Forget Windows 95, it’s 30 years since Doom was released on the Super Nintendo Entertainment System. And thanks to the Raspberry Pi RP2350 microcontroller, the game is back in cartridge form.
Doom on the SNES was a programming tour de force that arguably kicked off the “But can it run Doom?” meme. The first-person shooter, with its atmospheric soundtrack and 3D graphics, had taken the world by storm after its release in 1993. It was ported to many platforms, but squeezing it onto an SNES cartridge was a challenge.
The port was accomplished thanks to a Super FX chip, a graphics acceleration coprocessor developed by Argonaut and Nintendo, in the game cartridge, but the experience was not as fluid on SNES as on high-end PCs of the time and some content was missing.
Thirty years on, Limited Run Games (LRG) is having another crack at the old warhorse, this time with updated cartridge hardware, while still ensuring it will work on the vast majority of SNES consoles.
According to Randal Linden, who also worked on the original port in the 1990s, the update replaces the Super FX chip with an inexpensive RP2350. As well as simulating the Super FX chip, the RP2350 also performs “graphics format conversion operations” and provides a high-speed interface to multiple memory devices.
It’s important to note that RP2350 is not running the game itself, although Doom has been ported to the hardware. Rather, it is simulating and emulating the Super FX chip. The setup has allowed the developers to further optimize the code and reach a heady 20 frames per second.
Speaking to John Linneman of The Digital Foundry, Linden explained how it works:
Controls can be via a standard controller or mouse, and there’s also rumble force feedback. Missing levels have been added, as well as Episode 4.
As for why performance is capped at 20 frames per second, Linden explained: “A single screen’s graphics data requires about 30 kilobytes of transfer. With the reduced display size – thanks to letterboxing at the top and bottom, and pillarboxing on the sides – we’re able to transfer roughly 11 kilobytes per vertical blank.
“Multiply that by three vertical blanks (which corresponds to 20 frames per second), and you get around 33 kilobytes of data, which is exactly what’s needed to transfer the next frame. That’s how we’re hitting the theoretical throughput limit.”
The world of retro gaming and creating new games to run on old systems is thriving. There are few better ways to celebrate 30 years since Doom made its debut on the SNES than with a whole new cartridge containing a thoroughly-modern RP2350 taking on the duties of the innovative Super FX silicon. However, there doesn’t appear to be a delivery date for the cartridge and LRG did not respond to The Register’s queries. ®
Eddie AI has introduced a plethora of updates that should make the platform faster, more compatible, and easier to use. From broader NLE support to smarter media handling, the latest version is designed to fit more naturally into diverse editing workflows. Let’s see what’s new (again!) with Eddie AI.
Over the past year, we’ve covered how AI-powered tools are reshaping post-production, from automated rough cuts to text-based editing. We’ve written quite often about Eddie AI here at CineD, and noted its ability to accelerate the often time-consuming process of sorting, cutting, and syncing footage. In the last couple of months, they’ve added AI text-based editing to their rough cut mode, plus support for 30 languages. And it doesn’t seem to stop – we’ve got 50 new updates to take a look at! As a reminder, here’s Johnnie and Nino discussing what Eddie AI can do:
What is Eddie AI
So if you’re not yet familiar with Eddie AI, it’s a cloud-based platform that combines machine learning with text-driven video editing. You upload your footage, generate searchable transcripts, and quickly assemble rough cuts by working directly with text.
What’s in the latest release
Expanded NLE Integration and File Compatibility
Eddie now edits in 30+ languages. Import your interview in any one of dozens of new languages. (Except Klingon.)
Blackmagic Cloud and NAS Support: Eddie AI now supports Network Attached Storage (NAS) and Blackmagic Cloud. Boom.
NLE Proxy: If you make proxies in your NLE, Eddie now pulls those instead of the source media, making it swifter to import.
Updated Eddie extension for Resolve: Eddie is more stable for DaVinci Resolve imports and performs better with large projects.
FCP Simple Timeline: If you cull interviews on the timeline first in FCP, you can now drop in FCP timelines to Eddie’s Rough Cut mode. Support already exists for Adobe Premiere Pro and DaVinci Resolve.
