home *** CD-ROM | disk | FTP | other *** search
- Path: misa.extern.fh-hannover.de!boose
- Date: 21 Apr 1996 19:05:00 +0200
- From: boose@misa.extern.fh-hannover.de (Andreas Boose)
- Newsgroups: comp.sys.cbm
- Message-ID: <67HBDKtwmrB@misa.extern.fh-hannover.de>
- References: <4ktud1$4jh@madeline.INS.CWRU.Edu> <umcwikla.829689480@toliman>
- Subject: Re: Adding a 6522 to my c64
- X-Newsreader: CrossPoint v3.11 R/A768
- Organization: Private Site, Lehrte, Germany
-
- umcwikla@cc.umanitoba.ca (Tom Cwikla) writes:
-
- >I don't think you need to do anything special
- >to add a 6522. I added another 6526 on my expansion board without any major
- >difficulties. All you really need is to attach the data, address and clock
- >lines from the C64 to the appropriate pins on the 6526. Then just add some
- >address decode logic (I used a 74LS154) and it should work just fine.
-
- Tom, the problem is, that the 6522 uses the old style 6502 timing. That
- means it expects the address to be valid at the raising edge of phi2. This
- is hardly true for C64 systems since just 20ns before phi2 raises, AEC is
- switched by the VIC-II. That are just lousy 20ns to switch from VIC-II to
- CPU addresses, which is too tight for real life systems. The 6526 'knows'
- this problem and it expects the address to be valid 100ns *after* phi2 is
- risen, giving the system 120ns to setup up the proper CPU addresses.
-
- This makes the 6526 very universal, it operates in both 6502 and 6510
- based systems without a problem. But connecting a 6522 to a C64 requires
- the raising edge of phi2 to be delayed by ~100ns. In many applications this
- is done with a simple 7474 flip-flop:
-
- 7474
- ________
- | |
- ,--|-RESET |
- | | |
- phi2 -------*--|DATA Q|------------- phi2@6522
- | |
- dot-clock ----------|CLOCK |
- | |
- |________|
-
-
- MfG Andreas
-
-