48:04 for me. My coloring was a bit different than Mark's. I ended up with one color for each digit different than 5. It certainly paid off quickly for the disambiguation with the thermos.
Same here - and you can place the 5 in box 7 straight off - the 3 X dominos add to 30, so the other 3 cells are 15, 5 cant be in the corner as the others would add to 10 with no X - so 5 is in r8c6.
I did something similar. Since there are four unique Hi-Lo pairs that can be used on X dominoes, And there are two unique pairs that can be used on V dominoes, I pencil-marked High digits 6789 and low digits 1234 and chose one color each for each Hi-Lo pair So 1 and 9 were the same color (Red in my solve), 2 and 8 were the same color (Blue in my solve), 3 and 7 were the same color (Green in my solve), and 4 and 6 were the same color (Purple in my solve) I found this method much less confusing and it used two less colors than Mark did. He lost me around the 12 minute mark.
You've probably just inspired some constructor with a new puzzle idea: 'V's in the grid can be 'five' or 'greater than', and you have to figure out which each one is.
I think Mark made this harder than it needed to be by using only four colors. I just started with eight colors (no color needed for 5s) and the recognition that it wasn't too hard to figure out which cells in column 5 had to be the low digits. From there, fully coloring the grid was reasonably straightforward; and, I was able to finish in 21:10.
Indeed. I did the same. Well, to be honest I first did high-low colouring then split colours, and it got really painful, so then I just restarted and coloured r2c2-r2c5 and then r3c5-r6c5 all different colours, because a XVX line is always either 6-4-1-9 or 7-3-2-8 (or reversed) so one was one, and the other was the other, and from there it fell into place. (I contemplated just picking numbers instead of colours and then doing sudoku, and at the end swapping them to fix the thermos)
31:34, probably 5 of those minutes were figuring out how to color/notate the cells before solving the puzzle. EDIT: Oh, crap, multi-color mode was available!
I approached this by saying that green/red/blue/purple would represent two numbers (such that two adjacent cells of the same color would be joined by an X), and saying that dark and light grey represented low and high numbers, and then just ... did sudoku on that. I think that was the easiest way to reason about it. (And of course 5 was its own thing for when I had things like 5/blue pairs, then eventually I could disambiguate values via the thermos.)
Such a cool puzzle! I started straight away with the multicolor for each of the options, so my grid was much uglier than Mark's for most of the solve...
I love the update to the app where the instructions are posted in there as well. Makes it a lot easier for someone as absent-minded as I am. Plus the multi-color option is amazingly helpful.
Most colourful solve I've seen from Mark (in the literal sense). Very entertaining. [Thinks a while..] Ye Gods!! What the heck is the solve going to look like when Simon gets to grips with the new colour options? Avert your eyes now! It burns! It burns!
Yesss I beat Mark’s time! 24:38! I really love the ones where you end up coloring the whole grid, and then within like 30 seconds you go from almost no digits placed to a completed grid. Probably the most satisfying thing in sudoku for me haha
I was going crazy from 28:58 trying to figure out how greys could be 6. I was so relieved when Mark saw that they could not. I thought I was going nuts.
I love watching solves like these, but there is no way I could do it myself. I am always impressed by how these guys can work everything out bit by bit.
I'm somewhat colorblind too (orange and red are super hard for me to distinguish), and just wrote a comment, how I managed to solve it. Be invited to try it out.
I was so lost until he switched half of red to dark grey, and then so surprised that he didn't see the red bulb right away. Red can't be 9! Red can't be 9! For once I really was shouting at my screen. Nice work Mark, you basically solved two sudokus at once with the pairs of colours. I was truly bamboozled by this one.
25:48 I think I'm better at recognizing colors than numbers because I'm never this fast with a puzzle lol. It was really fun (XV is my favorite variant) and the coloring was therapeutic. Thank you to Quarterthru for a great puzzle and to Mark for sharing!
I did a different coloring scheme, and didn't use multicolor. I only used four different colors, and color the high and low of each X pair the same color. This seemed to be efficient and work well for this puzzle.
For coloring puzzles, I like to conceptually group warm colors (red, orange, yellow) and cool colors (green, blue, purple) and then the 3 grays if necessary. For this one specifically, I would have used warm colors for high digits and cool colors for low digits.
I absolutely LOVED this sudoku. I've been using a Colorku/Colordoku board since Christmas and it's been absolutely wild to finally apply that side of the brain digitally. Additionally I paired the X's by Green/Purp/Yellow/Orange and High/Low parity by Light Grey & Black. It lets you set every color to a color/shade pairing.
You're struggling around the 30:00 mark but it's so simple to resolve, all in box 3. You have low red and high red. Low red must be either 6 or 7. High red must be the matching one of 8 or 9. But high red can't be 9 because it's on the bulb of a thermo in box 3, so high red must be 8, and therefore low red must be 7. And the blues must be 6 and 9.
I'm not a setter myself, so I don't know if it's even possible, but I'd love to see an XV sudoku with no Xs nor Vs anywhere in the grid. It would be quite fun to play with effectively just the negative constraint active
You'll be pleased to know there are! (With a few givens or other combined constraints) Highly recommend the discord server where I have seen such puzzles.
There is something similar here on CtC; "The NEW miracle sudoku" by Phistomefel. This uses negative kropki rather than XV, but it's a very similar idea.
