Development Cycle Archive

Thread: Crafting Critical Failure Rates

RogueCloudwalker
Mon Feb 02, 2004 1:08 pm
#430

Here is another Jedi horror story


http://forums.station.sony.com/swg/board/message?board.id=jedi&message.id=246488


All that trouble for a critical failure.



Captain Trevick Cloudwalker of the YT-1300 "Corellia Star"

If you like my ideas, you would probably like the others too! Please support me!

My Instanced Starport Proposal GCW NPC Prisons in Space Proposal
My GCW/PVP/Player Bounty Proposal
A very feature rich Hoth
KelvinKlien
Mon Feb 02, 2004 1:54 pm
#431

Criticals should be removed for masters. On experimentation, sure it is acceptable to fail sometimes. You are gambling on a better version after all. However plain straight criticals is painful and further deflates the 'fun' of artisan crafting. I am Master Bio Engineer on one character and I expect 1/4 crafts concerning pets to critical. Usually at the final combine. Many friends have quit because of criticals, they got frustrated and went to other games....that is not a good thing.


I do not see how this would be broken. Masters should never fail!


The game is supposed to be 'fun'. Not torture.
Propolis
Mon Feb 02, 2004 2:02 pm
#432

Th, pardon if this has been hashed...



My only problem withexperimentation crits is on especially difficult schematics - such as Area Effect Mind Poison C. BUT!!! let me lay the caveat that it may already be well balanced. Note: my experience is from within an research center.


There are two ways you can experiment this - Severalbubbles at a time, or single bubbles at a time.


Several: The crafter decides to experiment more than one bubble, and chooses 2-8 bubbles to experiment on. On an easy schematic this can fail relatively easily. On a difficult schematic you are almost guaranteed a failure before you are done fully experimenting. This is most likely intended.


Single: Crafters often experiment one bubble at a time. I can think of two reasons for doing this - Either you are crafting something that can handle a little deficiency, but cannot afford the full loss of the item (large harvesters anyone?). OR, you are crafting something with such a high difficulty rating, it is the only possible way to get a perfect schematic. THE PROBLEM is that now you are experimenting TEN times. I don't know the math to figure out chances of 1 failure, but with a base 5% chance, after 10 rolls of a 20 sided, you are fairly likely to actually roll a 1 (crit fail) at least once. This may be as intended, but I certainly do not enjoy making 10 AE poison schematics before one comes out without crit or partial failures (moderate success IS a failure - it is usually worse that simply less success).




So Thunderheart, the question I would have you answer is this "Is it appropriate to have to burn 10ish items getting a 'great success'+ schematic on a difficult item, when the crafter is in a research center?".


Personally I think the average for the hardest item should be approximately 2imperfect schematics for every perfect one when experimented via single bubbles. I shudder to think what would happen if architects ever get a schematic this difficult (or anyone else with a very expensive experimentation).



Propolis,
Retired Character.
Contact Goa'uld in game, or GoaOld on these boards.
Dark_Prophet
Mon Feb 02, 2004 3:16 pm
#433


Krayt pearls... going for as much as 12 million on Bria (though you can probably get them for 6 mil)


I used the 4 best krayt pearls I had (each of them had to have been worth 10 mil) Here are the stats:



Pearl 1:

Color: Red
Damage: +189

Speed: -1.13

Wounds: +1

Force Power: -1


Pearl 2:

Color: Light Purple

Damage: +191

Speed: -1.11

Wounds: +2

Force Power: -1



Pearl 3:

Color: Dark Red

Damage: +192

Speed: -1.33

Wounds: +1

Force Power: -3



Pearl 4: (This is my favorite pearl)

Color: Light Green

Damage: +197

Speed: -1.15

Wounds: +1

Force Power: -9


Anyways, they're worth a fortune. So, I construct the very best lightsaber crafting tool I can, and the best crystal packs for the training saber, etc. I get amazing andgreat successes across the board until I hit the 5th Generation saber... and what does it do... it critical failures. That's 40+ million credits of resources down the drain for nothing. And they say us Jedi are having a great time, lol. I'm glad it was on Test Center 2....

