Wanted: A Very Accurate Timer

You've said that you just want a timer that will run over a period of six hours with 1/60 sec accuracy.

You've said that all you want is a simple clock display that reads out seconds.

You've said that it needs to just start at an arbitrary start time and count from there.

None of these goals is particularly hard, but to get that accuracy you'll need to buy some sort of commercial clock with a time display and mate it to a frequency source (which in this industry is also commonly called a "clock", further confusing this question) that is more accurate than such timers usually come with. Or, if you have real money to spend you can buy something with a real frequency standard (clock) inside which could be purchased with a digital clock display. The price for something like this could be anything from $2500 to $40,000. This whole range is much more accurate than you've asked for, but it seems unlikely that anyone makes something that meets just your minimum accuracy requirement.

The problem that I see is that you've not asked for any kind of electronic input or output for the timer, which makes most of us wonder how you expect to be able to use 1/60 sec accuracy while just doing this by eye.

If you're planning to use electronic start and stop signals, then you can get much better accuracy than 1/60 sec.

So this leaves us confused about what it is that you really want/need. It's not that people here are trying to be difficult; it's that they are trying to be helpful, but the specs of your request, taken as a whole, just don't seem to make sense.

So if you explained a little more, without giving away any of your secrets, then you will probably get the answer you're after.

-

----------------------------------------------- Jim Adney snipped-for-privacy@vwtype3.org Madison, WI 53711 USA

-----------------------------------------------

#2 on the list...

12 models with 1/100 displays and up to 100 hours. Whether they're really 99.99992%+ accurate is for you to determine.

Now, these were very easy to find. They meet your limited "simple" specs. Seiko is a name brand in sports timing, and the printer model has it all.

I'll politely assume that surely you searched Google first, found these, and determined they were inadequate by merely looking at them. So, what makes these unsuitable?

Richard

Well, you have already hinted at it your self.

What are the odds that a stand alone stopwatch will be anything close to 1/60th of a second of the correct time at the 6 hour mark?

And we all know that 1/100th of a second on a hand held stopwatch is nothing but a marketing gimic. :-)

Darren Harris Staten Island, New York.

Well, you have already hinted at it your self.

What are the odds that a stand alone stopwatch will be anything close to 1/60th of a second of the correct time at the 6 hour mark?

And we all know that 1/100th of a second on a hand held stopwatch is nothing but a marketing gimic. :-)

Darren Harris Staten Island, New York.

Well, you have already hinted at it yourself.

What are the odds that a stand alone stopwatch will be anything close to 1/60th of a second of the correct time at the 6 hour mark?

And we all know that 1/100th of a second on a hand held stopwatch is nothing but a marketing gimic. :-)

Darren Harris Staten Island, New York.

Yup.

So it seems. :-)

Let me repost something I wrote earlier in this thread:

"It is way too complicated to go into the details as far as why I need this, but basically this will involve conducting experiments/tests on the accuracy of human timing, and also the confirmation of certain conclusions drawn from studying the code contained within the hardware I'll be testing against."

"1/60th of a second is important because it is specific to that hardware and how it functions. It uses registers that change every 1/60th of a second to make certain occurances "random". If one could react with an accuracy of 1/60th of a second, then these occurances would follow a predictable pattern. But of course that kind of timing is not humanly possible with any kind of consistency."

Again, this involves "human timing".

I covered everything pertinent in my posts, and have no idea what else I can say(that doesn't throw everyone further into a state of cconfusion).

There are no "secrets".

This is actually much simpler than the original project which involves more than just a timer, but the ability to record the time of each of four (joystick)input activations/deactivations, which could number over

300 over the course of between 2 and 3 minutes. And then play them back the same way.(Thereby replicating my moves with that 1/60th of a second accuracy). But since I am having such difficulty with this, the original needs are definitely out of the question anytime soon.

I thought that there might be an affordable timer that would somehow keep it's accuracy by via 60Hertz AC. But I guess not.

Darren Harris Staten Island, New York.

As said before, the power company has something almost up to that standard, but that idea was probably also rejected in earlier discussion here?

--
Met vriendelijke groet,

   Maarten Bakker.

Google GPS stopwatch Job done.

On Mon, 27 Jun 2005 10:06:02 -0700, Searcher7 wrote: [and seems to have snipped all attribution]

We are "unfamiliar with the project" because you haven't _told_ us anything about the project. You've made some obscure reference to video games;

What do you need to time?

What are you trying to accomplish?

Yeah, the question was simple. "I need to time an interval to an accuracy of 1/60 second, over a span of possibly 6 hours."

People who have many years' experience have informed you of almost a half-dozen ways to accomplish this, but apparently they're unsatisfactory.

And newbies wonder why us crusty old farts get annoyed at newbies.

Get to the f---ing point, rather than bitching about the suggestions that have been offered based on nothing more than the above, with a dollop of mind reading thrown in. In Other Words, What Are You Trying To Accomplish?

Thanks, Rich

Me Too!

;-) Rich BTW, the email is richardgrise at yahoo dot com, but elide ard.

Well, Jesus Aitch! Why didn't you just say so? There's probably a dozen people here who could design a joystick tracker with millisecond accuracy, and record switch closures to microsecond tolerances. I'd lighten up a bit on looking for a coincidence detector, which is not going to happen unless you can physically hack the game you're trying to hack.

If you're looking at reaction time stuff, then you'll have to find a biology or anatomy group, although I'd still venture to guess that most people who are conversant with this level of electronics probably know something about nerve impulse propagation and electromyelographic interfaces. Heck, a year or so ago, I was being tested for neuropathy, and they taped some electrodes that look very much like EKG or EEG electrodes to my legs, and the nurse (or lady doctor - we didn't get into that) took a hand-held that looked so much like a stun gun that when I said, "Stun Gun???" she said, "That's what everybody says." They stunned me, and they took readings of my neural response. Diagnosis: Alcoholic Neuropathy.

Oh, well.

You might also look into the source code for "MAME" - Multiple Arcade Machine Emulator. I play Mr. Do! and Bubble Bobble regularly, and am considering something much like your project, to see how the software uses joystick/button actions to modify its own algorithm!

Good Luck! Rich

I think for you to measure an event with an accuracy of 1/60th of a second, you need to take measurements at least 120 times a second (well known theorem, I forgot the name).

--
Met vriendelijke groet,

   Maarten Bakker.

That would be Shannon.

No, that would by Nyquist. Shannon limits the data rate, based on bandwidth and S/N ratio.

--
  Keith

Nyquist.

As has been mentioned repeatedly in this thread, the short-term accuracy of power-system frequencies is several (many?) orders of magnitude worse than your stated requirement (they 'run slow' at times of high demand, but are carefully made to 'catch up' at other times so that domestic clocks, etc., don't develop cumulative gross errors).

Your confidence that the videogame's(!) registers 'will have undergone

1,296,000 increments over the course of 6 hours' is certain to be similarly misplaced, though if crystal-controlled perhaps only to the extent of a few hundred increments.

This, basically, is IMO why no-one can see the point of your accuracy-requirement - you seem to believe that you need it to 'keep in step' with a process that is proceeding at rate only approximately-known but from which you can't derive any synchronising-information.

--
Peter Duck

As I mentioned, I can't hack into the gameboard. This has to be a separate device. And this project is on the back burner anyway, since it is more complex than just the timer I am seeking for now.

MAME is not an option, because the original hardware must be used.

The original project involved a "Automatic Pattern Generator". People develop patterns to clear the mazes in the game Pac-man. I wanted to have a computer develop patterns through trial and error, but that would entail hacking into the game board.

So I came up with the idea to at the press of a button have my joystick movements recorded and then have the option of playing the sequence back through the joystick inputs.(It's a lot more complex than this, but those are the basics).

I've had to put these ideas on the back burner and concentrate on something else that requires the timer I posted about.

Darren Harris Staten Island, New York.