Multi color mode is SO USEFUL for this! I used light grey for "Low" and dark grey for "High", and then gave each pair a color. So Green/LightGrey + Green/DarkGrey = 10. It allowed easy scanning for both low/high, and pairs that go together. (It still took me over an hour to solve, but still) I love the multi-color mode so much
New software definitely came in handy for this one. I used two colors in every cell: green or red for low/high and white or black for {1,4,6,9}/{2,3,7,8}.
Very cool how different people used different color methods on this one... I used Red Blue (1 and 4 at the end) and Green Purple (2 and 3). I then added black to the base color for their respective high compliments. After watching the video, I'm glad I did it that way, since it made it much easier to follow. It also saved me from having to "split" the colors at the end and made the thermos far more obvious since it was PurpleBlack
27mn for me, I like these kind of puzzles it’s very fun to do !! I only used 4 colours for the four X-pairs, with pairs of similar colours four the X-pairs that match in the V-pairs (1-9 and 4-6 for instance). It was much easier to read than the multi colouring I think
Instead of coloring, you could just pick any fitting numbers in box 6 and solve the entire puzzle (ignoring the thermos), than replace the high numbers according to the thermos and than the lower ones accordingly. So, if 7 becomes 8, 3 becomes 2 etc. Worked much better for me, since you can just use small notation instead of all the colors.
This puzzle was lovely to solve! And I love the multi-color feature, I've used light and dark grey to determine low and high digits and then blue, green, red, and orange to determine the pairs. Only hassle was that in the end I couldn't double click my colors to instantly fill out the digits. But that's a fair price to pay for such a lovely feature and solve!
I ended up coloring the x's instead of high low like mark did so I ended up with a vastly different solve. I think it made the end a whole lot easier but the beginning a bit harder.
At around 30:00, disambiguation is actually fairly easy: the one thermo shows us that red is greater than grey, the other shows us that blue can be greater than red. Hence, red can't be 9, so the red/grey pair is 7/8.
I got fairly far with colouring, but lacking the split colour option it gets real tough. I just don't understand how the puzzle breaks, as I found the thermos are all high numbers, but as far as I can tell, you can still flip them and make it work. I am not getting something, and I am not sure I have the resilience to wade through it without the added colour options.
Managed to place several of the 5s as they cannot go in any X/V sum Box 6 is fairly limited as 1-2-3-4 are placed, the 5 is placed, the remaining cells must be 6-7-8-9, I decided to colour but I just wasn't grasping it.
First, I used blue for low numbers, and red for high numbers. Then I took advantage of the new multicoloring and split each of those up into 4, with orange, purple, green, and yellow. Here's what it looked like in the end: imgur.com/a/lr8pKAi
This puzzle had me feeling like someone with boxing gloves trying to pick up a credit card. I could see and feel the edges of the path to the solution, but could not grasp it, no matter what I tried. Watching Mark's solution was comforting - in that it was still tricky to understand what was going on! "I don't even understand what I'm deducing in this puzzle!" Me too, Mark. Me too.
I am assuming that when Mark said at the end that maybe the quicker way would be to bifurcate, he was referring to place-holding. That is - to use any combination of 14 vs 23 to start and then switch round at the end to get the thermos to work. I would be interested to know others thoughts, but I think there is a huge difference between placeholding and bifurcating. One is logical, the other is guessing. Most importantly, the placeholding never involves rubbing out and going backwards - it is something that you can build on at the end rather than destroy. I have said many times that it would be great to use letters in the grid to do this, but in reality, using 1-9 is no different to using colours, which itself is not bifurcating. In this case it is definitely much quicker to placehold, as you can fill the grid in without much hassle, and then the tweak at the end is pretty easy. The hardest bit is actually working out how to change all the digits as quickly as possible!
I used 8 colors for my solve. Made some of the logic a bit easier, the different pairs we more obvious. I probably couldn't have done it with 4. Also made the ending super fun to fill in all the numbers in about 10 seconds.
So I started coloring low/high. Then I switched to 2 color low and 2 color high. But then I realized I could go further starting in box one and went for 8 different colors with 5s uncolored. I achieved my goal of coloring the entire grid with 8 different colors with all 5s placed. So it took me a lot longer than Mark, but I finished the last 72 digits in nine moves and completed the puzzle. What amazes me when I've finished one of these hard puzzles is that when I'm watching Mark or Simon solve, even when they take a slightly different path as Mark did here, they seem to get stuck for so long over something that now seems obvious to me(because I've already labored over the puzzle thoroughly). Then I'll check how much time is left in the video and go "Holy smoke, he finishes from this point in 6 minutes?!" And I realize once again, they're better than me at this. Much better.
If a grid just had XV clues, then for any solution one could swap 1 with 4 and 6 with 9 and achieve another solution. For each of those two solutions, one could swap 2 with 3 and 7 with 8. For any of those four solutions, one could swap 1 with 2, 3 with 4, 6 with 7, and 8 with 9, yielding a total of eight solutions based on remapping the digits 1-9. The easiest way to solve the puzzle is to select any box that wholly contains at least one V, and at least partially contains another. Place 1 and 4 in the two cells of one V, and 2 in a cell of the other. Then ignore the thermos while symbolically solving the grid. This will yield a solution which would be correct if the digits were remapped to be consistent with the thermos. If there are two thermos in any puzzle of this form, it will only have a unique solution if, before remapping, one of the thermos has two digits that sum to 5 or 15 and the other does not. On the thermo where the digits sum to 5 or 15, the correct digits are either 2 and 3, or 7 and 8 (and one can tell which based upon the total, which will stay the same when the digits are remapped). On the other thermo, the digits must either be 1 and 2, 3 and 4, 6 and 7, or 8 and 9, and one of those digits will have been established by the first thermo (the first thermo establishes which digits map to 2, 3, 6, and 7). Thus, after solving the grid with numbers, one can use the highlight/replace function to swap groups of digits around, e.g. replacing all the 2's with 0's, then all the 3's with 2's, all the 0's with 3s, all the 7's with 0's, 8's with 7's, and finally all the 0's with 8's.
Great puzzle - really enjoyed doing it. I used four colours to colour the cells of the 4 types of x pairs. Then used digits 1 and 2 as variables to indicate, if known, a low or high number respectively - so giving the 8 possibilities.
45 minutes. Lost half of it figurine out I'd have to do intensive colouring - not just high and low ( yeah I didn't read the title nor watched the thumbnail). Pretty fun and negative constraint use was pretty clever
I spent about 50 minutes at the start working out the high-low parity of about half the board, getting six of the 5s in the process. THEN I switched over to 10-pair coloring, which allowed the rest of the board to be easily high-low marked, and was then able to work out which were which, though I struggled to select things like "all the low oranges," so it took me 80 minutes to get through. Ah, multicolor needs to be turned on in the options!
What a colour test. I'm watching for 3 minutes and everything is so beautifully colourful here and it's getting more and more colourful. I think I can fly ... Colours and more colours, but which one is which? Red is green and blue is orange but on the other hand orange could be blue and green could be red... This is really complicated! Nice Solve Mark!!! Well explained too... How would you solve this with pen and paper?
Remove the two thermos and the puzzle has 4 solutions, all of which are mathematically equivalent with 2-3/8-7 and 1-4/9-6 interchangeable. Quickest method therefore is to "guess" e.g. 1-4 and 2-3 in any box and follow the sudoku logic. Then the thermos tell you what needs to be swapped with what. Digits are easier to handle than colours (or letters for that matter) but they are equivalent. I'm sure Simon or Mark did something like this last year (was it with Kurt Hugo Schneider's puzzle?).
OMG. I made it without looking at the video... a first for me :) 75mins thought while working at the same time... I'm still happy! And that is because you guys helped me so much with your videos!
28 minutes for me but with multiple attempts and restarts. Got confused in the end because I forgot about the negative constraint... This was a lot of fun. I really like the multi-color option! :)
I used cell color split from the get go, using high/low parity and 4 colors for 1/9, 2/8, 3/7, 4/6. So 1 would be low green, 8 would be high red, etc. It was difficult to see towards the end, but it was useful and really pretty, i may say.
31:56 finish. I actually used the split colors for the cells. It slowed me down a bit, but made it easier to identify when it came time to actually fill in some non-5 numbers. So I colored blue for low numbers, red for high numbers, and then split each into four separate colors (yellow/green/grey/orange). The many colors are definitely hypnotic though. Don't stare at them too long! :-D
I split the colors at the very end, so I went from having the grid fully colored in green, orange, red, blue (and dark gray for 5) to one that was colored in all 9 colors. That made it easy to use the thermos to resolve all the digits. But I think it would have just been confusing to try using all 9 colors from the start.
57:34 because I forgot about the negative restraint. I couldn't be bothered colouring, so I just randomly picked the first number in the central box, and then made sure the thermos worked at the end.
It was too fun playing with multicolored cells, so I was slow and finished in 55:27. Nice puzzle and great update! Progress of my coloring pattern: GB => GY/GG BY/BG + GR/BR => G Y GR GRP + BG BY BR BRP
Colors seemed like a trap. I had success just using numbers for marking, starting with r1c1 being a 1, expecting if I it was wrong when I got to thermos (which of course it was) just to be able to swap 1/4 for 2/3 and 6/9 for 7/8.
Took me 1:12:27. Took a long time to figure out the best way to colour things and then I was really slow and methodical with colour sudoku. I have to say, my scanning is not great with numbers... but a thousand times worse with colours. Amd I coloured the entire non-5 part of the grid, basing my solve on simply looking at which cells are forced to be the equivalents of other cells from a different box.
Everyone has their own way of coloring, I guess. This is what I did ... . . . I grouped the ways to get 10 into two factions: 'hot' and 'cold' (one for the 1-9 and 4-6 pairs, the other for the 2-8 and 3-7 pairs) For 'hot' pairs, I use red and orange; for 'cold' pairs, I use blue and green. In each case, I distinguish btw. high and low by using center digits (so I get red-1234, red-6789, orange-1234, and so on ...) Doing this lets me color all 'X' pairs with a single color, and all 'V' pairs will either be red-orange or blue-green. In particular: I know that r2c2345 consists of either 'all hot' or 'all cold' and arbitrarily chose 'cold', and colored r2c23 green and r2c45 blue. This led to r3456c5 to have to be 'all hot', so I then colored r34c5 red and r56c5 orange, while using 1234 for all of {r2c3, r2c4, r4c5, r5c5} and 6789 for all of {r2c2, r2c5, r3c5, r6c5}. From there, I proceeded to identify all the other cells (including the 5s, which I left white).
I solved the puzzle, and I think you can change the 7-3 and 8-2 pairs, therefore the puzzle have 2 different solution. Am I right, or made a mistake somewhere?
Being a "brute force" person, after 2 pathetic attempts, I decided to attribute a random number to one cell, finish the grid by coloring only half of the cells, and finally correct the numbers according to the thermos. I shouldn't be, but actually I'm pretty proud of myself. At least I finished this puzzle
By the way, you can turn on multi-color in the settings. I didn't realize until after watching the video; I might have attempted the puzzle if I had known about that but I figured it would be too hard without multi-coloring.
really nice puzzle, liked it alot, especially for the negative constraint on Xs and Vs :) although it took me 73 minutes, it's always such a joy when finishing a puzzle like that without help
I did not get his coloring logic's deeper sense - i colored what would be 1/9 or 2/8 or 3/7 or 4/6 in four different colors and filled in 1234 or 6789 as it got revealed via the V and Sudoku logic. then I used the thermos to establish the link between the 4 colors and the 9876 and subsequently the 1234... worked fine for me
Took me 78 minutes, and had to throw away 20 or 30 minutes walking an error back. But I got 'er done. I colored the pairs that added to 10, and kept track of the high/low half of the pairs. (Center marked 1234 or 6789) Got the whole grid colored and marked as hi or low, then worked through the 2 thermos and the "v" pairs to get to the final solve.
I colored each sum of ten pair, ones and nines, twos and eights... and mark the small numbers with 1234 pencil marks and the big with 6789. it was a smooth solve for me this way!
My finale was a bit slower than it could have been, as I used pairs of colors to denote cells (light gray for "low", dark gray for "high", paired with red/green/blue/orange), which prevented me from just double-clicking a cell to highlight all the cells that matched (it highlighted all cells that shared a color with that cell instead). Still, I managed a time of 21:49 on this one. Lovely puzzle. Though maybe I shouldn't watch Mark's solve, given the length of the video...
I just picked numbers as place-holders for the 1,2,3,4 in box 6. It's equivalent to using four colours (and remembering that green+purple=5, etc). The plan was to solve the grid, and then re-label the numbers according to the thermos. But I actually got lucky!
I tried 2-coloring, 4-coloring, 2-coloring in multi-color mode and 4-coloring in multi-color mode but I always ended up making an error and getting stuck. Then I gave up colors and made a guess at two mutually exclusive V pairs to get started, using regular notation to complete the grid. I knew I would probably have to do a mass reassignment of digits at the end to fix the thermometers. I also knew I had a 50-50 change of having to backtrack because the initial guess had 3 binary degrees of freedom but the thermometers can only give 2 bits of information. So half of the possible guesses had to be impossible. And I did have to backtrack but only about a dozen moves, and I did have to reassign half of the digits. But I made it through.
I tried to use "hot" colors red and orange for high digits and "cool" blue and purple for the low, matching "fancy" purple and orange with each other, and "plain" blue and red with each other. I think that was more intuitive than Mark's choice. But color coding wasn't enough because you still had to keep track the two distinct options for each color. Early on, I had placed all 5s and that made latter attempts simpler but still not simple enough for me to complete.
Welp, I gave this a shot without watching the video and got the grid *almost* entirely colored before realizing I'd made a mistake. Gonna give it another go today.
Have they changed this since your comment? I'm on mobile and it only shows the rules, which shrinks the numberpad, in landscape orientation for me, so not really a problem.
Mark, to give Simon major joy, you should have pencil-marked the thermos with all possible candidates with the first step in your solve.
Nice update on the software! We finally get the rules in there, yay!
48:04 for me. My coloring was a bit different than Mark's. I ended up with one color for each digit different than 5. It certainly paid off quickly for the disambiguation with the thermos.
I wondered in Marks puzzle, if red couldn't also have been 6 and 8 instead of 7, 8?
Same here - and you can place the 5 in box 7 straight off - the 3 X dominos add to 30, so the other 3 cells are 15, 5 cant be in the corner as the others would add to 10 with no X - so 5 is in r8c6.
@@sarahr.1076 The reds were paired with 2 and 3 to make 10. So they were 7 and 8.
I did something similar.
Since there are four unique Hi-Lo pairs that can be used on X dominoes,
And there are two unique pairs that can be used on V dominoes,
I pencil-marked High digits 6789 and low digits 1234 and chose one color each for each Hi-Lo pair
So 1 and 9 were the same color (Red in my solve),
2 and 8 were the same color (Blue in my solve),
3 and 7 were the same color (Green in my solve), and
4 and 6 were the same color (Purple in my solve)
I found this method much less confusing and it used two less colors than Mark did.
He lost me around the 12 minute mark.
25:00 There's an orange X-Wing in r45 so in box 4 the orange must be in r6
(Not important, but it's a weird sudoku sentence and I wanted to say it)
Today's surprise treat for my officemates was the yell of "It's a naked red, Mark!".
If you replaced the thermos with 'greater than' symbols this would make an excellent puzzle for April 1st 😅
now that's just evil lmao
You've probably just inspired some constructor with a new puzzle idea: 'V's in the grid can be 'five' or 'greater than', and you have to figure out which each one is.
@@joadbreslin5819 That might actually work!
@@joadbreslin5819 My thoughts exactly!
@@joadbreslin5819 make it worse: In A x B v C x D, allow it to also mean A*B>C*D.
I think Mark made this harder than it needed to be by using only four colors. I just started with eight colors (no color needed for 5s) and the recognition that it wasn't too hard to figure out which cells in column 5 had to be the low digits. From there, fully coloring the grid was reasonably straightforward; and, I was able to finish in 21:10.
Indeed. I did the same. Well, to be honest I first did high-low colouring then split colours, and it got really painful, so then I just restarted and coloured r2c2-r2c5 and then r3c5-r6c5 all different colours, because a XVX line is always either 6-4-1-9 or 7-3-2-8 (or reversed) so one was one, and the other was the other, and from there it fell into place.
