Return to Project-GC

Welcome to Project-GC Q&A. Ask questions and get answers from other Project-GC users.

If you get a good answer, click the checkbox on the left to select it as the best answer.

Upvote answers or questions that have helped you.

If you don't get clear answers, edit your question to make it clearer.

+1 vote
1.7k views

This is about this trail: WRC Challenge trail in South Dakota

When every challenge on the trail (currently 14) has a checker, we will reward each tag-creator with 2 months of membership per tag.

If you find that a challenge is technically impossible, reply here and tell me. I will then review it, and add it as an exclusion if I agree.

Feel free to post here when you have created a checker, so that we can follow the process. When the trail is complete, do post here so that I notice, and can process the rewards.

Exception list:

CONTEST OVER, REWARDS SENT

in Miscellaneous by magma1447 (Admin) (241k points)
edited by magma1447 (Admin)
I'm working on a checker for the last remaining challenge (http://coord.info/GC5CR9B), the one where you have to fill the grid with unique cache owners and also make sure that all 36 alphanumeric initials are represented amongst those cache owners. I'm pretty sure this is a hopelessly intractable problem in the worst-case scenario (e.g., someone with a hundred finds for each of the 81 DT-combos and complicated overlap in the hiders represented), but the worst-case scenario is not all-that likely. In fact, I was able to use some tricks to get a script that can run in the allotted time for everyone who has logged the cache except for JBase. It even successfully gives negative answers to lots of people (everyone I've checked) who have filled their grids, but not with enough cache hiders to qualify. Unfortunately, JBase seems close enough to that worst-case scenario that it times out for them. I'm working to see if I can speed it up some more, but it might not really be possible.

7 Answers

0 votes
 
Best answer

Contest over and rewards will be sent to these developers:

  1. Target. => 7*31 days
  2. mole125 => 2*31 days
  3. ErWenn => 2*31 days No, I got so impressed by his fizzy checker that I changed it to 12*31 days
  4. Jakuje => 1*31 days
by magma1447 (Admin) (241k points)
+1 vote
This one http://www.geocaching.com/geocache/GC5CFC1_wrc-challenge-fizzy-not-in-my-state isn't possible - we don't know the cachers home state
by mole125 (Expert) (21.1k points)
+1 vote
I've added the following 2 tags:
Cache: GC5CA5K - "WRC" Challenge (Fizzy) Not Chosen Size

Config: { "challenge": [ "define 'cell' as expression difficulty * 1000 + terrain*100.", "find 81 caches with unique cell field where gccode <> 'GC5CA5K' and size == 'Not chosen'.", "output sorted by cell table gccode, difficulty, terrain, cache_name as 'Name', visitdate as 'Visit Date'." ] }

Cache: GC5CF74 - "WRC" Challenge Fizzy-No Traditionals, No Unknowns

Config: { "challenge": [ "define 'cell' as expression difficulty * 1000 + terrain*100.", "find 81 caches with unique cell field where gccode <> 'GC5CF74' and type <> 'Traditional Cache' and type <> 'Unknown Cache'.", "output sorted by cell table gccode, difficulty, terrain, cache_name as 'Name', visitdate as 'Visit Date'." ] }

Hint: there are some other very very similar caches to these in the series that other people may want to tag with the appropriate changes to the above config...
by mole125 (Expert) (21.1k points)
+1 vote

All exept the already mention are possiblet to create a checker for

I added a coupel of tags

"WRC" Challenge (Fizzy) with Multis
"WRC" Challenge (Fizzy) Micros Only
"WRC" Challenge (Fizzy) Smalls Only
"WRC" Challenge (Fizzy) Regulars
South Dakota's "WRC" (Fizzy) Challenge
"WRC" Challenge (Fizzy) with the lower 48
"WRC" Challenge (Fizzy) with a Jasmer   The checker resultis i a lot of not ok because it is a Jasmer challenge and a new neede month are added for each new month
by Target. (Expert) (104k points)
edited by Target. (Expert)
+1 vote

Hey, I just made a multi-region fizzy checker to handle this cache:

GC5CFC0 "WRC" Challenge (Fizzy) - Hat Trick States

 

 

by ErWenn (310 points)
I seldom check the source code, but this time I did. I really like that you commented the concept in the beginning. Much easier to know what to expect then.
+1 vote

Well, I can't think of a way to improve this enough to get the script to not time out on that one cacher (JBase), but it seems to work for everyone else, so I'm enabling it now.

Check out my checker for http://coord.info/GC5CR9B, which I believe is the last cache for this challenge.

If anyone can think of a way to improve the speed, I'd be willing to take suggestions. Similarly, if you find other cachers for whom the script times out, let me know. 

by ErWenn (310 points)
I am a week late (been busy), but the script is disabled now?
That might explain why no one responded. It should be enabled now.
Wow, lot's of source code. Tried to read it, but gave up when I realized the amount of code.

