Reverse Engineering Flaws

[andrewcn]

If there is a 20% chance that I will get a schematic for Reverse Engineering an item, why haven't I received a single schematic after I RE 17 of the same item?!?!

 

BioWare, you are not doing a very good job at making your customers happy. You are good, however, at angering people and getting them to leave this game and move on to obviously better ones

Edited May 15, 2012 by andrewcn
Title Change

[JediXMan]

I agree completely. 20% = 1 in 5. I shouldn't need more than 5.

 

Also.... LET BIOCHEMISTS RE! Come on. It's not fair to us. Armortechs and Synths can RE looted gear, but I can't RE looted medpacs and stims?

[jgoldsack]

20% chance per item.

 

It isn't cumulative. RE a stack of 5 does not give you a 100% chance... it is still 20% per item.

[tidmass]

If there is a 20% chance that I will get a schematic for Reverse Engineering an item, why haven't I received a single schematic after I RE 17 of the same item?!?!

 

BioWare, you are not doing a very good job at making your customers happy. You are good, however, at angering people and getting them to leave this game and move on to obviously better ones

 

do you understand what a 20% chance means?

 

1 shot in 5 may give you a schematic.

 

thats 1 shot in 5 on each..... not 1 shot in 5 over all 17.

 

you should have paid more attention in your math classes.

[Rekkx]

Calm down.

 

"Chance" does not work like that. You don't only need 5 and you should have it. You have a 20% chance EACH TIME you RE something.

 

The next one you RE does not have a memory of the previous one. Each RE has the same (20%) chance of giving you a schematic. 17 tries is a lot but not outside the realm of chance. It's just a long run. It's statistically likely as not.

 

Don't get mad, just keep on going and the next time you RE 3 in a row for a new schematic blue/purple remember just how fickle the "Lords of Chance" can be because that's almost as likely a run of the 17.

[Criosdh]

Not this arguement again. Seen the same thing in LotRO.

 

Sorry but 20% does mean 100% in 5. Hypothetically. Because if you have a 1 in 5 odds of something then doing it 5 times means you're theoretically gonna succeed. Sure, sometimes it may take 7 or 8 tries, and others you may succeed on 1 or 2 tries only. But in the end, if you do something 100 times with a 20% success rate then you should have 20 at the end.

 

The point is ridiculous sample sizes should not be needed to show the correct % success as a result.

 

Right now higher tier items are most certainly not reverse engineering at 20% no matter how you look at it. It's just plain obvious when comparing it to reverse engineering lower tier stuff.

[Innocente]

Guys, you ALL need to rethink statistical probability theory here. Yes, each RE has a 20% chance of succes. But, if it were an 'honest' 20% chance, then as you do REs, the odds of failure should go down as you progress through the rolls. Think of it as flipping a coin 20 times. Each flip is 50/50 for one side or the other. But, what are the odds of getting 20 Heads, or 20 Tails in a row? Very slim. So, to have a 20% chance of succes, and get 17 failures in a row is not a statistically probable event. And I can say I have experienced this sort of anomaly many, many times doing REs. There IS something wrong with the probablity generator they are using. In addition, I would point you all back to the 'bug' that was recently 'fixed' regarding the conversation win frequency problem where the first person to win a conversation roll would continue to win them.

[Sammm]

I always thought that the RE attempts were actually independent events. In other words, your Xth attempt result doesn't influence at all your Yth attempt one. They're supposed to be independent, right?

 

That being said, I bet that over 1000 RE attempts, you'll probably get around 200 successful REs.

 

P.S. After 1.2., I started to count my RE successes and so far it's at 18/87 ~ 20,69%. That's pretty close to 20%, don't you think! I remember that once I had 3 consecutive successes!

Edited May 15, 2012 by Sammm

[andrewcn]

I completely understand what 20% means. However, wouldn't it make just a tiny bit of sense to have received 1 schematic after REing 17 times.

 

Also, it would be much easier to stay calm if BW didn't continue to have so many gaming issues after 6 months of release. I don't expect a perfect game, that would be foolish to ask, but I do expect an improvement of some sort after 6 months. Why pay for a game that never improves?

 

At least there is D3 to try and GW2 comes soon

[siegeshot]

The RE's %'s need to have memory. I have a thread on it here.

 

http://www.swtor.com/community/showthread.php?t=454473

[Sammm]

I completely understand what 20% means. However, wouldn't it make just a tiny bit of sense to have received 1 schematic after REing 17 times.

 

Also, it would be much easier to stay calm if BW didn't continue to have so many gaming issues after 6 months of release. I don't expect a perfect game, that would be foolish to ask, but I do expect an improvement of some sort after 6 months. Why pay for a game that never improves?

