XIM Community

[APEX] PURIST -01- BASIC - DPI, Polling & Synchronization Performance & Accuracy  (Read 1831 times)

Offline mjfame

  • Member
  • *
  • Posts: 245
  • "Kolibri King" | Global Rank #1
    • View Profile
    • BF1 Kolibri King aka the BabyPistolPro: #1 in the world. (Semi-Retired)
  • Gamertag: MJ FAME (Active)
  • PSN ID: MJ-FAME (Retired)
Please, refrain from directly quoting this post in its entirety as it is rather lengthy. Thanks.

The focus of this topic is for building your configuration by the book (traditional) with recommended settings to get the most from your XIM device. This will provide your XIM with the highest mouse sensor accuracy & responsiveness which will result in a super-clean and accurate conversion. Taking a break from the extreme to produce a long overdue guide that is full of information that should appeal to a wider user-base. Retain and apply this information. As always, enjoy yourselves.

Most mice true native is nearly 800DPI (above 700DPI but below 800DPI). my mouse is equipped the PMW3366 sensor which uses a 40x40 array - 1x magnification = 800 Counts Per Inch (CPI). 800DPI is the true physical resolution of the sensor and therefore will track the absolute best (i determined this on my own using extensively thorough testing methodology the very first day that i acquired a device that was equipped with this sensor.. this was before i even learned the exact specifications of the sensor itself. so i am actually basing this statement off of those results. the specifications just so happen to confirm my findings to be correct and that's reassuring.)

We will be using true-native DPI of the mouse sensor (or as close as possible). This will ensure that the mouse will output user input as accurately as it is capable of tracking. DO NOT fall prey to the marketing that modern sensors are native across all DPI steps. This is false. Every sensor's true CPI (counts per inch) capability can be determined by its array which will always reflect the true-native DPI of the hardware. Anything above that means that the mouse firmware must add counts and estimate user-input. The mouse sensor is a camera that rapidly takes snapshots in order to track it's positioning.

Modern devices with 'flawless' sensors are being marketed as native across all steps due to the inaccuracies at extreme DPI levels being much lower than they once were. (say, for example 12kDPI tracking accuracy of 94%). This may not sound like much but there is a real-world performance difference that I am able to detect and you should be able to as well. I own a Logitech G Pro which is one of if not the snappiest mice at max DPI that is on the market and 800DPI still performs better each and every time I flip back to it. It's noticeably more responsive and much more accurate. It is enough to make a huge different while playing multiplayer shooters. Especially ones with longer time-to-kill and heavier aim-assist. In general it is simply more reliable overall.

Spend a moment to think about the way a digital camera photo looks when taken without zoom. It's clear and artifact-free. A zoomed photo will look slightly blurrier but not very noticeable if the zoom falls within the hardware's optical zoom range. Beyond that is the digital zoom area where photos will begin to look noticeably murky and pixelated. For the instant gratification aficionado, open an image using any photo viewer and zoom in slowly. Pay attention to how the photo loses detail as you scale it beyond its native size.

The same holds true for the technology that your mouse uses to track user-input. By comparison the optical range for the newer gaming mouse sensors typically would be above 800DPI up to 3200DPI and the digital would be 3200+. On top of that the increased counts add latency. Micromovements suffer greatly due to the additional latency and inaccuracy. This is the main cause of negative acceleration and by now I feel that most of us understand that negative acceleration makes aim-assist much tougher to deal with.


set your mouse for 800DPI and 1,000Hz. (best of both worlds)

800 is plenty of DPI if you are mindful of the way to use it properly with XIM. Frame-syncing makes this possible. Contrary to what others may believe, lower DPI will not make aim-assist more pronounced in fact it will make it easier to work with due to the fact that the mouse is translating with less haste and greater overall accuracy. Use 800DPI in order to feed your XIM with the cleanest possible user-input and let the XIM handle the multiplier, instead. It's better this way due to the XIM receiving a higher quality mouse output to work with and with less delay in between. Why raise the mouse multiplier when the XIM handles this anyway? All your doing is adding latency and degrading quality of the data from the mouse before XIM even receives it then the XIM also has to process delayed and degraded received data before it reaches the gaming platform. The XIM output can't be any more accurate than the data it has been provided with to work with and it can't output it any faster than how long it takes to receive it.  Think about that.. Personally, I would rather the XIM receive the cleanest possible data in shortest amount of time to work its magic on.

