Asenna Steam
kirjaudu sisään
|
kieli
简体中文 (yksinkertaistettu kiina)
繁體中文 (perinteinen kiina)
日本語 (japani)
한국어 (korea)
ไทย (thai)
български (bulgaria)
Čeština (tšekki)
Dansk (tanska)
Deutsch (saksa)
English (englanti)
Español – España (espanja – Espanja)
Español – Latinoamérica (espanja – Lat. Am.)
Ελληνικά (kreikka)
Français (ranska)
Italiano (italia)
Bahasa Indonesia (indonesia)
Magyar (unkari)
Nederlands (hollanti)
Norsk (norja)
Polski (puola)
Português (portugali – Portugali)
Português – Brasil (portugali – Brasilia)
Română (romania)
Русский (venäjä)
Svenska (ruotsi)
Türkçe (turkki)
Tiếng Việt (vietnam)
Українська (ukraina)
Ilmoita käännösongelmasta
Keskusteluiden säännöt ja ohjeistukset
2 
Ilmoita tämä viesti
It is extremely useful to input an atom inside a molecule in this puzzle(or the other way around), and preferably straight onto bonders as well.
If you have been worrying about waste or output inbalance, don't. If it is easier to output 2 of one molecule for every 1 of the others just do that, no harm in that.
Also consider using (partial) passthrough reactors. So a reactor does a thing, and then inputs and passes on stuff to another reactor; this can save reactors compared to splitting the stream early. Of course this is most useful when the reactors normal job only uses one output.
Wait! You mean: I can build up the outer shell, and then insert the center guy via an input?! That's just... cheating...? :)
1st reaction seperates the oxygen atoms and recycles them and sends the silicon onward.
2nd reactor: sends the silicon to one of two different reactors. In hindsight, this reactor isn't strictly necessary, and the 50/50 ratio I use isn't efficient.
The next three reactors assemble each molecule. Reactor 3 just does silicon, reactor 4 adds the boron and passes on the extra silicon atom, and reactor 5 assembles the arsenic.
This leads to a fairly middle-of-the-road solution.
Please don't tell me how, but you're saying that in Reactor #3 I should be able to construct a 3x3 silicon lattice for output from a 1-square-input silicon atom?
Yup, that's correct. Try to compartmentalize your approach to the final molecule. The final result is fairly simple. One bad habit I have with Zachtronics games is sticking with a very complicated approach, rather than starting over and trying a new angle at the problem, but this is one puzzle where I used a straightforward design all the way through.
Lol. Thank you. I think it's "impossible" to build a 3x3 in one go out of a 1-input, so clearly I need to think hard about this :) BTW, you do indeed construct a 3x3 == 9, not a 3x3 with a hole in the middle == 8, right?
(So far my approach had been to build 1 -> 3-in-a-line in one reactor, then 1 x 3-in-a-line -> 3x3 in the next reactor. And even then I ended up with the 3x3 only being able to come out of a single output, so then it would need splitting to multiple outputs later. Hence my need for about 10 total reactors!)
Yes, you have to build a complete 3x3 square. Without telling you how, I think it's clear that in order to meet the reactor limit you'll have to assemble the 3x3 square inside a single reactor. Injecting the the other atoms for the other two molecules is another challenge to overcome. At the moment, I forget exactly how I did it, especially since there's a certain command which I don't think is available for this level that makes things much easier.
Edit: Ah, actually getting the extra atom into the middle also have a fairly straightforward solution, too.
I'm just checking: I hope you're not a really mean person who knows that it cannot actually be done in one reactor, and have now set me off to live in torment racking my brain in vain to find a solution....
How'd it go, did you figure out a solution?
I hope you haven't been drive mad yet.
You say you can take a stream of element input from one square, and build a 3x3 out of it. I say poppycock. Let alone whether you can send it to more than one output.
The only approach I see is:
* I can build 3-in-a-row by "shunting" with 1 waldo. However note that to terminate the row I require a Sensor.
* I could adapt that by using 2nd waldo to "pull down" row on completion. Binding with row above beforehand. I'm sure this will require > 4 Melders. Which I can't have in a Sensor unit. Let alone, how I'll recognise to terminate the "move rows down".
And all of this is just for 1 Reactor, I've still got to use output to replace atom in center later.
I say this is ridiculously hard. I don't even know whether you're having me on and this is impossible anyway...
http://steamcommunity.com/sharedfiles/filedetails/?id=1140040541
Replacing the atom in the center makes it slightly more complicated, but not by much.
This was actuallly the last puzzle I completed in the game. I put it off until I'd already beaten it. When Zachtronics games get tough I remind myself of two things:
1) Even the creater of the game didn't beat the last levels
2) No matter how tough a puzzle seems, plenty of people out there have not only beaten it, but found solutions much better than I can come up with.
3) It's often best to delete a failing attempt and start over (I don't always follow this, to my detriment).
Without going into details, I created a factory that constructs the final product layer-by-layer.
It's somewhat innefficient and used all 5 reactors, but it got the job done.
To be slightly more detailed:
I build up the molecule in 3 parts.
One reactor sends in streams of Si-Si-Si, while the other input is alternating As and B.
The builder reactor then alternates between just building an all-silicon fragment, and removing one atom from the second layer to insert one of the other atoms.