Page 13 of 58
Re: The Hunting of the New Herschel Conduits
Posted: May 3rd, 2015, 8:51 pm
by dvgrn
simsim314 wrote:1. This is not a basic conduit - but rather uses some nice not trivial trick to combine two conduits. Two other conduits can be trivially combined to give the same result, so I would suggest to paint it yellow...
2. H->B that uses the first natural B from Herschel (like in the conduit above), is not spartan in your collection. It does the same as the newly spartan finding...
4. G-> G4 for some reason uses state3 of LifeHistory instead of state1 like all other conduits. 5. I've posted new Pi->QB see few messages above...
6. I'm very close to find something that works for QB (some annoying debris are always left, but this is just matter of time until we find some normal QB conduit), so please keep the track for those.
7. I'm somehow missing silver reflector. It should be considered basic spartan G->H at least for now.
Fixed, fixed, explained below, fixed, included, and fixed. One of your H-to-QBs was really a composite H-to-pi and pi-to-QB, so I moved that one. Silver reflectors aren't really elementary, but it's a pain to reconstruct them from elementary pieces, so I threw in the standard G-to-3G, the 575-tick Silver G-to-H, and a tricky handy 497-tick G-to-H variant.
The white state-3 H-to-G4s are definitely not elementary conduits -- there's a trivial prefixed H-to-B -- so they don't really belong in this collection. But they're the most useful Spartan transmitter/merge/splitter circuits in the toolkit, so I wanted them in the stamp collection anyway, right there in the row of H-to-G(small n).
Unlike the rest of the yellow stuff, the H-to-G4s are
not particularly oversized, so a different color seemed like a good idea. Whoever takes over the maintenance of the collection can change it, I don't mind (I've got my own copy the way I want it now...!)
simsim314 wrote:3. Why did you stopped at H->G8? There were few pretty cool above G8. I had one very simple G12. You posted G18->G so why stop at G8?
I included the G12 and another similar one in the plain H-to-G collection, blocking off the first natural glider. At around G10 the average complexity of the conduits started going up, and I didn't feel like making any more labels...!
I included H-to-G(n) for single-digit n, just as samples of what will eventually be a much larger collection that I don't want to include here. As a rough rule of thumb, I think H-to-G(n) should be included here mostly when a G(n) is a known input to another conduit. So I would include a G13 now if we had one, but we don't particularly need a G12 at the moment. Paul Callahan's G18 seems like a special case of a really useful piece of circuitry, and there's no other obvious place to put it.
Otherwise, all of these H-to-G(n) will end up in a separate
H-to-G(n) converter collection, whenever I get around to that (at this rate it may be a while...) It will include both elementary and composite conduits, so I won't have to worry about whether they should count as elementary or not.
Re: The Hunting of the New Herschel Conduits
Posted: May 3rd, 2015, 9:18 pm
by simsim314
@dvgrn - thx for this great compilation.
Maybe someone wants to try and make those full circuits:
Code: Select all
x = 67, y = 24, rule = LifeHistory
6.C42.C$6.3C40.3C$9.C42.C$8.2C41.2C4$20.2C$20.2C$63.2C$63.C.C$65.C$
65.2C$10.3C40.3C$9.C3.C38.C3.C$9.C3.C38.C3.C$7.2C.3C.2C34.2C.3C.2C$6.
C.C5.C.C32.C.C5.C.C$6.C9.C32.C9.C$3.C2.2C.C3.C.2C29.C2.2C.C3.C.2C$.3C
4.2C3.2C4.2C23.3C4.2C3.2C4.2C$C18.C.C21.C18.C.C$2C19.C21.2C19.C$21.2C
41.2C!
Re: The Hunting of the New Herschel Conduits
Posted: May 3rd, 2015, 9:52 pm
by Extrementhusiast
Scorbie wrote:Can somebody clean up? Catgl failed to remove that block. Maybe a transparent catalyst could clean them up.
Oh wait. Forgot to say that I only used spartan catalysts.
Code: Select all
x = 26, y = 36, rule = B3/S23
19b2o$19b2o8$5b2o$4bo2bo$5b2o3$2b2o$3bo$3o$o9$21b2o$21b2o3$22bo$21bobo
$20b2ob2o$21bo2bo$21bo3bo$21bo!
Not exactly what you're looking for, but that beehive can actually be made doubly transparent:
Code: Select all
x = 36, y = 27, rule = B3/S23
5b2o$4bo2bo$5b2o2$34b2o$2b2o30b2o$3bo$3o$o3$6b2o$5bobo$5bo$4b2o3$21b2o
$21b2o3$22bo$21bobo$20b2ob2o$21bo2bo$21bo3bo$21bo!
Re: The Hunting of the New Herschel Conduits
Posted: May 3rd, 2015, 9:54 pm
by Sokwe
dvgrn wrote:It sure looks like a nice new converter -- HR117B.... However, now when I try to give the output B some kind of future by connecting it up with something, I keep running into problems.... Anyone have a good workaround, or another connection that works?
The following transparent block reaction has the potential to follow this conduit (shown here as a B->G):
Code: Select all
x = 57, y = 50, rule = LifeHistory
31.2A$22.2A7.2A$22.2A$7.2A$8.A$8.A.A$9.2A7$9.C$9.C.C$9.3C$11.C17.2D2.
