George Hotz and his Tiny Corporation colleagues were pinning their hopes on AMD delivering some good news earlier this month. The development of a "TinyBox" AI compute cluster project hit some major roadblocks a couple of weeks ago—at the time, Radeon RX 7900 XTX GPU firmware was not gelling with Tiny Corp.'s setup. Hotz expressed "70% confidence" in AMD approving open-sourcing certain bits of firmware. At the time of writing this has not transpired—this week the Tiny Corp. social media account has, once again, switched to an "all guns blazing" mode. Hotz and Co. have publicly disclosed that they were dabbling with Intel Arc graphics cards, as of a few weeks ago. NVIDIA hardware is another possible route, according to freshly posted open thoughts.

Yesterday, it was confirmed that the young startup organization had paused its utilization of XFX Speedster MERC310 RX 7900 XTX graphics cards: "the driver is still very unstable, and when it crashes or hangs we have no way of debugging it. We have no way of dumping the state of a GPU. Apparently it isn't just the MES causing these issues, it's also the Command Processor (CP). After seeing how open Tenstorrent is, it's hard to deal with this. With Tenstorrent, I feel confident that if there's an issue, I can debug and fix it. With AMD, I don't." The $15,000 TinyBox system relies on "cheaper" gaming-oriented GPUs, rather than traditional enterprise solutions—this oddball approach has attracted a number of customers, but the latest announcements likely signal another delay. Yesterday's tweet continued to state: "we are exploring Intel, working on adding Level Zero support to tinygrad. We also added a $400 bounty for XMX support. We are also (sadly) exploring a 6x GeForce RTX 4090 GPU box. At least we know the software is good there. We will revisit AMD once we have an open and reproducible build process for the driver and firmware. We are willing to dive really deep into hardware to make it amazing. But without access, we can't."

NVIDIA's GTC 2024 AI conference will set the stage for another leap forward in AI. At the heart of this highly anticipated event: the opening keynote by Jensen Huang, NVIDIA's visionary founder and CEO, who speaks on Monday, March 18, at 1 p.m. Pacific, at the SAP Center in San Jose, California.

Planning Your GTC Experience
There are two ways to watch. Register to attend GTC in person to secure a spot for an immersive experience at the SAP Center. The center is a short walk from the San Jose Convention Center, where the rest of the conference takes place. Doors open at 11 a.m., and badge pickup starts at 10:30 a.m. The keynote will also be livestreamed at www.nvidia.com/gtc/keynote/.

Lately Tiny Corp. CEO—George Hotz—has used his company's social media account to publicly criticize AMD Radeon RX 7900 XTX GPU firmware. The creator of Tinybox, a pre-orderable $15,000 AI compute cluster, has not selected "traditional" hardware for his systems—it is possible that AMD's Instinct MI300X accelerator is quite difficult to acquire, especially for a young startup operation. The decision to utilize gaming-oriented XFX-branded RDNA 3.0 GPUs instead of purpose-built CDNA 3.0 platforms—for local model training and AI inference—is certainly a peculiar one. Hotz and his colleagues have encountered roadblocks in the development of their Tinybox system—recently, public attention was drawn to an "LLVM spilling bug." AMD President/CEO/Chair, Dr. Lisa Su, swiftly stepped in and promised a "good solution." Earlier in the week, Tiny Corp. reported satisfaction with a delivery of fixes—courtesy of Team Red's software engineering department. They also disclosed that they would be discussing matters with AMD directly, regarding the possibility of open-sourcing Radeon GPU MES firmware.

