Ah! So honored that you guys did my puzzle, I had a lot of fun composing it :) This may seem obvious, but sometimes thinking of an 8-unit thermo as the set 1-9 with 1 number removed makes visualization easier imo. For example, the thermo at c7r7 has to of different type to c3r4 because otherwise you'd have to remove 2 numbers from c3r4 thermo (whichever ones appear in c7r8 and c8r8) and it's also different from c9r9 thermo because otherwise you'd have to remove 2 numbers from c7r7 thermo (whichever numbers you put in c9r7, c9r8). Anyway, really glad you enjoyed it! I think it is possible to color thermometers by what 'group' they're in before imputing any numbers into the grid, but understandably imputing numbers and switching if you guess wrong is likely faster.
It’s nice to hear your feedback on the solve. I was wondering how necessary the bifurcation was in the video, and while Simon was on the right path, I’m in a way relieved that the same logic could’ve been visualized without all the theoretically inserted numbers if it was just thought about differently. You did a fantastic job, and I really hope to possibly see more of your work on here
@@Arocks014 yes, while admittedly the bifurcation does likely speed up a solve, here's is a complete solve path without much writing of numbers. It will involve some visualization and colors may be helpful: The long thermo beginning at c7r3 has to be of different type from the one at c3r4, and c3r4 must be of different type from c7r7 (else r8 breaks). c7r7 must be different from c9r9 (else you have 5 cells in block 9 that are on the ‘same side’ of the number 5). So far we know c7r7 and long one at c7r3 are same type, and c3r4 and c9r9 are same type. Now row 7 is very weak. Think of the 2 possible values for c3r4 (ie either 8,9 or 1,2 but it’s easier to not use numbers, let’s instead call them A for the most extreme and B for the one next to it). How can you put A in r7? The only cell is c1r7, what about B? It can only go in c3r7 or c5r7, but c3r7 would force A into c3r6 and now the A,B pair in c3r4 is broken, so B goes in c5r6 and thus c3r6 thermo is of type c7r7. Next the little thermo at c7r3 has to be of this type as well. Why? Well if it’s not then it’s like there is a 8-unit thermo starting at c6r2 that is of type c7r7. Now it might take a bit more visualization to see the next step, but observe that c2r6 and c8r6 must form a pair (since their thermos are of opposite type so they are both 4 units from the ‘A’ end), thus restricting the first 4 cells of c7r7 thermo to A’, B’, C’, D’ (where A’ is the opposite extreme from A). If you imagine a 8 unit thermo as the set 1-9 with 1 number removed, it’s clear you can't have a 8-unit thermo at c6r2 since this thermo must have both B’ and D’ removed. Ok home stretch, we’ve found 4 thermometers of the same type (c3r6, c7r7, and both at c7r3) either they are all working or they are all broken and every other thermometer works. The latter case is impossible because it breaks c2. Why? Well now c1r3 acts like the beginning of a 7 length thermometer but remember the value A is in c1r7 so c1r3 is restricted to B,C. Again, if you think of 8-unit thermos as the set 1-9 with 1 value removed then this is clearly impossible since the values of both c2r2/3 have to be removed from the 8-unit c3r4 thermo. Anyway now that you know 4 working thermos and one of those in the upper left will make the 5th, the one in the upper right must be a broken one and the rest of the puzzle flows, until a 1,3 in the upper right resolves the upper left thermos and the rest of the puzzle.
Thank you very much for the full explanation! You truly have a uniquely brilliant mindset for these puzzles. I could hardly wrap my head around the A,B strategy but it’s honestly such a clever technique. Even though Simon solved it somewhat differently, I’m glad he could still see the genius behind the puzzle
JNCDI the way they use notation, corner notation means that the number is restricted to two or three squares (but there may be many other numbers that could also be in the square) whereas middle notation means only those numbers can be in the square.
@@tovekauppi1616 I am aware of Snyder notation and use it extensively. However, in this particular case, it near impossible to see the center notation due to the color selection. Hence, the work around.
"only 45 thousand" a year ago and now you're well over 300k subs! hard to believe how far you've come, and I'm so glad I managed to find your channel early last year
hearing the number of subscribers said in this video and then looking at the current number....you guys are doing really well! I have always loved puzzles, and playing alongside your videos has helped me resist the desire to give up.
Hello! Just started the puzzle, but want to say that the purple lines really mess with my notations... Can you please stick with the normal grey lines? Thank you!
Yes please! Make sure the additional info in the grid doesn't make central notation invisible. Some puzzles had black parts in the grid that also made the notation invisible.
Paste this into your browser console to make them all gray: document.querySelectorAll('.sudoku-line').forEach(n=>n.style.backgroundColor = '#cccccc') By the way, usually you should avoid pasting in random code from strangers into your browser console. Only do this if you understand what you're doing or you really trust that I'm being honest!
I'm quite new to Sudoku, its probably the 8th puzzle I've solved. It might not have been the best choice to start with this one. It took me a bit over 6 hours with some sleep after 5 hours yesterday to crack this one, but it was quite fun. Thank you so much for this. I probably can't really judge how brilliant it is, but it was real fun to do!
~43:05 "I suspect I'm just missing something that's starring me in the face." I keep telling you to look at the 1, 3 in columns 8 and 9. It can't be a broken thermometer!!!
I was banging my head a littler earlier when he wasn't seeing the 4 and 3 in column 8. That said, first time I solved it, it took 5 hours and I got 5 broken thermometers so it wasn't a proper solve. I tried again after watching the first half of this video and it still took me an hour to get it!
In a way, I am very glad that Simon bifurcated using the incorrect polarity at first to help demonstrate the logic. If this was a live solve (and it appears to be) then it obviously wasn't intentional, but I would have found it very frustrating if the solve had begun with correctly assuming the direction of the first thermometer and I would've been compelled to try it the other way just to see what happens.
25 minutes just to prove something is wrong? Good Grief, what a lovely, yet disgusting puzzle. Edit: Kurt, if you're out there, you're some sort of crazed genius and i love it
I guessed the four thermometers that line up in 3 pairs of switched thermometers correctly the first time, and JUST squeaked under video time when I usually double that. you're right, it's a brutal, lovely puzzle.
@@eldiablo1467 but I imagine he's probably a better musician than Magnus. In fact I might even be willing to say he's a better music producer than he is a puzzle/chess master. Which places him in the top 1% of two entirely separate fields. (And he's ridiculously humble about it too) So throwing a "well he's not as good as X" doesn't really matter much
Anybody else took 4&1/2hours to finish this? (Sobs in corner) At least I was determined...... 2 hours in and found out all my thermometers were reversed and had to clear the whole board
You didn't have to clear the board though, just reverse the numbers. I don't do sudokus but I figured that part about 2 minutes in. It's like he says in the video, there would be two solutions if the number of normal and broken thermometers were the same, and therefore the only way to solve it is figure out which of the two solutions gives the right number of normal and broken ones, but otherwise each solution is the inverse of the other.
