NVIDIA GeForce GTX 1650 Returns to Top Spot According to April Steam Hardware Survey

[T0@st]

Valve has released the tabulated results and statistics of its April Steam Hardware and Software Survey - the key take away from last month's user generated data is that NVIDIA's trusty GeForce GTX 1650 GPU is once again the most popular graphics card. It dethrones last month's winner - the NVIDIA RTX 3060 graphics card which falls to third place where it sits below the second place GTX 1060 GPU. The RTX 3060 experienced an almost 6% decline in usership from the previous month, and the GTX 1650's userbase grew by 2% in the same period of time. It is interesting to note that the entry for the GTX 1650 encompasses both desktop and laptop variants, while the RTX 3060 gets divided into two separate entries on Valve's survey - the desktop version sits at third place and its laptop-oriented sibling trails slightly behind with a placement at position number four. NVIDIA absolutely dominates the field with lots of its budget and midrange cards (across several older generations) - AMD and Intel barely make it into the top 25 with a small sprinkling of iGPUs and one discrete model (Radeon RX 580) placed at position 24.

April's survey shows that Intel processors remain a favorite for many Steam users with a 67.14% share, and AMD follows in second place with a 32.84% share. AMD CPU popularity is on the rise (when compared to previous months) so a more even share of the market could be on the cards, if an upward trend continues. System RAM enthusiasts were upgrading to a smaller degree last month: 52.19% are on 16 GB, and 16.1 percent are on 32 GB - indicating slight declines (from March) of 4.73% and 6.61% respectively. The majority of users prefer to stick with Windows 10 64-bit - that OS has a 61.21% share, but its popularity has dropped by 12.74% within the survey period. Windows 11 64-bit is gaining ground with a 10.98% increase from March to April, and it sits at second place with a 33.39% share of the OS userbase. As always, the results indicated by the monthly Steam Hardware and Software survey are not considered to be pinpoint accurate due to the random nature of user responses, but overall and general trends can be discerned from the data on hand.



View at TechPowerUp Main Site | Source

 

[dj-electric]

This is very sad to see, especially after the more capable 1060 was already in this point years and years ago.

 

[Kirederf]

Time for LTT to build an absolute monster of a gaming PC using these components!

 

[nguyen]

rtx3000 has a total marketshare of ~30%, that's is some crazy adoption

 

[WorringlyIndifferent]

Oof, 1650 in the top spot? Times are tough.

 

[sethmatrix7]

rtx3000 has a total marketshare of ~30%, that's is some crazy adoption

Those stimulus checks went somewhere!

 

[Lew Zealand]

11 of the top 12 spots (40%) are 50 and 60 series cards, I hope game devs are paying attention.

"As always, the results indicated by the monthly Steam Hardware and Software survey are not considered to be pinpoint accurate due to the random nature of user responses, but overall and general trends can be discerned from the data on hand."

That's it? No comment about the bonkers numbers from the previous month when compared to the 6 month trend? Well the numbers are probably maybe reasonably close to 1-2 standard deviations of true.

 

[BorisDG]

11 of the top 12 spots (40%) are 50 and 60 series cards, I hope game devs are paying attention.

Yeah it's crazy. We are all here enthusiasts with good rigs, but the reality is different it seems.

 

[ZoneDymo]

76% Nvidia, 15% AMD, 8% Intel.....sigh solid competition yall...

 

[Tomorrow]

Nvidia RTX 3060 loses its crown as Valve's Steam survey returns to normal

In March, the RTX 3060 saw an unusual 6.31% increase among Steam survey participants to take the top spot in the GPU chart, replacing the GTX 1650,...

www.techspot.com

RTX 3060 and RTX 2060 saw losses of -6.01% and -3.6%, respectively

Intel also saw its previous month's massive 9.01% increase over AMD in the CPU section wiped out in April. Chipzilla fell -9.04% last month, pushing team red's CPU share to 32.84%, the highest since January.

The massive 25.3% rise that saw Simplified Chinese become the top language on Steam was also wiped out last month, pushing English to the top spot again. Windows 10 64-bit saw its March gains (11.62%) turn into April losses (-12.74%) as Windows 11 closed the gap on its predecessor to 27.82% thanks to the former's 10.98% rise.

