interesting topic. I think by the time GG Xrd comes out, players are going to want a quad mod on their sticks–PS3m PS4, 360, xbone. One that doesn’t lag on any system.
Followed this thread from the beginning and, sadly, the reaction to it was expected and unsurprising.
I’ve asked several times for something like this, but no one seemed to be interested. Lots of complaints about 1f links but 1f of input lag (not to mention the half a frame on Evo monitors)? No problem lol.
http://www.nicovideo.jp/watch/sm18717838 (the user has various comparisons on his account).
I’m glad to see you do this @Teyah. Keep up the good work.
Edit: BTW, when you test something with a “bypass” why do you attribute the value to the stick? For example the Fighting Edge with Chimp SMD bypass.
Sorry if this was posted already. How does PS360+ fair on PC? Is it similar to PS3 or 360?
I appreciate the feedback. I’d like to keep this about the actual testing & results, rather than the implications of the results, as well.
I’ve been following that website you linked, I actually have it linked on my overview page under the “Methodology” section. I used his testing as a reference during the planning stages of my test, and came up with the GGXX test over the SF4 test since there’s no need to slow down the video to see the timing differences, and GGXX also allows you to easily see timing differences of 2F while SF4 would not. I’ve browsed around his site in the past but wasn’t able to find any raw data numbers - do you know if these are on there?
Regarding the sticks with bypass dual mods - from my understanding of dual mods, they first route through the 360 PCB and then are passed to the PS3, so having an additional PS3 board should not affect the processing time of the 360 side which receives the inputs first and doesn’t need to wait for anything to begin processing inputs. My test results confirm this: I’ve further confirmed that the “unknown” Mad Catz PCB was a FightStick Pro, and also was able to test the FS Pro by itself with almost identical results:
360 FS Pro w/ChImp: 6.67 ms
360 FS Pro alone: 6.68 ms
I assumed the 360 Hori Fighting Edge is the same way, and will have the same input delay on 360 regardless of if it has a ChImp SMD inside or not.
I’m not sure, but I could give it a quick run through later today. One issue is that I’d have to rely on emulated GGXXAC to run this test, which introduces a (perhaps significant) degree of uncertainty. I may try running tests on both emulated GG and SF4 PC to see if there are any major differences.
Alright, pretty big site update today:
7/16/2014 changes:
-
Changed PS3 rankings to set the Qanba Q4 Black (original PCB) to 0.00 ms, re-ranked all PS3 sticks accordingly
-
Added a 95% confidence interval column which provides us with a range in which 95% of the data sets will fall within, relative to the average Delay (ms) reported. This is not a measure of input lag variance from stick to stick, but rather a number that represents how accurate the average reported delay (ms) is.
-
Added calculations for mean, standard deviation, and confidence interval (Raw data file only)
-
Added data for PS360+ firmware version 1.4 (XBox 360)
-
Added data for Mad Catz FightStick Pro (XBox 360)
-
Corrected rating for Toodles Cthulhu (PS3); incorrectly had the ‘Tie’ and ‘Wins’ columns for the Cthulu transposed
Revised charts:
On the way to be tested next week:
- Sega Virtua Stick High Grade (PS3)
- Qanba Q1 (PS3)
- Mad Catz TE Round 1 (PS3)
- Xtokki PS2 -> 360 converter + Sony PSX pad
- Mad Catz VS (360)
- Mad Catz TE Round 2 (360)
- Mad Catz CE Chun Li (360)
… And more!
That is irrelevant. Anything that does not have something to do with the testing, methodology etc, is just fluff. I have 0 interest in that.
Unfortunately, I don’t recall any numerical data from his website.
Thank you for the explanation of the bypass pcb issue :tup:
I’m curious - does PS3 still have that 1 frame of lag thing? I’m using my ps360+ stick on ps3 right now and it seems pretty damn responsive.
The 1 frame of lag on PS3 is with the SF4 engine, and only applies to the SF4 series on PS3. It’s a Capcom issue, not a PS3 issue. This is made worse by the fact that most people are playing on less responsive sticks on PS3, so they’re experiencing another +10-20 ms on top of that.