Great puzzle! The colour combination (blue numbers on the magenta lines) is a bit odd, maybe the software could paint a border around the numbers (white or the current colour of the cell)
I'm glad Simon enjoyed this puzzle, but I feel like his method of solving it did it a small disservice. I was utterly blown away by the discovery of what I ended up calling "Quantum Pairs". Like a square that could be 1 or 2 going upwards, or 8 or 9 going down, and having a brother have the same FOUR digits, and yet form a pair that took all four of those possibilities out of other cells (so long as they were on the thermometer and thus subject to "must be going this direction"). It gave an utterly unique feel that....well, I can only describe it as "beautiful logic". It seems that, watching this channel for so long, Simon has affected not just my skill, but my vocabulary as well :P Don't get me wrong, I enjoyed watching Simon's solve, and it's always interesting to see two different approaches that end in the same solution. But I was looking forward to seeing him notice and comment on the Quantum Pairs, maybe see what he called them.
Yeach, "Quantum Pairs" is pretty spot on name. They "entangle" the thermometers of the same direction (same "spin" perhaps). Still the Simon's solution and "quantum" solution for me at least moved more or less the same. In both uncertainty ends at the top left corner, where you finally can assign the actual directions to two "colors" of thermometers (I ended with proving that two top-left thermos must be of different directions and so there are five "green" thermos).
Like the 28 pair in column 7, kinda? (Speaking of which, I'm shocked that Simon didn't place the 19 pair in that same row as pretty much the first thing he did -- distractable! :) ) It is a good name, assuming we're talking about the same thing. I didn't get that same feeling as you did, though, at least not for more than a second or two. Don't get me wrong, I am so pleased to have been able to solve this without any assumptions. Thrilled, even. But I'm pleased now. During? Not so much, lol.
To better explain the Quantum Pairs, I'll use an example from early on in the puzzle. It's possible to tell from near the beginning that the thermometers starting at r4c3 and r9c9 must both be of opposite polarity from the thermo at r7c7, and as such must be the same polarity as each other. From there, we can see a strange phenomenon if we play through both options: If both of these thermos are "real", going from low to high, then we get a more regular 67 pair in row 8, columns 2 and 9. This pair eliminates 7 from the options one space down the left thermo, at r8c3, which was a 78 to start with. Do be careful though, because this same trick does NOT apply to the same 23/78 in r8c8, because, as part of a broken thermometer, it will go from high to low, making that square 23, not 78, making a 67 pair meaningless there. So that's one possible outcome of the quantum pair. The other is that both of the aforementioned thermometers are broken (which turns out to be the correct answer), but the logic plays out identically with the different numbers. r8c2+9 are a 34 pair, forcing a 3 out of r8c3 (and subsequently a 2 out of r8c4), but not touching the 78 in r8c8. The end result of this line of thinking is that the two squares with 3467 as possibilities end up being a pair, despite having four possible options. It's either a 34 pair or a 67 pair. A Quantum Pair. I hope that explanation was clear. Feel free to poke me if it was not :P
@@riluna3695 Ah, ok. Yeah, I did do something that was in a practical sense very similar in reducing the lower part of that thermometer to just the 789/123 options. (I think that's what I had by the time the ambiguity resolved, can't remember exactly.) And I had the 34/67 thing as well! So yeah, same logic, I just found it more painful than beautiful at the time. I love the way you conceptualized it.
I only started watching this channel a little under a year ago, but it's amazing to watch these great puzzles being solved. I love Kurt, always, and even more so now that I know he can make a mean sudoku!!!
I don't do sudokus but I started watching your channel recently. I like your passion and I can follow your logic without pausing most of the time, although I rarely think of something before you do. I guess this was a puzzle that's easier for people used to logical puzzles but not used to sudokus, because I figured out the "two solutions, inverted numbers, only one fits the number of normal and broken thermometers" about 1 minute in, while it took you about 8.5 minutes (not counting the 4.5-minute introduction)! Finally a puzzle where I can figure out the general line faster than you can, even though you still figure out the details much faster than I do of course.
This puzzle provides a clear example of how useful it is to fill in all thermometer candidates early (the "Goodliffe technique"). Simon does this early on, but then abandons the approach later, which creates significant obstacles, because you have to recalculate each cell individually, which means you have to track all restrictions continuously. eg. look at 41:10, where the 31 is staring at the box 2 thermometer.
46:17. And this reminded me of a previous video of yours, teaching one of the very few brand new techniques you introduce on your channel :D. Glad to see you spot the trick as well! Beautiful! Well done Kurt!
This is so great. I love it when you post puzzles I'm able to solve but at the same time, I love it when I throw my hands up and watch the genius at work.
48:24 - Holy hell that was rough. I made one mistake where I forgot to notate one number and spent at least 12 minutes with a messed up board because I was thinking that the one cell couldn't have been that number since I hadn't notated it as a possibility.
IIRC, math was one of Mr. Schneider's fields of expertise. That includes visualizing and understanding symmetries. So N ==> 10 - N would be a symmetry until something introduced that breaks the symmetry. As for switching, don't blank out all the numbers, just substitute 10 - N for N, for N = 1...9.
it took me 49 minutes to solve an expert level sudoku without any mistakes m so happy right now. i started sudoku 2 weeks ago and i learned a lot from you thank you for making such great videos
Thanks for the helpful hint at the start that there are effectively two answers to this puzzle, except for that one fixed thermometer more. It made it possible for me to reason to the solution.
As *@TheZzkar* pointed out in its reply to *@KurtHugoSchneider:* _"While I dislike bifurcation in general I really like the way it worked out here. The problem with bifurcation is finally proving something wrong and having to start over from the guess, but here no progress was lost when the guess broke."_ _"As Simon pointed out, all numbers could just be "inverted" (10-x) and no work was wasted."_ Definitely true, except that this trick is not *bifurcation* at all❗ It is, by definition *mono-furcation,* for the following reasons: 🔹It uses algebraic symbols that misleadingly look like real digits. They are typically called *placeholder digits.* (see below for details) 🔹From the very beginning you are perfectly aware that they are not supposed to represent themselves as digits. 🔹It does not require you to start from scratch if you are "wrong." 🔹Actually, you can never be "wrong" (since you never assume that X = N). You are using digits (except for 5) as algebraic symbols: *1* as a symbol for either *1* or *9* *2* as a symbol for either *2* or *8* *3* as a symbol for either *3* or *7* *4* as a symbol for either *4* or *6* *5* as a real digit (representing itself) *6* as a symbol for either *6* or *4* [...] From the very beginning, you know that for each cell the *final solution* of this algebraic problem will be either: *X = N,* or *X = 10 - N* By the way, you might also use letters for the same purpose: *A* as a symbol for either *1* or *9* *B* as a symbol for either *2* or *8* *C* as a symbol for either *3* or *7* [...] However, *placeholder digits* are no less symbolic than letters, and they are easier to sort along thermometers❗
Lots of comments, so unsure if this has already been posted; but I noticed that because the polarity of the thermometers depend on what the polarity surrounding thermometers are. So, as long as you can prove that two thermometers are opposite of each other, you can eliminate them, then the odd one out has to be real. Then you can work backwards from there.
i’m really glad that he guessed the polarity wrong at first and had to switch colors indicating correct/broken thermometers because green not being associated with correct makes much more sense in my head
The basis for the thermometer logic in this puzzle is the little break on C4. If it were the other way around, break in G3, the puzzle would be reversed.
