On Thu, 3 May 2001, Ruud Baltissen wrote: > Hallo Nate, > > > Use bus transceivers. These are bi-directional buffers of a sort, ... > > You probably mean the 74LS245 or equivalent. Let's assume we use one in the > databus. First question, when should it be enabled? One possible answer is: > whenever IO1, IO2, ROML and ROMH are active Low. But then what about the > REU? If the transciever set is the first thing out of the cart port, the DMA line can be used to reverse the direction of the address transcievers. Problem fixed ;) > I have thought about it many times but I never realy needed it. An other > important point: I had no friends to discuss it with. Now I have :) > Let's do some brainstorming. I assume the critical point is buffering the > data- and addressbusses and R/W. Buffering of incoming signals is not needed > IMHO. Buffering of outgoing signals can be realized by a 541. The 244 can buffer.. Why are you using a latch? > We'll use 245's to buffer the addressbus. Output Enable (OE) always (L). > Direction (DIR) depends on wether DMA is activated or not. RIght. and on the data bus, DIR is connected to R/W, and inverted by DMA. Hm. this could be a challenge. > We'll use 245's to buffer the databus as well. Pin 11-18 to the C64, pin 2-9 > to the external world. > OE should be (L) whenever: > - IO1, IO2, ROML and ROMH are (L) > - 6510 performs a write > - DMA is activated > > DIR should be (H), data external -> C64, whenever: > - 6510 performs READ > - when during DMA external device performs WRITE (REU, C64-CP/M-cartridge) > > Direction of R/W depends on DMA or not. Two 125 buffers will do I think. However, if a cart-port device needs to DMA from/to the I/O region (i.e. the cart slot), the data bus needs to be disconnected. So, the game and I/O lines must disable the transciever so it doesnt step on the data. (this is the challenge stated above) Is this a working solution? > It seems I covered everything, but I already know one situation where the > interface would not work: what if somebody wants to read from hardware NOT > using the I/O-lines but decoded the $DE00-$DFFF area itself? It is just a > question as IMHO it would be a very unwise thing to do because of > buscollisions when reading from the RAM under the I/O. Unless you also tweaked the cart port for a single cycle, triggering ultimax and disabling the ram/rom underneath ;) > Feel free to shoot on it. I also would appreciate it if you mentioned the > fact you found nothing. - This message was sent through the cbm-hackers mailing list. To unsubscribe: echo unsubscribe | mail cbm-hackers-request@dot.tml.hut.fi.
Archive generated by hypermail 2.1.1.