Information Fake Netflix App on Google Play Spreads Malware Via WhatsApp
Posted by: silversurfer - 11 April 21, 11:43 - Forum: Privacy & Security News - No Replies

Quote:Malware disguised as a Netflix app, lurking on the Google Play store, spread through WhatsApp messages, researchers have discovered.
 
According to a Check Point Research analysis released on Wednesday, the malware masqueraded as an app called “FlixOnline,” which advertised via WhatsApp messages promising “2 Months of Netflix Premium Free Anywhere in the World for 60 days.” But once installed, the malware sets about stealing data and credentials.
 
The malware was designed to listen for incoming WhatsApp messages and automatically respond to any that the victims receive, with the content of the response crafted by the adversaries. The responses attempted to lure others with the offer of a free Netflix service, and contained links to a fake Netflix site that phished for credentials and credit card information, researchers said.
 
“The app turned out to be a fake service that claims to allow users to view Netflix content from around the world on their mobiles,” according to the analysis. “However, instead of allowing the mobile user to view Netflix content, the application is actually designed to monitor a user’s WhatsApp notifications, sending automatic replies to a user’s incoming messages using content that it receives from a remote server.”
 
The malware was also able to self-propagate, sending messages to users’ WhatsApp contacts and groups with links to the fake app. To that end, the automated messages read, “2 Months of Netflix Premium Free at no cost For REASON OF QUARANTINE (CORONA VIRUS)* Get 2 Months of Netflix Premium Free anywhere in the world for 60 days. Get it now HERE [Bitly link].”
 
Over the course of two months that the app was live on Google Play, the malware racked up 500 victims, according to Check Point. The firm alerted Google to the malware, which took the app down. However, “the malware family is likely here to stay and may return hidden in a different app,” researchers warned.
 
“The malware’s technique is fairly new and innovative,” Aviran Hazum, manager of Mobile Intelligence at Check Point, said in the analysis. “The technique here is to hijack the connection to WhatsApp by capturing notifications, along with the ability to take predefined actions, like ‘dismiss’ or ‘reply’ via the Notification Manager. The fact that the malware was able to be disguised so easily and ultimately bypass Play Store’s protections raises some serious red flags.”

Read more: Fake Netflix App on Google Play Spreads Malware Via WhatsApp | Threatpost

Print this item

Bug AV-Comparatives: AV-Comparatives Awards 2020
Posted by: harlan4096 - 11 April 21, 09:15 - Forum: Independent Organizations Reports - No Replies

Quote:
[Image: avc-logo.png]

AV-Comparatives’ Awards Ceremony for 2020 honoured the security vendors whose products stood up to the year’s cyber-threats and kept the Internet safe. This year, for obvious reasons, it was held virtually rather than physically, but over 100 participants attended. In total, 27 different vendors were given awards for their security products.

As in previous years, there were major awards for the best Windows security products in the Consumer Main Test Series. This subjects consumer security products to a number of demanding tests, covering different protection scenarios, false positives, and impact on system performance.

The “Product of the Year” award for the best overall results went to Kaspersky, with Bitdefender and ESET receiving “Outstanding Product” awards. Avast and AVG were also honoured, getting “Top-Rated Product” awards.

2020 was an exceptional year in many ways, due to the biological virus known as Coronavirus. This had a knock-on effect with regard to computer viruses too. There was a huge increase in the numbers of people working from home, meaning that businesses needed to worry about their employees’ home-computer security as well. Sadly, cybercriminals exploited the situation by using various Covid-19-related scams to spread malware. Targeted attacks on enterprises and government networks continued too.

In 2020, AV-Comparatives performed more tests than ever before, to keep up with the cybercriminals. Tested products included security programs for Windows consumer devices, macOS and Android; endpoint protection and EPR (endpoint prevention and response) products for enterprises; anti-phishing products; parental control products.

All tested products underwent rigorous testing, using methodologies designed to emulate real-life scenarios. A certification by AV-Comparatives for one of its tests is a guarantee of an effective and reliable product.

In both the Consumer and Enterprise Main Test Series, security products were subjected to a number of different individual tests, to assess their all-round capabilities. These were: protection against Internet-borne malware; protection against malware on the local area network or external device; protection against targeted attacks; false positives; impact on system performance.

In line with its philosophy of public service, AV-Comparatives donated its surplus budget to the charity Red Noses Clown Doctors International.
...
Continue Reading

Print this item

Information Avast_Security_News: Cyberattack targets EU institutions
Posted by: harlan4096 - 11 April 21, 09:10 - Forum: Avast Blog News and Info - No Replies

Quote:
[Image: SNR_04.09.2021_banner.jpg]

Plus, Google launches a trial of its new ad targeting tech and hackers hide malware in a Call of Duty cheat

According to Bloomberg, a spokesperson for the European Commission confirmed that a number of European Union organizations “experienced an IT security incident in their IT infrastructure” last week, adding that no major information breach had been detected as of yet.

The spokesperson would not supply further details on the attack as the investigation is still in its initial stage. An inside source told Bloomberg that the attacks were bigger than average and serious enough to warrant an alert issued to all commission senior officials. Another anonymous source said that commission staff had recently been warned about potential phishing attempts, which suggests the attacks may have been expected. “Government institutions have to face way more risks than the average user,” commented Avast Security Evangelist Luis Corrons. “On top of the day-to-day attacks, they hold valuable information that is a target to cybercriminal groups and foreign intelligence agencies.” 

Google begins trial of new ad targeting tech

This week Google began testing its new ad targeting tech known as the Federated Learning of Cohorts (FLoC) with a random selection of Google Chrome users. The new tech is intended to replace cookies, assigning each user a FLoC ID number instead, which informs advertisers of the user’s behavioral patterns over the previous week, including the websites that were visited. The Chrome users participating in the trial do not know they are participating, and the only way to opt out is to turn off third-party cookies in the browser’s settings. Watchdog group Electronic Frontier Foundation (EFF) strongly criticizes the new tech. Read more on the EFF website

Malware hidden in Call of Duty cheat

Activision published a report this week warning users that hackers have been disguising malware as a cheat for Call of Duty: Warzone. The game is free to play and hosts millions of users. The malware installs a dropper on the user’s system, which is then poised to receive more malware from its command-and-control center. An anonymous source told Vice that one of the malware’s goals is to hijack the user’s computer power and use it to mine cryptocurrency. The hackers selling the malware even posted a tutorial video on YouTube to teach attackers how to use it. In the last year, Activision has banned over 80,000 cheaters from Call of Duty: Warzone.

Arguments for and against regulating Google

Tech giant Google is embroiled in three antitrust lawsuits, including one brought by the U.S. Department of Justice which alleges the company acts like a monopoly and elbows out its competition. The Land of the Giants podcast recently devoted an episode to discussing the pros and cons of breaking up Google. According to Vox, supporters of regulating Google believe that the internet titan has amassed too much influence over daily life and the economy, and that people are essentially forced to use its products. Those against any new regulations claim that because Google’s products are largely free, they are beneficial to consumers, and tighter restrictions could stymie some of Google’s greatest innovations. 

Russia implements Twitter throttling

Researchers learned this week that Russia is attempting a new form of censorship – instead of outright blocking Twitter, it is throttling all network traffic between the social media site and Russian users. Experts believe that by slowing network traffic down to a crawl, Russia is hoping its citizenry will grow too frustrated to use the site. The move came with some unintentional side effects, however. All Twitter content is hosted on t.co, so Russian regulators throttled all domains that had “t.co” in its string. This led to many other sites, including Microsoft.com and Reddit.com to be throttled as well. Read more about it on Ars Technica
...
Continue Reading

Print this item

Lightbulb About Data Encryption Software – And the Dark Side of This Moon
Posted by: harlan4096 - 11 April 21, 09:07 - Forum: Heimdal Security Blog Articles - No Replies

Quote:
[Image: MicrosoftTeams-image-18.png]

Data Encryption is Essential for the Cybersecurity of Your Company. Learn More about Data Encryption Software!

What do German WW2 machines and modern ransomware attacks have in common? Data encryption. How can you keep your company safe from ransomware attacks and various other threats? With data encryption software. How does data encryption software work? Let me explain. 

Data Encryption Software – Definition, M.O., TypesAs the people who developed the Enigma machine in the early 20th century knew, encryption (=the process of encoding information) is essential for protecting data and information. 

Data encryption software uses “a key and an algorithm to transform the accessible data into an encoded piece of information which can only be deciphered by authorized persons”, as illustrated in the image below: 
 
[Image: Public_key_encryption_keys-1-1-1.png]

How does data encryption software works?Encryption scrambles messages by using mathematical algorithms, so that only those who possess the sender’s key or cypher are able to decode the encrypted messages. 

The two main methods of encryption are called symmetric and asymmetric. In the case of symmetric encryption, the data is secured with a single private key. Asymmetric encryption uses a combination of multiple keys, both private and public

As SecurityBoulevard mentions, 
 
Quote:The public key is, as its name implies, public and open to anyone in the system. The public key is used to encrypt data. The private key however is private. It is stored on the user’s device and is used to decrypt data. What’s the difference between a public and private key? The public and private key are really large prime numbers that are mathematically related to one another but are not the same. Being related in this case means that whatever is encrypted by the public key can only be decrypted by the related private key. A person cannot guess the private key based on knowing the public key.

Because of this, a public key can be freely shared. The private key however belongs to only one person.

How many types of data encryption software are there?Data encryption can be achieved in various ways: 
  • Individual file and folder encryption. This type of encryption is used only for certain files and folders. 
  • Volume encryption. Volume encryption implies the creation of a container in which all the files created or saved will be encrypted. 
  • Full disk encryption. As its name suggests, full disk encryption encrypts all the files and folders on the disk and it’s very user friendly because there’s no need to save anything in a special place. 
Data Encryption Software – Ransomware Principle & Examples

As I have mentioned at the beginning of this article, the dark side of data encryption is ransomware. Ransomware is one of the biggest cybersecurity threats of our modern world and  represents
 
Quote:[…] a type of malware (malicious software) which encrypts all the data on a PC or mobile device, blocking the data owner’s access to it. After the infection happens, the victim receives a message that tells him/her that a certain amount of money must be paid (usually in Bitcoins) in order to get the decryption key. Usually, there is also a time limit for the ransom to be paid. There is no guarantee that, if the victim pays the ransom, he/she will get the decryption key. The most reliable solution is to back up your data in at least 3 different places (for redundancy) and keep those backups up to date, so you don’t lose important progress.

Some of the most dangerous ransomware forms that we’ve faced so far were, for example, Wannacry, Epiq, and Maze

WannaCry

WannaCry was a crypto-ransomware that targeted devices running on Microsoft Windows OS, encrypting the victim’s data and requesting payment in Bitcoin in exchange for their decryption. As my colleague Bianca wrote, “WannaCry used an exploit dubbed “EternalBlue”, which took advantage of a security vulnerability that allowed malicious code to propagate without the user’s consent across systems set up for file-sharing.” 

Epiq Ransomware

The Epiq Ransomware was a real team effort, because the attack started with a TrickBot infection that opened the gate to Ryuk actors. Ryuk ransomware targets large companies and government agencies, as well as restaurants, newspapers, public institutions. The first victim of the  TrickBot-Ryuk partnership was Epiq Global, a company that provides legal services to financial institutions and governments from 80 offices worldwide.

Maze Ransomware

Maze Ransomware was another interesting example because it not only encrypted its victims’ data and blocked access to it: the criminals behind Maze also had a public website where they posted the stolen data of the victims who refused to pay the extortion fee. The Maze ransomware cybercrime operation was shut down in September 2020, but, as it usually happens, its affiliates switched to another ransomware, Egregor
...
Continue Reading

Print this item

Bug New REvil Ransomware Version Automatically Logs Windows into Safe Mode
Posted by: harlan4096 - 11 April 21, 09:03 - Forum: Heimdal Security Blog Articles - No Replies

Quote:
[Image: Heimdal-Security-News-and-Updates-1030x360-6.png]

A Recent Change to the REvil Ransomware Allows Threat Actors to Automate File Encryption Via Safe Mode After Changing Windows Passwords.

