Good idea to jump around project?

That might not be very clear at all but heh, the long question being…

You have a major project you are spending an absurd amount of time on, you are on it day and night and implementing all these cool ideas you have planned and wrote down. You get the basics of something working and during the creation of this you finally realise a way to get another mechanic implemented.

This just happened, I am going through the painful process of getting a proper path finding algorithm implemented, which is simple on it’s own but finding a way to have your entities use the data, another aspect all together.

During the process of this, I quickly added the ability to add and remove tiles using the mouse, in order to test my paths to ensure it is not doing anything funky.

Now I am messing around with the idea of creating a level editor that I can use, currently as it is, I draw my levels in photoshop and simply read the pixel value. This works fine, but it’s not “live”, with this way I could alter my levels, as I play them.

This sounds amazing.

So do you guys think it is a good idea to work on one thing, get the basics down, go do something else then come back to it later? Seems like it would be best, to avoid boredom/frustration!