Subsequently, Hotz documented his interactions with Team Red representatives—he expressed 70% confidence in AMD approving open-sourcing certain bits of firmware in a week's time: "Call went pretty well. We are gating the commitment to 6x Radeon RX 7900 XTX on a public release of a roadmap to get the firmware open source. (and obviously the MLPerf training bug being fixed). We aren't open source purists, it doesn't matter to us if the HDCP stuff is open for example. But we need the scheduler and the memory hierarchy management to be open. This is what it takes to push the performance of neural networks. The Groq 500 T/s mixtral demo should be possible on a tinybox, but it requires god tier software and deep integration with the scheduler. We also advised that the build process for amdgpu-dkms should be more open. While the driver itself is open, we haven't found it easy to rebuild and install. Easy REPL cycle is a key driver for community open source. We want the firmware to be easy to rebuild and install also." Prior to this week's co-operations, Tiny Corp. hinted that it could move on from utilizing Radeon RX 7900 XTX, in favor of Intel Alchemist graphics hardware—if AMD's decision making does not favor them, Hotz & Co. could pivot to builds including Acer Predator BiFrost Arc A770 16 GB OC cards.

Today is an exciting day for the entire studio, and especially our ENDLESS Dungeon team that has worked super hard these past years to make this game a reality. We've put a lot of passion and dedication into this project, and we're extremely proud of releasing our game today! We hope you enjoy playing it as much as we enjoyed making it.

Endless Dungeon is out, at least the advanced access (aka early unlock) for those of you who bought the "Last Wish Edition". For those who own the Standard Edition you will be able to play starting Thursday (16:00 CEST / 15:00 BST / 7:00 PST).

Shutterstock, Inc., a premier partner for transformative brands, digital media and marketing companies, today announced that it entered into a definitive agreement to acquire GIPHY, Inc. from Meta Platforms, Inc. GIPHY is the world's largest collection of GIFs and stickers that supplies casual conversational content.

GIPHY's vast library of GIFs and stickers draws more than 1.3 billion search queries on a daily basis and powers more than 15 billion daily media impressions. These impressions are distributed via GIPHY's more than 14,000 API/SDK integrations and its owned-and-operated website and mobile app. GIPHY's content serves as a critical ingredient in text- and message-based conversations on platforms such as Meta, other social media platforms such as TikTok, Twitter and Snapchat and team collaboration platforms such as Slack and Microsoft Teams, in addition to integrations with most mobile devices. GIPHY's content library is fueled by both individual artists who contribute original content and top verified media partners such as NBC, Disney, Netflix, NFL, MLB, and NBA, ensuring a steady supply of fresh culturally relevant content which can be inserted into everyday conversations and shared via social media.

Logitech's competitive "G" gaming brand released a short teaser video on Friday (May 19) that alludes to an upcoming next gen product - with a cryptic message: "Built with Pros. Made for you. 05.24.2023." Gaming mouse enthusiasts were quick to leap into speculation mode and predicted that Logitech is set to reveal a successor to the G Pro X (GPX) Superlight (weighing in at 61-63 g) wireless gaming mouse - this model and its (sort of) predecessor 2018's Logitech G Pro Wireless (GPW) have been widely adopted by competitive multiplayer gamers. Others have tempered expectations, and reckon that a new lineup of gaming-oriented headphones is incoming.

The GPX and GPW models have topped mouse popularity charts consistently for the past four years, but hardcore FPS-heads have long wondered when Logitech would eventually catch up with the likes of Razer in terms of cutting-edge sensor and wireless technology. A retail listing on China's online shopping platform, Taobao, has seemingly broken Logitech's (intended?) teaser marketing method - leaked images show a new wireless model called "GPRO X Hyperlight," along with an alleged unit weight of 49 grams, a 35K Hero Sensor, an 8 Khz LIGHTSPEED wireless polling rate, and Lightforce hybrid switches. The Taobao page has since disappeared but the information has been preserved and shared on social media and Reddit. Debates have raged about the authenticity of this leak, so the images and spec details should be taken with a grain of salt.

CD Projekt declared in an investor regulatory announcement (issued on May 11) that its troubled "Project Sirius" multiplayer game was back on track with a renewed development focus. Their briefing is titled: "New framework for Project Sirius, decision concerning partial reversal of the impairment allowance for 2022, and write-off of part of the development expenditures incurred in Q1 2023." As reported back in March, the Polish gaming group made the difficult choice to reboot its multiplayer focused Witcher title and write-off a significant chunk of the development budget. Last week's update seems to indicate that their North American studio, The Molasses Flood, is still involved in the making of Project Sirius and that a smaller chunk of project expenditure has been written off in the mean time.