(I contemplated just picking numbers instead of colours and then doing sudoku, and at the end swapping them to fix the thermos)
31:34, probably 5 of those minutes were figuring out how to color/notate the cells before solving the puzzle.
EDIT: Oh, crap, multi-color mode was available!
I approached this by saying that green/red/blue/purple would represent two numbers (such that two adjacent cells of the same color would be joined by an X), and saying that dark and light grey represented low and high numbers, and then just ... did sudoku on that. I think that was the easiest way to reason about it. (And of course 5 was its own thing for when I had things like 5/blue pairs, then eventually I could disambiguate values via the thermos.)
Such a cool puzzle! I started straight away with the multicolor for each of the options, so my grid was much uglier than Mark's for most of the solve...
24:05 ... I *love* puzzles like these!
Wonderful puzzle!
20:57. Yay the rules are finally in the WebApp! Another Elma grid by the end which was easy enough to fill in once you used the restraints
I love the update to the app where the instructions are posted in there as well. Makes it a lot easier for someone as absent-minded as I am. Plus the multi-color option is amazingly helpful.
Most colourful solve I've seen from Mark (in the literal sense). Very entertaining.
[Thinks a while..]
Ye Gods!! What the heck is the solve going to look like when Simon gets to grips with the new colour options?
Avert your eyes now!
It burns! It burns!
Yesss I beat Mark’s time! 24:38! I really love the ones where you end up coloring the whole grid, and then within like 30 seconds you go from almost no digits placed to a completed grid. Probably the most satisfying thing in sudoku for me haha
I never had so much fun solving a Sudoku before. The multicolor setting is gorgeous and this puzzle was meant for it. I absolutely loved it!!
It looks like Mark and Simon swapped puzzles.
42min , first time i solve this kind of sudokus where u have to colour the entire grid first , i loved it !!!
I was going crazy from 28:58 trying to figure out how greys could be 6. I was so relieved when Mark saw that they could not. I thought I was going nuts.
!!! not the only one :D
I love watching solves like these, but there is no way I could do it myself. I am always impressed by how these guys can work everything out bit by bit.
Colorblind me: “Alright, so what does CTC have for me today?”
CTC: “Colour Test Sudoku!”
Also colorblind me: **Homer Simpson grunt** 😭
I'm somewhat colorblind too (orange and red are super hard for me to distinguish), and just wrote a comment, how I managed to solve it. Be invited to try it out.
This was such a genuinely fun puzzle.
I got started by using all nine colours in box 6 and was able to work my way from there quite smoothly. Spectacular setting.
Excellent solve Mark! Always a pleasure to watch you solve.
I solved this pretty fast by placing random 1v4, 2v3 pairs and at the end fixed them to match the thermos
I was so lost until he switched half of red to dark grey, and then so surprised that he didn't see the red bulb right away. Red can't be 9! Red can't be 9!
For once I really was shouting at my screen.
Nice work Mark, you basically solved two sudokus at once with the pairs of colours. I was truly bamboozled by this one.
25:48 I think I'm better at recognizing colors than numbers because I'm never this fast with a puzzle lol. It was really fun (XV is my favorite variant) and the coloring was therapeutic. Thank you to Quarterthru for a great puzzle and to Mark for sharing!
I absolutely love being able to see the instructions in the app. Thank you so much. Thank you, thank you, thank you.
I did a different coloring scheme, and didn't use multicolor. I only used four different colors, and color the high and low of each X pair the same color. This seemed to be efficient and work well for this puzzle.
I don't see how I could use multicolor cells... ah, in the options.
For coloring puzzles, I like to conceptually group warm colors (red, orange, yellow) and cool colors (green, blue, purple) and then the 3 grays if necessary. For this one specifically, I would have used warm colors for high digits and cool colors for low digits.
Absolutely love the new multiple color mode in the software! Was a nice coloring puzzle, used all colors and this new feature made it so much easier!
I absolutely LOVED this sudoku. I've been using a Colorku/Colordoku board since Christmas and it's been absolutely wild to finally apply that side of the brain digitally.
Additionally I paired the X's by Green/Purp/Yellow/Orange and High/Low parity by Light Grey & Black. It lets you set every color to a color/shade pairing.
You're struggling around the 30:00 mark but it's so simple to resolve, all in box 3. You have low red and high red. Low red must be either 6 or 7. High red must be the matching one of 8 or 9. But high red can't be 9 because it's on the bulb of a thermo in box 3, so high red must be 8, and therefore low red must be 7. And the blues must be 6 and 9.
I'm not a setter myself, so I don't know if it's even possible, but I'd love to see an XV sudoku with no Xs nor Vs anywhere in the grid. It would be quite fun to play with effectively just the negative constraint active
You'll be pleased to know there are! (With a few givens or other combined constraints)
Highly recommend the discord server where I have seen such puzzles.
I may be wrong but I remember seeing one in the channel
There is something similar here on CtC; "The NEW miracle sudoku" by Phistomefel. This uses negative kropki rather than XV, but it's a very similar idea.
@@joaoricardo9174 If my memory serves, I think it had only one x and one v.
I would also add the "XV" combo for 15-sum. Which would not exist because of the negative constraint.
