Support NeoGAF

[jimboton]

Very interesting article. It would have been nice to see tested different versions of the same game, pc too if possible. There was a thread a while ago where someone had supposedly tested xone vs ps4 versions of the same games and his takeaway was that xone games were a tad slower in most cases, he speculated this was due to some kind of OS level 'feature'. Would be nice to see DF look into this.

Also Destiny should have been included, it's a relevant fps I think?

 

[Renekton]

Feel this ordered list should be in the OP to avoid confusion.

I didn't want to take away views from the video by posting all their numbers here

 

[LordBaztion]

I'm surprised how much lag TF2 has. It just feel great to play, though

 

[JoeLT]

I didn't want to take away views from the video by posting all their numbers here

The OP is a bit misleading though, considering those are the three lowest rated inputs in the video.

 

[nOoblet16]

I would assume NXGamer's testing method includes things like TV monitor latency, which I believe the method DF is using bypasses.


Generally speaking, the XBO versions of CoD titles struggle to hold the same 60fps standard of the PS4 counterpart.

Why not include Destiny?


TV input lag added.

He said his numbers don't factor TV input lag.

Also Destiny has 86ms on Xbox one and 97ms on PS4
https://displaylag.com/console-latency-exploring-video-game-input-lag/

 

[Renekton]

The OP is a bit misleading though, considering those are the three lowest rated inputs in the video.

I don't understand. Can you explain what you mean by that?

Anyways people should just watch the video in the link.

 

[Shock32]

So COD IW has the least amount for the COD series. Here's hoping they continue that with WW2.

 

[ethomaz]

He said his numbers don't factor TV input lag.

So maybe something else because DF test in more in line with DisplayLag if you add TV latency.

Also Destiny has 86ms on Xbox one and 97ms on PS4
https://displaylag.com/console-latency-exploring-video-game-input-lag/

They include display latency... that is even why Killzone is way lower on DF than DisplayLag.

That means Destiny is probably close to 60-70ms being a 30fps game... that is crazy... it is inline with most 60fps games tested in DF... maybe that is why the gunplay fells so great.

 

[ResidentDante]

Pretty surprising such a high value (110 ms) for Killzone Shadow Fall SP. Hmmm.

Well it felt like unresponsive shit to play, just like all Killzone games, no suprise it's that high.

 

[nOoblet16]

So maybe something else because DF test in more in line with DisplayLag if you add TV latency.


They include display latency... that is even why Killzone is way lower on DF than DisplayLag.

That means Destiny is probably close to 60-70ms being a 30fps game... that is crazy... it is inline with most 60fps games tested in DF... maybe that is why the gunplay fells so great.

No.
Check the link, the number I quoted is without display latency.
They specifically mention that the numbers they have in their test is including 10ms of display lag (which I substracted from their original numbers)

A 30FPS game having 60ms latency seems practically impossible. Going below 100 is hard as it is for a 30FPS game (and yes this is without counting display latency).

 

[TheListener]

DF still doing the lord's work. It would be interesting to see if there are input lag discrepancies between platforms, including PC. And wireless vs wired.

 

[ethomaz]

No.
Check the link, the number I quoted is without display latency.
They specifically mention that the numbers they have in their test is including 10ms of display lag (which I substracted from their original numbers)

A 30FPS game having 60ms latency seems practically impossible. Going below 100 is hard as it is for a 30FPS game (and yes this is without counting display latency).

They will update with Destiny 2 soon... we will know how much.

 

[Paragon]

Why? There are lots of methodologies to measure this.
This article does it. You can get Overwatch down to 31 ms in an ideal scenario, and CS:GO down to 20 ms. (I think these numbers are inherently somewhat higher than those used in the DF tests due to the testing methodology)

I like the novel testing method, but I'm left wondering how those numbers compare to other testing methods, and if it might be possible to come up with something similar for a digital signal.
Those Call of Duty numbers seem surprisingly low if the game has V-Sync enabled, but perhaps this testing method is going to produce much lower results than others.
I'm also not sure if firing a shot is going to produce the best results. Battle(non)sense prefers to use strafe inputs for example.

