41: Gish

Gish was the first commercial game released by a team including Edmund McMillen, who would go on to actually make Jim care about platformers some years later with the punishing but brilliant Super Meat Boy. It garnered a fair bit of attention at the time – it was 2004 and the Indiepocalypse hadn’t yet begun, but there were the beginnings of these small-scale games heavily reliant on one mechanic.


The mechanic in question is a physics engine combined with an adaptable character. Gish is a ball of tar who in his regular state gloops and slops his way about the world – bounce him from side to side, time it right and you can use his elasticity to get reasonably high jumps. However, he can also take on some other properties at the push of a button, making him heavy and inflexible, or sticky so he can walk up walls, or slippy so he can fit through smaller gaps.

These different states can be combined in an intuitive enough way – when I needed to bring down a pillar, it seemed obvious enough that I’d be sticky to climb up it, before being both sticky and heavy to bring it down. If you need to move fast downhill, a combination of heavy and slippy did the trick. At times, early on, this seemed quite interesting, but there’s only so many ways you can combine three different states and that’s when the many, many problems begin to emerge. If problems were rocks, this game could sink a battleship. In zero-g.


To start with, there’s a confusion I think on what type of game they’re actually trying to make, as it oscillates between a one-off challenge in the vein of SMB and a regular platformer with lives and levels. You can pick a difficulty and that affects the amount of damage you can take, you have a limited number of lives, and that says regular platformer to me. But then, losing those lives doesn’t actually matter, it doesn’t send you back to the start, it just resets your points counter, so that says challenge. But then there are secret areas as well to find and that adds some weird aspect of exploration as well, and occasional bosses, and the whole thing is a big mess of conflicting impulses about how you should be approaching the game.

The levels themselves start off as a fair enough way of showing off the different ways you can use your properties – enemies to crush, gaps to leap, walls to climb and so forth, but this quickly degenerates into another mess. The difficulty curve is more of a series of difficulty speedbumps, with all the frustration that entails. Levels vary massively in length and challenge and include some truly infuriating, scream-out-loud moments. It’s possible to screw yourself over without realising it, blocking off exits with moveable objects by mistake or not making a jump when you had momentum. There’s many bits which were included which baffle me but I think the best and most complete example is level 3-5, which you should give a quick perusal here.

You fall down a slope, crushing little pink rats, and avoid blocking off your path with the columns you’ve got to ascend. Then comes the first headscratcher, a barely moveable rod thing that you’ve got to knock into a lever. I never figured out how to do it as quick as that video, but the best approximation I came up with was being heavy, sticking myself to it, and gradually rocking back and forth to shuffle the rod along. Eventually I got this down to about half a minute of mindless tedium.


You go up through the gap that pushing that lever has revealed and stay away from the spikes, and then comes the screaming moment, where you have to transfer between two moving blocks while surrounded by spikes. You can barely jump from a standing start, becoming sticky has a tendency to stick you both to the ground and the object you’re jumping to, so about 9 times out of 10 you just plummet and die on the spikes, and 1 time you’ll make it through on sheer luck. Even on that video, on that perfect run, the guy had to drag his screaming bag of tar along the spikes because it’s such an awkward maneuver to do accurately.

Two tedious, lengthy bits done and the level’s barely half over. There’s still the same amount of time to fuck up in and have to do those insanely frustrating bits again. And again. That’s the first unforgiveable thing, the insanely difficult jump should be right at the end or right at the start. The second unforgiveable thing is that this level very nearly broke me and it’s halfway through the game, there’s never anything that’s this hard again. The third unforgiveable thing is that you never have to use those skills of transferring between moving blocks or shuffling along a rod again, it’s learning tricks without any points. Overall, this is bad design, and it’s especially surprising coming from the guys that would make SMB’s compact, replayable masterpieces.

The colour palette may also have stuck out to you there – the first three chapters are entirely in black and grey and subdued dark reds, so it’s not even that much fun to look at, there’s no joy in the aesthetic beyond Gish’s maniacal face. Add in a background track that varies from Seattle jazz to screaming guitars and you have a pretty textbook case on how not to make a platform game. The physics engine may well be accurately modelled but it’s a bit of a random fucker to use effectively. The word for Super Meat Boy was precision. To succeed, you had to time your jumps and movement and momentum as close to perfect as possible. When you completed a level, you could sit back and justifiably think “Yes. I have mastered that level”. You may wish to go back and do it better now you know the tricks. In Gish, there is no such feeling, beyond one of weariness and a slight suspicion that it was all just blind luck that you got through this time.


All told, I’ve spent far too long plummeting off surfaces I should be sticking too, or trying to jam a tarball into a pipe it will only fit into if you attack it from the right direction, or trying to propel crates accurately using a vague direction and sudden tarball expansion. For what it inspired, and what it showed was feasible, Gish should perhaps be recognised. For every other aspect of the game, it should be burnt, crushed, burnt again, packed into a lightweight travelling case, compacted, and exploded. It’s an infuriating, exasperating, ugly game.

0 stars, 5 hours played

This entry was posted in Uncategorized. Bookmark the permalink.

3 Responses to 41: Gish

  1. This is a pretty good dissection – by which I mean it agrees with my opinion <_<. Like you say, it was a herald of better times to come. But not worth a shelf in the Museum Of Indie really.

  2. I remember when this came out. It was the original Indie darling. Linux support helped to push it in those circles a lot. However, when I actually got around to playing it, I discovered what you did. Also, I always felt the controls were a little off. Perhaps it would have felt better with mouse controls?

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s