A single Nvidia card can win or lose ~6% month-over-month. CPU market share can have 9% swing and Windows can have 12% swings and people still quote this survey as a reliable metric of AMD's market share being only 10% or something.

 

[Octavean]

Slim pick'ns out there. Not a god sign. Not a good sign at all,...

 

[wolf]

A single Nvidia card can win or lose ~6% month-over-month. CPU market share can have 9% swing and Windows can have 12% swings and people still quote this survey as a reliable metric of AMD's market share being only 10% or something.

This is a more dramatic variance than usual, by quite a bit, it seems like Valve had to fix something that was broken for last month perhaps. And yeah, it's still pretty much the most reliable metric for AMD's market share, certainly better than mindfactory sales numbers....

 

[Lew Zealand]

https://www.techspot.com/news/98521-nvidia-rtx-3060-loses-crown-valve-steam-survey.html

A single Nvidia card can win or lose ~6% month-over-month. CPU market share can have 9% swing and Windows can have 12% swings and people still quote this survey as a reliable metric of AMD's market share being only 10% or something.

According to this month's Steam guesses eh hmm... numbers, Nvidia to AMD proportion is about 5:1, which seems roughly plausible. Within an order of magnitude or so.

 

[LupintheIII]

It is incredible how everyone still produce articles on that chart without even noticing 25% of the userbase just disapeared overnight.
It's right there, -25% of Chinese users after a +25% the previous month... same for all the hardware that jumped up next month including the RTX 3060 that got headlines for being the N°1 GPU on Steam.

The only noteworthy new is:
Steam chart is broken and it always has been.

Not only the data is not "pinpoint accurate", but no "overall and general trends can be discerned from the data on hand", a chart that claim to represent milions (if not bilions) of users that can see a 25% swing month on month it's worthless.
If the March data was right, April data is BS and vice versa.
If Valve has manually corrected the data to fall more in line with previous trend it means we have no way to know what and how was altered, meaning all the data are skewed and worthless.
It is more sad since that stuff is regarded also along gamesevs.

The more probable reason of that 25% blip is chinese bots accounts used to farm in-game currencies, each machine can run hundreds If not thousends of accounts, just a few can skew the chart.
Regardless we have no way to know how many of that accounts are included in every months data... meaning Steam Hardware Survey is worthless.

This is a more dramatic variance than usual, by quite a bit, it seems like Valve had to fix something that was broken for last month perhaps. And yeah, it's still pretty much the most reliable metric for AMD's market share, certainly better than mindfactory sales numbers....

A 25% swing in userbase month on month show the polling system is broken making Steam chart completely worthless.
If Valve "fix" the data it mean the chart is worst than worthless, is riddled with bias by the inside with no way to know how the weight was assigned.
Stop using Steam survey to represent anything, AMD marketshare, Intel marketshare, Nvidia marketshare etc.
Is pointless.

 

[64K]

It is incredible how everyone still produce articles on that chart without even noticing 25% of the userbase just disapeared overnight.
It's right there, -25% of Chinese users after a +25% the previous month... same for all the hardware that jumped up next month including the RTX 3060 that got headlines for being the N°1 GPU on Steam.

The only noteworthy new is:
Steam chart is broken and it always has been.

Not only the data is not "pinpoint accurate", but no "overall and general trends can be discerned from the data on hand", a chart that claim to represent milions (if not bilions) of users that can see a 25% swing month on month it's worthless.
If the March data was right, April data is BS and vice versa.
If Valve has manually corrected the data to fall more in line with previous trend it means we have no way to know what and how was altered, meaning all the data are skewed and worthless.
It is more sad since that stuff is regarded also along gamesevs.

The more probable reason of that 25% blip is chinese bots accounts used to farm in-game currencies, each machine can run hundreds If not thousends of accounts, just a few can skew the chart.
Regardless we have no way to know how many of that accounts are included in every months data... meaning Steam Hardware Survey is worthless.


A 25% swing in userbase month on month show the polling system is broken making Steam chart completely worthless.
If Valve "fix" the data it mean the chart is worst than worthless, is riddled with bias by the inside with no way to know how the weight was assigned.
Stop using Steam survey to represent anything, AMD marketshare, Intel marketshare, Nvidia marketshare etc.
Is pointless.

