M2 MacBook Pro Can Hit 108C Under Full Load

Apple announced its upgraded M2 SoC at WWDC recently, and the first laptop to use it is the 13″ MacBook Pro. The laptop became available to testers a few days ago, allowing content creators to put it through its paces. This week YouTuber MaxTech compared the new 13″ M2 Pro to the older M1-based Pro with the M1 Pro SoC. In the tests, the newer M2-based Pro’s cooling performance, or lack thereof, is shockingly bad. He found when exporting 8K Canon RAW footage the M2 model hit 108C and experienced severe thermal throttling. This stands in stark contrast to the 14″ M1 MacBook Pro, which never throttled in the same test.

The video released by MaxTech features side-by-side comparisons of both notebooks; the 13″ MacBook Pro with M2 and the 14″ model with the M1 Pro. One of the major differences in the two laptops is the cooling system. The M2 model has only a single fan inside its 13″ chassis. The 14″ model with the Pro chip has two fans. It’s possible Apple determined a second fan was necessary due to the “Pro” designation of the SoC. However, as MaxTech discovered, that second fan is also needed in the M2 model. As noted above, he saw it hit outrageous temps under sustained load. He wrote on Twitter the temps were the highest he’d ever seen on a Mac. Adding insult to injury, he said that includes Intel-based Macs.

The high temps were achieved exporting 8K Canon RAW footage, which is as stressful as you can get for an SoC. That’s because it stressed both the CPU and GPU to the limit throughout the export. What’s surprising is the MacBook hit those temps while the single fan was spinning at its 7,200rpm peak. This means that since the fan was doing all it could do to keep temps in-check, there was only one way for the system to lower temps more: thermal throttle itself. This is when clock speeds are lowered to reduce temperatures. This reduces performance, but prevents a chip from overheating and damaging itself.

The way it did this was to instantly drop the P-core clocks from 3,200MHz to 1,894MHz. The E-cores went from 2,228MHz to 1,444Mhz. The GPU also dropped from 1,393 MHz to just 289Mhz. This resulted in a drastic reduction in power consumption. The entire package went from sucking down 29.46W to just 7.31W. This allowed the M2 chip to drop all the way down to 84C for a bit, before making the climb back up to over 100C. The YouTuber said this cycle repeated itself over and over until the export finished. This back-and-forth throttling lead to reduced performance, and it was crushed in the test by the M1 Pro laptop. The M2 took almost 20 minutes to finish the export, with the M1 Pro taking a smidge over 11 minutes. The dual-fan M1 Pro model also never throttled in the test either.

Since the M1 Pro is not directly comparable to the M2, he also tested the M1-based 13″ MacBook Pro. That’s the model the new M2-based machine is replacing, so it’s apples-to-apples. In the same export test, he says the M1 model maxed out at 94C. However, he says the CPU wasn’t running at 100 percent the entire time, but the GPU was. The M1 model also never throttled. This seems to indicate something changed with the M2 model, either in the fan curve, thermal paste used, or something else.

For now, we will have to wait and see what other reviewers have to say about this matter. Though MaxTech’s tests seem straightforward, it’s always informative to get confirmation from multiple sources before declaring something as “buggered.” However, this comes in the wake of multiple reports confirming Apple severely limited SSD performance in the 13″ M2 model. Apple went with a single 256GB NAND chip in the base model. The M1 version used two chips to hit the same capacity. The result is a drastically slower SSD, at least compared to the M1 model. Higher capacity SSDs on the M2 do not suffer this fate, so it is seemingly due to the single chip versus using two.

The M2 model of the MacBook Pro has just a single SSD, with an empty pad next to it that was populated on the M1 model. (Image: MaxTech on YouTube)

Overall, it makes the 13″ M2 MacBook Pro seem like a flawed laptop. We’re curious to see what other reviewers have to say about its thermals in the coming days. This revelation has people expressing doubts about the upcoming M2-based MacBook Air, which is fan-less. However, that model is not designed for exporting 8K footage, so it will probably fare better in the cooling department. Still, it’s not a good look for Apple’s newest SoC, especially since its M1 line was so successful.

Now Read:

• Apple’s New M2 Takes Serious Shots at Intel, x86
• Apple Unveils Redesigned MacBook Air With M2 SoC
• It’s Not ‘Too Late’ for Intel to Beat the Apple M1

New Utility Fixes Windows Defender Hogging CPU Time on Intel CPUs

Kevin Glynn, aka Uncle Webb at TechSpot, has developed several useful freeware utilities like ThrottleStop and RealTemp over the years. In the course of developing those programs he discovered a curious behavior in Windows Defender with Intel CPUs. Windows Defender is the software included with Windows to protect your PC from malware and viruses. Webb discovered that at random intervals Defender would suddenly begin using excessive CPU resources. In some cases it can result in up to six percent lower performance. Thankfully Webb has created a free utility to resolve the issue, and it’s called Counter Control. Note this behavior has been reported so far with Intel 8th, 9th, 10th, and 11th gen CPUs on Windows 10 and 11. AMD CPUs are not affected at all.

Here’s a simple explanation of the situation. Intel CPUs include three fixed function hardware performance counters for each thread. They are designed to be a shared resource, so temperature and performance tools can access them. They can be used either by the OS or the user. These three counters can be programmed to one of four modes reflecting different levels of privilege. Those include Disabled, OS access (Ring-0), User (ring>0), and all-ring levels. Most performance monitoring tools set this to “all-ring levels” or Mode 3. This allows any program to access them with no issues. However, Windows Defender’s Real-time Protection notification feature will try to change all three to Mode 2 at random intervals. This is the crux of the issue, as Defender will use CPU time trying to change the status of the counters. If you’re curious, you can load up HWINFO and put the CPU under full load. If might report a slightly lower maximum clock speed. The software’s author says this is likely Defender trying to use the counters, and interfering with HWINFO.

What I saw after a fresh boot. The 0x222 means Defender is using the counters.

As far as it affecting performance, it can have a noticeable impact, at least in benchmarks. One example according to TechPowerUp uses a Core i9-10850K running at 5GHz. It showed a decreased Cinebench R23 benchmark score of ~1000 points (16800 vs 15800). