PFM, SCSI, TIPI and the Geneve

[mizapf]

That is, you'd check the ID and decide for a particular protocol to work with the chip? In that case, I would have to implement this as a variant of the existing emulation. Hopefully we don't have too many compatible chips with different IDs.

[atrax27407]

I just finished an "all nighter" trying to find another equivalent for the AT29C040 and AT29C040A. There is NO functional equivalent among multiple manufacturers for the AT29C040. The ONLY equivalent for the AT29C040A is the Winbond W29C040 and it is exact. So, unless you check for the manufacturer/product ID in the programming algorithm, you could most likely get by by eliminating it completely. Just use either the W29C040 or the AT29C040A. A single programming algorithm would work for both. 

[+InsaneMultitasker]

Can't completely eliminate the original 29c040 as it is installed in a number of Geneves

 

The two equivalent chips (29c040a and Winbond 29c040) are also usable in the HSGPL and perhaps other cards, right @atrax27407

[atrax27407]

You could probably use the PLCC version of the 29C040 with an adapter to replace those already installed in a Geneve. The Winbond could be used in place of an AT29C040A PDIP in most applications. 

Edited December 11, 2020 by atrax27407

[+FarmerPotato]

1 hour ago, atrax27407 said:

I just finished an "all nighter" trying to find another equivalent for the AT29C040 and AT29C040A. There is NO functional equivalent among multiple manufacturers for the AT29C040. The ONLY equivalent for the AT29C040A is the Winbond W29C040 and it is exact. So, unless you check for the manufacturer/product ID in the programming algorithm, you could most likely get by by eliminating it completely. Just use either the W29C040 or the AT29C040A. A single programming algorithm would work for both. 

I'd like to hear what you found. Sector size, Are the programming commands different? What differences are there in the super common chips?

 

Are there schematics for the PFM? If it is just the Flash chip, there's that adaptor for the TL866/Xgpro. Also, I could easily make a PLCC adaptor board. I made one for a breadboard, using through-hole PLCC socket. 

 

[mizapf]

PFM also makes use of two CRU bits: https://www.ninerpedia.org/wiki/Programmble_Flash_Memory_Expansion

[+InsaneMultitasker]

1 hour ago, FarmerPotato said:

I'd like to hear what you found. Sector size, Are the programming commands different? What differences are there in the super common chips?

 

Are there schematics for the PFM? If it is just the Flash chip, there's that adaptor for the TL866/Xgpro. Also, I could easily make a PLCC adaptor board. I made one for a breadboard, using through-hole PLCC socket. 

 

If I recall correctly, I posted a scanned image of my very rough and crummy hand-drawn PFM512 "schematic".  At Cecure we did not write down certain aspects of hardware upgrades because of a few unscrupulous people and at least one who absconded with information that was not theirs. Richard is the only other person with whom I shared the specifics back in the late 90s/early 2000s, and I know he has successfully installed the PFM512.

 

At a high-level the PFM512 is typically installed as follows:

1. Remove the existing EPROM

2. Inspect the EPROM 28-pin socket.  If single-wipe or suspect, remove the old socket and install a new 28-pin socket

3. Prepare a 32-pin socket by soldering wires to pins 30,31,32,1,2 for connection elsewhere on the board; there may also be one other pin (CE?) to account for.

4. Insert the 32-pin socket into the 28-pin socket and tie-wrap it in place

5. solder the wires to the appropriate places for power, CRU, etc

6. Modify jumpers and add resistors where necessary to ensure proper power-up bank is selected

 

If the PFM512 is used with the 512K SRAM version of the 384K upgrade (see Fabrice's site for current instructions) there may be another jumper or two to change. 

 

The PFM+ is a similar animal but it required an extra logic chip and was more complicated, especially when coupled with the 384K memory upgrade.

 

 

 

 

[BeeryMiller]

Going to throw a question out there.

 

Is there anyone with a Geneve with only floppy storage capability, that would use (and buy if they do not already have it), a TIPI to load MDOS from the TIPI?  At this point, just trying to ascertain interest. 

 

I am wading through some Geneve Eprom code and some LOAD/SYS code to try and be able to boot the Geneve from the TIPI.  Neither the 0.98 Eprom nor the 1.00 eprom will permit booting from the TIPI, so if workable code can be written, it will require a Geneve boot eprom upgrade.  

 

As separate pieces, I already have eprom code loading load/sys from the TIPI on bootup, and separately have LOAD/SYS loading MDOS from an Editor/Assembler file.  Working on the tweaks so that both pieces work together which is a bit more complex due to the way memory is mapped and used during the bootup cycle.

 

Bootup times on the TIPI will be slower (~20 seconds) versus ramdisk, pfm, scsi, or pfm options.  It will also be slower than a Jumpboot floppy disk.

 

Once booted, the work Tim and I have been doing (mostly Tim's work), will give the TIPI file storage capability just like any other hard drive storage device.  That part does not require any Geneve eprom upgrade.  It's just to boot from TIPI would necessitate an eprom upgrade as neither 0.98 or 1.00 eprom anticipated the TIPI.  

 

Beery

[Nick99]

I am for the use of TIPI to load MDOS, I already have a TIPI for the Geneve and it would be great to have another way to load MDOS than just with a floppy disk.

do I need to do some more to the computer than change the eprom to get this to work?

I also have a IDE card, but I guess that will not be able to load MDOS?

 

You have my interest.

 

Niklas

[BeeryMiller]

45 minutes ago, Nick99 said:

I am for the use of TIPI to load MDOS, I already have a TIPI for the Geneve and it would be great to have another way to load MDOS than just with a floppy disk.

do I need to do some more to the computer than change the eprom to get this to work?

I also have a IDE card, but I guess that will not be able to load MDOS?

 

You have my interest.

 

Niklas

I don't have an IDE card, and I don't know if there was ever a LOAD/SYS written/modified to load from that device.  I've got a thought it may not be too difficult to do, but not having the original LOAD/SYS code for the Myarc HFDC, I can't say if a byte edit of the eprom code would allow a boot from the IDE.

 

At this point, unless other code would be forthcoming (and their is that possibility) that is workable, any eprom update would be floppy, HRD, or TIPI only and not IDE, nor SCSI.  MFM possibly.

 

Beery

 

[+InsaneMultitasker]

28 minutes ago, BeeryMiller said:

I don't have an IDE card, and I don't know if there was ever a LOAD/SYS written/modified to load from that device

Yes, I updated the SCSI load/sys last year and created a functioning IDE version based on the same code.    @Shift838 loaned me an IDE card this past weekend and we successfully tested IDE support (available via SCSMAP unit 7).  It's mostly just a matter of adding code to the EPROM to execute a LOAD (opcode >05) request for "IDE1.IDE_SYS" or comparable, and recovering from the error if not present.  Similar situation to the TIPI loader requirement.   [The same can be done for SCSI, TIPI, HDX, etc with appropriate loaders].  

 

I'lll gather what I have this weekend and send it your way for experimentation, along with the HFDC LOAD/SYS code reconciled to the last known good version.

 

Edit: when/if a new EPROM is released, at that time it would be appropriate to finally eliminate the obnoxious '/' character from the system files.  Just sayin'