![]() So after reviewing your data I have decided to make two changes to PARSEC:ġ) Combat end events will be handled the same way but calculations based on time will now use the last combat end event time instead of including the floating combat window.Ģ) Damage taken will now automatically include absorbed damage. Obviously this skews the numbers by 1 second. PARSEC opens a 1 second window after a combat end event in order to pick up any out of order combat events. Because of the out of order logging that occurs with the combat end event each parser has a different approach to combat end. This difference is explained by how PARSEC deals with the combat end event. Yes some of the other numbers are slightly different. Damage done, APM, healing recieved, etc is different. Unfortunately, they are still not identicle. And if you have any feelings about it, share them and convince me to change it ![]() As a tank myself I can see the value in both calculations. This difference in calculation affects all of the damage taken numbers. If you add the Abs and total column from PARSEC you get 303614 which is the same as TorParse's numbers. The difference you are seeing is that TorParse's damage taken calculations include absorbed damage. PM would probably be the best way to contact me.Īctually, they are identical. If there is more information I can give, I would be happy to. One of them reads horribly inaccurate, the question is which. While I can confirm they are the same fight. ![]() I am submitting the log here for your inspection, along with the Torparse link.Įdit: Upon further inspection, almost no information from the two fights are the same. While I found the tanking information quite nice I have found some pretty large discrepancies between parsec and torparse. You could have the database automatically, provisionally classify attacks based on circumstantial evidence (if you see it dodged it is M/R, if shielded it is M/R or K/E, if resisted it is F/T, etc.) and then let some of the other number crunchers post more authoritative classifications that get marked in the database as confirmed. I feel like a database of attack types would be so useful to be worth it. It is possible to build a database of attacks and their attack type based on mitigation results over time, but this is more work than it is worth since it would be useless on new content. With this in mind, i designed parsec to display total mitigation rates like this:Īs a tank myself i find these statistics usefull in determining my performance and gear effectiveness even though they are not ideal. These limitations along with the complexity of the mitigation rules make it impossible to accurately display mitigation in seperate buckets. Second, if an attack is avoided its damage type is not reported in the log. First, log files do not specify ranged/melee vs force/tech. Unfortunately, the log files have some limitations when it comes to damage types and attack types.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |