Reverse engineering 20% my foot

[Jeadeye]

One, i think you may be exagerating a little bit and Two, you are RE'ing BLUE level items to get the Purple level items right??? Cause if you are truely at 0% I could easily see it because you are RE'ing green level items thinking you are going to get a Purple....

 

No I am not exagerating ..i've been attempting to get a purple shield recipe for approx lvl 40 -from a blue item and i do understand how the system is supposed to work- ...

I have gone through at least 2 full stacks of the req. materials, bought more from GTM and had companions continually doing missions for more. It does say 10% RE chance for new recipe but it FEELS LIKE 0% as i have spent hours upon hours only to NOT get any new recipe.

 

I do admit RE on some items seems to be quite fair , you make 10-20 and most of the time get a better recipe version , but it seems that is not the case with shields as to get a new recipe it sometimes takes several days before

any new recipe is awarded.

I think the problem that developers overlook continually ( in a lot of games) is the % factor.

To give a % chance on something means that you do have a chance to NEVER get what you are aiming for , however unlikely this maybe it is possible.

[Samoth_Nomad]

and there you have it. 10% chance. like rolling a 1 on a 10 side die. Sure, you might hit, but you might have long runs of nothing. I encountered this preforming RE on offhands. Took many tries.

[Sorwen]

Basically it comes down to this if you can never get it then it doesn't matter if you had a chance or not.

[Teykos]

No I am not exagerating ..i've been attempting to get a purple shield recipe for approx lvl 40 -from a blue item and i do understand how the system is supposed to work- ...

I have gone through at least 2 full stacks of the req. materials, bought more from GTM and had companions continually doing missions for more. It does say 10% RE chance for new recipe but it FEELS LIKE 0% as i have spent hours upon hours only to NOT get any new recipe.

 

I do admit RE on some items seems to be quite fair , you make 10-20 and most of the time get a better recipe version , but it seems that is not the case with shields as to get a new recipe it sometimes takes several days before

any new recipe is awarded.

I think the problem that developers overlook continually ( in a lot of games) is the % factor.

To give a % chance on something means that you do have a chance to NEVER get what you are aiming for , however unlikely this maybe it is possible.

 

Actually the system's been improved over what was before.

 

That chance of you learning a schematic also used to apply to schematics you already knew. Pissed people off constantly.

[Endymionsama]

Actually the system's been improved over what was before.

 

That chance of you learning a schematic also used to apply to schematics you already knew. Pissed people off constantly.

 

How do you know they are just not surpressing the message instead of saying you already know the schematic. For all we know thats what they did.

[VanorDM]

Basically it comes down to this if you can never get it then it doesn't matter if you had a chance or not.

 

No. There's a difference between 0% and 20% chance.

 

Just to state it again, since so many people here don't bother reading most of the posts.

 

Having a 20% chance does not mean you will get 1 success for every 5 attempts, that's now how the math works.

[Zhiroc]

How do you know they are just not surpressing the message instead of saying you already know the schematic. For all we know thats what they did.

 

By keeping an accurate record of attempts and successes (and remember to count the 10-vs-20% separately). With enough rolls, that should approach the stated success rate. It does for me (actually, better, around 21% for the 20% items after about 150 rolls so far).

[GnatB]

Well, like I've said:

 

On an item with 10% chance of success, Each and every time you RE an item you have:

1. a 10% chance of successfully learning a schematic.

2. a 9.85% chance of this being the first failure in a string of at least 22 more failures.

 

Don't you just hate static x% probabilities?

[dufox]

the title made me laugh

 

this has already been said but 20% does not mean 1 for every 5 you make.

but i do love how u can reverse engineer a low level item and get the schematic almost instantly and then go to a high level item and takes forever to reverse engineer and get he schematic. (while they both say 20%)

[Sorwen]

No. There's a difference between 0% and 20% chance.

 

Just to state it again, since so many people here don't bother reading most of the posts.

 

Having a 20% chance does not mean you will get 1 success for every 5 attempts, that's now how the math works.

And you didn't read my post. If I do it 40 times(which I have before) and never get the schematic that 100% failure rate. Doesn't matter if I have a 20% chance each time. If you never get it then it doesn't matter what chance you have you still never get it.

 

I agree with what someone else said that there should be a increased chance each time you fail. Though I would stipulate that it is each time you fail on the same item.

[Endymionsama]

By keeping an accurate record of attempts and successes (and remember to count the 10-vs-20% separately). With enough rolls, that should approach the stated success rate. It does for me (actually, better, around 21% for the 20% items after about 150 rolls so far).

 

Let me just say this, with computers there is truely no thing as random. Psuedo random number generators mess up all the time. I have seen it happen in alot of other games. So much so that the developers had to put things in to compensate for that aside from add a higher chance of percentage to not fail. In City of Heroes I tracked drops for some the loot we got in that game. I had done it for months and never got what I wanted. Statically it should never have happened. However the developers in that game eventually changed the way the recipes were weighted as well as put in some direct buy methods to get the same thing the random the RNG would give you. In the game Aion the developers put a streak breaker in the game. However you would have to be an extreme masochist to ever continue crafting expensive items to get the streak breaker to activate. There was an exploit to force the streak breaker to happen after crafting hundreds of items.

