RuQu Posted February 6, 2012 Share Posted February 6, 2012 (edited) *edit* exact numbers changed after I found a bug in the code. Updated code doesn't show the same peaks, but it does show a loss in AP and MP casts as a percentage of total and a corresponding HPS drop as alacrity rises. Looking into converting the spreadsheet to C++ for more accurate simulation. Edited February 7, 2012 by RuQu Link to comment Share on other sites More sharing options...
Aritok Posted February 7, 2012 Share Posted February 7, 2012 (edited) Sadly you removed it. I wanted to see the AI you created, that could compensate the change of the casting time of the spells and smart change the rotaion . But yes, there will be a point, when your HPS will break down. It mostly happens when you burn your ammo to fast to get it back. You will have to use HS in this "extra" time, when all other is on cooldown. But one extra Hammershot is nothing compared to the insane Alacrity you need for it. Or you have to change the rotaion to optimize it to the changed circumstances. Edit: and yes, i know you can just macro/program a order of spells depending on your current ammo and cooldowns. Even with simple if;then;else. But there are a few skills, that i normaly never use, but suddenly becoming usefull when you have all on cooldown and still have time to spend. Edited February 7, 2012 by Aritok Link to comment Share on other sites More sharing options...
RuQu Posted February 7, 2012 Author Share Posted February 7, 2012 Well, the calculator and rotation logic used are still available here. I'm constantly trying to improve it, but currently its the only tool the CM community has for this purpose. It's extremely labor intensive to find the cast count for every alacrity value over a large range, as in the spreadsheet you either need a separate sheet to do the rotation for each value, which Google Docs can't do due to formula count limits, or you have to change the value used by the one rotation every time and log the result. Coding it in C++ should alleviate all of those, once I get the bugs worked out of that. Link to comment Share on other sites More sharing options...
Buckit Posted February 7, 2012 Share Posted February 7, 2012 So what does this mean in layman's terms? That commandos should try and set alacrity around 140-165? Link to comment Share on other sites More sharing options...
Jooji Posted February 7, 2012 Share Posted February 7, 2012 I looked at the thread before you removed it and I didn't quite understand where you were getting your "soft cap" from. What were your assumptions for there being a soft cap to begin with? Link to comment Share on other sites More sharing options...
Aritok Posted February 7, 2012 Share Posted February 7, 2012 Well, the calculator and rotation logic used are still available here. I'm constantly trying to improve it, but currently its the only tool the CM community has for this purpose. It's extremely labor intensive to find the cast count for every alacrity value over a large range, as in the spreadsheet you either need a separate sheet to do the rotation for each value, which Google Docs can't do due to formula count limits, or you have to change the value used by the one rotation every time and log the result. Coding it in C++ should alleviate all of those, once I get the bugs worked out of that. This sheet may work really well for crit, surge, power and so on. But it does not work well for Alarity. With the upgrade calculator i get as result on a change of 281 alacrity a -72 HPS value. At 300 its suddenly +45 HPS. With 350 its only +38, with 400 suddenly +421 HPS. For me my main goal on Alacrity is 800 (for the 1τ=63%). In long run it is 1450 to get the casting time down to 0,99. Link to comment Share on other sites More sharing options...
Buckit Posted February 7, 2012 Share Posted February 7, 2012 For me my main goal on Alacrity is 800 (for the 1τ=63%). In long run it is 1450 to get the casting time down to 0,99. I don't follow, but casting time down to 0.99 sounds very desireable. Can you please explain. I am but a simple caveman. Link to comment Share on other sites More sharing options...
Aritok Posted February 7, 2012 Share Posted February 7, 2012 I looked at the thread before you removed it and I didn't quite understand where you were getting your "soft cap" from. What were your assumptions for there being a soft cap to begin with? I could only see it once, and saw the diagram. Basicly he was calculating the HPS for each Alacrity value, and adding it to the diagram. This way you could really well see, that the HPS was jumping around (like i as well said in last post). You could really well see, that the HPS was really high at 140-165 and rapidly going down at 315. But the Diagram didnt went far enough to see that it will go up again after the value (like the 400-500 i said in last post as well). I could not read the whole post. It was to late yesterday and wanted to check it today. But it was gone already. But basicly you cant use a straight formular/script to calculate the stats. You only need it with a good combat log in the game. Everything else cant give you good data. Sure, i can easy calculate how much i can heal over a given time on a single target. But this is not the normal way in the game. You have to be able to burst heal one person if possible. And you have to hold back ressources for it. From my experience so far for example. I keep supercharge long at 30 charges for the +3%. Only if i have to, becauce of burst or great AE damage, i will fire it for the 10% and bonus Kobo Bomb or the free Charged Bolt for DPS. But in the end it all comes down to the same. There is no way to say what stat is the best without a combat log. From my understanding of the formular, it comes down to power> all. All other stats in best case evenly spread, so reduce the diminishing returns. Link to comment Share on other sites More sharing options...
Recommended Posts