I’ve always heard people compare console performance using teraflops, but I’ve never heard anyone mention that metric with regards to PCs. Why?
In: Technology
because computers have extremely high teraflop rates (stuff like 10 multiplied by itself 9 times, or 10^9)
edit: you aren’t meant to downvote shit you disagree with
Ideally, to compare gaming machines (pc or console) you want to run a certain scene (or multiple) in game with the same resolution and quality settings and compare the frame rates. Higher fps means that machine is better at playing that game. Repeat this with a bunch of popular games and you end up with an idea of each machine’s performance.
On PC, this is very easy. You can choose the exact settings and resolution the game runs at and you can unlock the frame rate. On console, you typically can’t choose the exact settings. And the frame rate is limited to 30 or 60fps. To make things worse, console settings and resolution are often dynamic. So a game might silently switch from 4k to 1080p when it has trouble maintaining 30fps (which is great if you’re just playing the game, but terrible when benchmarking). These dynamic settings make it difficult to even compare generations of consoles because a ps4 game on a ps5 might just run at a higher resolution instead of a higher fps.
Because of this, consoles have to be compared in other ways. As mentioned by other users, Teraflops are not a great way to measure gaming performance. [Compare AMD’s 5500xt graphics card (5.2 TFLOPs) and their RX 580 (6.2 TFLOPs)](https://www.anandtech.com/bench/product/2524?vs=2577). The 5500xt performs better in many games despite having less TFLOPs.
Because fanboys who talk about consoles generally don’t know what the fuck they’re talking about.
Teraflops can- and often are- used to quantify computing power in general, but it’s a very flawed metric with limited real-world meaning. They measure the total floating point operations per second that a CPU/GPU is capable of, but for a lot of reasons, real-world computing applications are often limited in how they can take advantage of that.
Generally, it’s more useful to look at actual benchmarks (preferably ones that are relevant to your particular usage) to measure how powerful a computer is. But for consoles, TFLOPS are a big number that you can put on your marketing material that fanboys will repeat and argue about (seriously, I was a veteran of the PS3 vs. Xbox 360 wars, and it was shocking how little most of the people talking about them actually knew about computers).
OP, imma be really honest with you.
Teraflops are a worthless way to compare the power of consoles. The vast majority of people talking about them do not know what they mean, or in fact what they do.
Anyone using it as a significant comparative metric is talking out of their ass and you should ignore them 🙂
Latest Answers