ten PCI devices where the PM9500 only has 7 (six slots + Grand Central). The ANS adds the 53C825 SCSI chips and the onboard video chip. Each PCI device gets a unique interrupt (routed through Grand Central) in the Apple universe, so the ten interrupts supported by Grand Central are arranged very differently on the ANS than on the PM9500, which is at least one reason the ROMs for one do not work on the other.
Report
trag ,
I was on the Education business development team for the Apple Network Servers. I had an ANS 700/200 that was used to demo to the Higher Education customers in the western states. I had dealings with the engineering and product marketing teams in Cupertino. I knew that there were efforts to support additional operating systems on the ANS servers, Windows NT and Classic MacOS. I was successful in obtaining a ROM that allowed MacOS 8 & 9 to boot on the ANS but could not get the NT ROM. My guess was that Apple was having issues with MicroSoft over the costs of development, MicroSoft wanted Apple to fund the entire project.
So this is how I ended up with a ROM that would boot Classic MacOS.
adios,
jt
Excellent! I will send you a PM. There is a nice newly recapped logic board and 150MHz CPU card here ready to go in it.