One mouse click, 2 PC's
in [Design]
|
From: PeterD on 14 Jun 2010 08:41 On Mon, 14 Jun 2010 00:09:38 GMT, keningram(a)overden.com (Ken Ingram) wrote: >Is there any practical way that would enable me to use a single mouse >click in order to start a sequence at exactly the same time on two >separate PC's (identical units)? > >I suppose this means hacking into the mouse lead itself, but how to >find the relevant wires? > >Ken Ingram Define "exactly the same time" more clearly. The answer is yes, there are ways... Also does it have to be a mouse click, or can it be some other type of signal? What are you controlling?
From: Tim Watts on 14 Jun 2010 08:49 On Mon, 14 Jun 2010 04:59:42 -0700, Archimedes' Lever <OneBigLever(a)InfiniteSeries.Org> wibbled: > On Mon, 14 Jun 2010 06:17:36 -0400, JW <none(a)dev.null> wrote: > >>On Sun, 13 Jun 2010 21:08:50 -0700 Archimedes' Lever >><OneBigLever(a)InfiniteSeries.Org> wrote in Message id: >><3qab16l70s0s0pn9rqoa5ull13dlspi8tp(a)4ax.com>: >> >>>On Sun, 13 Jun 2010 20:34:22 -0500, AZ Nomad >>><aznomad.3(a)PremoveOBthisOX.COM> wrote: >>> >>>>On Sun, 13 Jun 2010 18:20:24 -0700, Artemus <bogus(a)invalid.org> wrote: >>>> >>>>>"Ken Ingram" <keningram(a)overden.com> wrote in message >>>>>news:4c157293.3891718(a)news.tpg.com.au... >>>>>> Is there any practical way that would enable me to use a single >>>>>> mouse click in order to start a sequence at exactly the same time >>>>>> on two separate PC's (identical units)? >>>>>> >>>>>> I suppose this means hacking into the mouse lead itself, but how to >>>>>> find the relevant wires? >>>>>> >>>>>> Ken Ingram >>>> >>>>>Even if you succeed with the wiring the sequence start isn't going to >>>>>be that exact as the mouse is a polled device. Art >>>> >>>>Actually, it isn't. >>> >>> Mice are polled devices. >> >>AlwaysWrong. PS/2 mice use an interrupt. You big dummy. > > Do interrupts not also get polled in cyclic fashion? Can you > guarantee > that both machines will poll their respective interrupts at the same > moment? > > When was the last time you saw a new PC or laptop that had a PS/2 > mouse > port? They do, but more often, they are made without them at all. > > There are USB mice that plug into USB ports, and no PS/2 port is > anywhere to be found. Those to are polled, and any such polling would > be asynchronous with a separate device not triggered by the same clock. > > You might think it worked "at the same time" but it in fact cannot be. > There will always be some difference. All of this is utterly irrelevant when compared to the OS scheduling quantum (typically between 10mS to 1mS), assuming the OP is running the test systems on regular machines and not some fancy realtime bit of hardware. Anyway, back to the original point - I would have said time based would be easier - have the button "clicked" at a fixed time and time lock the machines using NTP. Or have a trivial network listener do the "clicking" and use a script on one to send the network signal to start to both machines from a script. -- Tim Watts Managers, politicians and environmentalists: Nature's carbon buffer.
From: Archimedes' Lever on 14 Jun 2010 09:15 On Mon, 14 Jun 2010 12:49:18 +0000 (UTC), Tim Watts <tw(a)dionic.net> wrote: >All of this is utterly irrelevant when compared to the OS scheduling >quantum (typically between 10mS to 1mS), assuming the OP is running the >test systems on regular machines and not some fancy realtime bit of >hardware. For the most part, if he had told us what the circumstance was, any one of us could have delivered a more appropriate response. In one view, one person and two mice could be 'clicked' 'simultaneously', FAIAP. In another, if he really needs more simultaneous synchronization than he himself could bring with two mice, then reliance on raw synchronization that falls inside your timing region may not be enough either. So, yeah, more would be needed. Generally the lay person would not see that as a difference, and if his application requires time stamps or other real time function, he would need something more appropriate, like two systems integrated into a chassis, being triggered by a third device installed into each of those two systems. Even then there can be latencies between pieces of gear, which have to be 'calibrated' against in the individual components. This pretty interesting considering that the systems I currently plan use a 10Mhz source and two GPS fed 10Mhz "switches" that then feed all the remaining gear in the receiver system, keeping them all synch'd up. The rest is just standard network switch gear. Fire up the analyzer though and see that we know how to make the 'eye' look real nice. Fun piecing together millions of dollars worth of equipment and watching that and five hundred or so wires and cables come together to allow a high bandwidth channel to the world, to be placed anywhere in the world.
From: Archimedes' Lever on 14 Jun 2010 09:17 On Mon, 14 Jun 2010 12:49:18 +0000 (UTC), Tim Watts <tw(a)dionic.net> wrote: >Anyway, back to the original point - I would have said time based would >be easier - have the button "clicked" at a fixed time and time lock the >machines using NTP. > >Or have a trivial network listener do the "clicking" and use a script on >one to send the network signal to start to both machines from a script. Only if your net cards are set to NEVER time out and sleep. Many do, and windows uses it, set default to sleep after a certain period. There are latencies, just none that should concern someone that asked how to "see" button clicks from the cable end.
From: Archimedes' Lever on 14 Jun 2010 09:24
On Mon, 14 Jun 2010 09:03:47 -0400, JW <none(a)dev.null> wrote: > > >In addition, you could not splice a USB device to two different computers. I never said you could. My posts was about timing, you idiot. |