 

At least there is D3 to try and GW2 comes soon

 

"... that never improves?"

 

*Master Oteg Facepalm*

[jgelling]

With all the bugs in this game, I don't have any confidence the 20% random number generator doesn't bug out on various items, either. That said, on the (generous) assumption it's working properly, there's still a better than 2% chance you could reverse engineer something 17 times and end up with nothing but sweet, delicious QQ tears.

 

Really the devs should cheat and game the system in the players' favor if they're getting particularly unlucky - keep the customers happy. Even casinos throw their whales a bone every now and again!

Edited May 15, 2012 by jgelling

[rjkishida]

Dead topic. Countless threads have already been posted about this. Bioware doesn't want to fix it. Cue: fanbois claiming that 20% doesn't mean you will succeed. Waste of time. BTDT

[whatthefett]

Guys, you ALL need to rethink statistical probability theory here. Yes, each RE has a 20% chance of succes. But, if it were an 'honest' 20% chance, then as you do REs, the odds of failure should go down as you progress through the rolls. Think of it as flipping a coin 20 times. Each flip is 50/50 for one side or the other. But, what are the odds of getting 20 Heads, or 20 Tails in a row? Very slim. So, to have a 20% chance of succes, and get 17 failures in a row is not a statistically probable event. And I can say I have experienced this sort of anomaly many, many times doing REs. There IS something wrong with the probablity generator they are using. In addition, I would point you all back to the 'bug' that was recently 'fixed' regarding the conversation win frequency problem where the first person to win a conversation roll would continue to win them.

 

If I need statistical probability theory to RE, then it's broke and dumb. Really?

[Unusualsuspect]

Guys, you ALL need to rethink statistical probability theory here. Yes, each RE has a 20% chance of succes. But, if it were an 'honest' 20% chance, then as you do REs, the odds of failure should go down as you progress through the rolls. Think of it as flipping a coin 20 times. Each flip is 50/50 for one side or the other. But, what are the odds of getting 20 Heads, or 20 Tails in a row? Very slim. So, to have a 20% chance of succes, and get 17 failures in a row is not a statistically probable event. And I can say I have experienced this sort of anomaly many, many times doing REs. There IS something wrong with the probablity generator they are using. In addition, I would point you all back to the 'bug' that was recently 'fixed' regarding the conversation win frequency problem where the first person to win a conversation roll would continue to win them.

 

20 heads in a row is an extremely unlikely statistical event: roughly .000095%, or so.

 

The likelihood of getting 17 failures in a row when there is an 80% fail rate rather than 50%, however, is a far less unlikely 2.25% or so. That is to say, around 1 in 45 items you try to RE will give you 17 failures in a row (with, obviously, only a 20% chance you'll succeed on the 18th, etc).

 

Thinking of it alternatively, roughly 1 in 45 people who attempt to RE a given item will have at least 17 failures.

 

Statistics are most valid across broad ranges, and one aspect of chance is that any PARTICULAR situation can happen in ways that most certainly do NOT fit the average.

 

Think about how many people play this game. Think about how many of those people craft. Think about how many items are being RE'd to a higher tier, sometimes multiple times.

 

Now, add all these together... and you'll find that a few (damnably) lucky people almost always RE's within a few attempts (There's roughly a 50% chance of 3 failures in a row, by and by), while a few (pitifully) unlucky people will fail 10, 20, even 30 or 40 times in a row. Those would be the outliers, the small percentages on either side of a Bell graph. The vast majority of us? We'll have a mix of luck, from REing in 1 attempt to failing after 40 REs, with most people successfully REing in a reasonable number of attempts.

 

That's the theoretical statistical side. It does require the 20% number to be accurate, and I'd wonder if even Bioware has numbers to confirm that their RNG is producing an accurate 20%. As is almost always the case in programming, the RNG is not completely random, but is a formula based on a seed number that SHOULD (but is not always) random that comes from outside sources. I'm not intimately familiar with RNG programming, but from what I've heard, something like the time of day on the computer (or some cycle of number of seconds, milliseconds, whatever) can be used.

 

I'm not saying Bioware's RNG is for sure not to blame for streaky RE issues, but I CAN say with a fair amount of certainty that even an extremely unlucky person does not necessitate that Bioware's programming is at fault. With as many people playing this game (many with multiple characters), a few unlucky folk are almost certainly going to fall in the "needs at least 10 attempts to RE anything)" category while a few lucky bastards will find 1-2 attempts REs practically anything.

 

Most of us are in between, and as time goes on, most of the exceedingly fortunate AND unfortunate will probably regress towards the mean. Such is life.

