Also if anyone is interested in tweaking any of the settings i have made a guide. Here is an image of how the kinzu copycat accel works.

- Black line is where curvelimit1 (exponential accel) starts.
- Yellow line is where curvelimit2 (logarithmic accel) starts.
- AccelerationParabolic and AccelerationLog is the strength of the curve.
- LowerSensitivityCap is the speed at which the mouse accel curve will keep curving down (logarithmic accel) even if is affecting the base sensitivity.

*If you want to alter the exponential accel (curvelimit1) to start later so that fine movement has no accel remember that it will make the transition to the logarithmic accel (curvelimit2) more dramatic as the accel hasn't being given enough time to start and therefore making medium speed movements harder to control.

*If you want to alter the logarithmic accel (curvelimit2) to start earlier so that the transition from exponential to logarithmic accel is less dramatic making medium speed movement easier to control remember that the act of transitioning from one accel curve to another makes tracking aim inconsistent so you only want these inconsistencies to happen in close range fights where tracking precisely is less important.

*If you want to alter the logarithmic accel (curvelimit2) to start later so that the logarithmic accel doesn't make tracking aim inconsistent, remember that the transition from exponential accel to logarithmic accel will be more dramatic as by making the logarithmic accel start later you have given it time to increase the accel more.

EDIT: NEW TWEAKS

*If you are using a mouse with a forward sensor position and were previously using a mouse with a further back sensor you will want to decrease the AccelerationLog value. A good rule of thumb is to never go below 10% of the AccelerationExponential value.

*You will have to find the above two settings on your own but you can use the WMO and Kinzu config as a reference for how much you should tweak each setting as they are polar opposites to each other.

I found out something very interesting today! Strenx likes to aim like this -
I tried using a low sens with the kinzu and found the accel was very difficult to control but if i aimed like strenx (with a claw grip or with a mouse that has a sensor further back like the kinzu) it made the accel way easier to control.

Just found out you will also need to download Povohat's mouse accel tool from here - https://dl.dropboxusercontent.com/u/4369386/i...est2.7.zip and install it using install.bat and also download Visual Studio 2013 Redistributable x64 from here - https://www.microsoft.com/en-us/download/details.aspx?id=40784

Great effort yoyoman! I dont have time to look into it atm but I have a related link worth checking out, bumping G100s together with kinzu v1.

https://www.reddit.com/r/MouseAccel/comments/...ion_curve/

"First: Scroll to the right, you can see the graph rather than just numbers.
Anyway, I got bored today and decided to bump two of the mice I own together for hours on end, here's the results I got. The Kinzu's acceleration curve should be easy enough to figure out now that the info on it is public, since it never was before. I didn't use all 185 events because it'd result in a graph that looked super compressed (like this) due to the Kinzu's sporadic malfunctions."

https://docs.google.com/spreadsheets/d/1Lh-kw...49SCA/edit

Btw, here is a recent mouse with a good sensor that has a similiar shape to kinzu http://shop.nixeus.com/product-p/rev-bkwh16.htm

Hi Ennea, I'm happy to see you found my post on Reddit. I was the one that originally posted that, spent way too long doing that, but the results came out well and allowed me to perfectly emulate the Kinzu's acceleration due to some other posts on Reddit that helped me figure out curve fitting and what not. I am currently using a Fourier curve to achieve it, it's pretty close actually.

I could release all the data to the public if anyone actually has any further interest in it, there's quite a bit on my PC all in a folder.

Good job, but one small thing, I don't know where u got info regarding evil's mouse sensitivity settings but the settings he was/is using with the kinzu is: "400" dpi (actually around 550 in practice), sens 2.4 and firmware 10.10. Strenx was using with same settings just sens 2 asaik.

Strenx's facebook says that he was using 800 (~1030) CPI iirc.

But he had mousesenscap set to 1.

The Kinzu acceleration acts differently on 400 and 800 CPI, but you're right and I forgot about that.

what does it matter he had 1 sens @ 800 dpi
hardware accel bypasses mousesenscap

Regarding hardware accel it's obviously irrelevant, but if he has sensitivity set to 2 and mousesenscap set to 1 then that makes his effective sensitivity 1, which is relevant.

So this whole time i've being replicating the accel for strenx when i wanted evils lol. I knew evil used 400dpi too but got it mixed up in the process D: . Anyway i'll try and replicate evil's accel over the month and keep you posted!

Also i bought a kinzu v3 the other day and found that the accel curve feels horrible with it so i will have to find a new setting for it as well. I assumed after converting all the settings for 500dpi that if i increased the sensitivity because the sensor is further behind, it would be enough for the kinzu v3 to feel like the wmo with a claw grip but it looks like i have to adjust the accel curve values and when each accel curve starts as well. Will keep you posted!

Hello I also bought a kinzu v3 recently. Is there any way to turn the natural accel off, and use in-game?

There is no hardware accel in the kinzu v3.

Ah. Well, any way to tweK the settings to dip of anything, instead of in 250 dpi increments

If you want to know so you can use my settings just convert to your desired dpi using this - http://www.funender.com/quake/mouse/index.html

Thank You for the link most of it works.

""
mouse movement recorder
mouserate checker
direct input mouse rate""

These however were 404 error

You just need to put my accel and dpi into the website and then get the real accel and save it in notepad somewhere then put your own dpi and change the accel until it's similar enough to the real accel you had before. I'll do it for you if you aren't sure.

cool, Im good for now

ON a side note Im looking at Evils config from quakecon2016 and i dont see cpi/dpi set, his is 400 correct for 2.4 sense and 0 accel?

