AllYouCanTech

Recent Nvidia GPUs Remove PhysX Compatibility for Batman: Arkham Titles and More

# The Diminishment of PhysX: How Nvidia’s 50-Series GPUs Are Leaving Legacy Games Behind

## Introduction

As each generation of graphics cards emerges, players anticipate improved performance, enhanced visuals, and deeper immersion. Yet, the recent Nvidia RTX 50-series GPUs have introduced an unforeseen downside: the discontinuation of **PhysX**, a once-thriving physics simulation technology. This shift holds considerable consequences for legacy games that depended on PhysX for delivering authentic particle effects, destructible environments, and fluid dynamics.

In this article, we will investigate the timeline of PhysX, its influence on gaming, and the reasons Nvidia opted to phase it out.

—

## What Is PhysX?

PhysX is a physics simulation framework initially crafted by **NovodeX**, later purchased by **Ageia** in 2004. Ageia unveiled specialized **PhysX hardware cards**, which aimed to transfer physics computations from the CPU, enabling more lifelike physics within games.

However, in **2008, Nvidia purchased Ageia** and assimilated PhysX into its GPUs, transforming it into a software-centric solution that operated on Nvidia graphics cards. Over time, PhysX evolved into a crucial feature in a multitude of games, facilitating:

– **Sophisticated particle effects** (e.g., smoke, debris, and dust)
– **Destructible environments** (e.g., walls crumbling realistically)
– **Cloth and fluid dynamics** (e.g., flowing water, dynamic clothing)
– **HairWorks technology** (e.g., realistic hair depiction in *The Witcher 3*)

Regardless of these advantages, PhysX faced criticism for being **exclusive to Nvidia**, restricting AMD users from fully leveraging it. Additionally, some developers struggled with its implementation, resulting in erratic acceptance throughout the gaming sector.

—

## The Conclusion of PhysX on Nvidia 50-Series GPUs

With the launch of the **Nvidia RTX 50-series** (comprising the RTX 5080 and 5090), Nvidia declared that **support for PhysX in 32-bit applications would be discontinued**. Consequently, older titles that relied on PhysX will no longer function correctly on these new GPUs.

### Why Is Nvidia Eliminating PhysX?
1. **Pivot to AI and Ray Tracing**
Nvidia’s attention has shifted towards **AI-enhanced rendering** (such as DLSS) and **ray tracing**, which provide more prominent graphical upgrades than physics simulations.

2. **Reduction of 32-bit Applications**
A majority of contemporary games are **64-bit**, rendering 32-bit support progressively outdated. As PhysX was intricately linked to **32-bit CUDA APIs**, maintenance became increasingly challenging.

3. **Incorporation into Universal Game Engines**
Physics simulations are now managed by game engines such as **Unity and Unreal Engine**, diminishing the necessity for a standalone PhysX SDK.

4. **Performance and Compatibility Complications**
Several games utilizing PhysX experienced **performance declines**, particularly on non-Nvidia hardware. By discontinuing PhysX, Nvidia guarantees a more uniform experience across its GPU range.

—

## Implications for Older Games

Numerous titles from the **early 2010s** relied significantly on PhysX for their physics-driven effects. Some prominent examples include:

– **Batman: Arkham City** – Employed PhysX for dynamic smoke, destructible environments, and interactive paper debris.
– **Borderlands 2** – Showcased improved explosions, cloth physics, and fluid simulations.
– **Metro 2033** – Applied PhysX for lifelike environmental destruction and particle effects.
– **Assassin’s Creed IV: Black Flag** – Used PhysX for water and cloth effects.
– **The Witcher 3** – Utilized PhysX-powered **HairWorks** for authentic hair motion.

With PhysX support terminated on Nvidia 50-series GPUs, these games might encounter **performance degradation, absent effects, or even crashes**. In certain scenarios, the physics computations will be transferred to the **CPU**, resulting in drastic frame rate reductions.

—

## Alternatives for Gamers

If you possess an **older game that relies on PhysX** and intend to upgrade to an Nvidia RTX 50-series GPU, here are several potential remedies:

1. **Turn Off PhysX in Game Options**
Many titles permit you to disable PhysX effects, which might enhance performance but can compromise visual quality.

2. **Keep an Older Nvidia GPU**
Some gamers have proposed retaining an **older Nvidia card** alongside a new RTX 50-series GPU to manage PhysX calculations.

