In a time where far too many people are being laid off, it’s nice to get a bit of good news for a change.  AR tech start-up Magic Leap, have had a bit of a rocky year cumulating in the announcement they would be laying off half of their employees.  Thankfully it was just announced that they received a $350M investment and those layoffs have been cancelled.

Details of the new funding from SiliconAngle:

Troubled augmented reality company Magic Leap Inc. has raised $350 million in new funding, according to an internal memo to employees obtained by The Information.

Details on the round, which depending on its form should be a late-stage Series F, are somewhat vague. The investors are described only as current and new investors with the suggestion that a key healthcare company may have been involved.

Although the fundraising itself is surprising, it was good news for employees as Magic Leap withdrew a notice to terminate about 1,000 staff issued in April — meaning that staff on the chopping block will now keep their jobs. Notably, at that time, Magic Leap said it was in the process of negotiating “revenue-generating partnerships.”

The news that a healthcare company may have invested in the company follows Magic Leap’s announcement in December that it was shifting to enterprise augmented reality after failing in the consumer market. The company’s first product, the $2,295 Magic Leap Creator Edition, sold only 6,000 units, a huge disappointment give that the company predicted sales of 100,000 units.

It is nice to see that Magic Leap lives to fight another day, although it’s less and less likely we will see Magic Leap in game development with the shift towards enterprise computing, where they will compete directly with Microsoft’s HoloLens.  More details of the investment and Magic Leaps rocky recent history check out the video below.

GameDev News VR

May 22, 2020 The giveaway has now ended and we’ve dispatched the free code to the lucky winners. Check your email inbox to see if it’s you!

Pascal’s Wager achieved what we all thought was impossible: making Dark Souls or Bloodborne work on mobile. And it achieved it with aplomb, providing a visually-stunning portable experience that’s absolutely dripping with darkness. The enemies are tough, the environments encourage you to explore, and the combat is just the right level of tricky and fun to encourage you to keep pushing on.

We absolutely loved the RPG, which you can tell from reading our Pascal’s Wager review. We called it “A bravely ambitious game. Worth playing if you are not easily frustrated and prepared to use a separate controller.” Slight control hiccups aside, you should forgive your friends for mistaking it for Bloodborne on your phone.

To celebrate our love for Pascal’s Wager, we gave it away for free to our lucky readers. We had 25 iOS codes up for grabs, and we invited you to fill in a simple form below to get in with a chance of winning. Android folk: Pascal’s Wager isn’t out just yet on your platform, but it is coming. Sorry you miss out on this particular giveaway; we’ll make it up to you.

The giveaway is now closed, and we’ve sent the iOS codes out to the lucky winners via the email address you shared when you filled out the form. Please check your inbox to see if it was you!

If you were unlucky this time, you can go ahead and grab Pascal’s Wager from the App Store right now.

A coalition of tech companies, including a group that represents Apple, is calling on Congress to protect user browser history from warrantless searches.

We reported back in March about how Racing Apex became Hotshot Racing; it’s been a long twisty road for this eagerly-awaited retro-styled racer to make it over to the Switch, but according to the most recent issue of EDGE magazine, the wait is almost over. It’s now slated for a June 2020 release.

From Lucky Mountain Games and Sumo Digital (with Curve Digital on publishing duties), Hotshot Racing boasts bright, polygonal visuals and boost-filled drift-style gameplay in a bid to channel classic arcade racers of times gone by. It also promises to achieve a “silky smooth 60 frames per second” on every platform – and yes, that includes Switch.

The game features eight drivers to choose from, each with their own personalities and cars, as well as 16 tracks and a number of game modes. You can play either alone or in multiplayer – both online and offline – including four-player local split-screen co-op.

There’s also a traditional Grand Prix mode where you can race on multiple tracks to take gold in a championship, a Time Trial mode, a ‘Cops and Robbers’ mode where criminals must evade capture from the police, and a ‘Drive or Explode’ mode where players need to race to checkpoints to avoid exploding.

Check out the trailer above and let us know if you are excited about the imminent release of Hotshot Racing with a comment below.

