Topics

1

General Discussion / Xim Nunchuk Survey

« on: 11:09 PM - 08/18/18 »

I've been working away on a product for you guys. Please take this survey to help me see what you want as a product.

https://www.surveymonkey.com/r/VDQDHVX

Some comments from things I've seen so far:
* Wireless support is hard. It adds not only cost, but a lot more variability and reliability issues with production. Battery charging circuitry, bluetooth connectivity, and cheap lithium ions make the product much more difficult. I don't want to bite off more than I can chew and go the way of the grifta, so am leaning towards not offering wireless, at least not this iteration. (Change my mind)

* It won't be hard to offer multiple products, especially since I am 3d printing the body of each controller. For this purpose, we could offer a few different bodies if you guys really requested it, and could build custom designs to accommodate anyone with special needs.

* The initial production will take a lot of time, think 1-2 months before I can ship out a large order. Once I get initial orders in, I'll order enough stock to reduce production time. As you all have learned from Apex stock - it is hard to predict how many people will order something. Production will be cyclical because ordering components and manufacturing will take a lot of time, but getting an order when I have stock means you'll get it in 4-5 days.

* The joystick keyboard add-on is a cool idea I had while making the survey. It would be the cheapest and easiest thing to produce, and would be able to add analog support to any keyboard if I can make it work. I will be prototyping this to see if I can make it ergonomic enough to use. If it is, it will be sold as separately alongside the nunchuk for those interested.

* I will be adding more thoughts as I have them

2

XIM Link / A very specific request for OBsIV

« on: 08:01 PM - 08/08/18 »

I've got a fun one for you today. I came across this first working on a script for overwatch, a guy wanted a button to rocket jump instead of having to spin a 180, look down, shoot, and look back with a button push (because doing consistent 180s at turn speed cap is difficult with xim).

I accomplished this using XimInputData commands to tilt the stick in the direction I needed, and the amount I needed, but came across an issue with doing this.

Xim translation does not account for XimInputData inputs when translating to the ST, and this is an issue for these scripts because any mouse input will send the cursor flying.

Say I wanted a fortnite script to build a box, it slowly turns and places walls as it turns.
In this script I will send

Code: [Select]

XIMInputData("RightStickX", 40). The issue is that while this is going, if I move my mouse to the right even just a tiny bit it will send my character spinning wildly.

Basically the Xim sees mouse delta, and says "okay for this speed I need to push the stick to the right 10," since the stick is already to the right 40, it pushes the stick to 50. The issue is that the turn speed difference from 0-10 is much smaller than the turn speed difference from 40-50, and so my cursor takes off much faster than I had anticipated with the mouse movement.

Is there a way to correct for this within the translation? Basically instead of calculating mouse delta as a set stick inflection, it would have to calculate stick delta from mouse delta, where stick delta varies over the current stick tilt.

This also could help to correct for issues such as players having messed up controllers that are stuck one way or the other, slight enough that it doesn't turn in game due to deadzone but enough to affect translation.

3

XIM APEX Discussions / Ballistic Curve Creation Tool

« on: 02:05 PM - 07/29/18 »

This is a work in progress. Over the coming days/weeks I will be adding features, optimizing performance, and adjusting the UI. The current build is intended for experimental use.

Update 1.01: I have added a section for Function Generation. This is not yet operational.
Update 1.02: The Function Generator should now be operational. I'll be implementing a better way to interact with these values, as well as some default values so the user doesn't need to experiment as much. Basically this adds an easier way to change function parameters and then output the new parameters straight to a curve.

Javascript curve creation tool spreadsheet: Latest Version

Here is what it looks like:

To use this, put any function of A5 into C5 and click the button, and your curve is ready to be pasted into manager from cell A1. The tool automatically corrects for negative values and values over one hundred, so the curves should still work in those cases.

The linked spreadsheet is read only, so press "File" > "Make a copy" to make your own spreadsheet to use however you'd like. I recommend getting the Google Sheets phone app, so that you can easily copy paste the script into your mobile manager.

If you guys have an ideas, requests, or contributions on how to make this more useful let me know.

4

XIM Link / Found out why XimInputData isn't working [BUG]

« on: 08:38 PM - 07/27/18 »

I think I may have found the reason the XimInputData isn't working for us

So this is the directory where the AHK_Additionals file actually is


But this is the specified directory within the actual file and within the header


I copied the Xim Link Folder from Roaming back to AppData and now it works as intended

5

XIM Link / Any Link users have a Grifta Solo?

« on: 08:23 AM - 07/26/18 »

I'm wondering if anyone has a grifta solo and can test the functionality with Xim Link.

I used to play Nav but switched to keyboard only about a year ago, but I want to go back to an analog joystick for CoD.

I never liked the Nav that much and the grifta solo looks perfect, but I've heard they have issues

6

XIM APEX Discussions / Question for OBsIV

« on: 10:31 AM - 07/23/18 »

In the ballistics editor, we are given points 0-50 on the X axis as mouse input values. What do these numbers actually correlate to?

We've been experimenting with custom mouse drivers, and they use "mouse movement in a single update" which is dots per update.

Are the ballistics editor units the same?

7

General Discussion / Acceleration Curves Testing - Made for Call of Duty, could work for all games

« on: 10:03 AM - 07/23/18 »

Mods: Can you move this to Game Support? I accidentally posted it here instead

On the Xim Link board we have been messing around with the Povohat mouse acceleration drivers, and some users are reporting very positive results in call of duty. It basically scales your sensitivity based on mouse speed, so you can have very slow and sticky aim on targets, but still snap quickly to other targets and cut through aim assist when you need to. Since most users don't use Xim Link, I'm going to be working on a tool that allows creation of Xim ballistics curves that accomplish the same thing without needing the tool itself. Aka - put in your own parameters and generate your own custom ballistics curve.

Here are some different exponential curves with different ballistics for you to test in the meantime.

Users have now tested them and have no complaints  None of these curves have been tested on an actual Xim and may be corrupt. I will edit this once users confirm the curves work

These are the same curves as before, but they have an added deadzone of 0.05 instead of the previous 0.8. I removed the logarithmic curves and re-numbered the others. I will work on a sort of mock-povohat tool in the near future to try and accomplish the same thing using only xim ballistics.

Code: [Select]

>>> XIM APEX [Translate Ballistics] START COPY >>>
X5BC:gAWACYAVgCmARYBpgJWAyYEFgUmBlYHpgkWCqYMVg4mEBYSJhRWFqYZFhumHlYhJiQWJyYqVi2mMRY0pjhWPCZAFkQmSFZMplEWVaZaVl8mZBZpJm5Wc6Z5Fn6mhFaKJpAWliacQ:X5BC
<<< XIM APEX END COPY <<<

Code: [Select]

>>> XIM APEX [Translate Ballistics] START COPY >>>
X5BC:gAWAJ4BPgH2AsYDsgS6BeIHMgiiCj4MBg36EB4SdhUKF9Ya3h4qIbollim+LjYzAjgmPapDiknWUIZXql8+Z0pv1njegnKMkpc+nECcQJxAnECcQJxAnECcQJxAnECcQJxAnECcQ:X5BC
<<< XIM APEX END COPY <<<

Code: [Select]

>>> XIM APEX [Translate Ballistics] START COPY >>>
X5BC:gAWAPIB9gMiBHYF8geWCWILVg1yD7YSIhS2F3IaVh1iIJYj8id2KyIu9jLyNxY7Yj/WRHJJNk4iUzZYcl3WY2JpFm7ydPZ7IoF2h/KOlpVinECcQJxAnECcQJxAnECcQJxAnECcQ:X5BC
<<< XIM APEX END COPY <<<

Code: [Select]

>>> XIM APEX [Translate Ballistics] START COPY >>>
X5BC:gAWASoCagPWBXYHSgleC7IOShEyFGYX8hvaICYk2in6L5I1pjw+Q2JLFlNmXFJl7nA2ezqHApOSnECcQJxAnECcQJxAnECcQJxAnECcQJxAnECcQJxAnECcQJxAnECcQJxAnECcQ:X5BC
<<< XIM APEX END COPY <<<

