Now that PAX is over lets take a quick look at what we're doing to tackle the last Status Report’s listed blockers, and what issues are between us and .60 hitting experimental (so you're all in the know).
- Loot Distribution: The QA, Design, and Gameplay Programming teams have isolated the cause of the issue, specifically tied to stalling of item respawn - while initial server spawn, and server cleanup are functioning properly. The Gameplay Programmers responsible for this are hard at work nailing down a fix to this issue, and a possible solution has been handed to QA - we will run it through the QA department for verification.
- "Stuck Magazines": This issue has been resolved, and the QA Team are no longer able to reproduce it. Strike one more issue from the list.
- Sliding Players: All of the reproducible methods for causing this issue have been flagged and resolved by the gameplay programming team, and have been verified by QA.
- Playability of New UI Inventory - The team has finished all remaining changes needed to be ready for .60 experimental - once the build is in the hands of the experimental userbase, we'll be able to make any required adjustments or improvements based upon how they interact with it.
In addition to these previous issues, the engine and gameplay programming teams are currently working on resolving two issues we'd like to see solved prior to hitting experimental:
- FPS Drop / Clouds: Internal Testing discovered a 100% reproducible frame drop to single digit factors on limited hardware configurations. While the issue only appeared on 2 out of 30+ PCs, we're taking it very seriously. That said, the engine programming team has committed a fix and QA are verifying it as of the time of this report.
- Server Crash: We've encountered an issue with unconscious players reconnecting / logging in to a fresh server sometimes causing the server to suffer an unexpected crash. Obviously we'll need to get this resolved before pushing to a large userbase because no one likes playing on servers that are consistently crashing - and we all know there will be plenty of unconscious survivors throughout Chernarus. That said, the gameplay programming team have a solid repro, and all the crashdumps and server logs they could want, and are tracking this issue down as I write this.
As you can see, we're knocking issues down one by one, and each day brings us closer to pushing the first .60 experimental build to Steam. Keep in mind, this will be just like every other experimental upload and will have issues. So lets all chip in and make sure to submit good bug data when encountering them - Quality Assurance Lead Ondrej Klima gave some excellent pointers in how to submit the best bug data possible in our last Dev Q&A video on the official Youtube Channel.
All of us here at the DayZ dev team cannot wait get .60 on to experimental branch. The excitement and support at PAX East was overwhelming, and there are still so many more improvements to the technology powering DayZ to come!
Brian, Viktor, Miroslav, and Raist give us an update on current development in this Status Report. Subjects covered by the guys are blocker issues, network code, the new AI spawner, and the new animation system.
It’s that time again - biweekly Status Report time. In addition to my update, we have some interesting reads from Miroslav our Lead Gameplay Programmer, Viktor our Lead Animator, and Raist the Senior Programmer responsible for DayZ's Central Economy. That aside, let’s look at what our current critical issues are for the release of .60 to experimental branch.
Loot Distribution: Current internal builds have an issue we're chasing down that drops globally spawned quantities of items to about 3,000 in certain situations/server states. This is around 17,000 to 18,000 less than the acceptable minimal operating value. This issue is one of the top priorities for the designers currently, as it effectively makes the title near unplayable.
"Stuck" Magazines: The gameplay programming team are working on an issue that causes magazines during the reload process to become "stuck" within the players inventory, thus causing several cascading failures with players inventories requiring them to disconnect and reconnect to resolve the issue.
Sliding Players: The QA Team encountered a 100% repro rate series of steps to cause players to become stuck in a sliding state - regardless of laying, prone, or erect. These steps have been provided to the gameplay programming team - and they are working on resolving the issue.
Playability of New UI Inventory: The design and gameplay programming teams are wrapping up the last of the playability issues moving to the new UI, as functional support for the legacy UI won't be possible moving forward with the new Enfusion renderer.
