Darkrasp

Darkrasp's Blog 9/4/2017

29 posts in this topic

Good evening,

Alright, so it's been a few weeks since my last post.  My backpacking trip was pretty good.  Rained a lot, but we still had a good time.  Nice to get away from the 9 to 5 anyways.  Since getting back we've been pounding out a lot of work.  We brought in another ten beta testers and got rid of a couple inactive ones.  The new testers are settling in well and churning out a lot of content reports.  More than enough to keep us busy for the time being.  We now have three scripters working away on writing quest and event scripts, and plans to bring on a couple more soon-ish.  The ones we have now have identified a few things they want added to the scripting system so Asura is working on those changes.

I've spent a lot of time with the new testers getting them set up with what they need, but I have to give credit to the existing beta team, who have been amazing with helping the new people get comfortable and productive.

For myself, since coming back I've submitted four core patches and a couple database updates.  The new reports coming in are mostly database related so I've been tackling the loot-related issues.  If the backlog gets too big I'll assist with some of the other ones as well, but I'm hoping that Nogar will be able to push through those and leave me to core work.

So, my core patches for this week have been a patch that fixed a problem with the level of Guardians when they get summoned, a fix for an issue with spell damage penalties, the big absorb patch that fixed four issues with absorbs, and implementation of the Daze mechanic when getting hit from behind.  Also a one-line fix to correct a typo that prevented mobs from getting tagged, but since that literally took 2 minutes I don't really count that as a fix.

First thing, Guardians.  This is a type of summoned creature that has a specific duration and "belongs" to the summoner.  The limited duration distinguishes it from a pet, and the ownership distinguishes it from a "wild" summon.  A "wild" summon is a creature that gets summoned and then behaves entirely independently with no connection to the summoner.  Some examples of Guardians are like.. the Compact Harvest Reapers summoned by the goblins in the foundry in Deadmines, or the Frost Spectres summoned by the last boss in RFD.  Players can also summon guardians, typically through using items like the Dog Whistle from Houndmaster Locksey, the Mechanical Yeti, or the Barov Peasant Caller.

These spells work differently for players and creatures.  A creature-summoned guardian typically matches the level of the creature that summoned it (with some exceptions - Morganth's minions are actually 3 levels above him), while guardians summoned by players also match the summoner level, but only to a maximum, defined by the level of the item itself.  So a Scarlet Tracking Hound will max out at level 35, a Mechanical Yeti will max out at level 58, etc.  I wrote up a patch that does this, Asura took it and made a few modifications, and I think it's done now.

The spell damage penalty patch has to do with spells below level 20 getting their spelldamage coefficient scaled down.  There was an issue with some new code that caused ALL damage to get scaled down instead of just the bonus damage.  It's a little more complicated than you might think because if you have mods that increase spell damage by a percentage, that applies to both the initial and bonus damage, where this penalty only applies to the bonus.  Anyways, little bit of reordering required but nothing major.

The absorb patch was a little more complicated because it had to fix four problems with absorbs.  First we had to research the actual correct implementation of them.  There was some argument over whether they were a simple First-On-First-Off system or if they would eat school-specific buffs before moving on to general buffs.  It was proven through video evidence that First-On-First-Off is the correct method (excluding Mana Shield - more on this later).  Imbaslap found a Sapphiron video from a mage perspective where he pops Ice Barrier, then Frost Ward, and the Ice Barrier is removed before the Frost Ward when the only damage being taken is from the frost aura.  Our system was already set up to work that way, so that was good, no problem there.  However, a typo and a missing increment meant that popping Ice Barrier put the server into an infinite loop, effectively crashing it.  That was the easiest fix.

Second problem was the removal of absorb auras when they were used up.  Basically we keep a running tally of how many points of absorb are left in each buff you have, but we weren't actually removing the aura when they ran out.  They'd stop working, and you would take damage, but it would LOOK like you still had Ice Barrier, or Power Word: Shield, or whatever, on your character.  Fixing that involved building a list of each "empty" absorb and then removing all the used up auras at the end of the damage calculation.

Third problem was Mana Shield, which is annoying because it's a special snowflake.  In Vanilla, Mana Shield was only capable of absorbing physical damage, and after some patch, I think 1.11, it would always be the LAST absorb used.  We had some handling for this but it was incorrect.  Basically the absorb would check to see if you had anything other than Mana Shield and take from that, then it would take from Mana Shield.  Which sounds right, except that it was missing the part that zeroed out what was left over, so the same amount of damage that got removed by Ice Barrier would ALSO get eaten by Mana Shield.  Took a little while to actually figure that out, but it ended up only needing one line to fix.

Last problem was that due to a typo in some code that was shuffled around recently, physical attacks would actually run through the absorption code twice.  While they wouldn't do any additional damage to your health, they ate through shields twice as fast.  That one was also a one-line fix that just took a little while to find.

Final fix was the implementation of Daze when being hit from behind by mobs.  Pretty interesting formula for this one.  Turns out Ascent already had a function for it, but it was kinda sloppy, the code was dated, and it wasn't actually called from anywhere else so it was just sitting around unused.  I tuned it up to make it more efficient and self-contained and put it into the autoattack code.  Tested and working beautifully.

 

Alright, so that was my week.  I just wanted to point out something quickly that came up in conversation with, and while getting reports from, our new beta testers.  We get asked a lot about the loot tables and drop rates.  I spend quite a bit of time checking over them, largely because a primary source used by people is the vanilladb page, which is notoriously unreliable.  I've talked before about how they show clam meat dropping from all kinds of creatures, among other weirdnesses caused by incorrect reporting from the wowhead client if you have a creature targeted while you open a container.  That's one thing and most people know about that by now.

The bigger problem with vanilladb is their quest item drop rates, and I think this has a lot to do with why levelling on private servers is so trivial compared to how it was on retail.  I'm going to give just a few examples of this so you can see what I mean in a moment, but basically we get a question where they might think a drop rate is set too low, or that they're only getting 25% on something when vanilladb says it should be 80%.

First off, vanilladb has a thing about setting virtually all quest item drops to either 80% or 100%.  Our quest drop database was parsed back in TBC and I've gone through and manually adjusted a lot of it, taking data from archived pages, comments, and in some cases, common sense.  Very few quest items have a high drop rate, especially in Vanilla.  I know that in some cases the drop rates are reported low due to items not dropping for people who aren't on the quest, but at the same time, people who aren't on the quest aren't typically grinding those mobs, either.  Not while levelling, anyways, obviously high level mobs are a different story, but we can look at STV as a pretty good example of some issues.  I'll show two of them so you can see what I mean.

First up, Split Bone Necklaces.  We can see here the vanilladb page showing the drop rates for them:

1.png.7b618babbcf65a780d6659a49e6b301f.p

Meanwhile, the comment section from Wowhead, dating back to TBC, has the following remarks:

2.png.893fe4befc9020f2e04fd34209ae6a33.p

Forgive the sloppy red lines, I'm using the same images I was using earlier to explain this in our beta chat.  I think you get the point though.  Clearly it should take quite a while to grind these items out, it shouldn't be nearly a lock to get one off every troll.  Our database has this item set to roughly 20%, in accordance with our old parse, tidied up by myself to round off the 18.52142 etc.

 

Next up, we'll have a look at the quest Mok'thardin's Enchantment:

Screenshot_5.png.bd4f2655fd1325300525205

Hm, from that quest text, it seems like they should be pretty rare and at least somewhat difficult to get.  Let's have a look at the comments and see what the experience of players from retail was on getting them:

Screenshot_3.png.9129c99f65b43099c4c26f9

Yikes.  I bet vanilladb has the droprate set pretty low, then.  Let's check it out:

Screenshot_6.png.97052b1b02f52f4e7e6090f

100%.  Well, I guess that certainly simplifies the process and gets you levelling faster.  We're going with a more Blizzlike 3%.

 

So yeah.  The point is that we've gone to great lengths to really recreate the experience of playing Vanilla WoW here.  Yes, it has taken us a long time to get this far, but we're really not overlooking any details.  We're pushing through reports and writing content every day.  The new testers are enthusiastic and busy.

