Does anyone have any actual numbers on this, with and without ASI, for various guns?
I tend to think full clip has higher DPS and shoot-pause is safer, but I haven't done any testing.
Shoot-pause or full clip?
Tue, 10/11/2011 - 06:59
Tue, 10/11/2011 - 08:18
#2
almost the same dps if you
almost the same dps if you have very high ASI. but shoot pause is slightly more
Wed, 10/12/2011 - 06:26
#3
i'm not sure that guy's
i'm not sure that guy's testing was accurate or complete.
this is what fossaferox had to say about it:
"If each shot is X seconds apart, the necessary pause to avoid a reload is Y seconds, and a reload is Z seconds:
Shot shot reload has a cycle time of X+Z.
Shot Pause Shot Pause has a cycle time of Y+Y.
Which method is better depends on which is less, X+Z or Y+Y. IAS only lowers the value of X, but it does it pretty dramatically. Y is already upwards of 0.5Z, though it is difficult to measure precisely. With a low enough X value, X+Z is not only less than Y+Y, it is dramatically less."
in this thread: http://forums.spiralknights.com/en/node/5286
also, when i shield cancel my shots and then shoot again, i have to wait or i'll get reload.
that seems like a wasted space of time when i could have hit reload faster and already been on my next clip by spamming.
Wed, 10/12/2011 - 08:34
#4
you forgot W, the time to
you forgot W, the time to takes to actually fire a shot starting from rest
so it becomes:
shot shot reload = W + X + Z + W + X + Z + ...
Shot pause repeat = W + Y + Y + Y + ...
and W and Y are both affected by ASI.
and i saw u have 2 alchemer lines, why don't u go to training and test if for us?
Wed, 10/12/2011 - 08:41
#5
[Redacted]
I'm tired.
Wed, 10/12/2011 - 11:10
#6
@Somenubcake
i think the point of what fossaferox was saying was that no matter what, Y is going to be Y and that if you are going to shot-pause, you don't need to max your ias.
sustained dps might be arguable, i don't know about that, i'm not much of a mathematician, but burst dps is definitely higher by hitting both shots on your clip.
it doesn't take much playtime to realize that.
i've also seen someone argue that sustained dps doesn't exist in this game because of all the movement.
that seems to be true, though i couldn't say with certainty.
however, if there is no sustained dps, then burst dps is even more important.
since i maxed my ias and go to the clip with alchemers, i have noticed that soloing is a lot faster.
when i'm shot-pausing, i get impatient.
Wed, 10/12/2011 - 13:33
#7
Raunwynn
Fossa's math is wrong. he didn't take into account of W nor did he take into account of the fact that Y is affected by IAS while Z isn't. and by doing that, he is effectively skewing his final results so that the shot shot reload scenario gives you more dps.
fern's shots per 30s is accurate. it's a theoretical amount of damage of course (there are times when u can't fire and have to run, and u can miss). i know you have 2-3 alchemers, just go to training hall and test it out yourself.
now to your point about burst dps:
burst dps is dependent on the amount of time you are given. being a gunner gives you a huge window of time to shoot at your enemies without being forced to shield/run. therefore sustained dps IS relevant for a gunner. and if you really want burst dps, use a sword lol...
Wed, 10/12/2011 - 15:03
#8
@Somenubcake
you know, at first i got pissed off at you, sending me to test [stuff] out like some flunky.
it's not even my argument!
it's just some stuff that fossa said that seemed to be right according to my experience.
then i got curious and went ahead and did it.
alright, i'm just a human, trying to press buttons on time while watching my computer's stopclock, so the guns could probably perform better than this (i got a few accidental reloads during shot-pause and i was a little slow on some of the full-clips), but the results seemed to be consistent. (as far as i know, the alchemers are all equal speed, so the variation in the results should come from errors in my timing.)
i fired 3 different alchemers at maximum asi for 1 minute on full clip and 1 minute on shot-pause then did it again with minimum asi.
the values i'm listing are the number of shots i was able to get:
umbra driver
max asi: 76 full clip, 66 shot-pause
min asi: 64 full clip, 66 shot-pause
magma driver
max asi: 74 full clip, 67 shot-pause
min asi: 64 full clip, 65 shot-pause
storm driver
max asi: 78 full clip, 70 shot-pause
min asi: 62 full clip, 67 shot-pause
my results seem to show what fossa was talking about.
asi is good if you are using your full clip and doesn't make a difference if you are using shot-pause because the time to cause reload stays the same.
Wed, 10/12/2011 - 15:47
#9
Thx Raunwynn!
I wonder at
Thx Raunwynn!
I wonder at what point it breaks even... but.... that's asking a bit much.
In the grand scheme of things, it doesn't make a big difference for DPS, it just affects your ability to quickly shield.
Thu, 10/13/2011 - 22:17
#10
Raunwynn
thanks for testing it out! sorry for sounding like a douche.
EDIT:
i got myself a shadowtech alchemer ASI med. got nameless set on so final ASI is maximum.
timed myself for 30seconds, 3 replicates. if i messed up the shot pause shot (like accidentally caused a reload), i discard and redo that shot pause shot trial
shot shot reload : fired 38 shots
shot pause shot: fired 38 shots +/- 1shot
conclusion: same number of shots in 30s with max asi
with jsut the shadow tech alchemer asi med, no nameless set. final asi = med
same experiment
shot shot reload: fired 34 shots
shot pause shot: fired 36shots +/- 1 shot
conclusion: shot pause shot lets me fire more with asi med
i have no clue how u got such a low amount raunwynn.
Thu, 10/13/2011 - 22:16
#11
bumped with more testing
bumped with more testing
Fri, 10/14/2011 - 00:25
#12
The testings are ...
The testings are ... confusing. When I did a "one minute test" i allways had some bullets more with shot-wait-shot at alchemers for ex... I also realized that you can do with magnus: shot-shield-shot-RELOAD (WTF!?). XD
How ever, sometimes you have to be safe, sometimes you need burst damage. But with 3+ shots per clip ALLWAYS DO NOT RELOAD!!
Fri, 10/14/2011 - 00:43
#13
i don't know, you might just
i don't know, you might just be better than me at shot-pause.
the best i did over a 1 minute test was 70, and even then i had a reload in there.
the time i got 65 during shot-pause with my magma driver i got like 4 reloads during shot-pause.
i don't know of any way to test the true speed of shot-pause because one tiny misjudgement in timing gives you a reload and messes everything up.
abathur, i get reloads after shot-shield with alchemers too.
on the test i did, i just timed single shots and didn't cancel them with shield.
i don't know if that makes difference or not.
Fri, 10/14/2011 - 01:22
#14
I didnt cancel them with
I didnt cancel them with shield - I just noticed this once at my Magnus ... also I noticed that Pummel Gun is the only gun with a 1 shot clip :D ROFL xD and while reloading with Pummel you don't move slower :O XD
btt :P I think m' gonna test this again ... maybe I made a mistake. But how ever - it's just theory - in a real fight you allways will 2 shot against mechas for ex, but never will against mewcats - they're just too fierce o.O same for greavers ... but there i farly prefer polaris due having one more shot per clip, and we all know (i hope) greaver cancel their attacks when beeing attacked. Piercing weapons are pretty useless against their T3 mobility.
Fri, 10/14/2011 - 04:39
#15
i don't know of any way to
i don't know of any way to test the true speed of shot-pause because one tiny misjudgement in timing gives you a reload and messes everything up.
@ Raun, I have a death adder that supports its own macros. What asi, and what line do you want tested? I'm pretty sure I can just do a type of action button < insert delay > action button untill I trim the delay of millisecconds to the smallest figure that wont trigger a reaload and spam it.
I have fooled with this functionallity in the past, and one interesting thing to note is that ASI seemed to skew and lose a small portion of value while moveing.
For example I trimmed the macro:
A being action button or shoot of course...
