68k Macintosh Liberation Army Forums
68k Macintosh Liberation Army Forums
Home | Members | Search | FAQ
 All Forums
 Lounge
 2300c - SRAM Card VM/Boot Disk Hack?
Author Topic  
Trash80toG-4
NIGHT STALKER


USA
2899 Posts
Posted - 17 Dec 2003 :  13:43:27
This is the continuation of my earlier "you guys are FREAKIN' KILLING ME!!!" response *thinks: wail of despair! * to a couple of you *s* comrades discussing PCMCIA boot disks and the benefits of using of SRAM cards as opposed to Flash Cards, you got me . . . . . . whatever . . . a solid state based block of Virtual Memory or Boot Disk could really speed things up on a 2300c. Here are some of the other inspirations for this current episode of DuoDementia

http://www.cosc.canterbury.ac.nz/~greg/hardware/pb150/

http://www.wau.nl/hemeltje/temporary/personal/miscelaneous/cfadapter.html

http://www.aemmenet.ch/Publikationen/CFtoIDEpinout.html

While poring over piles of PDFs about the NuBusPowerPBs and their IDE blessed 68k kin, I discovered several real gems. Apple's documentation of the Baboon ASIC's capabilities appear to be uniformly cryptic across the line. The 2300c DevNotes state:

quote:

deviceID

A short word that uniquely identifies an ATA device. The field
consists of the following structure:

struct deviceIdentification
{
UInt16 Reserved; // the upper word is reserved
UInt16 deviceNum; // consists of device ID and bus ID
};
typedef struct deviceIdentification
deviceIdentification;

Bit 15 of the deviceNum field indicates master (=0) /slave (=1)
selection. Bits 14 through 0 contain the bus ID (for example,
$0 = master unit of bus 0, $80 = slave unit of bus 0). The present
implementation allows only one device in the master configuration.
This value is always 0.


. . . so it seems likely a slave device "could" be supported by the Baboon ASIC, as it was in the development roadmap leading to the 1400.

AFAIK, Baboon supports a second IDE hard drive in the expansion bay of the 5300 and190, which were introduced in the same rollout as the 2300. Baboon's multiple device ATA support is a given in the case of the 1400c,  as detailed in the DevNote, the internal CD is an ATAPI/IDE device and the ATA documentation states:

quote:

Expansion Bay

Sequence of Control Signals

Specific signals to the Baboon IC and the Power Manager IC allow the computer to
detect the insertion of a module into the expansion bay and take appropriate action. For
example, when a module with an IDE device is inserted, the computer performs the
following sequence of events:

1. When a module is inserted, the /DEV_IN signal goes low, causing the Baboon IC to
generate an interrupt.

2. The Power Manager IC reads the three DEV_ID signals, which identify the device as
an IDE device.

3. System software responds to the interrupt and sets a signal that turns on the power to
the expansion bay.

4. When the expansion bay power goes high, the Baboon IC generates another interrupt.

5. System software responds to the power-on interrupt and asserts a signal to enable the
IDE bus in the expansion bay.

6. The software then releases the /MB_IDE_RST signal from the Power Manager IC,
allowing the IDE device to begin operating.

Essentially the reverse sequence occurs when a module is removed from the expansion
bay:

1. When the module is removed, the /DEV_IN signal goes high causing the Baboon IC
to generate an interrupt and set /MB_OE high, disabling the IDE bus.

2. System software responds to the interrupt by reading the device ID settings in the
Power Manager IC, setting the /MB_PWR_EN signal high to turn off the power to the
expansion bay, and asserting the /MB_IDE_RST signal to disable the IDE drive.


Hopefully, the Power manager, Device Manager, Slot Manager . . . whatever . . . will handle the updated capabilities of the 1400 in the 2300. It might only be a matter of using something like "wish I were" to install the required components of the proper O.S. level in order to enable these routines in the 2300.

In a worst case scenario, 2300 and 1400 DevNotes make it clear that their architecture, chipsets, ROMs and system software are nearly identical. A 2300 might require a Baboon/ROM transplant from a 1400. MadDog has demonstrated by transplanting the ECSC LCD controller used in the 1400 to replace the older CSC of a Duo 270c, that this might not be impossible at the chipset level if drastic measures are required.

In a best case scenario, one could cobble up a dual device cable adapter for the ATA bus of a PPC/NuBus/PB from some goodies from this page . . .

http://www.cablesonline.com/44pinlapinri.html

. . . . set the jumpers of an IDE HDD to slave, hook it onto the ATA ribbon cable along with the internal drive and power the sucker up!

Riiiiiight!!!!!!!  :rolleyes:

jt  

jt .
Trash Hauler: call sign: eight-ball
C.O. AC-130H SpecOps 68kMLAAF

Edited by - Trash80toG-4 on 17 Dec 2003 13:54:52

Trash80toG-4
NIGHT STALKER


USA
2899 Posts
Posted - 17 Dec 2003 :  16:13:08
On a somewhat more sane note . . .

New toys are on the way!

. . . one of these:

http://secure.oswaltsystems.net/gosurplus/10Expand.asp?ProductCode=PCMCIA.MCDISK

I'm hoping the mechanism was OEM'd by MPL, or is an earlier version of . . .

http://www.mpl.ch/fr2331.html

. . . unless of course it's not, but it's better at handling I/O cards than MPL says . . .

http://www.mpl.ch/fr2351.html

. . . because we ALL know that the Mac handles Ethernet over SCSI just fine . . .

*crosses fingers and sacrifices a goat to the SCSI gods with visions of wireless card support *

. . . and one of these . . .

http://secure.oswaltsystems.net/gosurplus/10Expand.asp?ProductCode=PCMCIA-DRIVE

. . . which, with any luck at al,l may turn out to be an IDE interface card using a SCSI ribbon cable to connect to the card cage and provide power and ground just like a removable drive bay . . .

jt .
Trash Hauler: call sign: eight-ball
C.O. AC-130H SpecOps 68kMLAAF

Edited by - Trash80toG-4 on 17 Dec 2003 16:43:56Go to Top of Page

wisof
Junior Member


USA
257 Posts
Posted - 17 Dec 2003 :  20:39:50
I am trying to wrap my head around the many possibilities of those pcmcia card readers. could be interesting cause there are some tight pcmcia cards. . .too many options. . .can't compute. . .the price doesn't seem alltogether too bad either. . .I am broke, but considering the possibilities. . .Go to Top of Page
   

68k Macintosh Liberation Army Forums

© 2001-2003 68kMLA

Go To Top Of Page

68k of the Week: kastegir's PowerBook 180.