After a controversial reveal that saw tickets sell out almost immediately, Ravensburger has announced it’s opening up over 1,500 new spaces for its inaugural Disney Lorcana Challenge competitive events.
18.03.2024 - 18:05 / wccftech.com / Muhammad Zuhair
A new research paper has discovered the usefulness of DRAM cache for GPUs which can help enable higher performance at low power.
The GPU industry, which involves consumer, workstation, and AI GPUs, is proceeding in a way that we are seeing advancements in memory capacities and bandwidth, but it isn't sustainable, and ultimately, we could hit the limits if an innovative approach isn't taken.
We have seen GPU makers advance this segment by either incorporating large sums of secondary LLCs (Last Level Caches) or increasing the size of L2 caches. Keeping this in mind, researchers have devised a new way of developing GPU memories, particularly HBM, to break the modern-day capacities and bandwidth limits, along with making data transfer and management much more efficient.
Based on a research paper published at ArVix, researchers have proposed using a dedicated DRAM cache on GPU memory, similar to what we see in modern-day SSDs. DRAM cache is a high-speed storage place for memory, allowing an effective "fetch-and-execute" process. However, this cache isn't what we see in SSDs, somewhat a bit different since it involves the use of SCM (Storage-Class Memory), which is a much more viable alternative to the modern-day HBM and has a lower per-bit dollar cost than DRAM as well.
Researchers have proposed a hybrid approach, utilizing both SCM and DRAM together, to reduce and avoid memory oversubscription & ensure a higher performance per capacity.
The research is pretty in-depth, as expected, and it involves multiple data-flow models as well to aid the SCM data fetching process, and one of those is Aggregated Metadata-In-Last-column (AMIL) DRAM cache organization, which is an attempt to hasten up the process of fetching "data tags," which tells where the data resides in each cache line. The AMIL method suggests keeping all the tags together in the last column of a single row within the DRAM cache for faster access, reducing tag probe overhead, and maintaining Error-Correcting Code (ECC) protection.
Now moving to the more exciting part, the proposed solution guarantees a significant performance improvement, increasing by 12.5x compared to HBM and being 89.3% more energy efficient. These optimistic figures could potentially mark the next-gen transition of the industry into more "innovative" solutions for GPUs, given that SCM combined with DRAM becomes an actual reality once it passes specific qualification tests.
News Source: @Underfox3
After a controversial reveal that saw tickets sell out almost immediately, Ravensburger has announced it’s opening up over 1,500 new spaces for its inaugural Disney Lorcana Challenge competitive events.
NVIDIA's GeForce RTX 4090D Gaming & H20 AI GPUs could soon be restricted from being imported to China as per the latest US trade regulations.
For players who have entered the Public Test Realm for Patch 1.4.0, an issue has been discovered that causes the Codex of Power to be reset, removing any progress made by players. After boosting a character to level 100 via the NPC in Kyovashad, the Codex of Power upgrade process will reset to zero. This means that if you have upgraded various Aspects in the Codex and made another character, boosting that character to level 100 will then wipe the Codex to its original state for all characters. Before boosting a character to level 100 vs after boosting a character to level 100. The only way to work around this currently if you wish to have numerous characters is to make all of your alt characters and boost them to level 100 before you start to upgrade Aspects in the Codex of Power. This will prevent a new character from accidentally removing all your progress as you battle your way through the Diablo 4 PTR.
In the vast world of Grand Theft Auto 5's Story Mode, players are often on the lookout for the coolest rides to steal and cruise around Los Santos. While the game offers an array of vehicles, some stand out for their performance, style, and rarity. Here are the top five cars you can find in free roam that are sure to elevate your gaming experience.
Blizzard has revealed Mythic Tindral and Fyrakk nerfs, significantly reducing the healing check needed on both encounters.
AI tool DrugGPT helps doctors prescribe medicines safely in England; Shikha Goel urges tech pros to develop AI solutions for cybercrime; Intel's Core Ultra Chips: Powering next-gen laptops for content creators and gamers; BJP to utilise AI translation tool for PM Modi's speeches in South India. All this, and more in our daily roundup. Let us take a look.
In an interesting turn of events, Apple has filed a lawsuit alleging a leak of classified information about projects including the recently launched Journal app. This development comes just days after the iPhone maker got hit with a lawsuit by the US Department of Justice for allegedly creating a walled garden with its devices and ecosystem. This time, Apple has alleged that Andrew Aude, a former iOS software developer, leaked details about its classified projects. Consequently, Apple seeks damages of over $25000 in addition to a jury trial.
Intel has some new cores in town and two new CPU families as revealed in the latest Linux Code patches which have unearthed the Adams Lake "Client" and Cooper Forest "Server" chips.
Take-Two will acquire developer Gearbox from Embracer for $460 million.
Intel Battlemage GPUs were recently leaked, revealing new details about two GPU models that could be officially unveiled by the company in the near future. The company has yet to reveal retail models of the new Battlemage GPUs, but Intel hopes to release the GPUs before the end of the year.
AMD is planning to refresh its Ryzen 5000XT CPUs based on the huge popularity of the AM4 platform with faster Zen 3 chips.
A few weeks ago, Toys for Bob unexpectedly split from Activision and Microsoft to become an independent developer once again. However, even in that announcement, the studio based in Novato, California mentioned that they were exploring a partnership with Microsoft for their next game.