|Welcome Guest | Login|
|Index | Recent Threads | Register | Search | Help | | View Unanswered Threads|
|Forums » List all forums » Forum: Tips & Tricks, Questions & Answers » Thread: Ryuken's Illustrated Distilling Guide|
Thread Status: Normal
Total posts in this thread: 9
|[Add To My Favorites] [Watch this Thread] [Post new Thread]|
Joined: Oct 31, 2010
Hello. Ryuken here. I've gone from Broad/Proficient to eventually obtaining Ultimate and now #1 experience on Obsidian so... I would like to think that I have learnt enough to shed some light on this puzzle that most people deplore. Here's hoping it helps you.
much thanks to chesster, hissul, galam, ozzy, possum and evilduck for scrutinising my play as I struggled through this puzzle
psst: my first post on the forums so i'm hoping gifs and images work fine. try a quick refresh if they're laggy for you
- Goal: Put Crystal Clear (CC) chains into the keg. Nothing else.
- Black > White = Furnace (down), anything else goes into the Keg (up). Yes this includes a full column of brown (revisited in Expert section)
- Whites that go down are burnt. 1 Burn piece is generated for every 2 Whites burnt. It is not worth it to generate a burn piece (revisited in Misc. section)
- Abandon your puzzle (revisited in Misc. Section)
- Be aware of the glitch (revisited in Misc. Section)
Junk = Browns & Blacks
PureBrown = A column consisting of only Brown pieces
Earlygame = Rescuing Whites
Midgame = Playing with the white CC block
Endgame = After the CC chain has started
- Basic, Skilled and Expert sections (arbitrarily defined by me). Ideally you'd stay in that section until you've mastered the techniques and hit the consistent CC chain average of the next section before moving on. More importantly anyway, practise each technique before expecting to be able to calculate and execute it when the situation arises.
- CTRL+F Index for the Movement and Planning bits.
M1: Crawling (implicitly horizontal)
M2: Vertical Crawling
M4: Accelerated Crawling
P1: Spice Walls + Problematic Pieces
P2: Spice Traps
P3: Keeping Value Pieces
P5: Tidying the Right Side
P6: Black Tips
P7: Move Gaps
Basic (CC5 and below)
Save your Whites so you don't burn anything. Ideally, you'd repeat this over and over, only sending Junk into the furnace thereby conserving Whites for the CC chain that we're after.
A) Legal moves. (image)
Not much to be said. The connecting mini-bubbles show you available moves. White climbs into Black and falls into Brown, etc. It'll become second nature with time. A piece with more legal moves has more value.
B) Performing moves.
Please click-drag instead of click-click or keyboard. No exceptions. A common beginner-level mistake here is not dragging pieces in a straight line. Just practise more, I can't really help you there sorry.
C) Dachimpy's simulator is noteworthy, but without the time factor you won't progress much. Definitely use it while you're learning movement, but the earlier you break away from it, the better. Suffer through the puzzle once you've learnt movement. Pressure makes diamonds.
Movement: Crawling (gif) [M1]
Crawling - Movement of 2 distinct pieces through a 3rd medium.
Eg: At mid/end game, High Browns and low Blacks are problematic because they have no value in a sea of Whites. Crawling offers a simple solution by trading value between both Junk pieces, using one to carry the other. Black climbs into Brown to +1 value then falls into White to -1. Brown does the opposite, repeat.
Movement: Vertical Crawling (gif) [M2]
Once comfortable, crawling can be performed in any desired angle through a strategic mix of both horizontal and vertical crawls. As it requires a medium (sea of Whites in this example), you'd find little use for it in the earlygame. Nonetheless crawling is a tool at your disposal when the need arises. Learn to do it fast, it is SO useful.
- Your focus at this stage is merely the Whites, all the other Junk is irrelevant. It's intuitive to work from the Right and save the most immediate White.
- Move stray Whites to the left. Once all the Whites are connected, move Junk from left to the right
- Pick a friendly starting board. In the rightmost 3 columns, look for 5 Whites or less. Immediately abandon any other board (no penalty). This is especially helpful when movement calibre is poor and thought process is slow. No shame in doing so.
Objective: More efficient planning and execution than Basic. Spice Knowledge + Early detection of problematic pieces.
Movement: Rows (gif) [M3]
Rows (aka Trains) - Moving rows of piece A using 1 piece B through medium C.
Over time, you'd find that Crawling alone isn't sufficient. We cannot always have 1 Black to carry 1 Brown, etc. Moving rows maximises the utility of the value piece by carrying a chain of problematic pieces.