I would never propose the Steam Survey to be a Scientific Survey because it isn't and it was never intended to be that. It's just for a general idea what gamers are using.

Can you offer an alternative to the Steam Survey?

 

[Tomorrow]

I would never propose the Steam Survey to be a Scientific Survey because it isn't and it was never intended to be that. It's just for a general idea what gamers are using.

Can you offer an alternative to the Steam Survey?

Adjustment to Q422 JPR dGPU Market Report

Super compute products skewed dGPU market share.

www.jonpeddie.com

JPR seems to have some data tho from what i understand it's based only on 31 countries and similarly seems to display some odd numbers for dGPU market with Intel somehow having 5% share in 2021 (Alchemist launched in 2022) and now sitting only at 6% while AMD being only at 9% despite having decades on head start.

More realistic estimates put AMD at around 30% and Nvidia 60% with Intel around 10%.

 

[64K]

Adjustment to Q422 JPR dGPU Market Report

Super compute products skewed dGPU market share.

www.jonpeddie.com

JPR seems to have some data tho from what i understand it's based only on 31 countries and similarly seems to display some odd numbers for dGPU market with Intel somehow having 5% share in 2021 (Alchemist launched in 2022) and now sitting only at 6% while AMD being only at 9% despite having decades on head start.

More realistic estimates put AMD at around 30% and Nvidia 60% with Intel around 10%.

I trust Jon Peddie Research much more than Steam Hardware Survey and you can catch some info for free but I think full quarterly reports cost $6,000 per year.

 

[LupintheIII]

I would never propose the Steam Survey to be a Scientific Survey because it isn't and it was never intended to be that. It's just for a general idea what gamers are using.

Can you offer an alternative to the Steam Survey?

Well that's good for you, unfortunately all the tech media is reporting that chart every month as ghospel also indulging in articulated market analysis (all based on wrong data).
We saw a lot of titles about RTX 3060 being N°1, but very few reported on the actual reason (chinese bots) or how odd a -25% is...

Speaking of alternatives, the easiest one would be ask Steam to be open about survey methodology and to improve polling system.

Adjustment to Q422 JPR dGPU Market Report

Super compute products skewed dGPU market share.

www.jonpeddie.com

JPR seems to have some data tho from what i understand it's based only on 31 countries and similarly seems to display some odd numbers for dGPU market with Intel somehow having 5% share in 2021 (Alchemist launched in 2022) and now sitting only at 6% while AMD being only at 9% despite having decades on head start.

More realistic estimates put AMD at around 30% and Nvidia 60% with Intel around 10%.

Q3 2022 Discrete GPU Market Share Report: NVIDIA Gains 88% Market Share Hold, AMD Now at 8% Followed By Intel at 4% (wccftech.com)
The numbers are based on OEM's sales and exclude all of retailers and especially e-tailers like Amazon.

The fact JPR itself just debunked its own numbers based on nothing but assumptions and wrong data after being called out by basically everyone tells you all there is to know about that report...

 

[wolf]

A 25% swing in userbase month on month show the polling system is broken making Steam chart completely worthless.
If Valve "fix" the data it mean the chart is worst than worthless, is riddled with bias by the inside with no way to know how the weight was assigned.
Stop using Steam survey to represent anything, AMD marketshare, Intel marketshare, Nvidia marketshare etc.
Is pointless.

I don't think it's worthless and pointless, the last month or two have been an aberration that was corrected, but by and large the long term trends shown in the survey over the years have been proven to be very closely linked to the 'real' values of market share and other statistical points of merit, when cross referenced against other analyses and repots. Things market share, like popular screen resolutions, amount of cores in a CPU and so on. The numbers don't need to be of ultimate accuracy and extremely large or all encompassing sample sizes either, they just need to be indicative.

While admitting it's not perfect and irreproachable down to the most minute (my-newt) details, It's really quite far from worthless and pointless, and one of the best data sources we have for these kind of things, so I'm going to go with a hard disagree on your pov.

I respect your pov, but I don't agree with it whatsoever on this topic, and reading your other replies I doubt there's much you could say that would convince me, but I'm always open to consuming and logic or rationale you can bring to the table that makes the claim more credible.