First results from the GZDoom 3.5.0 survey

[Graf Zahl]

Roughly 2000 users have reported by now, here's the first preliminary results:

67% use Vulkan compatible hardware
17% use hardware which can run OpenGL with all features enabled
12.6% can run the modern render path (i.e. they are OpenGL 3.3 compatible)
3.4% run the legacy (OpenGL 2.x) render path.

3.3% use a real 32 bit system.
There really was 1% of users which only ran the legacy build on modern hardware!
2.7% use a Mac
6.3% use Linux
3.3% used a real 32 bit system
5 users reported Windows XP

So what does this mean?
Obviously for Mac and Linux the numbers are still too small to make any conclusions why they have changed.

Overall, Vulkan capability went from 57% to 67% in a mere 4 months! I think this trend is more than obvious.
Non-Vulkan but modern OpenGL-capable cards went from 34% to 29%, unfortunately this wasn't split like here in the old survey so it's hard to say how this really developed.
This is even more obvious for real legacy hardware which was nearly halved in user share over the last couple of months.

I believe the trend this shows is very evident: The legacy segments are declining rapidly. I am quite certain that, if we do another survey at the beginning of 2019, after this year's Christmas business has run its course, that OpenGL 2.x will have been reduced to a little more than background noise, and this should be a clear warning to anyone still running such hardware: Support for this will end rather sooner than later. The same may be true for 32 bit. This has also nearly shrunk in half over the last 4 months.

[Cacodemon345]

Only a few people use Windows XP? Whatever, I expect the users to go down in just a matter of time. And as for OpenGL 2, I expect it to also go down. I honestly wish it goes down as soon as possible.

[Rip and Tear]

Are there any estimates of how many users are actually updating to the latest version? It’s anecdotal, but a lot of the people I have spoken to who play with GZDoom still use the version from when they first downloaded it.

[Rachael]

If they refuse to upgrade then obviously they have nothing to worry about.

That being said, it's going to narrow what mods they can run - but remember, GZDoom 1.8.6 can still run Vanilla Doom. So it's mostly going to affect their ability to run newer mapsets and gameplay mods, more than anything.

[Kinsie]

17% use hardware which can run OpenGL with all features enabled
12.6% can run the modern render path (i.e. they are OpenGL 3.3 compatible)

I wouldn't have called this being so comparatively small.

[Graf Zahl]

It's only 4.5% less as with the last survey. But why are you surprised? Most people actually upgrade their computers every now and then.

[drfrag]

Are there any estimates of how many users are actually updating to the latest version?

That's irrelevant once the sample is big enough.

That being said, it's going to narrow what mods they can run - but remember, GZDoom 1.8.6 can still run Vanilla Doom. So it's mostly going to affect their ability to run newer mapsets and gameplay mods, more than anything.

And it still runs Brutal Doom. Also they'll miss a lot of features.
But weren't the old versions supposed to be unsafe? Remember that there are better old versions with many bugfixes for old hardware: ZDoom32 and ZDoom LE.

[Graf Zahl]

People won't use some forks if they can get the real thing. If they go to zdoom.org's download page and the archive they find all official releases. They do not read the development threads in the forum.

[Rip and Tear]

Are there any estimates of how many users are actually updating to the latest version?

That's irrelevant once the sample is big enough.

Not necessarily, this could very well be a biased sample.

[Graf Zahl]

That entire question is misguided. Not all users upgrade at the same time - otherwise we'd have a large sample right after the release with no growth afterward. But in reality many users only upgrade when they really need to. This has to run for several weeks until the relative percentages stabilize before the numbers can be considered reliable.

With the 3.3 survey this required roughly 10000 reports.

One thing I definitely can see is some low end users trying to manipulate the survey. There's several who posted multiple reports right after another. Too bad that these are easily to detect.

[Kinsie]

It's only 4.5% less as with the last survey. But why are you surprised? Most people actually upgrade their computers every now and then.

For some reason I was thinking that the Vulkan support statistic was exclusive and didn't include every lower tier. Brainfart, sorry.

[Graf Zahl]

The Vulkan support statistic is exclusive. For the report I split video hardware into 4 exclusive categories and when you add up the numbers it'll be 100%.

[Graf Zahl]

Are there any estimates of how many users are actually updating to the latest version?

That's irrelevant once the sample is big enough.

Not necessarily, this could very well be a biased sample.

Just so you know: Over the last night 200 reports were added - 6 of which were legacy hardware. So either the bias is constant or the 3% user share of legacy hardware is real. The number has been relatively constant over the last few days with an occasional spike upwards or downwards.

[fakemai]

I'd assume from that it won't make a world of difference but there's no reason not to keep the stat checking going for a reasonable period of time, at least as long as the 3.5 versions are the current ones. Also, while this didn't overstep too badly, any time the type of data collected is changed, and especially increased, it should always prompt again instead of assuming it's OK, because it's that sort of scope creep that makes people wary of any form of data collection, and with good reason. It'd be as straightforward as using statsenabled either 0 (if they disagree they probably won't want to do it ever) with anything higher versioned in a similar way as the check for whether it's already been sent for a version.

[Graf Zahl]

The main reason for the new survey was that the last one did not differentiate between full support of all features and modern render path with minimum required feature set.
You may be right that it should have reset the user's setting from last time. But as things are, there won't be any pressing need for another survey.

OpenGL 2 has already been offloaded to a legacy build, and 3.6 will most likely do the same for 32 bit as there's only 1.5% of users running the non-legacy 32 bit build on a real 32 bit OS. That's a tad too low to pass on 64 bit-related optimizations just to maintain backwards compatibility.
Once that separation has been made the main build will have shed all baggage and can develop undisturbed for a bit longer. The next thing that could be dropped is OpenGL 3.3 support but that one's a lot less critical than the stuff we are talking about now. In all likelihood, once Vulkan works, the OpenGL backend will be mostly left alone and a few years down the line, when it isn't needed anymore, be removed.