Timer Accuracy/ Software

Discussions on buying or building timers, solenoid start gates, light trees, weigh scales, and other race related electronics.
Post Reply
TAL
Master Pine Head
Master Pine Head
Posts: 306
Joined: Wed May 17, 2006 8:47 pm
Location: N.C., USA

Timer Accuracy/ Software

Post by TAL » Sun Nov 18, 2007 9:38 am

We have a couple of Micro Wizard timers (K-3 series) that are very good timers. We also use GPRM software.

The micro wizard web site states the timers are accurate to .0002 and if two cars finish within the 2/10000 that a tie will be determined.
As most already know the micro wizard only displays .001 (thousands)

Well on the timer display the tie will be shown (.001 and .001 ), but when the results are sent to the computer it will send and display on the computer for example .001 and .002 even though the timer displays .001 and .001

What I'm wondering is some racing groups say their microwizard is accurate to 1/10,000 and actually show .0001,(2,3,4,5,etc) computer results even though the microwizard only displays .001 (thousands) and is only accurate to 2/10,000 .

Does the race management software have any effects on the times displayed on the computer and computer display results even though the timer is only accurate to 2/10,000 ?



User avatar
gpraceman
Site Admin
Site Admin
Posts: 4696
Joined: Fri Jun 20, 2003 12:46 am
Location: Highlands Ranch, CO
Contact:

Re: Timer Accuracy/ Software

Post by gpraceman » Sun Nov 18, 2007 6:00 pm

TAL wrote:We have a couple of Micro Wizard timers (K-3 series) that are very good timers. We also use GPRM software.

The micro wizard web site states the timers are accurate to .0002 and if two cars finish within the 2/10000 that a tie will be determined.
As most already know the micro wizard only displays .001 (thousands)

Well on the timer display the tie will be shown (.001 and .001 ), but when the results are sent to the computer it will send and display on the computer for example .001 and .002 even though the timer displays .001 and .001

The way the Fast Track timers handle ties does pose a problem for race management software. People do want the software to match what the timer is displaying. Fast track timers send data to the computer that indicates who won a tiebreaker, even though the tied racers will show the same time. There in lies the problem in matching what the timer displays.

Here's an example of the timer's data:

A=3.005! B=3.002" C=3.003# D=3.003$ E=3.004% F=3.006&

Two racers are tied with a time of 3.003 seconds. The character after the times indicates the finish order. Lane C (3) ended up in 3rd place and Lane D (4) got 4th place, taking the tiebreaker into account.
TAL wrote:What I'm wondering is some racing groups say their microwizard is accurate to 1/10,000 and actually show .0001,(2,3,4,5,etc) computer results even though the microwizard only displays .001 (thousands) and is only accurate to 2/10,000 .

Does the race management software have any effects on the times displayed on the computer and computer display results even though the timer is only accurate to 2/10,000 ?
With Version 7 of GPRM, it will check for tiebreakers and add 0.0005 seconds to the tiebreaker losers. All times in GPRM for this timer are now displayed out to 4 decimal places. All you will ever see in that last digit is a 0 or a 5 and the 5 would only be for a tiebreaker loser. So, in the above scenario, Lane D (4) would get a time of 3.0035 seconds.

I don't see how you would ever get the 4th decimal place to show anything else, so people claiming an accuracy of 1/10,000 second are incorrect.


Randy Lisano
Romans 5:8

Awana Grand Prix and Pinewood Derby racing - Where a child, an adult and a small block of wood combine for a lot of fun and memories.

User avatar
Darin McGrew
Pine Head Legend
Pine Head Legend
Posts: 1753
Joined: Fri Jul 18, 2003 1:23 pm
Location: Mountain View, CA
Contact:

Re: Timer Accuracy/ Software

Post by Darin McGrew » Sun Nov 18, 2007 6:11 pm

gpraceman wrote:With Version 7 of GPRM, it will check for tiebreakers and add 0.0005 seconds to the tiebreaker losers.
So just out of curiosity, what is displayed if the results are

A=3.005! B=3.005" C=3.005# D=3.005$ E=3.005% F=3.005&

?



User avatar
gpraceman
Site Admin
Site Admin
Posts: 4696
Joined: Fri Jun 20, 2003 12:46 am
Location: Highlands Ranch, CO
Contact:

Re: Timer Accuracy/ Software

Post by gpraceman » Sun Nov 18, 2007 6:25 pm

Darin McGrew wrote:A=3.005! B=3.005" C=3.005# D=3.005$ E=3.005% F=3.005&
You do pose a troublesome, but highly unlikely scenario.

To be honest, GPRM V7, as it is now, would actually display:

Lane 1 3.0050 - 1st
Lane 2 3.0055 - 2nd
Lane 3 3.0055 - 2nd
Lane 4 3.0055 - 2nd
Lane 5 3.0055 - 2nd
Lane 6 3.0055 - 2nd

In that scenario, GPRM would not match the finish order of the timer. Though, think how small the odds of two cars being tied is. Then think about how much smaller the odds of three or more cars being tied is. Anyways, I'd rather err towards showing a tie in such rare cases.

