What's new

How to reduce input lag with DS4 controller on pc by a lot Tutorial.

In short you will be overclocking your DS4 controller it reduce the interval between inputs and hardware by a lot i mean a lot. For example Jax Quadgrab KB is ez as fuck now.

Been doing this for about a week and its seems to be safe im not responsible if your controller explode.

 

Gooberking

Solidly Mediocre
Super interesting. I heard wired is slower. That's by quite a bit.

I don't use a ds4 very often for fighters, but it's worth knowing about.
 

Zviko

Noob
Wait so input lag for wired DS4 is 10ms? That's not even bad. I don't see how would OCing it really change anything. Even Jax's KB. Why is it easier? You still press the button on the same frame don't you? Or you were 1f too slow before.
 

Gooberking

Solidly Mediocre
Wait so input lag for wired DS4 is 10ms? That's not even bad. I don't see how would OCing it really change anything. Even Jax's KB. Why is it easier? You still press the button on the same frame don't you? Or you were 1f too slow before.
It's not bad, but the fact that the floor is under 3ms is nice to try and cozy up to. At 10ms you are more than half a frame deep. Depending on when the signal is sent you have a better than half a chance that the button push will actually end up in a later frame than if it was lower.

I guess I'm not really sure how the input polling would play out, but I'm assuming the input becomes available during one frame, and then is read as part of a new frame when the the event loop starts up again. If the input takes half a frame, and you miss the polling window, then you likely roll over to the next frame. At over half a frame you would probably very consistently end up in different frames of delay. Sometimes end up processed in the very next frame, and sometimes the one after. At 3ms you probably more reliably nail the next frame window, with more rare instances of being bumped the additional frame.

All that really focused on how FG's work, and the people interested in this may actually be running other kinds of games at upwards of 100+ fps.
 

Lady Tsuru

{EMPR klan}: EMPRSS_Tsuru
Premium Supporter
My husband and I were wondering about this as well. I can never do Skarlet's F2>Blood tenticule on DS4 controller on PS4, but can do so with ease on PC (using the Xbox controller). I am not sure if it's a PS4 vs PC thing, or DS4 vs. Xbox controller thing that I can better do this.
 

Gooberking

Solidly Mediocre
My husband and I were wondering about this as well. I can never do Skarlet's F2>Blood tenticule on DS4 controller on PS4, but can do so with ease on PC (using the Xbox controller). I am not sure if it's a PS4 vs PC thing, or DS4 vs. Xbox controller thing that I can better do this.
Lag in general between consoles and PC is a pretty layered. Lag sources aggregate and this is just one, small lag source. If you aren't using the same display or even if you aren't using the same kind of display the same way, you could have a major lag source on one that isn't on the other.

Not sure about MK, but many FG's using the Unreal Engine have had additional latency on PS4. Sometimes efforts have been made to lower the problem, and other times not. Like Tekken used to have much higher input latency on PS4 than PC and it was something they worked to improve. I think it's still higher but it's better than it was. Upper SC6 players say SC6 on PS4 still has a gross amount of input latency compared to PC, and just don't want to go near that version of the game.

On PC people naturally tend to end up using monitors which trends towards lower display lag over TV's. TV's are much better than they were even a couple years ago, but you still have to be deliberate about what you buy to make sure you don't have several frames of delay added by your display device. So, it's possible for someone to end up with a lot less lag on PC just by accident and virtue of PS4 tending to suffer more as a platform than PC.

If someone was noticing a difference between platforms is also more likely that they are noticing those things, which are larger sources of lag, then an additional 7ms on a controller.

To make things even more muddled, even with PC being the better option lag wise, Steam can add latency if it's applying controller support for a game.
 
Word of warning to anyone looking to try this: the YouTube comment section for the video is rife with complaints that it 'worked great' at first, but then days or weeks down the line they started experiencing noticeable delay with their DS4 controller and couldn't figure out a fix.

These things were not intended to be overclocked so proceed at your own risk and don't be surprised if you run into hardware malfunction after extended use. Not saying you will for sure, but yeah.
 
Word of warning to anyone looking to try this: the YouTube comment section for the video is rife with complaints that it 'worked great' at first, but then days or weeks down the line they started experiencing noticeable delay with their DS4 controller and couldn't figure out a fix.

These things were not intended to be overclocked so proceed at your own risk and don't be surprised if you run into hardware malfunction after extended use. Not saying you will for sure, but yeah.
most of the comments about the controller been slow or sluggish are from (soap bars in my mouth) that experienced it online* only input lag.
i havent seen a single comment talking about experiencing malfunction of the controller. and myself been doing this for months and my controller works better than Day1. Im not saying its impossible but i rather have a controller that works for 4-5 months perfectly that is laggy all the time.
 
Wait so input lag for wired DS4 is 10ms? That's not even bad. I don't see how would OCing it really change anything. Even Jax's KB. Why is it easier? You still press the button on the same frame don't you? Or you were 1f too slow before.
You still didn't answer this question. Stating "by a lot" in the title is a long shot when we're talking about 10ms.
 
