44 replies to this topic

[Sandpit]

yea RNG on a pc doesn't roll above or below average. It rolls randomly. Due to its nature a 2 has the exact same chance and probability of a 6. You run into issues in just about any game where the RNG is the main mechanic for this reason. It used to drive people nuts in megamek because that was a direct port of the TT game which required die rolls. The RNG would make some very bad runs in games but long-term it would average out.
It didn't prevent player frustration though when they lost 10 rounds of initiative or rolled TACs, 2s, etc. repeatedly for an extended streak in one or two or even three games

[Mister Blastman]

Canniballistic, on 07 January 2014 - 12:47 AM, said:

Why are we still using this rediculous random chance BS instead of the heat system that was used in MW4.

The work is literally already done and still someone thinks its a good idea to have a weapon be completely random sometimes shooting 3 bullets or a chain of over 10 before jamming and other times (more often the case) you fire once and it jams!

Guns broken... been broken this whole time... still not fixed...

Use the system MW4 used already.

Personally I like MW:LL's system the best. Random number generators have no place in Mechwarrior.

[MeiSooHaityu]

Just playing devil's advocate here. Sarna describes the UA/C5 as prone to jamming.

"Introduced in 2640 by the Terran Hegemony^[4]. The Ultra Autocannon/5 is an upgraded version of the standard Autocannon/5. Utilizing a different loading mechanism, the Ultra Autocannons are capable of maintaining a substantially increased rate of fire over traditional or LB-X autocannons at the cost of higher heat and the risk of jamming." - Sarna.net

Guns don't jam on a regular interval. They can jam on the 5th shot or as soon as you pull the trigger once. I interpret the higher heat as a cost of higher fire rate and not as some sort of ghost heat artificial limitation.

The way the system currently works seems to make sense in that regard. To me that seems to fit with the risk/reward attributes of the gun. Now, maybe the ratio of risk/reward needs to be shifted a bit towards the reward end (a smaller jam chance), but really the gun seems to function as it should.

Besides, be happy we no longer have to manually unjam the gun by assigning and un-assigning the gun to each of the 6 fire groups like previously.

Edited by MeiSooHaityu, 07 January 2014 - 09:17 AM.

[Mister Blastman]

MeiSooHaityu, on 07 January 2014 - 09:14 AM, said:

Just playing devil's advocate here. Sarna describes the UA/C5 as prone to jamming and not overheating or running hot (like MW4). Guns don't jam on a regular interval. They can jam on the 5th shot or as soon as you pull the trigger once.

The way the system currently works seems to make sense in that regard. To me that seems to fit with the risk/reward attributes of the gun. Now, maybe the ratio of risk/reward needs to be shifted a bit towards the reward end (a smaller jam chance), but really the gun seems to function as it should.

Besides, be happy we no longer have to manually unjam the gun by assigning and un-assigning the gun to each of the 6 fire groups like previously.

How often have you shot real firearms?

I've shot them plenty and they rarely jam. In fact, the main times I've seen them jam are due to:

1. Operator error.
2. Poor cleaning habits.
3. Way too hot.

Pulling the trigger one time and having it jam is not realistic. The gun isn't hot yet. The Battlemechs are well cared for and the mechanical loading system won't make mistakes.

Even the AH-64 Apache has chain-fed ammunition which will yank an unspent, jammed round out of the chamber to slot a new one in.

Heat meter gives us the best representation of knowing when the weapon is likely to jam.

[Sandpit]

MeiSooHaityu, on 07 January 2014 - 09:14 AM, said:

Just playing devil's advocate here. Sarna describes the UA/C5 as prone to jamming.

"Introduced in 2640 by the Terran Hegemony^[4]. The Ultra Autocannon/5 is an upgraded version of the standard Autocannon/5. Utilizing a different loading mechanism, the Ultra Autocannons are capable of maintaining a substantially increased rate of fire over traditional or LB-X autocannons at the cost of higher heat and the risk of jamming." - Sarna.net

Guns don't jam on a regular interval. They can jam on the 5th shot or as soon as you pull the trigger once. I interpret the higher heat as a cost of higher fire rate and not as some sort of ghost heat artificial limitation.

The way the system currently works seems to make sense in that regard. To me that seems to fit with the risk/reward attributes of the gun. Now, maybe the ratio of risk/reward needs to be shifted a bit towards the reward end (a smaller jam chance), but really the gun seems to function as it should.

Besides, be happy we no longer have to manually unjam the gun by assigning and un-assigning the gun to each of the 6 fire groups like previously.