Alright, it's started to thunderstorm here and I may lose power.  I want to get this posted before that happens so I'll wrap this up right now.  My next blog post will be on Monday, September 18th.  As always, feel free to leave comments or ask questions below.  Talk to you soon, and thanks again for your continued interest in Crestfall.

Share this post


Link to post
Share on other sites

This is awesome, I love that the drop rates are being closely looked at! 

80% drop chance on quest items?! Ha! Give me that real Vanilla grind. I look forward to spending many hours trying to get those boar livers, and various other items.

Share this post


Link to post
Share on other sites

ru translate.

Если в кратце, то один из одминов пишет, что он и его молодая команда много работает над тем что бы эллизиум был на 100% ваниллой и как много сил они вкладывают.
Из изменений:
-Проведена большая работа над "гвардами" и теперь они полноценно работают (на пример: гварды вызванные гоблинами в литейном цехе в Deadmines и т.д.) 
-Так же завезли работу абсорб щитов (а-ля мана-щилд у магов и т.д.)
-Корректная работа "ошеломления" (daze)
И много пишет о шансе дропа. (хочет что бы как на вовхеде всё) и жалуется на не достоверную инфу с других сайтов посвященным ваниле. Просит его читать и задавать вопросы. В следующий эфир выйдет 18 сентября. вроде всё)

Share this post


Link to post
Share on other sites
7 hours ago, Gwenet said:

Если в кратце, то какой-то из одминов пишет, что он и его молодая команда много работает над тем что бы эллизиум был на 100% ваниллой и как много сил они вкладывают.
Из изменений:
-Проведена большая работа над "гвардами" и теперь они полноценно работают (на пример: гварды вызванные гоблинами в литейном цехе в Deadmines и т.д.) 
-Так же завезли работу абсорб щитов (а-ля мана-щилд у магов и т.д.)
-Корректная работа "ошеломления" (daze)
И много пишет о шансе дропа. (хочет что бы как на вовхеде всё) и жалуется на не достоверную инфу с других сайтов посвященным ваниле. Просит его читать и задавать вопросы. В следующий эфир выйдет 18 сентября. вроде всё)

agreed

Share this post


Link to post
Share on other sites
1 hour ago, Gwenet said:

Valaquenta, ru translate.

They deleted my joke :s.

Are people not supposed to write in english here by the way?

Share this post


Link to post
Share on other sites
On 9/5/2017 at 4:03 PM, Darkrasp said:

Forgive the sloppy red lines, I'm using the same images I was using earlier to explain this in our beta chat.  I think you get the point though.  Clearly it should take quite a while to grind these items out, it shouldn't be nearly a lock to get one off every troll.  Our database has this item set to roughly 20%, in accordance with our old parse, tidied up by myself to round off the 18.52142 etc.

Instead of doing stuffs based on assumptions from some random wowhead comment dating during Wrath of the Lich King and The Burning Crusade period, start investigating things before you set some slippery slope numbers .

https://web.archive.org/web/20060319023653/http://thottbot.com:80/?i=11072

https://web.archive.org/web/20060508032157/http://www.thottbot.com:80/?qu=598

These comments are written 6 months before 8 May 2006 , which mean these are from December 2005

Quote

 

Drop rate
Score 0     Vote: [-] [+] by Fleaflicker, 6.4 months ago

The drop rate is BS.  They only drop if you have the quest.  Other quests have you to kill these pests.  Thus, the 10% drop rate is more like 30% or better.

 

Quote

 

?!?!?!?!
Score 0     Vote: [-] [+] by hijklm, 5.6 months ago

WTF are you bitching about? It took me 30 min tops to do this quest. Almost every troll dropped a necklace.

 

March 19 2006

https://web.archive.org/web/20050128041955/http://wow.allakhazam.com:80/db/item.html?witem=3916

Jan 28 2005

Enjoy your droprate chance with accurate numbers.

Even the 2nd item you mention "we are going to make more Blizzlike 3%" is completely wrong based on assumption.

https://web.archive.org/web/20070707141033/http://www.thottbot.com:80/i3862

Jul 7 2007
 

Written 1.2 years before Jul 7 2007 , it means the comment is written during May 7 2006

Quote

 

by pwoefkewfkpok, 1.2 years ago
Last edit: 1.2 years ago

Bear in mind that the droprate you see here is not correct, as these gorillas get killed by people who don't have the quest, and hence, the gorillas will never drop the quest items for them.