Precise Timecode and Reference: Relinking is smoother now that start timecode inconsistencies across vendors have been addressed.
Support for Apple App Store’s DaVinci Resolve: Compatibility issues with the App Store version of Resolve have been fixed.
Optimized Clip Importing: Large batches of clips for B-roll logging import more efficiently, avoiding timeout issues.
High Framerate Support: 90, 100, 119.88, and 120 FPS are now supported.
Fixed Premiere Pro Motion Effect: The latest Premiere Pro bug affecting Motion Effects on edit sequences is resolved.
Preventing Metadata Duplication: Eddie now detects and avoids re-importing FCP clips with existing metadata.
Enhanced User Experience & Account Management
Pay-as-you-go, No Sub required: Buy credits without a subscription and use them as needed.
Simplified Subscription Management: In-app unsubscribe and reactivation, plus direct billing portal access.
Broadened Language Support: Define a default language for transcription.
Multicam and Project Workflow Refinements
Improved Multicam Error Handling: Clearer error and warning messages with more flexible import options.
Robust B-roll and A-roll Processing: Smoother performance, even under heavy loads.
Automated Multicam Activation: Automatic switch to multicam mode when importing multicam clips.
Enhanced Multicam Podcast Sync: Better angle detection for podcast cuts.
Export in one click. Image credit: Eddie AI
Refined Export and Media Handling
Advanced MP4 Export and Captioning: Transparent outline styling and a new font for multilingual support.
YouTube Video Import: Direct import of YouTube videos for editing, including vertical clip creation for social platforms.
Better FCP Captions: Cleaner captions with fewer orphaned words.
Robust Export Stability: Dozens of backend updates for smoother exporting.
Smart Rough Cut Export: Chapter names now appear as ranged timeline markers.
Key Bug Fixes and Performance Enhancements
Improved B-roll Analysis: Better support for pro codecs.
Text Editing Reliability: More stable text-driven edits.
Multicam Project Stability: Fixes for crashes related to R3D files and multicam audio thumbnail bugs.
Mobile Upload Improvements: Better performance for Chrome and Android uploads.
Streamlined UI: Cleaner transcoding, uploading, and sharing interfaces.
Who is Eddie AI for?
One editor reflected that after feeding a 100 GB+ interview into Eddie, it isn’t trying to be your editor – it’s trying to be your assistant editor. And that sentiment echoes through all its updates: Eddie isn’t meant to replace your intuition, your narrative instincts, or your human touch. It’s built to enhance them, to hand over the groundwork so you can focus on telling the story that only you can tell.
Price and availability
All the new features are now live for Eddie AI users. To try it out and to get more information, please visit the Eddie AI website.
Are you currently using Eddie AI in your projects? Have you noticed a significant difference in your workflow? Let us know in the comments!
The road to SAP S/4HANA is paved with data, and for many organizations still navigating this digital transformation, the journey is about a fundamental rethinking of their most critical asset. The challenge isn’t just moving to the cloud, but what to bring to the new system and what to leave behind.
Speaking ahead of Data Migration International’s (DMI’s) Digital Lounge @Lakeside 2025 event in Switzerland, Thomas Iseler, Vice President of Strategic Partners at DMI, discussed the real-world data challenges on the path to a modern ERP.
“For the longest time data has been an afterthought,” Iseler explained. “Then the shortcut was always to take all data.” However, he added that this approach can create a ripple effect of problems, from sluggish system performance to bloated project timelines and costs. Moreover, in today’s AI-driven landscape, that old mindset is a recipe for failure.
According to Iseler, a successful SAP S/4HANA data strategy rests on three key pillars:
Embrace a Clean Data Core. The first rule of a modern migration is to be selective. “You should only migrate data that is relevant and that is needed,” Iseler emphasized. Bringing decades of historical clutter into a new, agile environment undermines the purpose of the transformation. Iseler added that focusing only on necessary data is key for system and process performance as well as the speed and complexity of the migration itself.
Prioritize Data Quality. While good data has always been necessary for reporting and transactions, AI raises the stakes exponentially. “With the growing importance of AI, the relevance and importance of data quality has grown significantly,” Iseler noted. Without clean, reliable, and harmonized data, any investment in AI-driven business outcomes is built on a shaky foundation.