Not arguing that it shouldn't jam first shot sometimes, or 3rd, 10th, etc. just that RNG is a bad mechanic to use as the ONLY mechanic

[Almond Brown]

Corbon Zackery, on 07 January 2014 - 08:02 AM, said:

stuff

The Jam percentage is to high for computer based games since a computer will always roll above avg.

Boy oh boy, that sounds pretty Tin Foil Hat based. A RNG is an RNG not matter who rolls the dice right? That is why it is called an "Random Number Generator" right?

[Trauglodyte]

Mister Blastman, on 07 January 2014 - 09:17 AM, said:

How often have you shot real firearms? I've shot them plenty and they rarely jam. In fact, the main times I've seen them jam are due to: 1. Operator error. 2. Poor cleaning habits. 3. Way too hot. Pulling the trigger one time and having it jam is not realistic. The gun isn't hot yet. The Battlemechs are well cared for and the mechanical loading system won't make mistakes. Even the AH-64 Apache has chain-fed ammunition which will yank an unspent, jammed round out of the chamber to slot a new one in. Heat meter gives us the best representation of knowing when the weapon is likely to jam.

I had my M4 jam a few times during basic when we were out on the range practicing for quals. Ah the good ol' SPORTS came in handy but it was annoying. But, your point is valid.

Anyway, RNG on the weapon is annoying and would much rather it be a building chance to jam based on number of weapons used and number of rounds fired. That solves the issue with the singular UAC even when used outside of the double fire rate. Which, btw, is a ******!

[MeiSooHaityu]

Mister Blastman, on 07 January 2014 - 09:17 AM, said:

How often have you shot real firearms?

I've shot them plenty and they rarely jam. In fact, the main times I've seen them jam are due to:

1. Operator error.
2. Poor cleaning habits.
3. Way too hot.

Pulling the trigger one time and having it jam is not realistic. The gun isn't hot yet. The Battlemechs are well cared for and the mechanical loading system won't make mistakes.

Even the AH-64 Apache has chain-fed ammunition which will yank an unspent, jammed round out of the chamber to slot a new one in.

Heat meter gives us the best representation of knowing when the weapon is likely to jam.

I get what you are saying. You would think by 3050 they could make a quick firing reliable ballistic cannon.

Think of it this way, the IS is practically a 19th Century era man (IS) trying to reverse engineer a modern 21 century ballistic firing system (Star League). It is a bit shoddy at best

The only heat scale the game has is mech heat and not weapon heat. That is unless they could do something similar to the Gauss charge bar....hmmm. I suppose that wouldn't be too bad. As the weapon fires, the weapon bar fills, jam occurence get very high when the bar fills....Maybe

Edited by MeiSooHaityu, 07 January 2014 - 10:57 AM.

[wanderer]

Funny thing is, the UAC's "double-tap" feature wasn't a deliberate part of it's original design. Basically, some wise-guy test pilot figured a way to glitch the normal loading cycle into delivering an extra cassette of ammo into the feed before it was supposedly seated properly, letting him snap off an extra round of fire in half the time. The Ultra was actually designed to just give you better range and accuracy at a slight increase in tonnage, but the "glitch" became a "feature".

It wasn't ever designed for it, story-wise. UAC's on ultra mode are overheating and running the feed faster than the gun's designed to fire them safely- basically, you're overloading the weapon at anything better than single-fire rate. Real "rapid-fire" autocannons don't come along until the later rotary-barrel designs, which also have better jam-clearing mechanisms as well.

[MeiSooHaityu]

wanderer, on 07 January 2014 - 12:12 PM, said:

Funny thing is, the UAC's "double-tap" feature wasn't a deliberate part of it's original design. Basically, some wise-guy test pilot figured a way to glitch the normal loading cycle into delivering an extra cassette of ammo into the feed before it was supposedly seated properly, letting him snap off an extra round of fire in half the time. The Ultra was actually designed to just give you better range and accuracy at a slight increase in tonnage, but the "glitch" became a "feature".

It wasn't ever designed for it, story-wise. UAC's on ultra mode are overheating and running the feed faster than the gun's designed to fire them safely- basically, you're overloading the weapon at anything better than single-fire rate. Real "rapid-fire" autocannons don't come along until the later rotary-barrel designs, which also have better jam-clearing mechanisms as well.

Well, there ya go. A good cannon (I assume without reference) reason that they jam.

[ReXspec]

Varent, on 07 January 2014 - 12:48 AM, said:

love the UAC. Not sure why so many still feel the need to hate on it.

Because the jam chance is outrageous.