That about wraps up the great DPI debate.

Moving on..

1,000Hz is a must if your mouse is capable of properly updating at this rate. The higher the updates per second the greater the tracking resolution, the lower the latency and the less aim-assist will interfere with user input.

Connecting an external power adapter to your USB hub is also recommended. It will provide performance stabulity by ensuring that your devices are receiving a consistent flow of energy.


Vanilla Smart Translation (No Ballistic)
800 DPI (Mouse)
1,000Hz (Mouse & XIM)
Synchronize to framerate: 60FPS=COMMON // 30FPS=SLOW
Smoothing=0

(Smoothing should not be necessary while frame-syncing and utilizing a reasonable sensitivity multiplier)


Using the above settings I use COMMON Sync + a sensitivity multiplier of 52.5 for Halo 5.
(That would be COMMON Sync + a sensitivity multiplier of 3.5 @12kDPI if anyone would like to create and test two separate configs back-to-back)

So there you have it. The best setup for the purist in all of us. It honestly doesn't get much better than this unless you dare to venture into the Extreme. If you're up for that then smack the link below for a follow up read which will detail everything for you.

https://community.xim.tech/index.php?topic=65974.0




I could have sworn you recently started yet another thread on this very same subject where you recommended 800dpi @ 1000hz polling but with sync set to off, and both smoothing and sensitivity set to their maximums, using a ballistic curve to alter the sensitivity.

Whats it to be?  You seem to come up with something new on a daily basis, making grand promises that it's perfect, and yet you clearly aren't satisfied with it, otherwise why keep altering it.  I have no doubts that you spend a lot of time testing your theories, but they seem to change on a daily basis!



I sure did however I am making attempts to offer the best method with explanation for getting the most out of the standardized method so that there is an offering to suit a variety of user preferences. Not everyone is comfortable enough to venture into the extreme. I get that. I also believe that anyone who refuses to venture into the extreme is cheating themselves out of something pretty special.

My R&D period with XIM APEX began from the very first day of ownership. It lasted roughly 10wks and has been over with since a few days prior to sharing "Perfect Match". No new R&D went into the Purist offering because it is already being used to lay the foundation for the Extreme variant. (800DPI / 1,000Hz)

So additionally you have the Purist offering which clearly was an afterthought but I figured why not share the stripped down model for those who will remain configuring that way. If you're going to stick to the basics then you might as well have the information to help you get the most out of it.

Bits and pieces (precursors) were shared as progress was made in order to let everyone in on it but nothing new is on the horizon. These last two topics are final and these two setups are about as good as it gets for the Purist and for the Extremist. 

For what it's worth I only use 800DPI / 1000Hz for every game. I use the traditional method with my recommended Purist settings for very few titles and for most others I use the Extreme approach. And once I get around to playing those select games that are still using the basic version they will be converted to Extreme which imo is the superior method.

Either way you can't go wrong however the extreme model already matches up better before even beginning to tune the sensitivity via the ballistic editor. This means that it will provide a much greater threshold of refinement and while spending less time doing so.
« Last Edit: 06:29 AM - 05/17/18 by mjfame »
XIM APEX // BETA : 20180402
Logitech G Pro // Swift-RX XL // XB1X // Sony GDM-FW900 (CRT) // BenQ RL2455HM (LCD)

"Kolibri King" | Global Rank #1 | 2017.07.27 (10k)

https://battlefieldtracker.com/bf1/leaderboards/all/Kills/weapon/33f9105c-40c7-3980-7982-add4b9f6311a

Offline undazpoon

  • Member
  • *
  • Posts: 8
    • View Profile
Thank you for sharing this. This is a big contribution. I'll try this today with destiny 2 (ps4). I have great expectations.

Offline mjfame

  • Member
  • *
  • Posts: 245
  • "Kolibri King" | Global Rank #1
    • View Profile
    • BF1 Kolibri King aka the BabyPistolPro: #1 in the world. (Semi-Retired)
  • Gamertag: MJ FAME (Active)
  • PSN ID: MJ-FAME (Retired)
Thank you for sharing this. This is a big contribution. I'll try this today with destiny 2 (ps4). I have great expectations.

your welcome.

use SLOW sync.