In this tutorial, you’ll learn the runtime complexity of different Python operations.

Then, you’ll learn how to calculate the complexity of your own function by combining the complexity classes of its constituents. This is called “static analysis”

The tutorial is loosely based on (source) but it extends it significantly with more practical examples, interactive snippets, and explanatory material.

Introducing Big-O

Definition: The complexity of an operation (or an algorithm for that matter) is the number of resources that are needed to run it (source). Resources can be time (runtime complexity) or space (memory complexity).

So, how can you measure the complexity of an algorithm? In most cases, the complexity of an algorithm is not static. It varies with the size of the input to the algorithm or the operation.

For example, the list method list.sort() has one input argument: the list object to be sorted. The runtime complexity is not constant, it increases with increasing size of the list. If there are more elements to be sorted, the runtime of the algorithm increases.

Complexity Function

Given input size n, you can describe the complexity of your algorithm with a function of the input n → f(n) that defines the number of “resource units” (e.g., time, memory) needed to finish it (worst-case or average-case).

Say, you’ve got three implementations to sort a list: implementation_1, implementation_2, implementation_3.

The figure shows the three complexity functions. The x axis measures the input size (in our example, it would be the list size). The y axis measures the complexity with respect to this input.

• implementation_1 has a quadratic complexity function f(n) = n².
• implementation_2 has a quadratic complexity function f(n) = 2n².
• implementation_3 has a logarithmic complexity function f(n) = n log(n).

You can see the code we used to generate this plot here:

import matplotlib.pyplot as plt

import math implementation_1 = [n**2 for n in range(1, 100, 10)]

implementation_2 = [2*n**2 for n in range(1, 100, 10)]

implementation_3 = [n*math.log(n) for n in range(1, 100, 10)] plt.plot(implementation_1, '--.', label='implementation 1')

plt.plot(implementation_2, '-o', label='implementation 2')

plt.plot(implementation_3, '-x', label='implementation 3') plt.legend()

plt.xlabel('Input (ex: List Size)')

plt.ylabel('Complexity (ex: Runtime)')

plt.grid()

plt.show()

Of course, it’s good to waste as little resources as possible so the logarithmic complexity function is superior to the quadratic complexity functions.

Big-O Notation

For large inputs, the runtime behavior will be dominated by the part of the complexity function that grows fastest. For example, a quadratic runtime complexity function f(n) = 1000n² + 100000n + 999 will be much better than a cubic runtime complexity function g(n) = 0.1n³.

Why? Because sooner or later the function g(n) will produce much higher values than f(n) as the input size n increases.

In fact, you can argue that the only important part of a complexity function is the part that grows fastest with increasing input size.

This is exactly what’s the Big-O notation is all about:

The Big O notation characterizes functions according to their growth rates: different functions with the same growth rate may be represented using the same O notation. (wiki)

Roughly speaking, you remove everything but the fastest-growing term from the complexity function. This allows you to quickly compare different algorithms against each other.

To show this, have a look at our two examples:

• Complexity function f(n) = 1000n² + 100000n + 999 grows like O(n²).
• Complexity function g(n) = 0.1n³ grows like O(n³).

By reducing the complexity function to its asymptotic growth, you can immediately see that the former is superior to the latter in terms of runtime complexity—without being distracted by all the constant factors in front of the constituents or the constituents with smaller asymptotic growth.

Examples Big-O of Complexity Functions

So, here are a few examples of complexity functions and their asymptotic growth in Big-O notation:

You can see that the asymptotic growth of a function (in Big-O notation) is dominated by the fastest-growing term in the function equation.

Python Complexity of Operations

Let’s explore the complexity of Python operations—classified by the data structure on which you perform those operations. A great coder will always use the data structure that suits their needs best.

In general, the list data structure supports more operations than the set data structure as it keeps information about the ordering of the elements—at the cost of higher computational complexity.

Python List Complexity

Assume that the length of the data type is defined as n (that is—n = len(data_type)). You can now categorize the asymptotic complexity of the different complexity functions as follows:

Tuples are similar as lists—with a few exceptions: you cannot modify a tuple because they are immutable.

