PCVGS retweeted
御冥福をお祈りいたします
クリスタルキング・中村公晴さん死去 ボーカル・田中昌之が公表【全文】北斗の拳「愛をとりもどせ！！」作詞
daily.co.jp/gossip/2026/10/2… #デイリースポーツ #DailySports
PCVGS retweeted
Mark Cuban says AI isn't smart, and he'd trust a seeing-eye dog over it to get across the street
"AI is not smart. It just knows how to use statistics basically to predict the next word. It’s predictive. It’s not deterministic. If you asked 4 different AIs the same question, you’re going to get 4 different answers"
"AI doesn’t know the consequences of its actions. AI doesn’t have inherent ethics"
"If I’m standing at a 4-way intersection and there’s cars going back and forth, and I have video in my phone and the best models ever, and I have a seeing-eye dog right here, I’m letting the seeing-eye dog take me across the street before I let the AI take me across the street. It’s always going to be behind"
"A 2-year-old sitting on a high chair with a sippy cup knows that if it pushes the sippy cup off the high chair, Mom’s going to come running, and it knows it’s going to crack up. AI is not going to figure that out ever"
Anyone more knowledgeable on this than me could explain?
Proof REA is is malicious.
This repository is simultaneously:
(1) the vehicle for an active social engineering campaign harvesting credentials and biometric data from users who install it,
(2) the subject of apparent artificial star inflation to manufacture legitimacy, and
(3) the host of an architectural vulnerability enabling prompt-injection-to-RCE in AI agent workflows.
The combination of pre-positioned @latest install instructions and a coordinated amplification network represents a complete supply-chain attack delivery mechanism should a malicious package version be published.
Readers added context they thought people might want to know
Security reviews of the official rea-agents npm package found no install hooks or telemetry endpoints, and Snyk lists zero direct vulnerabilities. Reports of malware stemmed from a separate fake repository (rea-tools/rea), not the official REA project.
senn-tech.com/en/blog/rea-re…
security.snyk.io/package/npm/re…
pasqualepillitteri.it/en/news/23845/…
PCVGS retweeted
Terry Davis built TempleOS without networking or memory protection. He kept hardware support limited because he wanted the entire system to remain simple enough for one person to understand.
He once said:
"An idiot admires complexity. A genius admires simplicity."
He spent years writing the 64-bit OS almost entirely himself, including its kernel, HolyC compiler, graphics engine, and even a 3D flight simulator.
PCVGS retweeted
🚨 A VKD3D-Proton bug was just found where an NVIDIA workaround was being applied to EVERY GPU, potentially breaking D3D12 games on completely unrelated hardware. 👀
• NVIDIA had a workaround for certain shader conversion operations
• The workaround could turn 32-bit operations into 64-bit ones
• That unnecessarily forced some shaders to require shaderInt64
• GPUs without Int64 support could then FAIL to create the D3D12 pipeline
• The issue was reproduced on Mali-G710 and Mali-G68 GPUs
• Those same shaders worked correctly WITHOUT the NVIDIA workaround
• A new dxil-spirv fix allows the workaround to be disabled depending on the GPU driver
• The fix is still under development and has NOT been merged
• This could improve D3D12 compatibility on Linux/ARM GPUs without shaderInt64
an NVIDIA-specific workaround was accidentally creating requirements for GPUs that never needed the workaround in the first place.
Credit to Utkarsh Dalal for tracking this one down!
PCVGS retweeted
Is your TV on the list? 👀
293 TV models from 83 brands!
All tested by people on their own TVs.
Free and open source. No root, nothing uninstalled.
ALT Tested Android/Google TVs using CleanYourTV
PCVGS retweeted
🚨 Intel Nova Lake-S is already showing some AMAZING single-thread performance.
A leaked 24-core Core Ultra 4000 desktop sample has appeared in PassMark, and the early numbers are interesting:
• Nova Lake-S 24-core: 5,210 single-thread
• Core Ultra 9 285K: 5,059
• Ryzen 9 9950X3D: 4,805
• Core i9-14900KF: 4,635
That puts the early Nova Lake sample around 3% ahead of the 285K and 12% ahead of the 14900KF in this test.
But multi-thread is a different story:
• Nova Lake-S: 61,509
• Core Ultra 9 285K: 67,274
• Ryzen 9 9950X3D: 72,065
Even crazier, an 8-core Nova Lake sample reportedly managed to slightly beat the 285K in single-thread too.
These are EARLY engineering samples with inconsistent results, so don’t believe this as final performance.