 

Quote

 

Drop Rate
Score 1.3     Vote: [-] [+] by onatoo, 1.2 years ago

I just did this quest in less than 10 minutes, took about 10 to kill!

 

The whole Wall of text written goes Titanic if things are not done properly with quality.

/Kind regards Killerduki

 

Share this post


Link to post
Share on other sites
10 hours ago, killerduki said:

<snip>


Irony:  Using old comments as an argument that old comments are unreliable. 

Obviously just because some guy decided to post that he was exceptionally lucky or exceptionally unlucky isn't definite proof one way or the other.  Blizzard never released any official drop rate info for these items, and the rates for many of them changed over time, so we have to use common sense and multiple sources to get as close as possible to what was probably correct, or at least what is reasonable for the quest in terms of it's reward and the time we expect a player to spend on it in order to obtain that reward.

If I have to make an educated guess, I'll happily err on the side of having lower drop rates and making things harder, and I won't apologize for that.

If you're going to make arguments though, try to be a little less of an obvious cherry-picker when compiling your "evidences".

From the same page you linked:

Screenshot_7.thumb.png.e5dd475560c81aeea1457afc4c98a51d.png

Share this post


Link to post
Share on other sites

The cherry picking comments are minority , if you read all of the comments, you will find out 70% of them got it within few kills.

image.thumb.png.2a7a55a0ffbd67b361ab6df63bb5a980.png

/Kind regards Killerduki

image.png

Share this post


Link to post
Share on other sites
4 hours ago, killerduki said:

groogle groggle

TBC was released Jan 2007, the Thottbot cache timeframe is on is set to July 2007, 6 months after TBC was released. The comments are timestamped at 4.5 to 3.7 months ago from the stated cache time, so they were still after TBC launch, most of the comments on the page are probably from about feb-march-april 2007. Also note the titles, content, and scores on the comments in this section, most are trolly and all are less than a 0.5:1 ratio in votes, most likely because vanilla players saw them later on while leveling an alt and gave it negs. My point is - don't call him out for linking late comments when you're doing the same thing.

If you're going to nitpick, pay a lot more attention to details. an effective cherrypick isn't about just taking things that LOOK like evidence, rather things that you can confirm are. 

Even in Darkrasp's pic, you can see that they were still taking 70 kills to get in TBC, yet some are still getting them in three tries. It's luck and RNG.

 

At this point, the cherry picking comments are a majority. All in all you are either testing Darkrasp, trolling, or you're just generally unobservant, most likely a mix of the two ladder.

 

/Kind regards Riplip

Share this post


Link to post
Share on other sites

How about instead of "arguing" and saying, "oh my Thottbot is better etc" ... Use both? But i'd only use Thottbot if the archive allows you to go before 2007, as RipLip stated. Judge the comments by their "last stated" and compared to the date that Thottbot is stuck at.

In the end we are all looking at somewhat biased comments and possibly manipulated data. In the end just go with what feels right. 

Share this post


Link to post
Share on other sites
On 9/7/2017 at 7:45 AM, riplip said:

TBC was released Jan 2007, the Thottbot cache timeframe is on is set to July 2007, 6 months after TBC was released. The comments are timestamped at 4.5 to 3.7 months ago from the stated cache time, so they were still after TBC launch, most of the comments on the page are probably from about feb-march-april 2007. Also note the titles, content, and scores on the comments in this section, most are trolly and all are less than a 0.5:1 ratio in votes, most likely because vanilla players saw them later on while leveling an alt and gave it negs. My point is - don't call him out for linking late comments when you're doing the same thing.

If you're going to nitpick, pay a lot more attention to details. an effective cherrypick isn't about just taking things that LOOK like evidence, rather things that you can confirm are. 

Even in Darkrasp's pic, you can see that they were still taking 70 kills to get in TBC, yet some are still getting them in three tries. It's luck and RNG.

