Please upgrade your browser for the best possible experience.

Chrome Firefox Internet Explorer
×

Nerf Mercenaries


Aetideus's Avatar


Aetideus
01.25.2013 , 04:21 AM | #111
Quote: Originally Posted by Kalifus View Post
I play one on both sides and perform way better on my commando...how is that possible?
Maybe it is the same with other classes like Sorcs/Sages. Maybe they just nerf the final damage output ex. madness for Assassins compared to Sorcs.

-Holmes's Avatar


-Holmes
01.25.2013 , 05:43 AM | #112
Quote: Originally Posted by Macroeconomics View Post
Again, just as was the case with your claims about how there was a negligible difference between how often Commandos/Mercs trigger Riposte/Retaliation, I can see why you'd think this. But you are simply WRONG. Again. As I said, send your combat log to a file. Use your 1 second ability buffer (that's what I had mine set at FYI). Then look at your combat log. What you will see is that even with a 1 second queue buffer, your abilities are NOT triggering 1.5 seconds apart. There is a variance of between .01 and .1 seconds for EACH ability queued. Thus the total variance for a long string of abilities becomes quite large. In order to maintain test accuracy you need shorter sequences.

And please, stop with the claims that Ion Pulse uses up 66.6% more energy than Flame Burst. If this was true you wouldn't see both abilities max out at 8 uses before the resource stack blows up. Anyone can go to a training dummy and test for themselves. You'll get 8 uses for those abilities on either class.
And in a long term fight, say, a ranked warzone or an operation boss? Those "8 even shots" do no good there. The longer the test runs, the further apart the two classes drift. THAT is what we are debating.

FREDDOSPWN's Avatar


FREDDOSPWN
01.25.2013 , 10:42 PM | #113
Quote: Originally Posted by Macroeconomics View Post
Again, just as was the case with your claims about how there was a negligible difference between how often Commandos/Mercs trigger Riposte/Retaliation, I can see why you'd think this. But you are simply WRONG. Again. As I said, send your combat log to a file. Use your 1 second ability buffer (that's what I had mine set at FYI). Then look at your combat log. What you will see is that even with a 1 second queue buffer, your abilities are NOT triggering 1.5 seconds apart. There is a variance of between .01 and .1 seconds for EACH ability queued. Thus the total variance for a long string of abilities becomes quite large. In order to maintain test accuracy you need shorter sequences.

And please, stop with the claims that Ion Pulse uses up 66.6% more energy than Flame Burst. If this was true you wouldn't see both abilities max out at 8 uses before the resource stack blows up. Anyone can go to a training dummy and test for themselves. You'll get 8 uses for those abilities on either class.
As I (and others) have already said, the ability to burn through the resource pool is not the issue. This issue has more to do with longer fights.

This is why the test with a filler was devised, to see a longer-term effect of the heat advantage.

If you are so adamant that he is doing the test wrong, why do you not do the test yourself? Do not try to sweep this issue under the proverbial rug by doing a burn test, but put a filler in and see what happens. The test with a filler move will more accurately reflect the discrepancy between the two resources. Yes, the issues you have mentioned may be an issue (unlikely), but should not the same person have the same timing disadvantages in both tests?

If you cannot be swayed in your assertion that the test is unfair, what test would you accept as viable (that is not a burn test)? You cannot say any that test that is designed to examine the long-term effects of the heat advantage is invalid because of timing issues.

Macroeconomics's Avatar


Macroeconomics
01.26.2013 , 01:48 AM | #114
Quote: Originally Posted by FREDDOSPWN View Post
As I (and others) have already said, the ability to burn through the resource pool is not the issue. This issue has more to do with longer fights.
If the Vanguard's ability truly cost 66.6% more energy than Flame Burst, then you wouldn't need a longer fight to see that. It would be patently obvious almost immediately.