When it comes to ingenious hacking strategies and techniques, there’s no stopping REvil Ransomware.

Once more, the well-known ransomware has elevated its attack vector to change the victim’s login password in order to reboot the computer into Windows Safe Mode.

While malicious groups are always updating their attack methodology to counter security measures, the threat actors behind the REvil ransomware are particularly adept at honing their malware to make their attack campaigns more efficient.

Last month, security researcher R3MRUM discovered a new sample of the REvil ransomware that refines the new Safe Mode encryption method by changing the logged-on user’s password and configuring Windows to automatically login on reboot. When the -smode argument is used, the ransomware will change the user’s password to ‘DTrump4ever.

Afterward, the ransomware configures the following Registry values for Windows to automatically log in with the new account information.

At the moment it is unknown whether or not the new samples of the REvil ransomware encryptor will continue to use the ‘DTrump4ever’ password, but at least two samples uploaded to VirusTotal in the past two days have done so.

This new tactic illustrates the way ransomware gangs constantly evolve their strategies to successfully encrypt users’ devices and demand a ransom payment.

Just last week, world-leading French EMS company Asteelflash confirmed it has been the victim of a cybersecurity incident, recognizing the involvement of REvil ransomware.

The attackers demanded Asteelflash to pay a whopping $24 million ransom after it was initially set to $12 million in Monero crypto. Because the negotiations didn’t reach a point of agreement in time, the actors raised the ransom to double the amount and leaked the first sample of the exfiltrated files.

REvil also launched a service for contact to news media, companies for the best pressure at no cost, and DDoS (L3, L7) as a paid service. Threat actors, or affiliated partners, will perform voice-scrambled VoIP calls to the media and victim’s business partners with information about the attack.

PC vendor Acer also became the victim of a REvil ransomware attack. The requested ransom might be the largest one to date, REvil asking for $50 million.
...
Continue Reading

Print this item

Information New Algorithm Makes CPUs 15 Times Faster Than GPUs in Some AI Work
Posted by: harlan4096 - 11 April 21, 08:46 - Forum: Hardware News - No Replies

Quote:
[Image: o6R5GnN6o45NnzVBuhB2DJ-1024-80.jpg.webp]

CPUs can beat GPUs in some AI workloads

GPUs are known for being significantly better than most CPUs when it comes to AI deep neural networks (DNNs) training simply because they have more execution units (or cores). But a new algorithm proposed by computer scientists from Rice University is claimed to actually flip the tables and make CPUs a whopping 15 times faster than some leading-edge GPUs. 

The most complex compute challenges are usually solved using brute force methods, like either throwing more hardware at them or inventing special-purpose hardware that can solve the task. DNN training is without any doubt among the most compute-intensive workloads nowadays, so if programmers want maximum training performance, they use GPUs for their workloads. This happens to a large degree because it is easier to achieve high performance using compute GPUs as most algorithms are based on matrix multiplications.  

Anshumali Shrivastava, an assistant professor of computer science at Rice's Brown School of Engineering, and his colleagues have presented an algorithm that can greatly speed up DNN training on modern AVX512 and AVX512_BF16-enabled CPUs. 

"Companies are spending millions of dollars a week just to train and fine-tune their AI workloads," said Shrivastava in a conversation with TechXplore. "The whole industry is fixated on one kind of improvement — faster matrix multiplications. Everyone is looking at specialized hardware and architectures to push matrix multiplication. People are now even talking about having specialized hardware-software stacks for specific kinds of deep learning. Instead of taking a [computationally] expensive algorithm and throwing the whole world of system optimization at it, I'm saying, 'Let's revisit the algorithm.'" 

To prove their point, the scientists took SLIDE (Sub-LInear Deep Learning Engine), a  C++ OpenMP-based engine that combines smart hashing randomized algorithms with modest multi-core parallelism on CPU, and optimized it heavily for Intel's AVX512 and AVX512-bfloat16-supporting processors. 