Personally, I don't like the tiebreaker process with these timers. I'd rather see the racers tied. Heat ties are extremely unlikely to pose a problem in the final standings, so why worry about them? Just worry about ties in the final standings. The only reason I even try to get GPRM to match the timer display in these cases (at least the vast majority of them) is that customers think there is something wrong with the software if it doesn't match the timer.


Randy Lisano
Romans 5:8

Awana Grand Prix and Pinewood Derby racing - Where a child, an adult and a small block of wood combine for a lot of fun and memories.

TAL
Master Pine Head
Master Pine Head
Posts: 306
Joined: Wed May 17, 2006 8:47 pm
Location: N.C., USA

Re: Timer Accuracy/ Software

Post by TAL » Sun Nov 18, 2007 7:31 pm

I agree that three way ties rearly occure but they do .
I also agree that a single heat tie will most likely not have drastic measures on the final outcome just as stated above .

My concerns were primarily when kids or parents of the kids see a tie on the timer and then see a display screen showing a different story some do question the matter and it is really nice to have at least a simple logical explanation if the matter is questioned.

Also there are some group(s) making accusations that their Microwizard is accurate to 1/10,000 and was wondering was/is it possible to modify the accuracy of the micro wizard with software. Just wanted to be sure.

Thanks for the quik and reasonable explanations.
I do like the addition you added of the .0005 display.

So version GPRM 7 in a three way tie will display 1st as .0010, and a tie for 2nd and 3rd as .0015 on the computer/display screen; Is that correct?



User avatar
gpraceman
Site Admin
Site Admin
Posts: 4696
Joined: Fri Jun 20, 2003 12:46 am
Location: Highlands Ranch, CO
Contact:

Re: Timer Accuracy/ Software

Post by gpraceman » Sun Nov 18, 2007 7:54 pm

TAL wrote:My concerns were primarily when kids or parents of the kids see a tie on the timer and then see a display screen showing a different story some do question the matter and it is really nice to have at least a simple logical explanation if the matter is questioned.
If the timer reports a true tie (within 0.0002 secs of each other) then you should also see a tie in the software. The software should not report something different in that scenario.
TAL wrote:Also there are some group(s) making accusations that their Microwizard is accurate to 1/10,000 and was wondering was/is it possible to modify the accuracy of the micro wizard with software. Just wanted to be sure.
There is no way to change the timing resolution of that timer via external software. The SmartLine timers do have a way for software to change the timing resolution, but not the Fast Track timers.
TAL wrote:So version GPRM 7 in a three way tie will display 1st as .0010, and a tie for 2nd and 3rd as .0015 on the computer/display screen; Is that correct?
Yes. I'll rerun some tests to verify, but that's what I'd expect to see.


Randy Lisano
Romans 5:8

Awana Grand Prix and Pinewood Derby racing - Where a child, an adult and a small block of wood combine for a lot of fun and memories.

User avatar
Microwizard
Merchant
Merchant
Posts: 6
Joined: Mon Sep 26, 2005 4:09 pm
Location: Union, KY USA

Re: Timer Accuracy/ Software

Post by Microwizard » Wed Nov 21, 2007 2:32 pm

Our Fast Track timers are only accurate to 0.0002 seconds. We did extensive testing on our sensors to see how accurate they could be to determine when was the best time to call a tie. If we were accurate to 0.0001 seconds we found that the timer could misreport the information as the sensors were not accurate enough to call so close of a race.

At the same time since so many people buy our timers for just the place finish we didn't want to limit our accuracy to 0.001 seconds and wanted to have less ties than that.

In your initial example TAL stated that he had a result of 0.001 and 0.001 on the timer, but a result of 0.001 and 0.002 on the computer. In very early K3 models this could be possible and this can be fixed by loading the latest firmware. You can load the firmware in the bootloader section of our software utility. You can download the firmware here for K3 105b.hex.

So in conclusion there is no way to alter our timer to be accurate to 0.0001 seconds and if you somehow did you could actually end up calling the wrong winner in an extremely close race. We strive to have the highest accuracy possible without producing an erroneous winner and that is why we have the accuracy to 0.0002 seconds even if it does cause some problems with software at times.

Ian Ferguson
Micro Wizard



User avatar
gpraceman
Site Admin
Site Admin
Posts: 4696
Joined: Fri Jun 20, 2003 12:46 am
Location: Highlands Ranch, CO
Contact:

Re: Timer Accuracy/ Software

Post by gpraceman » Wed Nov 21, 2007 2:49 pm

Hello Ian,

Thanks for jumping in and clarifying some things.
Microwizard wrote:We strive to have the highest accuracy possible without producing an erroneous winner and that is why we have the accuracy to 0.0002 seconds even if it does cause some problems with software at times.
Without challenges, life would get pretty boring :wink: With GPRM V7, the hardware interface was redesigned to allow greater flexibility with each timer supported. Essentially, we can tweak the interface to work with particular scenarios that you find with one timer but not another. With the variety of features from one manufacturer to another and how each brand behaves, that definitely helps to provide better support.


Randy Lisano
Romans 5:8

Awana Grand Prix and Pinewood Derby racing - Where a child, an adult and a small block of wood combine for a lot of fun and memories.

Post Reply