Comments for Przemysław 'rezoner' Sikorski

Comment on 15+ Useful HTML5 Game Developer Resources #1 by Jason

Jason — Thu, 06 Dec 2012 01:17:29 +0000

This is an awesome list :) thanks for posting retroships.com on there!

Comment on 15+ Useful HTML5 Game Developer Resources #1 by Amit Patel

Amit Patel — Wed, 05 Dec 2012 22:10:10 +0000

Thanks rezoner!

Comment on 15+ Useful HTML5 Game Developer Resources #1 by Michal Przybylowicz

Michal Przybylowicz — Wed, 05 Dec 2012 08:03:01 +0000

Very good collection. Thx.

Comment on HTML5 Text particle effect by rezoner

rezoner — Sat, 01 Dec 2012 15:11:05 +0000

Yay thanks

Comment on HTML5 Text particle effect by Tanner

Tanner — Sat, 01 Dec 2012 14:57:46 +0000

Missing “http://” in CSSDeck’s link ;)

Comment on 10+ Tips For Optimizing Websockets Bandwidth Usage in HTML5 Games by Terhix

Terhix — Mon, 19 Nov 2012 15:46:12 +0000

That’s a fast response :). Raycasting is not an issue on isometric tile grid and pathfinding can be done 100% on the client (validated on the server to prevent cheating, but that’s orders of magnitude less computation intensive), so that’s not an issue either. I do have some experience with game making, involving real time on ws, and my first intuition is to not get too crazy on optimizing packeting. You can say I picked a genre that’s not too demanding on either tech or content on purpose, want to just focus on gameplay and getting shit done. Most likely I’ll just go with JSON-RPC for transparency and debugging. BiSON just got my attention as I haven’t heard about it before hence it triggered some curiosity :).

As far as XHR is concerned – while the game will be turn based I intend to keep it rather intense (short time limits per turn), so it’s important to see what is going on and what moves your opponent performs as soon as he does them – which makes XHR as such not really an option. Long polling maybe but then I’m pretty sure ws has higher performance (and is easier to do, which is not to be ignored).

On a side note – it’s quite interesting how much friction HTML5 games are getting in Poland ;). Pozdr.

Comment on 10+ Tips For Optimizing Websockets Bandwidth Usage in HTML5 Games by rezoner

rezoner — Mon, 19 Nov 2012 15:08:52 +0000

I haven’t done any performance tests – however I bet that you will be implementing things like raycasting, pathfinding and such – and these will be far more bottleneck of your code than encoding objects to JSON/BSON — it is so fast that you shall not even consider it in the meaning of performance – for sure not at this stage of project.

Finally, if you are writing turn based game you probably will NOT have any problems with either bandwidth or performance.

You should have a talk with someone who can compare scaleability of XHR vs websockets in turn based games.

Comment on 10+ Tips For Optimizing Websockets Bandwidth Usage in HTML5 Games by Terhix

Terhix — Mon, 19 Nov 2012 14:33:54 +0000

Question on BISON – It says in the readme that it’s performance ranges from x0.5 to x2.0 of JSON – now it’s interesting to see since JSON is (finally) implemented natively, as it is on node, hence any format that’s done in pure JavaScript is at disadvantage. Have you done any performance tests of the encoding yourself in terms of CPU usage?

I’m planning on building a turn based tactics game and bandwidth doesn’t seem to be a huge concern of mine, but if BiSON does perform better, then there is no real disadvantage to using it.

Comment on Conceptual loops for puzzle game by Tanner

Tanner — Sun, 18 Nov 2012 15:32:10 +0000

Really good ;)

Comment on 10+ Tips For Optimizing Websockets Bandwidth Usage in HTML5 Games by rezoner

rezoner — Sat, 17 Nov 2012 09:53:59 +0000

Fixed structure is superior to any other concept – however it is much harder to manage than for example BISON.

As for the second part – yes it is for readability – probably I haven’t made it *bold* enough. Revised the article.