Thoughts: Sabers shouldn't decay (or have them decay at a limited rate comparable to the Jedi's skill level), and sabers shouldn't ever critical fail

Decadre
Mon Feb 02, 2004 3:54 pm
#434

This is something interesting that I think the Devs should look into perhaps. Between my brother and myself, we play this game on three different PCs. Without a doubt, one of them produces a significantly higher critical failure rate than the other three.


The other nite for instance, I decided for the hell of it to rid myself of 6 Volitor armor segments. BTW, I am a Master Armorsmith. I crit failed 2 of them, and crit failed on experimenting on another 2! My brothers a master BE, and he stopped crafting on that same computer as he was experiencing an extremely hi critical failure rate while making pets.


This may be a problem with some machines perhaps, and with the method of generating random numbers.


Now here come the rambling part... I remember one time writing a simple C++ program, and I seeded srand with the time() function which uses the time from the PC. My memory is a bit vague, but in addition I also divided the system time by a very small number like 5 or something. What happened on my PC at home, but not in school was that I would get the essentially the same numbers in a very familiar pattern.


Something like 5, 12, 34, 56, 79, 5, 12, 37, 56, 83, 7, 15, 32, 75, etc


Now, one thing I noticed on another of my PCs is that when im power grinding a crafting class, I can almost see a pattern at time if I do it fast enough with like 6 or more crafting tools and making something very simple, wherein I get 2 crit failures in a row! That could in essence show that perhaps the above pattern is happening on this PC.


Now this is all conjecture on there being a problem seeding srand with time(), as I'm not even sure how the crafting system works internally inside the programming code, but perhaps some people are seeing a problem on some PCs as time() is returning a number or numbers that are not incrementing or changing as fast as other PCs. This in turn can cause these users to end up getting not as random numbers as other users.


So what I am trying to say is that on most systems, this is what is happening on a range of 0-100

5, 56, 34, 78, 99, 65, 43, 23, 77, 51, 81, 100, 6, 36, ...


but on some systems this may be happening

5, 8, 13, 23, 26, 31, 37, 42, ...


Now factor in that programming code take milliseconds to execute, and that it can take quite a few seconds to fill up a schematic at times, if not longer, this may explain why people don't always experience a crit. And it could also explain why some people experience them when crafting very purposely and quickly. They fall into the zone where unfortunately they are going to hit those failure numbers because something is FUBAR.


Well, anyhow this is just a theory and my 2 cents on the subject.


Plageron
Mon Feb 02, 2004 5:24 pm
#435

So its great to take large numbers and to skew the results, but failing to calculate correct sequential data is very bad.


Yes, so anywise...its nice to say 4%....
seems like a low number, but when you do your calculations correctly you find that 4% is what the variable is set to, but not what the actual number is.

Here are some stats for you. Talking the above Posters experimentation totals and putting them into a realistic view for some people.

Assembly attempts: 1448
Assembly critical failures: 65
Assembly critical failure rate: 4.49%

Total Creature Objects Attempted By a Bio-Engineer using those statistics 724
With those numbers of failures that would make the failure rate percentage equal to: 8.97%

Total number of say an Architect Buildings created using those statistics: 181
And with those numbers we would have a nice Critical failure rate of 35.91%

Experimentation attempts: 1284
Experimentation critical failures: 54
Experimentation critical failure rate: 4.21%

Now lets take a look at these Experiemtal ones closer.
Lets say we have 10 Experimental points. Yes thats lower then what many people have. But you will see my point.

Total number of Items created = 128.4 Items
And with that critacl failure Number the rate is more like 42.05%.(Thats per object that you would get a critical failure during experimentation)

So anywise...I am very surprised no one has noticed this.

Yes it means we fail alot.And alot more for a Master proffession.