Since the last Status Report the team has nailed down the last blocking bugs on the renderer side, squashed some nasty virtual machine server exception issues, resolved blocking issues with navmesh generation, infected AI behavior, several critical client crash bugs, a new nasty issue with the desync of player positions, and rolled out the support for configuring all the initial required graphical options within the options menu for the new renderer release.
Looking at what the Engine Programming team have taken on and ahead of them for the Enfusion Renderer:
Alpha to Coverage Implementation
Resolved all issues flagged as "Must Fix" or "Blocker" for .60
Current Focus is on:
New UI and Renderer Major & Minor severity level issues
Aside from the current issues I've mentioned above, personally I'm a bit concerned of the network state of vehicles. With the team focused on the milestone goals for .60 we haven't been able to give the attention to iterating on network synchronization, performance, and physics on our current group of vehicles. I think they're far from where I'd like them to be - but we'll more than likely have to put off improving their state to somewhere more enjoyable for the Early Access playerbase to future builds.
Each day moves us closer to having a build we're confident and able to move to experimental branch, and no one group of people want it out there more than us, the developers. Lastly, before I sign off for this Status Report - I know I've said it before - but I can't say it enough, keep in mind this is just our first public iteration of this portion of Enfusion engine. In addition to iteration and progression from feedback and bug reports from the Early Access player base we'll also be working with GPU manufacturers to figure out how best to use their technologies to the maximum effect, and get DayZ optimized for use in future drivers from said manufacturers.
Hey everyone. While I don't have an experimental build for you today - I do have good news and an update on what is going on with the blockers, how far we've got with them, and so on:
User controls within the new inventory look to be at parity with the old inventory system. The last few remaining issues are a few nasty blocking bugs on the functional side, such as items missing their icons.
The QA team have worked with the design and animation teams to get a firm handle on exactly where the inconsistencies are with the new reload mechanics, and said teams are working on resolving the issue. I think we can all agree that when we push a large change to how weapons reload - they damn better all reload with the same mechanics ;)
Proper character loading/saving is back, and in the process some nice network optimizations in this area were made!
Renderer based blocking issues
We've cut the list of blocker flagged issues tied to the new Enfusion renderer in half since the last time we talked - and in the time since our last Status Report the Enfusion engine team has made some outstanding optimizations across Chernarus - which you can see the result of in the latest .60 Dev Log video that you see below.
In addition, while the Enfusion engine team was able to commit and test their latest optimizations, the automated performance benchmarking tool briefly discussed in the latest .60 Dev Log video has enabled the environment team to begin to isolate problematic issues with certain areas of the map that would not otherwise be noticed. We're excited to see these fixes give us yet another small bump up in Enfusion's new renderer performance.
As we get closer to bringing .60 to experimental branch, I lightly urge you all to temper your excitement with a few brief statements of reality about public development of a title, and its engine:
.60 is focused on UI, and Renderer - it does not feature the new animation system, player controller, or user actions
We still have plenty of issues ahead of us to tackle, the renderer is but one hurdle on the road to a complete DayZ
Network Optimizations, Server Performance Optimization (This is a big one, standing between us and more players, more infected, more animals), Animation System, Eden Update Audio Tech merge, and more are all ahead of us
So I encourage you to keep your eye on the proverbial prize here and report your bugs over on the official DayZ forums, be active in testing these new technology changes, and be a part of moving DayZ towards 1.0
"I frequently get asked why a specific bug might not be resolved yet, or why something hasn't been addressed yet in development that from a players perspective might seem critical.
There are a lot of things to take into consideration in development when weighing a bugfix, not to mention when you're dealing with a title transitioning from one engine, to another one that is *in development* Lets take a quick look into a brief break down of how the thought process goes:
- First off - what is the risk of this fix at the current time?
What are the goals for the upcoming build? Does this potentially represent larger issues that will push the build back?
Is this issue tied to a technology that is going to be replaced?
For example - Is this an issue tied with the legacy renderer?
