Xonotic Forums

Full Version: High Arc Accuracy + Damage & XonStats
You're currently viewing a stripped down version of our content. View the full version with proper formatting.
Recently I found out how to get 70-95% overall accuracy with the Arc within a match, by extension, having it deal a truckload of damage, and nearly every game I've done that, it seems to have been memory holed by XonStats.

Is it pure coincidence (the contingency it's not recorded), or is it intentional such that the system distrusts the data and discards it?

Image not related, but how would XonStats handle accuracies above 100%?
@Antares - Did *any* of them get saved? If so, can you point me to a game_id? If not, is this coming from your /v/ server?
Update: I analyzed my current set of logs for occurrences where the number of "fired" arc shots was less than the number of "hit" arch shots. I tabulated both damage (actual and potential damage) and count (shots fired/hit) numbers. I found one occurrence out of the 131 games I looked at and it was off on the damage numbers only. 

e acc-arc-cnt-fired 52
e acc-arc-cnt-hit 52
e acc-arc-fired 172
e acc-arc-hit 200

Edit: It's also presumptuous to imply that XonStat ate your data. There's nothing in my logs to indicate that was the case. If you have something that points to that situation, I'd be happy to investigate further.
(07-11-2017, 08:55 AM)Antibody Wrote: [ -> ]Update: I analyzed my current set of logs for occurrences where the number of "fired" arc shots was less than the number of "hit" arch shots. I tabulated both damage (actual and potential damage) and count (shots fired/hit) numbers. I found one occurrence out of the 131 games I looked at and it was off on the damage numbers only. 

e acc-arc-cnt-fired 52
e acc-arc-cnt-hit 52
e acc-arc-fired 172
e acc-arc-hit 200

Edit: It's also presumptuous to imply that XonStat ate your data .There's nothing in my logs to indicate that was the case. If you have something that points to that situation, I'd be happy to investigate further.

This is why I'm asking if it was pure coincidence, the chance(s) that the packet didn't make it to XonStats, or if XonStats has any mechanism to discard data...
In your server logs you would most likely have an HTTP 400 or 500-level error returned back to you in text format.