Is Windows misbehaving? Maybe it's freezing, crashing, or just not performing up to snuff. Diagnosing a general problem like this can be challenging. It could be software. It could be hardware. But one potential perpetrator you may not immediately think of is faulty memory.

RAM chips can go bad just like any other physical component, but you can scrutinize it by running a memory diagnostic. If your memory passes the test, you can rule out this problem. If it fails, you know one of your RAM modules likely needs to be replaced.

To scan your memory, turn to Microsoft's free Windows Memory Diagnostic, which runs a variety of tests on your PC's RAM to see if everything is in working order. It's built into all versions of Windows from the past several years, including Windows 7, Windows 8/8.1, Windows 10, and Windows 11. And it works the same in each version. Let's give it a whirl.

Open the Windows Memory Diagnostic Tool

First, press Win+R. At the Open field in the Run window, type mdsched.exe and click OK. Up pops the opening screen for the Windows Memory Diagnostic Tool. The tool needs to run when your PC starts up, so you have two options. You can restart your computer now, and the tool will execute at bootup. Or you can schedule the tool to run the next time you reboot your PC. Let's choose the option to restart your computer now.

Your PC restarts and launches the Memory Diagnostic Tool's RAM test. Right off the bat, the tool starts checking your memory for potential problems. By default, the Windows Memory Diagnostic begins in Standard mode, which includes eight different, successive memory tests, each of which uses a unique algorithm to scan for different types of errors. You can allow the Standard tests to run until completion, after which time Windows will restart.

Alternatively, you can press F1 at any time during the test. That takes you to a different screen where you can choose the type of scan you wish to run. The tool offers three types of scans: Basic, Standard, and Extended.

Basic runs only three tests, but it runs them very quickly. Standard runs the eight different tests but takes longer to run. And Extended runs 17 different tests and takes a long time to complete. Your best option is to stick with the Standard scan. If a glitch is detected, you can run the Extended scan to see if further details pop up.

At this window, you can also enable or disable the cache. Turning off the processor's cache during the scan forces the tool to access data directly from RAM, ensuring more thorough testing. You can leave this at the default value, which means the cache setting for each test remains as is. Finally, you can set the Pass Count, meaning the number of times each test will run. Two is the default value and should be sufficient to unearth any problems. After you’ve made your selections, press F10 to apply them and run the test.

After each test completes, the tool displays a Succeeded or Failed message. A Failed message is followed by details about the failure and typically indicates a fault with one of your RAM modules. Depending on your PC environment, the diagnostic may be able to narrow down a memory problem to a specific module. Also, if your PC contains more than one memory module, you should be able to remove each of the modules one at a time and run the Memory Diagnostic to see if you can pinpoint the errors to a certain module.

After the scan runs, your PC reboots. Log back into Windows and you'll see a small message briefly pop up, giving you the results of the scan.

You can also check memory test results in Windows using the Event Viewer. Press Win+R. At the Open field in the Run window, type eventvwr.exe and click OK; Event Viewer launches.

Expand the category for Windows Logs. Right-click on the System setting and then select Find from the pop-up menu.

In the Find what field, type MemoryDiagnostics-Results and click Find Next.

Click Cancel on the Find window to close it. In the Event window, double-click the first highlighted entry, which should be the test you just ran. The results will provide more details on the scan.

That’s according to a survey of business customers using Microsoft 365 for email commissioned by Cyren and conducted by Osterman Research, which examined concerns with phishing, business email compromise (BEC), and ransomware threats, attacks that became costly incidents, and preparedness to deal with attacks and incidents.

“Security team managers are most concerned that current email security solutions do not block serious inbound threats (particularly ransomware), which requires time for response and remediation by the security team before dangerous threats are triggered by users,” according to the report, released Wednesday.

Less than half of those surveyed said that their organizations can block delivery of email threats. And, correspondingly, less than half of organizations rank their currently deployed email security solutions as effective.

Protections against impersonation threats are viewed as least effective, followed by measures to detect and block mass-mailed phishing emails.

Thus, it’s perhaps no surprise that almost all of the organizations polled have experienced one or more types of email breaches.

In fact, 89 percent of organizations experienced one or more successful email breach types during the previous 12 months. And, the number of email breaches per year has almost doubled since 2019, according to the report, most of them due to successful phishing attacks that compromised Microsoft 365 credentials.

Overall, according to the survey, successful ransomware attacks have increased by 71 percent in the last three years, Microsoft 365 credential compromise increased by 49 percent and successful phishing attacks increased by 44 percent.