Took me several hours, but I completed this one using some different techniques for the middle part. There were more triples and other logic on the center thermo that were quite nice. Really liked this one. At the start, there were too many numbers, so I took I guess like you did. I guessed right. I was willing to switch the numbers around as well.
Well, I found a triple. I can't find anything else. I need to watch the video. Edit: Watched the video until 15:06 - Took a 50/50 guess and solved the rest on my own with a total time of about 70 minutes. The guess happened to be correct. Had the 50/50 been wrong, I would have opened a second window and just transferred my solution - turning every 9 into a 1 and vice-versa. Holy, this is some amazing logic.
@@BluesTravels487 You don't "need" the assumption chain he did - he even said so. He put the numbers in to make it more visible. But you can perfectly do the logic by just coloring the thermos two different colors. You will end up with a logic that forces all thermos to be colored either green or grey - and just by counting you will know. If there are five grey ones then they are the correct ones, if there are only four grey ones then they are the broken ones. From there on you can start with numbers ;D
@@QemeH I might come back to your comments on logic later. I agree there maybe a logical solution, but his comments were not reassuring, given the time spent watching his video. My interest is in programming, and I've seen to many 'brute force methods' that were nothing more than trial and error. in fact I wrote my own program to solve standard puzzles at what I call first level logic so as to practice on more complex logic. was looking to see if thermos could be included but given the missing numbers in thermos I don't think I would bother However, right now I think the covid-19 debates are more important, as our lives are at stake!
I think this is a good example of puzzle where, when stuck, you should employ the scientific method: make an observation, ask a relevant question, form a little postulate and apply that. Try to employ Ockhams Razor, an idea that claims simplification is ideal.
Went to school with him, and was in a lot of the same classes. He grew up without a TV in his house, and is just generally unreasonably smart. He was doing chess multis in middle school, I was part of it, and then after taking one year level advanced math (algebra in 6th grade), over the summer he took the next two years (geometry and pre-calc) just so he could start taking Calculus in grade 7. He was motivated and smart, basically, and grew up in a good enough school system that was able to support that. He just had to walk across the street to go take those high school classes, and drive less than 10 minutes to take classes early in the local community college. Great theater program, great music program, lots of chances to take AP classes, etc.
Watching this and screaming at my phone as if you could hear me as I noticed that the smaller of the two themometers in the top left coner was broken. That solved everything for me after. :)
I see a 40+ minute video, I instantly press like! :D Edit: This is mind-boggling... Also for you adults out there, if you want a quick drinking game to start of your next weekend, just watch the first 15 minutes or so and drink every time Simon says "good grief". ;D
Immediately before starting you can see that the thermos exiting box 3 then connecting with the thermo in box 4 both can't be ascending because their combined length is 13 squares. So right away you can prove that the thermo exiting box 3 has to be ascending and the one in box 4 has to be descending.
Hi Simon, long time viewer first time commenting. I love your videos in general because you are so good at explaining your logic as you solve. This video was particularly brilliant though, the solve logic of being able to swap the digits was brilliant! Thanks for the constant flow of great puzzle content.
Absolutely loved following on with the logic with this puzzle! Gave up on it myself but enjoyed Simon's solve anyway. Reminded me of the substitution logic video you put up awhile back-it's funny how logic grows on us
I used center marks for one set of possibilities and corner marks for the other to help keep things straight while I was determining which thermos had to be the same as each other and which had to be different. Once I had 5 of the same I could fill in the correct numbers for real and finish from there.
It seems, psychologically, that the two thermometers beginning in box 1 "have to" be one broken and one normal since, if they were the same kind, the setter would have just written a single thermometer.
4 роки тому
at 39:15 im here screaming about the free 3 on the top right, "JUST PUT THE 3 IN!!!" im screaming in my head, then 5 seconds later "if i can get this square...."--- ITS A THREEE
identifying the "gag" (gag = a pivotal tidbit used as basis for whole) that the puzzle can be unique only if an unequal number of thermometers are broken is an observation worthy of bestowing genius status upon yourself (imbo)
Really enjoyed this. Took me a couple of hours but I started by just checking through for classifying the thermometers, i.e. is this the same type as this? - and it didn't seem that I needed trial and error for that. There were 5 of one type and 4 of another when compared to each other - unless I just got lucky!
A fantastic puzzle and video. I don't have the skill to do the incredible logic this puzzle demands but it was a joy watching you work through the puzzle.
I think the trick to solving without trying any numbers is finding out the two sets of polarities and then using their size to determine which is which. For example if there's 5 green thermometers then the green thermometers are normal
30 minutes of staring at the puzzle before I watched the first part of the video, 1:15 of trying to work out which ones went which way (including a reset in the middle), 40 minutes of normal Sudoku. My head now hurts and it's half eleven at night....
I spend 2 1/2 hours on this until I realised that I made a mistake somewhere and I just gave up at this point. Good sudoku but waaaaay to complicated for me (but I just recently started solving those kind of sudokus, so no surprise here^^)
I find the idea behind this puzzle similar to the technique (which I don't remember the name of, naturally *sigh*) that you showed some time ago for solving hard sudokus where a subset of the numbers formed a common pattern through the grid (in this case it was the 7, 8, and 9). The technique was to just place an initial set of numbers, to find A solution, and then switch the numbers around to find THE correct solution. The difference in this puzzle is that you swap the numbers with their "polar" if/when they create a contradiction.
Brilliant puzzle, took me well over an hour to solve but thoroughly enjoyed it. To be honest, I think most other 'normal' thermo sudokus will feel extremely easy in the future after this brutal puzzle.
Bit of an epic - 4h 12m, very clever puzzle though. First thermo I have done and stared at a blank grid for ages. carried on putting all options in each cell until I realised they were all in complementary pairs (e.g. 1289, 34, 456) and I realised why the only given number was 5. by that time I had 4 thermos (coded green) going one way and 2 (red) going the other so guessed green was right and new I could swap out the complementaries if not. My grid wash awash with little digits by the time I spotted this so credit to Simon for spotting this early on. A case of getting inside the composers head.
The purple lines make it terribly difficult to read your pencil marks. Otherwise, thank you for another awesome video! Your content is often the highlight of my evenings during this long pandemic! Also got me deep into sudoku!
Indeed knowing the number of normal and broken thermos is the key, because otherwise the 10's complement of a solution is also a solution, since 10's complement of 5 is still 5.
i am really enjoying your channel, only just started playing Sudoku, already hooked, but come on, this puzzle was a doddle to come up with, you take a completed Sudoku game and add the thermometer lines, you can clearly see they cant go any further, so no one worked it out, just drawing lines through number that go up or down, but i was very impressed with your method to solve it. Of course to back up what i just said id better work on a puzzle myself and see if it gives you as much trouble
If you started with the opposite assumption (i.e. assume the correct orientation of the thermometers) you reach a slightly different way to check. I determined the 2 connected thermometers in the top left had to be opposite polarity, which meant I needed one more of each broken and unbroken, and I already had 4 unbroken thermometers so I new it had to be correct.
43:02 Yes. The 1-3 in the top right corner is forcing the polarity of that 4 thermo that you're working on. Still, I went down the rabbit hole of chaining, and got stuck way deep, so I'm not one to talk here.
