Do you post your GPU to Steam Survey?

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

Do you submit your GPU details to Steam for their online survey statistics?

  • Yes

  • No


Results are only viewable after voting.

Schmide

Diamond Member
Mar 7, 2002
5,581
712
126
Poll are just that polls. If steam encompasses near 60% of the people who game it would still be considered an incredible sample and probably spot on for any trend in gaming. Most polls are conducted with sub single digit percentage points.

Edit:

No, & this poll sums up exactly why the Steam survey isn't all that accurate.

I beg to differ.
 
Last edited:

ronnn

Diamond Member
May 22, 2003
3,918
0
71
Does steam say what percent don't allow their hardware to be surveyed?

The interesting questions: are anandtech forum members more paranoid then hardocp members?

Does steam really respect your wishes when you request no information be sent in?

Can they read your mind?
 

MrK6

Diamond Member
Aug 9, 2004
4,458
4
81
Poll are just that polls. If steam encompasses near 60% of the people who game it would still be considered an incredible sample and probably spot on for any trend in gaming. Most polls are conducted with sub single digit percentage points.
Edit:
I beg to differ.
Statistics 101 failure: this is not a random sample. However, this doesn't mean it's useless, it just means you have to understand the limitations of the survey.
 

Schmide

Diamond Member
Mar 7, 2002
5,581
712
126
Statistics 101 failure: this is not a random sample. However, this doesn't mean it's useless, it just means you have to understand the limitations of the survey.

How is this not a random sample? By that logic every time a sample is taken and someone says they don't want to participate in it, it becomes invalid?
 

MrK6

Diamond Member
Aug 9, 2004
4,458
4
81
How is this not a random sample? By that logic every time a sample is taken and someone says they don't want to participate in it, it becomes invalid?
There are many reasons it isn't but yes, that's one of them.
 

Schmide

Diamond Member
Mar 7, 2002
5,581
712
126
There are many reasons it isn't but yes, that's one of them.

That was not an answer to my question.

Look I took statistics in college. There is no hard fast rule on how to sample anything. If you have proof otherwise, present it and explain it.
 

myocardia

Diamond Member
Jun 21, 2003
9,291
30
91
Yes and I do log in on multiple machines. (mostly 2) So I cancel out myocardia. Ha. But I have 5 functioning machines, so I get 2/5 a vote?

You have five gaming machines? I call bullshit. Oh, and if we're going by how many computers we own, your 5 is only 5/9 of my current 9.;)
 

Schmide

Diamond Member
Mar 7, 2002
5,581
712
126
You have five gaming machines? I call bullshit. Oh, and if we're going by how many computers we own, your 5 is only 5/9 of my current 9.;)

No worries, I have more than enough spare parts to make up for it.

Side note: I pulled out my probability/statistics book, it was many years ago. I guess you would consider this a biased sample. So ok.

Edit: Quick Listing my machines currently running.

Athlon II 620 gtx260 <- logged onto steam
q9550 dual 4870s <- logged onto steam
Phnom II 720 9600gso Linux
e5200 4670 <- logged onto steam
Athlon 2800+ 9600gso <- logged onto steam 2 months ago

Just added: Shuttle K45 with a e5200 (been meaning to see if this works for a couple years)
and my old q6600 and a 4870 on a P45 <- this will go to my brother in law soon
No HD on these last two.

+ a laptop and an OLPC
 
Last edited:

Idontcare

Elite Member
Oct 10, 1999
21,118
58
91
No worries, I have more than enough spare parts to make up for it.

Side note: I pulled out my probability/statistics book, it was many years ago. I guess you would consider this a biased sample. So ok.

I was hoping it would be, as such the statistic would serve more as a conservative upper-limit on the actual ratio of survey reporters.

If I go to the densest population of gamers on the AT forums and ask this question and at best we get a 60:40 ratio of steam survey reporters then I have to question how or why I should expect the ratio to actually be even higher in favor of steam were I to start to make the sampling any less biased.
 

GaiaHunter

Diamond Member
Jul 13, 2008
3,628
158
106
I was hoping it would be, as such the statistic would serve more as a conservative upper-limit on the actual ratio of survey reporters.

