[jb_s]

I've unfortunately played way too many twitch shooters on PC and latency is really noticeable to me.

To the point where I can't stand using my work-supplied laptop because of slight mouse lag (I think the mouse travels maybe 30-40ms behind my hand) and 60hz screens. This is somewhat annoying coming from 165Hz, gaming peripherals. Then I have to connect to a VM to do development. Absolute hell.

[KaoruAoiShiho]

Yeah, it won't match the requirements for very high end competitive gamers. I was just talking empirical results found by Digital Foundry matching consoles.

> mouse travels maybe 30-40ms behind my hand

That sounds awful. But it reminds me that the chart I posted is tested using a very exacting tool and even the 3090 has a 30ms latency on 120hz and that's about as good as it gets. So you shouldn't take these numbers in a vacuum but more comparatively. GFN on 120hz is better than a 3060 on native in 60hz. I know it's not compelling for anyone who hates 60hz heh but for people who are okay with 60hz then GFN latency won't feel any different to you.

[blibble]

given GFN is running the same software as you'd run locally (the game), and using the same local hardware (keyboard, mouse, etc) to send the inputs, and then the same local hardware to display the server rendered output (your monitor/TV)... how can be true that GFN doesn't add at least the same input latency as running it locally?

unless there's some trick I'm missing it's going to be at least the same (plus the network latency)

[KaoruAoiShiho]

I said same as native console, not the same as the equivalent hardware on PC. If you look at the chart it's comparing different hardware not the same hardware.

The "trick" here is that native devices already have some amount of latency, just they are in an acceptable range for most people. However nvidia can optimize PC hardware to reduce the device's latency such that even with the network latency added it's still faster than the average person's native device. Hope that makes sense.

[nordsieck]

> and using the same local hardware (keyboard, mouse, etc) to send the inputs, and then the same local hardware to display the server rendered output (your monitor/TV)... how can be true that GFN doesn't add at least the same input latency as running it locally?

Something is a little weird with the numbers as shown. 120hz to 60hz only gains 18ms on PC, but it gains 45ms on xbox. Presumably with the same peripherals, tv, network connection, etc. Not really sure what's going on there.

[KaoruAoiShiho]

I'm not entirely sure either but I just go with what Digital Foundry tells me https://www.youtube.com/watch?v=jOcFSlniGrw&t=676s

Watch the video yourself.

[blibble]

I suspect it's an unfair test

they should be comparing the same input/output hardware with the xbox

then only switching out the xbox for a PC connected to the same inputs/output

what would be terrible: comparing xbox plugged into TV+controller with geforce now running on a PC with keyboard+mouse+PC monitor

the numbers don't make sense otherwise

[KaoruAoiShiho]

I'm getting the feeling that you want a test of the internet connection or something like that by your standard of what a fair test is. I don't think that's what they're going for, but a more holistic "typical usecase" test. That said I haven't bothered to check the video for exactly how the test is setup maybe someone will comment with that info.

[dbttdft]

An extra 40ms is absolutely terrible, and apparent to everyone. Anyone who says otherwise is accustomed to ultra bad games.

> GFN on 120hz is better than a 3060 on native in 60hz

False. 60Hz input lag is subtle and if someone can't notice an extra 40ms, they won't notice this.

[KaoruAoiShiho]

The input lag is measurable, I'm talking about empirical results.

[beebeepka]

You're both right. As some SLI/Crossfire enthusiasts found, high fps don't mean much when they're late. Latency is king

[KaoruAoiShiho]

Nobody mentioned fps, my post was entirely about latency.