Hey everyone, if you're struggling with the colors of the thermos, open the web app and paste this into your browser's console to make them all gray: document.querySelectorAll('.sudoku-line').forEach(n=>n.style.backgroundColor = '#cccccc') By the way, usually you should avoid pasting in random code from strangers into your browser console. Only do this if you understand what you're doing or you really trust that I'm being honest!
That took me about four hours, starting over twice. One of the mistakes I made was watching the beginning of the video when I thought I was really stuck - most of the numbers I'd found were all "wrong" - only later did I watch enough to figure out why. Fun puzzle regardless.
As a setter of exams, I don't like giving redundant or otherwise useless information; students tend to get creative with it. Thus, I was suspicious of the thermos in the upper right as both broken or correct. If that were so, why put the bulb in the middle? It doesn't do anything. Therefore, one should be correct and the other broken.
But that isn't proof, that is a preference. If they were both broken or both real, it would still force the puzzler to find more proof before filling in the numbers.
I'm months late to this - I'm a sub and a patreon for a year or more just have a LOT to catch up on. I loved the logic of the puzzle being non unique unless the polarity mattered, so when you found out that the polarity of the shared bulb was the same in both directions, doesn't that logically mean that they both have to be normal as there are 5 normal and 4 broken? Brilliant logic and really enjoyed this one, took me ages to solve but really was fun.
My solve process on this one: the two bottom right thermometers cannot be both broken, or both whole. one of each. cell 4-3 is both the bulb and end of a thermometer. These two thermometers must be one of each. the thermometers with bulbs at 4-3 and 7-7 cannot be both broken or both whole. they must be one of each. There are only two ways to set those 4 thermometers up. I arbitrarily assigned the thermometers with bulbs at 9-9 and 4-3 to be broken and the other two whole. I left all the other thermometers as 'undetermined', filled in the numbers I had, and went to work. This produced a unique solution. As i typically finish the puzzles in double the time of the video, and I got my solution in just under the video length, I conclude the other way may not produce a solution. Testing revealed as much, confirming the anomaly was my correctly deducing the correct thermometer pattern the first time.
*spoiler question* 34:06 - top left thermometer must be ascending, because the only way to solve the 349's on a descending thermometer would be to have a 5 in the bulb, right? And there's already a 5 in column 2,so therefore the thermometer must be ascending. Is my logic sound?
There was a contradiction in the beginning when you said the thermometer in block 3, r3c7-r2c5, is a broken one, because if the bulb is 7/8, and it is 2 blocks long after the bulb, those 2 blocks have to decrease from 9, so 9,8,7 which is can’t because there is an 8 in r7c5 which would clash with the 8in r3c5. Anyone else notice?
I listened to the premise and rules for this sudoku, and I'm certain that it would take me at least a few hours to solve it (assuming I ever did solve it). I think I'll just watch you go through it!
Practice, practice, practice. I'm at the point where I'm routinely doubling the fellows' video times on sudoku solves, unless they're doing a walkthrough, and not a solve (the Isolation Killer sudoku took me just under 2 hours, for an example of my times on walkthrough puzzles).
@@tashkiira7838 Oh I agree. I've bought both the iOS apps that CtC have put out, and between those apps and watching these videos I'm getting much better (faster) at solving them. But for this particular puzzle, it was really clear to me that it'd take me hours if I tried to solve it by myself, and I just don't have the patience (or the spare time) to spend that that much time on one puzzle. But I was certainly happy to watch to see how he solved it!
I just looked at the puzzle that the bulbs could go either away as you had 5 increase/4 decrease or vice versa seeing as the only given digit was a 5. (so you could have 4 increase/5 decrease and you can switch 1 for 9, 2 for 8, etc). Got messed up with some digits I had so needed a fixing of the right's thermometer for flush the rest of it out. Now to switch the pairs if needed.
I only saw this guy before on a video where he was blindfolded playing chess vs 4 other guys (different games) who could see the game at the same time .
Do you guys sell a bound book of all the sudokus? I like to work them out by hand and the only books they have on amazon are traditional sudoku, I can't find any books with the create rules you guys play with.
Please go back to the way ya'll were doing thermometers in your software before. All of these thick dark lines make it very hard to read central pencil marks.
1:15 Simon, buddy, you're gonna want to upgrade to Windows 10 at some point! Windows 7 just recently ended support so you won't have the latest security patches!
@@CrackingTheCryptic I had to switch from using old Windows 7 so I just started using a different computer with W10 and kept W7 as a backup till I got it all worked out.
@@ItsSansom It's funny how you say that,yet the security breach with the QR codes in discord happened on win10,but not on win7. Now why's that I wonder. No offense,but sometimes making a change too fast could be more dangerous than making a change when it's certified as BETTER FOR SURE option. Just wanna put my 2 cents.
Ah! So honored that you guys did my puzzle, I had a lot of fun composing it :) This may seem obvious, but sometimes thinking of an 8-unit thermo as the set 1-9 with 1 number removed makes visualization easier imo. For example, the thermo at c7r7 has to of different type to c3r4 because otherwise you'd have to remove 2 numbers from c3r4 thermo (whichever ones appear in c7r8 and c8r8) and it's also different from c9r9 thermo because otherwise you'd have to remove 2 numbers from c7r7 thermo (whichever numbers you put in c9r7, c9r8). Anyway, really glad you enjoyed it! I think it is possible to color thermometers by what 'group' they're in before imputing any numbers into the grid, but understandably imputing numbers and switching if you guess wrong is likely faster.
Wow you heart his comment but not mine :(
It’s nice to hear your feedback on the solve. I was wondering how necessary the bifurcation was in the video, and while Simon was on the right path, I’m in a way relieved that the same logic could’ve been visualized without all the theoretically inserted numbers if it was just thought about differently. You did a fantastic job, and I really hope to possibly see more of your work on here
@@Arocks014 yes, while admittedly the bifurcation does likely speed up a solve, here's is a complete solve path without much writing of numbers. It will involve some visualization and colors may be helpful:
The long thermo beginning at c7r3 has to be of different type from the one at c3r4, and c3r4 must be of different type from c7r7 (else r8 breaks). c7r7 must be different from c9r9 (else you have 5 cells in block 9 that are on the ‘same side’ of the number 5). So far we know c7r7 and long one at c7r3 are same type, and c3r4 and c9r9 are same type.
Now row 7 is very weak. Think of the 2 possible values for c3r4 (ie either 8,9 or 1,2 but it’s easier to not use numbers, let’s instead call them A for the most extreme and B for the one next to it). How can you put A in r7? The only cell is c1r7, what about B? It can only go in c3r7 or c5r7, but c3r7 would force A into c3r6 and now the A,B pair in c3r4 is broken, so B goes in c5r6 and thus c3r6 thermo is of type c7r7.
Next the little thermo at c7r3 has to be of this type as well. Why? Well if it’s not then it’s like there is a 8-unit thermo starting at c6r2 that is of type c7r7. Now it might take a bit more visualization to see the next step, but observe that c2r6 and c8r6 must form a pair (since their thermos are of opposite type so they are both 4 units from the ‘A’ end), thus restricting the first 4 cells of c7r7 thermo to A’, B’, C’, D’ (where A’ is the opposite extreme from A). If you imagine a 8 unit thermo as the set 1-9 with 1 number removed, it’s clear you can't have a 8-unit thermo at c6r2 since this thermo must have both B’ and D’ removed.