Also while we are at it, i would like to see a comparison between wireless and wired PS4/XB1 controllers. But i guess thats getting too granular.

DualShock 4 wired vs wireless. (wireless is faster)

 

[a Master Ninja]

Keep doing input lag videos, DF!

 

[ethomaz]

Keep doing input lag videos, DF!

Better... keep giving us articles

http://www.eurogamer.net/articles/digitalfoundry-2017-console-fps-input-lag-tested

Here you can see all info detailed.

 

[Dictator93]

Gun model has motion blur, you can disable it in settings on ps4.

There's definitely some really great motion blur, it's really noticeable on the gun bobbing when you sprint in multiplayer. It has the smoothest 60fps of any game I've played. Shame the MP is garbage IMO.

Is it just limited to the view model? Got any good screens of it?

 

[Chip N Chance]

I like the novel testing method, but I'm left wondering how those numbers compare to other testing methods, and if it might be possible to come up with something similar for a digital signal.
Those Call of Duty numbers seem surprisingly low if the game has V-Sync enabled, but perhaps this testing method is going to produce much lower results than others.
I'm also not sure if firing a shot is going to produce the best results. Battle(non)sense prefers to use strafe inputs for example.


DualShock 4 wired vs wireless. (wireless is faster)

Thats wild. I remember someone at microsoft mentioning around the time of the xbox 1 launch that the wired would be faster. But i guess sony has some great wireless technology. Or Blutooth...

 

[chronic_archaic]

COD MP has been almost flawless 60FPS since Black Ops 2 in almost all platforms. PS4 version of Ghosts and WiiU version of Black Ops 2 being the two exception. And even in campaign Advanced Warfare and Infinite Warfare provide a very solid 60FPS, more so in the Pro (where even Black Ops 3 campaign performs very solidly)

BO2 on Wii U is about the same as 360 in terms of graphics and framerate in MP - Ghosts on the other hand seems to have noticeable drops on the Wii U.

 

[EmSeta]

I wonder how Splatoon 2 fares.

 

[RoboPlato]

Titanfall 2 and Doom are higher than I expected. BF is lower than I expected. Figured CoD would be the lowest considering it has the most simple processing pipeline of all the games tested, although IW is more advanced on that front. Good work there.

Really hope that devs are able to refine rendering processes in order to lower input lag while maintaining a modern visual composition.

 

[Ardenyal]

Titanfall 2 and Doom are higher than I expected. BF is lower than I expected. Figured CoD would be the lowest considering it has the most simple processing pipeline of all the games tested, although IW is more advanced on that front. Good work there.

Really hope that devs are able to refine rendering processes in order to lower input lag while maintaining a modern visual composition.

Yeah, especially with regards to TF2 considering how customizable the controls are. I thought the game had some aim smoothing going on that made the control feel a little bit swimmy but I guess it's just the input lag.

 

[nOoblet16]

Titanfall 2 and Doom are higher than I expected. BF is lower than I expected. Figured CoD would be the lowest considering it has the most simple processing pipeline of all the games tested, although IW is more advanced on that front. Good work there.

Really hope that devs are able to refine rendering processes in order to lower input lag while maintaining a modern visual composition.

Yea I agree with this.
Battlefield always felt relatively sluggish (compared to COD, TF2 and Doom) due to its animation/movement system and the uneven framerate that is often closer to 30FPS than 60 (on base PS4) yet it has an input latency that's lower than near perfect 60FPS game like Doom and Titanfall 2 that also feel far more twitchy and responsive than BF.

 

[Orca]

I don't understand. Can you explain what you mean by that?

Anyways people should just watch the video in the link.

You listed it like Halo was the worst demoed even though it's not. It's led to people who clearly didn't watch the video thinking Halo's input lag is atrocious.

 

[Deputy Moonman]

Didn't this video literally just show Halo 5 has less lag than Titanfall 2, Doom, Overwatch, as well as any 60fps last gen game? Do people just read the three in the OP and assume Halo 5 is suddenly lag central?

