NVIDIA Reflex Coming to Counter Strike 2

erek

erek

8-D
2FA
Joined
Dec 19, 2005
Messages
19,071
Interesting for input latencies

"Spotted by The Verge, the NVIDIA official slide shows the benefits of NVIDIA Reflex in Counter Strike 2, with an improved latency on RTX 4070, RTX 3060, and the GTX 1060. It reduces latency down from 10 ms to 8 ms on the GeForce RTX 4070. As noted by The Verge, this means that 9 out of the top 10 most popular competitive shooters will have Nvidia Reflex support, including Apex Legends, Call of Duty: Warzone 2.0, Destiny 2, Escape from Tarkov, Fortnite, Overwatch 2, Tom Clancy's Rainbow Six Siege, Valorant, and Counter Strike 2. Previous rumors for the GeForce RTX 4070 confirmed the AD104 GPU with 5888 CUDA cores, 12 GB of GDDR6X memory, 200 W TDP, and a $599 price tag."

Dq5cLlvV48oKarcU.jpg


Source: https://www.techpowerup.com/306912/nvidia-reflex-coming-to-counter-strike-2
 
Given the mass number of GTX class hardware used to play Counter Strike 2 I see this as a blatant upgrade pitch.
The 1060 through 1660 super make up more than half the people playing that game, if the rumors about the 4050 are true that is a card aimed squarely at gamers in this category, as it would offer them a noticeable improvement from their existing hardware while giving them the 4000 series reflex scores at a price that could be tempting for them.
The RTX 3050 was and is by all accounts a turd, and offers very little over the 1660 super it was designed to replace as the features the card offers are wasted because it doesn't have the power to actually do them in a playable manner with the exception of DLSS which at the time looked bad at 1080p it may look better now but... I can't imagine the trade-off is worth it but if somebody wanted to do a review I'd read it.
 
Zarathustra[H] said:
Woo. 2ms. I guess this will benefit all of those scientific instruments that play competitive video games :p
Click to expand...

While this is basically nothing I guess it makes more of a difference for CS2 than it would CS:GO.

Servers in CS:GO would just take all the input from every player accumulated since the last tick and process it all at once. But in CS2 every input gets a timestamp and the server processes it in the order it happened.

So your input lag matters more than it did before. If you were 2 ms faster than someone else in CS:GO it would rarely make a difference. Now it will always make a difference.
 
sharknice said:
While this is basically nothing I guess it makes more of a difference for CS2 than it would CS:GO.

Servers in CS:GO would just take all the input from every player accumulated since the last tick and process it all at once. But in CS2 every input gets a timestamp and the server processes it in the order it happened.

So your input lag matters more than it did before. If you were 2 ms faster than someone else in CS:GO it would rarely make a difference. Now it will always make a difference.
Click to expand...

I'd guess in the past it would usually not make a difference, but when it did make a difference it would make a larger difference.

2ms difference is nothing really. I very much doubt it would have any measurable impact on in game performance.

That said, in the past, if those 2ms put you right over the edge into the next tick, I guess that difference would be larger.
 
Zarathustra[H] said:
Woo. 2ms. I guess this will benefit all of those scientific instruments that play competitive video games :p
Click to expand...
Because the game is probably already running at like 400+ fps at times on that hardware. The duration of a frame is already low single digits millisecond wise.
 
Zarathustra[H] said:
I'd guess in the past it would usually not make a difference, but when it did make a difference it would make a larger difference.

2ms difference is nothing really. I very much doubt it would have any measurable impact on in game performance.

That said, in the past, if those 2ms put you right over the edge into the next tick, I guess that difference would be larger.
Click to expand...

I disagree with that assessment. If a server tick was 16 ms and you're 2 ms faster it's a 1/8 chance it actually mattered. Now if you're 2 ms faster it always matters.

I agree though that a 2 ms difference would very rarely change an interacation.
 
sharknice said:
I disagree with that assessment. If a server tick was 16 ms and you're 2 ms faster it's a 1/8 chance it actually mattered. Now if you're 2 ms faster it always matters.

I agree though that a 2 ms difference would very rarely change an interacation.
Click to expand...

I guess my take is that consistency is usually less intrusive than irregularity, so if you every 8th tick or so have a random lag spike of <16ms that might be more noticeable than a constant 2ms input lag.
 
You must log in or register to reply here.
Back
Top