Ballistic-free: start with a sensitivity around 52.5 or

if you want the highest possible boost to make aim-assist even less hindering just open the ballistic editor and tap the right arrow once then slide the vertical knob up to 100% and save. if you go this route you would start with a sens of 1.05 (52.5x.02=1.05)
« Last Edit: 01:09 AM - 05/15/18 by mjfame »
XIM APEX // BETA : 20180402
Logitech G Pro // Swift-RX XL // XB1X // Sony GDM-FW900 (CRT) // BenQ RL2455HM (LCD)

"Kolibri King" | Global Rank #1 | 2017.07.27 (10k)

https://battlefieldtracker.com/bf1/leaderboards/all/Kills/weapon/33f9105c-40c7-3980-7982-add4b9f6311a

Offline Sl3vin

  • Member
  • *
  • Posts: 169
    • View Profile
I am now in 4000dpi and 500hz polling with ADS 17 and hip 12 (sync common)

With your settings (800dpi and 1000hz polling) wich ADS and hip i need?

I dosent understand the calculating

Offline mjfame

  • Member
  • *
  • Posts: 245
  • "Kolibri King" | Global Rank #1
    • View Profile
    • BF1 Kolibri King aka the BabyPistolPro: #1 in the world. (Semi-Retired)
  • Gamertag: MJ FAME (Active)
  • PSN ID: MJ-FAME (Retired)
I am now in 4000dpi and 500hz polling with ADS 17 and hip 12 (sync common)

With your settings (800dpi and 1000hz polling) wich ADS and hip i need?

I dosent understand the calculating


Old sens 17 >>> New sens 85

Old sens 12 >>> New sens 60


Only DPI needs to be factored when making adjustments to a sync-enabled value but you should still use 1,000Hz instead of 500Hz.

4000 / 800 = 5. So 5x your old COMMON value = your new COMMON value.



does not apply for your question but if for example your original 500Hz value was a NO SYNC value then it would be x10 to determine the new NO SYNC 1000Hz value. again, disregard this 10x multi since you are using frame sync.


« Last Edit: 06:38 AM - 05/15/18 by mjfame »
XIM APEX // BETA : 20180402
Logitech G Pro // Swift-RX XL // XB1X // Sony GDM-FW900 (CRT) // BenQ RL2455HM (LCD)

"Kolibri King" | Global Rank #1 | 2017.07.27 (10k)

https://battlefieldtracker.com/bf1/leaderboards/all/Kills/weapon/33f9105c-40c7-3980-7982-add4b9f6311a

Offline Sl3vin

  • Member
  • *
  • Posts: 169
    • View Profile
Big thx, i test it!!!

Offline lmulli

  • Member
  • *
  • Posts: 121
    • View Profile
I could have sworn you recently started yet another thread on this very same subject where you recommended 800dpi @ 1000hz polling but with sync set to off, and both smoothing and sensitivity set to their maximums, using a ballistic curve to alter the sensitivity.

Whats it to be?  You seem to come up with something new on a daily basis, making grand promises that it's perfect, and yet you clearly aren't satisfied with it, otherwise why keep altering it.  I have no doubts that you spend a lot of time testing your theories, but they seem to change on a daily basis!



Offline lmulli

  • Member
  • *
  • Posts: 121
    • View Profile
You mean this?

https://community.xim.tech/index.php?topic=65974.0

Yep!  And tomorrow, or in the next few days, there will be another one!

Offline mjfame

  • Member
  • *
  • Posts: 245
  • "Kolibri King" | Global Rank #1
    • View Profile
    • BF1 Kolibri King aka the BabyPistolPro: #1 in the world. (Semi-Retired)
  • Gamertag: MJ FAME (Active)
  • PSN ID: MJ-FAME (Retired)
I could have sworn you recently started yet another thread on this very same subject where you recommended 800dpi @ 1000hz polling but with sync set to off, and both smoothing and sensitivity set to their maximums, using a ballistic curve to alter the sensitivity.

Whats it to be?  You seem to come up with something new on a daily basis, making grand promises that it's perfect, and yet you clearly aren't satisfied with it, otherwise why keep altering it.  I have no doubts that you spend a lot of time testing your theories, but they seem to change on a daily basis!