Yes but kinzu at 400dpj is actually 540dpi.

dam :( so when adjusting old wmo sense vs kinzu i need to account for 140 dpi differnce!

It's basically just sens 3. And the real dpi for the wmo is around 430dpi although it seems to vary by mice but not by much.

I meant for adjusting my personal sense from 2010. nice to know on the wmo.

Ok my initial assumptions were correct. It doesn't matter if i used 400 or 800 dpi on the kinzu. My aim for this kinzu project wasn't to replicate the kinzu 100%. It was to figure out how the kinzu works and find a setting better than that using the kinzu as reference. 400 and 800 dpi kinzu basically work the same way, just different starting points for curvelimit1 and possibly curvelimit2. My method of finding the best setting was by -

1. Find out the basics of how the kinzu works (no accel/exponential accel/logarithmic accel)
2. Get help from Aiya and his friend on coding the program and the mathematics involved to create such a program
3. Do 100's of different bot tests for different situations, and then repeat those tests online against real people
4. Tweak the settings until eventually you narrow down the tweaks to near zero decimals and feel that my aim is better than the original kinzu

The reason my aim needs to feel better than the original kinzu is because the kinzu is an inconsistent sensor with inconsistent curves so in tbeory unless my aim is better, the curve is not close enough.

I just used evil's sens on the wmo with a claw grip and again my aim felt better than the kinzu. I converted evil's sens (2.4) by converting from 430dpi to 540dpi and just got a close enough figure of sens 3.

Ok just found out that this accel program is not made for palm grip users period. I bought a kinzu v3 and found even with the sensor further at the back, tracking still felt horrible no matter what settings i used. It did make the accel more controllable but it did not fix the tracking problem. I don't know why this is the case.

what kind of tracking problems isnt the 3050 just plain horrible on some pads?

Nothing hardware related, just difficulty in tracking aim. Anyway read my post below, i may have found a solution.

so essentially it's some mouseaccel with offset and senscap?

No, it's offset with exponential accel then logarithmic accel that curves beyond the sens speed until it reaches a malfunction speed. And also possibly 1x1 mouse movement may have a big impact too.

Well something growing logarithmicly basically doesnt grow after a bit, so...

You will never recreate the kinzu feeling with an offset and a sens cap so no, it is not the same thing. In fact i think unless your using something like this - http://www.esreality.com/post/2560986/use-dif...se-speeds/ i would go as far as to say your aim will never be as good as the kinzu copycat accel just because of the way it works.

And a bit offtopic but if i had to choose between my prev settings and what i got now i prefer these settings because even though my tracking aim is a bit worse it makes up for it with it's easier vertical aiming.

I understand it's not the same, my point is that it essentially is close enough given how the acceleration is distributed on the graph. Unless i dont get that picture.

Well i guess it looks the same, but if you zoomed in the graph you would see how curved and un-linear each accel curve is and the point is it feels completely different from just accel with a offset and sens cap.

As of current the claw grip config feels better than the kinzu for -
- high sens (18cm/360 + forwarded wmo sensor) with a claw grip.
- low sens (24cm/360 + further back kinzu v3 sensor + strenx aiming style) with a claw grip.

The only time the claw grip config felt bad for aiming was when i was using a low sens like the one above and not using strenx's aiming style or using a palm grip.

The project has come to an end. I will no longer keep bumping this thread. To quickly summarise for future users of this accel -

- After lots of trial and error i have found that the prediction thing was placebo.
- I have also found that the palm grip config is not an accurate representation of the kinzu accel.
- My research has led me to believe that the 800dpi and 400dpi work similarly enough and i do not need to find an accel curve for two different dpi's, just one is enough. I believe that their is certain differences in the accel such as Curvelimit1 is on the opposite side of the spectrum (400dpi low, 800 high) and maybe Curvelimit2 (400dpi high, 800dpi low) which makes a certain dpi better for a particular sens but i think it would be better to find a middle ground without reducing aim potential rather than trying to recreate the kinzu accel 100%.
- The biggest thing i have found is good aiming using the claw grip config using a palm grip is possible. The problem i have found is directly tied to the sensor and sensor positioning. I believe this because i used a kinzu v3 with a palm grip @ low sens and could not aim the same, i used a wmo with a palm grip @ low sens and could not aim the same, but only when i used a wmo backwards with a palm grip @ low sens was my aim better than any other mouse i ever tried. I believe the 1x1 movement of the wmo is the cause of this.

darn. well tks for updating it. Sounds like you got most of what you were looking to find out , though. Ill try to use your advice for future!

Guys, I finally did it! I figured out why i couldn't aim well using a palm grip or aim well using evil's sens without using strenx's aiming style. The kinzu accel is highly dependent on mouse feel. The sensor, shape, weight etc all have a huge impact on the feel of the accel unlike any other mouse accel. I know this because my aim improved a considerable amount to the point where i was able to palm grip it and use evil's sens once a i put my kinzu v3 sensor into a kinzu v1 shell. Even though i aimed relatively the same with both mouses without accel between the kinzu v3 and the kinzu v1 with a kinzu v3 sensor.

The only way to aim well using this program is to either use the proven mouses i have already listed or get a kinzu v1 shell with a decent sensor.

I think i can finally mark this project complete.