Solve the Legacy Data Puzzle. Historical information is often essential for compliance, audits, and analytics, and traditional archiving is not a smart and efficient option. Leaving it scattered across aging, expensive-to-maintain legacy systems is a huge risk. The goal, Iseler said, is to figure out “how to make that data readily accessible, and how to make storing it transparent and cost-effective.”
Ultimately, the move to SAP S/4HANA is an opportunity to break from the past. It’s a chance for organizations to transform data from a liability into their most powerful asset. As Iseler concluded, with the rise of AI, data has moved from the periphery to the core. “It’s a complete reversal from being an afterthought. Now it has to be the center of the strategy.” For organizations embarking on their transformation, making data the central character in their story is the key to a successful outcome.
Watch the full interview here.
What This Means for ERP Today Insiders
Seek a unified partner rather than a point solution. Modern digital transformation is a complex web of interconnected data challenges, encompassing migration and quality, as well as archiving and compliance. Engaging multiple vendors for different pieces of the puzzle adds complexity, risk, and cost. For example, DMI is a strategic partner with a single, unified platform, JiVS, that manages the entire data lifecycle. It handles migration, data quality, harmonization, masking, and legacy data retirement from a single point of contact. This approach simplifies governance, reduces project timelines, and ensures consistency, proving critical for organizations looking to build a flexible and future-proof data architecture.
Your AI and cybersecurity posture depends on data strategy. An organization’s future capabilities in AI and its current security resilience are directly tied to how it manages data today, especially historical data. AI is useless without high-quality, relevant, and accessible data. By focusing on what Iseler describes as a “clean data core” and ensuring data quality is paramount during the transformation, DMI’s approach helps build the essential foundation for reliable AI and analytics. Additionally, legacy systems are a huge cybersecurity risk. They are often unpatched, difficult to monitor, and managed by a shrinking pool of experts. Decommissioning these systems and moving historical data to a modern, secure, and audited platform dramatically reduces the attack surface and helps ensure regulatory compliance for decades of data.
The time to move to the cloud is now, and it’s more feasible than you think. The primary roadblocks to migration, such as cost, complexity, and downtime, are directly addressed by a smart data strategy. By migrating only a small fraction of relevant data and intelligently managing the rest, the migration project becomes faster, cheaper, and less risky. Furthermore, a lean, clean system in the cloud is more agile, scalable, and ready to connect to modern innovations like SAP Business Technology Platform (BTP) and Business AI.
Xiaomi is preparing for a big launch of the 16 Series next month. The company usually introduces two flagship models in the fall, but this year the plan is different. Instead of two devices, Xiaomi will unveil three. The new lineup will include the Xiaomi 16, Xiaomi 16 Pro, and Xiaomi 16 Pro Max. This is a major change because the “Ultra” model is not part of this launch. The Ultra is expected to arrive later, just like in previous years.
The latest information comes from certifications in China. These listings have revealed some important details about the charging speed of the upcoming phones. All three models will support 100W wired charging.
Xiaomi 16 Series to Launch Next Month With a Surprise
This is slightly faster than last year’s Xiaomi 15 series, which offered 90W wired charging. While the upgrade may not seem huge, it still shows that Xiaomi is pushing for improvements in battery technology and charging efficiency. Faster charging is always a welcome feature for users who want to spend less time plugged in.
Another exciting detail is about performance. According to leaks, all three devices will be powered by Qualcomm’s Snapdragon 8 Elite 2 chipset. Qualcomm is set to introduce this new chip in late September. Xiaomi is expected to be the first smartphone brand to launch devices with this processor. If the reports are true, the Xiaomi 16 series could be announced just days after Qualcomm’s event.
The Snapdragon 8 Elite 2 is likely to bring big improvements in speed, energy efficiency, and AI performance. This will make the Xiaomi 16 series one of the most powerful Android smartphones available at launch. For users, this means smoother multitasking, faster gaming, and better overall performance.
