Reminder: Update your PC info for the next round of statistics updates

I think the PC info should include if you use a VR headset under GNU/Linux and the model, just like with the gamepad.

I like so much your PC info statistics! :D

I think the PC info should include if you use a VR headset under GNU/Linux and the model, just like with the gamepad.

Done, added.

If you want more options let me know and I will add them in.

Also updated the code so the bit that automatically displays your own personal details in this article is now shared with the profile displaying, hooray for no code duplication and making it easier for me to expand it.


Last edited by Liam Dawe on 25 Oct 2019 at 11:36 am UTC

I think the PC info should include if you use a VR headset under GNU/Linux and the model, just like with the gamepad.

Done, added.

If you want more options let me know and I will add them in.

Awesome!

Literally just bought the components for my upgraded build; so have updated my specs to match. Good timing!

the way it currently displays "Desktop Environment: Window Manager Only" makes it sound like there's a desktop environment out there called "Window Manager Only".

I think you'd have to be very determined to misinterpret it like that. It seems very clear what that option is for. FWIW, back when the survey was a google form, there used to be more options for WM-only users, and it got pretty messy.

Personally, I think the architecture question could be retired given that a number of distros are dropping support for 32-bit and it's therefore unlikely to see a resurgence. The trend has been that ~99.8% (+/- 0.1%) of respondents are using 64-bit for as long as the charts stretch back. I'm not sure there's any value in continuing to measure that.

Done, added.

Done, selected. :D

Funnily enough I've also just upgraded :)

Updated.
Since yesterday I use an unsupported arch (removed antergos packages and repository).

Can I say "btw i use arch?" :D:D

Funnily enough I've also just upgraded :)

Well, it is that time of year -- there's nothing quite like strolling through the golden fields of RAM, swaying in the breeze; selecting a particularly ripe CPU from the processor bushes before wandering over to the GPU orchards to harvest a nice fresh one (before the wasps get to them). I'd give it a few more weeks before digging up a new motherboard though, they're better after a good hard frost gets to them.

Can I say "****?" :D:D

You just did.
Sometimes I wonder why that distro name is so weird and long (four words instead of one)…

I've noticed the total number of GOL users has been dropping for a while, month after month. Less people are using "Click here to continue to be included"?


Last edited by Shmerl on 25 Oct 2019 at 3:22 pm UTC

I second the session stat as well (i.e. "What desktop session are you using?" Wayland / X11).

Some notable points:

* KDE almost caught up to Gnome, after the later got a boost a while ago from the former Unity users.
* Dual booting keeps being replaced by Wine / Proton usage.
* AMD is now at 32.91% for GPUs and 36.67% for CPUs (and keeps growing).

I've noticed the total number of GOL users has been dropping for a while, month after month. Less people are using "Click here to continue to be included"?

A lot of people don't seem to bother updating their details after setting it initially.

For example, right now there's currently around 125 people who haven't updated since 2017 and we only keep people in the statistics if they've updated within two years after that they're wiped for privacy.

The most interesting bit I noticed in the latest statistics is that this is historically the first month when an AMD GPU is the single most popular GPU on GOL. The RX 580 has been rising through the statistics quite rapidly and is now the most popular GPU in the survey at 7.83%, beating the GTX 970 (7.23%) and GTX 1060 6GB (7.73%). The overall marketshare still favours Nvidia of course (62.28% vs 32.91%) but it's good to finally see some variety in the GPU model specific rankings.

I've adjusted the display code, to actually tell you right in the article if you are included currently or not. Hopefully that will make it simpler with much shorter text.

Also added a link to the statistic page on users profiles, below their PC info to hopefully make it even easier to find the page and perhaps pull a few more people in who see it :)

It also has a note in our rotating announcement bar at the top of the website. It can be dismissed by users just like the mailing list bar, hopefully again this will bring a little more attention to it!

On top of that, the Statistics Page will also now tell you when you're logged in if you're set to not be included.


Last edited by Liam Dawe on 25 Oct 2019 at 5:20 pm UTC

The overall marketshare still favours Nvidia of course (62.28% vs 32.91%) but it's good to finally see some variety in the GPU model specific rankings.

The trend however, for AMD is positive, while Nvidia is dropping. Hopefully that will continue.

For example, right now there's currently around 125 people who haven't updated since 2017 and we only keep people in the statistics if they've updated within two years after that they're wiped for privacy.

It actually makes sense not only due to privacy, but because stale data becomes obsolete. It makes sense to purge it if it's not updated, or at least ignore for current values. I.e. in reality those users could have changed their hardware several times already, while their profile if not purged, would still reflect their setup years ago, messing up aggregated values and trends.


Last edited by Shmerl on 25 Oct 2019 at 8:31 pm UTC

For example, right now there's currently around 125 people who haven't updated since 2017 and we only keep people in the statistics if they've updated within two years after that they're wiped for privacy.

It actually makes sense not only due to privacy, but because stale data becomes obsolete. It makes sense to purge it if it's not updated, or at least ignore for current values. I.e. in reality those users could have changed their hardware several times already, while their profile if not purged, would still reflect their setup years ago, messing up aggregated values and trends.

Yes, that is mentioned directly in the article. I was just making a point in this case about the privacy side of it too.