[Strum Wealh]

Sandpit, on 07 January 2014 - 08:57 AM, said:

yea RNG on a pc doesn't roll above or below average. It rolls randomly. Due to its nature a 2 has the exact same chance and probability of a 6. You run into issues in just about any game where the RNG is the main mechanic for this reason. It used to drive people nuts in megamek because that was a direct port of the TT game which required die rolls. The RNG would make some very bad runs in games but long-term it would average out.
It didn't prevent player frustration though when they lost 10 rounds of initiative or rolled TACs, 2s, etc. repeatedly for an extended streak in one or two or even three games

In the 2D6 system BattleTech uses, the probability of a result of 2 (1 in 36, or 2.78%) is NOT the same as the probability of a result of 6 (5 in 36, or 13.89%).
The assertion that "a 2 has the exact same chance and probability of a 6" is wholly incorrect.

In BT, the UACs (all of them) had a 0% chance of jamming when fired in "normal mode", and a 2.78% chance of jamming when fired in "Ultra mode", BUT they did not have the ability to unjam at all during a battle (that is, the weapon was rendered wholly non-functional until a technician could repair it).

In MWO, however, the UAC/5 currently has a 20% chance of jamming (in 2D6 terms, roughly equivalent to "result of 4 or lower" (which is actually 16.67%; "result of 5 or lower" is equivalent to a 27.78% likelihood)) if "double-tapped" AND can unjam itself during a match (which, in BT, was an ability reserved for the RACs).
(For further comparison, the likelihood of a RAC jamming in BattleTech (per pg. 114 of Total Warfare) is as follows: 0% in single-shot mode, 2.78% (roll result of 2) in double-shot mode, 8.33% (roll result of 3 or below) in quad-shot mode, and 16.67% (roll result of 4 or below) in six-shot mode; the current MWO UAC implementation actually jams more often than a max-salvo BT RAC! )

Personally, I would have preferred a UAC implementation that included a player-controlled toggle for "normal mode" vs "Ultra mode" (so that microbreaks/lag/packet errors/etc in one's mouse/joystick/whatever would not result in unintended double-fire commands being sent), along with setting the jam chance to 2.78% (what it would be in BattleTech) AND making it a perma-jam (to be repaired by spending c-bills while in the MechLab, like (or as part of) a repair cost - as it would be in BattleTech).

[Sandpit]

Strum Wealh, on 08 January 2014 - 06:10 PM, said:

In the 2D6 system BattleTech uses, the probability of a result of 2 (1 in 36, or 2.78%) is NOT the same as the probability of a result of 6 (5 in 36, or 13.89%).
The assertion that "a 2 has the exact same chance and probability of a 6" is wholly incorrect.

Uhm I think you misread or misunderstood my post. I said exactly what you're saying. What DOES have the exact same chance and probability of rolling a 2 or 6 is an RNG. That's the difference.
An RNG is truly random with same probabilities for each number.
Dice work differently. You have higher probabilities for some numbers than you do for others.

Maybe I worded it badly? I dunno......

The result probability isn't the same as the number generation probability

With an RNG you have a 1 in 6 chance to roll each number equally. There's no distinction
When you throw a 2d6 system into an RNG there is no better or worse chance for it to roll a 2, 6, or 12. That's what I was getting at. Physical dice have a completely different probability which is why I stated an RNG as the sole means for this kind of mechanic is bad.

[Noesis]

Sandpit, on 08 January 2014 - 06:19 PM, said:

Uhm I think you misread or misunderstood my post. I said exactly what you're saying. What DOES have the exact same chance and probability of rolling a 2 or 6 is an RNG. That's the difference.
An RNG is truly random with same probabilities for each number.
Dice work differently. You have higher probabilities for some numbers than you do for others.

Maybe I worded it badly? I dunno......

The result probability isn't the same as the number generation probability

With an RNG you have a 1 in 6 chance to roll each number equally. There's no distinction
When you throw a 2d6 system into an RNG there is no better or worse chance for it to roll a 2, 6, or 12. That's what I was getting at. Physical dice have a completely different probability which is why I stated an RNG as the sole means for this kind of mechanic is bad.

Might help

[Sandpit]

Noesis, on 08 January 2014 - 06:28 PM, said:

Might help

and............?
An RNG by its nature does not work on a bell curve of averages nor does it adhere to statistical probabilities

A computerized RNG has an equal statistical chance to roll 5 2's in a row as it does to roll 5 7's
I don't know how else to explain the nature of that. I've stated twice (at least) an RNG long-term will average out statistically. In the short-term that's not always the case.

This is exactly why you get players complaining of "I got xx map 12 times in a row!"
That short-term example shows an RNG performing exactly as I'm describing

It's also exactly why long-term you see the same players in the above example say "Well now that I look at map stats everything is pretty even with maybe an anomaly or two here and there"
That long-term example shows an RNG performing exactly as I'm describing

I know statistics. I know bell curves, median, mean, mode, averages, etc.