In terms of display sizes, the Xiaomi 16 and 16 Pro are expected to share a smaller form factor. The Pro Max, on the other hand, may come with a larger display and higher-end specifications. This strategy will allow Xiaomi to target different types of users. People who prefer compact phones can go for the regular 16 or the 16 Pro, while those who want a bigger screen and premium features can choose the Pro Max.
See Also: TECNO Spark 40 Pro+ Wins Over Tech Reviewers with Slim Design, Wireless Charging and G200 Processor
As for the global launch, things may follow Xiaomi’s usual pattern. The company generally introduces its flagship devices in China first. Then, several months later, the global versions are released. Based on this timeline, the Xiaomi 16 series could reach international markets in early 2026. However, it’s not clear if all three models will make it outside China.
The launch of three flagship phones at once shows Xiaomi’s growing ambition in the premium smartphone segment. By offering more choices, the company can compete directly with rivals like Samsung and Apple, who also provide multiple high-end options every year.
For now, fans are eagerly waiting for the official announcement. If the rumours hold true, Xiaomi will reveal its new flagship series in late September, making it one of the first to use Qualcomm’s next-generation chip.
Australian development house Click Studios has warned users of its Passwordstate enterprise password management platform to update immediately if not sooner, following the discovery of an authentication bypass vulnerability that opens the doors to an emergency administration account with nothing more than a “carefully crafted URL.”
“We have released [Passwordstate] build 9972,” Click Studio’s support team posted to the company’s announcements board, “which includes two security updates. We recommend customers upgrade as soon as possible.”
Those two security updates are detailed in the company’s changelog as modifications to prevent clickjacking attacks against the software’s browser extension and a “potential authentication bypass” which, at the time of writing, was pending the assignment of a CVE ID.
In a separate security advisories page, Click Studios goes into slightly more detail. Creating “a carefully crafted URL” allows attackers to access the Passwordstate Emergency Access portal, which is designed to provide ingress into the software when other accounts have been locked out or are otherwise inaccessible. “This account doesn’t allocate a license from your available license pool,” the company’s documentation explains, “and is not intended for use in day to day operations. It should be regarded as an account of last resort.”
Will passkeys ever replace passwords? Can they?
READ MORE
For attackers, though, it seems to be very much an account of first resort. With nothing more than the “carefully crafted URL” and a web browser, attackers can bypass the authentication requirement of the Emergency Access portal – giving full administrator-level access to the Passwordstate installation, albeit at the cost of triggering email alerts to all registered security administrators on the system.
The flaw impacts a not-inconsiderable number of users worldwide. According to the latest figures released by Click Studios, Passwordstate is used in over 29,000 organizations and by 370,000 “security and IT professionals” in fields including government, banking and finance, insurance, healthcare, legal, utilities, mining, and, perhaps most worryingly, defense. These include Tasmania’s Department of Health, which has been using the product since 2016 – though many of the company’s customers choose to keep their use of the Passwordstate software a private matter.
This latest vulnerability is the fourth authentication bypass flaw to hit Passwordstate 9 since its release, following the discovery of CVE-2022-3876 and CVE-2022-3875 in 2022 and CVE-2024-39337 in 2024. Its severity has, naturally enough given the ease of exploitation, been rated as “high,” though for partial mitigation Passwordstate advises – though does not require – that administrators restrict access to the Emergency Access portal by IP address specifically to avoid its misuse in the event of exactly this kind of issue.
Passwordstate 9 users looking to protect themselves from the vulnerability are advised to install Build 9972 or later to receive the patch, along with the related clickjacking fix. ®
A recently developed artificial intelligence (AI) model known as YOLO v5 is effective in classifying skin irritation following patch tests, new data suggest, allowing for swift and accurate evaluations.1
The data regarding this new AI model for classification of skin erythema from patch tests were authored by investigators such as Seoyoung Kim, from the Department of Integrative Biotechnology at Sungkyunkwan University in South Korea. Kim et al noted that recent studies have examined AI’s implementation as an assistive diagnostic tool in dermatology.
