There were quite a few comments from our visitors, especially on the Daily Killer Sudoku page, which makes that and other pages load slow. Therefore, I have moved some of the old comments to this new category.
If you’d like to read what other people had to say about our puzzles you may find the old comment by clicking on the “comments” link below or here.
UPDATE (some 5 years later): I reopened the comments on all pages. Feel free to add your comments there, it means a lot to me!
Really appreciate the daily killers. Thanks.
Unfortunately today’s text file is corrupt like it was a couple of days ago. v 0.4 of ‘Perfect Sudoko’ shows very large numbers instead of the usual ones.
Looking at the text in the file I think it’s a generator problem because all the numbers are big compared to previous versions.
Let me know if I can help more (I’m a VB.NET developer).
Andy
Andy, sorry for this. The reason is that this puzzle was generated by the new version of software, which uses slightly different file format.
I’ll try to avoid this in the future.
By the way, if you re-enter the cage sums by dividing the existing ones by 256 you will get the correct cage sums.
HI all. just wondering what is the order of difficulty for all this puzzles.. I assume COOL is the easiest what about the rest?
EASY -> COOL -> THINKER -> BRAIN -> IQ -> INSANE
Now I’m not sure if THINKER should be higher than BRAIN instead…
About my walkthrough… Great for JC to spot the hidden triple move but I must say hidden sets are not ordered too high in my “weapondry of techniques”… Cause for harder puzzles I must work with candidates/pencilmarks so hidden sets often get masked by other more obvious moves. Probably use them more when I deal with easier puzzles and apply the “pen only & no marks” approach…
Also, excellent for Matthew to discover the shortcut in nonet-4! Well if ND were the designer I bet he’d have clogged up that back door at the first place… No offence to JC because it’s a very well-thought-out puzzle nonetheless. So Mikey don’t make another fight suggestion please… ^_^
Hi!
Any chance of an Insane Killer any time soon? Are there any Insanes in the archives to be found?
cheers,
Myriam
I just updated the sums from the image. I am curious why the extra bits in the newer format. I have written my own killer sudoku interface that interprets your file format. The only thing I don’t have is a solver.
Well, I’ve been looking at J-C’s new, beta-test puzzle (“B2”) & so far it looks like it’s got most backdoors clogged up pretty nicely…. so I don’t think anyone’ll find a quick shortcut to it, though I & J-C may be surprised!
Just started drafting a 2nd set of tips for the site — I’ll add to it gradually over the next few days. At the moment it’s got an example of the use of “bound pairs” from puzzle #4, & then a quick rundown on other techniques.
heya, great puzzles!
i was just wondering could you put a list of the order of difficulty of the killers. i know you call them brain, IQ etc but im not sure which order they go in!
oops just read above!
I noticed from some of the comments that the number in parenthesis after the puzzle description mean something. For ex. INSANE (675028) ON 01/20/06. Could someone please enlighten me?
Michael
Dear djape,
Loved the Insane today…very elegant.
Keep up the good work!
Regards,R.
ne1 hav more old puzzles? teh arcive only has 5.
I was flying through today’s insane. What’s Dj talking about? This isn’t difficult! Nonet 2 completed – no problem. Nonet 8, got half the numbers but know the permutations for the rest. Same with nonet 9. Suddenly, bang! How can I make the 8 in nonet 5 with 3 numbers, when the only ones available are 1,2 and 3? Also, how can I make 19 in nonet 5 with 3 numbers when I only have two numbers, 6 and 3? Waaaaarrrrrggggggghhhhh!
Time for bed. Try it again tomorrow when I’m sober.
Mike Japan
im completely stuck on todays insane 675028
can someone give me a start please!!!
What’s the fuss with this insane puzzle? A little tricky but did it in an hour and a half. Once I’d sorted out the 13, 14, 15 combinations in nonet 2 and 8 the rest falls into shape.
raj,
Here are some hints in the Insane:
1. “45” on N2 -> r2c3/7=5, “45” on N8 -> r8c3/7=7, “45” on C1-3 -> r2/8c3=5, “45” on C7-9 -> r2/8c7=7, “45” on R1 -> r2/3c5=10, “45” on R9 -> r7/8c5=8, r1c4/5/6=18, r9c4/5/6=9.
2. The 8-cage in N1 contains a 1, so r2c3 is 2 or 3 (part of a 6(3)-cage), so the square r2c3, r2c7, r8c3, r8c7 is:
2 3 or 3 2
3 4 2 5
One of this options will lead to a conraction, we’ll see later which one…
3. “45” on N8 -> r7c46+r8c46=28= either {9874} or {9865).
4. 12-cage in N9 is {156},{246} or {345}. {345} is rejected because r8c7 is 4 or 5, furthermore, if the 12-cage is {246} then r8c7 is 5 (can’t be 4), and in no way is the 5 in N9 part of the 29-cage.
5. 29-cage in N69 is {9875}. So the 5 in the 29-cage is in r6c8.
5. Now we’ll find a contradiction for one of the options from step 2.
If r2c3=3, then the 8-cage in N1 is {125}, r2c7=2, r8c7=5, 12-cage in N9 is {246}, r7/8c9={13}, r6c9=9, r9c4/5/6={135}, r7/8c5={26}, r2/3c5={37} (all other options are rejected), 19-cage in N3 contains a 7 (no other 7 on row 1), and doesn’t contain a 5.
6. Now let’s look at the 14(3)-cage in N3/6. It must contain a 5 in r2c9 or r3c9 (r8c7=r6c9=5, & no 5 in the 19=cage), so the rest of it is 9. This 9 cannot be {18} or {36} (r7/8c9={13}). It cannot be {45} (Already got a 5 in the 14-cage) and it cannot be {27} (already got {27} in nonet 3. The 2 in r2c6, the 7 in the 19-cage).
7. And we get:
r2c3/7=[23], r8c3/7=[34], 12-cage in N9={156},…..
And you carry on.
Beautiful puzzle today. Worked out 1st time for me 😉
How I started today’s Insane…
1. 24/3 in R9123 = {789} -> 12/3 in R9C789 = {4{26|35)} (my lingo for “must have a 4” !)
2. 45 on R9 -> R9C456 = {1(26|35)}. R78C5 = 8/2 = {(26|35)} (can’t be {17} because R9C456 must have 1). -> 17/5 in N8 = (12356}.
3. R2C3 = [1..3]. Playing with 45 all around, we get R8C7 = [3..5] but it can’t be 4 because R9C789 must have one. -> R8C7={35}.
4. R78C6 add up to 13 or 15. -> R78C4 add up to 13 or 15 -> R8C3 = {24}
5. Each of R78C4 & R78C6 add up to 13 or 15 out of {4789}. So on is {78}, the other {49}. R78C4 can’t be {49} because 17-13 = 4, twice a 4 in the cage. -> R78C4 = {78}, R8C3=2, R78C6 = {94}, R8C7=5
6. In N5 : 8/3 = {125}, 19/3 = {9..}, 18/3={8..}
…
Oooops, forget about my previous message. Must be too tired, I forgot the case {156} for the 12/3 in R9C789.
DJApe, if you can, lease delete both theese messages :-S 😉
While Friday’s insane looked very pretty, I reckon that symmetrical puzzles are easier to solve than randomly designed puzzles because you can apply the same techniques twice.
Mike Japan
PS. Being sober is also a definite plus to puzzle solving.
Dear “Killerholics”, there was this interested killer variant posted in the sudoku.com forums, which I think everybody would probably be interested. First I’ll post the links here:
http://www.apathysketchpad.com/images/uploads/doublekiller.png
http://img359.imageshack.us/my.php?image=facerippergrid2sd.png
These 2 links are refering to the same pic but in case the 2nd link is down the 3rd will be the backup mirror. The creator of this is the brother of PaulIQ164, a very active member of the sudoku.com forums. Here is their blog page for this puzzle: (which might go down sometimes)
http://www.apathysketchpad.com/index.php?id=pic/killer
And this is the thread from where I found this great puzzle: (some minor spoilers included)
http://www.sudoku.com/forums/viewtopic.php?t=2914
I think this one is not too difficult to solve, but it definitely gives you a fresh taste about killer sudokus…
I’ve just been away for 2 weeks and was very disappointed to only find 5 puzzles in the archive (and those were very difficult to find). Any chance of archiving the ones from the week before last??
The Insane quad-symmetrical one is a nice one — took a long while for me to figure out the best approach, but once you do it falls into place quite quickly. (I haven’t run through the tips Shai’s posted but basically: you can use the 45 rule to find the sum of r19c4 = 8. I managed to locate 1 in r2c4 early on, I forget the exact logic now; so r3c4={23}, & with the hidden 8(2) in r19c4 = {26} or {35}, this left only two possibilities for the 19(3) in the middle, which both involved a 4, so in turn cracked the 8(3) cage to its right….)
Here’s how I started yesterday’s wonderfully Insane puzzle:
1. We can get R2C4=1 — it cannot go in R2C3, and since 45-rule on N2 gives us R2C3+R2C7 = 5, those two cells must be {2,3}. So R2C4=1.
2. Doing various permutations of the 45-rule will get us the following:
a. R2C3 & R2C7 are both {2,3}
b. R2C3 & R8C3 are both {2,3}
c. R2C7 & R8C7 sum to 7 (so the latter is {4,5}
d. So going clockwise around these four cells, starting with R2C3, they are either 2,3,4,3 or 3,2,5,2
3. In either instance, you find that R23C6 & R78C6 are, in some order, {49} and {86}. This is based on jiggling the handle a bit with R78C4, which eliminates some combos in N8.
4. So we’ve officially wiped out 4689 from C6. This tells us R46C6 is {125}, so R1C6 and R9C6 are {37}. And from what we know about R9C13…
5. So we now know that R1C45 = 11, and we know R1C4 + R9C4 = 8. Based on what we know about R1 and R9, we can eliminate all options for R1C4 except {36}. But making R1C4 a 3 (and R1C5 an 8) does not allow for both R23C5 and R23C6 to be filled in. This opens things up significantly.
ZD – New Mexico (USA)
That’s a very happy 8 up there, eh? 🙂
ZD
Thanks for the Insane Killer!
While Shai’s method works to solve Friday’s insane puzzle, I wanted to find a solution that didn’t involve lengthy test of hypotheses, and with enough persistance I found one. Here’s mine:
Some notation first: R2C3 = A, R2C7 = B, R8C3 = C, R8C7 = D.
Using the 45 rule gives A + B = 5, C + D = 7, A + C = 5, B + D = 7. (Also, use the 45 rule to split the large cages in N2 and N8. (they are 18/10 and 8/9, respectively.)) R1C123 must contain 1, so A is 2 or 3, B is 2 or 3, C is 2 or 3, D is 4 or 5. The naked 2,3 pair in A and B means that R2C4 is 1. R9C123 is {789}, so R9C789 is {156} or {246} (it cannot be {345} due to the 45 pair in D), so it contains 6 and R9456 contains 3. The 4 and 5 in N9 must be in D or R9C789, so R6C8=5.
R456C6 contains a 1, and R1C6 + R9C6 = 10, so R1C6={678}, R9C6={234}. R1C5 + R9C5 = 9, R1C5={45678}, R9C5={12345}. R1C4 + R9C4 = 8, so R1C4={3567}, R9C4={1235}. The naked 23 pair in R2 (A and B), means that R2C5={4678} R3C5={2346}. So, the 9 in N2 is in R23C6. So, the 9 in N8 is in R78C4. So, R456C4 contains 8. The 4 in N8 must be in R9, so R9C456={234}, and R9C789={156}, so D=4, B=3, A=2, C=3, R3C4=3. R9C4=2, R1C4=6, R9C6=3, R1C6=7, R9C5=4, R1C4=5, R2C5=8, R3C5=2.
We can now use the 45 rule to get more information:
R4C1 + R4C2 = 5
R4C8 + R4C9 = 15
R6C1 + R6C2 = 12
R6C8 + R6C9 = 13
So, R6C9=8. The 3 in N9 is at R7C9, R8C9=2. From here, you keep plugging away in a pretty routine manner.
I hope somebody finds this helpful and/or interesting. I had a lot of fun with this puzzle.
Thanks so much for the great puzzles you put up on this site!
-Cindy-
It helps to sleep… 😉
For insane 675028:
1. N1 : 8/3 must have 1 -> R2C3={23}, R23C4 must have 1.
2. 45 on C123 -> R28C3=5 -> R8C3={23}
3. 45 on N8 -> R8C37=7 -> R8C7={45}
4. 45 on R1 -> R1C456=18, R23C5=10.
5. 45 on R9 -> R9C456=9, R78C5=8
6. 45 on C5 -> R19C5=9
7. 9 of C5 can’t go in R78C5=8/2 nor R19C5=9/2 nor R23C5=10/2 because R23C4 must have 1. Thus it must go in 18/3 of N5.
8. N5 : 8/3 must have 1 -> 18/3 can’t be {189} -> 19/3 must have 8.
9. N8 : 17/5 can’t have 9 nor 8. 17/3 can’t have 8, thus it must have 9 -> 18/3 must have 8 and {45}. 18/3 can’t be {855} -> R78C6={86}, R8C7=4 -> R8C3=3 -> R2C3=2.
R78C4={95}, R78C5=10/2={17}, R9C456={234}
From here on, the rest is quite simple.
11. R9 : R9C123=24/3={789}, R9C789={156}
12. N9 : 29/4 = {5789} -> R6C8=5, R7C78+R8C8={789}, R78C9={23}, R6C9=8
…
A very quick-fire way to resolve the Insane last friday is noticing that r23c5=10, r78c5=8, so r19c5=45-10-18-8=9.
Now 9 could be safely eliminated from these 6 cells and must be locked within r456c5.
This, together with r2c4=1, r3c4={(23)}, r19c4=8={26|38} leave r456c4 to be {(45678)}, which has to be {478}.
Now all 3 cages in n5 are settled and the rest would be fairly easy…
Here is a walkthrough of the first few steps:
1. 6(3)-cage in n1/2 = {123}, 8(3)-cage of r1c123 = {1(25|34)} -> 1 eliminated from rest of r1/n1 -> r2c3=2 or 3.
45-rule on n2 -> r2c37=5={23} -> naked pair on r2 -> r2c4=1.
2. 45-rule on c1234 -> r19c4=8={26|35} -> 2,3 eliminated from r45678c4 (complex naked pair).
3. 45-rule on r1 -> r23c5=10={(234678)}, 45-rule on r9 -> r78c5=8={(123567)}, r19c5=45-10-18-8=9={(12345678)}.
Now 9 is locked in r456c5! -> 19(3)-cage of r456c4={(45678)}={(47|56)8} (sub-combo). But r19c4={26|35} elimininates {568}!
So r456c4={478} -> naked triple on c4/n5 -> 8(3)-cage of r456c6={125}, 18(3)-cage of r456c5={369} -> naked triples on c5/6!
4. Now r23c5=10={2478}, which could only be [82] -> r3c4=3, r2c37=[23], r9c4=2 (hidden single), r1c4=8-2=6.
Now r23c6=16-3=13={49} -> naked pair on c6/n2 -> r1c56=[57] -> r9c56=[43], r78c4/5/6={59},{17},{68} respectively.
From here on r8c3=17-5-9=3, r8c7=18-6-8=4 and the rest would follow smoothly…
ps1: Now that I look at it my route was a bit similar to J-C’s…
ps2: I think I get djape’s subtly expressed intention from the “deleted post”… Sorry!
Good puzzles
Trying to get to grips with the Killer Thinker_12_r-2 last week. I can understand why row 4 columns 4 and 5 are 468 and 642 but why can you say with certainty that it becomes 46 and 64 – from there I can solve it quite easily
Hi, Udosuk.
Thanks for the link to the ‘variant’ killer puzzle. I had fun with it and solved it first go. And it was useful to have on DJ’s day off!
Pommy Mike… give it a go but it’s definitely not a drinking puzzle!
Here’s another link for more killers. The website “Sudoku-A-Day” provides daily regular sudoku, but only two killers a week.
Although not as challenging as DJ’s, the 2 new killers come out on Sunday so you can have something to do until DJ posts the Monday puzzle.
http://www.angelfire.com/games6/sudoku/killer.html
their archive contains about 40 killer puzzles, rated up to 5 stars
Bob, USA
Cheers Bob.
Mike Japan
Could someone enlighten me to what the numbers in parenthesis, after the puzzle type, mean?
I asked this recently, but cannot find my question listed.
Thanks.
Michael, I try to answer to all comments, but sometimes I miss one or two.
The numbers in parenthesis are numerical representation for puzzle difficulty. This can be used to compare two puzzles with the same level of diffculty, but it really shouldn’t be used for that because it’s very arbitrary.
It’s mainly there for me, to determine the level and also it’s there so you can refer to a specific puzzle by it’s number (altho in theory two puzzles could have the same numerical difficulty).
