Forum Settings
       
This Forum is Read Only

Test Central: Log Parse Results = MAX Attribute+Gear EffectsFollow

#1 Jan 11 2011 at 1:40 PM Rating: Excellent
*
56 posts
I. POST UPDATES
II. INTRODUCTION
III. EMPIRICAL TESTING GUIDELINES
IV. GENERAL POINT MECHANICS
V. STATUS POINTS
VI. ELEMENTAL POINTS
VII. DAMAGE FORMULA THEORY
VIII. ADVANCED DAMAGE FORMULA
IX. CRAFTING ANALYSIS
X. GATHERING ANALYSIS


=========== SECTION I: POST UPDATES ===================================================

5/19 FFXIV FORUM UPDATE: Official FFXIV Forums: Dev Team Feedback and My Response

5/16 TEST RESULTS: The Effect of a Heavily Damaged Weapon on Conjurer Skills: See Advanced Damage Formula Section. Test with Deus Bahamut and Iris Akiha of Lifestream.

5/16 TEST RESULTS: The Effect of Output on Gathering: See Gathering section. By Ami Trubble of Lifestream.

5/16 POST UPDATE
--Added section on Gathering Analysis
--Removed sections on Major Game updates and Current Projects
--Updated and condensed all existing sections

5/12 TEST RESULTS: Effect of Magic Potency and MND on Cure Potency by Ami Trubble of Besaid and Lifestream LS.

4/29 TEST RESULTS: Log Parse Results! Effect of STR and DEX on Damage and Accuracy

4/25 TEST RESULTS: Log Parse Results! Effect of New NM Gear Stat Bonus on 30MRD

3/25 FINALIZED TEST RESULTS: Fully analyzed data and put in final recommendations for Stat Point Allocation and Appeal to Dev Team. This test case is now 100% complete!
* The Effects of INT PTY RANK and ELEMENTAL POINTS on Scourge Scourge II Shadowsear

3/24 FINALIZED TEST RESULTS: Fully analyzed the data and added several new items on the spreadsheet to illustrate the effects of the variables. This is the methodology I will be using in my tests from now on. This test case is now 100% complete!
* The Effects of INT PTY and RANK on Phantom Dart and Damnation

3/21 MAJOR POST UPDATE
*Massive content updates to every category in the entire post, added new categories
*Major changes to post format, organization, etc.
*FFXIVCore Exclusive Naoki Yoshida Interview, hints major damage calculation changes are underway in Upcoming Patches

2/7 NEW TEST RESULTS: My First Parsed Test Results:
* The Effects of INT PTY and RANK on Phantom Dart and Damnation


=========== SECTION II: INTRODUCTION ===================================================

PURPOSE
To provide a concise, global picture of FFXIV battle system mechanics and status allocation. To cut down on chatter around badly executed tests, emphasizing true empirical testing. To set baseline requirements for testing to include automatic log capture and parsing scripts. New theories, tests, and results will be documented here. Existing test posts in forums will be linked.

AUDIENCE
Advanced players who are interested in damage optimization, who can contribute theories and/or empirical test results. Read Section II for details how to automate your testing. It is VERY EASY to test using Crabby as an external log parser.

WORK IN PROGRESS
This post is a growing framework, that identifies what needs testing, and where it fits in the big picture. The format may change as test data is verified and integrated, and game updates are made. Please help out by posting links to any related testing results.

Enjoy and I look forward to your feedback and your own experience testing game mechanics.
With collaboration, we can get somewhere.


=========== SECTION III: EMPIRICAL TESTING GUIDELINES ===================================================

GUIDELINES
1) Maintain the same rank.
2) Fight one mob at a time. Do not fight mobs that do not give any EXP. The parsing script assigns each battle data to the next mob's exp that dies. Fighting two mobs at once will result in an abnormal EXP rating, and those results will be thrown out because they can not be accurately assigned to a particular mob rank. Killing a mob worth no EXP will result in those attacks being assigned incorrectly to the next EXP mob that dies.
4) Keep your equipment durability topped off. Equipment durability starts to have an impact at 35%
5) If a mob part breaks you will do more damage with subsequent hits. Buffs and debuffs should also not be used unless they are defined as variables. Data with more than 1 variable should be removed from averages. I can do this automatically in the script if it is brought to my attention, so do not feel that your data is irrecoverable.
6) Ensure only a single variable exists, and that all other factors are constant. Consider all equipped gear, status allocation, mob type and level, durability, etc. Use the experiment definitions in my own tests as a guide.

AUTOMATION
Crabby Thread, by golfandsurf
DOWNLOAD CRABBY HERE
**NOTE THIS IS A THIRD PARTY PROGRAM. USE AT YOUR OWN RISK. I DO NOT CONDONE THE USE OF THIRD PARTY PROGRAMS TO GAIN AN UNFAIR ADVANTAGE OVER OTHER PLAYERS**
1) Download Crabby.rar from the link above and extract the EXE to your own directory.
2) Start FFXIV and log in using windowed mode.
3) Run Crabby.exe. Check if it has synced up by looking at the bottom of the window. It will say "Attached". If you need to resync, just double click the icon at the lower left of the window.
4) You can create separate window feeds for each chat type. Click "Settings > Chat Tabs" and select one or more types. They correspond to what you "Show" or "Hide" In-Game under "Configuration > Chat Log > Settings". Best of all, there is no limit to the history or number of log entries Crabby remembers.
5) Execute your test with the guidelines above, and copy out the timestamped entries into an external text editor. Save it as TXT UTF-8. Please be aware that logging out of the game may crash Crabby and prevent you from copying out the logs. Be on the lookout for bugs and report in the Crabby Thread. Try a more stable version of Crabby if you experience problems.

CONTRIBUTION
If you would like to send me log data for any test, I am willing to parse your test with my Perl scripts. Please message me through the forum. It must be in Crabby's "RAW" format to include the chat codes. Please take out any personal information such as linkshell chat.
If you would like to contribute by performing your own test,
1) Follow the guidelines above, capturing your logs as explained above (submit Crabby's *RAW* data)
2) You also need to state all the constants and variables as shown in the first page of the spreadsheet in any of my own tests.
3) Please PM me for my email and I would be happy to analyze your data similar to the results in my tests.


=========== SECTION IV: GENERAL POINT MECHANICS ===================================================

STAT POINT REASSIGN
At Physical Level 50, it will take exactly 3 hours for Reassign to regenerate after use. Tested late January 2011.

STAT POINT CAP
In general, there is a point where it has been observed on a PER SKILL basis at a specific Player and Mob Rank, that increasing a stat will provide no significant or measurable benefit. The cap is somewhat inconsistent. I have lodged a complaint to the dev team regarding this.
**"Stat Caps Per Rank [WIP]" by Beregren. Last updated 1/22/2011
This thread tracks each rank where adding VIT stat no longer increases your HP.
STILL NEEDS TESTING

STAT POINTS ON GEAR
TEST RESULTS: Log Parse Results! Effect of New NM Gear Stat Bonus on 30MRD
This test showed a correlation where a decrease in "Attack AND STR" at the same time, resulted in a decrease in average damage and average critical damage.
Discussion: There is a distinction between adding stats in your stat allocation and stats in your equipped gear. The difference is clear that you do not receive HP from adding VIT gear, but unclear how else specific equipment stats may be ignored.
STILL NEEDS TESTING

SP GRINDING "FATIGUE"
Your log will tell you you have received "bonus skill points" when you are under fatigue. When fatigue starts or gets more severe, your log will tell you your "skill bonus rank has increased." Each step will reduce your skill point gain -10%.

It's easiest to think of fatigue as a slider that slowly resets to origin, and slowly increases based on the AMOUNT of SP you earn, and the maximum time the slider will take to fully reset is 1 week.

**3/21 Update: I've noticed while getting a consistent 50-60k SP per leve reset that it is enough to maintain a 1x level of fatigue (-10%) through the whole week, where at the beginning and middle of the week there is a short period of time where there is no fatigue. I never hit 2x fatigue. Since there is 4.6666 leve resets per week, this means earning roughly 234-280K SP per week will keep you safely within 1x fatigue. The amount of SP needed to hit fatigue from the origin may vary depending on your rank. Also some perspective, if you zerg any job from Rank 1 to 18 in a day, you will not hit 1x fatigue until you hit Rank 18. I tested this on 5 jobs 1 to 18.

"FFXIV Tactics: Understanding the Fatigue System" VIDEO


=========== SECTION V: STATUS POINTS ===================================================

Quick and Dirty Recommendation for Stat Balancing
STR: Max it as high as possible to affect basic attack damage. May cap out effect sooner for weapon skills (untested)
VIT: Maximize after you are happy with your other parameters. Affects Max HP and may also contribute to reducing raw damage.
DEX: In large volume, DEX will result in a decrease miss rate.
INT: Max it as high as possible to affect basic wand attack damage. Will cap out or not be relevant for weapon skills and nukes.
MND: Don't increase it past 40. You shouldn't ever have issues getting MP with Siphon MP. Also affects Cure potency.
PTY: Minimize, it's utterly useless. May still affect enfeeble accuracy and duration, but this is untested and would only matter for NMs.

STR
Log Parse Results! Effect of STR and DEX on Damage and Accuracy by myself, Deus Bahamut.
STR behaves much like INT such that an increase in STR will result in a steady and consistent increase in damage.

INT
The Effects of INT PTY and RANK on Phantom Dart and Damnation by myself, Deus Bahamut.
The Effects of INT PTY RANK and ELEMENTAL POINTS on Scourge Scourge II Shadowsear by myself, Deus Bahamut.
**See each thread for my conclusions.
**Increasing INT values does steadily increase Phantom Dart average damage. However may not affect Weapon Skill or Nuke (Scourge/Scourge II) damage. Either the cap is met at a lower value, or INT does not affect Weapon Skill or Spell damage. Phantom Dart does not yet appear to cap at all. Stick to using Phantom Dart on SP grind mobs with a maximized INT and minimized PTY.

DEX
Log Parse Results! Effect of STR and DEX on Damage and Accuracy by myself, Deus Bahamut.
Increasing DEX will result in a decreased Miss Rate. [/b] This test shows a correlation between increasing DEX by 120 points resulted in a consistent Miss Rate decrease in Light Swing, Skull Sunder II, and Trunksplitter II.

PTY
The Effects of INT PTY and RANK on Phantom Dart and Damnation by myself, Deus Bahamut.
The Effects of INT PTY RANK and ELEMENTAL POINTS on Scourge Scourge II Shadowsear by myself, Deus Bahamut.
**See each thread for my conclusions.
**Increasing PTY has ABSOLUTELY NO EFFECT OR CORRELATION TO INCREASING ACCURACY OR LOWERING RESIST RATES for Phantom Dart, Damnation, Scourge/II, Shadowsear
Stick to using Phantom Dart on SP grind mobs with a maximized INT and minimized PTY.

VIT
"Stat Caps Per Rank [WIP]" by Beregren. Last updated 1/22/2011
**NOTE: The following test was completed before the 11/25/2010 Patch.
This thread tracks each rank where adding VIT stat no longer increases your HP.
VIT may also reduce the raw damage you take, in addition to increasing your HP value.

MND
Should behave similar to VIT. Keep your MND at 40 unless youre increasing cure potency. You only need a small amount with abilities like Siphon MP.

Effect of Magic Potency and MND on Cure Potency by Ami Trubble of Besaid and Lifestream LS.
Ami was gracious enough to do this test and give me permission to post her results here. You will see in the following spreadsheet data for Cure II HP values at variable MND and variable Magic Potency. From these values she was able to derive the formula to conclude the following.
Conclusions:
1) Raising MND+2 will increase Cure II Potency by +1.
2) Raising Magic Potency+2 will increase Cure II Potency by +1.
3) Cure II base cure amount is 312.92


=========== SECTION VI: ELEMENTAL POINTS ===================================================

ELEMENTAL DAMAGE
How greatly does increasing elemental affinity affect elemental damage?
STILL NEEDS TESTING
How greatly does increasing elemental affinity affect rate of resistance?
STILL NEEDS TESTING

SCOURGE AND BANISH DAMAGE
Is there a correlation between the increasing elemental affinity and light and dark damage?
Water, Ice, Earth = SCOURGE
Fire, Wind, Lighting = BANISH
STILL NEEDS TESTING

ELEMENTAL BUFFS
Is there a correlation between buff power and elemental affinity?
STILL NEEDS TESTING
**An easy empirical test would be to test Stoneskin I+II power with lowest and highest earth against a Cactuar's 1000
needles.
Determine Elemental affinity:
Stoneskin = Earth?
etc

ELEMENTAL DEBUFFS
Is there a correlation between debuff resistance and duration and elemental affinity?
STILL NEEDS TESTING
**Logging/parsing is nearly required in order to use time stamps to determine how long the debuffs last. This will make for some interesting Perl script enhancements.
Determine Elemental affinity:
Sleep = Water?
etc


=========== SECTION VII: DAMAGE FORMULA THEORY ===================================================

NOTE
As stated above, fansite interviews with FFXIV developers indicate the damage formulas are broken and need to be completely revamped. This section is more of a placeholder and statement of theory until the underlying system is changed.

DEFINITIONS
PC = Player Character
MOB = Monster
RANK = Current rank
STR/VIT = As reported on Attribute screen
STR_BONUS/VIT_BONUS = Equipment attribute bonus total
Attack/Defense = Total Physical Attack/Defense rating on your character status screen
DUR = Object Durability. OBJ value is scaled by DUR based on the object's condition' or 'durability %'. Gear damage *should* only affect the Attack/Defense values of that item only, not the total Attack/Defense.

FORMULA
The formula is merely demonstrating a basic relationship between combat variables. The structure of the formula is loosely organized around the notion that the core damage is decided by subtracting the STR and VIT stats (Group B), scaled by the ratio of Attack and Defense values (Group C).

A,B,C,X = unknown constant values (may even be zero). Also denotes groups referred to below.

A*( MOB_Level / PC_Level ) *
B*( MOB_STR - (PC_VIT + X*PC_VIT_BONUS) ) *
C*( MOB_Attack*MOB_DUR / PC_Defense*PC_DUR )
= Physical damage taken by PC

A*( PC_Level / MOB_Level ) *
B*( (PC_STR + X*PC_STR_BONUS) - MOB_VIT ) *
C*( PC_Attack*PC_DUR / MOB_Defense*MOB_DUR )
= Physical damage taken by MOB

Group A:
This value is significant to soloing in FFXIV as opposed to FFXI. Damage seems to drastically reduce after a mere 2-3 levels whether it is Level 15-18 or Level 37-40 (i.e. MOB turns from Red to Green). The reduction rate appears to be similar in both cases. In FFXI, your level did not seem to directly influence damage much at all, as you could get killed by easy prey at a high level, while at low level you could kill mobs several levels above you in 2 or 3 hits.
**See my test results above for INT and PTY to infer the effect (DELTA)Player/MobRank has on output.