Is this issue related to ongoing investigation of a larger issue with wider sweeping impact on the title?
If this issue IS tied to a system scheduled to be rewritten or replaced - what is the estimated time involved in:
Work resolving the issue itself
Test pass on the issue
Overall BVT test pass to ensure related issues are not encountered
Regression testing on the issue
A lot of issues consumers see in the way of:
Duplicating (Old action system)
Glitching into models
Damage / Projectile cheats
Balancing / etc
Are all things that are slated to be resolved, or at least mitigated with new technologies that are being worked on *right now* - so the hard decision has to be made - Do we potentially waste valuable development time and resources on something that will *not* end up in the shipped title? Honestly - we try to do that as little as possible. We still end up going a very large amount of it to try and keep the consumer steam branches as playable as possible, but frequently the unpopular and hard decision has to be made to stay the course, and dedicated those resources towards the final product / final systems / final technology."
Hey guys, its that time again - Status Report day. I know you all wanted to hear some crazy hype tastic reveal like - FREE OCULUS RIFT HEADSETS UNDER EACH OF YOUR COMPUTER DESKS OMG YAY (but more specific to our upcoming .60 experimental push) but I'm afraid I don't have that for you today, and we don't do hype like that anyways. While we've made strides in client side performance functionality, we're still chipping away at blocking issues in the following areas:
With the move to Direct X 11, we'll no longer be able to support the legacy user interface. Thus, functional parity with the legacy UI -has- to be achieved on the new UI. The design team are currently working on resolving the last few bugs blocking us from having that parity.
The move in .60 to involving manual bolt cycling, as well as misfire and jamming management via the charging handle on weapons such as the M4 has an impact on every firearm in the game. We're currently working to resolve some functional hiccups in the switching between magazines, and the need to chamber twice before firing from a new magazine.
A very large volume of unneeded traffic was discovered when investigating issues with character loading - a rewrite of the methods used to save and load from the central hive. Once we restore proper functionality, this area will no longer be a blocker.
Tasks Completed: - .60 Build Optimization Pass (Performance in cities increased by 50% from initial build - optimization is ONGOING however)
- Cleared Rendering Test Pass list of visual/rendering based bugs
- Renderer settings UI
Current Focus: - MSAA Hardware Multisampling Support
- Alpha to Coverage GPU Feature (Which will increase quality of vegetation rendering)
- Finalization of In-Game UI Features
We've heard a lot of feedback wanting some changes to how we're communicating development information over the last year - and we started steps a few months back towards adjusting to try and meet some of these requests. We've expanded the community team and put processes in to ensure that regardless of development tasks, the Status Reports will come out on schedule (Tuesdays, BiWeekly). In addition, we've begun using our YouTube channel more - to show upcoming changes to DayZ as they are developed. (Status Reports talk about development progress, and what changes are planned in the future - where as the YouTube channel should show what is coming in the next build - when it is functional enough to demonstrate)
We'll also be trying a team Q&A type video where folks can ask questions about what its like to work on DayZ, work at Bohemia, and such - our first video on this from our Quality Assurance team should be coming shortly. Our Community Manager Smo55 is ramping up his presence on our subreddit and official forums, our new Brand Manager DHawkz is working with Bohemia Interactive publishing to expand the options for DayZ merchandise on the BI Store, and a redesign of DayZ.com, as well as working with our hardware partners to create opportunities for raising awareness of DayZ as it moves towards our Beta, and Release goals.
I'll be discussing some of the systems and pending changes tied to the new animation system and player controller over on the official forums later this week, and next up for the YouTube channel we're working on our next Dev Log video on .60 covering performance comparisons between Direct X 9, and Direct X 11 rendering.
Once .60 hits experimental branch, we'll start putting together our first "Whats new with .60" video - where DHawkz, Smo55, and I go over the changes in .60 and what to expect gameplay wise - for our YouTube channel.