Now to teach some of You who Dont understand and failed Math class.
Its called a Numeric statistic progression, because even though the percentages for each attempt do not change, the over all number of times you do something increase the odds. And since some objects require multiple attempts, the percentages increase the more times you do not fail.

Thats how you can tell if a random system is rigged or not.

Because the more times you do the system the better the chance you will win. But you do not have to be the one constantly genrating the numbers, othe rpeople can.(This is the reason I dont go to Casinos becasue weighted and rigged machines are uncool and the Lottery systems...lol)


Also I want to re-state that I no longer get Amazing successes on anything I do since I became a Master.

Except for when I critically fail which results in reducing some statistic to zero on an item and then when I try to experiment on that line I get Amzing Successes all the time, but get no actual results from the so-called amazing success.

I would estimate that its becasue some coder put a numeric number that is causing a certain value to become zero when a rating is reached.
(let me guess not able to calculate beyond 99 or with the 4% junk hmm 96..and everyone who is a Master is 100...LMAO)

Which means that the Failure Percentages may not be very accurate, because the equation might be messed up just enough to skew the results again.

So anywise...Stop posting large numbers of individual experiements and post the ones using multiple objects for one item.

If I where an Architect i would Be real ticked with these numbers.

OH gee....Thats right...Architects are ticked.......gosh I wonder why, with only a 50% chance to fail all the time.

Yes I didnt even go as far as to calulate Experimintal failures combined with whole object failure.

Becasue If I did we would see every object that anyone makes that requires multiple objects to make it would have a very high rate of failure on the finalized product.

but go ahead...Use the statistics posted and figure it out.



---"Back off Man!! I'm a Scientist."----
Peter Venkman from Ghostbusters
alioninwinter
Mon Feb 02, 2004 6:11 pm
#436

I think a big part of this is the numbers you have are just critical failures, where failures in other areasis an issue too. There is also an issue with the number of failures in a row which just magnify the issue to us crafters. For example, as a Doctor, if I want to make crates of the 6 stat buffs, I would make sure I had at least 1000 extra avian meat and reactive gas (above the required 138). With anything less there is little to no chance of making fairly similar buffs. I'm not gunning for "perfect" buffs with every experimentation point I put in coming out perfect. In reality the 10 experimentation point boxes added, I am hoping to get 7 "filled" boxes - I don't think that's a lot of ask. With this in place, just two weeks ago I had over 9,000 of each resource, to make crates50 I need only 6900, but I'm left with just under 800 of the resource left which tells me I had to craft about 56 times to produce 6 schs that did not have a failure or critical failure (very few would be due to falling short of my standard 7 of 10 boxes rule) during the creation process. This is with a station rated in the 44.xx and tools rated at 14.9x. I even have +4 medical assembly. Now things are not always this bad, but this is very had that in one session things could be so rough.



==================================================
Master Dragger, Drunken Friar
Gabriev
TheWizardKs
Mon Feb 02, 2004 7:05 pm
#437

Critical failures, ya they bite. But hey, I can live with a 4% failure, even as a master. But it is not fair in my opinion to let architects have *no* loss at all on installation combines when they crit. To be fair, all professions need to have this on expensive combines, like Ris, Kryat weapons etc. I would not recommend removing the no loss on arch's LOL, that would ruin the profession.


Now I really would like to air my real complaint about crafting. That being that there is such a hard limit on what can be crafted, that it rarely matters what quality of resources you use. I have crafted alot of armor experimenting with this, and I can take materials with alot lower stats and make the same armor as I can with extremely good quality resources. Given, it takes getting amazing successes on the lower end stuff, but it can be done. With the high resources, even with amazing success, I still end up with the same as the great success experiments. This is flat out wrong.


Until this issue is resolved the crafting professions are all broken.