But Nova Lake’s single-thread showing this early?
Intel might have something cooking. 👀
PCVGS retweeted
Prices aside, do we even need a generation upgrade? I haven’t played anything on these systems yet that I feel has really tapped into the power of the PS5 and Xbox.
Everything still feels like a ps4 game at a higher res/frame rate.
PCVGS retweeted
en.gamegpu.com/news/igry/fsr…
AMD has explained why FSR 4.1 works differently across Radeon GPU generations. The RX 9000 implementation uses RDNA 4's dedicated FP8 matrix acceleration for efficient neural-network processing. RX 7000 and RX 6000 cards require separate optimizations because their hardware capabilities differ significantly.
On Radeon RX 6000, AMD adapted the upscaling model to INT8 computation. Support is now available through Adrenalin 26.10.1, while Xbox Series X and S support is still coming. Some mobile devices also use a smaller neural network to reduce computational demands.
Sony's PSSR and QSSR technologies share related research with AMD, but they should not be treated as identical FSR 4.1 implementations. Image quality and performance depend on GPU architecture, model optimization, rendering resolution and the game itself.
#AMD #FSR41 #Radeon #RDNA4 #Xbox #PlayStation #Upscaling
PCVGS retweeted
Up now for all, is a slice of my complete QSSR vs PSSR2 vs IGTI vs FSR 3: Image Quality & Performance Cost Breakdown and Analysis.
Here I pit QSSR against ITGI to test how good it really is for PS5's lifespan.
NXG Technical Review
youtu.be/mBbiELIYW4Y?si=ygBj… via @YouTube
PCVGS retweeted
Clair Obscur: Expedition 33 has sold 8.7 million copies and generated around $366 million in revenue, according to Alinea Analytics.
-58.8% on Steam
-5.5% on Xbox
The game also had 4.3 million Game Pass players, which may have reduced Xbox sales but it also helped more people discover the game.
Make amazing games, sell like crazy? Who would have wondered this
Halloween is upon us. So here is a list of games and movies I recommend to play and watch for this spookiest time of the year!
pcvgs.com/article/halloween-…
PCVGS retweeted
Shigeru Miyamoto is the first from the video game industry to be awarded the "Order of Culture" honor in Japan.
It celebrates contributions to the advancement and development of Japanese culture.
news.denfaminicogamer.jp/int…
PCVGS retweeted
Original source: reddit.com/r/masseffect/comm…
IGN also reports that a Mass Effect 5 story trailer was ready to go:
ign.com/articles/mass-effect…
PCVGS retweeted
en.gamegpu.com/news/zhelezo/…
Hardware benchmark analysis reveals a massive computing gap in INT8 operations between current gaming platforms. The updated Sony system achieves an impressive 300 TOPS, which outperforms the Xbox Series X at 48.6 TOPS and the Xbox Series S at 16 TOPS by a tremendous margin.
This computational advantage stems directly from dedicated neural hardware inside the newer platform. Competing consoles utilize standard compute units without tensor acceleration for running machine learning tasks.
Massive inference capability enables advanced reconstruction algorithms and next-generation upscalers to process high-resolution frames with stable framerates across demanding modern titles.
#PlayStation #Xbox #PS5Pro #Hardware #Gaming #Technology #Console
I signed just for the lol's. I know it won't do anything. But it is funny.
Stop datacenters from monopolizing graphics card supply - Sign the Petition! c.org/yMvtJTXFsC
PCVGS retweeted
🚨 GOD OF WAR RAGNAROK just helped uncover another big KytyPS5 performance bottleneck.
• KytyPS5 was resetting around 8,000 shader states every frame
• Roughly 3 MILLION entries were being scanned per frame
• The new optimization removes those unnecessary repeated scans
• Cuts roughly 2.8–3.5ms of CPU/GPU-thread work in tested scenes
• Delivered about a 1ms frame-time improvement during the sled section
• Validated across 45 MILLION shader materializations without missing an entry
• Testing was done on Linux using a Radeon 8060S
• The fix is still in an open PR
Eliminating millions of pointless operations every frame is the optimization KytyPS5 can benefit from .
Huge credit to NeySlim for digging this deep into the shader recompiler.
PCVGS retweeted
Hey @AMDGaming @GPUOpen @UL_Benchmarks, now with this big milestone of FSR 4.1 maybe it's time to update the 3DMark benchmark? Add FSR 3.1 & 4.1, add FG. The current FSR benchmark is been obsolete and completely useless for a long time already.