 

At this point, the cherry picking comments are a majority. All in all you are either testing Darkrasp, trolling, or you're just generally unobservant, most likely a mix of the two ladder.

 

/Kind regards Riplip

 

On 9/7/2017 at 2:33 PM, Eyedie1 said:

How about instead of "arguing" and saying, "oh my Thottbot is better etc" ... Use both? But i'd only use Thottbot if the archive allows you to go before 2007, as RipLip stated. Judge the comments by their "last stated" and compared to the date that Thottbot is stuck at.

In the end we are all looking at somewhat biased comments and possibly manipulated data. In the end just go with what feels right. 

image.thumb.png.74b20decb69816180b4d4db2ed3b8e90.pngimage.thumb.png.104bed4e3d2e9d68ef37dfd73d9a1aed.png

Yeah , there is your 1.2 Years ago before Jul 07 2007 , this mean they are from May 7 2006.

You need Goggles next time and check links before you talk, i have clearly posted there links which include posts from 2006 (before TBC) compare to Darkrasp Wrath of the Lich king links.

I made my statement, the comments Darkrasp pointed are minority there , 70% of the comments says it dropped with few kills.

The yellow marked comment have the highest vote rate as you can see with score 4.4 and 1.3 , which mean everyone approve it.

p.s i like how Crestfall fanboys cry and thrown personal insults against me in their Crestfall website when i disapprove something incorrect with evidence.

http://forums.crestfall-gaming.com/index.php?/topic/2334-darkrasps-blog-942017/&do=findComment&comment=35677

Such professionalism by not censoring those who violate TOU, it gives such great negative and terrible impact to Elysium.

/Kind regards Killerduki

Share this post


Link to post
Share on other sites

Duk is there something terrible bothering you deep down? I don't mind you sharing. You used to be a "crestfall fanboy" but now you  not only bash Crestfall, who are making things like this quest's droprate better (not perfect, but better) than they already were, but you are also bashing elysium for a reason that is unbeknownst to me, maybe you seek "professionalism" from a volunteer project that is not even a business. . Is it because you got rejected from being a beta tester, or is it a bit more complicated?

Share this post


Link to post
Share on other sites
55 minutes ago, riplip said:

Duk is there something terrible bothering you deep down? I don't mind you sharing. You used to be a "crestfall fanboy" but now you  not only bash Crestfall, who are making things like this quest's droprate better (not perfect, but better) than they already were, but you are also bashing elysium for a reason that is unbeknownst to me, maybe you seek "professionalism" from a volunteer project that is not even a business. . Is it because you got rejected from being a beta tester, or is it a bit more complicated?

I am not sure if you either dreaming or you are simple delusional , i was never Crestfall fanboy and will never be.

How they are making things like droprate better outside of Blizzlike? It can be either Blizzlike or either Modified , no other option.

When i say unprofessional , i mean that, allowing kids insulting publicly is disaster for anyone who try to build project regardless if it is volunteer or not , it will simple go Titanic either way for that.

This is one way minded , where "you have to accept when they insult you without getting offended, neither them getting banned for doing it" , i am sorry but this doesn't work that way. Rules are rules and should be applied to everyone , not to "fanboys only" .

I never applied for being a beta tester and will never even bother doing that .

Now i'd like if you stick to the Topic instead to "Killerduki" , this Topic is not "Killerduki" , it is "Blog with discussion and wall of text about Droprate and stuffs related to game" .  This mean ,you derail this thread with "offtopic" discussion and trolling only.

If you have at least 00000.1% IQ then you will understand what i say.

/Kind regards Killerduki

Share this post


Link to post
Share on other sites
Quote

allowing kids insulting publicly is disaster for anyone who try to build project regardless if it is volunteer or not

You.

 

 

you're making a huge deal out of a single quest, which is an improved blizzlike version of what it was before. I don't see why you're crying so hard about it. 1,408 posts and a crestfall logo, and you still don't get it.

Share this post


Link to post
Share on other sites
5 minutes ago, riplip said:

You.

 

 

you're making a huge deal out of a single quest, which is an improved blizzlike version of what it was before. I don't see why you're crying so hard about it. 1,408 posts and a crestfall logo, and you still don't get it.