Code: [Select]

>>> XIM APEX [Translate Ballistics] START COPY >>>
X5BC:gAWAW4DAgTeBwYJggxaD5oTShduHBYhSicSLX40ljxmRP5OYliqY9ZwAn0yi3aa3pxAnECcQJxAnECcQJxAnECcQJxAnECcQJxAnECcQJxAnECcQJxAnECcQJxAnECcQJxAnECcQ:X5BC
<<< XIM APEX END COPY <<<
If you are curious how I made these curves, check out the spreadsheet I used to make them. https://docs.google.com/spreadsheets/d/1yF9A9i-qD1bDjBV4sI2FNf9UDX3aFC79CcqkWQ_XiEQ/edit#gid=0

8

Beta / Suggesting the same feature I've been asking for since Sync was added

« on: 07:59 AM - 07/23/18 »

Can we please have automatically scaling sensitivities for Sync modes? Changing the operating frequency of the xim doesn't affect sensitivity anymore, so why should Sync?

I saw Antithesis say the only reason he used sync was so that he had sensitivity numbers he was familiar with from xim 4.

The fact that I have to change sensitivity every time I try a different Sync reduces the usefulness of the feature, and having it automatically scale would be a huge quality of life improvement. It also would make settings easier for noobs to navigate, and there would be less confusion for everyone in general.

9

Game Support / H1Z1 New Config Testing Spreadsheet

« on: 07:26 AM - 06/29/18 »

Favorites so far: Deus Ex: Mankind Divided and Overwatch.

I made this spreadsheet so that we can have more coordinated config testing, simply test out the config and see what feels decent.

Simply put Yes/Maybe/No whenever you test out a config. After we get most of them tested I'll make another spreadsheet for people to rank all of the Yes configs so we can find the best one available.

If the deadzone is off - meaning if you move your mouse slowly and it doesn't affect the character - try putting the first point of the curve up to counter this effect. Some games may have similar curves but different deadzones, and that is the quickest way to try if the deadzone is okay. If you Test someone elses config and disagree with them, simply shift the rating one way. If they had No and you think Yes, put Maybe and vice versa, you get what I mean.

Thanks in advance for helping!

10

Game Support / H1Z1 New Update: Give Feedback

« on: 11:53 AM - 06/21/18 »

For the last 4 updates they are no longer posting that look mechanic fixes are a thing they're working on.

This weeks update came out today, and it does not include the look mechanic fixes.

Please visit the subreddit www.reddit.com/r/h1z1 and give them feedback that you want it fixed. It seems to not be a primary issue for the developers anymore because nobody is complaining about it. We need to remind them that this is a huge issue and if they don't fix it soon the game will die.

11

XIM Link / A Discussion: How hard would it be to support Titan 2 scripts with Xim Link.

« on: 07:46 PM - 06/13/18 »

I'm just curious. I'm wondering if there was a way to use AHK but have it automatically convert T2 scripts to AHK and run them.

Basically a way to interface the T2 language to control AHK, or use a different system entirely that will just natively run the scripts and send XInput.

I don't have enough programming experience to even know where to begin with this. In my opinion, the number one thing holding Link back is AutoHotKey. Very few people know how to or are willing to script with it. Even a simple method to convert Titan 2 scripts to AHK would help make Link a more viable replacement.

12

Game Support / H1Z1 Ultimate Curve Testing Facility

« on: 05:34 PM - 06/06/18 »

H1Z1 Overwatch ST Curves

Curve dump here for H1Z1. I'm using no smoothing on the Overwatch ST for H1Z1, in-game ADS sensitivity at 60%, all other sensitivities at 100%.

So far, the most popular curves seem to be Curve 4 for Hip, and Curve 5 New Curve 7 for ADS. Try them out and see what you think.