Let’s consider another important data structure:

Python Set Complexity

Assume that the length of the data type is defined as n (that is—n = len(data_type)). If there are two sets in a single operation such as s1 == s2, the lengths are given by the variables n1 and n2. You can now categorize the asymptotic complexity of the different complexity functions as follows:

I highlighted the set operations that are more efficient than the corresponding list operations. The reason for those being O(1) rather than O(n) is that the list data structure also maintains the ordering of the elements—which incurs additional overhead.

Python Dictionary Complexity

Now, have a look at the time complexity of Python dictionary operations:

Most operations are O(1) because Python dictionaries share multiple properties of Python sets (such as fast membership operation).

Composing Complexity Classes

Now, you know the complexity of different operations. But how do you obtain the complexity of your algorithm?

The concept is simple: you break the big problem (knowing the complexity of the whole algorithm) into a series of smaller problems (knowing the complexity of individual operations).

Then, you recombine the individual operation’s complexities to obtain the solution to the big problem.

How? Let’s start with a small example: you first get a list from a dictionary of list values. Then you sort the list:

d = {1: [3, 1, 2, 4], 2: [4, 4, 1, 9]} lst = d.get(1)

lst.sort()

print(lst)

# [1, 2, 3, 4]

Try it yourself in our interactive code shell (click “run” to execute the code):

You have four operations in the code snippet. Let’s annotate each operation with a complexity class.

# Operation 1: Dictionary Creation --> O(n)

d = {1: [3, 1, 2, 4], 2: [4, 4, 1, 9]} # Operation 2: Dictionary Get --> O(1)

lst = d.get(1) # Operation 3: List Sorting --> O(n log n)

lst.sort() # Operation 4: Print List --> O(n)

print(lst)

Knowing the complexity classes of the different operations, you can recombine it as follows:

O(n) + O(1) + O(n log n) + O(n)

(1) = O(n + 1 + n log n + n)

(2) = O(n log n + 2n + 1)

(3) = O(n log n)

You see in Equation (1) what you could call the “chain rule of complexity analysis”: O(f(n)) + O(g(n)) = O(f(n) + g(n)).

You can see in Equation (3) that the Big-O notation focuses only on the largest growing term. In our case, O(n log n) grows faster than O(2n).

Here are two important examples of Big-O recombination:

• O(f(n)) + O(f(n)) = O(f(n) + f(n)) = O(2f(n)) = O(f(n). In other words, Executing an operation a constant (fixed) number of times, doesn’t change the overall complexity of the algorithm.
• O(f(n) + g(n)) = O(f(n)) if the complexity function f(n) grows faster than g(n). An example is O(n³ + 1000n³) = O(n³).

In programming, you can also have conditional execution:

if condition: f(n)

else: g(n)

You can recombine the complexity class of the overall code snippet as follows: O(max(f(n), g(n)). Roughly speaking, (if the condition can be true), the complexity of the conditional execution is the maximum of both blocks f(n) or g(n).

Example:

if lst: lst.sort()

else: lst = [1, 2, 3, 4]

The complexity is O(n log n) because it grows faster than O(n) — the complexity of the block in the else statement.

Another possibility is to repeatedly execute a certain function (e.g., in a for loop).

If you repeat function f(n) m times, the computational complexity is m * O(f(n)) = O(m f(n)). If m is a constant, the computational complexity simplifies to O(f(n)).

Where to Go From Here?

Enough theory, let’s get some practice!

To become successful in coding, you need to get out there and solve real problems for real people. That’s how you can become a six-figure earner easily. And that’s how you polish the skills you really need in practice. After all, what’s the use of learning theory that nobody ever needs?

Practice projects is how you sharpen your saw in coding!

Do you want to become a code master by focusing on practical code projects that actually earn you money and solve problems for people?

Then become a Python freelance developer! It’s the best way of approaching the task of improving your Python skills—even if you are a complete beginner.

Join my free webinar “How to Build Your High-Income Skill Python” and watch how I grew my coding business online and how you can, too—from the comfort of your own home.

Join the free webinar now!

This week at Bungie, something unexpected happened on the way to the Moon Bunker.

