VOGONS


First post, by DNSDies

User metadata
Rank Member
Rank
Member

Just a weird idea I had, but would it be possible to put an interposer board on a CPU socket with 2 additional CPU sockets, and be able to have, say, a switch that prevents vcc to one of the CPUs so you could pick which CPU you're using before powering on the computer?
Like, say I wanted a Super Socket 7 machine with a Pentium 100mhz AND an AMD K6-III 450mhz, and with the push of a button, I could swap CPUs without opening the case.
Or say I had a motherboard that supported a 386 and 486 CPU socket, and wanted to have a way to isolate the 486 CPU via an interposer PCB and a switch that would physically disable the multiple VCC pins.

Would this work, or would the address/data lines and various other pins need to be separated as well?
Because I don't think it's possible to find a 200-300 input switch. But a 4-6 input switch? That's much easier.
This is a dumb idea.

Reply 1 of 3, by mkarcher

User metadata
Rank l33t
Rank
l33t
DNSDies wrote on 2021-01-26, 18:41:
Just a weird idea I had, but would it be possible to put an interposer board on a CPU socket with 2 additional CPU sockets, and […]
Show full quote

Just a weird idea I had, but would it be possible to put an interposer board on a CPU socket with 2 additional CPU sockets, and be able to have, say, a switch that prevents vcc to one of the CPUs so you could pick which CPU you're using before powering on the computer?
Like, say I wanted a Super Socket 7 machine with a Pentium 100mhz AND an AMD K6-III 450mhz, and with the push of a button, I could swap CPUs without opening the case.
Or say I had a motherboard that supported a 386 and 486 CPU socket, and wanted to have a way to isolate the 486 CPU via an interposer PCB and a switch that would physically disable the multiple VCC pins.

Would this work, or would the address/data lines and various other pins need to be separated as well?
Because I don't think it's possible to find a 200-300 input switch. But a 4-6 input switch? That's much easier.
This is a dumb idea.

Don't try to take a unused CPU off the bus by just disabling the power pins. It won't work, as the ESD protection diodes from the data pins to VCC will cause excess leakage and power the chip. Instead, you need to correctly power both CPUs, but have the unused CPU in bus hold or float mode where the CPU does not drive bus pins.

Reply 2 of 3, by adalbert

User metadata
Rank Oldbie
Rank
Oldbie

Doing that with bunch of clicky relays would be funny. And probably wouldn't work, but whatever 😁

This is just a joke, but I wonder if some relatively cheap digital switching IC could be used. 386s and 486s are not too fast, so maybe? That is, if we would like to really isolate these CPUs. If full isolation is not required, then it certainly would be easier.

Attachments

  • cpuswitch2lol.png
    Filename
    cpuswitch2lol.png
    File size
    385.85 KiB
    Views
    202 views
    File license
    Fair use/fair dealing exception

Repair/electronic stuff videos: https://www.youtube.com/c/adalbertfix
ISA Wi-fi + USB in T3200SXC: https://www.youtube.com/watch?v=WX30t3lYezs
GUI programming for Windows 3.11 (the easy way): https://www.youtube.com/watch?v=d6L272OApVg

Reply 3 of 3, by DNSDies

User metadata
Rank Member
Rank
Member

So, drive the BOFF pin and HOLD and AHOLD to their respective active voltage levels via a physical jumper or switch, and that might do the trick, huh?
Might be neat to design such a thing. It's a niche market, but it could also be used to provide voltage regulation to normal Socket 7 motherboards to use newer K6s, or allow a standard 486 socket take lower voltage DX models.
I'd pay $150~ for something like that.

adalbert wrote on 2021-01-26, 21:07:

Doing that with bunch of clicky relays would be funny. And probably wouldn't work, but whatever 😁

imagine the noise though.