Ok home stretch, we’ve found 4 thermometers of the same type (c3r6, c7r7, and both at c7r3) either they are all working or they are all broken and every other thermometer works. The latter case is impossible because it breaks c2. Why? Well now c1r3 acts like the beginning of a 7 length thermometer but remember the value A is in c1r7 so c1r3 is restricted to B,C. Again, if you think of 8-unit thermos as the set 1-9 with 1 value removed then this is clearly impossible since the values of both c2r2/3 have to be removed from the 8-unit c3r4 thermo. Anyway now that you know 4 working thermos and one of those in the upper left will make the 5th, the one in the upper right must be a broken one and the rest of the puzzle flows, until a 1,3 in the upper right resolves the upper left thermos and the rest of the puzzle.
This was utterly brilliant.
Thank you very much for the full explanation! You truly have a uniquely brilliant mindset for these puzzles. I could hardly wrap my head around the A,B strategy but it’s honestly such a clever technique. Even though Simon solved it somewhat differently, I’m glad he could still see the genius behind the puzzle
You forgot to mention the most interesting rule to the puzzle, all pencil marks are invisible.
That's why the thicc fading grey thermometers are superior
Blue numbers on purple lines make this harder than it should be.
Agreed!!! Used corner notation as a work around.
JNCDI the way they use notation, corner notation means that the number is restricted to two or three squares (but there may be many other numbers that could also be in the square) whereas middle notation means only those numbers can be in the square.
@@tovekauppi1616 I am aware of Snyder notation and use it extensively. However, in this particular case, it near impossible to see the center notation due to the color selection. Hence, the work around.
Just for the record, the sudoku app has these lines greyed now.
"excuse me, I'm just being a little slow.." I say "a little slow" is like 5 times my thinking speed.
That passion in your eyes "it's alive! it's alive". I'm scared.
"only 45 thousand" a year ago and now you're well over 300k subs! hard to believe how far you've come, and I'm so glad I managed to find your channel early last year
"It's not broken so it's annoying."- always wanted to hear someone say this sentence haha
hearing the number of subscribers said in this video and then looking at the current number....you guys are doing really well! I have always loved puzzles, and playing alongside your videos has helped me resist the desire to give up.
Famous people are watching your great channel. :)
Hello! Just started the puzzle, but want to say that the purple lines really mess with my notations... Can you please stick with the normal grey lines? Thank you!
Yes please! Make sure the additional info in the grid doesn't make central notation invisible. Some puzzles had black parts in the grid that also made the notation invisible.
This ^^^^ - So much!
Yea, I think it was done differently since there's thermometers that 'end' on another one's bulb but grey should still work here
It also makes it difficult to follow along on the video, not just trying to solve it ourselves.
Paste this into your browser console to make them all gray:
document.querySelectorAll('.sudoku-line').forEach(n=>n.style.backgroundColor = '#cccccc')
By the way, usually you should avoid pasting in random code from strangers into your browser console. Only do this if you understand what you're doing or you really trust that I'm being honest!
I'm quite new to Sudoku, its probably the 8th puzzle I've solved. It might not have been the best choice to start with this one. It took me a bit over 6 hours with some sleep after 5 hours yesterday to crack this one, but it was quite fun. Thank you so much for this. I probably can't really judge how brilliant it is, but it was real fun to do!
~43:05 "I suspect I'm just missing something that's starring me in the face." I keep telling you to look at the 1, 3 in columns 8 and 9. It can't be a broken thermometer!!!
You finally heard me. Good.
Haha same, i dont think i've screamed so passionately at a sudoku before xD
Dude, same. It was torture
I was banging my head a littler earlier when he wasn't seeing the 4 and 3 in column 8. That said, first time I solved it, it took 5 hours and I got 5 broken thermometers so it wasn't a proper solve. I tried again after watching the first half of this video and it still took me an hour to get it!
@@moonglum101 Well, if you had 5 broken thermometers but otherwise valid, wouldn't subtracting each number from 10 have gotten you the true solution?
In a way, I am very glad that Simon bifurcated using the incorrect polarity at first to help demonstrate the logic. If this was a live solve (and it appears to be) then it obviously wasn't intentional, but I would have found it very frustrating if the solve had begun with correctly assuming the direction of the first thermometer and I would've been compelled to try it the other way just to see what happens.
25 minutes just to prove something is wrong? Good Grief, what a lovely, yet disgusting puzzle.
Edit: Kurt, if you're out there, you're some sort of crazed genius and i love it
I guessed the four thermometers that line up in 3 pairs of switched thermometers correctly the first time, and JUST squeaked under video time when I usually double that. you're right, it's a brutal, lovely puzzle.
michael baker have you seen him beat 4 people at once while blind playing chess?
@@Pr2jay Still not as good as Magnus.
@@eldiablo1467 but I imagine he's probably a better musician than Magnus. In fact I might even be willing to say he's a better music producer than he is a puzzle/chess master. Which places him in the top 1% of two entirely separate fields. (And he's ridiculously humble about it too) So throwing a "well he's not as good as X" doesn't really matter much
@@Kelarys Such an irrelevant, wasted reply, considering we were talking about chess to begin with. Nice try though.
Anybody else took 4&1/2hours to finish this?
(Sobs in corner)
At least I was determined......
2 hours in and found out all my thermometers were reversed and had to clear the whole board
4 hours 12 minutes but a lot of that was false starts and staring at a blank grid.
You didn't have to clear the board though, just reverse the numbers. I don't do sudokus but I figured that part about 2 minutes in. It's like he says in the video, there would be two solutions if the number of normal and broken thermometers were the same, and therefore the only way to solve it is figure out which of the two solutions gives the right number of normal and broken ones, but otherwise each solution is the inverse of the other.
Took me 6 hours and about 3 oversights that broke it. No shame.
Once I'm past 3 hours and more than 1 starting hint from Simon, I stop counting. I'm just glad I like torturing myself.
Great puzzle! The colour combination (blue numbers on the magenta lines) is a bit odd, maybe the software could paint a border around the numbers (white or the current colour of the cell)
I'm glad Simon enjoyed this puzzle, but I feel like his method of solving it did it a small disservice. I was utterly blown away by the discovery of what I ended up calling "Quantum Pairs". Like a square that could be 1 or 2 going upwards, or 8 or 9 going down, and having a brother have the same FOUR digits, and yet form a pair that took all four of those possibilities out of other cells (so long as they were on the thermometer and thus subject to "must be going this direction"). It gave an utterly unique feel that....well, I can only describe it as "beautiful logic". It seems that, watching this channel for so long, Simon has affected not just my skill, but my vocabulary as well :P
Don't get me wrong, I enjoyed watching Simon's solve, and it's always interesting to see two different approaches that end in the same solution. But I was looking forward to seeing him notice and comment on the Quantum Pairs, maybe see what he called them.