And yes the PS360+ is pretty damn responsive on the PS3, as you can see from the chart above. 
Yea, I was asking because my xbox 360 gamepad used on xbox 360 its about equally as responsive as my ps360+ is on ps3. That being said I just assumed the 1 frame of lag was hammered out with the ultra patch or something.
Also, why the hell would anyone ever dual mod when the Paewang is that good?
Thanks for the update, Teyah.
I have a suggestion… maybe you could remove the letter-tier rankings in the table? A lot of people make a lot of assumptions without actually looking into the data. It might be better for everyone.
i second that suggestion. People should actually look at the numbers posted instead of going “A C?!?? THIS IS UNACCEPTABLE” -Asian Father disowning stick
This. Lay persons are going to glaze over the numbers and default to the letter rankings, thinking they are going to become superstars when they upgrade their “C” grade stick to an “A” (~4ms).
-ud
Hmmm just noticed the difference between the kitty and the cthulu
I don’t think Toodles has open sourced anything, but i guess cthulu could be improved with a faster microcontroller. This is just a random guess, and I don’t know enough about microcontrollers to say there would definitely be.
Why are people putting this down? Do you all hate empirical data?
no johns.
the only person’s opinion i can really trust on the reliability of my stick is mine.
his hori VX-SA rates at 0ms, but what about my pre-owned hori VX-SA purchased from ebay and shipped across the country during the holiday season that dacidbro played on a few times that i threw a PS1 PCB into? how many different VX-SAs were tested? how many variables could come into play? how reliable is the testing method?
even within the same production batch it’s likely that there will be some variance from stick to stick. unless you purchased 30 VX-SAs from multiple sources and from a variety of production batches and they all produced the same result (basically something only hori themselves could do) and your testing methods were reviewed and approved by a panel of the world’s most anal engineers, I would not put my VX-SA in a display case and treasure it as the holy grail of latency.
and even then i’d only do so because you went through so much trouble, and not because i really care about a 6ms difference between that stick and this other stick i’m playing on right now.
do you know how much lag is already present when your body interfaces with a mechanical device that interfaces with multiple electronics? buttons don’t just magically push themselves and sticks don’t instantly move to their correct positions as soon as you want them to. your video games don’t process everything in an instant, and the picture on your screen takes time to produce and render.
do you start changing the way you keep your fingers on the buttons to reduce input lag? rearrange your button layout so all the buttons are closer together? put tons of electrical tape on your stick to reduce deadzones? lift weights so you can move your stick faster? drop the stick entirely and switch to hitbox or pad? wire the stick directly to the nerves in your fingers?
the average person doesn’t have this perspective, they don’t understand what this data means in the grand scheme of things. they’ll trade in their qanbas for horis, and when they lose they’ll look at their opponents stick and say “you motherfucker, you installed a ps360+ in there didn’t you you cheap son of a bitch”
@Teyah - when testing the Virtua Stick High Grade please mention which pcb revision is is. The original pcb had serious problems with double inputs on games like Tekken 6 so it had a major revision in the later batches.
And as previously mentioned the TE Round 2 on 360 had the TE-S pcb installed in the final batch (same as found in the PRO, VS, TE-S) which is completely different to the Version B pcb found originally, so please identify which pcb you are testing.
A VX-SA with a PSX PCB inside is, for purposes of this test, no longer a VX-SA.
Also, regarding the reliability of the test and variance among sticks/PCBs from the same line, from the overview:
These all have the same PCB.
Also, you can see from the table above, that the two MCZ FightStick Pros that were tested on 360 have nearly identical delay (+6.67 ms and +6.68 ms). So I think the test is quite accurate when considering sticks/PCBs of the same type.
Milliseconds are not unreasonably small windows of measure in this context, and these are averages. You could say the same thing about displays. You can always test it yourself of course.
Yes, we have talked about this before, in this thread even. I’m interested in the results of your research and wish to subscribe to your newsletter.
No.
No they won’t. Who the hell are these hypothetical weirdos? The average person does this? No they don’t.