Lastly - as .60's time on experimental starts to draw to a close - we'll be bringing back the official DayZ survival gaming e-sport, The Survivor GameZ. The Survivor GameZ VII Qualifiers will be a fantastic way for us to stress test .60 before it hits stable, and to see how the new inventory, and reloading/chambering mechanics work in a high stress situation. Keep an eye on the official Survivor GameZ and DayZ twitter accounts for more info on this - twitter.com/survivor_gamez and twitter.com/dayzdevteam
For this week, Brian Hicks (Lead Producer), Peter Nespesny (Lead Designer), and Miroslav Maněna (Lead Gameplay Programmer) will let us in on details regarding the CLE, the new engine + renderer, and upcoming plans and work for the damage system and Enforce script among other things.
As mentioned as a high possibility, we ended up passing our internal goal for putting 0.60 in consumers hands. However, this was expected as a high chance. We are of course working hard on getting it ready, and on to experimental branch as soon as possible. Our work currently is focused on optimization of the performance inside major cities in Chernarus. New Chernogorsk has caused us some performance hickups we want to resolve to an acceptable playable frame rate. The engine team has a daunting task ahead of them, focusing on the last few hurdles to unlock the capabilities of the renderer while still combating and improving legacy engine scene management. With trouble areas in the DirectX 9 area dipping as low as the mid teens in some situations, we consider it critical to isolate those areas and ensure we can hold a stable 30fps (In Direct X 11) in them. Globally, frame rates on average can be in the 40 to 60 range on average - so obviously our focus right now is in those major cities. It is key to keep in mind as said many times before - this is just the first iteration. In addition to continuing bugfixes on into beta and the 1.0 release, the new renderer technology will open up many new particle effects to the design team, and we'll be working with GPU manufacturers for cross testing to isolate areas in which hardware specific optimizations can be made.
Implementation of all base renderer features
Simulweather/True Sky Implementation
Optimization of major cities
Finalization of New Renderer Settings UI/Options
Implementation of New UI
Bug Fix, Bug Fix, Bug Fix
I'm confident that a global increase in frame rate playability, as well as a huge upgrade in the visuals in DayZ are going to be well worth the work that has gone into the part of the Enfusion engine, and I can't wait to share it with you all. While .60 obviously focuses most heavily on implementing the engine's new renderer - .60 also has continued iteration on the Central Economy (something I think many people don't realize is VERY critical to the DayZ experience) with the audit mentioned in previous Status Reports complete, as well as changes to the tagging for weapons. Changes to vehicle spawn points, initial attachments, and finally - weapons are both spawning with random attachments AND a chance to spawn with magazines that have a random amount of ammo inside them.
Beyond .60 - the teams next major updates focus is primarily on the new animation system and player controller. For the end user, it is important to understand these large engine changes have been holding back fixes on a lot of legacy issues with the title, as well as much larger over all changes to how DayZ plays - including, but not limited to DayZ's new user actions - something that I feel paired with the renderer, and new UI complete a 3 part massive change to how DayZ *plays*. In addition, the gameplay programming team's work on the new damage system - which will be explained in a bit more detail below by Lead Gameplay Programmer Mirek. The members of the programming team focusing on the Central Economy also continue to work on the deployable version of the CLE & Database structure for use by both mod authors and private server operators alike - as well as functionality for DayZ's local offline mode.
On the Early Access / Community side - our forum transition is complete, and DayZ.com forums are back online. As some of you may know the feedback tracker is offline while we transition to a new software option for it. This shouldn't be down too much longer, as it is critical to providing an outlet to those testing DayZ's development builds to communicate the issues they encounter while doing so.
We know that members of the community have wanted to see an increase or change in the format of our outbound communication, and that too has received some love this month. We've done some restructuring in the processes behind these, as well as the structure of responsibilities. We're hoping that over the coming month you'll all enjoy the changes in this area - and we'll be keeping an eye on your reactions.