RNG's do not operate in the same manner or with the same probabilities as 2 physical 6 sided dice.

[Sandpit]

I don't know how else to explain it.

For this particular subject the short-term aspect of the examples I ahve given are a bad balancing mechanic to be used as the sole means.
It means that short-term you do get players that experience a UAC5 jam every time they pull the trigger. If we had prolonged games where the RNG had an opportunity to even out it wouldn't be nearly as big of a deal but because we are working on individual matches each time the short-term RNG examples I gave can wreak havoc in balancing a mechanic like this.

By long-term I mean hundreds of "rolls" to even out. With the RNG as the sole balance mechanic that means it is feasible and probable that a player could go through dozens of matches where the UAC jams on the first shot every single time. That's why I say it's a bad mechanic to use as the sole means of balance

[Noesis]

Sandpit, on 08 January 2014 - 06:34 PM, said:

and............?
An RNG by its nature does not work on a bell curve of averages nor does it adhere to statistical probabilities

A computerized RNG has an equal statistical chance to roll 5 2's in a row as it does to roll 5 7's
I don't know how else to explain the nature of that. I've stated twice (at least) an RNG long-term will average out statistically. In the short-term that's not always the case.

This is exactly why you get players complaining of "I got xx map 12 times in a row!"
That short-term example shows an RNG performing exactly as I'm describing

It's also exactly why long-term you see the same players in the above example say "Well now that I look at map stats everything is pretty even with maybe an anomaly or two here and there"
That long-term example shows an RNG performing exactly as I'm describing

I know statistics. I know bell curves, median, mean, mode, averages, etc.

RNG's do not operate in the same manner or with the same probabilities as 2 physical 6 sided dice.

I know RNGs also and you will get a good bell curve with its use over a good sample than just a flat line.

[Varent]

I still dont understand how people are complaining about this weapon. Its a good weapon.

[Sandpit]

Noesis, on 08 January 2014 - 06:46 PM, said:

I know RNGs also and you will get a good bell curve with its use over a good sample than just a flat line.

Agreed and understood. That has nothing to do with long-term versus short-term.

The example and mechanic you're talking about is the long-term. It's valid, true, and factual

What you're ignoring are the short-term affects I'm discussing

@Varent the weapon itself is great. The only issue I've ever had with it is using an RNG as the sole balancing mechanic. It leads to discrepancies in the short-term which is every match.

Now to compound that you can't just include a single player's experience. Long-term you get the curve and average Noesis is talking about. However, due to the sheer volume of use and the number of players who use it you still run into situations where some players may rarely run into an issue or repetitive jamming issues. In order for the averages to hold true you will also have players that experience the exact opposite and having it jam consistently on a regular basis and far more often than "average"

If they would throw in a slider type mechanic along side the RNG then you wouldn't get individual anomalies to the average as often.

Obviously most of this has to be based on theory crafting because none of us have the actual numbers but I can understand how one player will love this weapon and another will hate it due to the RNG mechanic.

[Strum Wealh]

Sandpit, on 08 January 2014 - 06:34 PM, said:

and............?
An RNG by its nature does not work on a bell curve of averages nor does it adhere to statistical probabilities

A computerized RNG has an equal statistical chance to roll 5 2's in a row as it does to roll 5 7's
I don't know how else to explain the nature of that. I've stated twice (at least) an RNG long-term will average out statistically. In the short-term that's not always the case.

This is exactly why you get players complaining of "I got xx map 12 times in a row!"
That short-term example shows an RNG performing exactly as I'm describing

It's also exactly why long-term you see the same players in the above example say "Well now that I look at map stats everything is pretty even with maybe an anomaly or two here and there"
That long-term example shows an RNG performing exactly as I'm describing

I know statistics. I know bell curves, median, mean, mode, averages, etc.

RNG's do not operate in the same manner or with the same probabilities as 2 physical 6 sided dice.

Though, that depends on how the system is programmed - "virtual dice" systems are hardly an unheard-of or undocumented employment of a RNG (see here, or here, or here), and are but one of a number of methods of creating the desired result (e.g. simulating the difference in likelihood between a 2D6 result of 2 and a 2D6 result of (any other possible 2D6 result)) that would be available to PGI.

The point being, a "proper" (that is, more true-to-BT) implementation of the UAC (as described in the closing paragraph of my previous post) could have been done with relative ease (or most of it could even be jury-rigged into the current system, since we know that PGI can alter the jam percentages and set the unjam timer to something exceeding the match time limit), instead of the "RAC in all but name" we have currently.