On Thu, Aug 2, 2012 at 3:48 PM, Sean Breheny wrote: > Are you considering only discrete gates or do you also want to know > about ones which are completely internal to a large IC (like a > processor)? Normal CMOS gate switching speed is very dependent on the > load capacitance which is usually much higher if that output line > actually comes out to a pin on the IC. So, the fastest CMOS gate (as a > part of a microprocessor or FPGA) is probably around 10GHz switching > speed but the fastest 74xx series variant would probably be only > around 300MHz to perhaps 1GHz. Note that this 10GHz figure I am giving > here is my guess for the maximum clock rate for a single gate - when > you put them together in a system the max clock rate is greatly > reduced because the system is usually operated synchronously so the > worst-case propagation delay path will determine the maximum system > clock rate. It doesn't have to be discrete, it can come in any package/platform/technology that really doesn't matter for me. I am just interested in the fastest response time, the product will be designed around this feature. > > I have a microwave prescaler which runs to 12GHz. It has flip-flops in > it which are being clocked by the input frequency. However, it is very > different from typical CMOS logic because the input and output logic > levels are much smaller and the output doesn't switch very "hard" ON > even when clocked at moderate frequencies. I say this because the > output is not very rich in harmonics when I view it on a 21GHz > spectrum analyzer. Sounds cool, which methodology did you follow while building the prescaler = ? > > I am fairly sure that there have been experimental electronic logic > gates which have switched at several hundred GHz. > > What do you mean by "non-contact magnetic gates"? > Imagine there are two coils (very rough example but same principle) and we apply a potential to first coil's terminals and a potential will be measured in secondary coil's terminals. Since the magnetic field will propagate at the speed of light, only capacitive behavior will effect the response time. Am I missing something here ? --=20 http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist .