| Videos |
MondSemmel
Kwon-Tom Obsessive
Puzzles: 6159
Best Total: 7m 47s | Posted - 2013.04.28 21:30:46
Three years ago, I recorded videos of me solving weekly Kwontomloop puzzles. I did 35 videos in total.
This year, someone asked me whether I could do more, so I've started recording more and will do at least 14 in total, more if there's some interest.
I've uploaded the first video today, and will upload the others as soon as they are no longer active weekly puzzles.
You can find the videos here.
EDIT: Feel free to ask me questions if you have any. When posting, you can post puzzles with the Formatting -> Puzzle option. Or you can add timestamps to my videos - simply add '#t=1m2s' or similar to the url, like this. Though I fear that option might be rather imprecise.
Last edited by MondSemmel - 2013.05.01 16:46:51 |
LoopGuy
Kwon-Tom Obsessive
Puzzles: 761
Best Total: 45m 59s | Posted - 2013.04.29 20:58:31
Thanks. This week's Friday and Saturday should be interesting. |
MondSemmel
Kwon-Tom Obsessive
Puzzles: 6159
Best Total: 7m 47s | Posted - 2013.05.03 19:51:23
A pattern someone asked about:
You get 4 additional crosses.
By the way, my thinking on this situation would work like this: a "1" can be seen as a tile in one of these four modes:
In particular, when it's in the mode in, say, the top left, it cannot be in the mode in the bottom right.
So, I often do basically a mini exploration (try both incompatible modes; if one leads to an error, the other one must be correct). That works remarkably often, and is one way to speedily solve the sample pattern I provided above. |
Darklady
Kwon-Tom Obsessive
Puzzles: 5369
Best Total: 9m 37s | Posted - 2013.05.03 23:18:12
I tend to think of that pattern in terms of "diagonal lines". For instance, this has essentially the same deduction:
as does any other situation where you know a line will be going between two diagonally adjacent 1s. It's also half of the reason why you can make the same deduction if you replace the middle 1 in MondSemmel's example with a 3. |
MondSemmel
Kwon-Tom Obsessive
Puzzles: 6159
Best Total: 7m 47s | Posted - 2013.05.04 09:34:34
Right, that works, too. Diagonal lines are probably one of the most important techniques overall.
I find it hard to write down my thought processes for some of these basic patterns, though, because I do so many of them mostly out of experience now...
EDIT: The first 8 videos are now uploaded and available to watch.
Last edited by MondSemmel - 2013.05.05 09:39:44 |
Darklady
Kwon-Tom Obsessive
Puzzles: 5369
Best Total: 9m 37s | Posted - 2013.05.05 17:33:43
Hah, you had a misclick at 0:49 on Sunday's puzzle. You're lucky that line ended up being correct anyway.
By the way, I've been recording a few videos of my own, starting with last Tuesday's. Once it's no longer an active puzzle I'll be posting them too. I seem to have a touch of performance anxiety, enough to lose me a few seconds purely from my hands shaking too much, and some of them ended up being rather slow, but I've managed to get a couple decent times in the lot.
(I wasn't going to say anything until after posting them, but I realized that I should mention it in case you want to keep recording so that we can get a couple weeks worth of videos on the same puzzles. I'm curious about what they'll look like side-by-side.) |
MondSemmel
Kwon-Tom Obsessive
Puzzles: 6159
Best Total: 7m 47s | Posted - 2013.05.05 18:40:29
Hah! Thanks for catching that misclick. How embarrassing^^. Also, how lucky...
Yes, recording definitely impacts one's time. Not by much, but still. I think I had some issues with this in 2010; not so much now.
You're right, it was a good idea to mention that you were also recording videos. I originally planned to stop recording today, but now I'll do at least another week. After all, once I had the recordings set up, it's not too time-consuming to do more. I just hate the uploading-to-YouTube part.
By the way, what software do you use for recording? I use CamStudio - http://camstudio.org.
Last edited by MondSemmel - 2013.05.05 18:42:03 |
Darklady
Kwon-Tom Obsessive
Puzzles: 5369
Best Total: 9m 37s | Posted - 2013.05.05 19:45:01
I have some basic familiarity with video processing, but I haven't done anything quite like this before, so I just looked up the older video topic where you describe your process and copied it. Except, I'm using Camstudio's lossless codec, which produces pretty small files (~21 MB for this Sunday's, just over 3:48 worth of video).
Given the simplicity of the video contents, and how different it is than the content most lossy codecs are optimized for, it's not surprising that a lossless codec would do well. I haven't gotten around to the uploading-to-Youtube bit, but I hear that it can process this codec just fine, and even if it can't it won't be a big deal to reencode it since it's lossless. (I've dealt with ~60GB video files before; 20MB is nothing.) |
MondSemmel
Kwon-Tom Obsessive
Puzzles: 6159
Best Total: 7m 47s | Posted - 2013.05.06 21:47:48
That's embarrassing for me again^^. See, I forgot my process from back then, or that I apparently posted it (it was three years ago, after all), so when I decided to record videos again, I had to search for video recording software again until I eventually recognized the software I used last time...
I'm using the lossless codec now, too. I don't know why I didn't do that back then.
Last edited by MondSemmel - 2013.05.06 21:48:42 |
Darklady
Kwon-Tom Obsessive
Puzzles: 5369
Best Total: 9m 37s | Posted - 2013.05.07 00:36:06
My first video is up. Watching it now, I see that I miss a highlander deduction that could've let me complete it without guessing. (And later, I noticed a much sillier mistake...)