D$30.3D$31.D3$2.2A13.A.2A4.2A$3.A11.3A.A5.2A$3A11.A4.A.A$A13.A.A.A.2A
$15.2A.A$18.A$18.2A$16.2A$17.A$17.A.A$18.2A35.2D$54.D.D$56.D5$19.2A$
18.A.A$18.A$17.2A2$28.2A$28.A.A$29.A$44.2A$44.A$45.3A$47.A!
This transparent block reaction can follow most X->B converters, but unfortunately we don't yet have a nice way to get a non-glider signal from it. There is this bulky B->R->H, but it's too big to be practical:
Code: Select all
x = 81, y = 70, rule = LifeHistory
46.2A$37.2A7.2A$37.2A$22.2A$23.A$23.A.A$24.2A6$60.2A$24.C36.A$24.C.C
34.A.2A$24.3C33.2A.A2.A$26.C17.2D2.D16.2A$45.3D12.3A$46.D13.A2.A$62.
2A2$17.2A10.A2.A.2A4.2A$18.A10.4A.A5.2A30.2A$15.3A16.A.A34.A2.A$15.A
15.2A2.2A34.A.A$31.A40.A$32.A$31.2A35.D$66.3D$67.D2$29.2A$30.A$30.A.A
$31.2A4$29.2A$29.2A4$74.2A$74.A.A2.2A$76.A2.A$76.2A.A$3.A62.3D10.2A$
2.A.A63.D10.A$2.A.A62.3D7.A.A$.2A.2A16.2A53.2A$22.2A44.2A$.2A.2A62.2A
$2.A.2A34.2A$A39.A.A$2A40.A$42.2A2$8.2A$2.2A4.2A$.A.A27.2A$.A29.A.A$
2A31.A$8.2A23.2A$8.2A3$14.A$13.A.A$14.A!
Re: The Hunting of the New Herschel Conduits
Posted: May 3rd, 2015, 10:11 pm
by Extrementhusiast
Sokwe wrote:dvgrn wrote:It sure looks like a nice new converter -- HR117B.... However, now when I try to give the output B some kind of future by connecting it up with something, I keep running into problems.... Anyone have a good workaround, or another connection that works?
The following transparent block reaction has the potential to follow this conduit (shown here as a B->G):
This transparent block reaction can follow most X->B converters, but unfortunately we don't yet have a nice way to get a non-glider signal from it.
Could this then be a possible partial to getting an LWSS out of it?
Code: Select all
x = 34, y = 43, rule = LifeHistory
31.2A$22.2A7.2A$22.2A$7.2A$8.A$8.A.A$9.2A7$9.C$9.C.C$9.3C$11.C17.2D2.
D$30.3D$31.D3$2.2A10.A2.A.2A4.2C$3.A10.4A.A5.2C$3A16.A.A$A15.2A2.2A$
16.A$17.A$16.2A14$26.2A$26.2A!
Re: The Hunting of the New Herschel Conduits
Posted: May 4th, 2015, 1:19 am
by Kazyan
Gah, I wish this worked, but the leftover loaf and the fact that the FNG kills this is unfun.
Code: Select all
x = 37, y = 29, rule = LifeHistory
19.3D$20.D$18.3D2$28.2A$28.2A5$32.2A$32.2A5$3.2A29.2A$3.A2.A26.A2.A$
4.3A27.2A2$2A.A.2A$A.2A.A$5.A$5.2A3$11.3A$12.A$12.3A!
Re: The Hunting of the New Herschel Conduits
Posted: May 4th, 2015, 8:39 am
by dvgrn
Kazyan wrote:Gah, I wish this worked, but the leftover loaf and the fact that the FNG kills this is unfun.
Hmm, yes, it's impressive how that loaf crawls back to that location without ever leaving its reaction envelope, so there's not much hope of getting hold of it, unless some improbable spark adjustment to the north can be made to leave a block-or-something that happens to suppress it.
It would be a slow recovery, but there might end up being a left turn that knocks out the loaf -- at least there are cleanup lanes available, one of each color (this one, and three lanes to the right):
Code: Select all
x = 81, y = 109, rule = LifeHistory
60.3D$61.D$59.3D6$67.2A$67.A.A$69.A$69.2A7$59.2A$59.2A2$50.2A.A$50.A.
2A$69.2A$69.2A$79.2A$79.A$77.A.A$77.2A9$73.2A$73.A.A$75.A$75.2A$31.2A
$31.A.A$33.A4.2A$29.4A.2A2.A2.A$29.A2.A.A.A.A.2A$32.A.A.A.A19.2A$33.
2A.A.A19.2A$37.A28.2A$57.3D6.A$23.2A33.D8.3A$24.A7.2A22.3D10.A$24.A.A
5.2A38.D$25.2A39.2A2.D$66.2A6.D$72.D$76.D$74.D2$70.2A$35.2A33.2A$35.A
$36.3A$38.A2$41.2A29.2A$41.A2.A26.A2.A$42.3A27.2A2$38.2A.A.2A$38.A.2A
.A$43.A$3.2A38.2A$4.A$2.A$2.5A14.2A26.3C$7.A13.A28.C$4.3A12.A.A28.3C$
3.A15.2A$3.4A$.2A3.A3.2A$A2.3A4.2A$2A.A$3.A$3.2A2$29.A$11.2A16.3A$12.