Group B:
This value carries the most weight, and can be heavily scaled by Group C. Against mobs equal or greater to your level, increasing STR will scale your damage better than increasing Attack. Inference from FFXI. This will all change with future updates and require more testing.
STILL NEEDS TESTING

Group C:
Can heavily scale Group B. Against mobs lower than your level, increasing Attack will scale your damage better than increasing your STR. Inference from FFXI.
STILL NEEDS TESTING


=========== SECTION VIII: ADVANCED DAMAGE FORMULA THEORY ===============================================

DIRECTION AND BODY TARGETING
It's clear from the logs what part of the mob you are facing when attacking is noticed by the battle system. It make it more likely to target certain body parts. Does it have any other effects such as increased damage or critical rate?
**3/21 Update: I have confirmed from fighting various open-world NMs that the direction you are facing the mob directly impacts which mob part you are likely to hit. For example, standing in front of Uraeus and using Flashfreeze II will sometimes result in breaking his eyes. Standing in front of Tarbh Usige's legs will make him fall if Incapacitation is triggered. The horns can be broken with certain WS standing diagonally and closer to his side. You will see in the logs that your WS hits the horn, but not normal attacks. Breaking his horn purportedly increases the drop rate of his horn.

DURABILITY
Group D is tied to item durability; a broken weapon of <25% durability will result in a damage penalty.
Conversely, when an armor part breaks, if it is hit again, it will result in increased damage.
Durability only affects Attack or Defense based on the 25/35%, and the status color change of the equipment box on the Gear screen, NOT based on the status color change of the bar indicator. The logs may prematurely indicate the item is damaged, when it has not yet crossed the threshold.

YELLOW DURABILITY EFFECTS
Log Parse Results! Effect of New NM Gear Stat Bonus on 30MRD
Conclusions: Average damage very slightly decreases. Accuracy was inconclusive due to few trials. Compare TEST2 VS TEST3.

RED DURABILITY EFFECTS
Conclusions: Average damage was significantly affected across the board for Spirit Dart, Levinbolt, and Tier II Magic. Accuracy was also slightly affected, resulting in more misses and resists with red durability. A decent amount of trials were conducted to represent accuracy.
Effect of Red Durability on R30 Conjurer Skills test and results logged by Iris Akiha of Besaid and Lifestream LS, parsed by Deus Bahamut.

SPECIFIC TO MOB BODY PARTS
Which part is hit appears to be directly to related to the exact position your character is standing and what skill is used. Breaking a mob part will result in increased damage to that part with subsequent hits. How quickly the part breaks may be random, probability may increase after successive attempts, or it may have a hidden HP bar.
**3/21 Update: A hidden HP bar seems unlikely, because only be executing the appropriate weapon skill in the mob incapacitated. However, there still might be a hidden HP value that must ALSO be met for incapacitation. This theory may be disproven by incapacitating on the first hit of battle.

SPECIFIC TO PC BODY ARMOR
When the monster attacks you, it's probable specific gear is targeted each hit. The damage may just be evaluated on the whole, or may also be independently influenced by that targeted armor's Defense value. The mob might even target more than 1 armor parts per hit.
STILL NEEDS TESTING


=========== SECTION IX: CRAFTING ANALYSIS ===================================================

MEASUREMENT
Full automated log measurement is not straightforward primarily because the COLOR of the orb is only reflected visually, not in the logs. This means that during a crafting test, crafts should only be done on WHITE orbs or only on YELLOW orbs, etc. when observing the success of the craft. The PROGRESS, QUALITY, and DURABILITY can be directly recorded with no issues. Hidden attributes seem to be a RISK value, which increases the probability of sparking or element destabilization. Sparking is rare and usually from doing too many bold synthesis, but sparking does not appear in the logs, but only visually. Element destabilization does appear in the logs, so this can be measured.

So assume a scenario where an item is being crafted equal to your rank (for a more telling distribution of passes/failures). You would exclusively do a standard synthesis ONLY when the orb is white. If the orb is not white, you can WAIT or selectively use abilities to turn it white (Preserve) but NOT abilities that MIGHT affect SUCCESS or RISK rates. Until these abilities are tested they might just throw off the results.

The cool thing is it doesn't matter if you fail the synthesis, each individual 'standard synthesis' attempt is all that matters. If you can optimize each individual synthesis attempt, imagine the impact that will have on your overall item success rate.