Last edited by Darklady - 2013.05.08 22:11:03 |
Darklady
Kwon-Tom Obsessive
Puzzles: 5369
Best Total: 9m 37s | Posted - 2013.05.08 22:21:39
Here are some links for our videos side-by-side:
Tuesday
Wednesday
Thursday
Friday
Saturday
Sunday
Monday
Tuesday
Wednesday
Thursday
Friday
Saturday
Sunday
All done now!
Last edited by Darklady - 2013.05.26 11:36:51 |
MondSemmel
Kwon-Tom Obsessive
Puzzles: 6159
Best Total: 7m 47s | Posted - 2013.05.10 19:05:38
This Sunday, I'll have recorded three weeks worth of videos, so we have about two weeks to compare. Do you think that is enough for interesting comparisons, or should I do more? (Today, I forgot to turn off the flashing rectangles during recording, which soured my attitude towards it quite a bit -.-.)
Last edited by MondSemmel - 2013.05.10 19:05:57 |
Darklady
Kwon-Tom Obsessive
Puzzles: 5369
Best Total: 9m 37s | Posted - 2013.05.10 23:41:14
I think that sounds good. |
MondSemmel
Kwon-Tom Obsessive
Puzzles: 6159
Best Total: 7m 47s | Posted - 2013.05.11 09:56:33
You can't essentially answer "yes" to an "or" question =(. |
Darklady
Kwon-Tom Obsessive
Puzzles: 5369
Best Total: 9m 37s | Posted - 2013.05.11 22:08:27
Yes!
Haha, I meant that the two weeks should be fine.
Looking at the videos so far, it's interesting how close our times tend to be despite having distinctly different styles. The big differences seem to come down to particular causes, such as me spending a minute chasing a wrong guess on Friday's puzzle, while you got the right guess first try. In comparison, on Thursday's puzzle, I was doing more mental explorations while you were doing more explicit trial-and-error, yet we finished within seconds of each other. |
MondSemmel
Kwon-Tom Obsessive
Puzzles: 6159
Best Total: 7m 47s | Posted - 2013.05.12 20:35:23
Comparing our videos:
Wednesday: How'd you get the line above the 3 in this pattern:
Mistake or a pattern I don't understand? I'm really puzzled by this^^.
In that video, I also find it interesting that for a long time, neither of us noticed that the loop in the bottom right closed early.
Thursday: The tradeoff between deduction and trial&error exploration is an interesting issue in general. We usually don't have the time to estimate which approach will be faster - there'd be no time savings left.
Friday: The time difference between the right t&e guess and the wrong one can indeed be huge, and I essentially got lucky. My attitude towards explorations would differ a bit in that I try to "cut my losses early", so to speak: I try an exploration, and if it isn't fruitful within seconds, I try another one, until something works out. That strategy sometimes fails (when none of the explorations end soon), but usually serves me well.
I do everything I can to avoid nested explorations. It's a question of memory: In my videos, you can sometimes see me retry an exploration because I forgot what line started the exploration...
And in general, I find it interesting how you rarely use crosses, while I use them obsessively. But I think we already discussed that a few years ago. On average, I'd guess that difference makes me a bit slower in weekly puzzles, and a bit faster in weekend puzzles?
Last edited by MondSemmel - 2013.05.12 20:37:18 |
Darklady
Kwon-Tom Obsessive
Puzzles: 5369
Best Total: 9m 37s | Posted - 2013.05.13 02:56:23
Quote: Originally Posted by MondSemmel Comparing our videos: Wednesday: How'd you get the line above the 3 in this pattern: |
I think that I saw that not having a line there leads to the loop below getting cut off, but it might have been a mistake.
Quote: My attitude towards explorations would differ a bit in that I try to "cut my losses early", so to speak: I try an exploration, and if it isn't fruitful within seconds, I try another one, until something works out. |
I used to do that, but now I try to keep explorations going and push it through as long as it seems like it's getting me somewhere. See, I figure that if I don't get a great time, it's not so important whether I get a merely "good" time or a bad time - so I'll go for broke and hope I'm on the way to getting a great time. It probably does make my times more volatile, though. (It worked out well for me on the most recent Friday puzzle.)
I'd also originally pick guesses that seemed likely to lead to dead ends quickly, whereas now I try to pick guesses that seem likely to be correct.
Quote: And in general, I find it interesting how you rarely use crosses, while I use them obsessively. But I think we already discussed that a few years ago. On average, I'd guess that difference makes me a bit slower in weekly puzzles, and a bit faster in weekend puzzles? |
For a while I was trying to use zero crosses, but more recently (now that I've broken my original habit of using them obsessively) I've started throwing them in again in spots where I wouldn't be able to see them easily from looking at the lines. I'm really not sure what the relative impact of our techniques is - this last week, I was faster on most of the weekday ones and slower on the weekend, but I've had many weeks where things have been the other way round, too.
Last edited by Darklady - 2013.05.13 03:00:02 |
MondSemmel
Kwon-Tom Obsessive
Puzzles: 6159
Best Total: 7m 47s | Posted - 2013.05.20 20:09:27
All my videos are uploaded and visible now. Altogether 35 videos from 2010 and 22 videos from 2013. |
Darklady
Kwon-Tom Obsessive
Puzzles: 5369
Best Total: 9m 37s | Posted - 2013.05.26 11:45:54
All the side-by-side links are up now too. |