A19.A$9.3A19.2A$9.A6$41.2A$34.2A5.A.A$34.2A7.A$43.2A2$30.A$29.A.A.2A$
29.A.A.A.A$26.2A.A.A.A.A2.A$26.A2.A2.2A.4A$28.2A4.A$34.A.A$35.2A!
What the...? But -- but these things
never work out when you try them!
Code: Select all
#C R321, repeat time 441 ticks (ouch) or various overclocked periods
x = 67, y = 61, rule = LifeHistory
60.A$58.3A$57.A$20.A26.2A8.2A$19.A.A26.A$19.A.A26.A.A$20.A5.2A21.2A$
26.2A35.2A$35.2A26.2A$36.A$32.2A2.A.A$32.2A3.2A$63.2A$63.2A5$39.2A$
39.2A$65.2A$65.A$63.A.A$63.2A2$9.C$9.C.C$9.3C$11.C5$2.2A10.A2.A.2A$3.
A10.4A.A25.2A.A$3A16.A.A23.A.2A$A15.2A2.2A$16.A$17.A43.2A.A$16.2A43.A
.2A2$54.2A$54.2A16$53.3C$53.C$52.2CD!
#C [[ THUMBNAIL AUTOSTART LOOP 600 ]]
Well, so it won't set any records. But we can celebrate with a p154 gun anyway:
Code:
Select all
x = 81, y = 98, rule = B3/S23
7bo$7b3o$10bo$9b2o8b2o26bo$19bo26bobo$17bobo26bobo$17b2o21b2o5bo$3b2o
35b2o$3b2o26b2o$31bo$29bobo2b2o$29b2o3b2o$3b2o$3b2o4$36b3o3bo$27b2o5b
5o2bobo28b2o$27b2o4b2o8bo2bo25b2o$b2o31b2o4b3o2bobo$2bo32bo3b2o4bobo$
2bobo41bo31b2o$3b2o73b2o$74b2o$74b2o2$47bo$9b3o34bobo$9bo35bo2bo30b2o$
10bo35b2o31b2o2$70bo$47b2obo2bo15b3o$19bob2o25bob4o14b2o2bo$19b2obo23b
obo23bo$13bo32b2o2b2o19bo$12b4o35bo15bo$3bob2o4bo3b2o33bo16bobo$3b2obo
5b5o14b2o17b2o15b2o$13b2o17b2o$13bo17bo$13bo5$80bo$78bobo$79b2o4$5bo$
5b3o$8bo58b2o$7b2o58b2o2$29b2o43bob2o$30bo43b2obo$29bo$29b2o2b2o$32bob
o23bob2o$27b4obo25b2obo$27bo2bob2o3$2o18bo$2o17b2o$19bobo2$32bo$5b2o
26b2o$5b2o21b2o2b3o$b2o25b2o5bo40b2o$b2o26bo4bo41bobo$30bo3bo30b2o11bo
$24b2o6bo31bob2o10b2o$7b2o15bo2bo24b2o6b2o3bo2bo$7b2o16bobo24b2o6bobo
2b4o$62bo4bo$62bo$56b4obo5bo$66b3o$65b2ob2o6b2o$64b3ob3o5b2o$45b2o3b2o
13b2ob2o$45b2o2bobo14b3o$49bo17bo$48b2o26b2o$39b2o35b2o$33bo5b2o21b2o$
32bobo26bobo$32bobo26bo$33bo26b2o8b2o$70bo$71b3o$73bo!
[[ THUMBNAIL AUTOSTART ZOOM 4 LOOP 1540 ]]
Re: The Hunting of the New Herschel Conduits
Posted: May 4th, 2015, 6:02 pm
by Kazyan
So one of the only conduits that can actually follow that thing
just so happens to delete the loaf with its output FNG, so it's semi-usable after all?
That's it; I'm making the BTS my forum avatar.
EDIT: Known lane; different timing.
Code: Select all
x = 31, y = 24, rule = B3/S23
5bo$5b3o$8bo6b3o$7b2o8bo$16b3o11bo$28b3o$27bo$27b2o9$3b2o$3bo2bo$4b3o$
22b2o$2obob2o15b2o$ob2obo$5bo$5b2o!
EDIT2: *angry noises at edge of reaction envelope*
Code: Select all
x = 37, y = 30, rule = LifeHistory
35.2A$35.A$7.2A24.A.A$7.A25.2A$2.A.2A.A$2.2A.A.2A2$6.3A$5.A2.A7.C.C$
5.2A9.C.C$16.3C4$2.D$.2D$2D24.2A$.D24.2A$2.D9$12.D.D$12.D.D$12.3D!
EDIT3: Pi-to-C using that big Pi catalyst I found with Bellman a while back.
