65 replies to this topic

[Just wanna play]

OneEyed Jack, on 23 December 2013 - 12:55 AM, said:

COULD HAVE SWORN HE SAID UAC5 IS THE BURST WEAPON.....

[D04S02B04]

TL:DR - Your method of calculating the UAC5's DPS is wrong and you should use a probability tree analysis. Watch/read carrioncrow's guide

] to using UAC5 and the UAC5 is actually

pretty balanced where it stands.

--- --- ---

Click on *Show* for long reply and explanation.

Spoiler

The UAC5 should be analysed via using a probability tree analysis, accounting for all possible scenarios in an appropriate pre-determined time frame.

To determine the time frame for testing, I used the Lowest Common Multiple (LCM) of the 2 Scenarios. A successful double shot resulting in 1.5sec CD and a Jammed Attempt which Results in 1.5sec + 5sec = 6.5sec. The LCM is 19.5 sec.

The main problem with this model of analysis is that if you keep adding 0.5 sec to the time frame of recording shots fired, you'll always find that you can squeeze in another set of probabilities and skewing the test result. Therefore, I set the cut-off for placing addiitonal shots in at 18.5 sec because it takes slightly more than .5 sec to fire off 2 shots in double shot mode on the UAC5.


UAC5 Probability Table by D04S02B04, on Flickr

You will derive a mathematical probability average of 0.76989 shots per second or 3.849449 damage per second. This puts the weapon on par with the AC2 for DPS.

Utilising a simple statistical observation collection of shots fired over a 10 second time frame, the most common scenario is 7 shots fired and 1 jam, correlating closely to the predicted 7.6989 shots per second.

Take note that the mechanism for testing for jams is a Random Number Generator and as computers are all logical, they can never be truly random, therefore there will be biases and skewing from a normal distribution in determining jams and shots fired. Given a small opportunity of fire window (anywhere from 3 -10 seconds), this results in higher than "average DPS" and lower than "average DPS"... One of the best scenarios firing 6 shots in a 3.5 second window (Total CD time 4.5sec) beating even the AC20 (30 dmg vs 20dmg) as long as you put your shots on target.

As explained by carrion crows in his post on one of the (yet again) UAC jamming thread:

Carrioncrows, on 28 November 2013 - 09:11 AM, said:

Gentleman's guide to using the UAC5

You only use the UAC5 to doubleshot when the opportunity presents itself (e.g. When people shut down, present a full target, when enemies charge etc.)

Picking the UAC5 over the AC5 is trading off an extra ton for the ability to squeeze off another shot when the situation calls for it.

So hopefully you understand that the UAC5 is in fact, a superior and highly skilled weapon with a small element of luck due to randomness.

Edited by D04S02B04, 24 December 2013 - 01:54 PM.

[Varent]

D04S02B04, on 24 December 2013 - 01:53 PM, said:

TL:DR - Your method of calculating the UAC5's DPS is wrong and you should use a probability tree analysis. Watch/read carrioncrow's guide

] to using UAC5 and the UAC5 is actually

pretty balanced where it stands.

--- --- ---

Click on *Show* for long reply and explanation.

Spoiler

The UAC5 should be analysed via using a probability tree analysis, accounting for all possible scenarios in an appropriate pre-determined time frame.

To determine the time frame for testing, I used the Lowest Common Multiple (LCM) of the 2 Scenarios. A successful double shot resulting in 1.5sec CD and a Jammed Attempt which Results in 1.5sec + 5sec = 6.5sec. The LCM is 19.5 sec.

The main problem with this model of analysis is that if you keep adding 0.5 sec to the time frame of recording shots fired, you'll always find that you can squeeze in another set of probabilities and skewing the test result. Therefore, I set the cut-off for placing addiitonal shots in at 18.5 sec because it takes slightly more than .5 sec to fire off 2 shots in double shot mode on the UAC5.