EFFECTS OF CRAFTSMANSHIP VS MAGIC CRAFTSMANSHIP
From my research online and personal experience, this is still a great unknown. It is unknown which items and trends apply as to which specific recipes have higher success rate with one stat higher than the other.
STILL NEEDS TESTING

EFFECTS OF CONTROL
The hidden risk value seems to effect element destabilization and aether sparking.
STILL NEEDS TESTING

EFFECT OF DESTABILIZATION
The presence of destabilization seems to dramatically reduce standard synthesis effects (increased durability loss and reduced progress). This, however, needs to be more accurately measured.
STILL NEEDS TESTING

EFFECT OF AETHER SPARKING
I'm guessing the sparks have a similar effect as destabilization.
STILL NEEDS TESTING


=========== SECTION X: GATHERING ANALYSIS ===================================================

EFFECT OF OUTPUT STAT ON MINING AND LOGGING
For mining and gathering, Ami Trubble of Besaid and Lifestream LS has calculated the formula which determines the effect Output and Rank on the number of attempts allowed at each node before you have to move to another node.

Effect of Rank and Output on Mining Attempts by Ami Trubble of Besaid and Lifestream LS.
Conclusions:
1) Each Player Rank adds 4 to the Effective Output, allowing more extractions.
2) Most of the time, an additional 23 Output is required for +1 extraction attempt. Exceptions are 22 more output to go from 4 to 5 extractions, and 30 more Output to go from 6 to 7, or 7 to 8.
3) Minimum extractions is always 1, and maximum extractions is always 8.
4) Each Grade requires 60 more output to get the same number of extractions as the previous grade.
5) 60 output is 20 more than what is received from just ranking, so the extra 20 output needs to come from gear, meaning that a Miner needs to add +20 to output per grade to result in the same number of extractions.


EFFECT OF OUTPUT STAT ON FISHING
For fishing the Output affect is still not defined to a science, but appears to lower the amount of time you wait for a bite.
STILL NEEDS TESTING

EFFECT OF GATHERING STAT
It is believed that the rate at which the remainder bar decreases is reduced. In this way, you are able to strike more times to attempt to procure higher rank items. Higher rank items clearly start with a smaller remainder bar, which makes them more difficult to obtain, but also gives you a clear indicator what item you are about to get before you even get it. The rate at which Gathering scales and affects the remainder bar needs testing.
STILL NEEDS TESTING

EFFECT OF PERCEPTION STAT
It is believed that the quality of the item (+0 +1 +2 +3) is affected by perception (Faerie Apple +3 vs Faerie Apple +1). It may also affect the yield, or number of items you get in each single attempt to get an item (yield of 3x Faerie Apples vs 1x Faerie Apple). The rate at which Perception scales these affects needs testing.
STILL NEEDS TESTING

Edited, May 19th 2011 2:20pm by DeusBahamut
____________________________
Deus Diabolus, Gstar Raw, Besaid Server
Leader of Lifestream LS
http://lifestreamls.enjin.com
#2 Jan 11 2011 at 1:41 PM Rating: Excellent
*
56 posts
**RESERVED**
#3 Jan 11 2011 at 1:41 PM Rating: Excellent
*
56 posts
**RESERVED***
#4 Jan 11 2011 at 4:08 PM Rating: Decent
**
518 posts
I wish I could add here,

but i think until we can get some kind of Log parser or SE just comes out and tells us these things, testing like this is borderline torture.

I can tell you this, STR in the stats does affect DMG, that has been tested and proven. Now to what extent +YSTR=XDMG no idea.
Pre-patch there was no direct correlation between +DEX and a higher hit rate. Even though SE says there is such a correlation, I saw four tests with over 600 samples each that either showed with more DEX came a VERY small increas in ACC, no increase in ACC, or a lose in ACC.

When a parser comes out all this data will be soooo much easier to do.
#5 Jan 11 2011 at 4:25 PM Rating: Excellent
Avatar
**
676 posts
Quote:
Before the late November patch, I did some empirical testing for 5 hours and 400 data points, and noticed a very strong correlation between raising INT and basic THM or CON attack damage. +40 INT points translated to about +20% basic attack damage at level 15 on Green mobs. This should not come as a surprise to anyone. I am trying to locate post-patch tests. Specifics will be linked here.

This information would be a lot more useful if we knew what your INT was beforehand. If you got a 20% increase, but only had about 12 INT to begin with, it's not exactly earth-shattering but if you had 100 INT before, its a whole lot more impressive.

Still, this does seem to be the case. INT, while having some effect, is hardly as amazing as it really should be for nuke damage. I piled everything I had into INT to try and maximise my damage but have come to the realisation that im sure the majority of conjurers have come to. The tiny increase in damage doesn't justify the loss of HP you could be getting by putting those points into VIT. I'm all for hitting hard, but when its a case of beating an enemy with one fewer dart or surviving three more hits, I'll go with the HP every time.

I really hope some solid figures come up to show how this works, because I've lost a little faith in the stats system to be honest.
____________________________

FFXI: Siren Server: Seiowan Lvl 99 WHM, SCH, BLM
FFXIV: Ragnarok Server: Lemuria Glitterhands All Classes 50
#6 Jan 11 2011 at 5:58 PM Rating: Excellent
*
56 posts
@EzellLangor

Agreed. I hope by showing everyone the need here for a log parser contributors would be more motivated to create parsers, to find easy empirical answers to these unknowns.

