John Romero posted some interesting stuff about a casual game his old company (MonkeyStone) made about three years ago – Congo Cube.

Congo Cube Post-Mortem
Congo Cube Design Doc
Congo Cube Playable Demo/Purchase Page (on Real)

I really like the music and the art style of the game. It definitely feels a bit dated, but that’s to be expected. The gameplay is innovative, but not quite there, IMO (a bit simple/repetitive, but then again, that was more typical 3 years ago). Still, it’s cool to be able to see a bit into someone else’s development process.

BTW, I do not do detailed design docs for my casual games. I just start coding and have a rolling to-do list…

If you’re in the Banana Bay beta program (and have gotten a user id and password previously), there is a new version out today. Download it from the same location, with the same user id and password as last time.

At this point, MOST of the people who asked to be in the beta program have been invited to test the game, but not quite everyone (a few potential competitors I’m leery of including, at least until the game is closer to shipping, and a few unlucky folks I just haven’t gotten to yet).

But the point is, I’m running low on fresh beta testers (those who can look at the game with fresh eyes), so drop me a line at psteinmeyer A T newcrayon D 0 T com if you’re interested in beta testing. Please indicate if you work in the software/games industry in some capacity (again, it doesn’t rule you out, but may affect the timing of when you get to see the beta.)

And many thanks to the beta testers who’ve already sent feedback. The game is better because of it – most of the changes in the last couple weeks are the result of beta feedback.

Just a quick progress update. I am slowly but surely making progress on my next casual game. I’ve seeded it out to a couple of beta groups, but not yet to everyone who asked to be in the beta. I plan to use most of the folks who asked to be in the beta, but I want to break it into stages, so that there’s ‘fresh eyes’ looking at the game every week or two.

I’m definitely taking longer on this game than I did on my last one (Bonnie’s Bookstore). Perhaps I have the second system problem going on – I really want this one to be as perfect as possible. We’ll see. The beta testers so far have been quite positive, and that’s encouraging.

The last few days the web host (Dreamhost) for my company site (www.newcrayon.com) has been, to put it charitably, inconsistent. The main site was up and down, and when it was up, was often very slow.

I also run my blog aggregator (www.casualgameblogs.com) on the site, and noticed this weekend that it was down. Whereas the rest of my sites were solid by this morning, this site was still down. I went to check it out in an ftp program and discovered that all the permissions had been set such end users couldn’t see the content on the pages, and that I (the admin) couldn’t change anything, including the permissions themselves. I’m guessing this was a hangover from a server rebuild or something on Dreamhost’s side. It was frustrating, though after an e-mail to Dreamhost’s tech support, things were fixed within a few hours.

I’m not the only one experiencing problems with Dreamhost lately.

My company site isn’t really mission critical at this point. I’m only generating a small amount of sales through it (most sales are through portals, not directly from my site). But my next game has a global high score feature that must access the server, and the server has been spotty ever since I implemented the feature several months ago.

It’s hard for non-experts to determine who to use as a web host. There are hundreds of providers, all offering virtually the same service, at very similar prices. But the important variables you can’t readily determine from looking at their sales pages are those relating to reliability and speed. Web hosts offer cheap packages because they put many customers on the same server. A bad host might put 3X as many customers on a server as a good host, but customers can’t readily tell who will be good or bad until you go through the pain of signing up and getting all set up.

I finally found a useful, objective site for comparing quality of web hosts. This site does an effective job of using user reviews to separate the wheat from the chaff. The first host I used, Host Excellence, gets lousy reviews that match my lousy experience with them. Dreamhost is more middle-of-the-pack. And the host I use for this blog, Lunarpages, gets a very good score, in line with the very good experience I’ve had with them.

Since I have a personal account with Lunarpages and a biz account with Dreamhost, I was able to do a comparison. I put up a simple PHP script on both servers, then tested time for it to execute and appear on my browser (using the excellent firefox plug-in Fasterfox).

The script was really short and simple, and didn’t access a database at all. On Lunarpages, in 10 tries, the average time was 244 ms (milliseconds). On Dreamhost, the average was 449. That includes some time shuttling back and forth across the net and for Firefox to render the page, which should be the same for either one. If you subtract 150 ms for THAT, then the host-side time difference is even more dramatic.

I then reran the test 50 more times on each server, only counting the times the query took abnormally long (over 500 ms). On Lunarpages, that number was 0 – even in the worst case it was under 500 ms. On Dreamhost, it took > 500 ms (i.e. half a second) 7 times, with a worst case of over 6 SECONDS.

So I set up a second account on Lunarpages for my biz stuff, and should be migrating shortly.

[edit – I just set up the mySQL databases on Lunarpages, and can now do a more complete test, including database access. The results are even more strongly in favor of Lunarpages now. The SQL queries (isolated from all the other stuff) consistently take < 10 ms on Lunarpages, versus 50-100 ms on Dreamhost] [edit 2 - fixed an error - I set up the new account on Lunarpages, of course (earlier version I said I'd set it up on Dreamhost)]