Ineffective Defensive Approaches

Digging into where email defense breaks down, the firms found that, surprisingly, use of email client plug-ins for users to report suspicious messages continues to increase. Half of organizations are now using an automated email client plug-in for users to report suspicious email messages for analysis by trained security professionals, up from 37 percent in a 2019 survey.

Security operations center analysts, email administrators, and an email security vendor or service provider are the groups most commonly handling these reports, although 78 percent of organizations notify two or more groups.

Also, user training on email threats is now offered in most companies, the survey found: More than 99 percent of organizations offer training at least annually, and one in seven organizations offer email security training monthly or more frequently.

“Training more frequently reduces a range of threat markers Among organizations offering training every 90 days or more frequently, the likelihood of employees falling for a phishing, BEC or ransomware threat is less than organizations only training once or twice a year,” according to the report.

Further, the survey found that more frequent training results in more messages being reported as suspicious, and a higher share of these suspicious messages proving to be malicious after analysis by a security professional.

So far so good. So where’s the breakdown? One concerning finding: Only about a fifth (22 percent) of organizations analyze all reported messages for maliciousness.

“How employees should determine the maliciousness of reported messages by themselves when they do not receive a verdict from security professionals is unclear,” according to the firms.

Across the board, the survey also showed that organizations using at least one additional security tool to complement the basic email protections offered in Microsoft 365. However, their implementation efficacy varies, the survey found.

“Additive tools include Microsoft 365 Defender, security awareness training technology, a third-party secure email gateway or a third-party specialized anti-phishing add-on,” the report explained. “There is a wide range of deployment patterns with the use of these tools.”

The firms concluded that these kinds of holes and ineffective defenses in general translate into major costs for organizations.

“Costs include post-incident remediation, manual removal of malicious messages from inboxes, and time wasted on triaging messages reported as suspicious that prove to be benign,” according to the report. “Organizations face a range of other costs too, including alert fatigue, cybersecurity analyst turnover and regulatory fines.”

Well, is it? It's not a question easily answered by anyone but there are those in the know with a better idea of what's going on. That's why I've had the pleasure of once again speaking with Dr. Thomas Goldsby, the Haslam Chair of Logistics at the University of Tennessee’s Master's of Science in Supply Chain Management online program.

Last time we spoke we were in one of the worst periods of shortages for GPUs, but this time things are looking up, and I want to know whether a certified supply chain expert like himself also feels like the situation is improving.

"I do think that things are better," Dr. Goldsby says. "You may recall that we broke the problem down in terms of supply and demand. And most people focus on the supply part, but you cannot overlook the demand implications a pandemic when people were altering lifestyles, altering work, play and entertainment.

"And that certainly led to this incredible surge for all things electronic, but on the supply side of things it also affects us in our ability for people to show up to work, to feel confident in the work that they perform."

Between the last time I spoke with Dr. Goldsby and now, the new variant of the Covid-19 virus, known as Omicron, surged across the globe. Some areas still have high cases of Covid-19, including Hong Kong, which has led to lockdowns and similar measures in China's high-production and shipping areas. However, many restrictions and measures for mitigating Covid-19 have since been lifted or removed.

"By virtue of Omicron presenting this ugly head and fortunately starting to relent. It helps us on both sides. It helps us in terms of people getting back and starting to live a little bit more in a sense of normal life. And then also our ability of our supply chains to perform at a kind of pre-Covid level. And so there is some normalising, which I think would help to explain why maybe supplies a bit improved, maybe prices are starting to relent a bit.

"I am seeing some glimmers of hope, just as you are," Goldsby continues.

That insight might help to explain why graphics card stock levels have appeared to increase in recent weeks. However, prices have yet to drop to their MSRPs with this new influx of cards. 

AMD's RX 6500 XT is often spotted at or under its $199 MSRP—few others are ever so close to affordable. Though it appears prices are at least much lower than they have been over the past 12 months.

WHEN WILL PRICES COME DOWN?

"I would expect prices to come down when it becomes apparent that supply is catching up and that they are readily available," Goldsby says. "A supplier out there will get nervous that they are sitting on considerable supply for an item that has a limited shelf life (with newer cards entering the market) and will make the move to clear out that excess inventory.

"Peer suppliers will then have to follow that lead. And back to equilibrium we fall."