Advances in machine learning (ML) and deep learning (DL) have also led to certain studies highlighting the feasibility of utilizing ML to automate skin reaction detection.2
“These investigations highlight the revolutionary potential of AI-powered diagnostics in dermatology, particularly for improving the accuracy, consistency and accessibility of epicutaneous patch testing,” Kim and coauthors wrote.1 “In this study, we developed a model that overcomes the limitations of the convolutional neural network (CNN) to classify skin irritation from images using the object detection algorithm YOLO v5 (You Only Looking Once version 5).”
Trial Design Details
The investigative team gathered patch test data from trial participants between 2020 – 2023, assessing patients who had been recruited in Seoul and the Gyeonggi Province who had also voluntarily enrolled in their analysis. Individuals deemed eligible for enrollment in the trial were adults ≥18 years who were shown to have healthy skin and were able to provide written informed consent and commit to follow-up visits as well as all other study requirements.
The patch testing required for Kim and colleagues’ research was carried out using Van der Bend chambers, which consisted of hypoallergenic, medical-grade non-woven polyester and contained chromatographic filter paper. The investigators would apply cosmetics or cosmetic ingredients to the upper back of study subjects and leave these in place for 24 hours. Participants’ skin responses were then evaluated 1 hour and 24 hours after removal. Reactions were graded on a 5-point scale and were adapted from the Frosch and Kligman method and CTFA safety guidelines:
0: A lack of visible reaction
1: Mild erythema, either scattered or diffuse
2: Moderate, uniform erythema
3: Pronounced erythema with edema
4: Pronounced erythema with edema and vesicles
The team, for the purposes of documenting responses, gathered standardized clinical photographs under controlled conditions, regardless of whether a reaction occurred. All images were captured by Kim and colleagues at a fixed resolution of 4160 × 2768 pixels to allow for the detection of any subtle erythematous shifts. Preprocessing of images was performed to prepare them for analysis with the YOLOv5x deep learning algorithm.
Each of the collected images was assessed independently by 4 evaluators. In cases of discordance, the investigators instructed a senior reviewer to provide the final score. Annotations were later mapped to corresponding bounding boxes within the dataset.
Findings on AI Model for Skin Irritation
Kim and coauthors’ training dataset was made up of 83,629 images, with 1,312 and 1,536 images being utilized for evaluation and validation, respectively. Overall, the AI model attained a classification accuracy of 0.983 at both the 24 and 48-hour mark.1 The model’s F1 score, indicating a balance between precision and recall, was shown to be 0.982.
In terms of individual grades, the areas under the curve (AUCs) were shown to be 0.914 for score 0, 0.838 for score 1, and 0.865 for score 2. Sensitivity for detecting score 0 was particularly high at 0.997. Such conclusions would indicate that this AI model supports and classifies skin irritation effectively, thus allowing for quicker and more accurate skin assessments.
“The AI-based erythema reading model developed in this study demonstrates significant potential to enhance the efficiency of evaluations while minimising inter-rater variability, thereby enabling more objective and consistent assessments,” they wrote.1 “Moreover, the integration of the proposed future improvements is expected to further increase the accuracy and reliability of patch test reaction grading, ultimately broadening the model’s applicability across a variety of clinical environments.”
References
Kim S, Hwang H, Oh M, Han J, Park S, Lee S, Kim G, Cho S, Lee DH, Cho JY. Evaluation of Artificial Intelligence-Assisted Diagnosis of Skin Erythema in a Patch Test. Contact Dermatitis. 2025 Aug 27. doi: 10.1111/cod.70011. Epub ahead of print. PMID: 40859876.
WH Chan, R Srivastava, N Damaraju, et al. “Automated Detection of Skin Reactions in Epicutaneous Patch Testing Using Machine Learning,” British Journal of Dermatology 185, no. 2 (2021): 456–458.
Vuzix is revolutionizing smart glasses software thanks to the guidance of their visionary CEO, Paul Travers. His experience in the wearables space is vast, beginning with Virtual Reality (VR) headsets in the early ‘90s for gaming interfaces like Quake, Doom and nearly a hundred other titles. On the ground floor adapting mixed reality technology, this entrepreneur has witnessed computing evolve over the years.