@Glitterhands

I chose to omit the details of my test because I would like to conform to stricter standards regarding the conditions I gathered the log data. I will check my old spreadsheet and will post more specifics later as an FYI.

I can't tell you how many times I survived with 7 HP so I understand the value of VIT. However there is a point where you just have to accept death every now and then to grind mobs in fewer casts. Or on the other hand to convert more INT to VIT while fighting an HNM you would lose to if you died. With more empirical data we would get the most out of every single attribute point and be confident with it.


Edited, Jan 11th 2011 6:59pm by DeusBahamut
#7cornyboob, Posted: Jan 12 2011 at 12:33 AM, Rating: Sub-Default, (Expand Post) try not to divide by zero lol
#8 Jan 13 2011 at 6:14 PM Rating: Good
*
56 posts
*Bump* Trying to get more visibility over less productive posts
#9 Jan 13 2011 at 11:08 PM Rating: Excellent
***
1,335 posts
Deus, I believe this thread may have some helpful info for ya.
____________________________
(clickable sig, see this thread for info)
#10 Jan 13 2011 at 11:56 PM Rating: Good
Scholar
*
92 posts
Someone has already released a parser, it's called Crabby.
#11 Jan 14 2011 at 3:53 PM Rating: Decent
*
56 posts
Awesome links, thanks guys, I'll look into it.
#12 Jan 14 2011 at 9:39 PM Rating: Decent
*
127 posts
The stat caps thread only matters for HP/MP gains from VIT/MND.

A very easy way to test this is to cap out your MND MP cap, and then cast Cure on someone. Now add 2 points to MND. You just healed for more.

Fatigue comes in "levels", and the further you venture into the fatigue, the more drastic of an effect it has on your SP gain.

Level 1: -10% SP
Level 2: -20% SP (I think.)

It seems to go down in increments of 10, but I personally haven't done further tests. You can easily test this by leveling a caster a bunch, then equip a shield. Once your caster class gets into fatigue, the shield will receive full SP and the caster class will receive diminished returns. Compare the two numbers to get the exact SP loss.

If you actually go further with this, I can contribute more.

Edited, Jan 14th 2011 9:41pm by Crevox
#13 Jan 15 2011 at 9:54 PM Rating: Decent
*
56 posts
Thanks, yes I will include more details for fatigue. You are correct it is -10% each "bonus rank". The most interesting aspect of the system is how the cool-off period works. That will be a very useful test.
#14 Jan 16 2011 at 10:15 PM Rating: Decent
*
56 posts
I've confirmed that Crabby is EXACTLY what I was looking for. With this program, testing should be extremely easy. All you do is download the main package, and run the exe after you have logged into FFXIV in Window Mode. You may need to double click the icon in the lower left of the window to sync up. Once synced you have granular control over creating individual feeds for all the types of log entries corresponding to what you "Show" and "Hide" in the "Configuration > Chat Log > Settings" window. The feed can timestamp data and have an ID stamp as to what type of log entry it is. This data can be copy pasted out. I will work on creating perl scripts that evaluate parameters such as average non-critical and average critical damage.
#15 Jan 17 2011 at 7:04 PM Rating: Decent
****
9,526 posts
If you do testing I would like to see some food tested.... I know it is complicated but would like to see how much effect non-tasty food has versus tasty, delicious and divine food
____________________________
lolgaxe wrote:
When it comes to sitting around not doing anything for long periods of time, only being active for short windows, and marginal changes and sidegrades I'd say FFXI players were the perfect choice for politicians.


#16 Jan 17 2011 at 8:34 PM Rating: Good
Sage
**
447 posts
Is this thread meant to be battle centric, or will it include craft effects? If not, can you or someone else create a similarly dedicated thread on attribute/gear effects toward crafting?
____________________________
Djigga, please. Highland Hyurs can't jump.

#17 Jan 18 2011 at 12:11 PM Rating: Excellent
*
56 posts
I have nearly finished an INT and PTY test. All while getting 8k SP/hr. I will post the results within a couple days. I have been thinking about ways to test food and craft. I'm not too familiar with how food works as I've never used it. If you could point me to a good thread.

As far as crafting goes, there is a probability whether the light turns white, yellow, red, or flicker. And at each of those stages there is a probability of success or failure for standard, rapid, and bold. Variables are stat attributes, elemental attributes, craftsmanship+, control+, and all the selectable bonus abilities. Unfortunately the logs do not tell you the color of the light, so testing via parsing logs may not be as straightforward.

For now, I'd like to focus on battle mechanics. I also hope that once I post and everyone can see how easy it is to conform to what I've done without spending much time analyzing the data, that I will have people contribute their own test results to corroborate what I've done or test something new.

Edited, Jan 18th 2011 1:52pm by DeusBahamut
#18 Jan 18 2011 at 12:53 PM Rating: Decent
****
9,526 posts
Best food info I have found in english is on Baxter Mergous' lodestone blog:

http://lodestone.finalfantasyxiv.com/rc/diary/entry?e=102969
____________________________
lolgaxe wrote:
When it comes to sitting around not doing anything for long periods of time, only being active for short windows, and marginal changes and sidegrades I'd say FFXI players were the perfect choice for politicians.