If I go to the densest population of gamers on the AT forums and ask this question and at best we get a 60:40 ratio of steam survey reporters then I have to question how or why I should expect the ratio to actually be even higher in favor of steam were I to start to make the sampling any less biased.

Guess you can try the PC gaming sub-forums and then compare.

I would be surprised if the difference was significant though - AT forums is a hardware place to start with.
 

MrK6

Diamond Member
Aug 9, 2004
4,458
4
81
That was not an answer to my question.

Look I took statistics in college. There is no hard fast rule on how to sample anything. If you have proof otherwise, present it and explain it.
No worries, I have more than enough spare parts to make up for it.

Side note: I pulled out my probability/statistics book, it was many years ago. I guess you would consider this a biased sample. So ok.
Basically. And I reread what I wrote, I wasn't trying to be difficult, sorry if I was, but just more so to ensure that it's clear what the limitations are on the this survey as well as the Steam survey. I have a funny feeling that this thread and it's associated poll will turn up again in some "great video card forum debate" at one point or another :p.
I was hoping it would be, as such the statistic would serve more as a conservative upper-limit on the actual ratio of survey reporters.

If I go to the densest population of gamers on the AT forums and ask this question and at best we get a 60:40 ratio of steam survey reporters then I have to question how or why I should expect the ratio to actually be even higher in favor of steam were I to start to make the sampling any less biased.
Exactly. It just goes to show you how little weight such a survey carries as far as serving as indicative of anything, except maybe trends. The trends portion of the survey might have some statistical significance, as it's more of a "random sampling" than the rest of the data presented. Ultimately though, it's more "gee whiz" data than anything else.
 

0roo0roo

No Lifer
Sep 21, 2002
64,862
84
91
its the best data we have.
steam encompasses many games now so the sample population is great.
steam doesn't lie
steam survey is a click ok and you are done thing increasing likely hood of response.
hardware junkies are not shy about their rigs
in all likelyhood that would bias the numbers towards more powerful hardware.
but the numbers show plenty of old stuff as well. so it looks like folks aren't that shy in general.
 

0roo0roo

No Lifer
Sep 21, 2002
64,862
84
91
Statistics 101 failure: this is not a random sample. However, this doesn't mean it's useless, it just means you have to understand the limitations of the survey.

indeed, but all surveys are not random, people self select whether to bother. and phone/other types are so onerous only nitwits respond. steams survey is different being so simple for the user. so as surveys go, its probably better than most.
 

dflynchimp

Senior member
Apr 11, 2007
468
0
71
That was not an answer to my question.

Look I took statistics in college. There is no hard fast rule on how to sample anything. If you have proof otherwise, present it and explain it.

There are alot of factors that can effect the randomness of this poll, first of which being that it's posted on a tech forum in which most of the members are above average in their interest of the hardware in their systems.

I can't say with 100% certainty, but I would put money on it that there is a demographic of PC gamers that aren't interested enough in their hardware to frequent tech sites unless they're here to troubleshoot and cry for help. The same people probably would generate a even lower statistic in this poll because on average they wouldn't care enough about their hardware.

Even some of the members here admit that they didn't bother voting, or weren't prompted with the survey itself, which throws off the results.

In any case, this is one of those statistics that I don't really see any use for.
 

Schmide

Diamond Member
Mar 7, 2002
5,581
712
126
There are alot of factors that can effect the randomness of this poll, first of which being that it's posted on a tech forum in which most of the members are above average in their interest of the hardware in their systems.

Ironically my comment was already accepting the 57-60% as the figure for those who would allow steam to gather their statistics. As much as we bias ourselves by posting on this one particular site, there is enough diversity here and obvious variance of opinions to at least quantify this as a ballpark figure. It is more an insight into what privacy concerns we as a group would have and makes you wonder how much that would differ from the general population?
 

dflynchimp

Senior member
Apr 11, 2007
468
0
71
Ironically my comment was already accepting the 57-60% as the figure for those who would allow steam to gather their statistics. As much as we bias ourselves by posting on this one particular site, there is enough diversity here and obvious variance of opinions to at least quantify this as a ballpark figure. It is more an insight into what privacy concerns we as a group would have and makes you wonder how much that would differ from the general population?

yeah I wasn't really arguing with you or anything, just posting my caffiene-fueled take in stats here. XD
 

Keysplayr

Elite Member
Jan 16, 2003
21,209
50
91
After everyone votes, feel free to post why you care (if at all) about the Steam Survey. What does it represent to you, and why it matters, or matters not.