What we cannot discount is the apparent falling demand from other GPU users, largely cryptocurrency miners. The past few years have been mighty profitable for cryptocurrency miners, or at least those with cheap enough electricity, and that has seen them snap up graphics cards for profit. That is still often a viable option today, but there's no denying that the profitability of ethereum mining especially has dipped for most, which may have seen a dip in interest for brand new gaming GPUs.

Similarly, there could be a general dip in interest for technology as Covid-19 measures are sidelined for now.

"Just how much stuff including hardware, including electronics, do we need?" Dr. Goldsby says. "You can look at some interesting segments: Peloton did the exercise bike—did they think that every household was going to have two Pelotons? I don't know, but that market kind of saturated and people realised, 'oh, I guess I can get out and ride a real bicycle. I don't have to be captive to my basement or something.'

"You know, you just kind of reach a point of saturation. And that's true with any product, it's only going to be appealing to a certain subset of the population. And once they get theirs, you know, they're kind of happy."

This also comes down to the Bullwhip Effect, which says that small increases in demand will be interpreted as much larger than they are once they get further down the supply chain. You know, like a whip: from a small movement results a loud crack.

Customer demand must be forecast by the retailer, distributor, and manufacturer. Since demand is rarely static, there's a relative imbalance at any point during the supply chain. Sometimes you won't really notice it as a consumer, when supply chains aren't in crisis, but when they are you may find yourself paying more attention to the ebb and flow of supply.

"The other one is something I find fascinating, I was teaching this with my students last night in class, is an 11-year aged Tennessee whiskey. That's a tough one to forecast, because the stuff you're putting in a barrel today, you're not selling until 2033."

Now think of a graphics card supply chain. These PC components have thousands of components and not all sourced from anything close to a single supplier. The chip, the capacitors, the MosFETs come from various locations around the world and are assembled into a single unit someplace else. These make not only forecasting for demand difficult, but it's also not an overnight job to suddenly adjust production to compensate.

If a baker sells more croissants one day, they might make more the next. If Nvidia can't meet the orders of its partners, and that demand is sustained (which is a whole other risk assessment it will need to make), it might be a long time before they meet that demand as it has to reorganise production and manufacturing.

"I think folks have just started to kind of come to grips with the fact that you can't just snap your fingers and produce or deliver a product. It takes an incredible amount of planning and execution to pull that off. It's a simple premise, but anything but simple to execute."

This is especially true of high-performance computer parts, and no more so the GPU. The prime component in our PCs for performance today, these are also often some of the most dependent on cutting-edge process nodes. Those of which are in incredible demand. CPUs, too, are often at the whims of the latest 7nm, 5nm, 3nm process nodes, which AMD often champions with its Ryzen processors built in TSMC's fabs, though Intel is also making significant progress towards catching up with new fabs in the US and Europe.

"It's among the slowest forms of production to change," Dr. Goldsby says. "... you announce that you're going to embark on a new fab, and maybe three years from now you're actually producing chips, in good quality. So some things just can't turn on a dime. And certainly high tech among the toughest."

GRAPHICS CARD SUPPLY IS IMPROVING

Though Goldsby's words are hardly an indictment of any improvement to the market we're seeing today. TSMC chairman Mark Liu today noted that it appears consumer demand for electronics is starting to slow, and UK retailer Box told our friends at TechRadar that it expects "things to return to a more attractive price" by the end of April/beginning of May.

During my regular rounds of browsing the web for graphics cards in stock and at a decent price, I have begun to notice a gluttony of GPU stock at most major retailers. The decent price bit is usually missing, but they are dropping a touch in that regard. And if this level of stock can be maintained, price drops are sure to follow.

There are signs of improvement in terms of semiconductors generally, and that's promising for graphics card supply in 2022. Though for all these heartening signs, we are still not out of the woods yet. GPU prices remain above MSRP and major chipmakers, such as Intel, have voiced their concerns of tight supply until the lights flicker on at their new fab facilities, and those like them built by competitors. 

At least through times of difficult supply, manufacturers are reportedly learning to adapt, and perhaps there's a lesson being learned for today's lean inventory supply chains in avoiding such a crisis in the future. General Electric's additive manufacturing is one example of a company learning to do more with less reliance on supply links that could one day jeopardised its own profits.

"It's kind of a neat example where having the vision of a problem led them to innovate in very meaningful ways," Goldsby says.

But just as much so are technology and automotive companies looking into ways to shore up their supply chains and get as much of their supply chain in-house or close to it as possible.

