Ez Computer Setups At Major Matches?

[bgary]

1) Set up the whole match on one computer. The registrations, the stages, everything. This will be your "master" computer.

2) Go to Setup / Match Info and tell it YES this is part of a multicomputer scoring system, and NO, it is not a slave.

3) Go to the File menu and "create a slave system disk". You can do this on a floppy drive or on a USB thumb-drive

4) Take that "slave system disk" to another computer running EZWS and copy the "db" file onto the computer. This will be your "slave" computer (note that you can set up more than one "slave")

5) On the slave computer, go to Setup / Match Info, and tell it YES this is part of a multicomputer scoring system, and YES it is a slave.

Now you have a master and a slave set up.

As Bill Noyes notes, at this point it is CRITICAL to have good "process management" to keep from screwing up data. The most natural way to do it is to set things up so that some of the stages are entered ONLY on the master computer, and others are entered ONLY on the slave. For a 10-stage match, for example, you might enter stages 1-5 on the Master, and stages 6-10 on the slave. Put big signs up on the computer monitors to indicate which stages to enter on which machine, and/or set up workflows so that the wrong scorecards do not end up in front of the wrong machine, etc. If you get this tangled up, it can be a nightmare to sort out (ask for details. I have all the stories -)

So, a typical process is:

-- enter scorecards from stages 1-5 on the master

-- enter scorecards from stages 6-10 on the slave

-- periodically, put the USB thumbdrive on the slave computer, go to the Match menu, and pick "export to master". Make sure to point the data to the USB drive. This copies all the scorecard data on the slave onto the drive.

-- take the USB drive over to the master computer, go to the Match menu, and pick "import from slave". This copies all the scorecard data on the drive into the "master" scoring system.

-- Now, all the stuff is on the master. Print verifications, print results, whatever.

-- If there are any errors or corrections to be made on stages 6-10, they MUST be made on the "slave" computer. Any changes you make to those stages on the master will be LOST (actually, over-written) next time an import from the slave is done.

-- you don't have to copy anything back to the slave before you start entering again. A person can be entering on the slave as soon as the "export" is done.

Hope that helps!

Bruce

[bgary]

Well, Bruce Gary (I think) has a program that does the cross-checking for you, but yeah, I'm sorry, I still think double-entry is nuts.

You're probably right, and I have gotten quite spartan in my stats setups over the last couple of years.

When I learned "big-match stats", it was under the tutelage of Judy Sunderlin and Anita Augsberger, the original "stats babes". They had worked out a system of double entry where they each entered the entire match on two separate computers, and they verified by [manually] comparing computers - they would display results on-screen, one would read off the names in order, and if the names were in the same order on both machines, they were "verified".

This system worked for them because they were *SO* darn fast at entering. Each of them could easily process 1000 scorecards per half-day, so double-entering was way faster than "paper verifying". The downside was that if names were not in the same order on the two screens, it could take a lot of work to figure out what the problem was. And reading 500 or so names, every time you wanted to verify, was a lot of time.

When I came along, I was pretty amazed at how fast they could enter, but sceptical about whether or not the match was "verified" when the names were in the same order on screen. And I figured there had to be a faster - and more definitive - way of doing it. So I wrote a program that mechanically compared the two scoring files. It was pretty much instantaneous, and it told you *exactly* where the problem was. It would say "shooter 112 has a stage-5 score on computer 1 but not on computer 2" or "shooter 112's time on stage 5 is 4.82 seconds on computer 1 but 4.92 seconds on computer 2." Really a nice way to do verifying for the double-entry approach.

Without keypad wizards of their caliber, however, I'm pretty convinced the double-entry method is not worth it. I generally figure that I can handle upwards of 750 scorecards per half-day by myself (assuming that the upstream workflows are in place - that the scorecards get sorted and checked and all that before they are ready to enter). 750 scorecards per half-day works out to a 300-shooter, 12-stage, 2-1/2 day match... so, for matches like the Area-1, I just set up one computer, and have helpers who collect, sort, check and file scorecards, and do paper verification.

For a bigger match, these days I tend to have the same processes, but set up master/slave computers. Even with the little extra work involved to export and import, two good entry people can average well over 600 scorecards per half-day each, so it works out OK.

ETA: the goal is to get *verified* results out as rapidly as possible. I generally get "finals" out on the board within a couple of minutes of the last shot, and these days they are on their way to the USPSA website, via cellphone, while the paper is still being posted. That is a LOT more about good process (and good help) than it is about computer configuration, IMHO.

B

Edited September 19, 2006 by bgary

[Shooter Grrl]

Okay - after reading all the replies, I'm really comfortable with how I'm going to set this up now - thanks everyone!

[robomanusa]

Nick,

Could you explain the process you used last year to get the scores from the input computer(s) into the main computer? (I am avoiding the terms master and slave computers, as I don't think you used EZ's master/slave way of doing things?)

Also, you had to be careful about a few other things (DQ's, division changes and such..) ??

Kyle, I couldnt explain it any better than bgary just did, I used the master with 3 slaves setup, with only the slave machines doing score entering. The master did not do any scoring entering at all.

I did not use a slave 'disk' as I setup a network and just retrived the db file from the slaves via the network when I wanted to sync the stats for printing. I also setup all the machines on the 2 printers we had and could print from any of the computers when needed to either of the printers.

[Shooter Grrl]

2 computers, no network (unless I can find my extra router), 1 printer, no UPS needed - we'll use a laptop as the main computer for entering in case of power failures. I'm thoroughly convinced anything more than that is total overkill (and a waste of time and money!)

With 2 of working stats, one of us can enter and one can pre-sort and verify and then doublecheck after entering - I'm not worried anymore after hearing Bruce's method, cause I KNOW I TYPE FASTER than he does

[bgary]

Heh. I think my dog can type faster than I can.

BTW, just a quick extra note....

I am a big fan of "process" in stats stuff. I like to be able to tell what part of the process a scorecard is in, just by where it is. So I tend to set up my stats operations like this:

-- runner brings in a bunch of scorecards and puts them in the 'incoming" place (may be a table, a bin, whatever)

-- somebody sorts them out (by stage, and by shooter number), and moves them to the "ready to check" place (table, stage bins, whatever), where we have little boxes or racks or whatever for each stage.

-- somebody grabs a stage-stack, and checks the scorecards. That means making sure all the *columns* add up to the right number, making sure all the *rows* add up to the right number, and making sure there is a time. I also like to have the checker "highlight" (yellow marker or whatever) Mikes, No-Shoots, Procedurals, etc, to make it harder for the entry person to miss them. When a stack is checked, it goes to the "ready to enter" place.

-- Entry person grabs a stack, enters them, prints out a verification sheet, and puts the stack and the verification sheet in the "ready to verify" place.

-- Verification person grabs a stack and a sheet, verifies them, and (assuming no problems), folds the verification sheet around the stack and puts it on the "done" table, organized by stage. If there are problems, the scorecard goes back to the "ready to check" place, to have any problems highlighted just as if it was a new scorecard coming in the door. From then on, it flows through the system like any other scorecard.

-- At the end of each half-day period, we throw a big rubber band around each half-day worth of scorecards (ie, a bundle for stage 1, a bundle for stage 2, etc), and file them away.

For a section or state match, it is easy to do all this with just a few people (2-4), including the runner). Trading off jobs keeps it from getting too boring. The only thing I (try to) insist on, is I don't like to have the person who entered a scoresheet also verify that scoresheet. I like to get "different eyeballs" on the scoresheets when they get verified.

It seems like a lot of steps, but it really isn't. And, knowing "what has been done" to a scorecard, just by seeing where it is, can save a *lot* of time over the course of a weekend.

B

[shred]

I've never done a big match, but by far the best hint I've heard is to take a stage-stack and put all the ones with penalties on top. Then enter them. Once you get to the first one without penalties, you can stop looking for them.

[robomanusa]

bgary is almost describing how I had the setup at ohio state last year. Only I didnt use the master to enter any scores on. I had...

master

slave1 - scored stages 1-3

slave2 - scored stages 4-6

slave3 - scored stages 7-9

run in that configuration we had 1 person running for scoresheets, then they were sorted by stages and then entered by 3 people.

setup in this way we were able to stay ontop and not get behind, volunteers were not working there ass off, were able to carry on a conversation to pass the time without fear of pressure and making a mistake along the way......everything went smoothly and we were able to get the scores posted in about 6 minutes or so after last shot fired.

Just make sure you set yourself up so not to be in a rush here rush there all-day long ordeal......you can make it a nightmare or make it so everyone has fun.

[wgnoyes]

2 computers, no network (unless I can find my extra router),...

You can also network 2 machines together with a $10 or so crossover cat5 cable. More than that, and then obviously you need the router. At Area 6 this year (12 stages, 367 competitors), as previously described, we set up a wireless router in one corner of the room with nothing attached to it, then the 3 scoring systems connected and happily moved stuff back and forth, with the master machine hosting the shared printer. On the last day, the MD's system joined into the network from the next room so she could develop and print her prize distribution stuff. Bn

Edited September 21, 2006 by wgnoyes