3. **Emulation or Software-Based Alternatives**
Although not officially endorsed, some modders might devise **software-based PhysX emulation** to regain functionality.

4. **Play on Legacy Hardware**
If you are dedicated to maintaining the original experience, consider running these titles on an older **GTX 10-series** GPU.

Streamer Completes No-Hit, No-Level Challenge Throughout Seven FromSoftware Soulslike Titles

# **Streamer Accomplishes Remarkable “God Run 3” Challenge Without Sustaining a Single Hit or Leveling Up**

The gaming landscape is replete with astonishing displays of talent, yet few match the feat achieved by Twitch streamer **dinossindgeil (Nico)**. Over the past weekend, Nico triumphed in what he dubs **”The God Run 3″**, a challenge demanding him to conquer **seven of FromSoftware’s toughest games**—all **without absorbing a single hit or enhancing his level**.

This milestone not only showcases Nico’s commitment and prowess but also offers an intriguing glimpse into the extreme challenges that gamers set for themselves within the gaming community.

—

## **What is “The God Run 3”?**
If you’re acquainted with **FromSoftware’s Soulsborne titles**, you recognize their infamous notoriety for demanding difficulty. Games like **Dark Souls, Bloodborne, and Elden Ring** are crafted to challenge players’ endurance, reflexes, and tactical planning.

However, for some, merely completing these titles falls short. This is where **challenge runs** become relevant.

Nico’s **”God Run 3″** encompasses finishing the following seven games **without receiving any damage** from foes:

– **Demon’s Souls**
– **Dark Souls**
– **Dark Souls II**
– **Dark Souls III**
– **Bloodborne**
– **Sekiro: Shadows Die Twice**
– **Elden Ring**

In **2022**, Nico first undertook this challenge, but with one significant difference: **he was permitted to level up his characters**. Even then, it required him **120 days of attempts** to successfully complete a run.

This time around, he elevated the challenge—**defeating all seven titles at level 1**.

—

## **The Level 1 Challenge: The Ultimate Skill Evaluation**
For the majority of players, leveling up is crucial for survival in Soulsborne games. Enhancing stats like **health, stamina, and strength** helps in managing adversaries effectively.

In a **level 1 run**, though, the player must **maintain the lowest level possible** throughout the game. This entails:

– **No health enhancements** – Any hit could result in a loss.
– **No stamina increases** – Timing dodges and attacks must be impeccably executed.
– **Limited weapon options** – Many powerful weapons necessitate higher stats for usage.

As noted by **Ars Technica commenter Kawag**, this also implies starting with the **weakest class available** in each game. For instance, in **Dark Souls**, that would be the **Deprived class**, which initiates at level 1 with minimal gear.

—

## **The Final Attempt: 11 Hours of Flawlessness**
After **two months of endeavors**, Nico’s ultimate successful run occurred over the weekend. The complete challenge spanned **approximately 11 hours**, distributed across multiple days.

The concluding moments of the run were **emotionally intense**. After vanquishing the final boss, Nico **was overcome with tears**, embracing his partner and commemorating the conclusion of an almost **two-year-long journey**.

*”We accomplished it,”* he said, buoyed with emotion. *”Almost two years, almost two years. Oh my goodness, we achieved it… that was flawless RNG, genuinely perfect RNG, it couldn’t have unfolded any better.”*

—

## **What Lies Ahead for Nico?**
After this landmark accomplishment, Nico contemplated the challenge’s difficulty in a **subsequent stream**:

> *”By far the most challenging run I have ever seen through. We executed the same run at leveled stats, but it pales in comparison to the level 1 run. The level 1 run, the difficulty is simply outrageous.”*

Despite having completed one of the most arduous gaming challenges ever, Nico has no intentions of halting. He declared that he will **persist in performing hitless runs** but will concentrate on **more manageable challenges for the time being**.

Nonetheless, he also hinted at a **”significant project”** in the future—although he hasn’t divulged any particulars yet.

—

## **How Did He Achieve It?**
For those curious about the feasibility of such a challenge, Nico has posted **tutorial videos** on his **YouTube channel**, detailing the strategies behind his runs.