Edited May 15, 2012 by Unusualsuspect

[GnatB]

Dead topic. Countless threads have already been posted about this. Bioware doesn't want to fix it. Cue: fanbois claiming that 20% doesn't mean you will succeed. Waste of time. BTDT

 

Actually they've ALREADY said they're looking into doing something to cut short the long failure streaks. Which is the other hand of why all these threads are amusing.

[Syadasti]

BW promise a lot in he last weeks (just saying).

 

For the topic i wonder when some people will understand, that this is not an statistical problem, it is simply not FUN to RE 17 items without procc-lerning. RNG was/is to overused by WoW, so very little people still enjoy there skinnerboxes, that´s the reason this topic shows up again and again.

[Blackardin]

If there is a 20% chance that I will get a schematic for Reverse Engineering an item, why haven't I received a single schematic after I RE 17 of the same item?!?!

 

BioWare, you are not doing a very good job at making your customers happy. You are good, however, at angering people and getting them to leave this game and move on to obviously better ones

 

RE should be a minimum of 50% success for anything under level 30, and the mats and missions need to be rebalanced so as to allow one to craft items at the proper level when leveling. I've tested and tested this situation. Most recently I hit level 18 before my character could RE a level 13 pattern.

 

Between the difficulty obtaining mats in a reasonable amount of time, having to jump in and out of your ship to get decent enough missions to pop, returning the less needed mat more often, and the fail rate, it has to be discouraging for new players. Heck, players will hit their first wall as early as level 1 mat farming.

[Blackardin]

BW promise a lot in he last weeks (just saying).

 

For the topic i wonder when some people will understand, that this is not an statistical problem, it is simply not FUN to RE 17 items without procc-lerning. RNG was/is to overused by WoW, so very little people still enjoy there skinnerboxes, that´s the reason this topic shows up again and again.

 

In a nut shell.

 

When one develops an aspect of the game, the first question should always be, "will the players find this fun", not where or not to make it challenging or difficult. There are times when "challenging" is fun, but there is never a time when tedium will be and right now its just tedious.

[Blackardin]

Actually they've ALREADY said they're looking into doing something to cut short the long failure streaks. Which is the other hand of why all these threads are amusing.

 

I suggested a long time ago that a scaling progression was needed. If your first attempt is set at 20%, the second should be at 25%, then 30% ect. Something along those lines would have been good.

[Blackardin]

20 heads in a row is an extremely unlikely statistical event: roughly .000095%, or so.

 

The likelihood of getting 17 failures in a row when there is an 80% fail rate rather than 50%, however, is a far less unlikely 2.25% or so. That is to say, around 1 in 45 items you try to RE will give you 17 failures in a row (with, obviously, only a 20% chance you'll succeed on the 18th, etc).

 

Thinking of it alternatively, roughly 1 in 45 people who attempt to RE a given item will have at least 17 failures.

 

Statistics are most valid across broad ranges, and one aspect of chance is that any PARTICULAR situation can happen in ways that most certainly do NOT fit the average.

 

Think about how many people play this game. Think about how many of those people craft. Think about how many items are being RE'd to a higher tier, sometimes multiple times.

 

Now, add all these together... and you'll find that a few (damnably) lucky people almost always RE's within a few attempts (There's roughly a 50% chance of 3 failures in a row, by and by), while a few (pitifully) unlucky people will fail 10, 20, even 30 or 40 times in a row. Those would be the outliers, the small percentages on either side of a Bell graph. The vast majority of us? We'll have a mix of luck, from REing in 1 attempt to failing after 40 REs, with most people successfully REing in a reasonable number of attempts.

 

That's the theoretical statistical side. It does require the 20% number to be accurate, and I'd wonder if even Bioware has numbers to confirm that their RNG is producing an accurate 20%. As is almost always the case in programming, the RNG is not completely random, but is a formula based on a seed number that SHOULD (but is not always) random that comes from outside sources. I'm not intimately familiar with RNG programming, but from what I've heard, something like the time of day on the computer (or some cycle of number of seconds, milliseconds, whatever) can be used.

 

I'm not saying Bioware's RNG is for sure not to blame for streaky RE issues, but I CAN say with a fair amount of certainty that even an extremely unlucky person does not necessitate that Bioware's programming is at fault. With as many people playing this game (many with multiple characters), a few unlucky folk are almost certainly going to fall in the "needs at least 10 attempts to RE anything)" category while a few lucky bastards will find 1-2 attempts REs practically anything.

 

Most of us are in between, and as time goes on, most of the exceedingly fortunate AND unfortunate will probably regress towards the mean. Such is life.

