HTML5 games are still a bit too slow for me. I'm on a modern machine, but if I'm doing a heavy compile in the background it shows how slowly HTML5/canvas rendering actually is. [If it could only start using my GPU?]

I'm not sure if Flash uses my GPU, but it's fast enough so that I can play a similar game while doing background compilations.

I thought Flash only uses the GPU for video rendering.

Have you tried it in IE9 or one of the recent builds of FF / Chrome?

Regrettably no. I was not aware there was a FF or Chrome available for my OS (which is OS X) that takes advantage of the GPU.

Is there anything in the W3C spec to have the option to treat a canvas image differently than a regular image?

When I am playing a canvas game I do not want to drag, zoom, or otherwise disturb the area that makes up the display.

Having a separate, more helpful, context menu and keeping focus are both the best things and the worst things about flash.

In what way is it innovative?

Heh, looks like we might be past the point where somebody can remake a game in <canvas> and call it innovative. 10/11/2010, save the date.

HTML 5 and Line Rider can both be considered innovative. Remaking Line Rider in HTML 5 isn't innovative.

It's not just a remake. This is interactive and has more physics to it, from what I can see. Reminds me of Ski Stunt.

Reminds me of Cyclomaniacs. Except no points for stunts...

I'm guessing anigbrowl has never seen Line Rider.

Well it was the use of Canvas that I thought was innovative - but I didn't think to see if LR has been reimplemented in HTML5.

I must admit I posted partly because I was sure this would have found its way to HN already and wanted a lazy way to find an existing thread about it!

To me, merely reimplementing an old idea using a new technology doesn't full under the umbrella of innovation.

Regardless, I'm going to be playing it for the next half-hour.

'Z' switches the bikes direction, the help should probably mention this.

For having 60 points and being on the home page still, it doesn't have a lot of positive feedback in the comments.

I for one am enjoying it, but had not seen linerider and co before. Cheers to the developer.

The first level presented really should be one you can just press gas and watch as you fly around, not one that requires you figure out how to lean.

The level chosen seems to be random. Those in http://canvasrider.com/tracks/popular seem to be more reliable.

Although after playing a few levels, the game is partly to blame. It's far too easy to die - I don't think I've had more than a few seconds of play before dying and having to restart. Gets frustrating.

The game has great physics and fun game play, but it fails socially. It really needs a few learner slopes to start with that introduce the concepts.

Crowd-sourcing of track design is a great idea, but the game needs a better way to filter out the crud. Most of the designs are personal experiments, and even the better ones could really use a few checkpoints.

Elastomania, anyone?

I wasted so much time at college playing Elastomania from (I think) a floppy disk. Great fun.

XMoto for me, but yeah.

Classic "genre". Most recently Trials HD: http://www.xbox.com/en-us/games/t/trialshdxboxlivearcade/def...

I keep stabbing myself to death on that fir tree by the start >.<

Yeah same here. How come I can't accelerate as fast as the ghostrider with the up key? Does the autorepeat rate matter or something?

Damn you, I've lost the whole morning on trying to master this game! :)

Feels a lot like Trials HD for XBox 360: http://www.youtube.com/watch?v=sQAmbTEOJG8

Or X-Moto.

http://en.wikipedia.org/wiki/X-Moto

I wish the physics model worked more like cyclomaniacs.