Many of you have been chasing a Lie. Last week, you began your hunt with Ana, successfully defending Seraph Towers in the Winding Cove, Anchor of Light, and the Rupture (or watched as your fellow Guardians carried the weight…). You burned through a thousand enemies with your trusty Shotgun. Sadly, your quest for answers came to a halt on Sunday evening, all thanks to a faulty door on the Moon’s bunker.

Earlier this morning, Destiny 2 Hotfix 2.8.1.3 was released, unblocking your journey in finding truth behind The Lie. While we hate to see a bug ship, we’re excited to right the ship as quickly as possible.

Now, we look to what comes next. Season of the Worthy is winding down, which means it’s time for a round of previews from the team. This week, we’re hyper focused on Legendary Engrams and weapon tuning.

Decrypting the Future

Last week, we did a deep dive on how loot will be evolving Season over Season. With the upcoming addition of a Max Power Level to your gear, players have been asking how acquisition methods will be changing. We have a breakdown from the Destiny Dev Team on how we’re keeping the items you earn from Legendary Engrams and core activities relevant during a given Season.

Today we’re going to talk about some changes that are coming in Season 11 to the way the world loot pool works.  Because the world pool serves as a baseline of legendary gear that can be earned across different activities, we want to update and refresh it seasonally with meta relevant weapons, some of which were previously exclusive to specific activities like Gambit or Crucible.

This shared pool can also be unwieldy — if you’re looking for a specific roll of a specific weapon, a pool too large can make it statistically challenging to find, much less with the roll you want. Starting in Season 11, it will be curated both for relevance and overall size, to strike a balance between number of rewards and the chance to get a sought-after drop. Every season gear from previous seasons will be added, and any gear that would no longer have a Max Power Level greater than or equal to the current Season cap will be cycled out to ensure that drops are Power relevant in the current Season.

The World Pool for Season 11 will consist of the following weapons:

• Uriel’s Gift
• Elatha FR4
• The Old Fashioned
• Mos Epoch III
• Hawthorne’s Field-Forged Shotgun
• Last Hope
• Timelines’ Vertex
• Interference VI
• Dire Promise
• True Prophecy
• Jian 7 Rifle
• Enigma’s Draw
• Distant Tumulus
• Escape Velocity
• Honor’s Edge
• Gnawing Hunger
• Arsenic Bite-4b
• Main Ingredient
• Outrageous Fortune
• Truthteller
• Nature of the Beast
• Last Perdition
• Bad Omens
• Night Watch
• Toil and Trouble
• Wishbringer
• Last Dance
• Lonesome
• Long Shadow
• Steel Sybil Z-14

If you still need a weapon or weapon roll from the current Vanguard, Crucible, or Gambit playlists that isn’t on that list, go grab it now!

Lock and Reload

Alongside the changes to the World Loot Pool, we have some changes coming to Destiny 2’s Sandbox. This week, we’ll be focused on weapons – specifically perk tuning. Here’s a quick breakdown from the team on what to expect:

Reload Perk Updates 

We’ve observed that players lean towards picking perks that increase their damage (e.g. Kill Clip) or reduce weapon downtime (e.g. Outlaw).

The damage perks were adjusted quite a while back, but reload perks still don’t feel like choices. This tuning pass aims to keep them feeling powerful without being so dominant that no other choices seem viable.

First, some information on the stats that impact reload:

• Reload Stat: 0-100, and maps onto an archetype-specific reload animation speed.
• Reload Duration Scale: Most reload perks also apply a small multiplier to the reload animation, so that if reload stat is capped, you still see a small speed bump.
• Reload Empty Duration Scale: As above, but only applies if the magazine is empty.

Note that because most weapons have fairly high reload stats, decreasing a reload stat bonus from +100 to +50 will still max out the reload stat most of the time, and so doesn’t slow down the actual reload as much as it seems.

The following perks that alter these reload stats are being updated:

• Outlaw – Slowed this down a little bit.
• Reload stat from +100 to +50.
• Reload Duration Scale from 0.8 to 0.9.
• Feeding Frenzy – This was strictly better than Outlaw, so we’re changing functionality to be based on number of rapid kills.
• Functionality changed to increase reload speed based on number of rapid kills, up to 5.
• With 2 kills, reload speed is equal to updated Outlaw.
• With 3+ kills, is faster than Outlaw.
• With 4+ kills, is the fastest reload in the game (except Alloy Mag).
• Max possible reload duration scale from 0.83 to 0.8.
• Max possible reload stat unchanged at +100.
• Rapid Hit – Gave a huge amount of reload and stability for very little work, front-loaded onto the first precision hit. We have updated it so that it still feels good, but isn’t quite as powerful.
• Adjusted stacking bonus to give less benefit for the first hit and more with subsequent hits.
• Max possible reload stat from +100 to +60.
• Max possible reload duration scale from 0.8 to 0.925.
• Max possible stability stat from +50 to +25.
• Drop Mag – Situationally really powerful without a massive drawback, as good as a perk, but in a magazine column.
• Reload duration scale from 0.85 to 0.9 (just brings it in line with Outlaw).
• Field Prep – Unchanged, but for reference:
• Reload stat +50.
• Reload duration scale 0.8.
• Alloy Mag – Unchanged, but for reference:
• Reload empty duration scale 0.6666.

Additionally, we’re fixing an issue where perks that grant partial weapon ammo did not respect shot count for burst weapons. This means that these perks will now work correctly on Pulse Rifles, Fusion Rifles, and Burst Sidearms, so these perks will be more common on those weapon archetypes in the future:

• Slideways
• Slideshot
• Ambitious Assassin
• Subsistence
• Overflow
• Lead from Gold
• Clown Cartridge

General Perk Retuning 

With opening up space by adjusting reload perks, we also retuned some other, underused perks, based on our internal testing and perk popularity and effectiveness data from our internal analytics.

• Dynamic Sway Reduction — Adds 10 Stability over time in addition to accuracy (this is way more powerful than it sounds).
• Reduced reticle movement from Stability should now tell players this is working.
• Pulse Monitor — Take critical damage and you can quickly switch to your now-fully loaded Shotgun!
• Reload amount from 0.35 to 1.0.
• +50 Handling.
• 5% faster swap speed.
• This works on stowed weapons too, which was already the case but makes it a lot more useful.
• Hipfire Grip — Now helps you hit shots closer to ADS ranges, still doesn’t affect damage dropoff or magnetism.
• 1.2x aim assist falloff.
• +15 aim assist.
• +1.7 degrees precision hip fire angle threshold.
• By default, when hip-firing a weapon, the center of your reticle must be over a target in order to get a crit, otherwise aim assist will give you a body shot. This change gives you a little leniency, so if the center of your reticle is not directly over a target you will still get the crit if you’re within this angle.
• Has no effect on Sniper Rifles.
• Sneak Bow — Actually makes you sneaky now. Stealth buff!
• Now doesn’t ping radar when shooting.

New Perks

Next Season, we’ll have new perks rolling exclusively on the Season of [REDACTED] weapons. Additionally, Iron Banner will be receiving two new perks on two reprised weapons. In future Seasons, these perks will begin to roll on other weapons.

While we don’t want to spoil the fun of discovering and testing new perks, we will leave you with the names of upcoming Iron Banner perks: Iron Grip and Iron Gaze. We are excited to see the community break these perks down when the new Season begins.

Upcoming Archetype Buffs 

Next Season, we’re taking a quick pass on a few weapon archetypes that have been underperforming.

• Slug Shotguns are underused in PvE, because they require crits to maximize damage, and didn’t reward that precision with higher damage per second.
• Slug Shotgun PvE Damage + 30%.
• High-Impact Pulse Rifles are underused in Crucible. While they have a very fast optimal TTK, that TTK was very hard to achieve, so we decided to soften that a little.
• High-Impact Pulse Rifle damage per bullet from 21 to 22. This changes it from 6 crits to 5 crits/1 body to kill a Guardian in PvP at most Resilience values.
• In analytics we see Bows are underused in challenging content, and locally we’ve observed that it feels terrible when a Bow leaves a red bar enemy at low health.
• Bow PvE damage VS minors + 10%.