I sure did however I am making attempts to offer the best method with explanation for getting the most out of the standardized method so that there is an offering to suit a variety of user preferences. Not everyone is comfortable enough to venture into the extreme. I get that. I also believe that anyone who refuses to venture into the extreme is cheating themselves out of something pretty special.

My R&D period with XIM APEX began from the very first day of ownership. It lasted roughly 10wks and has been over with since a few days prior to sharing "Perfect Match". No new R&D went into the Purist offering because it is already being used to lay the foundation for the Extreme variant. (800DPI / 1,000Hz)

So additionally you have the Purist offering which clearly was an afterthought but I figured why not share the stripped down model for those who will remain configuring that way. If you're going to stick to the basics then you might as well have the information to help you get the most out of it.

Bits and pieces (precursors) were shared as progress was made in order to let everyone in on it but nothing new is on the horizon. These last two topics are final and these two setups are about as good as it gets for the Purist and for the Extremist. 

For what it's worth I only use 800DPI / 1000Hz for every game. I use the traditional method with my recommended Purist settings for very few titles and for most others I use the Extreme approach. And once I get around to playing those select games that are still using the basic version they will be converted to Extreme which imo is the superior method.

Either way you can't go wrong however the extreme model already matches up better before even beginning to tune the sensitivity via the ballistic editor. This means that it will provide a much greater threshold of refinement and while spending less time doing so.
XIM APEX // BETA : 20180402
Logitech G Pro // Swift-RX XL // XB1X // Sony GDM-FW900 (CRT) // BenQ RL2455HM (LCD)

"Kolibri King" | Global Rank #1 | 2017.07.27 (10k)

https://battlefieldtracker.com/bf1/leaderboards/all/Kills/weapon/33f9105c-40c7-3980-7982-add4b9f6311a

Offline Grarrg

  • Member
  • *
  • Posts: 113
    • View Profile
I am going to give your extreme method a try when I get home.

I use a G502, why do you ustilize 100 mhz over 500 mhz?
I have read reported jitters at 1000 mhz and have always stayed at 500 mhz so to not encounter this issue.

Thank you.

Offline antithesis

  • MVP
  • *
  • Posts: 5453
    • View Profile
Jitter tends to be environmental, ie mousepad surface, dirt particles, pet hair, hand shakiness etc can all cause jitter. Or it could be the mouse sensor itself doesn't truly poll at 1000Hz consistently.

So jitter affects some users, but not others. Try 1000Hz and see if it makes any difference over 500.

Offline Grarrg

  • Member
  • *
  • Posts: 113
    • View Profile
Will do.
Thank you very much for the response.

Offline mjfame

  • Member
  • *
  • Posts: 245
  • "Kolibri King" | Global Rank #1
    • View Profile
    • BF1 Kolibri King aka the BabyPistolPro: #1 in the world. (Semi-Retired)
  • Gamertag: MJ FAME (Active)
  • PSN ID: MJ-FAME (Retired)
Jitter tends to be environmental, ie mousepad surface, dirt particles, pet hair, hand shakiness etc can all cause jitter. Or it could be the mouse sensor itself doesn't truly poll at 1000Hz consistently.

So jitter affects some users, but not others. Try 1000Hz and see if it makes any difference over 500.


thanks brother

i did have a rather poor experience with my limited time using the g502 compared to 500hz but i thing the weight of the @#$% thing was more to blame.

i also began to feel performance degradation beyond 6400DPI while using the G502 @1kHz

jitter shouldn't be an issue at native CPI of 800 though.

« Last Edit: 10:00 AM - 05/17/18 by mjfame »
XIM APEX // BETA : 20180402
Logitech G Pro // Swift-RX XL // XB1X // Sony GDM-FW900 (CRT) // BenQ RL2455HM (LCD)

"Kolibri King" | Global Rank #1 | 2017.07.27 (10k)

https://battlefieldtracker.com/bf1/leaderboards/all/Kills/weapon/33f9105c-40c7-3980-7982-add4b9f6311a

Offline Cizque

  • Member
  • *
  • Posts: 62
    • View Profile
Highly recommend these settings, was already dropping people with 12kdpi but i switched up to this and im just annihilating people. Feels so smooth and the aim is great. This is on cod ww2 not tried bf1 yet i will today.