"As a friend of mine reminds me, every business decision we make virtually manifests in the supply chain. And I think that CEOs have realised that hey, you know, supply chains aren't just the things that we say, 'hey, produce this product and do it at this cost.'"

"Experts have to be privy to, and be participating in the conversation, about what products we devise, how much it's realistically going to cost, and what are the prospects for actually being able to deliver those products."

Can we avoid a semiconductor shortage in the future? Only time will tell, but our brush with flaky supply over the past few years has shown how reliant we are on these clever blocks of sand. Graphics cards being just one part of a wider industry shaken by global events, and all will no doubt be taking note of how to avoid the worst of it in the future.

WHEN WILL WE KNOW SUPPLY IS BACK TO NORMAL?

In the here and now, I wanted to ask Dr. Goldsby one last time if he believed everything would come back to pre-pandemic supply.

"We're reading the same things you are, you know, particularly as I look at gaming systems and semiconductor, they're saying that 2022 is going to continue to be a challenge, probably into 2023. These are going to be certain sectors that are just going to take longer than others. And for the reasons we discussed earlier, I do think that, you know, consumer electronics are going to tend to lag a bit."

We’re now six months into the release of Windows 11. You’ve probably asked yourself at least once, “Should I consider upgrading my devices yet?”

Maybe you’re somewhat skeptical of new OS updates, especially of entirely new builds. Oftentimes new OS builds are buggy, and the kinks haven’t quite been worked out just yet. Plus, you’re considering if the investment is worth the payout – Windows upgrades at scale can be costly, both in time and capital. We’ve helped plenty of our clients to do them, so we know this is a reality.

And so here you are, wondering whether to dive into this new OS.

In this post we’re going to give you our take on this question as of March 2022. We’re going to specifically consider three relevant topics related to the new Windows 11 build, particularly:

• Security Features
• User Experience (User Interface)
• Required Specifications*

Then we’ll reflect on who we think should consider upgrading to Windows 11 at this point and who can ride the Windows 10 train for another year or two.

*Note: There are many factors to consider when considering an upgrade, and we have not covered all of them in this post. Instead, we’ve focused our analysis on those that we think are relevant to endpoint management professionals who are making a decision to upgrade or not. We encourage anyone to consider this post along with other assessments that they may read elsewhere and not restrict their decision to just the three categories covered in this post.

Security Features

With the rapid move into cloud-based and remote work, security is more top-of-mind than ever. Several recent high-profile data breaches have increased national awareness of the importance of cybersecurity, and it’s easy to see that the COVID-driven shift to remote work has helped shape Windows 11 both in features and in marketing.

According to Microsoft, Windows 11 is “the most secure version of Windows ever.” Windows 11 provides “chip-to-cloud” enhanced security to keep up with the rapid growth or remote work in the modern workplace. According to Microsoft, “Every component of the Windows 11 technology stack, from chip-to-cloud, is purposefully designed to help ensure ultimate security.”

One particularly notable set of features is related to hardware-based security. Microsoft writes, “we have worked with our chip and device manufacturing partners to integrate advanced security capabilities across software, firmware, and hardware to create tight integration that protects from the chip to the cloud.”

Some of these notable features are:

• Trusted Platform Module (TPM) 2.0: Trusted Platform Module is a firmware root-of-trust that enables security and privacy improvements for system hardware. The 2.0 version, according to Microsoft, includes ” important enhancements” that “enables stronger crypto algorithms and the ability for customers to use preferred alternative algorithms.”
• Virtualization-Based Security (VBS): VBS manages OS kernel-based attacks by isolating a secure region of memory and inhibiting malware from executing code or accessing information inside this secure environment.
  • Hypervisor-protected code integrity (HVCI): HVCI runs Kernel Mode Code Integrity inside the secure VBS region rather than the main Windows Kernel. This prevents malicious kernel-mode code from being injected even when drivers and other kernel-mode software have been compromised.
• Pluton Security Processor Support: Support for a security-oriented processor series. It allows for a “a flexible, updatable platform for running firmware that implements end-to-end security functionality authored, maintained, and updated by Microsoft.” Updates are delivered through Windows Updates rather than a third-party alternative.
• Secured-Core PCs Support: Supports the Secured-core PC series shipped for high-risk industries with additional protection enabled at the firmware layer. These PCs include features such as memory protection, firmware protection, and Dynamic Root of Trust for Measurement (DRTM).

But are these hardware-based security features worth the cost to upgrade your devices?

