It's a fact I experience on both linux and windows platforms, whenever Spring is minimized or otherwise made not to render, it has to catch up when you display it again. The catching can take up to a minute if it's been hidden for a long time.
I assume this is a known "bug", what's the reason for it?

i can usually instantly alt-tab between spring and other stuff. (and do it all the time)
sometimes here is a delay/hang but i think that is not supposed to be the norm.

that's actually different; alt tabbing doesn't actually stop it from rendering, it works fine for me too; can you try minimizing it for half an hour and then restoring it again?
the reason i wrote "otherwise made not to render" is because on linux you have the concept of multiple views/workspaces which cause the same adverse effect as minimizing if you're not on them

It does invoke a modifier key bug though, the number of times I've alt tabbed back into spring and found my scroll wheel behaving as if I still have alt pressed

This issue has existed for years. It's been reported on the forums, on mantis and in various other places a number of times but it still hasn't been fixed

I have the same behaviour as gajop.

sorry, i used "alt tab" to mean switching to different programs/minize.
Normal minize is no problem too though.
I often press win-key+d to minimize everything and spring plays along nicely.
As said, sometimes there is delay/hang (spring becomes a black window) but that is rare.

Sometimes game goes in 'Not Responding' state when is minimized or alt-tabbed.

I find it weird how my laptop catches up to games faster than my desktop yet my laptop is inferior in every way.

I remember my AMD K6 380Mhz machine with 128MB ram outperforming a Duron 1400 with 512mb. It was only when you tried to run games with better graphics that you saw the Duron move faster

Probably this issue?

http://springrts.com/mantis/view.php?id=3073

unlikely as it happens on both singleplayer and multiplayer games;
the way i see it, the bug (which manifests an error in the form of hanging spring until it catches up) is that the simulation isn't run while it's minimized

That is a different issue! The catching up after minimized is because of a messaging system that is part of the engine design. The Simulation sends various messages to the rendering thread. When minimized, no rendering occurs, so there will be a lot of unprocessed messages waiting in the pipeline. It may take a couple of seconds for the rendering thread to process all these messages when the window is opened.

A more serious consequence of this design is that it generates a memory leak. I.e. if minimized for very long, the message queue may eat so much memory that it leads to crash.