Recent content by Melkhior

Did you find a viable option? At 5V, 16 Mbit SRAM are expensive and with big MOQ, while 4 Mbit are more viable but you'd need 32 of them - that's a large PCB, and I'm not sure what the signal integrity would look like with all those extra chips/traces. I've thought about it as well, but couldn't...

Yes, tag SRAM (which include comparators so they can do both storing and checking of tags) haven't been made in years, are expensive if you can find them, and won't help making bigger/better caches. Using a programmable device (i.e. FPGA) with embedded SRAM for tags (and external discrete SRAM...

An information that isn't very useful because of the cost, but that delights me that it simply exists, you can still get them from the continuation manufacturer: MC68882EI25AR (they even have 50 MHz parts in PGA if you're willing to spend >$120 on each...).

The best way is to use the onboard socket, but beware that a lot of '882 for sales are remarked and not necessarily what they pretend to be :-( For a stock 25 MHz LCIII, most PLCC '882 should be fine, but there's always a risk if the source is not reputable (either lower frequency '882 remarked...

ACT would probably work as an alternative for ALS, though they have different driver properties. @Addicted : A supply issue you might have had on the 74ALS640 seem to have been transient, as I see significant stock at MOQ=1 at Mouser. ALS are generally fairly available.

If you want authentic vintage, I don't know; if you want the functionality, someone did a redesign some years ago. So you can build your own. Original discussion thread here.

As far as I know, all XC68LC040 have the errata and none can support a FPU emulator to run regular 68k FPU code. You want to use either a full '040 (to get the hardware FPU, best option), or a newer MC68LC040 that can support FPU emulation (why do that if you can get a full '040 instead, I don't...

Common PC ones are well documented (the wikipedia links has some of them, though I can't vouch for reliability). For others, usually you need to dig out the manufacturer's documentation - bitsavers and similar are valuable resources, but searching for the appropriate document is usually tough...

DE-9 (being a bit pedantic here sorry) was common for video connectors at the time, including Sun's ECL monochrome devices and early PC standards (CGA, EGA, as you mentioned early VGA also used DE-9 rather than the since-ubiquitous DE-15). They may have wanted to share electronics and cabling...

On MC68000 (i.e. the SE), MC68030 (other than the IIfx, so SE/30 and IIsi) and Quadra, very direct. It exposes most of the CPU bus directly. The IIfx has some buffering to do weird things with the CPU clock, as the CPU run at 40 MHz but the PDS at only 20 MHz. The LC PDS is missing some...

Does the NuBusFPGA count as a "real-word NuBus card" ? :-) (and coming 30+ years after the fact, I don't think it counts, but honestly having that much space to play with is convenient for modern HW designer!). Anyway, there's (at least) two way of seeing the large size of the superslot (256...

Using a PDS slot instead of the 68k original socket doesn't make of a difference, maybe a bit more signal integrity issue but at the 68000 speed not really an issue. The PiStorm was designed for the Amiga, hence some of the choices. Using the approach to make a peripheral instead of a CPU is...

That's the idea behind the PiStorm and similar. The main issue is that without reconfigurable hardware, the bus must be bit-banged, and at the speed of a 68000 it's not trivial, and for faster things it becomes extremely difficult. Unlike some bit-banged protocols that can live with the...

@eharmon Thanks for sharing your experience!

Bit-banging NuBus might be possible, but stuff from that era are already somewhat fast and sensitive to timings. Using dedicated hardware through a FPGA or CPLD is likely to be a lot easier and much more reliable. One in a billion cycles with the wrong timings is already a crashing error every...