Yeach, "Quantum Pairs" is pretty spot on name. They "entangle" the thermometers of the same direction (same "spin" perhaps). Still the Simon's solution and "quantum" solution for me at least moved more or less the same. In both uncertainty ends at the top left corner, where you finally can assign the actual directions to two "colors" of thermometers (I ended with proving that two top-left thermos must be of different directions and so there are five "green" thermos).
Like the 28 pair in column 7, kinda? (Speaking of which, I'm shocked that Simon didn't place the 19 pair in that same row as pretty much the first thing he did -- distractable! :) ) It is a good name, assuming we're talking about the same thing. I didn't get that same feeling as you did, though, at least not for more than a second or two. Don't get me wrong, I am so pleased to have been able to solve this without any assumptions. Thrilled, even. But I'm pleased now. During? Not so much, lol.
To better explain the Quantum Pairs, I'll use an example from early on in the puzzle. It's possible to tell from near the beginning that the thermometers starting at r4c3 and r9c9 must both be of opposite polarity from the thermo at r7c7, and as such must be the same polarity as each other. From there, we can see a strange phenomenon if we play through both options: If both of these thermos are "real", going from low to high, then we get a more regular 67 pair in row 8, columns 2 and 9. This pair eliminates 7 from the options one space down the left thermo, at r8c3, which was a 78 to start with. Do be careful though, because this same trick does NOT apply to the same 23/78 in r8c8, because, as part of a broken thermometer, it will go from high to low, making that square 23, not 78, making a 67 pair meaningless there.
So that's one possible outcome of the quantum pair. The other is that both of the aforementioned thermometers are broken (which turns out to be the correct answer), but the logic plays out identically with the different numbers. r8c2+9 are a 34 pair, forcing a 3 out of r8c3 (and subsequently a 2 out of r8c4), but not touching the 78 in r8c8.
The end result of this line of thinking is that the two squares with 3467 as possibilities end up being a pair, despite having four possible options. It's either a 34 pair or a 67 pair. A Quantum Pair. I hope that explanation was clear. Feel free to poke me if it was not :P
@@riluna3695 Ah, ok. Yeah, I did do something that was in a practical sense very similar in reducing the lower part of that thermometer to just the 789/123 options. (I think that's what I had by the time the ambiguity resolved, can't remember exactly.) And I had the 34/67 thing as well! So yeah, same logic, I just found it more painful than beautiful at the time. I love the way you conceptualized it.
I only started watching this channel a little under a year ago, but it's amazing to watch these great puzzles being solved. I love Kurt, always, and even more so now that I know he can make a mean sudoku!!!
This puzzle is phenomenal! I was so impressed with your solve. I wouldn't have even known where to begin!
Not sure why you ended up in my recommend videos. But I have been bing watching your videos like you're the tiger king. Love the videos.
I don't do sudokus but I started watching your channel recently. I like your passion and I can follow your logic without pausing most of the time, although I rarely think of something before you do. I guess this was a puzzle that's easier for people used to logical puzzles but not used to sudokus, because I figured out the "two solutions, inverted numbers, only one fits the number of normal and broken thermometers" about 1 minute in, while it took you about 8.5 minutes (not counting the 4.5-minute introduction)! Finally a puzzle where I can figure out the general line faster than you can, even though you still figure out the details much faster than I do of course.
One thing that might help is to use the two notation modes for the different directions the thermometers can take.
This puzzle provides a clear example of how useful it is to fill in all thermometer candidates early (the "Goodliffe technique"). Simon does this early on, but then abandons the approach later, which creates significant obstacles, because you have to recalculate each cell individually, which means you have to track all restrictions continuously.
eg. look at 41:10, where the 31 is staring at the box 2 thermometer.
46:17.
And this reminded me of a previous video of yours, teaching one of the very few brand new techniques you introduce on your channel :D. Glad to see you spot the trick as well! Beautiful! Well done Kurt!
This is so great. I love it when you post puzzles I'm able to solve but at the same time, I love it when I throw my hands up and watch the genius at work.
48:24 - Holy hell that was rough. I made one mistake where I forgot to notate one number and spent at least 12 minutes with a messed up board because I was thinking that the one cell couldn't have been that number since I hadn't notated it as a possibility.
IIRC, math was one of Mr. Schneider's fields of expertise. That includes visualizing and understanding symmetries. So N ==> 10 - N would be a symmetry until something introduced that breaks the symmetry. As for switching, don't blank out all the numbers, just substitute 10 - N for N, for N = 1...9.
it took me 49 minutes to solve an expert level sudoku without any mistakes m so happy right now. i started sudoku 2 weeks ago and i learned a lot from you thank you for making such great videos
"Thermometer-ness" is my new favourite word.
Thanks for the helpful hint at the start that there are effectively two answers to this puzzle, except for that one fixed thermometer more. It made it possible for me to reason to the solution.
As *@TheZzkar* pointed out in its reply to *@KurtHugoSchneider:*
_"While I dislike bifurcation in general I really like the way it worked out here. The problem with bifurcation is finally proving something wrong and having to start over from the guess, but here no progress was lost when the guess broke."_
_"As Simon pointed out, all numbers could just be "inverted" (10-x) and no work was wasted."_
Definitely true, except that this trick is not *bifurcation* at all❗ It is, by definition *mono-furcation,* for the following reasons:
🔹It uses algebraic symbols that misleadingly look like real digits. They are typically called *placeholder digits.* (see below for details)
🔹From the very beginning you are perfectly aware that they are not supposed to represent themselves as digits.
🔹It does not require you to start from scratch if you are "wrong."
🔹Actually, you can never be "wrong" (since you never assume that X = N).
You are using digits (except for 5) as algebraic symbols:
*1* as a symbol for either *1* or *9*
*2* as a symbol for either *2* or *8*
*3* as a symbol for either *3* or *7*
*4* as a symbol for either *4* or *6*
*5* as a real digit (representing itself)
*6* as a symbol for either *6* or *4*
[...]
From the very beginning, you know that for each cell the *final solution* of this algebraic problem will be either:
*X = N,* or
*X = 10 - N*
By the way, you might also use letters for the same purpose:
*A* as a symbol for either *1* or *9*
*B* as a symbol for either *2* or *8*
*C* as a symbol for either *3* or *7*
[...]
However, *placeholder digits* are no less symbolic than letters, and they are easier to sort along thermometers❗
Lots of comments, so unsure if this has already been posted; but I noticed that because the polarity of the thermometers depend on what the polarity surrounding thermometers are. So, as long as you can prove that two thermometers are opposite of each other, you can eliminate them, then the odd one out has to be real. Then you can work backwards from there.
i’m really glad that he guessed the polarity wrong at first and had to switch colors indicating correct/broken thermometers because green not being associated with correct makes much more sense in my head
I watched this on 1.5 speed and couldn't tell a difference. I never really knew what Sudoku was until I watched one of your videos. Brilliant!
I don't want to admit how much time I lost going crazy before a realised that thermometer numbers don't need to be consecutive
Same here, dude. Same here.
I keep doing that, they're hard enough as it is without doing them wrong!
The basis for the thermometer logic in this puzzle is the little break on C4. If it were the other way around, break in G3, the puzzle would be reversed.
Took me several hours, but I completed this one using some different techniques for the middle part. There were more triples and other logic on the center thermo that were quite nice. Really liked this one. At the start, there were too many numbers, so I took I guess like you did. I guessed right. I was willing to switch the numbers around as well.