Multi color mode is SO USEFUL for this! I used light grey for "Low" and dark grey for "High", and then gave each pair a color. So Green/LightGrey + Green/DarkGrey = 10. It allowed easy scanning for both low/high, and pairs that go together. (It still took me over an hour to solve, but still) I love the multi-color mode so much
Absolutely loved solving this one, thank you for all the amazing content CTC!
New software definitely came in handy for this one. I used two colors in every cell: green or red for low/high and white or black for {1,4,6,9}/{2,3,7,8}.
how to multicolor.. i cant figure it out . please. thanks :))
@@colleenscottcarmello5103 There's a new button for settings
@@martinepstein9826 Thanks Martin!
Very cool how different people used different color methods on this one... I used Red Blue (1 and 4 at the end) and Green Purple (2 and 3). I then added black to the base color for their respective high compliments. After watching the video, I'm glad I did it that way, since it made it much easier to follow. It also saved me from having to "split" the colors at the end and made the thermos far more obvious since it was PurpleBlack
39:04. my first real coloring book solve! (and also one of the solves ive been proudest of)
Same here. 👍
32:02 For the uninitiated, here is a translation: "Oh, poo!" = Bobbins.
🤩, the audio is definitely better!
Much better.
27mn for me, I like these kind of puzzles it’s very fun to do !!
I only used 4 colours for the four X-pairs, with pairs of similar colours four the X-pairs that match in the V-pairs (1-9 and 4-6 for instance). It was much easier to read than the multi colouring I think
Instead of coloring, you could just pick any fitting numbers in box 6 and solve the entire puzzle (ignoring the thermos), than replace the high numbers according to the thermos and than the lower ones accordingly. So, if 7 becomes 8, 3 becomes 2 etc. Worked much better for me, since you can just use small notation instead of all the colors.
38:53 it's really easier to work with multicolor these hard XV puzzles. I really enjoyed it so much, second XV puzzle I solve from Quarterthru!
I love the idea that constructors are now making puzzles to exploit new features in the solving software. Technological feedback loop!
20:20 for me! Super colorful puzzle, the thermos cleared up the numbers
This puzzle was lovely to solve! And I love the multi-color feature, I've used light and dark grey to determine low and high digits and then blue, green, red, and orange to determine the pairs. Only hassle was that in the end I couldn't double click my colors to instantly fill out the digits. But that's a fair price to pay for such a lovely feature and solve!
This was a delight to watch. Thank you.
I ended up coloring the x's instead of high low like mark did so I ended up with a vastly different solve. I think it made the end a whole lot easier but the beginning a bit harder.
19:03! Really happy with that time. I definitely might have been slightly slower on the start without the title, but it was very enjoyable
At around 30:00, disambiguation is actually fairly easy: the one thermo shows us that red is greater than grey, the other shows us that blue can be greater than red. Hence, red can't be 9, so the red/grey pair is 7/8.
I got fairly far with colouring, but lacking the split colour option it gets real tough. I just don't understand how the puzzle breaks, as I found the thermos are all high numbers, but as far as I can tell, you can still flip them and make it work. I am not getting something, and I am not sure I have the resilience to wade through it without the added colour options.
It's in the web app now if you click the settings gear. (not sure about mobile)
Managed to place several of the 5s as they cannot go in any X/V sum
Box 6 is fairly limited as 1-2-3-4 are placed, the 5 is placed, the remaining cells must be 6-7-8-9, I decided to colour but I just wasn't grasping it.
First, I used blue for low numbers, and red for high numbers. Then I took advantage of the new multicoloring and split each of those up into 4, with orange, purple, green, and yellow. Here's what it looked like in the end: imgur.com/a/lr8pKAi
Very satisfying that you also made the "distance from 5" consistently coloured as a result of encoding the matching via an X the same colour!
I started coloring, but quickly decided it was easier to just use number and fix things using the thermos and strategic x and v's in the end.
This puzzle had me feeling like someone with boxing gloves trying to pick up a credit card. I could see and feel the edges of the path to the solution, but could not grasp it, no matter what I tried.
Watching Mark's solution was comforting - in that it was still tricky to understand what was going on!
"I don't even understand what I'm deducing in this puzzle!" Me too, Mark. Me too.
Having the rules visible during the solve is so helpful.
60:14 for me, this was a really fun one
I am assuming that when Mark said at the end that maybe the quicker way would be to bifurcate, he was referring to place-holding. That is - to use any combination of 14 vs 23 to start and then switch round at the end to get the thermos to work.
I would be interested to know others thoughts, but I think there is a huge difference between placeholding and bifurcating. One is logical, the other is guessing. Most importantly, the placeholding never involves rubbing out and going backwards - it is something that you can build on at the end rather than destroy. I have said many times that it would be great to use letters in the grid to do this, but in reality, using 1-9 is no different to using colours, which itself is not bifurcating.
In this case it is definitely much quicker to placehold, as you can fill the grid in without much hassle, and then the tweak at the end is pretty easy. The hardest bit is actually working out how to change all the digits as quickly as possible!
I used 8 colors for my solve. Made some of the logic a bit easier, the different pairs we more obvious. I probably couldn't have done it with 4. Also made the ending super fun to fill in all the numbers in about 10 seconds.
So I started coloring low/high. Then I switched to 2 color low and 2 color high. But then I realized I could go further starting in box one and went for 8 different colors with 5s uncolored. I achieved my goal of coloring the entire grid with 8 different colors with all 5s placed. So it took me a lot longer than Mark, but I finished the last 72 digits in nine moves and completed the puzzle.