 

Long story short in this game what I feel is that they are just suppressing the message instead of just giving us a so called higher percentage.

[Zhiroc]

Let me just say this, with computers there is truely no thing as random. Psuedo random number generators mess up all the time. I have seen it happen in alot of other games. So much so that the developers had to put things in to compensate for that aside from add a higher chance of percentage to not fail. In City of Heroes I tracked drops for some the loot we got in that game. I had done it for months and never got what I wanted. Statically it should never have happened. However the developers in that game eventually changed the way the recipes were weighted as well as put in some direct buy methods to get the same thing the random the RNG would give you. In the game Aion the developers put a streak breaker in the game. However you would have to be an extreme masochist to ever continue crafting expensive items to get the streak breaker to activate. There was an exploit to force the streak breaker to happen after crafting hundreds of items.

 

Long story short in this game what I feel is that they are just suppressing the message instead of just giving us a so called higher percentage.

 

True, computers use pseudo-random number generators. But that said, there are library implementations of such that generate statistically valid random numbers. In fact, computers are used in real-life design and analysis utilizing monte-carlo simulations where the basis of the results depend on the RNG.

 

So, are RNGs useless? By no means NO. Can RNGs be used incorrectly? DEFINITELY.

 

A few weeks ago, I was THIS CLOSE to filing a bug because I thought the number of long failure strings were unrealistic. That's when I wrote my simulator and eventually did the analytical estimate that proved to me that my results were well within the expected results (at least from an eyeball look--I only know enough statistics to be dangerous )

 

Luckily, because I DO keep exact track of each and every success and failure, I was able to do this. So if you are claiming a bug, do the same, and when you have enough results (100 or more rolls), you can be more convincing. Maybe there's a bug with certain items. Maybe there's a bug between the single- and multiple-schematics items (i.e., mods vs. other items, I mostly do the former).

Edited May 23, 2012 by Zhiroc

[Maughann]

I feel your pain.

I've been grinding out Redoubt Powered Ultramesh Leggings for several days now and REing them. No success so far. Think I'm upto 25 REs now.

[Zhiroc]

I feel your pain.

I've been grinding out Redoubt Powered Ultramesh Leggings for several days now and REing them. No success so far. Think I'm upto 25 REs now.

 

The way you said that, it sounds like you're looking for a purple? If so, then that is probably a 10% success rate. According to my analysis, about 93% of the time you should have succeeded by the 25th. While that seems "very likely", it's still only happens about 14 out of 15 times, or in other words it'll take you at least that many tries every 1 out of 15 times you go for a success. You'll only cross the 1 out of 20 level at about 29 tries, 1 out of 100 is at 44.

[Endymionsama]

The way you said that, it sounds like you're looking for a purple? If so, then that is probably a 10% success rate. According to my analysis, about 93% of the time you should have succeeded by the 25th. While that seems "very likely", it's still only happens about 14 out of 15 times, or in other words it'll take you at least that many tries every 1 out of 15 times you go for a success. You'll only cross the 1 out of 20 level at about 29 tries, 1 out of 100 is at 44.

Thats another thing I wanted to ask. Before at the lower levels going from green to blue was 20% and blue to purple was 20%, why is it now 10% when going from blue to purple? Why not make it 20% always? To me they are making it too hard just due to that, lets not forget there are 5 possible recipes you could end up getting when it does crit. So that 20% is actually makes your odds even lower to get what you want. If anything the mod recipes should be 10% all all around just to be fair. You only need to unlock one for your character where as with the others like synthweaving and armortech you need to unlock the whole suite up to purple.

[VanorDM]

If you never get it then it doesn't matter what chance you have you still never get it.

 

Just because you did it 40 or 400 times, doesn't change the odds. Each time you do it, you have a 20% chance of sucess. The fact that you failed 40 rolls in a row doesn't change anything.

 

I agree with what someone else said that there should be a increased chance each time you fail.

 

Yeah I'd agree with that, because even though you could fail 400 times in a row, doing so would suck and it would be nice to know that after X tries you'll get a sucess.

[Klyvanix]

I think they need to have a bonus to getting a purple if you disassemble an item with an augment slot on it, a 5% bonus would be nice, because those are extremely nice quality items. there has been several times where I am trying to get a focus item on my artificing. now granted these are not 20% chance, but a 10% chance, and there are several different purples to unlock from these. Hawkeye, Tempest, Redoubt, etc. Now when I crit on the blue focus and get an augment slot, I cringe because those are extremely good quality. So it wouldn't hurt to have a 15% chance to crit on those items, instead of a 10% chance.

 

because of the diversity on the focus items I have probably crafted over 50 blues and unlocked 3 purple recipes. And there is still a chance to unlock something, so I keep on trying.

Edited May 23, 2012 by Klyvanix
forgot to add something

[sroster]

I think they need to have a bonus to getting a purple if you disassemble an item with an augment slot on it, a 5% bonus would be nice, because those are extremely nice quality items. there has been several times where I am trying to get a focus item on my artificing. now granted these are not 20% chance, but a 10% chance, and there are several different purples to unlock from these. Hawkeye, Tempest, Redoubt, etc. Now when I crit on the blue focus and get an augment slot, I cringe because those are extremely good quality. So it wouldn't hurt to have a 15% chance to crit on those items, instead of a 10% chance.

