News Nvidia study links FPS to kill-death ratio

I feel like a more accurate study doesn't try to imply causation through correlation at all.
Take a sufficiently large test group or varying abilities, set them on a slow system, medium system, fast system and then talk about results. Anything else is just making it up.

 

And given the extreme influence luck has... who spots the opponent first, does a third player turn up and try to score some cheap shot while you are dueling, how long it takes you to find shield after landing etc a sufficiently large sample size would need to be utterly absurd in size.

 

I don't get why something that's been know since the early days of competitive FPS' is still debatable almost a generation later.

 

so TLDR is, turn off all Antialiasing, set graphics options to minimum and buy the fastest monitor you can afford. Don't bother with a new GPU because you'll be turning off the shineys anyway, they're just a distracion.

 

Where is their methodology? How many participants were in the study? What extraneous variables did they control for and how? I.e. maps used, player experience, other demographics. How reliable and valid is their testing? What is the possibility of bias and conflicted interest in this "research"?

Every part of that little study reeks of bias and general scientific hokum.

 

Bingo, with the exception that faster GPUs continue to shave off render time even once you hit the FPS cap.

 

Lies lies lies lies lies. You need the maximum FPSs, so you need a 2080Ti to unlock your full potential. It's the weapon of choice. Just do it......

 

Because games are a lot more complex than in the old days.

In a game with extremely simplistic mechanics like UT99 it may be 92% git gud and 8% technical stuff like framerate, ping etc while in more complex games like Fortnite it may be 98% git gud and 2% technical stuff like framerate, ping etc.

 

Because Ngreedia said it, so it must be wrong!

 

It might, if that Speak & Spell runs nVidia spyware.

 

The complexity of the game doesn't change the maths though, the sooner you perceive 'an' event the sooner you can react.

 

I like how the blue line for the 1070 starts curving DOWN at 35 hours a week.

If gaming is a full-time job, get literally any other card.

 

All else being equal and the difference in framerate being big enough to make a statistically significant difference? Sure.

But since steps 1 - 1000 of git gud scrub are "be the attacker with a carefully laid out plan of attack rather than the reacting victim" especially when combined with the extreme prevalence of one shots in Fortnite... you'd need a crazy difference in fps (say 30 vs 144) to reach statistically significant territory.

Also lets not ignore the issue that nvidia used K/D data from BR games like Fortnite and Pubg in the first place, because killing fools isn't even the goal in them (sit around until the fools kill each other and picking your fights carefully to have the balance tilted heavily in your favour is the easiest way to an umbrella / chicken dinner, as a consequence the K/D data isn't actually a valid measure of success in BR games to begin with).

 

You don't need to be in statistically significant territory or a carefully laid out plan of attack rather than the reacting victim, yes the former would useful if you were collating statistics and the former useful no matter what, but none of that changes the maths.

 

No one is trying to change the math.

But common sense makes it obvious that nvidia used massively flawed data to exaggerate a small impact into a big one.

 

This study reminds me of studies that monitor companies have funded to show that more monitors increases productivity. That being said, I have three monitors (nothing fancy just happens to be 3) and nvidia graphics in all my computers. I'm really rebelling against these studies.

And I still very much suck at FPS games.

For the record, the monitors aren't new or anything fancy, most were purchased used.
The graphics cards are not the bees knees either, one is an advanced card, the other is an older card that is also Nvidia.

 

There we have it, if you have a big, fast sausage, you're going to win...