2/66Mhz vs 33Mhz

[Moogle!]

I was playing with my big discrete logic board 486 again, and discovered that an unmarked jumper selects how the oscillator crystal is used. The normally fixed position that the board was sold with runs a 66Mhz oscillator which is divided somewhere and appears as 33Mhz at pin C3 on the CPU socket. The other position, (I had to desolder the the solid wire jumper and solder pin headers to the motherboard so I could use the other position), sends the full crystal frequency to C3. As far as I can tell, there is no difference at all between 2/66 and full 33, so why was the board designed with the clock divider circuitry in the first place? There is no further clock division anywhere else as far as I am aware, and the ISA bus has its own oscillator, a socketed 16Mhz and a soldered down 14.318 Mhz.

 

[mcs_5]

The divider might provide a more stable duty cycle

 

[GiGaBiTe]

It could be a recycled 386 motherboard design slightly modified to support 486 CPUs. Most 386 motherboards had an oscillator that ran twice the host CPU clock.

 

[Moogle!]

The divider might provide a more stable duty cycle

How so?

It could be a recycled 386 motherboard design slightly modified to support 486 CPUs. Most 386 motherboards had an oscillator that ran twice the host CPU clock.

I have been wanting to see if the BIOS from my late model 386 board (another discrete logic design, sadly defunct) would operate it, but I don't know how to join the BIOS files.

 

[mcs_5]

How so?

The guaranteed duty cycle of the oscillator cans can be as bad as 40-60%. A divider that flips on the falling or rising edges will get you (close to) 50% exactly.