Why isn't there a single 3080 benchmark?
Announcement
Collapse
No announcement yet.
No 3080 benchmarks?
Collapse
X
-
We don't run the benchmarks ourselves so results will only start appearing when PerformanceTest users with this new card submit their results.
Given the price and stock scarcity of this card it may be some time before there are enough results submitted for an accurate benchmark result for the card. Currently there are 2 3080 submissions (baselines 1296473 and 1296583 ).
-
It is up in the charts now
https://www.videocardbenchmark.net/high_end_gpus.html
It is the fastest card at the moment by a significant margin.
Comment
-
The distribution for the 2080 TI is pretty broad. See graph below. This is due to many factors (slow CPU, slow RAM, monitor frame rate limits, driver versions, missing DX12 support, no 4K support due to monitor, etc..). Average is ~21500 however. So you are a long way over the average with your 2080TI.
So a highly optimized 2080TI setup can probably match a average 3080 setup (based on the very limited numbers we have so far for the 3080). The performance advantage of the 3080 isn't so great at 1080p resolutions.
Comment
-
Graph is screen shot from the PerformanceTest software. Anyone can download the software and make the same graph.
Would be cool to overlay the distribution data for the 3080 as well. But we don't have enough data for that yet.
Comment
-
Turbo12R, My guess is that most of review sites are using the fastest CPUs available to run their tests. Where as in real life the people buying the 3080 have a mix of mid range to high end CPUs. Some probably only have 1080p monitors as well. So the real life performance difference between the 2080TI and 3080, on the average machine, isn't as great as the reviews make out. The reviewers are seeing the best case, not the average case.
Comment
Comment