At this stage I just wish to point out something. If you caught on, the way i maneuvered the Brown row and the Black row were different. Apart from demonstrating how to move rows, this illustrates examples of good and bad rowing. Notice how none of the Browns were clicked when moving the Brown row?
If I clicked the Brown, a simple over-movement would cause it to slide too far and ruin the row; it is safer to drag a White because it cannot slide inappropriately regardless how poorly I flick my cursor. The Black row is the bad example.
Planning: Spice Walls + Problematic Pieces (gif) [P1]
I meant to illustrate a Spice example but as it turns out there's multiple things going on here so let's review it chronologically.
1) Noticing that my bottom-left corner is dead/problematic, I borrowed a Black from the right (see P4: Rescue). Now that the Blacks are connected, I can Row them out
2) This is a smaller trick. Borrowed a Brown from the right just to help the row move a bit quicker. Not crucial, I could've moved the row normally
3) Instead of replacing the most immediate White column, I flush the Blacks behind the Spice. It's a pre-emptive measure: It would be impossible to get behind the spice wall if it was at the end of the board, so I'd rather deal with it now. In the same way, flush Browns behind a spice wall that's at the top.
Planning: Spice Traps (image) [P2]
Planning is not the best word for it since you can't really plan for a Spice Trap. For the most part, try to Furnace all columns till you eventually burn the Trap as well. If you have to burn Whites, by all means. However, sometimes the Trap is located too deep into the CC, as follows:
Your best course of action is to drown the Trap. Never opt to send up a Trap, that wastes 7-8 Whites which is complete blasphemy in Distilling.
Your CC chain is already broken, it's better to settle for 4+TrapFurnace+8 than 4+TrapKeg+7 because your Trap went up. Or worse, 4+TrapKeg+6 because you wasted so many Whites that you can't even get 4+TK+7. TLDR; Trap = Furnace
Planning: Keeping Value Pieces (gif) [P3]
It's a good habit to store a few value pieces (Blacks at top, Browns below). This allows for easy manipulation when the spawn column gives problematic pieces like these 2 Blacks. There's no rush to move these value pieces into the last column unless you desperately need to save a White from burning, so keep them where they are.
Planning: Rescue (gif) [P4]
Perhaps you didn't keep enough value pieces, or simply none spawned. Performing a Rescue involves pulling pieces from the Junk pile on the right to extract a stuck problematic piece. Technically one Black piece here would've been enough to get them out, but I like to match the 3 Browns with an equal 3 Black, for aesthetics.
- Don't overdo the Keeping Value Pieces unless you're comfortable with it - 2 is usually sufficient. One for moving a row and one for backup.
- If you're still in the starting bits of the Midgame, you can afford to ignore one or two problematic pieces. There's a good chance the spawn can pick them up. But don't do this for too long!
- Previously, we've been picking boards with no more than 5 whites in the rightmost 3 columns. If your average CC has increased to 6ish, you can increase the margin progressively until you're comfortable regardless the number of whites
Objective: More productive and defensive play
Movement: Accelerated Crawl (gif) [M4]
Accelerated Crawling - Combination of Crawling and Keeping Value Pieces.
A high Brown-spawn that was easily dealt with via accelerated crawling. You'll find that the Blacks were given -1 value then restored back to their original position (which means they can be used for the next high Brown-spawn!)
But how do we set it up?
Sorry I don't have an eloquent way of explaining this. In essence, if a value piece spawns at the very edge, don't touch it. Alternatively, you can crawl on the edge and create your own value piece to set up the acc.crawl, as illustrated.
Movement: Combination (gif) [M5]
Not actually a technique. Just a display of movement combination. Again, chronologically:
1) Rowing the 2 Browns down, targeting the top Whites
2) Horizontal crawling to target the Black into the mid White
3) Quick circle to put the Black into place (Vertical Crawling)
4) Had a bit of leftover furnace time so I decided to crawl the Black, targeting the lowest White
5) Finish up with the Brown
On hindsight, I had enough time to rescue the other Black too, but I couldn't calculate that in time.
What if the fresh spawn had insufficient Junk? Retract the Junk from the last column.
What if you can't retract them? Then it's your fault for putting a Black at the very bottom or Brown at the very top. Which leads us to:
Planning: Tidying the Right Side (image) [P5]
For the most part, you only need to tidy up the first column of Junk. Tidying up involves ensuring every Junk single piece has value. This way, you can always retract them instead of sending up a bad column. More importantly however, they can dive back into the action and help out with misclicks/spices/anything in the middle.
But BE VERY CAREFUL with this! Tidying the right side significantly improves your gameplay by opening up more options, but tendency for PureBrown increases towards the end. Stay alert.
Planning: Black tips (image) [P6]
Avoiding the PureBrown again. Since "Black>White" condition is not fulfilled, PureBrown columns goes up and tarnishes your clean Keg.