Code: Select all
x = 39, y = 25, rule = LifeHistory
8.2A$9.A$9.A.A25.2A$4.2A4.2A25.A$4.A30.A.A$6.A28.2A$2.4A.A$.A4.A$.5A
2$.3A$A2.A$2A$13.3C19.2D$13.C.C17.D2.D$13.C.C17.3D6$25.2A$26.A$23.3A$
23.A!
Re: The Hunting of the New Herschel Conduits
Posted: May 5th, 2015, 1:52 pm
by simsim314
Clean and useful QB -> G
Code: Select all
x = 34, y = 13, rule = B3/S23
30bo$28b3o$2o25bo$bo25b2o$bobo12bo$2b2o11bobo$14bo3bo$15b3o$13b2o3b2o$
30b2o$30bobo$32bo$32b2o!
Here it's as part of Pi->G
Code: Select all
x = 34, y = 24, rule = B3/S23
30bo$28b3o$2o25bo$bo25b2o$bobo$2b2o4$30b2o$30bobo$32bo$32b2o6$15b3o$8b
o6bobo$6b3o6bobo6b2o$5bo18bobo$5b2o19bo$26b2o!
Another pretty rare QB->G
Code: Select all
x = 25, y = 22, rule = B3/S23
20b2o$20bo$18bobo$18b2o2$b2o$2bo$2bobo7bo$3b2o6bobo$10bo3bo$11b3o$9b2o
3b2o2$22b2o$21bo2bo$22b2o3$2b2o18b2o$3bo18b2o$3o$o!
And with pi:
Code: Select all
x = 41, y = 35, rule = B3/S23
3$25b2o$25bo$23bobo$23b2o2$6b2o$7bo$7bobo$8b2o5$27b2o$26bo2bo$27b2o3$
7b2o18b2o$8bo18b2o$5b3o$5bo10b3o$9bo6bobo$7b3o6bobo6b2o$6bo18bobo$6b2o
19bo$27b2o!
Re: The Hunting of the New Herschel Conduits
Posted: May 5th, 2015, 4:06 pm
by Sokwe
Kazyan wrote:EDIT2: *angry noises at edge of reaction envelope*
Sacrificing the output pi allows the B to escape cleanly (done by hand, so it can probably be reduced):
Code: Select all
x = 37, y = 36, rule = LifeHistory
35.2A$35.A$7.2A24.A.A$7.A25.2A$2.A.2A.A$2.2A.A.2A2$6.3A$5.A2.A7.C.C$
5.2A9.C.C$16.3C4$2.D$.2D$2D24.2A$.D24.2A$2.D7$23.2A$23.2A5$5.2A14.2A$
4.A.A14.A.A$4.A18.A$3.2A7.2A9.2A$12.2A!
It might also be possible to have the debris delete the B's natural block, resulting in a clean Pi->H.
Re: The Hunting of the New Herschel Conduits
Posted: May 5th, 2015, 8:19 pm
by Kazyan
Sokwe wrote:Sacrificing the output pi allows the B to escape cleanly (done by hand, so it can probably be reduced):
Awesome; I couldn't figure out how to delete the TL with PI myself. Another conduit involving a BTS, then.
Here's a Pi-to-G, the last thing I'm getting from the level of investigation I want to do on the Pi catalyst:
Code: Select all
x = 33, y = 25, rule = B3/S23
26bo$8b2o16b3o$9bo19bo$9bobo16b2o$4b2o4b2o$4bo$6bo$2b4obo$bo4bo$b5o2$b
3o$o2bo$2o11b3o$12bo3bo$12b2ob2o14b2o$31b2o6$26bo$25bobo$25b2o!
...and, Simkin, are you going for Spartan QB conduits, are other SLs okay? Because the QB's evolution involves TL predecessors pointing in just the right way to use Mike Playle's TL eater in the modern p94 gun, which messes with the reaction quite a lot, no matter the handedness of the eater with respect to the TL. Here's a QB->G and a QB->B using one of them, with the hook pointing "up".
Code: Select all
x = 89, y = 19, rule = B3/S23
17b2o$17b2o2$33bo26b2o$31b3o27bo20bo$30bo30bobo16b3o$30b2o30b2o15bo$
79b2o$42bo$40b3o$39bo$b2o36b2o10b2o$o2bo2b2o42bo2bo2b2o$b2o2bobo43b2o
2bobo$3b2o14bo33b2o14bo$3bo14bobo32bo14bobo$2obo13bo3bo28b2obo13bo3bo
15b2o$o2bob2o11b3o29bo2bob2o11b3o16b2o$2b2ob2o9b2o3b2o29b2ob2o9b2o3b2o
(Unfortunately, the QB->B starts within the reaction envelope and doesn't like the next FNG...I don't think anything can actually follow it yet! Drat!)
There are four more options: the hook pointing down, and then using two of these (one on each side), pointing either both up, both down, or in opposite directions.
EDIT: Another QB->G, this time with one "down" TL eater.
Code: Select all
x = 43, y = 28, rule = B3/S23
17bo$17b3o$20bo$19b2o11$19bo$18bobo$2b2ob2o10bo3bo$o2bob2o11b3o$2obo
12b2o3b2o$3bo$3b2o$b2o2bobo31b2o$o2bo2b2o31bobo$b2o38bo$9b2o30b2o$8bob
o$8bo$7b2o!
