Runescape Dev Tracker

Happy Monday everyone!

As a heads up, this update deploys at 11:00 Game Time today - but as the update is deployed, the servers will remain offline for approx 30 minutes as we perform some additional server maintenance.

Qualcomm Snapdragon or Exynos model?

Fix for this issue is in the next engine update

Greetings!
This is not available for Hardcore Ironmen or F2P. You can read more about this here: https://secure.runescape.com/m=news/a=12/runescape-in-february

Appreciate the comment :) hope you enjoy and learn something new!

It's an incredibly long and difficult job. You first have to design how the new core mechanics will work and then finetune those. You have to do a huge amount of balancing to make sure it all lines up as you intend and that going for 99/120 is still similar to before. Then you have to design how literally every other thing that plugs into a skill would work. For Mining & Smithing this took more time in development than the core skilling just because of how much content there is in this 20+ year old game. You have to change quest requirements and any M&S mechanics in those, you've got to consider things like invention perks and how those must be redesigned, you've got treasure trails to consider like clue that require M&S armours, there will be a million random miscellaneous items from around the game that somehow plug into M&S like a collector's insignia, you'd have to adjust the entirety of the skill guides, and then you've got all the drops from basically every combat NPC ever to...

How do you choose what scenarios to test for an update?

Context dependant, we'll usually look at what changes are made and to what pieces of content it affects. We'll look at existing game mechanics and functionalities to see how they interact now. We'll look for any mysterious ways something could interact and so on.

And how do you automate testing?

As i've detailed elsewhere about automation, it's not something we've typically had but overtime are introducing it more and more. Feel free to check out my other answers in this thread for some more info on it :)

Are there some tests that are performed before every game update or is it always case specific?

Yeah absolutely, we'll look at typical flows of play and golden paths to ensure things still work as expected. Even this might depend on what changes have been made though.

Can you detail the lifecycle of a project (EG: how does a project go from idea to release)?

Ideas may come from a few places. First we'll look at what type of update we need to deliver. Then we'll take ideas from past brainstorms, player feedback, data etc. and developers will begin designing the high level concept, then core elements and such. Once pitched and greenlit we'll begin development that may go from skeleton > core mechanics > polish for example, and then release. That's a very abstract and high level overview.

How large is an average project and how long does it take on average from conception to release?

It varies a lot though so none of this is set in stone, but i'd say the average skilling update/quest ranges from 2-4 months. Bosses tend to take anywhere around 3-6+ months. Skill reworks and new skills have previously taken around 18+ months.

Does Jagex (specifically R...

We work in scrum teams that consist of devs and QA, as well as any artists necessary for that project. We have a well established workflow at this point so it's easier for us to plan ahead with what assets we may need from each discipline (character, animation, audio, marketing etc.).

Our producers ensure things are going as smooth as they can at all times, although the responsibility also falls on everyone to be raising any concerns or issues and apprehending anything coming up. Some of my other answers in this thread will give you some insight as to how our teams work too :)

Beta servers actually take a lot of effort and time to setup and maintain, on top of already having to keep one live game updated and maintained.

You also need more resource and time to collate and parse any feedback you get. In the past we've also not seen an incredible amount of participation. Considering everything involved, we have to weigh up the effort required vs. the output gained.

That's not to say they're useless or we don't like them by any means. I personally think in the past we've used them exactly when necessary (M&S rework, weapon diversity) and i know there's people internally who are advocates for it too. It's just finding the right balance to deliver it.

There are a few different ways we go about doing this.

One way is just via talking to players. We'll either jump in game and discuss content, or take feedback and discuss on various social media sites.

Another is via our analytics and data science team, they'll often provide data and attribute mechanics to that data.

We do actually A/B test various things too though obviously not with content updates for bosses and such. One popular place for this has been with new players around Burthorpe/Davendale/Tut Island etc.

In pretty much every scenario i've seen this occur it's just talked out. Pros and cons are weighed up, design intention and objectives are considered etc.

Sometimes we might prototype something to see how A or B work, or pitch it to a council or the wider RS team to get opinions and consensus. Don't think i've ever seen something get so serious that people have "called it a day".

I actually enjoy when design problems occur because i love problem solving and it's very fun to work with others and explore ideas and consequences, and come to a conclusion that works (or works best).

There's a distinction to be made here between testing for QA purposes, and playtesting for playability, for example.

In terms of QA testing, typically mechanics and structures are tested as a solo player or using multiple accounts and checking to see if you get expected results from whatever you're doing. Checking to see what occurs when you do an action or that numbers look correct and so on.

When it comes to playtesting, we'll group up with others and try to take the boss on as players would. This is done pretty regularly, Raksha, Rex Matriarchs, and all 4 fronts of EGWD had a wider team playtest once a week during development to ensure it was getting the attention it needed at all times throughout development. In the past i know we've even brought players in to try them out, and sometimes even outside of team playtest folks such as the combat council will try taking group bosses on together.

See my answer here for unit testing https://www.reddit.com/r/runescape/comments/sttlkk/game_development_reddit_qa_with_mod_breezy/hxgdb3t/

RE: Integration testing, that's typically done in our "Release Control" or RC phase. Content will be merged with an environment as close to the live game as possible and tested with its own criteria to suit.

Has anyone worked on AI for testing content or even for the actual game but has been scrapped off?

We do have "bots" in a sense but they work differently to how you'd imagine them. Generally AI would probably be too bespoke or too general to suit what we'd need for testing.

Within content QA we don't typically have unit tests no. I think unit testing within games generally is rare simply due to how games work. That's not to say it doesn't happy at all, and in some departments it may be more common such as engine/platform.