The engine employs Locality Sensitive Hashing (LSH) to identify neurons during each update adaptively, which optimizes compute performance requirements. Even without modifications, it can be faster in training a 200-million-parameter neural network, in terms of wall clock time, than the optimized TensorFlow implementation on an Nvidia V100 GPU, according to the paper. 

"Hash table-based acceleration already outperforms GPU, but CPUs are also evolving," said study co-author Shabnam Daghaghi. 

To make hashing faster, the researchers vectorized and quantized the algorithm so that it could be better handled by Intel's AVX512 and AVX512_BF16 engines. They also implemented some memory optimizations.  

"We leveraged [AVX512 and AVX512_BF16] CPU innovations to take SLIDE even further, showing that if you aren't fixated on matrix multiplications, you can leverage the power in modern CPUs and train AI models four to 15 times faster than the best specialized hardware alternative."

The results they obtained with Amazon-670K, WikiLSHTC-325K, and Text8 datasets are indeed very promising with the optimized SLIDE engine. Intel's Cooper Lake (CPX) processor can outperform Nvidia's Tesla V100 by about 7.8 times with Amazon-670K, by approximately 5.2 times with WikiLSHTC-325K, and by roughly 15.5 times with Text8. In fact, even an optimized Cascade Lake (CLX) processor can be 2.55–11.6 times faster than Nvidia's Tesla V100.
...
Continue Reading

Print this item

Information Nvidia's Virtualization Unlocked On Gaming GPUs via Hack
Posted by: harlan4096 - 11 April 21, 08:44 - Forum: Hardware News - No Replies

Quote:
[Image: jk6rLcyfthMceB6u9ewCvd-1024-80.jpg.webp]

First the mining limiter, now virtualization has been hacked

A group of enthusiasts has unlocked vGPU (GPU virtualization) capability, which is only supported on select datacenter and professional boards, on standard consumer Nvidia GeForce gaming graphics cards. Since the vGPU capability is supported by the silicon but locked out by software, it was only a matter of time and effort before enthusiasts unlocked the feature. As it turns out, according to a Reddit post, that time has come, potentially saving some users the thousands of dollars they would otherwise have to shell out for a Quadro or Tesla GPU that supports the feature.

GPU virtualization, which allows more than one user to use a GPU simultaneously, is one of the differentiators between GPUs for data centers and those designed for consumer PCs. Nowadays, many workstations and even high-end desktops are located remotely so the users can share the GPUs. Modern hardware is so powerful that its performance is sometimes excessive for one user, so sharing one graphics card between multiple users makes sense.

From a GPU hardware perspective, virtualization is just another feature, so the silicon supports it. But this capability requires a lot of software to work properly (i.e., how companies that buy workstations expect it to) and validation with ISVs since virtualized GPUs are in many cases used for professional applications.

All of these things cost money, so vGPU support comes at a price, and Nvidia has a handful of expensive Tesla, Quadro, and some other GPUs it recommends for virtualization (partly because it does not make a lot of sense to validate a broad fleet of hardware with ISVs). Nvidia's vGPU software does not support most client GPUs. Nvidia did recently open up GPU passthrough via a driver update, but this only allows a single VM to access the GPU, whereas full vGPU support allows multiple VMs to share the same GPU.

The code for the vGPU unlocker is available at Github, and the principle behind it is fairly simple: it replaces the device id of a graphics card with a device id of an officially supported GPU that has the same feature set. For now, GP102, GP104, TU102, TU104, and GA102 GPUs are supported, and the capability works on Linux and with KVM virtual machine software. 

While the new unlocking technique deserves some attention, the big question is whether your typical consumer actually needs GPU virtualization. Linux users can virtualize their high-end graphics cards and use them for gaming, video encoding, and cryptocurrency mining simultaneously on different virtual machines. 