Paul Travers worked for the U.S. Navy and other companies, eventually building electronic binoculars for Hitachi before founding his own enterprises. Focused on technology for both consumers and businesses, he always respected the way cameras looked out into the real world while offering augmented versions of what users see. Having successfully built and sold businesses, he feels the promise in Vuzix technology is beyond belief.
Travers started Vuzix in 1997, recognizing the potential offered to computers by adding cameras. Today, he stands at the inflection point for Augmented Reality (AR) and sees benefits across numerous industries when combined with Artificial Intelligence (AI). Now AI-enabled smart glasses are slashing training times, eliminating costly errors, and delivering real-time guidance to frontline workers.
“It’s really hard to fathom how big a deal this is going to become,” said Travers.
Enhancing User Experience
Smart glasses are used for two-way communication. Users are able to see HD video feed off each pair, along with a display for others to inform processes and operations in real-time. Everything is interactive because of two-way visual communication which pops up right in the field of vision and is paired with voice.
”I’m walking around in the building, and you can see what I’m seeing because the camera’s running,” Travers explained.
Paul Travers, CEO of Vuzix
No longer any need to pull up a PDF. Instead, just draw or point an arrow in real time on a video feed while that arrow stays there in space, visible on a screen. When gathering information, no need to write up or take pictures on a phone in order to type up a work order for parts. Everything is being automated now, and smart glasses are changing the world in enterprise.
“There’s no back and forth, training becomes a few hours—just put the headset on and get started,” advised Travers.
Vuzix has partnered with Quanta Computers to make their patented Waveguide lenses. This technology creates a flat, transparent display that measures only one millimeter in thickness. They look like lenses that go in a pair of glasses, but they are actually the windows into a computer from which the information is pushed through to the user.
Wearables Transform Industries
The promise of smart glasses is in the workplace, not in gaming or entertainment. Vuzix welcomes the future of warehousing as it stands in relation to smart glasses. Certain industries stand to benefit most: logistics, healthcare, manufacturing and field service. Across all these sectors, AI assistants in smart glasses deliver hands-free training and instant operational support.
“Augmenting humanity is going to be the big next frontier where humans get better because they have access to tools like an AI engine and an AI assist agent moving things forward,” emphasized Travers.
Vuzix is working with Nadro, Mexico’s largest pharmaceutical distributor, and the company has reported 93% faster employee onboarding with 30% faster order picking. This wearable technology is directly helping ease the company’s labor shortages and supply chain bottlenecks.
Companies in the defense space are also using Vuzix lens technology. One part of the business is focused on OEM and custom builds, along with their enterprise product business. The next-gen stuff that they are working on enables spatial computing along with the current features.
They have almost 450 plus patents and patents pending, with every intention to provide this technology to their high volume customers in the broad markets. They also welcome competitors to come purchase Waveguides from them all day long.
Before electronic computers became widespread, NASA’s most advanced technology wasn’t a machine. It was people. Teams of human “computers”–many of them women–performed intricate calculations using nothing but pencils, paper, and chalkboards… Continue reading
AR Requires AI for Success
AR doesn’t just mean you put your hand out in space and you type on a virtual keyboard, or that you point to things and slide them around. Vuzix knows that it is exhausting to put your hands up in space like that without a physical feedback mechanism.
“If you have an AI engine and a camera running, it just closes the loop and makes it all work,” said Travers.
When you bring AI to a pair of AR glasses, all those things are just going to be simply automated and voice controlled. At that point, things get so much easier. AI is accelerating the possibilities so much faster because it provides capabilities that are essentially omniscient. Important work across every part of life will occur without the hours of research they would otherwise require.
The AR component is augmenting the real world, inserting a visual element in like a simulation. Smart glasses provide that capability. The power lies in combining AR with the Large Language Models (LLMs) of AI that use data they point to the real world, using augmented reality to enable the AI engine to drive the human-computer interface.
The Future of Smart Glasses
Smart glasses are starting to look and feel like real glasses. Fashion-forward designs are now very important as users want sleek, sexy designs that are comfortable on the face and don’t feel like ski-goggles.
Housing features developed by Vuzix have enabled wearable smart glasses for more standard use. Things started out big and bulky, but these products need to be practical for employees to wear them for a long shift. Also, the battery has to be able to last.
