New Steam survey results

I hope you might find that the Steam hardware survey isn't monitoring a majority of the computers out there. These are also "old" numbers already, you'll typically find the numbers lagging a month and this is data from february ... and DiRT2 only coming with Cypress and Juniper helps the numbers.

Yes yes of course, steam is just a subset of the total users and not all users are online at a given point of time. But statistically the numbers are valid, its just like doing a survey in a city or town. What does Dirt2 being bundled with Cypress and Juniper have to do with anything though?

And Juniper has been on sale since october, just three weeks after cypress. I dunno about you but i was very surprised to see that cypress had outsold it.

Of course this happpened with the previous gen too with 4800 series far outselling the 4600 series(in steam user stats anyway) but the 4800 series sold for less than 200 USD for a majority of their life
 
I'm surprised that Cypress outsold Juniper as well. I guess they priced it well at release, I assume that was when a majority of sales occured (before all the retailers raised it $50+ over MSRP)
 
I'm surprised that Cypress outsold Juniper as well. I guess they priced it well at release, I assume that was when a majority of sales occured (before all the retailers raised it $50+ over MSRP)

Indeed, in January the numbers were 2:1 in favor of Juniper (when they announced 2 million sold)
 
Yes yes of course, steam is just a subset of the total users and not all users are online at a given point of time. But statistically the numbers are valid, its just like doing a survey in a city or town. What does Dirt2 being bundled with Cypress and Juniper have to do with anything though?

And Juniper has been on sale since october, just three weeks after cypress. I dunno about you but i was very surprised to see that cypress had outsold it.

Of course this happpened with the previous gen too with 4800 series far outselling the 4600 series(in steam user stats anyway) but the 4800 series sold for less than 200 USD for a majority of their life

The problem is the cards don't offer a good value. You get vastly better performance going from the 5670 to the 5750 or 5770. You can find 1 gig 5770s for $150 which is sometimes just $30 more than a 1 gig 5670.

I would think the vast majority of 5670s are sold to oems or non gamers
 
Something to note is that Cypress doesn't exist in notebook land. Unfortunately. So the top end is Juniper and it fills in a bunch of models, and Redwood has quite a few too. I wish ATI had the guts to just market them with desktop model numbers.

Non gamers go with IGPs because it costs more to find a system that has more than an IGP. Wander around a brick and mortar for a bit and see what the boxes have in them.
 
Last edited by a moderator:
The problem is the cards don't offer a good value. You get vastly better performance going from the 5670 to the 5750 or 5770. You can find 1 gig 5770s for $150 which is sometimes just $30 more than a 1 gig 5670.

I would think the vast majority of 5670s are sold to oems or non gamers

I agree.
 
What does Dirt2 being bundled with Cypress and Juniper have to do with anything though?

The Dirt2 game bundled with Cypress and Juniper is a Steam coupon code. You have to download it from Steam and play it through Steam. This skews the usage numbers in favor of Cypress and Juniper. I assume that if you play Dirt2 through Steam at least once in a month, this adds to the usage number for that month.
 
The Dirt2 game bundled with Cypress and Juniper is a Steam coupon code. You have to download it from Steam and play it through Steam. This skews the usage numbers in favor of Cypress and Juniper. I assume that if you play Dirt2 through Steam at least once in a month, this adds to the usage number for that month.

Steam only checks once a month for hardware specs, game stats are accumulated real-time and averaged over a two week period.
 
AFAIK it should only be once a year - on the anniversary of your account creation.

I found a new install will also trigger the HWS opt-in.. I also have multiple machines running on the same account.. it seems to me it should trigger more often as it would only allow one machine to be surveyed that way. Same would go fo Cyber Cafés right?

Interesting though.. I can't find a thing on the frequency of the survey except what valve describes on the page itself "Each month, Steam collects data about what kinds of computer hardware our customers are using."
 
AFAIK it should only be once a year - on the anniversary of your account creation.

That would be horrible. Cypress/Juniper are out for about half a year, so only a small percentage has been counted so far? If sales were linear, about 25% would have been counted.

If true, these steam numbers are even more difficult to interpret than I thought.
 
AFAIK it should only be once a year - on the anniversary of your account creation.

That explains why the RV770 based GPUs are still seeing increased adoption.

Do you know this for absolute certainly? Its kind of important when considering these statistics, or if you don't could you find out? Thanks!
 
That explains why the RV770 based GPUs are still seeing increased adoption.

Do you know this for absolute certainly? Its kind of important when considering these statistics, or if you don't could you find out? Thanks!

What do you think about G92's massive, unexplainable jump last year?
 
What do you think about G92's massive, unexplainable jump last year?

Since I don't have that data on hand my presumption is that a lot of Steam accounts are created in Q4 and a lot of GPUs are bought in Q4 as well, so therefore a lot of the Steam accounts will give their yearly update (if true) in Q4.
 
IMO Steam HW survey would be more accurate if they simply asked for users permision and after that data would be automatically sent every month.
 
Since I don't have that data on hand my presumption is that a lot of Steam accounts are created in Q4 and a lot of GPUs are bought in Q4 as well, so therefore a lot of the Steam accounts will give their yearly update (if true) in Q4.

Well that explains a lot, and I agree with Rebel44 that I was under the assumption that it worked that way. I agree on the scavenging every time I install Steam.
 
What i find intriguing is that according to steam data Cypress is outselling Juniper! Who would have though that a part that sells for double the price sells more?

In DX11 GPU's the breakdown is as follows

5800 series 49.9%
5700 series 43.85%
5900 series 2.56%
5600 series 0.99%

Heck the 5900 series has sold more than the 100$ 5670. I thought the general consensus and the many studies from Jon peddie etc showed that majority of sales are below the 200$ pricepoint

Well, Cypress has been out longer also. If you watch the trend, from Oct - Dec, Juniper had extremely rapid growth in install base compared to Cypress rapid losing install base. IE - Oct (Juniper was launched in Nov. or Oct.?) 5800 had almost 90% of the install base. Fast forward to march and now it's only ~46% with the majority of install base since Oct. going to Juniper.

Looks like from Oct. - Dec., Juniper was probably outsellfing Cypress 4:1 or perhaps 3:1.

Regards,
SB
 
Well, Cypress has been out longer also. If you watch the trend, from Oct - Dec, Juniper had extremely rapid growth in install base compared to Cypress rapid losing install base. IE - Oct (Juniper was launched in Nov. or Oct.?) 5800 had almost 90% of the install base. Fast forward to march and now it's only ~46% with the majority of install base since Oct. going to Juniper.

Looks like from Oct. - Dec., Juniper was probably outsellfing Cypress 4:1 or perhaps 3:1.

Regards,
SB

Without actual numbers it's of course impossible to say exactly, but in January it was 2:1 total Junipers sold vs Cypresses
 
It seems that the Video card usage details are already updated for March, even though that month has not yet ended. February is missing from the chart.

Anyway, the trend for the 5700/5800 cards remains the same for 4 months in a row: +0.5% usage per month. I'm not sure why 5700 does not catch up to 5800 more quickly. Availability seems better, and I would expect sales volumes to be much higher in the mainstream segment than in the performance segment.
 
Back
Top