As of 41:13 there is an X-wing on 6 in r2 and r3, forcing a 6 into r1c6. I think that would speed the solving of the puzzle considerably.
Well, I found a triple. I can't find anything else. I need to watch the video.
Edit: Watched the video until 15:06 - Took a 50/50 guess and solved the rest on my own with a total time of about 70 minutes. The guess happened to be correct.
Had the 50/50 been wrong, I would have opened a second window and just transferred my solution - turning every 9 into a 1 and vice-versa. Holy, this is some amazing logic.
my understand typical rule for Sudoku is no guessing, is not trial and error
@@BluesTravels487 You don't "need" the assumption chain he did - he even said so. He put the numbers in to make it more visible. But you can perfectly do the logic by just coloring the thermos two different colors. You will end up with a logic that forces all thermos to be colored either green or grey - and just by counting you will know. If there are five grey ones then they are the correct ones, if there are only four grey ones then they are the broken ones. From there on you can start with numbers ;D
@@QemeH I might come back to your comments on logic later. I agree there maybe a logical solution, but his comments were not reassuring, given the time spent watching his video. My interest is in programming, and I've seen to many 'brute force methods' that were nothing more than trial and error. in fact I wrote my own program to solve standard puzzles at what I call first level logic so as to practice on more complex logic. was looking to see if thermos could be included but given the missing numbers in thermos I don't think I would bother However, right now I think the covid-19 debates are more important, as our lives are at stake!
i've never been more excited in my life, when I just got the first number before him :D :D :D :D :D :D :D
Got it in 1:25:00 with the same parity argument that Simon came up with. @Kurt Hugo Schneider this is a brilliant puzzle, thanks!
I think this is a good example of puzzle where, when stuck, you should employ the scientific method: make an observation, ask a relevant question, form a little postulate and apply that. Try to employ Ockhams Razor, an idea that claims simplification is ideal.
Wow, what a great puzzle! Took me 100 minutes, but worth it. How is Kurt so talented at everything??
Went to school with him, and was in a lot of the same classes. He grew up without a TV in his house, and is just generally unreasonably smart. He was doing chess multis in middle school, I was part of it, and then after taking one year level advanced math (algebra in 6th grade), over the summer he took the next two years (geometry and pre-calc) just so he could start taking Calculus in grade 7.
He was motivated and smart, basically, and grew up in a good enough school system that was able to support that. He just had to walk across the street to go take those high school classes, and drive less than 10 minutes to take classes early in the local community college. Great theater program, great music program, lots of chances to take AP classes, etc.
Watching this and screaming at my phone as if you could hear me as I noticed that the smaller of the two themometers in the top left coner was broken. That solved everything for me after. :)
I see a 40+ minute video, I instantly press like! :D
Edit: This is mind-boggling... Also for you adults out there, if you want a quick drinking game to start of your next weekend, just watch the first 15 minutes or so and drink every time Simon says "good grief". ;D
Immediately before starting you can see that the thermos exiting box 3 then connecting with the thermo in box 4 both can't be ascending because their combined length is 13 squares. So right away you can prove that the thermo exiting box 3 has to be ascending and the one in box 4 has to be descending.
Very nice. thank you for walking us through this one.
Hi Simon, long time viewer first time commenting. I love your videos in general because you are so good at explaining your logic as you solve. This video was particularly brilliant though, the solve logic of being able to swap the digits was brilliant!
Thanks for the constant flow of great puzzle content.
I wish I was as excited about anything in life as Simon is about sudoku
What kurt wants us to see is that one of the thermometers is in the form of a "9"
Well done, Simon. Thanks for sharing, Kurt.
Absolutely loved following on with the logic with this puzzle! Gave up on it myself but enjoyed Simon's solve anyway. Reminded me of the substitution logic video you put up awhile back-it's funny how logic grows on us
I used center marks for one set of possibilities and corner marks for the other to help keep things straight while I was determining which thermos had to be the same as each other and which had to be different. Once I had 5 of the same I could fill in the correct numbers for real and finish from there.
It seems, psychologically, that the two thermometers beginning in box 1 "have to" be one broken and one normal since, if they were the same kind, the setter would have just written a single thermometer.
at 39:15 im here screaming about the free 3 on the top right, "JUST PUT THE 3 IN!!!" im screaming in my head, then 5 seconds later "if i can get this square...."--- ITS A THREEE
identifying the "gag" (gag = a pivotal tidbit used as basis for whole) that the puzzle can be unique only if an unequal number of thermometers are broken is an observation worthy of bestowing genius status upon yourself (imbo)
Really enjoyed this. Took me a couple of hours but I started by just checking through for classifying the thermometers, i.e. is this the same type as this? - and it didn't seem that I needed trial and error for that. There were 5 of one type and 4 of another when compared to each other - unless I just got lucky!
A fantastic puzzle and video. I don't have the skill to do the incredible logic this puzzle demands but it was a joy watching you work through the puzzle.
My time was a little less than 5 minutes, before I realized that I couldn’t cope with the magenta thermometers 🌡
I managed to deduce which thermos were broken without having to guess, still took me 2 hours
That's even more amazing than solving the puzzle after you know whih way is up.
41:56 All that way around to try to understand if the four in the broken thermometer in row 1, when he has 1 and 3 in the row so it is not possible...
I think the trick to solving without trying any numbers is finding out the two sets of polarities and then using their size to determine which is which. For example if there's 5 green thermometers then the green thermometers are normal
30 minutes of staring at the puzzle before I watched the first part of the video, 1:15 of trying to work out which ones went which way (including a reset in the middle), 40 minutes of normal Sudoku. My head now hurts and it's half eleven at night....
I spend 2 1/2 hours on this until I realised that I made a mistake somewhere and I just gave up at this point. Good sudoku but waaaaay to complicated for me (but I just recently started solving those kind of sudokus, so no surprise here^^)
I find the idea behind this puzzle similar to the technique (which I don't remember the name of, naturally *sigh*) that you showed some time ago for solving hard sudokus where a subset of the numbers formed a common pattern through the grid (in this case it was the 7, 8, and 9). The technique was to just place an initial set of numbers, to find A solution, and then switch the numbers around to find THE correct solution. The difference in this puzzle is that you swap the numbers with their "polar" if/when they create a contradiction.
I think the lightbulb color is fixed! thank for whoever did it
Brilliant puzzle, took me well over an hour to solve but thoroughly enjoyed it. To be honest, I think most other 'normal' thermo sudokus will feel extremely easy in the future after this brutal puzzle.
Bit of an epic - 4h 12m, very clever puzzle though. First thermo I have done and stared at a blank grid for ages. carried on putting all options in each cell until I realised they were all in complementary pairs (e.g. 1289, 34, 456) and I realised why the only given number was 5. by that time I had 4 thermos (coded green) going one way and 2 (red) going the other so guessed green was right and new I could swap out the complementaries if not. My grid wash awash with little digits by the time I spotted this so credit to Simon for spotting this early on. A case of getting inside the composers head.
That was a HELL of a puzzle.
The purple lines make it terribly difficult to read your pencil marks. Otherwise, thank you for another awesome video! Your content is often the highlight of my evenings during this long pandemic!
