A catalyst is a stable object (usually a still life) which can interact with an active region and later return to its original state. Catalysts are very useful in conduits and in many other applications. Here is an example of a stable catalyst:
The green still life interacts with the white object, and recovers to its original state by generation 15. In this tutorial, you'll learn the main types of catalyses and their applications. But first, we need to cover some important definitions:
Stable catalyst: a catalyst which is a still life. Most catalyses are of this form. In this tutorial we will only focus on stable catalyses, as periodic catalyses are as of now quite underdeveloped.
Periodic catalyst: a catalyst which is an oscillator. These are generally considered inferior to stable catalyses because they usually only work in a fraction of the generations as a stable catalyst, since they usually only work in one phase of the periodic catalyst.
Transparent catalyst or transparent object: an object which is destroyed completely (all of its cells being off at the same time) then restored (in the same position and with the same orientation)
Pseudo-transparent catalyst: a catalyst that loses all of its cells but not at the same time (i.e. for each generation, at least one of its cells is alive)
Recovery time: the time it takes for a catalyst to return back to its state before interacting. In the example above, the recovery time is 15.
Replacement catalyst: When a common catalyst fails, another, less common, catalyst must be used.
Rock: A catalyst which does not lose any of its cells, even temporarily, during the interaction process. The snake-type catalyses below are examples of rocks.
Let's explore the different types of catalysts now!
The marked white cells should always be on in a fishhook-type catalyst. Sometimes, the fishhook may engage in a more unusual manner, but these are rarer:
Recognizing when to use a fishhook catalysis is really a matter of experience. For practice, try placing an eater 1 somewhere within this blue field to eat the middleweight spaceship (click on the thumbnail below to draw cells in LifeViewer using the pencil icon):
They typically occur when there is a two cell leading edge, which is marked. Boat catalyses fail very often, so see the "Replacement catalysts" section below for potential replacement catalysts. As a general rule, boats are most likely to succeed when a cell adjacent to the leading edge (shown in blue) will turn on in the following generation.
There is a separate boat catalysis that works on objects with a line of three or more cells, but it's important here that the red cell does not turn on:
Snake catalyses are examples of rocks, or catalysts which do not change even temporarily when interacting with an object. They must function this way due to the fact that the two cells in the center would die from overpopulation due to already having three live neighbors.
Three example snake catalysts are shown below. The white cells should always be on to trigger the important next generation, where the snake functions as a rock:
Conduit 1 is an example of a conduit using a snake catalyst. Try to catalyze the reaction below with a snake in the blue field (as a hint, the cell which should trigger the rock interaction is marked):
Tub catalyses can be tricky to find, but they are most likely to work when the frontend of the active region is two or three cells long, like one of the examples below:
The reactions start differently but end up the same way through the same general mechanism:
On generation 1, one cell (marked in white) is born next to the tub;
On generation 2, two more cells are born next to the tub. This causes the three white cells to die;
By generation 3, the cells in the center have died and what remains is a 5-cell parent of the tub. In all tub catalyses, the tub should turn into this 5-cell parent.
The ship is turned into a relative of the I-heptomino (specifically a great-uncle) then is turned back into a ship by a perfectly-timed temporary induction. The ship rarely recovers by its own, but it's worth trying because it has replacement catalysts that recover much more often.
Replacement catalyses
So, you've tried to catalyze a reaction, but the common catalyst fails! Luckily for you, this is not a complete disaster, since many replacement catalysts exist exactly to assist you when your fishhook or block doesn't just work out.
Fishhook replacements
The most common way for a fishhook to fail is something like this:
This was a bit cheating, however. Sometimes, we need a genuinely new catalyst. Here are two almost-catalyses where the fishhook almost recovers, but an extra spark destroys it or an extra cell gets attached to its recovering phase:
The most common replacement catalyst for a fishhook is the handy eater 2. An eater 2 recovers differently from an eater 1, making it more versatile (and its symmetry means that it can eat from two directions). It can handle both of our problems:
This trick works in a variety of circumstances where a block fails because unwanted cells are born to one side. Also, the fishhook can often be replaced by a block. This sometimes works even if a fishhook doesn't work.
Also, if the fishhook recovers too quickly but otherwise works, then like other examples where a fishhook recovers too quickly but otherwise works, it can sometimes be replaced with an eater 2.
Another example where a block catalyst can be fixed by supporting the block with another catalyst comes from the following example, where the block is turned into a first cousin of the century then crashes into a boat.
This trick only works if there is something for the century cousin to crash into (and even then, not always), but luckily, there is an actual replacement catalyst for when the block is turned into a century cousin with nothing to hit.
It should be noted that when a block fails in one of the most recent two ways, it can often be replaced with a hook with tail (although the hook with tail is not a replacement catalyst for the block in the strictest sense).