With 8 Flame Bursts/Ion Pulse giving the same resource stack impact, we are already down to a feasible resource consumption difference between the two abilities of around 10-15% at worst - and it's not even clear which uses more, if any. Sure, you can test with even longer ability usage sequences. But as I said, the longer the sequence, the more prone your test is to random error from the abilities not firing with optimal timing. You want me to create a test that is immune from that effect? Sorry, I can't do that. I simply don't have the uber ability to alter micro-second game timing on the fly....

FREDDOSPWN's Avatar


FREDDOSPWN
01.26.2013 , 02:08 AM | #115
Quote: Originally Posted by Macroeconomics View Post
If the Vanguard's ability truly cost 66.6% more energy than Flame Burst, then you wouldn't need a longer fight to see that. It would be patently obvious almost immediately.

With 8 Flame Bursts/Ion Pulse giving the same resource stack impact, we are already down to a feasible resource consumption difference between the two abilities of around 10-15% at worst - and it's not even clear which uses more, if any. Sure, you can test with even longer ability usage sequences. But as I said, the longer the sequence, the more prone your test is to random error from the abilities not firing with optimal timing. You want me to create a test that is immune from that effect? Sorry, I can't do that. I simply don't have the uber ability to alter micro-second game timing on the fly....
I do not want a test that is immune to error, I just want a test with an error margin that you would find acceptable. Then we can go and do that test. The test would give results that show heat is a better resource than ammo (because it is, as previously shown). Since you would have agreed that the test would be a good and conclusive way of determining the difference, you would then not be able to argue that heat and ammo are the same. This would hopefully have a knock-on effect and convince other members of the community (and possible Bioware) that there is an issue that needs resolving.

The person who you are quoting about the 66.6% difference is mistaken. They are comparing Ion Pulse ammo cost (2 / 12 = 0.1666) to Flame Burst heat cost (16 / 100 = 0.16).

0.1666 - 0.16 = 0.0066, or 0.66% difference (a decimal error on their part).

EDIT: Dr_Kid replied with his calculations, so feel free to ignore mine on the matter. I'll keep them in for posterity.

PerinnAybara's Avatar


PerinnAybara
01.26.2013 , 02:10 AM | #116
There's plenty of ways to do it.

The main thing is you don't just use one example to prove conclusively your data is right.

You have a larger sample size. Don't bother about when you start, just start randomly and spam that flame burst.

As long as you do the same thing each time for each character, it should average out to be the same results. Now if a powertech consistently gets 9 shots while a vanguard gets 8 every single time. (Out of say 100 trials) then you have a point. To measure the difference between the two abilities, (which from what everyone has said is miniscule) the more and more trials you'll need to do to actually get a statistically significant result.

With a sample size of 1, or 2, or 3 it truely means nothing. Imagine you're flipping a coin. YOu flip a coin once. It's heads! You now believe the odds are 100% heads? What happens you flip it again and it's heads again? 100% chance for heads? That's now how statistics works.
Character: Mathrim Cauthon, 50 Mercenary on Prophecy of the Five

Dr_Kid's Avatar


Dr_Kid
01.26.2013 , 02:26 AM | #117
Quote: Originally Posted by Macroeconomics View Post
If the Vanguard's ability truly cost 66.6% more energy than Flame Burst, then you wouldn't need a longer fight to see that. It would be patently obvious almost immediately.

With 8 Flame Bursts/Ion Pulse giving the same resource stack impact, we are already down to a feasible resource consumption difference between the two abilities of around 10-15% at worst - and it's not even clear which uses more, if any. Sure, you can test with even longer ability usage sequences. But as I said, the longer the sequence, the more prone your test is to random error from the abilities not firing with optimal timing. You want me to create a test that is immune from that effect? Sorry, I can't do that. I simply don't have the uber ability to alter micro-second game timing on the fly....
It costs 4.1% more energy. A short fight isn't going to show that. The difference is Flameburst costs 16% heat, and Ion Pulse costs 16.6666% Ammo. Spamming Flameburst or Ion Pulse themselves won't give the most definitive results.