Using two TL eaters seems not to be productive, by the way.
Re: The Hunting of the New Herschel Conduits
Posted: May 6th, 2015, 2:43 am
by simsim314
Kazyan wrote: are you going for Spartan QB conduits, are other SLs okay?
This is great! Think that we're now discovering (or proving usefulness of) new "basic element" that comes with H, Pi, B, C, R etc. it's not a conduit but a new group of unexplored conduits. We already have two H->QB and Pi->QB and a bunch of useful QB->G. Some useful QB->H/B/R/Pi would be the final step to incorporate it into the circuitry.
Here is something that looks like a nice start:
Code: Select all
x = 34, y = 19, rule = B3/S23
17b2o$8b2o7b2o$9bo$9bobo21bo$10b2o19b3o$30bo$30b2o5$b2o$o2bo2b2o$b2o2b
obo$3b2o14bo$3bo14bobo$2obo13bo3bo$o2bob2o11b3o$2b2ob2o9b2o3b2o!
Something that looks interesting:
Code: Select all
x = 39, y = 27, rule = LifeHistory
6$10.2C19.2C$11.C19.C$11.C.C15.C.C$12.2C15.2C$3.2C$2.C2.C2.2C$3.2C2.C
.C$5.2C14.C$5.C14.C.C$2.2C.C13.C3.C$2.C2.C.2C11.3C$4.2C.2C9.2C3.2C!
Code: Select all
x = 58, y = 36, rule = LifeHistory
4$32.2C$32.2C$23.C$23.3C$26.C$25.2C15.2C$42.C$40.C.C$40.2C2$13.2C$12.
C2.C2.2C$13.2C2.C.C$15.2C14.C$15.C14.C.C$12.2C.C13.C3.C$12.C2.C.2C11.
3C$14.2C.2C9.2C3.2C!
Re: The Hunting of the New Herschel Conduits
Posted: May 6th, 2015, 6:07 am
by Kazyan
I've got Bellman working on the queen bee right now. In the meantime, have yet another C-to-C.
Code: Select all
x = 25, y = 18, rule = B3/S23
19b2o$19b2o7$3b2o$3bobo$4bo7b3o$11bo2bo$2b5o4b2o$bo4bo$bob2o18b2o$2obo
19b2o$3bo$3b2o!
Re: The Hunting of the New Herschel Conduits
Posted: May 6th, 2015, 7:48 pm
by simsim314
And finally as expected, QB->H (two configurations of the same conduit):
Code: Select all
x = 81, y = 93, rule = LifeHistory
7.4B$8.4B$9.4B$10.4B$11.4B$12.4B$13.4B$14.4B$15.4B$16.4B$17.4B7.2C$
18.4B5.B2C2B$19.4B5.4B$20.4B5.4B11.C10.2C$21.4B3.6B8.3C8.2B2C2B$22.4B
2.9B4.C10.7B3.B$23.4B.10B.B.2C9.8B.5B$24.19B8.17B2.2B2.2B$25.16B10.
26BD$26.15B4.B5.26BDBD$25.16B2.4B5.14B2C9B3DB$12.2C9.26B2.15B2C11BD$
11.C2.C2.2C4.39B.15B$12.2C2.C.C4.37B4.3B.3B.B$14.2C.B6.6BC29B$14.C2.
2B.9BCBC28B$11.2C.C.12BC3BC27B2.B$11.C2.C.2C11B3C28B.B2C$13.2C.2C9B2C
3B2C28B2C$17.2B.43B$20.28B.B.B3.5B$20.22B.4B8.2B$20.20B$21.19B$22.16B
$29.3B$29.3B$30.B9$B$2B$3B$4B$.4B$2.4B$3.4B$4.4B$5.4B$6.4B$7.4B$8.4B$
9.4B$10.4B$11.4B$12.4B$13.4B$14.4B$15.4B$16.4B7.2C$17.4B5.B2C2B$18.4B
5.4B$19.4B5.4B11.C10.2C$20.4B3.6B8.3C8.2B2C2B$21.4B2.9B4.C10.7B3.B$
22.4B.10B.B.2C9.8B.5B$23.19B8.17B2.2B2.2B$24.16B10.26BD$25.15B4.B5.
26BDBD$24.16B2.4B5.14B2C9B3DB$11.2C9.26B2.15B2C11BD$10.C2.C2.2C4.39B.
15B$11.2C2.C.C4.37B4.3B.3B.B$13.2C.B6.6BC29B$13.C2.2B.9BCBC28B$10.2C.
C.12BC3BC27B$10.C2.C.2C11B3C28B$12.2C.2C9B2C3B2C27B$16.2B.40B$19.28B.
B.B3.3B$19.22B.4B9.3B$19.20B15.B2CB$20.19B16.2C$21.16B$28.3B$28.3B$
29.B!
Almost a conduit (such a waste of pretty rare transparent ship):
Code: Select all
x = 52, y = 19, rule = B3/S23
17b2o$17b2o$42b2o$33bo8bobo$31b3o9b2o$30bo$30b2o5$b2o$o2bo2b2o$b2o2bob
o42b2o$3b2o14bo29bobo$3bo14bobo29bo$2obo13bo3bo$o2bob2o11b3o$2b2ob2o9b
2o3b2o!