#19 Jan 21 2011 at 1:08 PM Rating: Decent
*
56 posts
I'd like to update that I completed testing INT and PTY with 100 mobs worth of data, and have drawn some conclusions with completed perl scripts, however, I am doing further testing at the next job rank, and have also hit phys level 50, so I am continuing further tests before I post all the results. I am testing multiple parameters and variables. Expect the results next week.
#20 Feb 07 2011 at 8:41 PM Rating: Excellent
*
56 posts
I have published my first test results here:
The Effects of INT PTY and RANK on Phantom Dart and Damnation:
http://ffxiv.zam.com/forum.html?forum=152&mid=1297132715260171115&page=1#m1297132715260171115

Stay Tuned for my next post about the effects of INT PTY and RANK on Scourge, Scourge II, and Shadowsear: Avg Unresisted DMG, Avg 1/2 Resisted DMG, Avg 1/4 Resisted DMG, and Resist Rate
____________________________
Deus Diabolus, Gstar Raw, Besaid Server
Leader of Lifestream LS
http://lifestreamls.enjin.com
#21 Feb 08 2011 at 9:25 AM Rating: Decent
Scholar
35 posts
Crevox wrote:
The stat caps thread only matters for HP/MP gains from VIT/MND.

A very easy way to test this is to cap out your MND MP cap, and then cast Cure on someone. Now add 2 points to MND. You just healed for more.


Tested this the other day. I had an r16 CON, and while MP stopped increasing at the listed cap, my Cure kept increasing. I got up to 130MND before I ran out of points and it was still going up, which is more than twice the MP cap. Considering 130MND was still giving effects at r16, I doubt this effect of MND caps. It also seemed to steadily increase about 1 point of healing for every 2-3 points of MND. So it's not really worth it to go past the MP cap, but there it is.

About a day after that, I did some basic STR tests, but it did seem like the attack bonus from STR capped out around the same area as the MP/HP. I was on r41 ARC, and saw increases from 100-140 STR, but didn't see any significant increases from 140-180 STR. I didn't test long enough to gather any rate of increase stats, just enough to see if the damage was trending upward.

I'd really like to see a lot more DEX testing done. All the more rigorous testing I've seen seems to suggest it doesn't do much of anything, which is unfortunate.
#22 Feb 09 2011 at 12:09 AM Rating: Good
*
56 posts
I have confirmed the soft cap is higher than what was proposed in the Stat Caps thread. It's possible that there might be NO soft cap, or a soft cap on certain functions of the stats on a case by case basis. The effect of MND on Cure is an easy test if anyone would like to contribute. The HARD cap on all stats is 184. You CANNOT raise higher than 184.

Edited, Feb 9th 2011 1:09am by DeusBahamut
____________________________
Deus Diabolus, Gstar Raw, Besaid Server
Leader of Lifestream LS
http://lifestreamls.enjin.com
#23 Feb 11 2011 at 5:48 AM Rating: Decent
Scholar
**
844 posts
Quote:
-----DURABILITY
Group D is tied to item durability; a broken weapon of 0% durability will result in a damage penalty.
Conversely, when an armor part breaks, if it is hit again, it will result in increased damage (confirmed).
It is unclear if durability only affects Attack or Defense based on the % or based on the status change of the color of the indicator, or the status color change of the equipment box on the Gear screen.


Quote:
Q.
What are the Gear Damage and Heavy Gear Damage statuses I sometimes see?

A.
These statuses reflect the current condition of one or more pieces of equipped gear. When an item's condition falls below 50%, a character will be afflicted with the Gear Damage status. Should further damage be sustained without repairing, the Heavy Gear Damage status will appear, indicating that an equipped item's condition has fallen below 20%. As with other statuses, the icons will appear in the upper portion of the screen. Currently, gear stats will decrease at a certain rate up until the Heavy Gear Damage status appears, after which point they will decline much more rapidly.


Seems to point at durability only affecting gear stats if < 50%, at least for B and C purposes.
____________________________
See your face upon the clear water. How dirty! Come! Wash your face!
#24 Feb 20 2011 at 2:44 PM Rating: Good
*
56 posts
Yeah good catch, it wasn't until that update I was sure it was not 60%, when the item's bar turns yellow, but the outline around the item is not yellow yet. I guess that is just a warning. Good to have a buffer anyways. A message will come up in the log at 60% saying the item is damaged. This is how I knew when to throw out some data.

Edited, Feb 28th 2011 6:56pm by DeusBahamut
____________________________
Deus Diabolus, Gstar Raw, Besaid Server
Leader of Lifestream LS
http://lifestreamls.enjin.com
#25 Mar 09 2011 at 1:14 PM Rating: Good
*
56 posts
The entire battle system will be redesigned. Finally, Square Enix got the message:
FFXIVCore Exclusive Naoki Yoshida Interview


On the topic of monster attribute changes and possibly player attribute changes, can you go more into detail on this? A lot of players feel everything is still a bit out of whack when it comes to the current attributes like dex barely affecting Accuracy or Shell barely reducing an Imps Blizzard damage.


