Z's Blog

Welcome to my blog. This is a technical site focused on game development. My open-source projects on GitHub: github.com/hxhb and Open-Source projects.
The articles in the blog include a C++ series, Hot Update for UnrealEngine, Plugins and tools development, Game Resource Management series, and thoughts series, etc.

In my spare time, I research and develop technical solutions and preview the results: Showcase. If you are interested in any of the projects, feel free to email me at imzlp@foxmail.com.

UE as Lib

Dictionary-based Shader Compression

Remake PAK at Runtime

Multi-stage Resource Inspection

Splitting Game Resource

Human Pose Tracking in UE

In addition to this site, I also created a community-driven Chinese knowledge base site for Unreal Engine https://ue5wiki.com/, hosted on GitHub Pages. Contributions of technical content are welcome to help build the Chinese Unreal Engine technology community! Please see the submission rules: Wiki Content Writing Format and Rules .

For personal knowledge management, I am very keen on creating my own local-first and open source self-hosted services. This way, I can avoid relying on any platform, have complete autonomy over my data, and be able to migrate services at any time. Options like Obsidian, Hexo blog, and the memos introduced in this article all meet these criteria.

The protagonist of this article, memos, is an open-source lightweight note-taking service that allows you to take notes in a way similar to posting on Weibo, supporting TAG marking and citations. It features an account system and permission management, allowing web access at any time, and notes can be public or private, offering great flexibility. I chose it because it complements Obsidian; Obsidian is still too heavy for my needs—it works well on PC but has a poor mobile experience. Thus, I need a lightweight, always-available note-taking service.

This article will introduce how to deploy a Memos service on a VPS using Docker, along with Nginx to bind the domain name, Certbot to issue and automatically update SSL certificates, and regular backups of the memos database, as well as some optimization configurations I made for the service.

In previous articles, I primarily introduced how HotPatcher can be applied in the hot update process. However, HotPatcher is not just capable of fulfilling the needs for hot updates; it is also a flexible resource packaging framework.

As the project development reaches a certain stage, a massive amount of resources needs to be packaged, and waiting for a complete package to be built can take a long time. If a real device encounters issues, the serial process of repair -> package -> verify can become exceedingly lengthy. There are often cases of script, data, or other resource errors that lead to package abnormalities.

In such cases, waiting hours to generate a complete package prevents smoke testing from proceeding. Even if you rebuild the package, there’s no guarantee that there won’t be other overlooked issues; otherwise, you have to start over again.

Based on this pain point, we urgently need a method to quickly fix resource issues within packages and validate them during the development phase. HotPatcher provides this functionality, allowing you to utilize it fully without waiting for a complete package.

This article will introduce how to utilize the editor features provided by HotPatcher to quickly package preset configurations and specific resources, accelerating the validation process of resources on real devices and significantly improving testing efficiency.

Since the last version release of HotPatcher over half a year ago, its support for UE4 has been commendable, but its support for UE5 has been lacking. There have been frequent inquiries in the group about adaptation.

The last Release version of HotPatcher was v81.0, released on 20230604, which only supports UE5.1.0 and does not support WP and Nanite. Subsequently, Epic released UE5.2 and 5.3, but the plugin has not been adapted. Although one can modify compilation errors themselves, functionality support in UE5 is somewhat in a beta state.

Taking advantage of the Qingming holiday, I worked intensively for a few days to comprehensively optimize and adapt HotPatcher to be compatible with the latest versions of UE4.21 to UE5. A new Release version has been published: v82.0. The latter part of this article will discuss the adaptation of HotPatcher in UE5, precautions, and the update log.

Additionally, in 2020, I wrote an article: Some Thoughts and Ideas on Open Source, and over the past two years, I’ve had some new thoughts on maintaining open source projects. Let’s summarize together.

During the New Year, my iPhone 12 dropped and caused some malfunctions, barely usable. But a while ago, the battery also swelled and could easily shut down automatically, and it has become completely unusable. After switching to a backup device, I left it aside for a while.

Recently, I remembered it because it can still boot normally, and major functions can still be used normally, so I speculated that there should still be partial failures instead of complete damage. It seemed a pity to scrap it entirely, and I could only resort to using a stainless steel bowl.

I took it apart myself to analyze the failure reason and tried to buy parts to fix it. This article documents the disassembly and resolution process.

This year, I plan to celebrate the Spring Festival in Shenzhen and the next stop for my Greater Bay Area trip is Macau. Traveling from Shenzhen to Macau is relatively convenient, so I am making arrangements for the trip. I’ve done some research and itinerary planning that can serve as a reference for those visiting Macau for the first time.

In the previous article Design and Implementation of Resource Self-Correction in UE, I introduced the implementation plan of filtering using the asset inspection of ResScannerUE, followed by automated processing.

Normally, if you want to check a certain property within a resource, you need to load the asset and retrieve the object:

However, if you want to check the properties of many resources in bulk, loading resources one by one in this way takes a long time, especially in the absence of DDC, where loading resources triggers the construction of DDC cache, which is time-consuming and has a high performance overhead.

As the size of the assets increases, scanning all resources in a project completely can be extraordinarily time-consuming. Therefore, I wonder if it’s possible to implement a method that does not require loading resources but can still retrieve properties within resources.

After research, I found a clever way to achieve this by combining reflection, asset serialization, and the features of AssetRegistry. Moreover, this is a non-intrusive approach that does not require modifying any existing resource type code.

This article will introduce the implementation principles and analyze the related logic in the engine, as well as present the practical application of this mechanism in the project.

In a previous article, I introduced the implementation and efficiency optimization of multi-stage and automated resource checks using ResScannerUE in UE.

• Multistage Automated Resource Check Solutions in UE
• Automated Resource Check Practices Based on ResScannerUE
• Resource Compliance Check Tool ResScannerUE in UE

This already allows for very flexible configuration of resource checks and timely alerts, with very powerful rule expression capabilities. However, they are still limited to the stage of discovering problems for reminders; although there are constraints to prevent submission, manual processing is still required. Especially in cases with a large amount of existing resources, the manpower involved in manual processing is very time-consuming.

Based on this situation, I designed a mechanism for automating resource corrections that can automatically fix resources after rules are established, ensuring that all scanned resources are set to the correct state.

This completes the final piece of the puzzle in my resource check solution and can be applied throughout the scanning process as needed. This article will introduce its design concept and implementation mechanism, as well as some practical application scenarios in projects.

If I had to choose one recent experience in working with UE that felt the most meaningless and torturous, it would definitely be trying to find a resource in the ContentBrowser based on a long string path.

The inability to quickly jump directly to a directory or resource makes me feel like I’m wasting my life every time I have to navigate deeply nested folders one step at a time.

To address this pain point, I wrote a small tool that significantly alleviates this manual anxiety. I named it j2 (jump to), which is a minimalist ContentBrowser jump tool, open-source on GitHub: hxhb/JumpTo.

This article will introduce the usage of this project, the thought process behind conceptualizing the solution, and the step-by-step code implementation. Although the functionality itself is a very simple plugin, the discovery, analysis, and solution process for the actual pain points is worth documenting.

In 2020, I wrote an article How to Build Your Own Knowledge System?, which mainly reflects on the daily accumulation of technology, leaning towards methodology.

This article will explore the possibilities of personal knowledge management from the perspective of tools and practical use, based on Obsidian, and attempt to integrate it with the Hexo blog. Of course, this preference for tools and practices has a strong subjectivity; this article is entirely based on my personal practice and experience; there is no best tool, only the one that suits you best.