*Eater3 doesn't work.
EDIT2 Only the blinker left:
Code: Select all
x = 62, y = 35, rule = B3/S23
53bo$53b3o$56bo$55b2o6$17b2o41b2o$17b2o41b2o2$33bo$31b3o$30bo$30b2o5$b
2o$o2bo2b2o$b2o2bobo$3b2o14bo$3bo14bobo$2obo13bo3bo$o2bob2o11b3o$2b2ob
2o9b2o3b2o4$48b2o$48bo$49b3o$51bo!
Re: The Hunting of the New Herschel Conduits
Posted: May 6th, 2015, 7:57 pm
by Extrementhusiast
simsim314 wrote:*Eater3 doesn't work.
Eater 2 does:
Code: Select all
x = 52, y = 28, rule = LifeHistory
17.2A$17.2A$42.2A$33.A8.A.A$31.3A9.2A$30.A$30.2A5$.2A$A2.A2.2A$.2A2.A
.A42.2A$3.2A14.A29.A.A$3.A14.A.A29.A$2A.A13.A3.A$A2.A.2A11.3A$2.2A.2A
9.2A3.2A3$43.2A.A$43.2A.3A$49.A$43.2A.3A$44.A.A$44.A.A$45.A!
Re: The Hunting of the New Herschel Conduits
Posted: May 6th, 2015, 8:02 pm
by Scorbie
Ek.... too late
Anyway nice to see that ship:)
Re: The Hunting of the New Herschel Conduits
Posted: May 6th, 2015, 8:19 pm
by simsim314
Extrementhusiast wrote:Eater 2 does:
Cool! So we have another QB->H now, Great!
This one should be simple to complete:
Code: Select all
x = 58, y = 21, rule = B3/S23
37b2o$38bo$17b2o19bobo$17b2o20b2o2b2o$43b2o$33bo$31b3o$30bo$30b2o4$56b
2o$b2o53bo$o2bo2b2o46bobo$b2o2bobo46b2o$3b2o14bo$3bo14bobo$2obo13bo3bo
$o2bob2o11b3o$2b2ob2o9b2o3b2o!
I think we've proved here sufficiently that QB is a good conduit material that can be converted into H (two conduits), or glider (many ways were found including two pure spartan), and from Pi and H.
Re: The Hunting of the New Herschel Conduits
Posted: May 6th, 2015, 10:56 pm
by dvgrn
Extrementhusiast wrote:simsim314 wrote:*Eater3 doesn't work.
Eater 2 does...
Very nice! So among other things, HF145QB plus this new QBT216H makes a good clean L361 Herschel conduit:
Code:
Select all
x = 54, y = 71, rule = LifeHistory
49.3D$50.D$48.3D2$39.2A$38.A.A$30.A8.A$29.A.A$29.A.A$27.3A.2A$26.A$
27.3A.2A16.2A$29.A.2A16.A.A$50.2A9$49.2A$49.A$47.A.A$20.A26.2A$20.3A$
23.A$22.2A5$35.D$35.D.D$36.D.D$13.2A21.D2.D$13.2A21.D.D13.2A$35.D.D
14.2A$35.D8$7.2A$8.A31.2A$8.A.A24.2A4.A$9.2A16.2A6.2A3.A$27.2A10.A$
35.5A.A$35.A4.A.A$37.A2.A.A$27.2A7.2A3.A$27.2A$9.C$9.C.C$9.3C$11.C5$
2.2A$3.A$3A$A$20.2A$20.2A!
#C [[ THUMBNAIL AUTOSTART STOP 361 ]]
Now, if someone can CatForce an output B or R or H or something from the nice edgy pre-traffic light that you get when you remove the bottom block in the above circuit, then we'd have a few more H-to-2H conduits as well (and yet another possible source/target pattern for elementary conduits...!)
Re: The Hunting of the New Herschel Conduits
Posted: May 7th, 2015, 10:57 am
by dvgrn
One of the things I remember needing a tub-with-tail eater for, that is now
suddenly Spartan, is the following Fx119+R64+F117 combination:
Code:
Select all
x = 99, y = 111, rule = LifeHistory
78.B2A$76.2BA2BAB$76.3B2A3B$74.12B$73.14B$74.13B$74.14B$73.15B$73.14B
$73.13B$73.B3D4B.3B$73.2BD4B$73.2B3D2B$73.6B$72.7B$71.8B$70.8B$69.9B$
68.4B.6B$43.A23.4B.7B$43.3A20.4B2.6B$46.A18.4B3.6B$45.2A17.4B4.6B$45.
4B14.4B4.8B$47.3B12.4B6.8B$47.4B10.4B6.9B$45.7B8.4B7.9B$43.9B7.4B8.
10B$42.11B5.4B9.5B2A3B$43.13B.4B10.5B2A4B3.2A$43.17B11.11B3.A$42.17B
12.4BD7BA.A$41.17B15.B3D4B2.2A$41.18B14.D2B2D2B$40.19B14.5B$40.20B5.