I am not making a huge deal , i am just pointing out the whole wall of text goes Titanic when evidence is presented as opposite of what is claimed.

Blizzlike or Modified , can't be 3rd way.

/Kind regards Killerduki

Share this post


Link to post
Share on other sites

What 3rd way? It's either blizzlike or an attempt at blizzlike, or else nobody will buy it.  No reason to bash an attempt at blizzlike, they are trying to bring nostalgia and it seems like you have malice in your intentions, else you wouldn't be trying to "sink the titanic" and instead be actually providing constructive criticism and productive posts.

Share this post


Link to post
Share on other sites

Makes sense to me duki. Thanks for pointing that out to the developers, hopefully they will get it right.

Share this post


Link to post
Share on other sites

There's a big difference between bashing it, and giving advice. No one knows anymore what Blizzlike is with many things. It's getting as close as possible. Thottbot comments were just as unreliable as Wiki or Wowhead because anyone can write them or upvote them. Even on that Gorrila page themselves, the comments are literally divided 50/50 between forever and one kill. Anyone can get lucky, some aren't. Some even have gained like 4-5 greens and a blue from farming for that quest. In the end it's an estimation which 3-5% seems close enough. 

I did the quest 4 times in vanilla and my brother and father probably attempted it that many times aswell. They never did the quest because they believed that the shit never dropped and would rather grind (too many people) or do another quest/dungeon. 

So no there is no point autistic picking through comments to fit your cause to prove that "This guy IS wrong." No one is ever 100% sure as there is no Blizzard Database for it.

Share this post


Link to post
Share on other sites
20 hours ago, Eyedie1 said:

There's a big difference between bashing it, and giving advice. No one knows anymore what Blizzlike is with many things. It's getting as close as possible. Thottbot comments were just as unreliable as Wiki or Wowhead because anyone can write them or upvote them. Even on that Gorrila page themselves, the comments are literally divided 50/50 between forever and one kill. Anyone can get lucky, some aren't. Some even have gained like 4-5 greens and a blue from farming for that quest. In the end it's an estimation which 3-5% seems close enough. 

I did the quest 4 times in vanilla and my brother and father probably attempted it that many times aswell. They never did the quest because they believed that the shit never dropped and would rather grind (too many people) or do another quest/dungeon. 

So no there is no point autistic picking through comments to fit your cause to prove that "This guy IS wrong." No one is ever 100% sure as there is no Blizzard Database for it.

There is not "Autistic picking" through comments , because your logic goes against Darkrasp whose based knowledge was taken from Wrath of the Lich king and TBC comments.

I chose to agree with the most reliable comments who was older , majority 70% of them , upvoted with rating 1.1 and 4.1 (highest score) compare to other comments who had barely 1% rating or less and was minority.

When you create things, you want them to base according to Blizzlike , outside of that , the cherry picking and deciding 3% is modified and non blizzlike according to self believes denying the evidence .

Speaking out of memory , yes i also did personally this quest and it dropped on 1st kill , so this kind of Logic goes Titanic.

/Kind regards Killerduki

Share this post


Link to post
Share on other sites
19 hours ago, Eyedie1 said:

There's a big difference between bashing it, and giving advice. No one knows anymore what Blizzlike is with many things. It's getting as close as possible. Thottbot comments were just as unreliable as Wiki or Wowhead because anyone can write them or upvote them. Even on that Gorrila page themselves, the comments are literally divided 50/50 between forever and one kill. Anyone can get lucky, some aren't. Some even have gained like 4-5 greens and a blue from farming for that quest. In the end it's an estimation which 3-5% seems close enough. 

I did the quest 4 times in vanilla and my brother and father probably attempted it that many times aswell. They never did the quest because they believed that the shit never dropped and would rather grind (too many people) or do another quest/dungeon. 

So no there is no point autistic picking through comments to fit your cause to prove that "This guy IS wrong." No one is ever 100% sure as there is no Blizzard Database for it.

I support this message.

Share this post


Link to post
Share on other sites

Guys just ignore killerduki. The rest of the people here usually do.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now