The company's investor announcement coincided with emerging rumors of employee layoffs - gaming news outlets started to pick up on social media declarations last Friday (May 12). Yesterday CD Projekt confirmed that the refocused and restarted development process has resulted in a round of headcount cuts on both sides of the Atlantic. In a statement issued to PC Gamer, a company spokesperson says: "Because the project changed, so has the composition of the team that's working on it - mainly on The Molasses Flood's side. The concrete number of employees we parted ways with is 21 team members in the US and 8 in Poland (working on the project outside of the US)."

Facebook's parent company, Meta, has announced that revenues generated by advertisement sales have improved fortunes, following on from three quarters of consistent slumps. It has reported an operating profit of $5.7 billion for the first quarter of 2023, this announcement arrives as a surprise to business analysts - their calculations predicted yet another decline in advertising sales for Q1 '23. Mark Zuckerberg - co-founder, executive chairman and CEO of Meta Platforms - has informed his investors that artificial intelligence-assisted systems have directed more user traffic to its social media platforms. Reels and Instagram have experienced the most user population growth, thanks to Meta's AI-driven recommendation system boosting engagement by 24% for the latter platform. The company's leader stated that AI-related operations are set to expand into generative tasking, visual creation tools (for Instagram) and business chatbots.

Meta is continuing to streamline its operations in 2023 - Zuckerberg has previously outlined plans for a year of "efficiency" with extensive plans already underway to reduce company-wide headcounts. 10,000 positions are set to be cut this year, and press coverage from last week revealed that Meta's gaming divisions being hit hard in terms of layoffs. Reports suggest that a third of the team at Ready at Dawn, an Oculus Rift-oriented development studio, were let go throughout that period. The social media and technology let go of 13,000 employees last year due to reduced profits margins. A return to better fortunes has not resulted in an immediate slowdown of headcount reduction - Meta is expected to carry on implementing its 2023 efficiency initiative.

Meta Platforms, Inc. (formerly known as Facebook) has begun implementing widespread staff layoffs across multiple company departments. The cutback operation was announced last month, with the social media giant targeting 10,000 positions (throughout 2023) in an effort to become more efficient. 13,000 staffers were released from their jobs last year, representing 13% of the workforce at the time - advertising revenues had declined in 2022 and Meta said that the change was required in order to navigate economic downturns. A repeat of that sentiment has been issued this year and two internal games development studios have been affected quite heavily by the latest layoff initiative, reports suggest. Ready at Dawn and Downpour Interactive are getting a lot of press coverage - due to former staffers divulging details of Meta's cutbacks via social media.

Thomas Griebel, a (now former) Senior engine programmer at Ready at Dawn, took to Twitter two days ago and made claim that: "One third of the studio was laid off today, including the studio head." He also observes that the studio has been shrinking over time: "Also lost some really great people just due to attrition. Think we're down almost (down to a) half since when I started in August (2022)." Former Ready at Dawn technical designer Colin McInerney has also released a string of information about co-workers being let go. Michael Tsarouhas (senior designer) and Daan van Zelst (level designer) have confirmed that they were released from their roles at Downpour Interactive.

According to Business Insider, Twitter has made a substantial investment into hardware upgrades at its North American datacenter operation. The company has purchased somewhere in the region of 10,000 GPUs - destined for the social media giant's two remaining datacenter locations. Insider sources claim that Elon Musk has committed to a large language model (LLM) project, in an effort to rival OpenAI's ChatGPT system. The GPUs will not provide much computational value in the current/normal day-to-day tasks at Twitter - the source reckons that the extra processing power will be utilized for deep learning purposes.

Twitter has not revealed any concrete plans for its relatively new in-house artificial intelligence project but something was afoot when, earlier this year, Musk recruited several research personnel from Alphabet's DeepMind division. It was theorized that he was incubating a resident AI research lab at the time, following personal criticisms levelled at his former colleagues at OpenAI, ergo their very popular and much adopted chatbot.