How much does your arcade stick lag? Arcade stick input lag testing & results thread

Pretty sure the HRAP V Silent uses the same PCB as the Kai.

It depends on who you ask honestly. I feel even if the response rate factors in, a good player knows how to get around such minor limitations.
Others will completely disagree feeling that lag time and frame counting is everything. I call it finding a scape goat for your own losses.

Thanks Teyah…
I’ll be waiting for someone gives an atrox for your tests.
=)

Dunno, I’m still hitting my 1 frame links in KOF98 on my laptop using PS3 mode of both Kuro VLX and Silent V. Is it because of years of emulation, netcode, and lolDelay5MeltyBlood in which I can adjust to latency?

Then again all I can say is fuck CVS2 on PS3 and I can’t adjust to that. So I don’t know.

These lag tests are kinda making me feel bad. At the time I had a Hori V4 and TE2. The results didn’t look too hot and eventually I sold them so I could get a Kuro VLX and Silent V. Now I get fucked even harder. I guess if ever, whenever PS4 mode gets figured out on PS360+, I’ll pick one of those up and have it modded into my sticks. For now, I guess I’m fine since I’m hitting everything just fine in Xrd, 98, 98umFE, 02um, and the other games I play. It just kinda stings knowing my sticks have about a frame of lag. Part of the reason I wanted those sticks were for the touchpads in them so I could use them in Hot Shots Golf if ever need be.

Well, I’ve used both ends of the spectrum, the Akishop and the HRAP Pro V Silent. Decision making and practicing the right reflex responses will always be more important, but at the very least you don’t want your opponent to have too much of a latency advantage over you. It does skew things quite a bit. Getting the latency advantage over them felt kinda cheap.

I have an Atrox and a new TE-S+ on the way. Is there any way for me to replicate the current test method easily?

Are you sure? I don’t own a HRAP.V Silent. But with the 4 Kai and VLX Kuro, the Assign Mode allows you to map up to 3 functions to the L1, L2, and Options buttons on the 4 Kai, but on the XLX Kuro you can only map up to 1 function for those buttons. That’s all I have to indicate there might be a difference in PCBs in those two.

So does the V Silent have 3 function remapping or 1 function remapping?

Also, I wish it was more readily available to determine which sticks have the same PCBs. For example, does the HRAP.VX-SA Kai have the same PCB as the HRAP.VX-SA?

The V silent has the same side panel as the kai. It would make no sense to use a different pcb. Even has the same case. I think all us kai owners got shafted.

Curious about the Hori Pad FPS Plus coming out soon so I guess I’ll wait for those results before any padhacking…

The VLX opens up and you can manually remap many of the buttons by changing the wiring.

Naw, that’s not what I meant. I meant the Assign Mode function that the newer HRAP series have. Where you flip a switch to turn it on, press the desired button, and hold the desired function. The function that is also indicated by the “Assign Mode” LED on the panel.

I know what you mean, I just saying it’s redundant when you can physically yet almost toolless change the wiring to remap the buttons to your exact specifications.
It’s 3 thumb screws holding the top panel shut and the wiring in side is all quick disconnects.

Yeah, but I wasn’t asking if the V Silent has the 3 or 1 button remapping function because I was concerned about remapping. I was trying to figure out if the V Silent’s PCB was more similar to the 4 Kai or VLX Kuro.

My whole point is that it’s hard to know what stick has which PCB without (most of the time) voiding the warranty and opening up the stick. I wish that information was more readily available. How are we supposed to know if revisions to PCBs were made? That kind of info isn’t provided when buying sticks.

Damn it. I just ordered a HRAP V4 Kai, and now I am regretting it. -_-

I can’t say I dislike my Hori VLX now (in all aspects other than input lag I think it’s fucking awesome), but… I’m quite sad. :frowning: I’ve always been happy about the input lag of the PCBs in my other sticks…

Hey. I’m a new player looking to get a stick, but I’m a bit confused by these results. A lot of these sticks are highly recommended and widely used without much complaints (from what I’ve seen from reviews and “what stick to buy” guides), so are these results contradicting that? Seems like there’s only a handful of sticks worth getting, or I’d have to mod in a PS360 PCB.

