5000, Views of Modern Rome

Views of Modern Rome, Ravensburger Jigsaw Puzzle, 5000 Piece.

Me and my wife together finished a 5000 pieces of jigsaw. We attempted this project couple of years ago before we had our son, but failed since none of us had a clue how we could finish it back then.

But this year is different, I guess time is all we have. So we gave it another shot, of course this time we were mentally prepared before we started. That is very important.


Solving puzzle isn’t easy especially when it comes to the magnitude of 5000, because taking 500 pieces as a typical task, dealing with 10 of them at the same time could be a bit overwhelming.

We do what we normally do to the pieces this time, ex. sorting them by colors and shapes. One of the challenges is that it would end up with handful of pieces, for instance, 100 minimum. In some way at the beginning it made me think sorting only won’t work.

In the end I applied my engineering thinking. BTW, I don’t think that is the best skill to have when it comes to jigsaw, but that’s all I had. So I see there’re some pillars in the picture, therefore I just focused on major pillars and picture frames and turned it into a grid looking map. I know not all jigsaw have this type of feature, so I know this might not be a generic approach.

What it gave me lights later is that we could sort the pieces smarter. At one point I wish there’s a robot handing me any piece with any feature I ask for it. Of course that machine doesn’t exist, even it does, I don’t really know how to describe the feature I want precisely most of time. However you just have to sort them in a way which makes sense to you at that moment.

The reason is that, only after that, you can end up with less amount of pieces. And if I’m lucky after narrowing down to 10–20 pieces for one feature, I can solve that feature under 30mins. We used our guts feeling and wiggled and wiggled into these lucky moments in the end. I think this might be what’s called optimism, but sorting is the survival manual.


Now it comes to the better part, how do we sort? I watched an amazing girl on YouTube who did 20K pieces. To be honest, I don’t think I learned much from that video except to aware of her hard work. One interesting skill that she demonstrated is that she always prepared couple of board of pieces arranged in specific color and shape before she attempt to fill in the blank of the feature. Of course I’m talking about hundreds of pieces at a time. It’s quite a skill that she can quantify the metrics of pieces for the feature and stick with it throughout the entire process. This requires years of training, no doubt about that, especially after I know her a bit better through her other videos.

We of course didn’t take her approach, that also explains why we were pretty slow as well. Though me and my wife had no systematic approach throughout the entire process, we used a mixture of approaches driven by our six sense tinted by each of our strength. In short, we used our best knowledge at that time.

By Feature

I already mentioned, my specialty is to identify things by feature. So I normally look at the pillars, the window frames, the lines which can separate things a bit into smaller chunks, therefore manually creating some pattern, even it’s temporary. IMHO this is a quick way to reduce number of pieces drastically. Most of time it does work, and it worked for this case. However it introduces high chance of error, and the error could be very big sometime. For instance if I identified one piece wrong at the beginning and mistakenly categorized it, this could be very problematic down the chain.

Either you consciously or subconsciously doing it, I believe sorting by feature is a very nice move to exercise if you manage to apply at the beginning of any project.

By Color

Most of time the difficulty of a puzzle is measured by the uniformity of the color range. If the range is small, it’s very difficult. Of course, if the color doesn’t show a pattern to the human eye, it could be very difficult as well. Of course this statement is very subjective, it varies from person to person.

The challenge of color is two folds.


How close a range would you like to define for this feature? If too narrow, you’ll end up with small amount of choices, but you might miss the right choice. If too broad, you’ll take more time to sort and end up with large amount of choices. Not to mention human eye isn’t very good at differentiating subtle change of color, especially under different background.


This actually turns out to be a surprise to me. When the color reaches dark side of spectrum, everything looks similar. My wife’s eye is much better here, she can tell the subtle difference in green, yellow, and red even when adding them to a black. Large chunk of sky, ocean, a sheet of scarf, becomes her specialty. We were joking, if we don’t want anyone noticing something in the future, we’ll just add some dirt or darkness to it.

By Shape

I used to not doing puzzle by shape except corner pieces, because I thought it’s a bit of cheating. Of course I was wrong. When dealing with tons of pieces, the only way to know you are doing it right is matching two shapes via edges. And even that is not 100% guaranteed, since it turned out there were 10 pairs out of 5000 pieces were interchangeable. For instance, you can swap one with another and make perfect shape fit with all four neighbors. Amazing!

This is where we did differently, we only started to apply sorting by shape when it came to final touchup, the last 500–1000 pieces. My wife favored more towards this way early on, but i think the main reason we didn’t use this as our main sorting way is that we couldn’t afford with this extensive amount of work. If we sort them with color and shape simultaneously, we might just have to kill ourselves, since we can’t agree upon the color, and we still haven’t come up a way to store that many sorted pieces. Damn. Something to think of.

Anyway, this jigsaw has 5 shapes, 2 out of 5 is not as common as the rest 3. Especially one of them is almost the most popular one (two inward and two outward in opposite direction). So I applied a rigorous looping method in the end only using shape.

  • find the possible shapes for that location
  • try with one shape with least amount of choices
  • if not working, try the next shape with larger amount of choices

The idea is very simple, I used brutal force, which my wife called me cheating this time. But I don’t care and don’t want to exhaust myself either, so I first make an educated (mostly 90% correct) guess on the shape and then try one shape only to rule it out until I find the right one.

Though effective, I didn’t want to apply that from the very beginning. Because I need a computer to do that, and I don’t have a computer. That’s the problem. No, that’ not my only problem, I don’t know how to build that algorithm either.


When we were struggling with the jigsaw, we were excited to notice solving it might not be our only challenge. The picture is almost 2 meters tall and wide, we run into couple of problem including transport and mounting.


You might start this on the floor, on the table, but eventually it can’t be there forever. So you need to move it, and during the moving, you don’t want to lose any chunk, or even worse the entire thing might collapse and fall apart.

Most likely you need to flip the picture to glue them together at some point. This step is very challenge. Make sure you know what you are doing before you do it :) What we did in the end were a newbie one, we put glues on the top and then the bottom by inserting sheets of paper. The end result is not very pleasing to look at, but at least we can flip it and carry it up and downstairs.


Mounting turns out to be a profession. Although there’s people to help you commercially, the picture is out of the normal size that common business supports, nor do we want to spend a fortune in mounting, considering the jigsaw itself is about $100.

Once again I watched some videos online, and followed tutorial in terms how to build a small picture frame from scratch. I skipped the framing part (as you can see in the attached picture), since I figured that’s the money you can pour in if you prefer to spend more.


It took two of us months to finish this jigsaw, maybe one, maybe two, we weren’t really interested at counting weeks. After it, we weren’t very excited or proud either, believe it or not. Since then we worked on some other projects but none were comparable in the scale as this. I think it might take us sometime to recover and be mentally prepared again. All in all it is a time-consuming process.

During the process, I did learned a few thing.

  • complicated thing can be done, of course, and by me as well
  • keep learning from others, keep inventing and wiggling
  • solution need time to unfold, working with time is a skillset

When I grow older, the weapon that I find essential is time. You need to know when to wait, when to prepare, when to fight, when to waste, when to … It’s not just about timing, but the right sense of estimation of length of time that you need to do any of them, not to mention there’s ripples of time frames inside each other.

Ironically time is something I had a lot when I was young, but had no idea how to use it back then. Now it made me think, does it make sense to teach kids use time wisely?

Front-end Engineer

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store