Our Lead Battle Director, Matsui-san, is currently undergoing--they're basically going to change the equations for how stuff is calculated, all of these attributes. As they balance each of the classes with the new balance system, they're going to go through and check each action. And then see how that needs to be adjusted, and reassign each attribute as they go along.

Since it's going to be such a huge undertaking, it's going to happen in steps--it's not all going to change in the next patch, but over the next few patches we hope to change it. But we want to reassess everything, so it's not like we're going to be looking at one broken thing, we're going to be changing pretty much everything.

One of the reasons that you have some of the things not affecting something at all--like Shell not working on a lot of spells--is because right now a lot of the calculations for that stuff is too complex. Because it's so complex, it's difficult to balance. They want to make it simpler, not just so the players know what's going on--having to have some super program going on in the background to understand the calculations--but also so when the devs do balancing, it's easier for them to balance.
____________________________
Deus Diabolus, Gstar Raw, Besaid Server
Leader of Lifestream LS
http://lifestreamls.enjin.com
#26 Mar 09 2011 at 2:20 PM Rating: Decent
Scholar
**
844 posts
Awesome news IMO
____________________________
See your face upon the clear water. How dirty! Come! Wash your face!
#27 Mar 21 2011 at 3:48 PM Rating: Good
*
56 posts
Major updates to the post, and a NEW TEST CASE was completed. See Updates section for details.
____________________________
Deus Diabolus, Gstar Raw, Besaid Server
Leader of Lifestream LS
http://lifestreamls.enjin.com
#28 Apr 08 2011 at 12:50 PM Rating: Good
*
56 posts
Recovered enough energy to post my most influential conclusions directly on the FFXIV forums, adding dev tags. Hopefully they will respond directly. I need your help to rate up the post and respond to it to get their attention:

http://forum.square-enix.com/ffxiv/threads/6355-dev1027-dev1032-Status-Allocation-System-Damage-Calculation-Broken-Unbalanced
____________________________
Deus Diabolus, Gstar Raw, Besaid Server
Leader of Lifestream LS
http://lifestreamls.enjin.com
#29 Apr 25 2011 at 4:18 PM Rating: Good
*
56 posts
UPDDATE--TEST RESULTS: Log Parse Results! Effect of New NM Gear Stat Bonus on 30MRD

The effect of the New NM Gear's STR and Attack on average damage and average critical damage on 30 MRD.
____________________________
Deus Diabolus, Gstar Raw, Besaid Server
Leader of Lifestream LS
http://lifestreamls.enjin.com
#30 May 01 2011 at 12:52 AM Rating: Good
*
56 posts
Completed testing on STR and DEX. Found results with STR consistent to INT. Found results with DEX such that DEX lowered Miss Rate. Thus inconsistent with PTY. See post for details.

http://ffxiv.zam.com/forum.html?forum=152;mid=1304231587296262403;page=1#m1304231587296262403

Edited, May 1st 2011 2:56am by DeusBahamut
____________________________
Deus Diabolus, Gstar Raw, Besaid Server
Leader of Lifestream LS
http://lifestreamls.enjin.com
#31 May 12 2011 at 1:15 PM Rating: Good
*
56 posts
Added test results from Ami Trubbles of Besaid. Effect of MND and Magic Potency on the effect of Cure II.
____________________________
Deus Diabolus, Gstar Raw, Besaid Server
Leader of Lifestream LS
http://lifestreamls.enjin.com
#32 May 16 2011 at 2:35 PM Rating: Good
*
56 posts
5/16 TEST RESULTS: The Effect of a Heavily Damaged Weapon on Conjurer Skills: See Advanced Damage Formula Section. Test with Deus Bahamut and Iris Akiha of Lifestream.

5/16 TEST RESULTS: The Effect of Output on Gathering: See Gathering section. By Ami Trubble of Lifestream.

5/16 POST UPDATE
--Added section on Gathering Analysis
--Removed sections on Major Game updates and Current Projects
--Condensed and updated all existing sections for accuracy
____________________________
Deus Diabolus, Gstar Raw, Besaid Server
Leader of Lifestream LS
http://lifestreamls.enjin.com
#33 May 19 2011 at 12:21 PM Rating: Good
*
56 posts
5/19 FFXIV FORUM UPDATE: Official FFXIV Forums: Dev Team Feedback and My Response

The Lead Battle Planner has responded to address the concerns laid out in these tests. I am satisfied with their responses. See the link above for the correspondence.
____________________________
Deus Diabolus, Gstar Raw, Besaid Server
Leader of Lifestream LS
http://lifestreamls.enjin.com
This forum is read only
This Forum is Read Only!
Recent Visitors: 29 All times are in CST
Anonymous Guests (29)