PCGamingWiki will use a Single Sign On (SSO) system to bridge wiki and forum accounts which is ready for testing. You may login using the 'Login with PCGamingWiki' button on both the wiki and the forum, which will soon be the only option. If you have any issues please message Andytizer on Discord.
By Caleb Wysor
As Valve's Steam Labs launch three new experimental features today, one has caught the interest of many Steam users: its new algorithm for game recommendations based on Valve's machine learning technology.
Valve says the Interactive Recommender uses a "neural-network model that is trained to recommend games based on a user's playtime history, along with other salient data." The data is modified by two sliders that users can edit: one ranges from "popular" to "niche," while the other slider ranges from "older" to "newer" games.
Rather than base recommendations around genre or category, the Interactive Recommender instead scans through Valve's data sets to find other Steam users with similar tastes. The model then recommends titles the user might enjoy based on other games played by like-minded Steam users.
Valve also says they discard most category information about the game when entering it into their model.
"We don't explicitly feed our model information about the games. Instead, the model learns about the games for itself during the training process. In fact, the only information about a game that gets explicitly fed into the process is the release date, enabling us to do time-windowing for the release-date slider. It turns out that using release date as part of the model training process yields better quality results than simply applying it as filter on the output," Valve said. They also discard information about review scores and tags, relying only on popularity and age variables.
Users worried about this experimental technology replacing their regular Steam recommendations have nothing to fear for the time being. Rather, Valve says users who want to try the Recommender will have to specifically choose it under the Steam Labs experiments section. Regular Steam recommendations will still function as before.
Since their algorithm discards the categories most other game recommendation algorithms operate by, Valve also claims that developers won't have to worry about optimizing their game description to make it more likely to be recommended.
"The best way for a developer to optimize for this model is to make a game that people enjoy playing. While it's important to supply users with useful information about your game on its store page, you shouldn't agonize about whether tags or other metadata will affect how a recommendations model sees your game," Valve said.
If you want to try the Interactive Recommender, head over to the Steams Labs experiments section.
(via PC Gamer)
Valve have started a new initiative with PC gaming's flagship application named Steam Labs. As the name might suggest, it is centred around adding experimental new features to Steam for people to test.
Steam Labs has started life as three experiments for us to jump into. The first, Micro Trailers, are game trailers limited to just six seconds in length displayed on a single page.
The second - and perhaps most eye-catching - experiment is the Interactive Recommender. This is designed to take a look at your Steam preferences determined by your library, wishlist, playtime and other parameters to recommend your next purchases. You can choose to exclude or include releases by popularity or age, as shown below.
The final experiment is 'The Automated Show' which, rather than being something straight out of an episode of Black Mirror, is a 30 minute video which showcases all of Steam's latest popular releases, with a link to the store page for each game that interests you.
Some have been quick to point out the similarity between Steam Labs and Google's own ill-fated labs experiment. Time will tell as to just how much attention Valve will pay to this new addition. My personal favourite is the Interactive Recommender, although it's not without its initial problems. Its tag-based exclusion system means that games which deliberately mis-tag themselves may still show up in your list despite not being a game that interests you. There also doesn't seem to be a way to exclude any NSFW results. These would be simple fixes though, and with a few iterations the recommender could be the go-to tool for scouting your next purchase.
What experiments or features would you like to see added to Steam Labs in its next round of updates?
Canonical, the developer of the popular Linux distribution Ubuntu, recently announced their upcoming removal of support for 32-bit applications. Many Linux application developers were extremely displeased with this development as it would mean a large percentage of older Linux games would no longer be supported on the most popular Linux distribution.
Valve has been at the forefront of development for gaming on Linux, and 32-bit support has been critical to their initiative. Pierre-Loup Griffais, a major programmer on Valve’s Linux team, had previously implied on Twitter that Steam would be “switching our focus to a different distribution” in response to Canonical dropping 32-bit support from Ubuntu.
After backlash from Valve and from Linux gamers, Canonical backtracked on their stance. They announced on their blog that “We will change our plan and build selected 32-bit i386 packages for Ubuntu 19.10 and 20.04 LTS. We will put in place a community process to determine which 32-bit packages are needed to support legacy software, and can add to that list post-release if we miss something that is needed.”
In the aftermath, Griffais posted on the “Steam for Linux” discussion forum, stating:
He also states:
At the very least, Steam will be compatible with Ubuntu for the time being. Thankfully Canonical was willing to listen and accommodate to community backlash.
Are you a Ubuntu gamer? If so, do you plan on sticking with Ubuntu?
Hi, my name's Voxarp. I learned of PCGamingWiki from the late TotalBiscuit's videos back in the day and I used to help contribute from time to time but now days I'm mostly just a user. Thanks to everyone involved for providing such a great wiki!
Recently Valve has introduced a new feature of SteamPlay called Proton which is a variant of WINE that allows Linux and possibly OSX to run Windows-only games. It's an exciting time!
What the community seems to desperately need is a resource to:
Report game compatibility, possibly with a WineDB type rating of Bronze/Silver/Gold/Platinum - right now all we have is a spreadsheet on Google Docs. Present fixes and workarounds for games that don't work out of the box. Provide clear delineation between Proton versions, a game might work properly on one Proton version but not another. Possibly provide an overview page with statistics on the percentage of games working, with a table that lets you sort by criteria such as Proton version or compatibility rating. I know this is a lot of effort and not to be taken lightly, but I feel it fits right in with PCGamingWiki's objective of providing fixes and workarounds for every single PC game.
Valve's Announcement: https://steamcommunity.com/games/221410/announcements/detail/1696055855739350561
SteamPlay on Reddit: https://www.reddit.com/r/SteamPlay
linux_gaming on Reddit: https://www.reddit.com/r/linux_gaming
Compatibility Spreadsheet: https://docs.google.com/spreadsheets/d/1DcZZQ4HL_Ol969UbXJmFG8TzOHNnHoj8Q1f8DIFe8-8
Who's Online 0 Members, 0 Anonymous, 55 Guests (See full list)
There are no registered users currently online
Recently Browsing 0 members
No registered users viewing this page.