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

Now I’m wondering how Playstation 2 sticks like the Hori Real Arcade Pro 2 stacks up. Also curious how much these sticks lag on PC.

I think you’re probably right, I just want to hash this out early so this doesn’t get brought up again and again throughout the life of the thread.

Does anyone else have any input on the wiring issue?

The problem with PC’s is the drivers. With a console it’s dead set and perfect for comparisonsm but you may find the same stick performs magically worse on your PC based on crappily written drivers that poll in funny ways or have lowered timings than what the controller says it can handle to compensate for a broad range of host hardware.

Lag test like these are great stack ups to see obvious flaws, but overall, to me, it becomes a witch hunt. Toodles always felt the same way AFAIK, since the ultimate equalizer is the human, wherein you will compensate for delays so long as they’re not ridiculously long.

The PRO stacks up as god awful, yet people win on them consistently. That’s just one example. This type of thing is great for research purposes but I’m betting we’ll get some folks in here going “zomg, I totally lost because my stick has a 1f delay” even though they’ve practiced on it for hundreds of hours, and won tons of matches.

Props on the work put into this, this was no small feat and you’ve covered quite a lot, including any loose end explanation

Not really. Having them isolated is good practice to avoid crossing grounds between ports, but for this it’s not going to skew results.

One thing to say is that the I/O voltages may interact in odd ways. For example, the I/O on the older Qanba sticks are akin to 3.3v when in 360 mode, while many others operate at 5v which can draw it overall to a lower level. However, in the sense of activation it won’t cause anything meaningful (harmful) in terms of this test.

Good deal, thanks for the input.

ZERO latency for the VX SA? That’s like optical level speeds through a USB? Or does the 0 because it was the control and nothing beat it? Sorry, I’m sure it’s covered in that link but I need to sit down and go through all that stuff properly haha really interesting work!

I tried digging around but I couldn’t find a very clear picture of the PCB for the Joytron Exchanger. Does it happen to be the same as the Paewang Revolution boards? Was hoping to see results for my sticks but thanks for the work. I really appreciated the detailed overview to help get some insight into your specific goals and mindset.

It’s 0 because it is the reference.

It’s true in a sense. If you’re used to your stick that has +1.4F delay and can hit all of your combos on it and still reaction DP hops in KoF and everything, then more power to you. :stuck_out_tongue:

The issue I’m more concerned with is hopping between sticks - the first time I hopped onto a friend’s Mad Catz FightStick Pro, I wasn’t able to land a certain (super important) 2 frame throw link at all in GGXX. Of course I was coming from my black Q4RAF, which is a really large difference. If I was used to other PS3 sticks it wouldn’t be such a big deal.

Things look worse and worse for PS360+ all the time…

To be fair, the PS360+ should be at the top of the list for PlayStation only players, since it should be future upgradeable to work on PS4. If you don’t care at all about XBox then it’s a great PCB. Maybe PS34+ would be a more fitting name.

It’s ok on PS3 :stuck_out_tongue:

But one of the main draws of it is that it’s a solderless 360 pcb.

I’ll believe all that PS4 shit when I see it.

Do the results for ps3 include the default latency on inputs that the ps3 has?

Is this a better test than hooking up a signal to an LED & filming it against a 60fps/120fps camera?

It’s notable for both stick switching and also other types of lag testing. For example, we had some conflicting results with SF4 input lag testing. I did dozens of tests and concludedthat 360 had 4 frames inherent lag, and PS3 5 frames, using wired OEM controllers. This synced with a previous test that in addition pegged the arcade setup at 4 frames.

But later on in various threads and IRC we had numerous claims of PS3 at 6 frames, with video, using 3rd party PCBs. This could easily jibe with the poorer PS3 results seen here.

That is a developer-specific problem, not an inherent PS3 input issue. It affects the SF4 series.

In-game mechanics tests with known outcomes are better than video tests because the results are determined automatically by the system. This allows for less human judgment and a higher frequency of tests; it’s very time consuming and error-prone to record results from 1000 video tests, frame by frame.

This is very interesting data and would confirm my suspicions with my PS360+ Hitbox. It always just felt slightly “off”. I thought it might have been the buttons, they have a stronger spring in them than I’m used to, but maybe it was just PS360+ lag.

Would like to know this too, unless it’s already been answered and I didn’t see it.

I’m wondering how the brawl/fight pad pcb scores.