In addition, there is a third way in which a block is sometimes turned into a P-pentomino. In this case, it can be replaced with a claw with tail supported by a fishhook.
For one particular type of block catalysis (specifically where the block is turned into a grin but not through the six-cell grin parent in Pentoad 2), the block can be replaced with a long hook with tail. For example, if a block is used with the eater 5 (which will be explained next) in this case, the chaos hits and destroys it (then destroys the tub with tail).
In this example, the white cell at generation 4 gets turned on when it shouldn't. We have a fix for this, as usual. Replace the boat with a tub with tail and block (this catalyst is called an eater 5 or tub-with-tail eater:
Sometimes, the snake needs to be replaced them with a similar catalyst which does the same thing but might have better clearance. For example, if a snake is required to be in close proximity to a fishhook, the fishhook's interaction might destroy the snake (and subsequently the fishhook). In this case, replacing the snake with a higher-clearance second fishhook can prevent the interaction:
A problem that sometimes occurs with snake catalyses is that a certain cell above the snake (marked in red) will be born when the snake is supposed to be done interacting. When this happens, try replacing the snake with a hook with tail.
Sometimes, a single dot at the end of the tub parent is replaced with a line of three cells, causing the tub to turn into a block instead of recovering:
The ship family is fairly unique as a catalyst family. For most catalyst families, one places the namesake catalyst expecting it to work, and the replacement catalysts are reserves. However, a ship will almost always fail when attempting a ship-type catalysis, and it is much more often the case that one of the replacement catalysts works. However, I think that it is worth initially placing lone ships (as opposed to initially placing a replacement catalyst) for three reasons:
It takes less time to draw a ship than to draw any of the replacement catalysts.
It is easier to see which replacement catalyst, if any, will work by observing a lone ship than by observing any single replacement catalyst.
A lone ship will occasionally work, and I don't want to miss when that happens. (If a lone ship works, none of the replacement catalysts will work.)
Often, the ship is turned into a loaf. Adding a fishhook turns it back into a ship:
If the loaf parent has an extra cell in a certain position (so that it becomes a great-uncle of the I-heptomino), there's another replacement catalyst that works (which was discovered fairly recently to use in the R49 conduit, so it is often called the "R49 catalyst"):
Lone ships, ship-fishhook pairs, and R49 catalysts are more likely to work when the cell that "activates" them survives into the next generation, while a fishhook bridge fishhook catalyst is more likely to work if by the next generation, the cell that "activated" it has died, and it has a two-cell edge to "slam" into. Common examples of the latter are a grin (like in 30P6.1) and a certain edgeshootable beehive predecessor.
For this reason, as well as the fact that a ship takes less time to draw than a fishhook bridge fishhook (and I can still tell whether or not a fishhook bridge fishhook will work by observing a ship), I place a long ship instead of a fishhook bridge fishhook even in situations that look like a fishhook bridge fishhook is the most likely ship-type catalyst to work.
Occassionally, instead of simply becoming the aforementioned blockade predecessor, the ship will become that blockade predecessor with three additional cells, and if one uses a fishhook bridge fishhook, it will mostly recover, but there will be three extra cells in the same position. When this happens (One only needs to observe this for one of those two catalysts.), a beehive works.
This is quite inconvenient for us because we can't use the loaf again (unless we flip it back around with another glider as shown in the viewer, which is unlikely), but a stable catalyst does exist to turn the loaf back around. Here is the eater 3:
The eater 3 has some other applications besides eating gliders, for example the rectifier. I typically use the eater three when a fishhook fails on a corner (in at least one orientation), and there aren't any replacement catalysts that work.
This may seem pretty niche, but it is actually a fairly common application.
Loaf
Besides its usage in the eater 3, the loaf has a pseudo-transparent catalysis that can eat a certain beehive predecessor or two block predecessors that are similar to each other.
This can also come in handy sometimes; for example the recently-discovered L122 conduit.
In order for the eater 4 to recover, it needs to stay diagonally symmetric. If there is a line of two on each side, it will receive dot sparks on both sides, stay symmetric, and recover properly.
However, if there is a line of two on one side and a line of three or more on the other side, it will receive a dot spark on one side but not the other, so it will lose its symmetry and fail.
Mike Playle's Bellman is a search program specifically designed to search for new catalysts, given a reaction to be perturbed. It discovered the Snark, as well as many other highly notable patterns. This part of the tutorial hasn't been written yet, so consult the discussion thread and original documentation.
Another catalyst finding program is CatForce, which is a brute force placement program of certain given catalysts in a given area, and can also find transparent catalysts if they're restored quickly enough. It has also discovered quite a few interesting things, like B60.