Haha yes apparently

You listed it like Halo was the worst demoed even though it's not. It's led to people who clearly didn't watch the video thinking Halo's input lag is atrocious.

this too. in general, the response times, Battlefield 1 ~ halo 5 and CoD is the clear winner. Good numbers for all 3 really.

 

[Magwik]

Wonder what the difference is between Halo 5 on X1 and Halo 5: Forge on PC.

 

[Bornstellar]

Input latency matters.

 

[rokkerkory]

Dynamic resolution has its drawbacks!

 

[Orayn]

Is this directly tied to the frame rate of the games?

Yes and no. Framerate dictates the fastest possible time between pressing a button and having something happen onscreen, but you can have a 30 FPS game that polls inputs much faster than that or 60 FPS games that respond more slowly than others.

 

[singhr1]

Dynamic resolution has its drawbacks!

Uhm, Infinite Warfare, Battlefield 1, Titanfall 2, and Halo 5 all use dynamic resolution

And, they're all running with their lowest input lag in their series. So..

 

[Kalentan]

Uhm, Infinite Warfare, Battlefield 1, Titanfall 2, and Halo 5 all use dynamic resolution

And, they're all running with their lowest input lag in their series. So..

IIRC Infinite Warfare only uses dynamic in SP but uses a locked in MP.

At least when playing many hours of IW MP I've never noticed any resolution changes.

 

[Jhn]

I would really like a video that runs this test on more games from this generation that are not first person shooters.
It'd be very interesting to see if my subjective perception matches reality.

 

[SebastianM]

Didn't this video literally just show Halo 5 has less lag than Titanfall 2, Doom, Overwatch, as well as any 60fps last gen game? Do people just read the three in the OP and assume Halo 5 is suddenly lag central?

Yeah, the numbers show a different story but the truth is that Halo 5's aiming still feels terrible and IMO worse than Titanfall 2 and Doom 2016 (can't speak for Overwatch as I don't play it).

I can't quite put my finger on it but I'm pretty sure it's a mixture of both input lag, and some weird acceleration going on. Quite sad because past 30fps Halos actually felt better to me. Can't aim for shit with the DMR now.

 

[Ricerocket]

Why? There are lots of methodologies to measure this.

This article does it. You can get Overwatch down to 31 ms in an ideal scenario, and CS:GO down to 20 ms. (I think these numbers are inherently somewhat higher than those used in the DF tests due to the testing methodology)

Very nice article on input latency. I've been wondering what the "Reduce Buffering" setting on Overwatch PC yielded and got more info.

 

[nOoblet16]

Yeah, the numbers show a different story but the truth is that Halo 5's aiming still feels terrible and IMO worse than Titanfall 2 and Doom 2016 (can't speak for Overwatch as I don't play it).

I can't quite put my finger on it but I'm pretty sure it's a mixture of both input lag, and some weird acceleration going on. Quite sad because past 30fps Halos actually felt better to me. Can't aim for shit with the DMR now.

Input latency is more than just aiming. It also affects movement, firing, jumping, throwing etc all of which Halo 5 does exceedingly well in terms of responsiveness.

The reason Halo 5 aiming feels off is not due to input latency but rather the aiming curve cause it gave in to the esport fad.

 

[MrCunningham]

Always had a feeling that this was the reason why peole still play COD. Peoplecan say whatever they want about the series, but it still holds a plac as being a solid twitch styled FPS multiplayer game.

Would love to see a PC verion, but there are so many more variables involed there.

 

[Dictator93]

Dynamic resolution has its drawbacks!

I do not think that is the moral of the story here. IW has the lowest input latency in this comparison, and it has tons of internal dynamic resolutions according if their recent, future presentations on their tech.

 

[-TK-]

Well it felt like unresponsive shit to play, just like all Killzone games, no suprise it's that high.

Yeah, hah hah. It is so bad. H5 feels like a shit too. Siege felt bad at first, but they fixed the input lag problems.

Not really any surprises here. Cod gets lots of hate, but the controls are the best I've seen seen in any console fps.

