Talk:Shipwrightery

Piece frequencies
It'd be good to have a rough idea of how often each piece occurs, and how this changes with difficulty level. --KrisInJapan 18:19, 28 December 2006 (PST)

Odd look
The Berth piece in the diagram doesn't look right, at least from my browser (Firefox). Looking at the page source, it is laid out correctly. -- Thunderbird
 * Fixed. The Shipwrightery4x2 template was actually 2x4.  23:32, 13 August 2005 (PDT)

Redirect
I think a search for "Shipwrighting" should re-direct to this page. Featherfin 08:04, 20 November 2005 (PST)
 * Go for it :) --Guppymomma 14:46, 20 November 2005 (PST)

Ranks
Am I missing something, or are the ranks listed here completely out of date? As far as I know, the ranking system is the same for all the puzzles.
 * Perhaps ranks is not the best term for the combo scoring words. -- Faulkston 10:10, 7 May 2006 (PDT)
 * These aren't the combo scoring words either, unless they're different on the various oceans. The combo scoring is the of standard bingo, Donkey, Vegas, Vegas^2... type.  I'll check them now to be sure. -- Tom_the_red 18:18, 7 May 2006 (GMT)
 * Yes you're right. There are two parts to the "combo score" in Shipwrightery: 1) the combo chain number which goes how you said and 2) the grading for the sequence based on size of patterns used to make that combo chain e.g. 3-3-4-4 is a better combo than 3-4-4-3. I'll look some more for a better term. -- Faulkston 10:28, May 7, 2006
 * Let's try superlative. -- Faulkston 10:32, 7 May 2006 (PDT)
 * Well, I don't think "superlative" is really the right term here (it's the one Tedv used, but it doesn't really fit linguistically); superlative generally means "the best," so that's a slightly strange term. At any rate, the article could probably use a rewrite to incorporate the chains and the "superlatives" based on the information in Tedv's post into the scoring section (with a link to said post, even).  I had no idea that the size of the piece was so much more important than the chain number until I tried playing the puzzle after reading that post. :P --Emufarmers 11:47, 7 May 2006 (PDT)
 * I'd go with "feedback score" or "feedback phrase", but that's just me. --54x 12:39, 21 May 2007 (PDT)

New layout?
I've got a new layout for the pieces at User:Barrister/Sandbox. Comments welcome. --Barrister 00:45, 17 May 2006 (PDT)
 * It's certainly much less of a space-hog; it does cut out all those little descriptions on what the parts actually do. Would anybody miss that? (I suppose someone could always link each piece type to a Wikipedia article describing its real-life function, if we want to keep that little historical flair.) Looks good, at any rate. :) --Emufarmers 01:33, 17 May 2006 (PDT)
 * I think a link to each wikipedia page would be good, and is there anyway to center the cringle, gaff, hatch, jib, knee, and thimble? They seem forced to the left-bottom of each cell somehow. --Rixation(t/c) 06:39, 17 May 2006 (PDT)
 * I've fixed the alignments. I considered adding wiktionary (not wikipedia) links, but not all the terms were defined there.  --Barrister 16:20, 23 May 2006 (PDT)

Questioning The Scoring
Though the page speaks of gaining more points from not continuing your combos, and instead using just the larger pieces, I decided to try out the formula through a whole session, rather than just a couple of lines. For consistency I did not add in the bonus for after the 20 pieces were placed. As long as a player makes sure to end their combo on piece 19, starting again at that point to build a bonus, things should work out for the best using combos, not quick scoring. Note that both methods can get you an incredible, and the method suggested on the page is faster, most of the time, as it saves on hunting for other small piece matches.

These are the numbers from actual sessions I ran, the first using the method the page suggests, the other ignoring it, and going for max combos. Each new line marks moving pieces and start over with the combo.


 * (3x1)+(4x2)+(5x3)+(3x4)+(5x5)=63
 * (4x1)+(4x2)+(5x3)+(5x4)=47
 * (4x1)+(5x2)+(5x3)+(4x4)=45
 * (4x1)+(5x2)+(5x3)=29
 * (4x1)+(5x2)+(5x3)+(44)=45
 * Suggested Method Total is 229.


 * (3x1)+(4x2)+(5x3)+(5x4)+(3x5)+(3x6)+(3x7)+(3x8)=127
 * (3x1)+(4x2)+(5x3)+(5x4)+(3x5)+(3x6)+(3x7)= 100
 * (3x1)+(4x2)+(5x3)+(5x4)+(3x5)=61
 * Max Combo Total is 288.

If anyone has some session results that differ greatly from the above examples, I would like to see them. That is, if they show the pages suggested method to actually be a better method to use other than working on max combos. --Lanya 03:36, 23 October 2006 (PDT)


 * I wrote most of the stuff you are referring to. The scoring example in which a piece is scored simply by (number-of-squares) x (position played) was just to demonstrate that it is better to play larger pieces later in the chain than sooner. As I wrote, it doesn't matter exactly how much the scoring increases - any scoring system that does increase will have this property.


 * When it comes to working out how valuable it is to extend a chain, it suddenly becomes extremely important how much the scoring increases, so we can no longer use such a simple model. We have to look at the superlatives (or whatever we should call them) to get an idea of what is worthwhile.


 * If you've placed 6 pieces and you then have the opportunity to place a three, this will score you an 'Excellent' (not to be confused with the puzzle score of the same name). But you can score an excellent just by placing a five as the 2nd piece, or a four as a 3rd piece, and you would then be able to score a lot higher a lot easier with subsequent pieces.


 * The idea of this kind of optimisation came out of a discussion thread on the forum a while ago, and was borne out by my own experiences of the puzzle. In particular, playing it again recently, I've found if I make a lot of chains like 4, 4, 5, 5, 3 then I get an excellent; sticking to 4, 4, 5, 5 keeps it at incredible. If things work out such that you can carry it on to 4, 4, 5, 5, 3, 3, 3 as in your example then it could well be fine, but if you can't see those threes in advance then I don't think it's worth the risk. -- Ruby spoon 03:11, 29 November 2006 (PST)

Learning the puzzle messages/triggers
As I'm playing SW again for the first time in a long time, I hit an area of the puzzle that was really confusing me. When I first pulled it up today, I got the tutorial box saying that "I'm doing well, so now I get two 5-piece patterns in my toolbox. Playing yet further, I hit Solid experience, and was drawing consistent excellents. I was wondering what needed to be done to kick it up to incredible, as I was also already at grand-master, and making longer chains seemed to take a lot of luck. To cut more directly to the point, several games later, I got what I'm hoping will be the last tutorial, which tells me that more gold pieces are going to start showing up, and suddenly it's a completely different (and much more fun) game. Should there be a note explaining how the game changes as the system 'eases' you into it? I know I was surprised that I was still getting tut windows. --AtteSmythe 23:22, 15 November 2006 (PST)

Edits 3/2/09
I added the scoring table as determined by screenshots to this point. And changed the example text to the scoring system that Shuranthae determined ((size - 2) * (chain multiplier)).

I also removed the incorrect references to the extra time being a "bonus." The final score in SW is the average of all moves - the extra time would only be a bonus if you happened to do better on those moves. Drc500free 17:12, 2 March 2009 (UTC)