You still didn't answer this question. Stating "by a lot" in the title is a long shot when we're talking about 10ms.
the polling rate goes from 250 to 1000mhz so yeah its a lot. and the video demonstrate exactly how many ms i dont feel like repeating what been demonstrated in two different video that i already posted.
 

Zviko

Noob
the polling rate goes from 250 to 1000mhz so yeah its a lot. and the video demonstrate exactly how many ms i dont feel like repeating what been demonstrated in two different video that i already posted.
No it's not a lot. That would maybe be noticable for a mouse not a controller. Videos didn't demonstrate shit. Just some useless numbers but no real comparison in practice. Why? Because there's no noticable difference.
 
No it's not a lot. That would maybe be noticable for a mouse not a controller. Videos didn't demonstrate shit. Just some useless numbers but no real comparison in practice. Why? Because there's no noticable difference.
huge difference for me and its false the video prove it with graph and the guy gived his inputs. i tested for months and it feels way better. Nobody is forcing you dont believe it? then move on.
 

Zviko

Noob
huge difference for me and its false the video prove it with graph and the guy gived his inputs. i tested for months and it feels way better. Nobody is forcing you dont believe it? then move on.
It's actually me who's trying to stop you from forcing it on others especially if it can break controllers faster. Going from 25-30+ms to 10ms would be a huge difference, you actually miss frames in that case. 3ms-10ms is still within the same frame. So why would it be A HUGE difference.

"It feels good" is not a proof. You are experiencing placebo mate. It can't feel better and fighting games are perfect way to prove it. If I can be frame perfect on my punish with 10ms input delay then there's NO WAY 3ms input delay will feel better. Save your controllers and don't do this, it's useless.
 
It's actually me who's trying to stop you from forcing it on others especially if it can break controllers faster. Going from 25-30+ms to 10ms would be a huge difference, you actually miss frames in that case. 3ms-10ms is still within the same frame. So why would it be A HUGE difference.

"It feels good" is not a proof. You are experiencing placebo mate. It can't feel better and fighting games are perfect way to prove it. If I can be frame perfect on my punish with 10ms input delay then there's NO WAY 3ms input delay will feel better. Save your controllers and don't do this, it's useless.
Well said. I don't want to come off like I'm shitting on deathstroke, as I appreciate these types of 'hacks' as much as the next guy and anything that potential reduces input lag is going to catch my interest, even if it's minuscule. I just feel like there wasn't due diligence done (even in the video itself) as to what this can do to your controller - and if anything it seems like it's being downplayed. The guy in the video grazes over it saying only that he isn't responsible for anything that happens, but more emphasis needs to be put on the fact this can (and there's a good chance it will, in some form, and at some point in time) damage your controller.

I'm not saying not to do it - just know what you're getting into. No different than overclocking a PC which can have seriously negative effects if you don't know what you're doing (but imo in this case it's arguably worse, since there aren't really diagnostics where you can monitor how your controller is handling the workload as with a computer - least as far as I'm aware).
 
Well said. I don't want to come off like I'm shitting on deathstroke, as I appreciate these types of 'hacks' as much as the next guy and anything that potential reduces input lag is going to catch my interest, even if it's minuscule. I just feel like there wasn't due diligence done (even in the video itself) as to what this can do to your controller - and if anything it seems like it's being downplayed. The guy in the video grazes over it saying only that he isn't responsible for anything that happens, but more emphasis needs to be put on the fact this can (and there's a good chance it will, in some form, and at some point in time) damage your controller.

I'm not saying not to do it - just know what you're getting into. No different than overclocking a PC which can have seriously negative effects if you don't know what you're doing (but imo in this case it's arguably worse, since there aren't really diagnostics where you can monitor how your controller is handling the workload as with a computer - least as far as I'm aware).
Can't disagree on your statement on this, as warning people for caution tends to do so, right? Except for the part about said creator of said video just grazing iver this. IMO he explains on these subjects/topics better than any1 i've watched on YT. And i research these things quite regularly. This vid-'s are just a slice of a bigger collection of content diving deeper as you go. But that's just my humble opi. and could be the odd1out on this. Might be the combination of said topic and using games as medium to demonstrate that captures me haha

edit; example on my reasonings
 
Last edited:

Gooberking

Solidly Mediocre
You aren't overclocking your multi hundred dollar CPU or GPU which people do on the regular.

It's an already disposable(*) piece of equipment being put under minor if any risk. If someone can afford to take the risk to improve something they care a lot about, then let them have their cake.

*disposable in that the contact pads wear out faster than anything else. Easily repaired, but few people do, a fact I assume is pad makers like.
 
Yeah I wouldn't be too concerned about the life expectancy of my controller, since I go through one every 4-6 month since I play a lot of fighting game and platformers and shit and they don't make those d-pads as durable as in the NES to PS1 eras anymore.

I'd just be concerned about wasting my time doing something that really doesn't do anything.