 

[defghik]

Input latency is more than just aiming. It also affects movement, firing, jumping, throwing etc all of which Halo 5 does exceedingly well in terms of responsiveness.

The reason Halo 5 aiming feels off is not due to input latency but rather the aiming curve cause it gave in to the esport fad.

Huh? Competitive players have complained about it more than anybody. It's just a bad design decision, nothing to do with "giving in" to anybody.

Halo 3 had a similar problem - the aiming feels like shit compared to the other games in the series (obviously with Halo 5 being the exception). There was a ton of bitching about it back in 2007.

 

[nOoblet16]

Bn

Well it felt like unresponsive shit to play, just like all Killzone games, no suprise it's that high.

Pretty surprising such a high value (110 ms) for Killzone Shadow Fall SP. Hmmm.

Uh wut? 110 is par for the course and rather above average quality for a 30fps game (check BF3 on PS3 that is 157ms). Killzone 2 had 160-170ms of lag, now that is bad. Battlefield 4 on PS4 is almost 100ms and that's higher than Killzone Shadowfall MP and just a hair less than the SP that runs at lower framerate.

 

[BigEmil]

Just played cod black ops 3 and it's super responsive I reckon it would be one of the best input lag here

 

[SmokedMeat]

Redbox gave me a free game rental, so I used it to try CoD IW on PS4. Gotta be honest, I'm enjoying the campaign and the controls feel great.

 

[Ahasverus]

Input lag is gonna be the next big thing along with frame pacing

 

[RoboPlato]

Redbox gave me a free game rental, so I used it to try CoD IW on PS4. Gotta be honest, I'm enjoying the campaign and the controls feel great.

IW is a really solid campaign with great visual design.

 

[BriareosGAF]

We spend a lot of time doing this analysis internally and instrumenting our frame to understand the latency from end to end. You might find this talk interesting if you like the topic:

http://www.gdcvault.com/play/1023220/Fighting-Latency-on-Call-of

 

[Durante]

Why? There are lots of methodologies to measure this.

This article does it. You can get Overwatch down to 31 ms in an ideal scenario, and CS:GO down to 20 ms. (I think these numbers are inherently somewhat higher than those used in the DF tests due to the testing methodology)

Correction: I was actually misreading the article - you can apparently get Overwatch down to 19 ms, and CS:GO to 14 ms.

 

[Head.spawn]

Input lag is gonna be the next big thing along with frame pacing

Where ya been?

 

[Fata1moose]

Yeah, the numbers show a different story but the truth is that Halo 5's aiming still feels terrible and IMO worse than Titanfall 2 and Doom 2016 (can't speak for Overwatch as I don't play it).

I can't quite put my finger on it but I'm pretty sure it's a mixture of both input lag, and some weird acceleration going on. Quite sad because past 30fps Halos actually felt better to me. Can't aim for shit with the DMR now.

Confirmation bias makes fools of us all.

 

[Renekton]

You listed it like Halo was the worst demoed even though it's not. It's led to people who clearly didn't watch the video thinking Halo's input lag is atrocious.

Disagree. I don't think my spoilered text would lead to that conclusion about Halo, unless I put a (worst/best) like I did for 1st-place COD.

Anyways, people should just watch the video.

 

[RoboPlato]

One thing that's neat about these tests is that it answers some of my questions about how checkerboard and adaptive resolution impact input latency. Turns out very little. CoD IW and BF1 both use adaptive res in conjunction with checkerboard and have great results.

 

[Head.spawn]

Disagree. I don't think my spoilered text would lead to that conclusion about Halo, unless I put a (worst/best) like I did for 1st-place COD.

Anyways, people should just watch the video.

That's honestly how I read it when I clicked on this thread earlier today, you have no qualifiers or data given or context other than CoD>BF>Halo 5, so it's initially vague as to why you chose those three and nobody will know that unless you had already sat and watched the entire video.. kinda baffled me at first because Halo 5 had always felt great to me.

Luckily others in the thread had filled in the data by the time I scrolled down a bit while I was watching it. Probably could've just added "Top 3" for a bit of clarity, the ">" kind doesn't quite say that IMO.