I'm just curious.
 

Rhoxed

Golden Member
Jun 23, 2007
1,051
3
81
Voted no, don't use steam, and have never sent my system information to them.

Computers/hardware in my sig ~_~
 

jvroig

Platinum Member
Nov 4, 2009
2,394
1
81
After everyone votes, feel free to post why you care (if at all) about the Steam Survey. What does it represent to you, and why it matters, or matters not.
I've been a rabid PC gamer ever since Prince of Persia (the original!) came into my first PC - green screen, 5.25" floppy disk only, XT processor. Prince of Persia was so advanced for that 6Mhz PC that it took almost a minute to load the scene at the end where he runs to the princess.

Despite being a PC gamer through and through, I never had any dealings with Steam. I always buy games at the store.

Perhaps Steam is crazy popular in the US, but that doesn't mean the rest of the world follows suit. (I'm part of the "rest of the world"; if sometimes my English fails me, I apologize for any inconvenience or head-scratching it may have caused you :) English isn't my native language)

So I don't hate Steam or anything. I actually like the idea, and I like Valve for providing that service because I was told that it actually helps independent developers get some action.

So with very real and profound respect to Valve and for Steam, I'm just voicing out my opinion here, not to say Steam sucks or anything. Just to be part of the voice that reminds people that there are certainly groups of gamers who, without even having to be paranoid, don't necessarily submit info to Steam or even use Steam. We therefore cannot really know "what gamers prefer", or "what majority of the gamer market uses what". All we can determine from the Steam survey at best is "what do Steam users have" or "what do Steam users prefer".
 

ScorcherDarkly

Senior member
Aug 7, 2009
450
0
0
I'm actually surprised at these numbers, considering 2 almost must have games (cod mw2 l4d2) require it, I thought it would be much more skewed.

As I said in the other thread, these games are almost must-haves for people that are fans of the FPS genre. Not everyone plays shooters. For people who's main shtick are RTSs, RPGs, MMOs, sims, almost anything but FPS, the likelihood of having Steam installed is slim. I wouldn't use the service if I hadn't bought Orange Box and/or L4D.

As for the 2 games you mentioned, I have no intention of getting L4D2 because I don't see how it's any different that the first one, and I'm not going to pay $40 to play a slightly different version of something I already have. MW2 isn't on my radar either as I don't want to give the publisher the idea that charging me $10 over standard for a computer game is ok.

I didn't notice, does Steam count SLi or CX setups? Do they count them as a separate category or as 2 distinct cards?
 

Idontcare

Elite Member
Oct 10, 1999
21,118
58
91
Yeah I am not a FPS game player, its not that I avoid steam its just that I have not had any reason to use them to date. I'm more an oblivion/civilization/etc strategy role-play gamer (will get into dragon age eventually I'm sure)
 

sandorski

No Lifer
Oct 10, 1999
70,085
5,618
126
At times I Cancel/refuse to do the survey, but at other times I complete it and include my GPU info.
 

jvroig

Platinum Member
Nov 4, 2009
2,394
1
81
oblivion/civilization/etc strategy role-play gamer
+1, same here, perhaps that's also a big factor why I haven't been attracted to Steam?

I play the occasional FPS as well (Doom, Doom II, Doom III, Counterstrike, Half-life ... that's about it... oh yeah, also the craptastic Die Hard: Nakatomi Plaza haha it was funny because it was terribly corny), but RPGs and strategy games are closer to home. Hmmm... played the father of all RTS games, Dune II; played Civ I and Civ II; naturally, got enamored with StarCraft; played all of the Elder Scrolls titles, from Arena to Oblivion, even the other Elder Scrolls derivatives like Redguard (cool, but hated the platform-jumping parts) and Battlespire (terrible); Eye of the Beholder I to III, Baldur's Gate, Neverwinter Nights, Icewind Dale, Planescape Torment, and perhaps some other AD&D titles I've missed.

Hmmm... could it be that most Steam users are FPS-leaning gamers? In that case, Steam could very well be representative of FPS gamers preferences... just thinking out loud, throwing it out there, you know...