PDA

View Full Version : Turbo buttons and mobo headers



offensive_Jerk
September 29th, 2009, 03:54 PM
This is regarding the computer I got off craigslist here: http://www.vintage-computer.com/vcforum/showthread.php?t=17459

I took the mobo out of the case, so I could see what's going on to configure it for a 486DX2.

Put the board back in and had an issue with the turbo button switch.
The header on the board is two pin, and the connector itself is 3 pin. The middle wire is black.

Couple of questions:

1. How does that hook up? Is a 3 pin header required?

2. Should the clock speed change on post if the button is pressed? It's hard to say if the buttons works or not. I don't know what the button actually does to the hardware, if it really does slow the clock or what.

Maybe someone with more turbo experience could chime in. As I said before, I'm a n00b with these vintage motherboard jumpers.

Unknown_K
September 29th, 2009, 05:42 PM
I can't tell in your picture (linked to thread) if you have a LED for speed on that case or not. I have a mini tower that has a turbo button and a LED display that goes from one speed to another using that 3rd line (normally the 2 line version just opens and closes a circuit that power up a case LED light).

offensive_Jerk
September 29th, 2009, 06:20 PM
Yes there is a LED display that goes from 66 to 33 when the button is pushed.

kishy
September 29th, 2009, 06:33 PM
The intention behind the turbo switch took a rather dramatic turn at some point.

It was originally to speed the system up...by the time the 486 (and I think even 386) were around, it was used to slow the system down for compatibility with older CPU-speed-dependent programs.

Though I may just be spitting out info that is already known, I figured it a valuable piece of input either way.

As for how it works, I'm not 100% sure. I believe on a properly "turbo" supporting system, it just slows down the bus speed (thereby slowing down CPU, expansion cards, memory, everything). This may not be (probably isn't) reflected on the POST screen but would probably be dependent on a specific motherboard/BIOS/chipset perhaps to show it.

southbird
September 29th, 2009, 08:04 PM
This may not be (probably isn't) reflected on the POST screen but would probably be dependent on a specific motherboard/BIOS/chipset perhaps to show it.

I've never seen it displayed on the POST per se, but I have done it DURING post and watched the RAM count slow to a crawl. :) Fun with the old (AMI/Award?) boards that would make that ticking sound which was audibly slower. Usually however there is an LED (or, in your case, a speed display, likely tied to what would be the LED pin.) Best find a CPU speed tester (I used to MAXSPEED.EXE by Maxim computers) and see how it is effected by the Turbo.

offensive_Jerk
September 29th, 2009, 08:11 PM
I can't tell in your picture (linked to thread) if you have a LED for speed on that case or not.

Oh, I totally misread what you wrote.

Yes, I belive there is a little LED that lights up when the turbo button is pressed, along with the LCD display changing from 66 to 33.

I'll have to take another look and see what's up. I believe there was a header for the Turbo switch and the light.

modem7
September 29th, 2009, 08:20 PM
1. How does that hook up? Is a 3 pin header required?
Apart from Unknown_K's suggestion (third wire activates a LED), there is another possibility:

My preference is to have the turbo function active when the turbo button is in (turbo button on). I'm sure that if I look through my collection of IBM clones, I'll find at least one where the turbo function is activated when the turbo button is out (off), opposite to what I like. And if I look, there will be nothing I can do about it; a 2-wire lead from the turbo switch down to a 2-pin header on the motherboard.

What you may find with your 3-wire lead from the turbo switch is that the center wire is the common contact, one of the other two wires is the normally-closed contact, and the final wire is the normally-open contact. This means that by changing which 2 of the 3 wires connects to the 2 pin header, one can change the behavior of the turbo switch to suit one's preference (either turbo = switch in, or turbo = switch out).

Chuck(G)
September 29th, 2009, 08:45 PM
What you may find with your 3-wire lead from the turbo switch is that the center wire is the common contact, one of the other two wires is the normally-closed contact, and the final wire is the normally-open contact. This means that by changing which 2 of the 3 wires connects to the 2 pin header, one can change the behavior of the turbo switch to suit one's preference (either turbo = switch in, or turbo = switch out).

Since I've got the same case, I can attest that you're right on the money. Simply pick two pins, one of which of necessity is the center one. One way, your turbo button will be "turbo" when up; the other way, when down.

Me? I set the display to say "On" and forget about the turbo switch. But you could also do "Hi" and "Lo" if you felt the turbo was worth preserving. Frequency displays are pretty much meaningless, given all of the other factors. Fact is, "turbo" mode was obsolete with the 386s, yet the silly tradition was continued well into the Pentium systems. Go figure.

It would be kind of cool to have an analog meter registering instructions/second x 10K, like the old GE mainframes (Was that meter on the 635 operator's console? It's been too long for my tired brain..)

"I want my system to run slower." Yeah, right.

offensive_Jerk
September 29th, 2009, 11:42 PM
That makes sense.
Now to see if it actually does anything.
Thanks guys.

offensive_Jerk
September 29th, 2009, 11:47 PM
Apparently it works.

Running Damn Small Linux live CD.

Conky is running on the desktop which displays the CPU usage.
When I hit the turbo button do deselect it, the CPU usage goes to 100percent and the screen has a hard time refreshing.. :p