When you look closely at these highlighted features, what you’ll notice is that many of them are already available in Windows 10.

For example, TPM 1.2 and 2.0 were optional in Windows 10. TPM 2.0 in Windows 11 is now required, and TPM 1.2 is no longer available. This is an improvement to security as the feature is now standardized. However, unless your organization has implemented SecureBIO or Auotpilot in a self-deploying mode , simply standardizing a feature lacks the kind significant improvement that would warrant an upgrade for most teams from Windows 10 this early in the game.

The same is the case with other security features including VBS, HVCI, Secure Boot, and Container Isolation in Microsoft Office and Edge. Each of these features is optional in Windows 10 and comes enabled out-of-the-box in Windows 11.

This is also true of some of the Cloud Security features. While integration with Microsoft Azure AD and Modern Endpoint Management in Endpoint Manager has been improved, most companies who use Microsoft Intune and Endpoint Manager already have access to many of their features using Windows 10.

All this considered, Windows 11 is more-secure out of the gate because of the standardization of many security features. However, they may not be significant enough to warrant most IT professionals to upgrade this early in the life cycle of the OS. If, on the other hand, you’re looking to increase the “default” or “baseline” security posture, Windows 11 would be a way to do that.

User Experience / User Interface

User interface (UX) is also important to IT pros. One of your goals is to enable productivity of your end users, and Microsoft wrote that their goal is to “empower every person and every organization on the planet to achieve more.” UX design is a part of this.

As such, the interface has been significantly updated in Windows 11 to optimize productivity and user experience. In place of the Windows 10 interface, Microsoft has opted for a new UX that prioritizes simplicity and ease of use.

Here’s just a few of the features that have been changed:

• Start Menu: Redesigned to be centered, app-centric, and easily customized to access your most-used apps. Customized start menu layouts can also be deployed to endpoints using policies.
• Taskbar & Active Applications: The taskbar has undergone a significant redesign, apps and widgets can now be pinned, and it can also be customized and deployed using policies as well.
• Virtual desktops: A feature available in Windows 10, it’s now standard on the Windows 11 taskbar, and they can be easily deployed with a single click.
• Windows File Explorer: The first significant update since Windows 8, the file explorer has a simplified, modern makeover.

Some of the additional features that were updated are:

• Alert boxes
• Settings menu (Windows Key +A)
• Search
• Widgets Panel (updated News feed)
• Task view/Snap assist updated features

The overall feel of the new UX is that it’s warmer and friendlier. It’s obvious that Microsoft has taken a hint from Mac UXUI in style and simplicity.

And in our opinion, it works. As one reviewer summarized, the design is “a breath of fresh air for Windows that attempts to throw out much of the old UX in favor of a more modern, fluid, and simplistic interface. I think it does a good job at achieving this goal.”

That said, is it worth an upgrade?

Our ability to enable productivity is central to our responsibilities as IT professionals. Microsoft has obviously taken this responsibility seriously with the new Windows 11 design, attempting (in many cases successfully) to simplify and streamline operation for the end user while going about their daily tasks.

For certain companies, these changes might warrant an upgrade. For companies who have significant stock in their end users’ ability to complete complicated tasks and navigate their OS in the least amount of time – industries like healthcare – the improved design and ease of use might be enough to warrant the significant investment of time and budget to upgrade to Windows 11.

That said, maybe not. Windows 10 isn’t unbearable. Most end users can navigate and be productive using Windows 10, making the improvements in user experience a welcome change to the Windows OS, but not necessary enough to warrant upgrading. For most companies, they can probably afford to wait awhile.

System Requirements

The required hardware is more substantial for Windows 11 than Windows 10, which is another factor to consider.

The required RAM for a Windows 11 build is 2x – 4x that of Windows 10, and Windows 11 requires more than 2 cores on a 64-bit processor. These requirements are significant considering many machines had a hard time meeting even the requirements for Windows 10 when it came out.

For many lower-powered machines, these requirements will be too great, necessitating companies not only update their OS but also the hardware to support it. This additional budgetary hurtle will, for many companies, make this upgrade currently out of reach.

So, Should I Upgrade?

Windows 11 is, to me, Windows 10 in a nicer skin. While it does come out-of-the-box more secure because of quite a bit of feature standardization, there’s few truly new features unavailable to a savvy IT team that can’t be enabled in Windows 10. Plus, the new build is resource-heavy enough to significantly increase the required specs, disqualifying many endpoints unless you upgrade hardware concurrently.