What amazes me when I've finished one of these hard puzzles is that when I'm watching Mark or Simon solve, even when they take a slightly different path as Mark did here, they seem to get stuck for so long over something that now seems obvious to me(because I've already labored over the puzzle thoroughly). Then I'll check how much time is left in the video and go "Holy smoke, he finishes from this point in 6 minutes?!" And I realize once again, they're better than me at this. Much better.
I used different colors for the different "X" dominos and numbers for high and low, which made the solve much easier.
That was a puzzle and a half, I found it incredibly tough
I’m sure Simon will be jealous of all the colouring
If a grid just had XV clues, then for any solution one could swap 1 with 4 and 6 with 9 and achieve another solution. For each of those two solutions, one could swap 2 with 3 and 7 with 8. For any of those four solutions, one could swap 1 with 2, 3 with 4, 6 with 7, and 8 with 9, yielding a total of eight solutions based on remapping the digits 1-9.
The easiest way to solve the puzzle is to select any box that wholly contains at least one V, and at least partially contains another. Place 1 and 4 in the two cells of one V, and 2 in a cell of the other. Then ignore the thermos while symbolically solving the grid. This will yield a solution which would be correct if the digits were remapped to be consistent with the thermos.
If there are two thermos in any puzzle of this form, it will only have a unique solution if, before remapping, one of the thermos has two digits that sum to 5 or 15 and the other does not. On the thermo where the digits sum to 5 or 15, the correct digits are either 2 and 3, or 7 and 8 (and one can tell which based upon the total, which will stay the same when the digits are remapped). On the other thermo, the digits must either be 1 and 2, 3 and 4, 6 and 7, or 8 and 9, and one of those digits will have been established by the first thermo (the first thermo establishes which digits map to 2, 3, 6, and 7).
Thus, after solving the grid with numbers, one can use the highlight/replace function to swap groups of digits around, e.g. replacing all the 2's with 0's, then all the 3's with 2's, all the 0's with 3s, all the 7's with 0's, 8's with 7's, and finally all the 0's with 8's.
Great puzzle - really enjoyed doing it. I used four colours to colour the cells of the 4 types of x pairs. Then used digits 1 and 2 as variables to indicate, if known, a low or high number respectively - so giving the 8 possibilities.
45 minutes. Lost half of it figurine out I'd have to do intensive colouring - not just high and low ( yeah I didn't read the title nor watched the thumbnail).
Pretty fun and negative constraint use was pretty clever
I spent about 50 minutes at the start working out the high-low parity of about half the board, getting six of the 5s in the process. THEN I switched over to 10-pair coloring, which allowed the rest of the board to be easily high-low marked, and was then able to work out which were which, though I struggled to select things like "all the low oranges," so it took me 80 minutes to get through.
Ah, multicolor needs to be turned on in the options!
This was WILDLY Entertaining!
What a colour test. I'm watching for 3 minutes and everything is so beautifully colourful here and it's getting more and more colourful. I think I can fly ... Colours and more colours, but which one is which? Red is green and blue is orange but on the other hand orange could be blue and green could be red... This is really complicated! Nice Solve Mark!!! Well explained too... How would you solve this with pen and paper?
Remove the two thermos and the puzzle has 4 solutions, all of which are mathematically equivalent with 2-3/8-7 and 1-4/9-6 interchangeable. Quickest method therefore is to "guess" e.g. 1-4 and 2-3 in any box and follow the sudoku logic. Then the thermos tell you what needs to be swapped with what. Digits are easier to handle than colours (or letters for that matter) but they are equivalent. I'm sure Simon or Mark did something like this last year (was it with Kurt Hugo Schneider's puzzle?).
Very fun! I coloured the 1234 possibilities in 4 colours, then went through with their 6789 pairs! Excellent puzzle!
I really enjoyed this one, cheers Mark!
Very enjoyable puzzle, not sure I would have been able to solve it without the hint in the video title, but sudoku with colours are almost more fun
OMG. I made it without looking at the video... a first for me :) 75mins thought while working at the same time... I'm still happy! And that is because you guys helped me so much with your videos!
28 minutes for me but with multiple attempts and restarts. Got confused in the end because I forgot about the negative constraint... This was a lot of fun. I really like the multi-color option! :)
I used cell color split from the get go, using high/low parity and 4 colors for 1/9, 2/8, 3/7, 4/6.
So 1 would be low green, 8 would be high red, etc. It was difficult to see towards the end, but it was useful and really pretty, i may say.
With hindsight, it is easier to use numbers 1 to 4 and 6 to 9 as placeholders and swap them at the end than colors. Colors are more fun though.
31:56 finish. I actually used the split colors for the cells. It slowed me down a bit, but made it easier to identify when it came time to actually fill in some non-5 numbers. So I colored blue for low numbers, red for high numbers, and then split each into four separate colors (yellow/green/grey/orange). The many colors are definitely hypnotic though. Don't stare at them too long! :-D
I split the colors at the very end, so I went from having the grid fully colored in green, orange, red, blue (and dark gray for 5) to one that was colored in all 9 colors. That made it easy to use the thermos to resolve all the digits. But I think it would have just been confusing to try using all 9 colors from the start.
57:34 because I forgot about the negative restraint. I couldn't be bothered colouring, so I just randomly picked the first number in the central box, and then made sure the thermos worked at the end.
I used all nine different colours from the start. It's madness at first but allows to skip all which-is-which deductions in the end.
This was a long but fun solve, and my first stab at multicoloring the grid!
It was too fun playing with multicolored cells, so I was slow and finished in 55:27. Nice puzzle and great update!