Some of those who happen to have servers with hundreds or thousands of consumer Nvidia GPUs could try to offer commercial remote desktop services to earn money, but the quality of such services would be something to worry about. Since the hack does not work with Windows and Vmware, it is useless for most users. 
...
Continue Reading

Print this item

Lightbulb Intel Core i9-11900K and Core i5-11600K Review: Rocket Lake Blasts Off
Posted by: harlan4096 - 11 April 21, 08:41 - Forum: Hardware News - No Replies

Quote:
[Image: BHvGVYdvVK56WuFsZJS8ye-1200-80.jpg.webp]

14nm flies again

Our Verdict

The Core i9-11900K is impressive in gaming and lightly-threaded work, but it trails the Ryzen 9 5900X tremendously in threaded work and doesn't cement itself well enough in gaming to justify the premium price tag.

For
  • + Snappy single-threaded performance
  • + Gaming performance
  • + Overclocking enhancements
  • + PCIe 4.0
  • + Faster iGPU
Against
  • - High pricing
  • - High power consumption
  • - Needs CPU cooler
  • - Eight cores
  • - Lackluster threaded performance
  • - Gear 2 Memory Mode
  • - Limited PCIe 4.0 Support
Intel's 11th-Gen Rocket Lake processors have finally been cleared for liftoff, with the eight-core $539 Core i9-11900K taking on AMD's potent Ryzen 9 5900X that leads our CPU Benchmark hierarchy, while the six-core $262 Core i5-11600K slots in with more palatable pricing as the mainstream gaming chip to challenge AMD's Ryzen 5 5600X, our current Best CPU for gaming.

The 14nm Rocket Lake family arrives during dark times for Intel in the desktop PC market. AMD's Zen-fueled assault on the desktop has culminated in its powerful Zen 3-powered Ryzen 5000 chips taking the uncontested lead in nearly every metric that matters, including in Intel's long-held gaming stronghold.

Intel says Rocket Lake will retake the gaming crown, but the devastating fallout of the company's failure to transition to 10nm desktop PC chips now ripples through a fifth generation of its processors. As a result, Rocket Lake comes fabbed on the seventh and presumably final iteration of Intel's 14nm process for desktop processors. This after having soldiered on since 2015 as Intel's longest-lived leading-edge node.

However, Rocket Lake comes with a powerful new addition — Cypress Cove, Intel's first new architecture for desktop PC chips in six years, which Intel says grants a 19% increase in IPC. But Cypress Cove comes with a big tradeoff: Rocket Lake tops out at eight cores and sixteen threads, a step back from the previous-gen 10-core Comet Lake i9 models that pales in comparison to AMD's beastly 16-core Ryzen 9 5950X flagship. 

Intel Rocket Lake Flagships
Suggested Price Cores / Threads Base (GHz) Peak Boost (Dual/All Core) TDP iGPU

RKL-S Core i9-11900K (KF) $539 (K) - $513 (KF) 8 / 16 3.5 5.3 / 4.8 125W UHD Graphics 750 Xe 32EU
RKL-S Core i7-11700K (KF) $399 (K) - $374 (KF) 8 / 16 3.6 5.0 / 4.6 125W UHD Graphics 750 Xe 32EU
RKL-S Core i5-11600K (KF) $262 (K) - $237(KF) 6 / 12 3.9 4.9 (TB2) / 4.6 125W UHD Graphics 750 Xe 32EU

The lowered core count is an unavoidable side effect of Intel basing its new architecture on a design that originally debuted with its 10nm Ice Lake processors, but the decision to etch it onto the 14nm process left its chip designers with a leaner transistor density budget that ultimately resulted in fewer cores.

Intel says this tactic, which is called backporting, was necessary because 10nm couldn't support the higher frequencies needed for desktop PCs. Intel claims the increased IPC and faster frequencies should offset the reduced core count in most work. However, as you'll see in our testing, the reduced core count can lead to lower gen-on-gen performance in some heavily-threaded applications. That certainly hasn't stopped Intel from charging a premium for its new 11900K flagship, though, which comes with a $51 upcharge over the prior-gen model.