That said, considering the effort and cost often required by an upgrade, we think that most companies might want to wait awhile before pulling out their corporate credit cards.

Also, most companies aren’t at the maturity of endpoint management where upgrading your OS will improve your security and productivity to a large degree. Why? Because new software isn’t the best way to secure or improve your environment. Instead, it’s adept management and use of the software that you have.

If you’re concerned with things like security and user experience, in our opinion your budget would be better spent on benchmarking your IT maturity and making process improvements rather than being an early adopter of a new OS. That’s where you’re going to see significant improvements to your security, efficiency, productivity, and the time and capital investment of your team.

A shiny new pair of shoes won’t prevent an injury if your running stride is poor. If you’re having knee pain, instead of heading down your local Foot Locker and dropping 90 bucks on “corrective” shoes, you should ask someone experienced to help you to improve your stride. That’s where the biggest improvements will come, and it probably will cost less too.

All that said, there might be a few cases where your company might consider upgrading to Windows 11 now. For example:

1. If you’re in an industry where security and end user productivity failures are catastrophic. Industries like financial services or government could make a compelling case for the standardized out-of-the-box security features of Windows 11 being worth the cost. Also, healthcare companies who need their OS to work 100% of the time and as quick as possible could also see benefit from the new UX in the form of improved end-user efficiency.
2. If your company has a high infrastructure maturity rating because of proficient endpoint management. If you’ve already optimized your processes, the improvements you could see from Windows 11 might be the cherry on top, improving your security and end user experience and keeping your already optimized infrastructure top of the line.
3. If you fit into the other two points and can afford the hardware required for this upgrade. Windows 11 hardware requirements will exceed many endpoints’ ability to run it, and these endpoints will need to be updated if they’re going to run the new OS. However, if your company has the budget for hardware upgrades, and security and end user experience are a priority, Windows 11 might be within your reach.

The Model Maturity Assessment

If you’re reading this, it’s highly likely that you’re thinking about improving endpoint security and efficiency. If so, we think one of the best investments you can make at this point is to benchmark your IT health and improve from there.

Here at Model Technology Solutions, we’ve assembled some of the best endpoint management engineers to help to assess and improve infrastructures like yours. We love the process of getting to know IT teams all over the world and partnering with them to improve their infrastructure health. It’s what we do.

Through our Maturity Assessment, our top-tier engineers will measure the health of your infrastructure and identify the most cost-efficient projects you could undertake to improve your security, efficiency, and reduce your costs. Upgrading to Windows 11 might be a part of that, or maybe not. Whatever those projects are, you can be sure they’ll be tailored to your particular industry and designed to give you the most significant returns for the least amount of resources.

If you’re interested in learning more about this maturity health check, click here or use the call to action below this post. You can learn about our proprietary maturity model that we use to frame the assessment and download a questionnaire that will show you exactly what kind of information that we’ll gather as a part of it.

Until next time!

DDR5 is simply the 5th generation of DDR Memory and the natural evolution following DDR4. Basically, it’s the newer, faster RAM. Bandwidth (speed) and density (RAM capacity per DIMM) have increased over previous gen, while power consumption has decreased.

Performance

With increased RAM bandwidth comes higher default RAM speeds. Though higher speeds like 4133MHz were available for DDR4, 3200MHz was the highest default speed available – anything higher was factory overclocked. DDR5’s higher bandwidth means 4800MHz or even 5200MHz speeds without overclocking. That’s a significant jump, but how does that translate to real world performance? That answer depends on exactly what you use your PC for.

With gaming, graphics cards are normally the performance bottleneck, so modern processors and RAM upgrades typically only offer incremental frame rate gains. Check out this data from Tom’s.

As you can see, going from DDR4-2133 to DDR5-6000 produces only a marginal FPS gain of about 8%. 3200 to 6000 was closer to 2%. It’s not nothing, but certainly not worth investing in a whole new platform.

The story does change somewhat with benchmarks that more heavily rely on RAM performance. Again, from Tom’s:

Here we do see gains in the 10% range going from DDR4-3200 to DDR5-6000. Nothing spectacular, but still significant.

What hardware supports DDR5?

As of this post, only 12th Gen Intel Core processors support DDR5, and only with specific Z690 chipset motherboards designed for it (yes there are also DDR4 only Z690 boards just to add to the confusion). Upcoming Intel and AMD processors set to be released later this year will all support the new memory too though, so we will eventually start to see full integration of DDR5 just as we did with prior generations of memory.