UAC5 Probability Table by D04S02B04, on Flickr

You will derive a mathematical probability average of 0.76989 shots per second or 3.849449 damage per second. This puts the weapon on par with the AC2 for DPS.

Utilising a simple statistical observation collection of shots fired over a 10 second time frame, the most common scenario is 7 shots fired and 1 jam, correlating closely to the predicted 7.6989 shots per second.

Take note that the mechanism for testing for jams is a Random Number Generator and as computers are all logical, they can never be truly random, therefore there will be biases and skewing from a normal distribution in determining jams and shots fired. Given a small opportunity of fire window (anywhere from 3 -10 seconds), this results in higher than "average DPS" and lower than "average DPS"... One of the best scenarios firing 6 shots in a 3.5 second window (Total CD time 4.5sec) beating even the AC20 (30 dmg vs 20dmg) as long as you put your shots on target.

As explained by carrion crows in his post on one of the (yet again) UAC jamming thread:




You only use the UAC5 to doubleshot when the opportunity presents itself (e.g. When people shut down, present a full target, when enemies charge etc.)

Picking the UAC5 over the AC5 is trading off an extra ton for the ability to squeeze off another shot when the situation calls for it.

So hopefully you understand that the UAC5 is in fact, a superior and highly skilled weapon with a small element of luck due to randomness.

agree on the fact that uac5 is very balanced where it stands.

Disagree on using this system to judge it.

[Nik Van Rhijn]

Varent, you disagree with using DPS to judge weapons as it "doesn't apply to shooters". What system (and data) do you use to judge weapons?

[Varent]

Nik Van Rhijn, on 24 December 2013 - 02:18 PM, said:

Varent, you disagree with using DPS to judge weapons as it "doesn't apply to shooters". What system (and data) do you use to judge weapons?

The problem is that DPS is a system based on prolonged firing. Issue being prolonged firing is something that rarely occurs in a shooting game. And specifically rarely occurs for extended periods of time. At a static range value.

Example - Mech A has really high dps based on his weapons, Mech A goes to start firing at mech B, who promptly takes cover.

Example - Mech A has really high dps at 600 meters. He is firing at 1100. His shots are doing minimal amounts of damage.

Example - Mech A has really high dps he exposes himself to start firing at mech B and is promptly killed because he is standing out in the open to fire for an extended time.

Example - Mech A has really high dps but is attacked by a light mech whom he has trouble hitting with his high dps weapons, he cannot effectively hit the mech and therefore cannot dps at all.

I could go on with further examples but I think this gets my point across. And this doesnt even go into the possible versatility options that certain weapons have as well as the bullet speed, etc etc...

The worst thing you can do to judge a weapon is look at DPS alone.

Playtesting a weapon is much much better. That said there is NO mathematical system out there capable of effectively calculating the worth of a weapon.

[Caleb Brightmore]

Varent, on 24 December 2013 - 02:27 PM, said:

The problem is that DPS is a system based on prolonged firing. Issue being prolonged firing is something that rarely occurs in a shooting game. And specifically rarely occurs for extended periods of time. At a static range value.

Example - Mech A has really high dps based on his weapons, Mech A goes to start firing at mech B, who promptly takes cover.

Example - Mech A has really high dps at 600 meters. He is firing at 1100. His shots are doing minimal amounts of damage.

Example - Mech A has really high dps he exposes himself to start firing at mech B and is promptly killed because he is standing out in the open to fire for an extended time.

Example - Mech A has really high dps but is attacked by a light mech whom he has trouble hitting with his high dps weapons, he cannot effectively hit the mech and therefore cannot dps at all.

I could go on with further examples but I think this gets my point across. And this doesnt even go into the possible versatility options that certain weapons have as well as the bullet speed, etc etc...

The worst thing you can do to judge a weapon is look at DPS alone.

Playtesting a weapon is much much better. That said there is NO mathematical system out there capable of effectively calculating the worth of a weapon.

Well played sir!

DPS execution is largely a matter of pilot skill.
Just because you carry a high dps does not mean you can hit me.
It just means that if we trade shots chest to chest AND we happen to have the same mech with the same armor left etc that you will win assuming that my lower dps isnt to your head in which case your DPS goes right out the window along with your mech.

Skill, teamwork, and using ALL weapon systems correctly wins matches not DPS.

Ask all those smoking Atlas Pilots that met my Spiders.

UAC/5 seems to be working as intended it just may not fit some pilots play styles any longer.

[Deathlike]

It's amazing how we do TheoryWarrior instead of playing MWO to see how the mass proliferation of UAC5s are in force (well, not always, but more often than not).

[Khobai]

Quote

What is the DpS compared to the AC5? An Ultra should be doing abourt +50% damage DpS wise at least.

Currently, UAC/5s do about 10% more dps. Which is a little low. But 50% more would be way too high. UAC/5s were overpowered when they were doing 30% more damage... 50% would be broken as hell.


IMO this is how UACs should work:

1) UACs should do about 20% higher dps than regular ACs. Offset by the following:

2) Instead of random jam chance, UACs should have an overheat bar, which increases while the weapon is being fired and decreases while the weapon isnt being fired. If the overheat bar maxes out the UAC will jam and may even suffer critical damage.

3) UACs should have a spool up period, where they start firing slowly, and take 1-2 seconds to reach their full rate of fire. This spool up period would help prevent abuse of the overheat bar mechanic.

Edited by Khobai, 25 December 2013 - 10:09 AM.

[Caleb Brightmore]

Khobai, on 25 December 2013 - 09:52 AM, said:

Currently, UAC/5s do about 10% more dps. Which is a little low. But 50% more would be way too high. UAC/5s were overpowered when they were doing 30% more damage... 50% would be broken as hell.


IMO this is how UACs should work:

1) UACs should do about 20% higher dps than regular ACs. Offset by the following:

2) Instead of random jam chance, UACs should have an overheat bar, which increases while the weapon is being fired and decreases while the weapon isnt being fired. If the overheat bar maxes out the UAC will jam and may even suffer critical damage.

3) UACs should have a spool up period, where they start firing slowly, and take 1-2 seconds to reach their full rate of fire. This spool up period would help prevent abuse of the overheat bar mechanic.

I would say 15% and leave as is but that is just my "off the top" number.

This idea is quite good at least in theory.

I wish we could test this.

It would be really cool if PGI would let us test things since they seem to "test" different than we tend to play.

Not to play Devils advocate but without being able to actually test anything all of our ideas are just theory.

I do however think your idea is well thought out and may have some merit as a solution but honestly think it's working as intended people just need to change their tactics to what we have instead of what we want.

The "Press R to Win" (League Of Legends reference to Karthus) should not exist in this game and unfortunately thats what some people think the UAC/5 is.

Adapt or be overrun seems to be the best choice.

Deathlike, on 25 December 2013 - 09:42 AM, said:

It's amazing how we do TheoryWarrior instead of playing MWO to see how the mass proliferation of UAC5s are in force (well, not always, but more often than not).

I wholy agree sir but at times its all we have due to the "WHAT ITS NOT AN AUTO WIN WEAPON OMG ITS BROKEN!!!" mentality of some players.

[xMEPHISTOx]

Egomane, on 23 December 2013 - 04:48 AM, said:

Accusing others of trolling with meme pictures, not realizing, you yourself are often enough the trolls.

Trust me...if I am trolling I am fully aware of it, even if slandering another whom himself is trolling as well. TROLOLOLOL.

[Axeman1]

Just another baddy that sucks at the game.

3 Uac/ 5's serves me fairly well. 500-700+ damage a game.

Single Uac 5 is also very strong.

[Caleb Brightmore]

Trauglodyte, on 24 December 2013 - 11:34 AM, said:

Hrm, I had (past tense) a BJ-1 build with a UAC5 and a PPC and a (past tense) SHD-5M with a UAC, PPC, and Streaks. I still run my 3D with a PPC, UAC5, and 2 Lrg Lasers and is the only mech that I don't worry about the jam issue.

