RCT remaining tasks

Pavgran · Post by **Pavgran** » October 31st, 2022, 9:35 am

calcyman wrote: ↑
October 31st, 2022, 9:01 am
I think that it's helpful to rewrite the last few steps, to handle the worst-case scenario where the ATBC is a maximum-speed spacefiller and to ensure that it ends up near the epicentre (for aesthetic purposes, and so that the ATBC is within Golly's 10^9-cell selection limit)

I also like the idea of epicentre-located ATBC.
However, I slightly dislike that ATBC starts coexisting with something else, namely SE-directed BSRD-deleting glider.
I propose the following modification to the calcyman's modification:

(27) NOT DONE: Far-southeast pseudo-BSRD's self-desturct mechanism emits an epicentre-directed glider. That glider travels back to the epicentre and triggers the ATBC seed.

Also, I think that's not the first time that a question of "where to put ATBC" has popped up.

Edit:

dvgrn wrote: ↑
October 31st, 2022, 9:30 am
If we really want the ATBC at the epicentre, then we need a small adjustment to the BSRD seed to send a glider back to the center to trigger the ATBC seed. That's awkward, because Pavgran has already calculated the number of gliders for the N-glider absorber.

Well, small modifications to the BSRD seed are entirely possible. Even large adjustments are not a major problem, as recompiling the seed and N-glider absorber is quite a small task.

dvgrn wrote: ↑
October 31st, 2022, 9:30 am
-- we really just have to have the ECCA fire one single slow^2 glider southeastward to trigger the pseudo-BSRD cleanup -- early on, at some opportune time probably when the slow^2 salvo for the far northwest is being sent out, so an elbow will be in the right vicinity.

But that change would definitely require seed recompilation, because the current input self-destruct glider's lane is close to the ECCA's elbow lane. So either new OTTs has to be constructed or existing should be recompiled. I'm fine with the latter option.

Post by **dvgrn** » October 31st, 2022, 9:40 am

Pavgran wrote: ↑
October 31st, 2022, 9:35 am
I propose the following modification to the calcyman's modification:

(27) NOT DONE: Far-southeast pseudo-BSRD's self-desturct mechanism emits an epicentre-directed glider. That glider travels back to the epicentre and triggers the ATBC seed.

Also, I think that's not the first time that a question of "where to put ATBC" has popped up.

Ha, we were both writing the same objection at the same time (see previous page).

My alternate suggestion would be to leave the pseudo-BSRD's self-destruct mechanism alone, since you've already done it and calculated the "N" for the N-glider absorber mechanism and so on.

Instead, add two still lifes to the last Corderabsorber in the far northwest, to send a glider back to the epicentre to trigger the ATBC seed. (The recipe for that far-northwest construction is still a NOT DONE item, (23).)

There will definitely be nothing left in the universe except the ATBC seed, by the time that signal gets back to the center -- even if we keep the current plan of waiting unnecessarily long to trigger the pseudo-BSRD cleanup. (But why not clean up the pseudo-BSRD as soon as we can, really?)

I had suggested moving the ATBC seed to the southwest corner because that way (if the pseudo-BSRD cleanup was done "properly") the ATBC construction could begin right away, just as soon as the last cleanup was complete, with no unnecessary four-aeon-long delays. But I do agree that having all the "good stuff" in the center is more important than getting the job done slightly sooner.

Post by **calcyman** » October 31st, 2022, 2:16 pm

Pavgran wrote: ↑
October 31st, 2022, 9:35 am

dvgrn wrote: ↑
October 31st, 2022, 9:30 am
-- we really just have to have the ECCA fire one single slow^2 glider southeastward to trigger the pseudo-BSRD cleanup -- early on, at some opportune time probably when the slow^2 salvo for the far northwest is being sent out, so an elbow will be in the right vicinity.
But that change would definitely require seed recompilation, because the current input self-destruct glider's lane is close to the ECCA's elbow lane. So either new OTTs has to be constructed or existing should be recompiled. I'm fine with the latter option.

It wouldn't require seed recompilation at all! We can take advantage of the fact that we haven't built the ECCA's one-time turners yet, and just attach a one-time turner that taps off the first input carrier glider (arriving from the BSRD) into the ECCA, and reroutes it to point straight back to the BSRD to destroy it.

Pavgran · Post by **Pavgran** » November 1st, 2022, 9:45 am

(103) DONE

I've added compile_ecca_recipe function to the repository and tested it on several recipes.
Now, ECCA operations can be added in without additional effort.
There is a possible discrepancy in even/odd parity/lane convention, but it could be changed, if necessary. Although care should be taken so that every recipe uses the same convention.

Post by **calcyman** » November 2nd, 2022, 8:19 am

I've almost finished task (5).

Quick question: based on the following comments (which I totally agree with by the way)...

Pavgran wrote: ↑
October 31st, 2022, 9:35 am
I also like the idea of epicentre-located ATBC.
However, I slightly dislike that ATBC starts coexisting with something else, namely SE-directed BSRD-deleting glider.

dvgrn wrote: ↑
October 31st, 2022, 9:30 am
The advertising for the RCT clearly states that it is capable of constructing anything that can be constructed with any number of gliders. There's nothing in that description that says that when X gets' constructed, there can also be some unimaginably distant garbage in the universe that we're supposed to ignore because it will be gone after an unimaginably long time.

...I conclude that the ECCA's elbow in the far north-west will be destroyed by some temporary circuitry in the north-west, and not destroyed by a glider from the ECCA's self-destruction? In other words, at the time that the ECCA's seeds of destruction are activated, the elbow block (and indeed everything else away from the epicentre) has already disappeared?

Pavgran · Post by **Pavgran** » November 2nd, 2022, 9:09 am

calcyman wrote: ↑
November 2nd, 2022, 8:19 am
Quick question: based on the following comments (which I totally agree with by the way)...

<more quotes>

...I conclude that the ECCA's elbow in the far north-west will be destroyed by some temporary circuitry in the north-west, and not destroyed by a glider from the ECCA's self-destruction? In other words, at the time that the ECCA's seeds of destruction are activated, the elbow block (and indeed everything else away from the epicentre) has already disappeared?

Yes, that should be the case. I think point (23) should be modified slightly to reflect that. Something along these lines:

(23) NOT DONE: Using slow^1 gliders, the ECCA builds the NW bank of Corderabsorbers (the approaching NW-directed Corderships have already been built, but they're safely to the southwest of the construction arm), the last of which will trigger a small seed when it absorbs the Cordership, resulting in two gliders: one is directed SE back to the epicentre, and one destroys ECCA elbow, as it's no longer needed.

Post by **dvgrn** » November 2nd, 2022, 9:41 am

Pavgran wrote: ↑
November 2nd, 2022, 9:09 am
Yes, that should be the case. I think point (23) should be modified slightly to reflect that.

Agreed, and done! I've moved some of the pieces around a little bit, to get everything in strict chronological order (I think).

Pavgran · Post by **Pavgran** » November 2nd, 2022, 12:20 pm

(15) DONE

I've compiled simeks's recipe, slightly changing it and adding several gliders to clean the rest of aperiodic ash. The center now looks like this: