SteamVR Developer Hardware

SteamVR Developer Hardware

PersonGuy 2. juni 2016 kl. 17:49
SteamVR Play Area Size Stats
A unique requirement of room-scale VR is the physical space necessary to interact with the virtual space. Early developers have asked us what Play Space sizes to target and support, but we've hesitated to make recommendations based on speculative research and surveys. Now we have SteamVR data to share from the first 2 months of Rift and Vive usage!

If you're new to VR, here's the basics for how a Place Space is defined. It's a rectangular area that players keep clear from floor to ceiling and developers keep their required interactive objects within. Room-scale VR requires a minimum of 2 x 1.5 meters, but VR applications may specify a larger minimum required size. Some VR content is designed to scale based on the Play Area size, effectively giving the best experience to players with a large Play Area while still supporting a smaller minimum size. SteamVR also supports Standing-Only VR which is simply defined as a 1 meter diameter circle.

First let's look at the most common specific Play Area sizes used in SteamVR (all units are meters):
18.5% - 1 x 1 (Standing Only)
14.1% - 2.5 x 2
12.4% - 2 x 2
8.0% - 2.5 x 2.5
7.3% - 2 x 1.5
7.0% - 3 x 2.5
6.5% - 2.5 x 1.5

As implied by 18.5% using Standing Only, there are 81.5% using a Play Area 2 x 1.5 meters or larger. And since a larger Play Area supports any smaller Play Area that fits within it, we find it's more helpful to visualize percentage supported at each size in 2 dimensions:

1.5 2 2.5 3 3.5 4 2 81% 65% 2.5 62% 52% 28% 3 33% 30% 20% 9% 3.5 14% 13% 10% 6% 2% 4 5% 5% 4% 3% 1% 0.4%

Here you can see that the most popular Play Area size (2.5 x 2) is used by 52%. We've also compared this graph from April to May and see no notable percentage changes.

In conclusion, an 81% majority of Play Areas used in SteamVR meet the minimum room-scale size and more than half of all Play Areas support 2.5 x 2 meters. 1 in 5 support 3 x 2.5 meters, but increasing in either dimension leaves only 1 in 10 or less.

Numbers updated Feb 27th here:
http://steamcommunity.com/app/358720/discussions/0/350532536103514259/?ctp=2#c133258092253222557
Sidst redigeret af PersonGuy; 27. feb. 2017 kl. 11:03
< >
Viser 31-45 af 51 kommentarer
Tenagra 14. juni 2017 kl. 19:52 
PersonGuy, what kind of rounding is done in those results? If someone has 2.4 x 1.9, do they get counted as 2.5 x 2 or 2 x 1.5?
PersonGuy 14. juni 2017 kl. 20:46 
It doesn't round, it always pessimistically truncates downward. So 2.4 x 1.9 would be in the 2 x 1.5 bucket.
kimusan 4. nov. 2017 kl. 8:46 
Any info on when you plan do give us updated statistics PersonGuy? They are super helpful for us creators and developers in designing room scale VR. Thanks for all the great work so far!
BOLL 4. nov. 2017 kl. 11:30 
Oh yeah, there's been another 8-9 months now :o would be nice to get a refresh! Actually, is there a newsletter to sign up to? ;)
kimusan 4. nov. 2017 kl. 12:35 
I saw the one from 8-9 months ago :-) wondered if when the next one is coming up :)
BOLL 4. nov. 2017 kl. 14:14 
Yeah, that's what I tried to say :p "another 8-9 months" since the most recent update which was after 8 months.

@PersonGuy I put these percentages into a spreadsheet to prepare for an eventual nice stacked graph, and noticed that the first set sums up to ~75% and the second to ~80%. Are the missing users using so weird setups they are not included in the data? Or are those seated users?

Perhaps the explanation is in the description and I'm too tired to see it :P
PersonGuy 4. nov. 2017 kl. 14:21 
It doesn't add up because only a few of the top 'specific' setups are listed. Those stats aren't as useful as the encompassing support percentages. I'll find some time to post a new data set next week.
PersonGuy 10. nov. 2017 kl. 14:00 
Here's SteamVR sessions from the last 30 days. The high-end percentages remain nearly the same as in February with their total install base increasing. The biggest deltas were at the low end with the Standing-Only percent increasing to 31%. More details on how these numbers work in the original post.

1.5 2 2.5 3 3.5 4 2 69% 52% 2.5 45% 39% 20% 3 21% 19% 12% 5% 3.5 8% 8% 6% 3% 1% 4 3% 3% 2% 1% 0.7% 0.3%
kimusan 18. nov. 2017 kl. 5:03 
Thank you very much PersonGuy!
GlockSoftware 12. aug. 2018 kl. 22:52 
Thanks for the stats, great work! Any updates so far?
BOLL 12. sep. 2018 kl. 3:21 
So, a refresh here? 😋 It's been longer since the last update than between the prior ones 😉
PersonGuy 12. sep. 2018 kl. 9:50 
Sorry for the lack of updates. I don't know where things stand since GDPR.
BOLL 12. sep. 2018 kl. 11:18 
Oh yeah, we actually had to implement a few opt in features in our products for GDPR. Collecting anonymized statistics should still be OK, but, it's also about making sure it's actually anonymized even if the data would accidentally leak to a third party. Pretty much.

That said it has been a while since I was at the GDPR seminar 😅 There they did mention that there is a 2 year implementation window though, which it seems nobody even knew considering the online panic, so the hard date was actually pretty soft 😐 If the law firm that held the seminar knew what they were talking about 😋

Edit: Apparently they didn't 🤣
Sidst redigeret af BOLL; 24. sep. 2018 kl. 8:23
jashan 24. sep. 2018 kl. 6:38 
Oprindeligt skrevet af BOLL:
There they did mention that there is a 2 year implementation window though, which it seems nobody even knew considering the online panic

The bad news is that the online panic was due to that 2 year implementation window ending - not due to its beginning. GDPR has been a thing since 2016 (I believe) ... but very few people apparently cared because, well, there was that 2 year window implementation until May 2018.
BOLL 24. sep. 2018 kl. 8:01 
@jashan if so that's hilarious and the law firm that had the seminar had misinterpret something 😅 They told us to take it easy...

Reading the document myself I find references that it is taken into effect 25th of May 2018, but the impression I got was that the regulation in itself included a two year implementation time. I see articles mentioning the two years were indeed between 2016 and 2018 though, so we were tricked I guess 🙄

That said, we did implement our changes anyway, because we had already prepared for it. What I do wonder is why not everyone prepared sooner... I didn't work at my current job when GDPR was announced in 2016 so not sure if they got any information about it at the time.

Thanks for clearing it up 👍
< >
Viser 31-45 af 51 kommentarer
Per side: 1530 50

Dato opslået: 2. juni 2016 kl. 17:49
Indlæg: 51