 

Yanno what? I appreciate the statistical analysis, but I don't give a rats arse. ;p I'm logging on to play a game for fun and relaxation. I don't want to delve in any deeper then pushing a button and, within reason for the level, getting the item I'm attempting to learn or craft so that I may continue playing and enjoying the game. Ats not happening right now and by the time I finish, quite often, I just log out because the frustration has built to a level where I'm not enjoying myself any longer.

Edited May 16, 2012 by Blackardin

[Unusualsuspect]

So noted. I was mostly responding the statistical aspect of that quote, which was sending all sorts of misinformation out there.

 

Personally, I would much rather see a progessively better chance to RE as failures incur, as stated by a previous poster. It fulfills all the requirements of a variable ratio rewards system (The most efficacious form of rewards to encourage more participation) while still providing a practically guaranteed success eventually to ward off the most frustrating of RNG streaks.

[Joonbeams]

20 heads in a row is an extremely unlikely statistical event: roughly .000095%, or so.

 

The likelihood of getting 17 failures in a row when there is an 80% fail rate rather than 50%, however, is a far less unlikely 2.25% or so. That is to say, around 1 in 45 items you try to RE will give you 17 failures in a row (with, obviously, only a 20% chance you'll succeed on the 18th, etc).

 

Thinking of it alternatively, roughly 1 in 45 people who attempt to RE a given item will have at least 17 failures.

 

Statistics are most valid across broad ranges, and one aspect of chance is that any PARTICULAR situation can happen in ways that most certainly do NOT fit the average.

 

Think about how many people play this game. Think about how many of those people craft. Think about how many items are being RE'd to a higher tier, sometimes multiple times.

 

Now, add all these together... and you'll find that a few (damnably) lucky people almost always RE's within a few attempts (There's roughly a 50% chance of 3 failures in a row, by and by), while a few (pitifully) unlucky people will fail 10, 20, even 30 or 40 times in a row. Those would be the outliers, the small percentages on either side of a Bell graph. The vast majority of us? We'll have a mix of luck, from REing in 1 attempt to failing after 40 REs, with most people successfully REing in a reasonable number of attempts.

 

That's the theoretical statistical side. It does require the 20% number to be accurate, and I'd wonder if even Bioware has numbers to confirm that their RNG is producing an accurate 20%. As is almost always the case in programming, the RNG is not completely random, but is a formula based on a seed number that SHOULD (but is not always) random that comes from outside sources. I'm not intimately familiar with RNG programming, but from what I've heard, something like the time of day on the computer (or some cycle of number of seconds, milliseconds, whatever) can be used.

 

I'm not saying Bioware's RNG is for sure not to blame for streaky RE issues, but I CAN say with a fair amount of certainty that even an extremely unlucky person does not necessitate that Bioware's programming is at fault. With as many people playing this game (many with multiple characters), a few unlucky folk are almost certainly going to fall in the "needs at least 10 attempts to RE anything)" category while a few lucky bastards will find 1-2 attempts REs practically anything.

 

Most of us are in between, and as time goes on, most of the exceedingly fortunate AND unfortunate will probably regress towards the mean. Such is life.

 

Great post. Couldn't have said it any better.

 

Relatedly, I think RE'ing is just about right (even if the percentage chances are a bit off) - perhaps even high. The rarity of purple materials coupled with the relatively difficulty in RE'ing for purple schematics, helps keep purple items rare and valuable. This helps crafters - both profiteerers and manufacturers - which helps keep the economy thriving. This is because RE failures help burn through a lot of blue/green materials, which helps keep the demand for those mats up. Any increase in RE success rate will absolutely hurt crafters and ultimately the economy. I know this isn't what people want to hear (and I've had my frustrations too), but ultimately, this helps all of us.

[JediXMan]

20% chance per item.

 

It isn't cumulative. RE a stack of 5 does not give you a 100% chance... it is still 20% per item.

 

Ah.

 

Still quite annoying.

[Blackardin]

So noted. I was mostly responding the statistical aspect of that quote, which was sending all sorts of misinformation out there.

 

Personally, I would much rather see a progessively better chance to RE as failures incur, as stated by a previous poster. It fulfills all the requirements of a variable ratio rewards system (The most efficacious form of rewards to encourage more participation) while still providing a practically guaranteed success eventually to ward off the most frustrating of RNG streaks.

 

Yeah, don't get me wrong. its good information, and I fully appreciate and understand the issue with misinformation. I'm just burnt out on having it be such a distraction, taking me out of the game so often. Just want to log in an have fun without trying to work around so many issues.

 

I enjoy the game, but this issue is like someone constantly poking you in the side with a stick.

 

Hope it came across that way. ;p

Edited May 16, 2012 by Blackardin