During the midgame, it's possible to be engrossed in the action and overlook the Junk pile on the right. Tipping each Junk column with a Black safeguards against PureBrowns as those Blacks are in deadlock. The earlier you establish your columns' Black tips, the quicker you establish safety.
Planning: Move gaps (gif) [P7]
All whites are identical - Change the target instead
Here, after moving the White-spawn up, I noticed there was no easy way to fill the 3 remaining gaps. I did however have somewhat of a Brown passage for Whites to flow through, so I Instead moved the gaps towards the Brown passage.
- Tidying up additional columns with your excess time in the earlygame would allow more time during your midgame to do other operations - not a bad idea but only do this with excess time. Organising additional columns of Junk should not be a priority if your board has other problems.
- In the endgame, stuck Whites must be rescued immediately as every White counts. Connect more Whites to it and fish out the entire row at once. Borrow a White from an organised CC if you must.
- Sometimes in the endgame, there are far more Browns or Blacks. Switch gears from endgame structure pattern (if you adopted one) and recognise that this is now a Black medium, for example. Crawl and Row accordingly using your limited Browns to transport Whites over into CCs
- Fill in the CCs diligently (thanks Galam). I'm not sure why but I fail at this far too often. The next CC isn't fully constructed and I place Whites at the next column. I have no idea sometimes - just don't do it. It's not hard to fix even after the error is committed but why go through all that.
1) Abandon your puzzle. This is arguably the most important takeaway if you're not already doing it. Do your best CC chain that you can pull off, and leave the puzzle immediately. Unless it's a CC12+, abandon. You'll see the results after a few sessions.
2) Generating a Burn isn't worth it. A random number of 1-30 is rolled for each piece. 30 spawns a Spice, every other value is divided by 3. That division remainder determines the spawn for that roll, with 0 being White. That's 2.4 Whites for a short column and 2.7 Whites for a long column, statistically. Further, Burns don't "spawn", they merely take the place of a White that does spawn. Thus every Burn effectively counts for 3 Whites (the 2 you burnt to generate it, and the 1 it replaces). TLDR: Don't burn more than 1 White unless you want to gamble, or simply just want to push yourself and burnt some in the process
3) Playing from Left to Right in the earlygame. Instead of rescuing Whites from the right, one could instead focus on the Junk from the left and eradicate problematic pieces right from the get go. This does strain the nerves a bit as Whites would be close to the Furnace for a while but it makes the midgame a bit easier. Somewhat of a tactical gambit if you're confident in your piece movement.
4) Half Push your endgame. If you cannot make a Junk row (eg. Only 3 Junk pieces remaining), you could opt to move them to the centre of the board first. Depending on the next spawn, you can assess whether you have enough Junk to send another Junk row down. This relieves some time pressure as you don't have to transport every single Junk from left to right, but you lose out on potential rowing and saving problematic pieces since you have already committed some pieces to the middle. It's yet another gamble, but worth considering if you lack speed
5) Pick your starting board based on Spice count. This is for when you're at the Leg/Ult level as it seems to be a very minor effect. All spices from your starting board would burn away, and every Wasted Spice deducts from your score. Pretty self-explanatory. Nonetheless putting the score-argument aside, why would you want to play on a board that starts with 7 Spices anyway?
6) The Glitch.
This occurs when you hold a piece and literally fling it the moment a Furnace occurs (regardless automatic or triggered). You get to watch an amazing slow-mo and your piece is stripped of all its legal moves. Solution: Bring a piece near the glitch. This poses somewhat as salient stimuli which causes a graphical update which "refreshes" the piece. Being aware helps maintain composure and solve the glitch with ease.
Q&A for what I've been asked up to 21Dec
1) How do I set up accelerated crawling in the first place? Added.
2) Not enough value pieces are spawning for me to even keep. Added.
3) What those random spices in the middle of the board that block row movement.
A blend of rowing and crawling would surmount most isolated spices, as shown. Alternatively, I could've just moved the row by bringing the Black underneath instead of over it, but it didn't occur to me at the time. Regardless, no special movement is required for an isolated spice. Hope this illustration helps establish that.
4) I like to do (image), have you considered this endgame structure? Why do you feel endgame structures don't offer enough?
This was my board immediately after the CC16 went up. The speckled idea allows for more freedom for Crawling, but eliminates all opportunity for long dragging of Whites. Up to interpretation whether this counts as a structure I guess. In any case, I've reviewed my games and most of them take one of these "forms":
in an abundance of Blacks
in an abundance of Browns
Some of the structures you guys have shown me are rather interesting as well, just be mindful that you can handle whitespawns at the extreme corners.
5) Why abandon a board with >5 Whites? It's more for beginners. As you progress, abandon boards with multiple spices. Further up, look for boards without multiple spices and with LOTS of whites. The latter reduces the probability of more spices spawning since your CC Chain would start earlier.
That's about it for this write-up for now. There's a lot more that can be said and demonstrated but these would cover the main bulk of things. At the very least, hope this gives you the motivation to give Distilling another shot.
Ryuken on Obsidian (active) and Emerald (not really)
[Edit 2 times, last edit by Kyura94 at Dec 21, 2017 9:31:41 AM]
Joined: Mar 2, 2010
This is a detailed and comprehensive post on what I consider to be the most clever and most under-rated puzzle in the game. Thank you!
-Porglit on Emerald
Joined: Oct 31, 2010
Appreciate the kind words! It is indeed a terribly underrated puzzle.
Been receiving /tells in game and PMs via the forums, some of which are very aged (I am so sorry I didn't even know they were a thing, overlooked the inbox notification somehow). I take it that they would like to remain anonymous so I've just taken the general issues raised and addressed them accordingly. There were many repeats and I hope I didn't overgeneralise.
Also added more, in general. Merry X'mas!
Ryuken on Obsidian (active) and Emerald (not really)
@Ryuken Very good Distilling guide lucky I do most of these techniques myself xD
Joined: Jan 5, 2007
Hi there, I'm a Leg/Ult distiller but I thought I'd take a look at your guide; it looks very thorough and I like that you added animated gifs, that's really helpful. To learn puzzles I would always suggest watching Youtube videos, if possible with commentary ;)
Just one question: although I've been a quite active player back in the days, I've never really bothered about high-scoring techniques and tricks, in Distilling or any other puzzles.
Could you elaborate a bit more about this part? If I get only a CC10 or CC11, is it better to leave the puzzle after getting the last CC, than to send down the last row and wait for the DR?
Barbecul, on Emerald.
Joined: May 26, 2006
Yes. If I remember it right, it's because your score is based on the average of rows you send instead of total score? Any CC6+ -> Abandon is better than finishing the session.
TriplePat, Joining the great obsidian migration.
Joined: Jan 17, 2006
You have to send up one BIG row with whites and spices. There is no real average of rows. Abandon CC6+ gives a small plus for lower levels, up to master level. If you go for real high levels like leg/ult on elite oceans, these small plus is not important because you have to play CC12+.
Platia - Captain of Kraken Hunt - Monarch of Kraken Down (Emerald)
Platia - Senior Officer of Lords of Darkness - Monarch of Dark Shadows (Opal)
Platia on all other oceans!
Joined: Oct 31, 2010
Empirical evidence is consistent with this idea, and I pretty much just accepted Dismissing as the way to go. I've recently learnt of the underlying rationale however, which is briefly summarised by Pat above. Turns out Dismissing isn't always correct.
A simplified model would be:
CC1 column = 1point, CC2 column = 2points, etc.
The averaging would work out to give:
Pulling off a CC1 = 1point / 1col = 1 Score
Pulling off a CC2 = (1+2)points / 2cols = 1.5 Score
Pulling off CC10 then CC2 = (55+3)points / 12cols = 4.8333 Score (same as doing CC2 then CC10)
I'll skip the math:
The blue points assume you complete a full keg of 12 cols, which is the only way to get the DR you mentioned
So yes pat is correct, CC6+ -> Dismiss is better.
Note: criticising the model, it's more likely that the scores increase, increasingly. A Vegas scores significantly better than a Donkey, whereas a Triple does little more than a Double. Hence the point gain from CC5 to CC6 should intuitively be higher than CC2 to CC3. Nonetheless, applying an exponential model to the points reaches the same conclusion pat outlined.
Further, there are other qualitative factors to consider. Suppose you get a CC6 and chose to continue:
1) May not get enough White spawns to get another CC6
2) Failed pushes (Blecch! / Smooth) when setting up the next CC chain
3) Spice Trap spawn
4) More Wasted Spice! instances which deduct from your score
5) More time spent (Lag / Disconnects / Server Reboots / Competition Distilling)
This does point to something else however - a small flaw in what I initially wrote. Always abandoning after the first chain is wrong: a CC3 should definitely be followed up with a CC9 (eg. Spice Trap forced a CC3). But if a Trap forces a CC5... then it's complicated. You'll have to weigh the score boost against the factors above.
Ryuken on Obsidian (active) and Emerald (not really)
Joined: Jan 5, 2007
I see, thanks a lot for the explanation!
Barbecul, on Emerald.
|[Show Printable Version of Thread] [Post new Thread]|