 

because of the diversity on the focus items I have probably crafted over 50 blues and unlocked 3 purple recipes. And there is still a chance to unlock something, so I keep on trying.

 

This is a great idea. Or I think a simple 1% increase after every failure would be nice. Not huge so it wouldn't be making it a cakewalk to RE (especially the 10% blue to purple rate) but you would know that eventually you will get to a 100% success rate. And after 30 or so failures you would know you have got that up to a coin flip and wouldn't feel so bad about sinking more mats into the gamble. It would reset back to 10% once you had a success.

[ekimmak]

I think a simple 1% increase after every failure would be nice.

 

Make it a 1% increase after every failure after 5 failures. That would be when people would start getting angsty about failures.

[Atlanis]

Hey I reversengineered three T4 RElICS before getting a schematic (Call me INSANE)

After that I built 10 of these and could buy myself all Rocketbootupdates^^

 

Now I already destroyed another 4 T4 RELICS and still no schematic, do I whine? No, cause I know hard days of farming all the daylies and space missions will pay off through a lot of credits:D

 

Just get a little bit more calm. Its a win win situation for me, dooing daylies I get money if I get a schematic of a RELIC I get more money, if I not I am back to doing daylies again.

 

Well 1 DD RELIC Down still a lot more to farm:D

[Calitri]

It's pretty irratating, i'm trying to keep my armor mmods and mods up to date for next level, and i've RE'd about 15 green plates and no blue schematic. Got cybertech for 2 reason, 1 was so I could just keep my gear up to date, as well a my friends while leveling up thru PvP and not have to touch comms. Gets expensive after so many fails.

[Blackardin]

Write a simulator like I did (see a post above), and you will see that long strings of failures are certainly possible, in fact, likely over time. I ran my simulator for a million dice rolls, and after doing that 10 times, each experiment had one string that required over 50 rolls to succeed, and about half of the experiments had one that took more than 60.

 

In fact, the 90th percentile is at around 10 rolls, meaning that 1 in 10 times you try to get a specific schematic, you will take 10 or more. The 95th percentile is at around 14, so 1 in 20 will take that many. And the 99th percentile is at 21, so 1 in 100 will take over 20.

 

Like you, I thought the RE chance was broken, but luckily, I keep complete records for every RE attempt I make. I was surprised to find out that my actual success rate is slightly over 20%. It goes to show that we humans suck at having a feel for probability.

 

All well and good, but I'm at 30+ tries and five hours (between log offs for missions, unbalanced return of silica over laminoid) for a level 13 barrel for an alt.

 

The problem is not whether the overall success rate is valid, it is those very streaks that cause the frustration that many feel. Should never happen, not once, not ever for anything below level 50. They have a great system here. Great design, just a very poor implementation of such. Looks good on paper, but not in practice.

 

I suggested the increasing probability change per attempt shortly after release. If it was possible to code it in, it is something they should have jumped right on.

Edited May 25, 2012 by Blackardin

[Blackardin]

Final tally, turned on about 34 attempts.

 

Trying for the nylasteel needed for the pattern just turned. Jumped in and out of my ship over 20 times trying to get a rich mission in hopes of a crit. Never happened. Not a one. Two abundant in all the attempts. no crit, no item, no fun. Logged with that frustrated feeling once again.....

[Thundergulch]

I'm amused by how people think that a 20% chance means 1 time out of every 5.

 

Go to Vegas, put down $100 on Green 00 for 35 times in a row.

When you don't win,and you've lost $3,500, complain to the pit boss that it's 35:1 chance it will hit.

See what he says.

[MadMaxamillions]

My point has been that if I was a pit boss, and someone ran on a 1:5 odds game and over 1000 tries only succeeded 66 times, that I would be arrested and investigated for gambling fraud.

 

Off chances are VERY possible. But after I noticed a consistent pattern of 1 in 15 for several hundred attempts, then made a log, and after 200 attempts still ended up with a 1 in 15 chance of schematic crafting (oddly, for both 20% and 10% schematics, both had a 6.6% success rate over time). Note that this is for Cybertech, and I think a primary issues is that SW devs may not be looking at them individually.

 

I'm the opposite of whiny - I like a challenge, and failure is expected, but when it starts getting to .03% odds of failing as much as I am, I gotta ask, for how long until normalization?

 

If you flip a coin 1000 times and it lands on heads only 170 times, isn't something awry?

 

SO.... btw, to everyone who said we're full whine?

 

http://www.swtor.com/community/showthread.php?p=2613451#edit2613451

 

The reverse engineering return rate is not currently working as intended. In an upcoming patch, we are improving the reverse engineering rates – especially for the higher-level items. We are also looking into the issue where a player receives "You already know that schematic" instead of learning a new research variation.

 

Patrick Malott

Systems Designer

Edited May 27, 2012 by MadMaxamillions
shtr