Progress of my coloring pattern: GB => GY/GG BY/BG + GR/BR => G Y GR GRP + BG BY BR BRP
Colors seemed like a trap. I had success just using numbers for marking, starting with r1c1 being a 1, expecting if I it was wrong when I got to thermos (which of course it was) just to be able to swap 1/4 for 2/3 and 6/9 for 7/8.
Took me 1:12:27. Took a long time to figure out the best way to colour things and then I was really slow and methodical with colour sudoku. I have to say, my scanning is not great with numbers... but a thousand times worse with colours.
Amd I coloured the entire non-5 part of the grid, basing my solve on simply looking at which cells are forced to be the equivalents of other cells from a different box.
Everyone has their own way of coloring, I guess. This is what I did ...
.
.
.
I grouped the ways to get 10 into two factions: 'hot' and 'cold' (one for the 1-9 and 4-6 pairs, the other for the 2-8 and 3-7 pairs) For 'hot' pairs, I use red and orange; for 'cold' pairs, I use blue and green. In each case, I distinguish btw. high and low by using center digits (so I get red-1234, red-6789, orange-1234, and so on ...)
Doing this lets me color all 'X' pairs with a single color, and all 'V' pairs will either be red-orange or blue-green.
In particular: I know that r2c2345 consists of either 'all hot' or 'all cold' and arbitrarily chose 'cold', and colored r2c23 green and r2c45 blue. This led to r3456c5 to have to be 'all hot', so I then colored r34c5 red and r56c5 orange, while using 1234 for all of {r2c3, r2c4, r4c5, r5c5} and 6789 for all of {r2c2, r2c5, r3c5, r6c5}. From there, I proceeded to identify all the other cells (including the 5s, which I left white).
I solved the puzzle, and I think you can change the 7-3 and 8-2 pairs, therefore the puzzle have 2 different solution. Am I right, or made a mistake somewhere?
The thermo in box 6 dictates the 7-8 order. If you swap them, you will break the thermo.
Being a "brute force" person, after 2 pathetic attempts, I decided to attribute a random number to one cell, finish the grid by coloring only half of the cells, and finally correct the numbers according to the thermos. I shouldn't be, but actually I'm pretty proud of myself. At least I finished this puzzle
very lovely puzzle, enjoyed it very much! and great additions to the software as well
By the way, you can turn on multi-color in the settings. I didn't realize until after watching the video; I might have attempted the puzzle if I had known about that but I figured it would be too hard without multi-coloring.
really nice puzzle, liked it alot, especially for the negative constraint on Xs and Vs :) although it took me 73 minutes, it's always such a joy when finishing a puzzle like that without help
I did not get his coloring logic's deeper sense - i colored what would be 1/9 or 2/8 or 3/7 or 4/6 in four different colors and filled in 1234 or 6789 as it got revealed via the V and Sudoku logic. then I used the thermos to establish the link between the 4 colors and the 9876 and subsequently the 1234... worked fine for me
Took me 78 minutes, and had to throw away 20 or 30 minutes walking an error back. But I got 'er done. I colored the pairs that added to 10, and kept track of the high/low half of the pairs. (Center marked 1234 or 6789) Got the whole grid colored and marked as hi or low, then worked through the 2 thermos and the "v" pairs to get to the final solve.
34m 23s. i always enjoy these puzzles where you color the whole grid.
I colored each sum of ten pair, ones and nines, twos and eights... and mark the small numbers with 1234 pencil marks and the big with 6789. it was a smooth solve for me this way!
My finale was a bit slower than it could have been, as I used pairs of colors to denote cells (light gray for "low", dark gray for "high", paired with red/green/blue/orange), which prevented me from just double-clicking a cell to highlight all the cells that matched (it highlighted all cells that shared a color with that cell instead). Still, I managed a time of 21:49 on this one. Lovely puzzle. Though maybe I shouldn't watch Mark's solve, given the length of the video...
I just picked numbers as place-holders for the 1,2,3,4 in box 6. It's equivalent to using four colours (and remembering that green+purple=5, etc). The plan was to solve the grid, and then re-label the numbers according to the thermos. But I actually got lucky!
I tried 2-coloring, 4-coloring, 2-coloring in multi-color mode and 4-coloring in multi-color mode but I always ended up making an error and getting stuck. Then I gave up colors and made a guess at two mutually exclusive V pairs to get started, using regular notation to complete the grid.
I knew I would probably have to do a mass reassignment of digits at the end to fix the thermometers. I also knew I had a 50-50 change of having to backtrack because the initial guess had 3 binary degrees of freedom but the thermometers can only give 2 bits of information. So half of the possible guesses had to be impossible.
And I did have to backtrack but only about a dozen moves, and I did have to reassign half of the digits. But I made it through.
I tried to use "hot" colors red and orange for high digits and "cool" blue and purple for the low, matching "fancy" purple and orange with each other, and "plain" blue and red with each other. I think that was more intuitive than Mark's choice. But color coding wasn't enough because you still had to keep track the two distinct options for each color. Early on, I had placed all 5s and that made latter attempts simpler but still not simple enough for me to complete.
Welp, I gave this a shot without watching the video and got the grid *almost* entirely colored before realizing I'd made a mistake. Gonna give it another go today.
How do you partly color cells?
Love the rules on the puzzle screen, though it shrank the number pad for us mobile users.
Have they changed this since your comment? I'm on mobile and it only shows the rules, which shrinks the numberpad, in landscape orientation for me, so not really a problem.