Future Archetype Updates 

We’ve been evaluating feedback and our analytics data, and while we’re not done yet, we wanted to touch on three community pain points. Our current goal is to touch on these in Season 12, but we’ll be sure to let you know if plans change.

• Sniper Rifles – Community feedback has been that Revoker and Beloved dominate, and looking at analytics they account for 86% of Sniper Rifle usage in the Crucible – and if we include other low-zoom Sniper Rifles, the number gets even higher.
• We’re specifically looking at how zoom translates into ease-of-use for Sniper Rifles. We’re investigating changes that make choosing a Sniper Rifle zoom more of an interesting choice.
• Hand Cannons – We’re evaluating the Hand Cannon subfamilies.
• As an example, we’re breaking out Aggressive Hand Cannons to let us tune their range independently of the others.
• Adaptive Auto Rifles – Auto Rifles in general are highly represented in Crucible, but generally feel balanced. Adaptive Auto Rifles are a little higher, so we’re looking at them.
• We’re not rolling back the Season 10 buff, but we’re adjusting the tuning a little to give other Auto Rifle subfamilies a chance to shine.

Moving forward, we plan to make more regular weapon balance changes, we’ll have more to share on cadence and scope later.

In the know

You may have noticed, @BungieHelp on Twitter is freshly verified. If you don’t follow this account, get to it! Player Support, partnered with the Bungie Network Operations Team (BNOC for short), runs this account to keep you up to date on maintenance timelines, bug investigations, and more. Accept no substitute. If you want more details on what’s happening outside of a 280 character limit, read on.

This is their report.

HOTFIX 2.8.1.3

Earlier today, May 21, Hotfix 2.8.1.3 was released to players. This hotfix was designed to resolve the issue with “The Lie” quest not progressing, and the bunker door should now open as expected.

As a result of today’s hotfix, the Felwinter’s Lie Shotgun will not appear in player Collections once acquired. This will be resolved in a future update.

If players are continuing to observe issues with the quest “The Lie”, please report them to the #Help Forum.

ITEMS LEAVING AT END OF SEASON

As the end of Season of the Worthy approaches, several items will be removed from Destiny 2 on June 9, 2020, at the weekly reset. Some of the items leaving at the end of the season include:

• Access to the Seraph bunkers on the EDZ, Moon, and Io.
• Warmind bits.
• 4th Horseman Exotic quest.
• The Lie quest.
• Legendary Lost Sectors.
• The Warmind Khanjali Seasonal Artifact.

CURRENT KNOWN ISSUES

While we continue investigating various known issues, here is a list of the latest issues that were reported to us in our #Help Forum:

• We are continuing to monitor increases in BEETLE, RABBIT, BEAVER, and ANTEATER error codes.
• Some players still have the Guardian Games milestone visible on their characters.

For a full list of emergent issues in Destiny 2, players can review our Known Issues article. Players who observe other issues should report them to our #Help forum.

Clip it

Good play? Clip it. Team wipe? Clip it. Idea for a montage? Write that down, do the thing, clip it. Got a bunch of clips? Splice ‘em. Got a Creation? Submit it.

Movie of the Week: Brightside

Honorable Mention: Wonderful World(s)

And once again, we come to the end of a TWAB together. How many have we shared together now? The last few have been filled to the brim with news on upcoming changes, and we’re only a few short weeks until the next Season of Destiny 2.

Wait, what’s that? The Season’s almost over? Sorry, gotta cut this short. I have a Conqueror title to earn.

Cheers,

Dmg04

This snapshot brings some exciting new changes for custom world and dimension settings.  Piglins still really like their gold, so don’t hurt them when admiring it.  As always, have fun and report any bugs to Mojang Studios and consider giving them feedback on the new settings.