HyKsOs24
Tue Feb 03, 2004 3:16 am
#438

Want one more Jedi CF?


http://forums.station.sony.com/swg/board/message?board.id=jedi&message.id=250513


sadly it's MY story


Now I'm using about the worst saber a Jedi can have, I can't do special attack it drain all my force power... so why the DEV decided to hateme so much? This light saber is the most important thing for a Jedi and this is not a crafting proff, we don't sell them !! So why DEV put a random generator that decide "ok... the NEXT Jedi that craft a saber gona get a CF because we gota respect the % of CF on the server or else all the god **edit** server will be unbalanced, so letsmake a Jedi really madMwouHaHaaAHaHaH".


And belive me... it made me EXTREMLY mad, I did master BH before the patchand was getting lot of bugued mission and it's far less frustrating then what just happened to me....



Every man dies, not every man really lives.

Coldghost ~ Master Slayer

Valin Horn ~ Master Tank

Amosis Sowo ~ Master of the Force

Endemoniado
Tue Feb 03, 2004 6:31 am
#439

Chrysalide,


while i believe you got those figures, I don't think they are the whole story.


I am a master doc, my tools areabout +14%(at work right now) and my crafting station is +36%. On Saturday night I had 6 crit fails in a row. It was only making advanded bio effect controllers, so no big deal at all, but at a chance of just under 5% (according to your results) the odds against 6 in a row crit fails are over 60 million to one against. If we still have 300,000 subs then everyone on the servers would have to craft 200 times to get one such result.


What makes me even more suspicious is that after the 6th one (I rememberit was 6 because (being a geek with a geek gf) I called over the gf in question to point it out) I gave up and tried to produce an advanced solid delivery shell. That worked fine, and then when I wend back to adv BECs they worked fine too. Sounds like it is possible to get into a certain "state" which had to be reset with a different craft.


OK, so 60,000,000 to 1 is not impossible. But neither are those gambling odds.


Endie
Ryche_Mykola
Tue Feb 03, 2004 8:29 am
#440

I think Plageron is right on.


As a Master Armorsmith, if I get even 1 Critical Failure (or moderate success, those are just as bad) in Experimentation the piece is garbage.


Numbers should be done with # of items created, instead of experimentation attempt. Where in if you get 1 critical failure - moderate success in any box of experimentation, no items is created. I'll bet the numbers get into the 40%-60% range then.


I personally think Critical Failures on the combine phase should be taken out or made to be like architects for all professions, where you get your resources back.





Ryche Mykola
Outer Rim Collective (ORC) High Council


Ariakus Mykola
Outer Rim Collective Master Rifleman
Cancelled accounts for Wow
Be0Wulfe
Tue Feb 03, 2004 10:24 am
#441

I've been crafting (artisan & weaponsmith) since beta3 - and I can unequivocally say that the effect of better FR tools & stations is minimal or non-existent.



Ackepawa Ackepawi (AA Kriegswerk)
At the Old Masters Hall of Vendors - Weapons, Armor, Droids, BE Clothes, Vehicles, Resources & more!
Bestine, Tatooine -1260 -2990

Mor-Dan
Tue Feb 03, 2004 1:30 pm
#442

i agree with one of the first posts (and it has prob been echo'd, but this is my voice = )

ther eis failure and there is success. there is no in between. why? because f'd up weapons dont sell. period. 4%-5% is WAY TOO HIGH for a master. this should be closer to 1%. one out of every 100 tries. you should drop the "moderate success" and "failure". there should only be crit fail, good success, great success, and amazing. in addition, the failure should be component specific in a final combine. i am making a republic blaster. i add my resources, my blaster power handlers, my pistol barrel, and my scope. complete. critical failure in the blaster power-handler. i get all my other stuff back. see? right now Architect has an extremely unfair advantage just because of resource requirements. do i think everyon should get two tries? no. but it is not right for them to keep everything in tact and all other professions to lose everything in one attempt. that is not 'balance'.

before any arch gets on here and flames me about how rough they had it, can it. i dont care to hear to your sob story. i have my own. i was an arch for 6 months.

I prefer my flames in Limerick form.



Vendor Locations:
-1560 120 in Soal Valley, Corellia
-4700 5600 north of Theed, Naboo
Page 34 of 49