RTX 50 GPU usage drops on Steam as the RTX 5070 sheds its lead and ties the RTX 3060

RTX 50 GPU usage drops on Steam as the RTX 5070 sheds its lead and ties the RTX 3060

Several Nvidia RTX 50-series graphics cards lost ground in Steam’s October hardware survey, with the GeForce RTX 5070 showing the largest drop among the listed RTX 50 models. Its share fell from 6.15% in September to 3.78% in Valve’s PC video-card table.

That puts the RTX 5070 level with the older RTX 3060 at the two-decimal precision shown in the table. The September result had briefly put the newer card in first place on its own.

The change describes the makeup of an optional survey, not graphics cards disappearing from gaming PCs. The RTX 5070’s August share was 3.77%, almost identical to its October figure, after a sharp September increase.

The RTX 5070 fell furthest, but not every RTX 50 model declined

Valve’s PC video-card breakdown also shows drops for the desktop RTX 5060, RTX 5060 Ti and RTX 5080. The RTX 5070 Ti and RTX 5060 Laptop GPU moved in the other direction.

Graphics card September October Change
GeForce RTX 5070 6.15% 3.78% −2.37 percentage points
GeForce RTX 5060 Ti 4.06% 2.70% −1.36 points
GeForce RTX 5060 4.40% 3.15% −1.25 points
GeForce RTX 5080 1.98% 1.78% −0.20 points
GeForce RTX 5070 Ti 1.87% 1.91% +0.04 points
GeForce RTX 5060 Laptop GPU 2.45% 2.68% +0.23 points

These are individual entries from the same PC table. Laptop GPUs are listed separately from their desktop namesakes, and the differences above are percentage points rather than percentages of units sold.

The September spike had put the RTX 5070 well ahead of the RTX 3060, as MIXED reported at the time. In October, the RTX 3060 rose from 3.66% to 3.78% in the PC breakdown.

Steam’s overview and detailed GPU table tell different stories

There is a wrinkle in the figures. Steam’s main survey overview lists the RTX 5070 at 3.56% for October, down 2.30 percentage points. The separate PC video-card table shows 3.78% and a 2.37-point fall from September.

The overview and the PC breakdown present different figures, so the table above uses the PC breakdown consistently. Mixing its September value with the overview’s October value would produce an unsupported comparison.

A survey swing is not proof that gamers changed cards

Valve says participation is optional and anonymous. The group represented in one month can differ from the next, so a sudden change in a model’s share does not establish how many cards were bought, sold or removed from PCs.

September’s survey also saw sharp changes in other categories. The share using 32GB of system memory fell from 42.22% to 37.06% in the October overview, while 16GB returned to first place. System memory and graphics-card memory are different measurements.

Valve’s published survey does not identify a reason for the October RTX 5070 change. It also does not establish that a survey fix caused the reversal, so the figures are best treated as a snapshot of respondents rather than a measure of every Steam gamer.

Leave a Comment

Your email address will not be published. Required fields are marked *