What’s the deal? Wouldn’t people have noticed these delays from the start and spoken out against them more, or am I just not looking for complaints hard enough? Are these delays just not that big of a deal unless it’s a really low score?

The Japanese and American pros use and are even using as we speak the ‘laggy’ hori sticks and some less laggy TE2 at EVO right now, it’s not an issue, or you would be hearing about it and you don’t, people like Daigo won with the laggy Fightstick Pro’s and the only time people complained about lag was the type of monitors they used and the setup(PS3 vs Xbox 360 version of SFIV that runs at different speeds) in the past.

I agree, this thread with it’s bad math, doggy tests results, loose definition of whats even 0ms. The 0ms control for one system is completely different then another.
It’s doing nothing more than spawning this witch hunt of “laggy sticks”. With people blaming their sticks for their bad game play.
Have this been submitted to a scientific journal the entirety of work here been rejected and all the data declared unusable.

If I had it my way this thread would be closed long ago. The data is not accurate and it isn’t serving the community, only feeding on the fears of others.

The extra lag is not horrible - the key reasons sticks get recommended are reliability (ie. good body build, PCB that doesn’t break), high quality stick and buttons, and a good, hefty body kit that will stay still and can take some punishment. I regularily play with a new PCB Qanba Q4 on a PS3, and it’s just fine - likewise, people win tournaments with PS3 Mad Catz sticks. It’s not that huge of an issue. That said, I do feel a difference with a lower-lag PCB, like a 360+ and the Cerberus. It’s just more smooth and enjoyable. Not a deal breaker, but if I get to choose between a laggier stick and a faster one, I’ll naturally buy the faster one.

Fuck, I clicked on a Darksakul post and was greeted with bs. Man, if you don’t like it, could you perhaps leave us alone? Teyah has clearly stated his methods, and they do work. He is also explicit that the lag measured is only in reference to the fastest stick on the platform.

You keep saying his math is faulty, never saying why. Best argument against his methods I have seen in the whole thread is “accuracy of single measurement is 1f, so we can’t be more accurate than that” (the other arguments have been about designing a better experiment and outlining the limitations of the current approach, which I do agree with, but within those limits, this method works and is very useful), which is wrong.

Repeating a measurement like relative lag allows us to be more accurate:

A: Frame 1
B: Frame 2
X: Control Stick
Y: Slow stick

If Slow stick has, say, 5 ms of lag and we’re able to press a button during each ms of the frame:


AAAAAAAAAAAAAAAABBBBBBBBBBBBBBBB
X
    Y

We can’t see shit, they’re on the same frame. But as we move the button press timing:


AAAAAAAAAAAAAAAABBBBBBBBBBBBBBBB
    X
        Y


AAAAAAAAAAAAAAAABBBBBBBBBBBBBBBB
         X
             Y


AAAAAAAAAAAAAAAABBBBBBBBBBBBBBBB
              X
                  Y

We can see how it starts to lag behind a frame as we delay our button press. If we have perfect control of our timing within a frame, we can see how many ms the second stick lags by looking at when it starts to lag behind a frame in-game.

Obviously, we can’t do that. What we can do, is press the double-wired button a shitton of times. The more times we press, the more uniform the distribution of presses starts to be. As a consequence, the same thing happens to the 1 frame behind behavior - we can see how many times the slower stick should be behind the control - and we can start to see where the threshold is. The more times we press, the more accurate and certain we get.

So please, STFU. The method is valid even if you don’t understand it, and the limitations have been outlined on the page. People exaggerating the importance of lag is the fault of the people who fear an extra ms of lag, not the work being done.