Thanks djape,
I thought I was missing a clue or something.
I promised you all a new year present, so see:
http://www.diceboard.co.uk
This site presents my first Killer Modulo Sudoku.
This is targetted purely at yourselves and as I do not intend to be anymore active than Nate or Jean-Christophe please comment here at Djape’s site.
I have 15 or so more puzzles solved, which I will put up if you show interest in this first one. They vary in interest/difficulty from being slightly harder than the basic killer to pretty well impossible.
Happy New Year
HATMAN
Sudoku-a-day now provides a killer everyday in addition to the 2 on Sunday, so it’s now 9 a week, but recently he/she had a schedule change??? so now it gets updated late at night, but since the change, he/she has been inconsistent in updating, so hopefully soon it’ll be back to normal.
HATMAN,
I absolutely looooooove the concept. But the construction, visually, is difficult to interpret. Any chance you might go with the Times’ or DJApe’s design? Or does someone have an applet that he or I could easily use to convert it to a more familiar format?
This actually does bring up a possibly interesting point: I would (monetarily?) appreciate a program that, by dragging boxes here and there, would allow me to construct my own killer puzzles, at least when it comes to cages and nonet-boundaries looking unambiguous. Does DJApe’s software allow this? If not, does anyone else have that capability?
Dave, 1 a day plus 2 on Sundays would be actually be 8, not 9 ! Anyway, plenty on the acrhive for me to be going on with !
ZD – I agree, I’ve printed off the puzzle (no colour printer), and it is very difficult to see where the cages end.
After a busy weekend I’ve been catching up on the comments. I enjoyed last Friday’s INSANE (675028) – took me quite a while, but under the influence of alcohol I’m afraid. (Not alone there, I see, after reading the comments!!!) Also enjoyed yesterday’s BRAIN (91206) – which I believe had the wrong file name – and today’s IQ (201827). It is COLD here, so no golf – hence lots of Sudokuing time. So thanks for all the sites with more killers, I’m about to explore them.
Gaynor, UK, wishing to emigrate to warmer climes
PS How are the sudoku’s going, Adele, up to starting on the killers yet?
I just phrased it wrong. Actually it is 9, one for each day, plus 2 in the weekly special sudoku section, but like I said, it’s been inconsistent. It was last updated Saturday and before that last Tuesday.
ZD
Appologies for the lack of visual clarity. I had just been putting these puzzles together in Excel and had not prepared any drawing setup. Then Udosok’s posting of the unknown-number puzzle on Saturday pushed me into going public rather hastily.
I have now put up the Excel workbook, hoping that that might make it (visually) clearer.
On On
Hatman
Hatman
Hi DJ,
Excellent IQ (201827) today. Really had to work the grid after the initial start, making it quite fun!
Gaynor
Where are you at? I’m from RI. Played golf 5 times last week & twice this week. A very unusual winter for us. It takes away from my Killer times.
Hatman:
I have posted a clearer version of your first modulo puzzle at:
http://members.shaw.ca/fdkr2/
Can you check to see if this is correct.
Don
I live in Hertfordshire, England – whereabouts are you – obviously a golf addict too (as well as Killer Sudoku addict) to play 5 times in a week! I normally play 3 times, sometimes 4, though often only 9 holes especially at this time of year. Can’t see me getting out this week with the forecast (not keen on carrying) – hence my comment about emigrating!
I see that it’s an Easy killer today so will have to download some others to keep me busy! Of course, I could do some housework . . . . .
For Hatman’s kimo 1, fdkr’s djape-style pic is nice but I think it’s better for the cage sums to be on top-right or bottom-right corner so when solving on paper you could add the ten’s digit in front of it… So I made another one using Excel:
http://i1.tinypic.com/mjyhj5.png
I have some experience myself at generating killer puzzle pics using Excel so I could share with you guys here. First you better choose cell widths at 0.5 and heights at 5. That gives you tiny small squares. Then you group each 8×8 squares to be a single cell in the puzzle grid, leave an empty “layer” on all 4 sides, and merge the central 6×6 cells. That way you could border the cages with thick dotted lines which will not interfere with the grid. For walkthrough pics I use 10×10 for each cell for more space to put the candidates in…
Hatman,
I loved your mod-10 puzzle. Please post more when you can.