Curve 1:

Code: [Select]

>>> XIM APEX [Translate Ballistics] START COPY >>>
X5BC:gFCAVIBggHSAkIC0gOCBFIFQgZSB4II0gpCC9INgg9SEUITUhWCF9IaQhxaHwoh2iTKJ9orCi5aMco1WjkKPNpA1kTaSQpNWlHKVlpbCl/aZMpp2m8KdFp5yn9ahQiLspJWnECcQ:X5BC
<<< XIM APEX END COPY <<<
Curve 2:

Code: [Select]

>>> XIM APEX [Translate Ballistics] START COPY >>>
X5BC:gFCAlYDlgUCBqIIdgqKDN4PdhJeFZIZHh0GIVImBismML420j1qRI5MQlSSXX5nGnFifGaILpS+nECcQJxAnECcQJxAnECcQJxAnECcQJxAnECcQJxAnECcQJxAnECcQJxAnECcQ:X5BC
<<< XIM APEX END COPY <<<
Curve 3:

Code: [Select]

>>> XIM APEX [Translate Ballistics] START COPY >>>
X5BC:AFAEHQaACE4JywsVDDsNRw4/DygQBBDWEZ4SYBMbE9AUgBUsFdQWeRcaF7kYVRjvGYYaGxqvG0Eb0RxgHO0deR4EHo0fFh+dICQgqSEuIbIiNSK3IzkjuiQ6JLolOSW4JjYmsycwJ6w=:X5BC
<<< XIM APEX END COPY <<<
Curve 4:

Code: [Select]

>>> XIM APEX [Translate Ballistics] START COPY >>>
X5BC:gFCAh4DIgROBaIHHgjCCo4Mgg6eEOITThXiGJ4bgh6OIcIlHiiiLE4wIjQeOEI8jkECRZ5KYk9OVGJZnl8CZI5qQnAediJ8ToKiiR6PwpaOnECcQJxAnECcQJxAnECcQJxAnECcQ:X5BC
<<< XIM APEX END COPY <<<
Curve 5:

Code: [Select]

>>> XIM APEX [Translate Ballistics] START COPY >>>
X5BC:gFCAlYDlgUCBqIIdgqKDN4PdhJeFZIZHh0GIVImBismML420j1qRI5MQlSSXX5nGnFifGaILpS+nECcQJxAnECcQJxAnECcQJxAnECcQJxAnECcQJxAnECcQJxAnECcQJxAnECcQ:X5BC
<<< XIM APEX END COPY <<<
Curve 6:

Code: [Select]

>>> XIM APEX [Translate Ballistics] START COPY >>>
X5BC:gFCHSotXjjCQYpIpk6eU75YNlwyX8Ji/mXyaKZrJm12b55xonOGdUp29niOegp7dnzSfhp/UoB+gZqCroOyhK6FnoaGh2KIOokGic6KiotCi/aMno1GjeaOfo8Wj6aQLpC2kTqRt:X5BC
<<< XIM APEX END COPY <<<
Curve 7:

Code: [Select]

>>> XIM APEX [Translate Ballistics] START COPY >>>
X5BC:gDKAiIDtgWSB7oKNg0OEE4T/hgiHMoh/ifGLjI1Sj0aRbJPFlleZIpwtn3mjCqbkpxAnECcQJxAnECcQJxAnECcQJxAnECcQJxAnECcQJxAnECcQJxAnECcQJxAnECcQJxAnECcQ:X5BC
<<< XIM APEX END COPY <<<
If it becomes a tossup between two curves, or if you have any suggestions for curves to try or create, I can easily generate more.

To see graphs of the charts, or see how I made these, check out this link

ALL OF THESE WERE TESTED AND DO NOT CORRUPT YOUR XIM DATA
If that doesn't give you peace of mind I don't know what will 

13

XIM APEX Discussions / How are curves encoded?

« on: 05:43 PM - 06/04/18 »

I am trying to test out a bunch of curves at a time, and am generating those curves in a spreadsheet using functions. I have 50 numbers that I want to put into a xim paste instead of having to manually enter each number.

Basically, I'm wondering how the data from the copy pastes is encoded, so that I can build a spreadsheet that will decode and encode scripts and graphically display them.

I thought I would get lucky and it would just be 100 digits, 50 different hexadecimal points or something. That obviously wasn't the case considering each point has 2 decimal places. The code is 136 to 138 letters, and it is caps sensitive.