Here's where the 66.66% figure comes from.
Flameburst + Rapid shots is 3 seconds (5 Heat Regen a second)
Flameburst uses 16 Heat, and gains 15 heat (5 heat regen for 3 seconds). Net Loss: 1 Heat.

Ion Pulse uses 16.666% Ammo, and gains 15% Ammo. Net Loss 1.6666 Ammo

Vanguards lose 66.66% more Ammo for every Ion Pulse (assuming the filler is included)

Go Test it yourself for once and you'll see the results. Any random lag is going to affect both characters equally. Want to reduce randomness? Run multiple tests, average the results. I got very similar results both times. Powertech was able to use about double the Flamebursts than compared to my Vangaurd.

According to you, my Powertech must somehow be permanently stuck with extra lag than my vanguard adding delay to only his attacks, despite using the exact same internet connection for both. Apparently this bug only affects Bounty Hunters, with absolutely no effect on Troopers, which you're basically now implying Troopers > Bounty Hunters.

Stop denying the facts and go test it yourself, If the lag is as bad as you'll make it, then you'll get significantly different results (something like 30 Flamebursts run 1, 50 Flamebursts run 2). Run it multiple times on your Powertech and then your Vanguard and compare the results. If the costs are working as intended, then you shouldn't see any pattern.

FREDDOSPWN's Avatar


FREDDOSPWN
01.26.2013 , 02:36 AM | #118
Quote: Originally Posted by PerinnAybara View Post
There's plenty of ways to do it.

The main thing is you don't just use one example to prove conclusively your data is right.

You have a larger sample size. Don't bother about when you start, just start randomly and spam that flame burst.

As long as you do the same thing each time for each character, it should average out to be the same results. Now if a powertech consistently gets 9 shots while a vanguard gets 8 every single time. (Out of say 100 trials) then you have a point. To measure the difference between the two abilities, (which from what everyone has said is miniscule) the more and more trials you'll need to do to actually get a statistically significant result.

With a sample size of 1, or 2, or 3 it truely means nothing. Imagine you're flipping a coin. YOu flip a coin once. It's heads! You now believe the odds are 100% heads? What happens you flip it again and it's heads again? 100% chance for heads? That's now how statistics works.
I know what you are saying. My point is that it is very hard to do a test 100 times and put it on video to prove that it has been done. It is then even harder to get enough people to watch the video of the test being done 100 times.

It is unreasonable to expect 100 videos (Holmes' shortest was 113 seconds, so 11300 seconds in total). You would not watch the 100 videos, and you refuse to take our word on results we have without videos. The solution?

I want a test:
That can be easily reproduced.
That is easy to perform.
Is long enough to show the resource discrepancy.

Since a burn test is not long enough to show the difference, what Holmes did is probably the next best thing.

Spam Flame Burst or Ion Pulse until you run out of resources, hit your resource regain ability, spam some more (Holmes, please correct me if I am wrong). He did his test twice (that we have proof of) to convince people that there is an issue. Both times came up in favour of heat.

My point is that you can easily do this test to your heart's content. If you think the test is wrong, repeat the test until you get a result that is inconsistent with ours. We will have to supply many videos to prove our point, you have to supply very few to disprove it.

If you do not think what Holmes did is a valid test, suggest one so that we can prove we are correct.

PerinnAybara's Avatar


PerinnAybara
01.26.2013 , 02:43 AM | #119
Apply the scientific method!

The null hypothesis!

Ecetera.
Character: Mathrim Cauthon, 50 Mercenary on Prophecy of the Five

PerinnAybara's Avatar


PerinnAybara
01.26.2013 , 02:44 AM | #120
Actually now that I think about it, you guys are basically saying that Mercenaries get 100 heat while powertechs get 96 heat. 12*8. If they don't have any rounding it's true. If they do, then it's not.
Character: Mathrim Cauthon, 50 Mercenary on Prophecy of the Five