Also got me deep into sudoku!
Is anyone else watching with no idea what's going on, but just came because Kurt?
Ok, just me.
Also, I came to watch confirmed that Kurt is a GENIUS
Indeed knowing the number of normal and broken thermos is the key, because otherwise the 10's complement of a solution is also a solution, since 10's complement of 5 is still 5.
i am really enjoying your channel, only just started playing Sudoku, already hooked, but come on, this puzzle was a doddle to come up with, you take a completed Sudoku game and add the thermometer lines, you can clearly see they cant go any further, so no one worked it out, just drawing lines through number that go up or down, but i was very impressed with your method to solve it. Of course to back up what i just said id better work on a puzzle myself and see if it gives you as much trouble
Yeah, that one was completely beyond me.
Wow!!!! So awesome!!! 😍💗💗💗 i used to play sudoku!!!!! 😍😍😍😍😍 this is so great!!!
If you started with the opposite assumption (i.e. assume the correct orientation of the thermometers) you reach a slightly different way to check.
I determined the 2 connected thermometers in the top left had to be opposite polarity, which meant I needed one more of each broken and unbroken, and I already had 4 unbroken thermometers so I new it had to be correct.
That was amazing to watch. Thank you to the creator and you
43:02 Yes. The 1-3 in the top right corner is forcing the polarity of that 4 thermo that you're working on. Still, I went down the rabbit hole of chaining, and got stuck way deep, so I'm not one to talk here.
Schrödinger's Sudoku! I love it!
what a beautiful puzzle!
Hey everyone, if you're struggling with the colors of the thermos, open the web app and paste this into your browser's console to make them all gray:
document.querySelectorAll('.sudoku-line').forEach(n=>n.style.backgroundColor = '#cccccc')
By the way, usually you should avoid pasting in random code from strangers into your browser console. Only do this if you understand what you're doing or you really trust that I'm being honest!
That took me about four hours, starting over twice. One of the mistakes I made was watching the beginning of the video when I thought I was really stuck - most of the numbers I'd found were all "wrong" - only later did I watch enough to figure out why. Fun puzzle regardless.
As a setter of exams, I don't like giving redundant or otherwise useless information; students tend to get creative with it. Thus, I was suspicious of the thermos in the upper right as both broken or correct. If that were so, why put the bulb in the middle? It doesn't do anything. Therefore, one should be correct and the other broken.
But that isn't proof, that is a preference. If they were both broken or both real, it would still force the puzzler to find more proof before filling in the numbers.
I'm months late to this - I'm a sub and a patreon for a year or more just have a LOT to catch up on. I loved the logic of the puzzle being non unique unless the polarity mattered, so when you found out that the polarity of the shared bulb was the same in both directions, doesn't that logically mean that they both have to be normal as there are 5 normal and 4 broken?
Brilliant logic and really enjoyed this one, took me ages to solve but really was fun.
My solve process on this one:
the two bottom right thermometers cannot be both broken, or both whole. one of each.
cell 4-3 is both the bulb and end of a thermometer. These two thermometers must be one of each.
the thermometers with bulbs at 4-3 and 7-7 cannot be both broken or both whole. they must be one of each.
There are only two ways to set those 4 thermometers up. I arbitrarily assigned the thermometers with bulbs at 9-9 and 4-3 to be broken and the other two whole. I left all the other thermometers as 'undetermined', filled in the numbers I had, and went to work. This produced a unique solution. As i typically finish the puzzles in double the time of the video, and I got my solution in just under the video length, I conclude the other way may not produce a solution. Testing revealed as much, confirming the anomaly was my correctly deducing the correct thermometer pattern the first time.
*spoiler question*
34:06 - top left thermometer must be ascending, because the only way to solve the 349's on a descending thermometer would be to have a 5 in the bulb, right? And there's already a 5 in column 2,so therefore the thermometer must be ascending.
Is my logic sound?
There was a contradiction in the beginning when you said the thermometer in block 3, r3c7-r2c5, is a broken one, because if the bulb is 7/8, and it is 2 blocks long after the bulb, those 2 blocks have to decrease from 9, so 9,8,7 which is can’t because there is an 8 in r7c5 which would clash with the 8in r3c5.
Anyone else notice?
3h 6min, messed it up ones and had to start over again. What a nice puzzle :)
just finished that puzzle... man what a remarkable difficult problem... took me over 2 hours :(
I listened to the premise and rules for this sudoku, and I'm certain that it would take me at least a few hours to solve it (assuming I ever did solve it). I think I'll just watch you go through it!
Practice, practice, practice. I'm at the point where I'm routinely doubling the fellows' video times on sudoku solves, unless they're doing a walkthrough, and not a solve (the Isolation Killer sudoku took me just under 2 hours, for an example of my times on walkthrough puzzles).
@@tashkiira7838 Oh I agree. I've bought both the iOS apps that CtC have put out, and between those apps and watching these videos I'm getting much better (faster) at solving them.
But for this particular puzzle, it was really clear to me that it'd take me hours if I tried to solve it by myself, and I just don't have the patience (or the spare time) to spend that that much time on one puzzle. But I was certainly happy to watch to see how he solved it!
@@garanceadrosehn9691 Fair dinkum. It's nice to be able to pick up a puzzle and put it down for when you have time again.
39:33, after I figured out which ones were opposites
Wow!!! I just found a UA-camr I followed years ago(from StarCraft 2) and now he showed here!
How do you change the colour on the grid? It doesn't work for me.
I just looked at the puzzle that the bulbs could go either away as you had 5 increase/4 decrease or vice versa seeing as the only given digit was a 5. (so you could have 4 increase/5 decrease and you can switch 1 for 9, 2 for 8, etc). Got messed up with some digits I had so needed a fixing of the right's thermometer for flush the rest of it out. Now to switch the pairs if needed.
I didn't know Kurt from anything but the Banelings song, turns out he's a big deal :D
I only saw this guy before on a video where he was blindfolded playing chess vs 4 other guys (different games) who could see the game at the same time
.
Do you guys sell a bound book of all the sudokus? I like to work them out by hand and the only books they have on amazon are traditional sudoku, I can't find any books with the create rules you guys play with.
Please go back to the way ya'll were doing thermometers in your software before. All of these thick dark lines make it very hard to read central pencil marks.
1:15 Simon, buddy, you're gonna want to upgrade to Windows 10 at some point! Windows 7 just recently ended support so you won't have the latest security patches!
I know! But I haven't dare do it in case my computer goes wrong and I can't make videos :(
New windows are still a security mess anyway, i would still stick to 7.
@@CrackingTheCryptic Ah, a very fair point, I don't think any of us want that lol
@@CrackingTheCryptic I had to switch from using old Windows 7 so I just started using a different computer with W10 and kept W7 as a backup till I got it all worked out.
@@ItsSansom It's funny how you say that,yet the security breach with the QR codes in discord happened on win10,but not on win7. Now why's that I wonder. No offense,but sometimes making a change too fast could be more dangerous than making a change when it's certified as BETTER FOR SURE option. Just wanna put my 2 cents.
A bit of logic was missed in row 7. Because 1 and 9 can't be on the middle of thermometer, there are only two locations for 1 and 9.
that was a crazy one to watch