A < delay > A < delay > A
Down to the smallest delays I could fit between shots in millisecconds and was able to spam off all three shots at a standstill, however while spamming the same macro while moveing it was very problematic about getting all three shots sequenced off and frequently ( 1/10 ) times would only fire two.
Fri, 10/14/2011 - 09:49
#16
@luke
wow, that's cool!
try an alchemer with max asi and min asi over a minute and compare shot-pause to full clip.
i'm interested in what you'll find.
Fri, 10/14/2011 - 15:55
#17
Testing.
In order to preform the timed test I would like a seccond available to time and count as I dont feel like I can do both accurately along with the steady shot macro, if I spam it too much it loops ahead and crashes XP (lol). However I do have some interesting preliminary findings that may put some more pieces in place (im rlly bad at math) for someone who can combine them useing Fossa's formula.
A) Maximum! ASI, fully leveled Hail Driver.
B) No ASI what-so-ever, same weapon.
Quickest time between <----------->prone to reload at least 1/100 volleys of 2<---->Threshold of 0 reloads in 100 clips of 2
shots without anomalys
(extra or missing
sound/animation)
A)-----------------235MS ----------------------------579MS<------------>618MS-------------------------------------------619MS -------------------------------
B)-----------------415MS--------------------------------(not really importent?)----------------------------------------------766MS-------------------------------
It seems to me that server latency plays a large part in these findings, for instance the last figure 766, when I had the delay set to 765, I got off a good 45 2-shot clips before a reload occured. I'm almost sure even though I had got off 100 2-shot clips with a 766MS pause without having to reload, had I tried 1000, or even 500, I would have gotten a reload sooner or later with the kanky way Java runs the game.
My personal opinion is that this is a mountain of haystacks with many needles because as Abathur and others have mentioned, if you are a decent shootist, during the majority of gameplay your surroundings and immeadiate situation will dictate wether you unload or plink away.
~Luke
Fri, 10/14/2011 - 16:10
#18
hmm.well, it might come down
hmm.
well, it might come down to what the individual shooter is best at, given his familiar latency.
also, there is a way to get a very fast two-shot clip by clicking attack and then holding attack.
i don't know if the people who favor shot-pause are doing this.
i noticed that if i just spam the attack button, i'll get two shots that aren't very close together and a then a reload sometimes, but if i click and then hold, i'll get two shots very close together every time.
also, i agree with you about this being situational.
if i have room to move around and there are no enemy projectiles, i just spam.
if i am coming in close to some mobs or dodging bullets, i interrupt single shots with shield.
Sun, 10/16/2011 - 15:46
#19
Some mre spading for you guys.
Might be a little late here, but I decided to try this out as well.
Equipment used: Hail Driver, Ultra ASI (Sorry, didn't have anything for that last point.)
Process 1:Used an Auto-clicker program to find the lowest millisecond setting that caused consistent reloads (at least 20 sets without a shoot-pause-shoot)
Shots per minute for comparison:
0 ASI = 60
5 ASI = 73
Reload Threshold
0 ASI = 743 ms
5 ASI = 717 ms
reloads per 60 sec ( 60 * .[clickspeed])
0 ASI = 80.754 sets
5 ASI = 83.645 sets
This shows that there should be a difference of 3 shots between ASI 0 and 5 using Shoot-pause-shoot
Was going to post this sooner, but ran out of time writing it up. Came back later to check results.
Process 2: Used auto-clicker to find lowest millisecond setting that allowed for Shoot-Pause-Shoot. Restarted if it reloaded, repeated several times with same results.
Minimum speed
0 ASI = 786 ms
5 ASI = 760 ms
Shots per 60 seconds (SPS/Reload)
0 ASI = 74/60
5 ASI = 77/73
So yeah, 5 ASI gives a whopping 3 extra shots per minute, if you can do it perfectly, and SPS does allow for more shots, although this is less helpful at higher ASI.
Might come back later and try it with a few other guns, as I do have an Antigua and a Magnus on hand. Nothing that wasn't figured out already, but at least it's hard numbers.
there is a write up in the "ask a master gunslinger anything" thread that proves what you've said.
shoot-pause is safer, full clip is higher dps.