Utopia Developers’ Update 2nd September
This week the blog is from Sean :) Please post comments in the thread as normal.
Hi everyone,
This week, it’s me, Sean, bringing your weekly report. It’s been a long week of bug fixing and performance testing. At this point I think we are both either working, eating and thinking about work, or sleeping and dreaming about work.
Last week, to help us try to catch up with the 260 or so bug reports that had come in, we took the alpha down. The rationale for this was that we were receiving so many new bug reports each day that we didn’t have any time left to get any of the existing ones fixed. We’d hoped that we could make a big enough dent in them to get the alpha back up this week, but realistically it’s going to take some more time. Of the 260 reports mentioned above, there are fewer obvious duplicates than we expected, and more than expected are turning out to be very time consuming to properly investigate. Often, after investigating a reported issue it turns out to be a symptom of another bug, but each one still takes a lot of time.
At the start of this WoL age we were aiming at mid-September to re-launch on new code, but that’s not going to be possible now if we want to get these bugs fixed.
As things stand, we are aiming to get the current bugs fixed and the alpha back online next week. We will then want to orchestrate a few tests with everyone’s cooperation, and then lift the white-list and try to encourage a lot more people into it so that we can get some real-world performance figures that we can extrapolate from. If things go well, this means that we will start a new Genesis age on the new code in 2-3 weeks (16th-23rd September), which is about when we were hoping to end the current WoL age. If Genesis goes smoothly, then we can restart WoL more or less at will. However, assuming that it doesn’t go entirely smoothly, and that we want to leave the two server ages staggered, it will be sensible to assume that WoL can’t be started on new code until week 1 or 2 of October, in total 2 or 3 weeks later than expected.
We’re sorry to break the news of a further delay, yet again, but hopefully people can start planning now for this longer-than-anticipated but shorter-than-normal age. We feel it is a better option than launching the game in a buggy state, as many companies would.
In other news, we’ve made lots of improvements to site performance. We’ve made a 10-fold increase to the tick speed, which was one of the major factors slowing down page loads before. We’ve also made a lot of other small speed improvements, that are adding up to much improved site performance. The ranking process was also causing problems when it ran every midnight. We’ve completely changed the way this works to actually make it several thousand times faster. In fact, it’s fast enough now that it runs in real-time, with new ranks written to it every tick (instead of every 24 hours).
I’m going to continue to hunt down the last of the performance gains, and will probably be doing so for the next day or two. After that I’ll be back to solving bugs with Brian. Working together on the remaining bugs that we’ve prioritized is going to take us into next week, at which point we’ll open the alpha again.
On to questions…
… There were a couple of general points I can address to start with:
1. Starting resources in the Alpha. When we re-open the alpha we can reset everyone, so that everyone starts at the same time. Hopefully this will help.
2. Site visuals. We understand people’s concerns about visual issues, and are aware of them ourselves. We will improve things when time permits, but getting the behaviour of the game correct is an issue with higher priority. There are some improvements that we plan to make sooner - see below.
3. Frames. As Brian said the other week, we’re going to take a look at the possibility of pinning the nav bar using a different approach. Frames are nowadays considered bad web design, which is why you see less and less of them around. Some of the reasons for this are listed here:
http://www.htmlhelp.org/design/frames/whatswrong.html.
Palem: That’s a lot for 3 weeks. Care to comment? *set alarm for 1 week*
Well observed.
Allanon KT: This should be a major concern. [re: mobile version]
Thanks for your feedback. There’s a lot of major concerns at the moment, but we’re doing our best to give all of them time as soon as possible.
Bishop: I do agree that the splitting up of spells is something that i still dislike.
This is a consistent theme in a lot of the bug reports we’ve received. We are planning on getting both kinds of spells back onto one page.
The past two blogs have been either Sean or Brian, but never both. Is everything ok between you two?
LOL. Since we ended up actually taking turns to write the blog, we thought we might as well drop the impersonal pretence and write a bit more directly. Maybe if I ask him nicely, Brian might say something now. [Brian: Can't talk, bug fixing...].
Cheers,
Sean