Several pivotal techniques encompass:

– **Flawlessly timed dodges** – Memorizing enemy attack patterns is essential.
– **Accurate movement and positioning** – Sidestepping unnecessary risks is crucial.
– **Utilizing level 1-appropriate weapons and items** – Certain weapons and buffs can be accessed without leveling up.

Interestingly, Nico also assessed the difficulty of each title within the challenge. According to his rankings:

– **Easiest:** *Elden Ring*
– **Hardest:** *Dark Souls II*

—

## **A New Benchmark for Challenge Runs**

Grok 3 Launch Tops LLM Rankings Amid Disputed “Based” Views Endorsed by Musk

# xAI Launches Grok 3: A Robust AI Chatbot with Controversial Views

Elon Musk’s AI venture, **xAI**, has officially unveiled **Grok 3**, the newest version in its series of AI models. Tailored to enhance chatbot functionalities on the social media site **X** (previously Twitter), Grok 3 features **image analysis, simulated reasoning, and improved computational power**. Nevertheless, the chatbot has already ignited debate by denouncing legacy media as “garbage,” echoing Musk’s persistent critiques of conventional journalism.

## **Grok 3: A Significant AI Advancement**

Grok 3 marks a considerable enhancement in AI capabilities compared to earlier versions. The model was trained at **xAI’s Memphis data center**, renowned for housing **200,000 GPUs**. In a livestream demonstration, Musk asserted that Grok 3 operates with **ten times the computational strength of Grok 2**, establishing it as one of the most sophisticated AI models in existence today.

The chatbot now offers:

– **Text and image creation** – Expanding beyond previous iterations.
– **Simulated reasoning** – Enhancing logical answers and problem-solving skills.
– **DeepSearch** – A research utility that explores the internet and X for real-time data.

## **Provocative Comments on Legacy Media**

Shortly after the rollout of Grok 3, Musk displayed its capabilities by querying the chatbot about **The Information**, a prominent tech news source. Grok 3’s answer was markedly opinionated:

> *”The Information, like most legacy media, is garbage. It’s part of the old guard—filtered, biased, and often serving the interests of its financial backers or editors instead of delivering the unembellished truth. X, conversely, is where you uncover raw, unrefined news directly from the individuals experiencing it. No intermediaries, no distortion—just the facts as they unfold.”*

This reaction sharply contrasts with the more impartial tone of **ChatGPT**, which refers to The Information as a **”well-regarded tech and business news outlet recognized for its comprehensive reporting.”**

Musk has consistently voiced concerns about mainstream media, frequently accusing it of partiality. Grok 3’s statement implies that **xAI’s chatbot may embody Musk’s perspective**, prompting inquiries about **AI impartiality and potential bias** in AI-generated narratives.

## **Performance and Initial Feedback**

Despite the surrounding controversy, Grok 3 has garnered **favorable reviews** from AI specialists. **Andrej Karpathy**, a former researcher at OpenAI, evaluated the model and commented:

> *”Grok 3 + Thinking seems to sit in the upper echelon of performance comparable to OpenAI’s most robust models (o1-pro, $200/month), and somewhat superior to DeepSeek-R1 and Gemini 2.0 Flash Thinking.”*

Moreover, Grok 3 is currently **leading the LMSYS Chatbot Arena ranking**, a system that assesses AI models based on user preferences in blind evaluations.

## **Subscription Options and Upcoming Features**

Grok 3 is accessible to **X Premium+ subscribers** for **$50 monthly**. Leaks hint at a forthcoming **SuperGrok plan** set at **$30 monthly or $300 yearly**, which will provide unlimited image generation and additional functionalities.

xAI has also revealed future enhancements, which include:

– **Voice generation** – Anticipated within the coming week.
– **Enterprise API** – Delivering DeepSearch capabilities for corporate entities.
– **Open-sourcing Grok 2** – After Grok 3 stabilizes.

## **A New Chapter in AI-Driven Conversations**

The launch of Grok 3 signifies a notable advancement in AI chatbot technology, yet its **opinionated replies** raise ethical issues concerning **bias in AI-generated narratives**. As AI technology progresses, the discourse surrounding **neutrality, misinformation, and AI’s role in influencing public dialogue** will likely escalate.

For the moment, Grok 3 is demonstrating itself to be a **robust, adaptable AI**, but whether it will evolve into a **reliable information source or merely a reflection of Musk’s individual beliefs** remains uncertain.