• Added the ability to toggle redstone wire between a cross and a dot
• Added import/export world settings functionality
• Added a new disableMultiplayer command line option – when used, the Multiplayer button is disabled
• Added a new disableChat command line option – when used, receiving and sending online chat is disabled

REDSTONE TOGGLE

• When right clicking a single piece of redstone, it toggles between a the cross and the dot
• A dot of redstone will not power its surrounding blocks

• Piglins no longer walk around when admiring gold
• If you hurt a piglin while it is admiring an ingot, the ingot will now disappear
• Increased the chance of ores to generate in the Basalt Deltas to balance against the reduced amount of Netherrack in this biome
• Lily pads are now considered junk fishing loot rather than treasure
• Updated logos for Mojang Studios and Minecraft
• Temporarily removed structure settings from flat level preset strings (workaround: use import/export settings functionality)

• Added the ability to control the height of the spreadplayers command
• Added custom worlds and custom dimensions

SPREADPLAYERS

Added an optional argument to specify maximum height.

New syntax: spreadplayers <center> <spreadDistance> <maxRange> [under <maxHeight>] <respectTeams> <targets>

• maxHeight – Specifies the maximum height for resulting positions

LOCATE

• Structure names are now specified as lowercase

CUSTOM WORLDS

• Added experimental support for new custom worlds
• Edit world screen now has an option to export world settings to a JSON file
• During world creation you can import previously exported world settings
• A bunch of new parameters are exposed, but marked as experimental, meaning you can play with them, but there is no guarantee that any of them will continue working (even in the next snapshot!)
• Flat level preset string lost structure settings (they are now stored in a common format for all worlds), we will restore some form of this functionality soon
• Buffet world can no longer be created from server.properties, flat level settings changed a bit; better support for custom world creation on the server is coming soon

• MC-97247 – Map making sound uses equipment sound subtitle
• MC-97507 – Using item which modifies the held item shows “Gear equips” subtitle
• MC-107103 – Trying to interact with entity 3 or more blocks away without seeing its eyes only interacts client-side
• MC-113068 – Zombie banging on door subtitle is called “Block broken”
• MC-123155 – When any non-player entity enters an End portal the obsidian platform is not regenerated and the entity can fall into the void
• MC-147516 – Hostile mobs will sometimes stop attacking/following their target when directly next to it
• MC-166292 – Subtitle from foxes eating chorus fruit: “Player teleports”
• MC-169764 – Fish flopping sound shows ‘Footsteps’ subtitles
• MC-170930 – Nether Sprouts item not visible when held in the hand in first person
• MC-172090 – Piglins would rather hold a golden item instead of a crossbow
• MC-172207 – Parrots don’t imitate hoglins, zoglins and piglins
• MC-172255 – Piglins immediately dismount any entity that they are riding
• MC-177075 – Taking product out of stonecutter produces no subtitle
• MC-177088 – Polar bear step sound does not have subtitles
• MC-177438 – Turtle egg’s name inconsistently capitalized across different subtitles
• MC-177830 – Blackstone and basalt prevent Ancient Debris and other Nether ores from generating
• MC-178130 – Zombie Villagers are called “Zombie” in some subtitles
• MC-178212 – Nether planks, their variants, and hyphae are harder than stems
• MC-178298 – Ancient debris and nether gold ore can’t generate in Basalt Deltas
• MC-180257 – Crash when setting LodestonePos of compass to something other than NBT compound
• MC-180852 – There is no space on the beacon’s secondary power effects in the UI
• MC-181270 – Trying to open old worlds (in Region format?) without version information in snapshots does not show warning screen
• MC-181298 – Game crashes when trying to generate a nether biome superflat world with ruined portals
• MC-181463 – The buttons are not selected in the right order using Tab in the new world creation menus
• MC-181499 – Buckets cannot be emptied into a source block, but still produce a sound when trying
• MC-181511 – Chunks with redstone are removed
• MC-181522 – Chunks that haven’t been explored since 1.13 are resetting
• MC-181611 – Vindicators stop attacking you when they are standing really close to you
• MC-182514 – 15 Block States per SubChunk Limit during World Upgrade
• MC-182743 – Wither skeletons don’t spawn inside of wither roses
• MC-182912 – Piglins still do not change their current crossbow for a new enchanted one
• MC-183249 – Carving Mask BitSets created for lower chunk statuses
• MC-183786 – Named seeds that are not numbers are not being recognized