Travers insists that smart glasses are going to be the next replacement for the phone. Ride share accounts will be connected to smart glasses through AI agents. Mapping services will become obsolete as users will no longer have to look down at their phones. Users will just ask their glasses how to get somewhere, and they will do all the work.
“The whole concept of an application is going away,” said Travers. “An AI agent is going to be your best friend.”
Shanghai, China — August 29, 2025 | XD Games is thrilled to announce that the highly anticipated Sword of Convallaria x Delicious in Dungeon crossover event is now live from August 29 to September 25, 2025. This limited-time event fuses the strategic depth of Sword of Convallaria’s tactical RPG combat with the irresistible charm and culinary creativity of Delicious in Dungeon.
From delicious monster feasts to tactical dungeon battles, this is an adventure fans of both worlds won’t want to miss. The Laios Party has arrived in the Kingdom of Iria, teaming up with familiar allies to explore a mysterious dungeon brimming with adventure and flavour. Along the way, they cook fantastical monster-based dishes, each one strengthening bonds, inspiring hope, and revealing hidden truths.
Event Highlights:
Chew It, Iria Food! (Aug 29 – Sep 23)
Take on crossover-themed stages packed with flavour and challenge. Completing event quests and achievements will earn players Astral Shards, Secret Fate, exclusive furniture such as State of Walking Mushroom and Castalia, Hope Luxite, and Random Legendary Trinket Chests.
Dungeon Adventure Guide (Aug 29 – Sep 25)
Complete designated tasks to unlock the legendary collab character Senshi and his Memory Shards entirely free, along with the Encounter in Dungeon avatar frame and Primordial Powder.
Adventurer’s Gift (Aug 29 – Sep 23)
Log in for seven days to claim Hope Luxite, the Roast Basilisk furniture, and additional Legendary Trinket Chests.
Collab Summon Gift (Aug 29 – Sep 25)
Summon in the “Adventure: Delicious in Dungeon” crossover banner to unlock the Pot of Memory interactive furniture, collab avatars Marcille and Falin, a dynamic avatar background, the Let’s Dig In! avatar frame, and the Unnamed Immortal Recollection. Marcille is available from Aug 29 to Sep 25, while Falin joins from Sep 12 to Sep 25. The first 10-pull is 50% off, and one featured collab legendary is guaranteed within 100 pulls.
Proof of the Dungeon Adventure Pack (Aug 29 – Sep 25)
A premium bundle featuring 680 Astral Gems, Mimic interactive furniture, Dryad Fruit and Red Dragon Full Course furniture, and the Dungeon Collector dynamic title.
Founded in 2003, XD Games. is a global video games developer and publisher with extensive experience across development, publishing and operations. As of 2022, XD has 38 free-to-play and paid games with over 50 million monthly active users worldwide. In addition, TapTap — a mobile games community and app store — was launched by XD in 2016 and is estimated to currently have 44 million monthly active users.
XD’s mission is to assemble like-minded talented games industry craftspeople with a common goal of touching the hearts of players around the world. XD produces games of the finest quality with passion, offering a pure gaming experience, and encouraging players of all ages to embrace the unique joy of video gaming.
More than 450 game developers on Blizzard’s Diablo team have formed a union, making them one of the largest and latest groups of Microsoft workers to do so. The new union includes game developers, artists, designers, engineers, and support staff across the Diablo franchise, according to the Communications Workers of America (CWA) press release, which voted “strongly in favor” to be represented by the CWA.
The Diablo team is seeking to improve compensation, job security, and workplace conditions under its Microsoft parent company. Over 3,500 employees at Microsoft are now in CWA unions, including more than 500 World of Warcraft developers who formed Blizzard’s first wall-to-wall union last year. Microsoft laid off as many as 9,100 employees earlier this year, and 1,900 employees at Activision Blizzard and Xbox in 2024.
“With every subsequent round of mass layoffs, I’ve witnessed the dread in my coworkers grow stronger because it feels like no amount of hard work is enough to protect us,” said Kelly Yeo, an organizing committee member for the Diablo union. “This is just the first step for us joining a movement spreading across an industry that is tired of living in fear.”