Rocket Lake has plenty of notable advances, though: Intel stepped forward to faster DRAM speeds (though that comes with a big caveat), finally adopted the PCIe 4.0 interface, added AVX-512 support and AI-boosting DL Boost technology, and also moved to the integrated UHD Graphics 750 engine that hails from the company's 10nm Tiger Lake chips. Rocket Lake also has other gaming optimizations with added support for Resizable Bar, which boosts gaming performance with supported discrete GPUs. The chips also cater to the enthusiast crowd with a host of overclocking features to wring more performance from the silicon, including unlocked memory overclocking with cheaper B- and H-series motherboards.

However, while much of AMD's successful formula has consisted of more cores, a newer architecture, and a denser 7nm node, Intel has launched its new flagship on an older, less-efficient 14nm node with fewer cores. As a result, Intel has attempted to offset the reduced core count by dialing power consumption to the extreme to maximize performance. The Core i9-11900K is impressive in gaming and lightly-threaded work, but it trails the similarly-priced Ryzen 9 5900X by large margins in threaded applications and doesn't cement itself well enough as a gaming leader to justify its premium price tag. 

Meanwhile, the six-core twelve thread Core i5-11600K lands with a much friendlier $262 price point that's much more competitive with AMD's comparable chips. In light of its price point, it has a very competitive price-to-performance ratio with the $300 Ryzen 5 5600X in a broad swath of games and applications. While the Core i5-11600K might not claim absolute supremacy, its mixture of price and performance makes it a solid buy. Provided, of course, that you can find any of these chips at close to sane pricing. 
...
Continue Reading

Print this item

Lightbulb AMD's Ryzen 9 5900 Blurs The Line Between Performance And Value
Posted by: harlan4096 - 11 April 21, 08:31 - Forum: Hardware News - No Replies

Quote:
[Image: tfJsnVZ5SHwkhhxgdw67gb-1024-80.jpg.webp]

Unattainable Zen 3 monster

Hardware detective Tum_Apisak has spotted AMD's Ryzen 9 5900 in a new benchmark. The 12-core processor appears to be a complete monster, but it's a shame that the chip is exclusive to OEMs.

Coming out of TSMC's 7nm furnace, the Ryzen 9 5900 is just like any other Ryzen 5000 (Vermeer) processor. Wielding the mighty Zen 3 cores, the Ryzen 9 5900 is essentially the power-optimized version of the Ryzen 9 5900X. The Ryzen 9 5900 retains the same 12-core, 24-thread design with 64MB of L3 cache as its counterpart. Unlike the Ryzen 9 5900X's 105W TDP, the non-X model is restricted to 65W.

With a stricter power limit, the Ryzen 9 5900 obviously arrives with lower operating clock speeds. The processor has a 3 GHz base clock, which is 700 MHz lower than the Ryzen 9 5900X. The boost clock, on the other hand, didn't suffer a huge reduction. The Ryzen 9 5900 easily boosts to 4.7 GHz, only 100 MHz lower than the Ryzen 9 5900X.

It's normal to wonder how the Ryzen 9 5900 performs in comparison to the Ryzen 9 5900X given the lower TDP. Being an OEM-exclusive processor, it's hard to find benchmarks of the Ryzen 9 5900 until Tum_Apisak managed to unearth one for us. UserBenchmark doesn't have the best street cred, but it's what's available so we'll have to settle for that benchmark for now.

It's important to highlight that the Ryzen 9 5900X's results are based on over 30,000 user submissions, while the Ryzen 9 5900 results is from a single user.

For now, approach the results with an open mind.

The Ryzen 9 5900X reportedly delivers 5% higher single-threaded performance than the Ryzen 9 5900. As for multi-threaded performance, the Ryzen 9 5900X seemingly outperformed the Ryzen 9 5900 by a margin of 7%. However, the Ryzen 9 5900X didn't win all the rounds of the fight. The Ryzen 9 5900 appears to have a 7% lower memory latency as opposed to the Ryzen 9 5900X.