A hint how it is made or how it is represented would help me out a ton.


I believe roads asked for something similar to this awhile back, but nothing came of it.

EDIT:
If anyone is feeling bold and wants to help me with this, feel free. I'm on the H1Z1 channel of the discord working on this.

So far I've found that if you have a constant curve, it repeats every 8 characters, so 16 times it repeats 8 characters. It has a leading character as well, I'm assuming some sort of check bit?

I'm just making random curves and then analyzing the character strings that are output and trying to find patterns so far. No major breakthroughs.

EDIT 2:
Found something, if the ascii value of the character goes up by 1, the value on screen goes up 10.24, 2^10/100. I don't think there is a check bit like I previously thought

EDIT 3:
I'm going to stop updating this, but basically found a weird system of values working, not sure yet how it all works. First digit is 2^10, and after that it seems like every digit is almost a random 2^x, not sure yet how it works.

This is the spreadsheet I'm working out of, if anybody is interested. https://docs.google.com/spreadsheets/d/1yF9A9i-qD1bDjBV4sI2FNf9UDX3aFC79CcqkWQ_XiEQ/edit?usp=sharing

EDIT 4:
Thread Canceled. I found out it is encoded with Base64, I wasted a lot of time trying to manually figure it out lmao

14

Game Support / A Crazy Idea: Training My Own ST

« on: 08:27 PM - 06/01/18 »

I've spent the last hour or so messing with H1Z1 configs, running pretty much random equations through the old ballistics converter to try and make it feel more linear.

That's when I came up with a crazy idea.

What if I tried to do what Mist does on my own?

So here is the plan:

I have an old PS4 controller, and I can take it apart and attach leads to the pins of the potentiometers in the right analog stick.

Just realized this won't work for me... Unfortunately the only arduino I have on hand is a nano so I can't actually output specific voltages. I planned on using this to output specific joystick amounts, and then recording the screen and going frame by frame to analyze the amount of time it takes to do a 360.

But I have another idea! If I make a ballistic curve that is a horizontal line, the output should be constant as long as the mouse is moving, shouldn't matter the velocity. This way, I could set the output to say 50% and then do the same screen method.

Then, I should be able to get a plot of the radians/sec vs the stick %, and then make a curve according to those values.

Just realized another issue, since I don't know of a game that has a perfectly linear ST, I won't be able to know how far the stick is actually held. 50% of maximum speed is most likely not equal to 50% stick output on most STs.

I saw this video as a reference https://www.youtube.com/watch?time_continue=1&v=5oT1qDUGxIw. I'm assuming Mist uses a method similar to this to train STs. If I had to gamble, he uses a microcontroller to automatically output every possible stick value (each voltage across the potentiometer) over time, and records the screen while doing this. He then records the amount of frames required to do a 360 for each part, puts those numbers into a spreadsheet and whoila he has the new ST data.

Obviously I am way oversimplifying the process. How does he do vertical sens? Diagonal? Every combination of those? How does he deal with time based-accel? How does he deal with inconsistent framerates? (possible why the PUBG ST is incomplete?)

I know you can't give away your secrets, but at least tell me if I'm onto something here.

Optimally, you would let me know of a game with a perfectly linear ST (or very close but with a deadzone) so that I can experiment.


Explanation for this post: I saw the video I linked, and immediately thought to myself "I can do this, I have arduino experience and can record gameplay, lets do it" and then started making this post. This post is me thinking out loud about the sorts of problems mist and obsiv solved with the xim. Maybe it is just the engineering student talking, but I am really interested in how you guys did it.

15

XIM APEX Discussions / Is it Possible to Adjust Right Stick Deadzone?

« on: 05:02 PM - 06/01/18 »

With all the options for mouse aiming, I was wondering if it is possible to adjust the deadzone of the right stick in relation to the mouse.

Basically playing H1 with no ST, I haven't found an ST that has a deadzone as large as this game. I can move my mouse very slowly, and it doesn't "push the stick" enough to actually turn the character.

I tried doing a step at the first point on the ballistics editor but it has no effect. The actual deadzone editor does not affect mouse translation.