Path: utzoo!utgpu!news-server.csri.toronto.edu!mailrus!uwm.edu!zaphod.mps.ohio-state.edu!sdd.hp.com!elroy.jpl.nasa.gov!lll-winken!ames!decwrl!nsc!voder!dtg.nsc.com!waggoner From: waggoner@dtg.nsc.com (Mark Waggoner) Newsgroups: comp.sys.amiga.hardware Subject: Re: Overdriving 68030 CPUs Message-ID: <398@icebox.nsc.com> Date: 3 Apr 90 21:08:54 GMT References: <1990Mar31.223641.13301@aucs.uucp> Reply-To: waggoner@icebox.UUCP (Mark Waggoner) Organization: National Semiconductor, Santa Clara Lines: 37 In article <1990Mar31.223641.13301@aucs.uucp> 880274d@aucs.UUCP (Ralph Doncaster) writes: >I'll be buying a 68030 based accelerator for my Amiga, without the >68030. What I am wondering is how much can I overdrive a 68030 >beyond it's spec. I'm getting a heatsink and small fan to keep it >cool, so I imagine 20% above spec should be no problem, but I'd like >to know how much more can I go than that. (Could I run a 16Mhz '030 >at 25Mhz an not fry it?) First off, I assume your accelerator is an asynchronous design or you can't run it faster at all. I don't believe that frying the chip is what you need to worry about. Generally, the reason a chip is spec'd at a certain frequency is because the timing parameters and logical operation of the chip can only be guaranteed at that frequency. If it is guaranteed over a commercial temperature range (often 0-70 degrees C) then, assuming that the part operates slower at high temperatures (CMOS does this), you have some margin if the device is kept at room temperature (about 25 degrees C). How much margin you have is something you can only find out by trying it, or perhaps by working for Motorola. You also need to know how critical the timing specifications of the chip are to the particular design you are using. It may be that the part will logically function at higher speeds but that the timing gets out of specified limits. If the design can tolerate this, then you can go faster. Of course it is also possible that the board was designed for room temperature operation and never tested under hot conditions and so you have little or no margin at all. When you run parts out of spec, experimentation and a willingness to risk unreliable operation are required. -- Mark Waggoner Santa Clara, CA (408) 721-6306 waggoner@dtg.nsc.com Unofficially representing National Semiconductor Local Area Networks Group Officially misrepresenting myself.