2A5.6B$32.2A7.20B5.A6.6B$33.A5.21B6.A.AB2.8B$33.A.AB2.21B7.2AB2.9B$
34.2AB2.19B11.12B$36.3B.17B12.12B$36.21B12.10B.B2A$37.20B13.8B2.BA.A$
35.21B13.8B6.A$35.2BC18B14.8B5.2A$34.3BCBC14B17.7B$35.2B3C13B15.11B$
34.5BC13B14.12B$33.20B14.12B$32.4B.16B14.11B$31.4B.7B2.8B14.B3D4B.4B$
30.4B.7B4.8B13.2BD4B4.2A$29.4B.7B6.6B14.2B3D2B4.A$28.4B.7B6.6B15.6B6.
3A$27.4B.7B7.7B13.7B8.A$26.4B.7B7.9B11.8B$25.4B.7B9.9B9.8B$24.4B.7B
10.10B7.9B$23.4B.7B9.22B.6B$22.4B.7B8.B.2B2A11BD5B.7B$21.4B.7B8.2A3B
2A9B3D4B2.6B$20.4B.7B9.2A14BDBD4B.7B$19.4B.7B11.15BD15B$18.4B.7B10.5B
2.B2.2B2.19B$11.2A4.4B.4B.2B11.2A4.3B7.18B$10.B2AB2.4B.4B2.3B11.A4.B
2AB5.19B3.B$10.4B.4B.8BA2B7.3A6.2A5.4B2.15B.B2A$11.2B.4B.8BABAB.2B4.A
14.4B2.18B2A$11.6B.4B2.2BA2BA3B2A17.4B3.12B2A3B.B$9.7B.4B3.3B2A2B.B2A
16.4B5.11B2A2B$6.14B3.AB2.5B.B16.4B6.10B2.2B$.B3.14B3.A.A4.2B18.4B9.B
.6B3.2B$2AB.9BA4B4.2A4.2B18.4B14.4B2.B2AB$2A10BABA4B8.B2AB16.4B14.B2A
2B3.2A$.B.9BA2BA4B8.2A16.4B16.2A$4.9B2A6B24.4B$5.11B2.4B22.4B$6.8B5.
4B20.4B$6.10B4.4B18.4B$7.3B4.2A5.4B6.2A8.4B$6.4B4.A7.4B4.B2AB6.4B47.
2B$5.4B6.3A5.4B3.3B6.4B26.A20.5B$6.B2AB7.A6.4B3.B.B4.4B25.3A19.7B$7.
2A16.10B2.4B25.A21.8B$26.14B26.2A20.9B$27.12B18.2A8.B20.10B$28.12B17.
A.A7.3B19.8BA$29.12B17.AB6.6B16.8BABA$28.3B2D9B15.2B3.B2.10B11.10BABA
$28.2BD2BD10B10.B2.19B3.2B2.14BAB$22.9B2D12B8.12BD3B2A15BD11B$21.17B
2A6B6.13B2D2B2A15BDBD9B$22.16B2A8B4.B2D11B2D18B3D7B$21.29B2.2B2D10BD
21BD7B$21.33BD10BD26B2.B$23.34B.4B5.13B.B.2B$23.39B5.7B.B7.B$24.B.B2.
31B$25.3B2.28B21.2A$24.B2AB2.27B23.A$25.2A4.24B22.3A$29.2AB.B.17B25.A
$28.A.AB2.6B.9B$28.A6.2B5.6B$27.2A11.8B$40.2A5B$40.2A3B$42.2B$41.4B$
41.B2AB$42.2A!
#C [[ ZOOM 4 AUTOSTART THUMBNAIL STEP 4 LOOP 750 ]]
The output G5 glider pair can be caught either by a loaf2GtoG reflector as shown here, or directly by an old standard Herschel receiver, also shown at the bottom (and the second receiver variant can now be made Spartan as well -- see the above link.)
It doesn't have a particularly good compression rating, but the recovery time is less than the 497 ticks that a Silver reflector needs -- and until a better Spartan G-to-H comes along, that's still the good-enough standard for self-constructing circuitry.
Re: The Hunting of the New Herschel Conduits
Posted: May 7th, 2015, 11:14 am
by chris_c
dvgrn wrote:One of the things I remember needing a tub-with-tail eater for, that is now
suddenly Spartan, is the following Fx119+R64+F117 combination
Cool. I'd just like to point out that the eater can also eat gliders from the other direction. When you look at it closely the eating action is the same as in the tub-with-tail eater for the first two generations:
Code: Select all
x = 63, y = 25, rule = B3/S23
13b2o$13b2o$7b2o$7bobo$8bo45b2o$12b2o4b2o35bo$11bo2bo3b2o35bobo$12bobo
41bobo$13bo43bo3b2o$61b2o5$15b3o41b3o$15bo43bo$16bo43bo6$3o41b3o$2bo
43bo$bo43bo!
And here is a B->G involving a transparent beehive. It's compatible with simsim's recent B->B but not much else as far as I can tell.
Code: Select all
x = 21, y = 17, rule = B3/S23
16b2o$o15bo$3o11bobo$3bo10b2o$2b2o2$10bo$9bobo$9bobo$10bo3$4b2o$5b2o
10b2o$4b2o11bobo$4bo14bo$19b2o!