I am a bit surprised I don't fulfil it myself since I have 10 complete DT-loops. In fact I am so surprised that I might consider making an alternative version in an offline program. My 4 DT's that I "only" have 10 of seems to have pretty unique cache owners.
Ok, I miss-read the rules. I didn't understand that I HAVE to use cache owners for every alphanum character. 0-9 probably makes it quite hard.
(I only have 6/10 digits, two of those I only have in 1 DT each)
To improve the checker not in speed but in usability for a cacher do a A-Z0-9 test first and A full DT test first. Display the result.
In both my and ganja1447 case the problem was missing CO starting with 0-9
I cloned the code and relaxed the requirement to only A-Z and we both make it
http://project-gc.com/Challenges//12927

After a quick read trough of the code that I would not say that I totally understand I am quit sure you are doing the partially wrong order.
I think you are testing CO in order of the DT entries they have with fevers finds.
You shall also sort the alphabet in order of how many CO there exist on the letters. CO with 0-9 are not common and that will lock them down immediately if I am not mistaken
And if the DT matrix in not filled in order of entries with few finds it should do that too. I thin you fill them in 1/1 1/1.5 now and that is the wrong order.
The DT matrix entries in unique CO:s in each DT and not number of finds on the DT

Or I dont realy understand the code. Or my ordering guess is wrong but it seams natural to fill the unusial entris in the DT matrix and Alphablet list in order of uniqnes. I would do it that way if i did it on paper.

The DT matrix order in unique CO:s in each entry and not number of finds
Displaying a DT matrix with number of CO if failure and number of CO on each letter if failure would be a good idea. Thos are the limiting factor and would give the cacher an idea where to look. Entries with low number is where to look for the problem
I suspect to the challenge is really two problem.
1:A different CO on each matrix entire
2:A CO alphabet with unique DT on each alphabet.
Both are easier to solve and combining the is probably easy
the first is a problem on high D and/or T caches
and the second is likly solved on low DT cache when only 36 entires are neded for A-Z0-9 And the unusual liky are on entries wih many finds
The code is complex and not as easy to read as my usual code. For that, I apologize. It's not a simple problem, and for once, I was trying to code for efficiency instead of readability.

It already does some of the things you ask.

The alphabet always goes in the same order, but the order it goes in is from least common as a starting geocacher name to most common, so that it tries for the hardest letters (3, 5, 7, ...) first, and puts off the easy ones (..., M, C, S, T) until later. I looked at a sample of a few thousand cache hider names and did my calculation there.

On second glance, I realize that I did that calculation but never put it into the code! Ha. I'll fix that soon.

But I definitely do reorder the DT-combos in order from least hiders to most hiders. That (combined with filling the letters first and the "sanity" check on the number of letters left) is the only reason the script terminates at all. If you're like ganja1447 with ten finds for every DT-combo, that's 10^81 combinations to check. Even if you could check a million combos a second, that's orders of magnitude longer than the age of the universe. At that speed, you could only check about a dozen DT-combos in a year.

Anyway, the script also checks for people without a completed DT-grid yet, but I haven't yet implemented a check to make sure that all the letters and numbers are represented somewhere (although if they aren't, it will fail the very first "sanity check").

The sanity check is weird. It makes things run longer when finding a positive, but without it, finding negatives take too long.
I hope you didn't take our comments about "lots of code" as criticism. It was rather the opposite, it's impressing that you have managed to solve this for most cases in the time frame and with the amount of cpu power given.
Ugh, well now that I went and put in the letter order optimization, it *does* terminate for my one exception JBase. But unfortunately, it tells me that JBase does not meet the requirements. Which either means there's a bug, or somebody went back and changed the DT or cache owner for one of the caches JBase needed or deleted one of his old logs. I'm leaving it enabled for now so people can still look at it, but I'll have to check out the problem myself later.
It should be fairly easy to confirm that his screenshot fits the data that Project-GC has. I can probably go through that today.
His 1.5/3.5 (GC2ZG14) is hidden by six2magoo, but has placed by as "ukillitwegrillit". This is gray zone of how to interpret "cache owner", but in my opinion, it is the one owning the cache, ie, your code does the right thing.

Same with 2.0/2.5 (GC3ERAA), placed by Qrtne, owner is CreepyJoe. My assumption is that this one is why he doesn't complete it with your code. Q is quite uncommon.

That was the only two I found that didn't fit his screenshot. Or it did, depending on how one interpret the requirements. For future reference, I will post my data here.
AND gccode IN (
'GCPG8W', 'GC42V1B', 'GC2H2EZ', 'GC3C5AP', 'GCA87D', 'GCEC40', 'GC21N2T', 'GCP29C', 'GC3H2Z3',
'GC1RFBE', 'GC2RQNK', 'GC1XPTV', 'GC1NV3J', 'GC1RCQA', 'GC2ZG14', 'GCP3MD', 'GCHVBq',  'GCK8XH',
'GC66CF', 'GC1ACJX', 'GCXKAH', 'GC3ERAA', 'GC9594', 'GCT2G9', 'GCR3B1', 'GC17XYB', 'GC1RVMK',
'GC1BN4Z', 'GC345BW', 'GC550M8', 'GCT2DB', 'GC3MBWN', 'GC3B1', 'GC36R8T', 'GCZY9J', 'GC1HB74',
'GCPNJG', 'GC1Y2VQ', 'GC3384C', 'GC1BPNC', 'GCXMKT', 'GC17D26', 'GCJYF0', 'GCVJZR', 'GCVYKT',
'GC11R4X', 'GCGHKB', 'GC20DF', 'GC1JG8H', 'GC8334', 'GC32PT4', 'GCZ8KT', 'GC25XAG', 'GCDB76',
'GCQCXH', 'GC1A6E7', 'GC1HFH4', 'GCGYF0', 'GC334ZA', 'GC1MJDM', 'GCNRY6', 'GC585E', 'GC20',
'GCWZ8V', 'GC13V6M', 'GC111W1', 'GC3BG93', 'GCNHPK', 'GC29XGN', 'GC2KHY6', 'GCJE26', 'GC3EHXK',
'GC12QFK', 'GC2FEX5', 'GCZYBR', 'GC14B96', 'GC369AR', 'GC100XK', 'GCDB55', 'GC17PV2', 'GC1GVN6'
)

+---------+------------+---------+----------------------+
| gccode  | difficulty | terrain | name                 |
+---------+------------+---------+----------------------+
| GCPG8W  |        1.0 |     1.0 | 2bugs                |2
| GC42V1B |        1.0 |     1.5 | boyercrew            |b
| GC2H2EZ |        1.0 |     2.0 | 7-Continents         |7
| GC3C5AP |        1.0 |     2.5 | HudsonAug5           |h
| GCA87D  |        1.0 |     3.0 | IBFun                |i
| GCEC40  |        1.0 |     3.5 | King Boreas          |
| GC21N2T |        1.0 |     4.0 | joukkusisu           |
| GCP29C  |        1.0 |     4.5 | dawnandgary          |d
| GC3H2Z3 |        1.0 |     5.0 | Aceey                |
| GC1RFBE |        1.5 |     1.0 | 3RoadMonkeys         |3
| GC2RQNK |        1.5 |     1.5 | 08rocklady           |0
| GC1XPTV |        1.5 |     2.0 | 83pony               |8
| GC1NV3J |        1.5 |     2.5 | xrobmn               |x
| GC1RCQA |        1.5 |     3.0 | blobabee             |
| GC2ZG14 |        1.5 |     3.5 | six2magoo            |<- "placed by " ukillitwegrillit
| GCP3MD  |        1.5 |     4.0 | Billy-Bob1993        |
| GCHVBQ  |        1.5 |     4.5 | Shrek & Fiona        |
| GCK8XH  |        1.5 |     5.0 | AstroDiver           |
| GC66CF  |        2.0 |     1.0 | Yellow Jacket        |y
| GC1ACJX |        2.0 |     1.5 | vtwinspin            |v
| GCXKAH  |        2.0 |     2.0 | 9Key                 |9
| GC3ERAA |        2.0 |     2.5 | CreepyJoe            |<- "placed by" Qrtne
| GC9594  |        2.0 |     3.0 | RainDog              |r
| GCT2G9  |        2.0 |     3.5 | mr.pike              |m
| GCR3B1  |        2.0 |     4.0 | cub scout pack       |
| GC17XYB |        2.0 |     4.5 | Hike and Seek        |
| GC1RVMK |        2.0 |     5.0 | OC1GuyWI             |o
| GC1BN4Z |        2.5 |     1.0 | Justus4geoteam       |j
| GC345BW |        2.5 |     1.5 | kc9gbo               |k
| GC550M8 |        2.5 |     2.0 | Trekkie79            |
| GCT2DB  |        2.5 |     2.5 | PCFrog               |p
| GC3MBWN |        2.5 |     3.0 | Love2Labs/kmhiker    |
| GC3B1   |        2.5 |     3.5 | kbraband             |
| GC36R8T |        2.5 |     4.0 | six2magoo            |
| GCZY9J  |        2.5 |     4.5 | Geoboss              |
| GC1HB74 |        2.5 |     5.0 | StaticTank           |
| GCPNJG  |        3.0 |     1.0 | annakido             |
| GC1Y2VQ |        3.0 |     1.5 | SEAM Smiley          |s
| GC3384C |        3.0 |     2.0 | 6OnaHunt             |6
| GC1BPNC |        3.0 |     2.5 | G Force1             |g
| GCXMKT  |        3.0 |     3.0 | treehugwi            |
| GC17D26 |        3.0 |     3.5 | akahix               |a
| GCJYF0  |        3.0 |     4.0 | zoejam72             |z
| GCVJZR  |        3.0 |     4.5 | Mythy                |
| GCVYKT  |        3.0 |     5.0 | GizFinder            |
| GC11R4X |        3.5 |     1.0 | 4Adventurers         |4
| GCGHKB  |        3.5 |     1.5 | Whitewater Guy       |w
| GC20DF  |        3.5 |     2.0 | raslas               |
| GC1JG8H |        3.5 |     2.5 | myside               |
| GC8334  |        3.5 |     3.0 | Team Honeybunnies    |
| GC32PT4 |        3.5 |     3.5 | Stu31                |
| GCZ8KT  |        3.5 |     4.0 | labrat_wr            |l
| GC25XAG |        3.5 |     4.5 | 10_is_enough         |1
| GCDB76  |        3.5 |     5.0 | JacobBarlow          |
| GCQCXH  |        4.0 |     1.0 | cachedasher          |
| GC1A6E7 |        4.0 |     1.5 | norski42             |n
| GC1HFH4 |        4.0 |     2.0 | So What!             |
| GCGYF0  |        4.0 |     2.5 | MR.SAR               |
| GC334ZA |        4.0 |     3.0 | habu!                |
| GC1MJDM |        4.0 |     3.5 | campwebb12           |c
| GCNRY6  |        4.0 |     4.0 | kent1915             |
| GC585E  |        4.0 |     4.5 | OffTrailHikers       |
| GC20    |        4.0 |     5.0 | Lakebum              |
| GCWZ8V  |        4.5 |     1.0 | 53M38A1              |5
| GC13V6M |        4.5 |     1.5 | tvgs                 |
| GC111W1 |        4.5 |     2.0 | Amarakana            |
| GC3BG93 |        4.5 |     2.5 | passmhi              |
| GCNHPK  |        4.5 |     3.0 | ArmyMen              |
| GC29XGN |        4.5 |     3.5 | mejadiaz             |
| GC2KHY6 |        4.5 |     4.0 | SuperGoober          |
| GCJE26  |        4.5 |     4.5 | beaner rat           |
| GC3EHXK |        4.5 |     5.0 | EskoClimber          |e
| GC12QFK |        5.0 |     1.0 | Team GCHound         |
| GC2FEX5 |        5.0 |     1.5 | snosho1              |
| GCZYBR  |        5.0 |     2.0 | charles_city_biker   |
| GC14B96 |        5.0 |     2.5 | Zemmy                |
| GC369AR |        5.0 |     3.0 | fanasfreaks          |f
| GC100XK |        5.0 |     3.5 | Math Boy             |
| GCDB55  |        5.0 |     4.0 | BlueDeuce            |
| GC17PV2 |        5.0 |     4.5 | LDove                |
| GC1GVN6 |        5.0 |     5.0 | Trekkin' and birdin' |t
+---------+------------+---------+----------------------+
81 rows in set (0.01 sec)
"Hidden by" and variations on the "birthday" of a cache is always confusing in checkers.
Especially when groundspeek, gsak and pgc uses slightly different name on the data. Groundspeak use different name on the webpage and in the GPX.
And really fun when the CO of a challenge dont know the difference hidden and published are sometime both used in the description referencing the same thing but they can be diffrent
Thanks for doing that extra work, ganja. I checked a few of the obscure ones, but I hadn't checked close enough to catch those.

I think that means that the script works, and *that* means that we've finished the requirements for the reward!
Oh, and I was more just expressing my own unhappiness of the messiness of the code. I might be a bit of a neat freak when it comes to my code.
0 votes

Depending of the acceptable false negative/positive ration this is not possible

http://coord.info/GC5CRJ8  "WRC" Challenge (Fizzy) Lucky 13~A Fizzy Challenge

The problem i to identify a Fizzy challenge(.  You could a fizzy and challenge match on the cache name but there are many challenges that are equivalent with other names. Se for example the example list on the cache page
You might get some false positive if fizzy is in the name but the challenge is different requirements.  Depending of the CO definition of a fizzy challenge (if all 81 DT are needed) GC44B3E - Non-Trad Cache Challenge #5 - mini fizzy is not ok becuse only D,T<=3 is required

by Target. (Expert) (104k points)
I'll just assume you are right. I would expect the fizzy concept to be quite fuzzy. No need for me to dig around to find out what you already know. (added to the list)
...