I have a question Traug.

I noticed that most of your builds center around ONLY using 2 weapon groups NOT two weapons just that according to your builds The PPC & UAC/5 can be grouped & chained with no BIG issue if you are DHS'd correctly.

Do you have issues in game with managing 3 or more groups?

I know that some poeple do because it requires the "driving hand" to fire the third group.

If so and this is completely understandable I would make the UAC/5 the 3rd group for you since it doesnt work well when you use it and I would make the second group for streaks/paired with mls possibly chained depending on how many etc.

Example: SHD-5M with a UAC, PPC, and Streaks. <--- your build. left mouse ppc. right mouse streaks. center mouse uac/5.

Again this is just a suggestion based on the types of builds you showed.

Personally I would dump the single UAC/5 because the way you use it, it is a regular AC/5 with jamming issues if you think about it.

Two UAC/5s chained works quite well if you position correctly but one by itself imo doesnt work well.

Or even better dump a bit and AC/10 if the build will allow it.

Hopefully this helps if not I apologize.

Edited by Caleb Brightmore, 25 December 2013 - 02:03 PM.

[DarkBazerker]

I still don't like the new firing rate of the ultraAC5 but I understand why they changed it.

[Just wanna play]

the uac5 is intended to have a chance to have an ac/10 for the weight of an ac5 but with a chance to jam, why give it more dps then anything except ac/20 with out even chance to jam/double firing?

Edited by Just wanna play, 22 January 2014 - 12:28 PM.

[ATao]

UAC5 is finely balanced against AC5.

3,66 vs 3,33 = ~9,9% more dps
9 vs 8 = 12,5% more weight

Now too many people forget how to properly use UAC5. It is NOT meant to always run in ultra mode. Most of the time you should try to shoot from middle ranges (400-600m) with single pinpoint alphas. Only when you are absolutely sure that you can land all your ultra into the desired mech location ONLY THEN you spam it for that extra burst. That's it.

Too many pilots use it as machine gun on middle to far ranges, splash it all over the target and get frustrated by jams. That is not weapon's fault, only their's.

So l2p 1st and whine 2nd. In fact after you've l2p you'll never whine. Almost .

Edited by Alexander Malthus, 22 January 2014 - 01:00 PM.

[Rhaythe]

Thorasta, on 22 December 2013 - 10:14 PM, said:

There have been some changes to the UAC5

Several months ago:
It jammed 25% of the time, and took 3 seconds to un-jam.
the mean DPS was just over 5.0 making it the highest damaging weapon in the game.

Now:
It jams 20% of the time for 5 seconds.
in 7 tests to get a rough mean, it is doing 3.6 DPS (the average is also 3.6).

So quit spamming it.

[Just wanna play]

Alexander Malthus, on 22 January 2014 - 12:51 PM, said:

UAC5 is finely balanced against AC5.

3,66 vs 3,33 = ~9,9% more dps

both are 3.33, just a lil heads up...

[Ghogiel]

Just wanna play, on 22 January 2014 - 12:58 PM, said:

both are 3.33, just a lil heads up...

The UAC can double shot with a chance to jam. Which is why it's average DPS is higher.

> though I think his math might be a bit wrong

Edited by Ghogiel, 22 January 2014 - 01:02 PM.

[ATao]

Just wanna play, on 22 January 2014 - 12:58 PM, said:

both are 3.33, just a lil heads up...

It's 3,66 counting extra shots and jam chance. Average of two averages .

Edited by Alexander Malthus, 22 January 2014 - 01:02 PM.

[Just wanna play]

Ghogiel, on 22 January 2014 - 01:00 PM, said:

The UAC can double shot with a chance to jam. Which is why it's average DPS is higher.

ohh hes talking about that dps, not a vary reliable;e number though....