Re: The Hunting of the New Herschel Conduits
Posted: May 7th, 2015, 2:24 pm
by simsim314
dvgrn wrote:Now, if someone can CatForce an output B or R or H or something from the nice edgy pre-traffic light that you get when you remove the bottom block
Well not B or R, and yet nice glider. Here it's as part of Lx367:
Code:
Select all
x = 54, y = 88, rule = LifeHistory
43.B$42.3D$42.BDB$41.2B3D$41.5B$41.6B$41.6B$41.5B$40.6B$41.6B$40.7B$
40.6B$40.6B$41.B2C3B$40.2B2C3B$40.8B$40.8B$35.2C4.7B$34.B2CB4.7B$34.
3B4.7B$34.2B5.6B$33.15B$33.17B$33.17B$32.17B2C$32.17B2C$34.16B$34.16B
$34.15B$33.10B.3B$34.8B$33.9B$34.9B$33.10B$32.12B$32.13B$32.12B5.2C$
32.12B5.C$33.10B3.BC.C$20.C11.11B3.B2C$20.3C9.15B$23.C6.18B$22.2C6.
17B$22.26B$22.27B$21.28B$20.29B$21.29B$19.32B$14.B.37B$13.2CB.36B$13.
2C37B2C$14.B.34B.B2C$16.31B5.B$16.12B.19B$16.28B.4B$16.12B.14B3.4B$
16.10B3.10B.3B4.4B$16.10B3.10B9.4B$15.10B5.9B10.4B$15.10B6.8B11.4B$7.
2C6.10B10.3B13.3B$8.C7.9B9.5B.2C10.2B$8.C.CB4.9B.2B6.B2C3B.C11.B$9.2C
B.14B2C6.2CB2.C$11.16B2C10.C$11.14B.2B7.5C.C$12.13B10.C4.C.C$11.14B.
2B9.C2.C.C$9.18B2C7.2C3.C$7.18B.B2C$7.2BC15B2.B$6.3BCBC4B.8B$7.2B3C4B
2.7B$6.5BC4B2.7B$5.10B3.7B$4.4B11.6B.BC$4.3B12.7BC.C$2.4B13.8BC$2.2C
15.6B.B$3.C15.5B$3C16.6B$C16.2CB.5B$16.C.CB3.4B$16.C7.4B$15.2C8.4B$
26.4B$27.4B!
#C [[ THUMBNAIL AUTOSTART STOP 367 ]]
Re: The Hunting of the New Herschel Conduits
Posted: May 7th, 2015, 2:56 pm
by dvgrn
simsim314 wrote:dvgrn wrote:Now, if someone can CatForce an output B or R or H or something from the nice edgy pre-traffic light that you get when you remove the bottom block
Well not B or R, and yet nice glider. Here it's as part of Lx367...
Looks like a good start. I've taken the drastic step of adding "TL" as a source and target in the elementary conduit collection, along with "QB" -- update coming maybe over the weekend.
With any luck this will remind people to notice promising edgy TL outputs in new conduits, instead of just blocking them off (pun somewhat intended.)
Re: The Hunting of the New Herschel Conduits
Posted: May 7th, 2015, 3:21 pm
by simsim314
dvgrn wrote: I've taken the drastic step of adding "TL" as a source and target in the elementary conduit collection
I just want to point out that TL is not "proven" conduit material, and we don't have very good TL->H or TL->G converters that work well. Unlike QB which is currently integrated pretty nicely into the existing circuitry, with few full conduits from and to QB.
I would speculate that HF, is much better "conduit material candidate" than TL, here is an example of QB->HF (and we have few HF->G I think).
Code: Select all
x = 49, y = 19, rule = LifeHistory
17.2C$17.2C2$33.C$31.3C$30.C$30.2C4$47.2C$.2C44.C$C2.C2.2C37.C.C$.2C
2.C.C37.2C$3.2C14.C$3.C14.C.C$2C.C13.C3.C$C2.C.2C11.3C$2.2C.2C9.2C3.
2C!
Re: The Hunting of the New Herschel Conduits
Posted: May 7th, 2015, 4:21 pm
by Kazyan
I've seen a few TLs ad HFs on the edge of my conduit-hunting envelopes. I'm not convinced that a TL would be a particularly good intermediate--and for that matter, I'm skeptical of the HF--but I'll be on the lookout. Bellman is doing some analyses right now.
(Incidentally, has anyone noticed a steep reduction in Bellman's speed on Linux platforms? Usually it took me a few hours to get 1 billion trials on Windows; now it's taking me a few days...maybe I just complied it wrong or something...)
Re: The Hunting of the New Herschel Conduits
Posted: May 7th, 2015, 4:24 pm
by Scorbie
Kazyan wrote:(Incidentally, has anyone noticed a steep reduction in Bellman's speed on Linux platforms? Usually it took me a few hours to get 1 billion trials on Windows; now it's taking me a few days...maybe I just complied it wrong or something...)
There's no Makefile in BellmanWin.(But there is a batch file instead) Did you compile with the same options? i.e. gcc -O3? (Yeah, Just checking.)