With all things considered, the Ryzen 9 5900 is a very attractive processor, and it's a mystery while AMD won't offer it to the retail market. The Ryzen 9 5900X has a 61.5% higher TDP, and it's less than 10% faster than the Ryzen 9 5900. Of course, we'd have to put the Ryzen 9 5900 through its paces to see whether the performance delta from UserBenchmark is valid.

Officially, the Ryzen 9 5900 is only available to OEMs, but that's what AMD also said with its Ryzen 4000 (Renoir) APUs, and we found them all over the place. It's only a matter of time before the Ryzen 9 5900 hits the grey market if it hasn't already.
...
Continue Reading

Print this item

Information AMD's Zen 3, Navi 2 APUs With DDR5 Support May Arrive Next Year
Posted by: harlan4096 - 11 April 21, 08:27 - Forum: Hardware News - No Replies

Quote:
[Image: vC3MWTREqtqXmFXKx3fpF7-1024-80.jpg.webp]

The best is yet to come

A Twitter user has shared an alleged AMD roadmap that details the chipmaker's plans for its next-generation mobile APUs. Respected hardware leaker kopite7kimi got word that the roadmap looks accurate. Nevertheless,  approach the information with caution since roadmaps are known to change as time progresses.

The roadmap reveals AMD's upcoming mobile APUs for four categories: HP (High Power), LP (Low Power), ULP (Ultra-Low Power) and ELP (Extreme Low Power). Starting from the top, Ryzen 6000 (Rembrandt) will reportedly land next year to replace the existing Ryzen 5000 (Cezanne) lineup. If the the data is accurate, Ryzen 6000 should be based on TSMC's 6nm manufacturing process.

On a microarchitectural level, Ryzen 6000 will utilize an optimized version of Zen 3. The Zen 3+ parts will likely arrive with optimizations in the form of a small frequency bump. According to the leaked roadmap, Ryzen 6000 will support DDR5 and LPDDR5 memory. In terms of graphics, Ryzen 6000 will finally leverage AMD's Navi 2 (RDNA 2) graphics engine, and it'll be the first APU to do so. As usual, Ryzen 6000 will compete in the 45W and 15W power brackets with Rembrandt-H and Rembrandt-U, respectively.
 
[Image: 8GBYGWDzxfeBUJrgKiLvrm.jpg]

In the LP class, AMD currently utilizes the Ryzen 5000 branding for both its Zen 3 (Cezanne) and Zen 2 (Lucienne) chips. Lucienne-U can be considered a refresh of Renoir, though AMD isn't fond of people calling it that. Lucienne-U uses the Zen 2 cores like its predecessor. It would appear that AMD will replace Lucienne-U in 2022 with Barcelo-U. Barcelo-U will represent a huge upgrade since AMD will endow it with Zen 3 cores. Sadly, the graphics engine is still Vega.

AMD's Van Gogh APUs, which compete in the ULP (Ultra-Low Power) segment, will debut sometime this year. Linux patches have revealed that the Zen 2 chips will sport Navi 2 graphics and support LPDDR5 memory. Dragon Crest is the rumored successor to Van Gogh. Since Dragon Crest seems to utilize the same recipe as Van Gogh, it's rational to assume that Dragon Crest could be a simple refresh.

Apparently, AMD doesn't have anything planned for the ELP category. Pollock, which is based on the Zen and Vega microarchitectures, will continue push forward through 2022.
...
Continue Reading

Print this item

[-]
Welcome
You have to register before you can post on our site.

Username/Email:


Password





[-]
Recent Posts
Website X5 Go 2024.1
Website X5 Go 2024.1...zevish — 09:32
Apple's rules to allow third-party app ...
Apple has announ...alison30 — 09:28
Intel: Microsoft AI PCs need a Copilot K...
Microsoft hopes th...harlan4096 — 08:55
Synchredible 8 Professional Edition v8.2...
          Synchredib...zevish — 08:54
GFYI [Official] EaseUS Data Recovery Wi...
I utilize EaseUS Par...zevish — 08:10

[-]
Birthdays
Today's Birthdays
No birthdays today.
Upcoming Birthdays
No upcoming birthdays.

[-]
Online Staff
There are no staff members currently online.

>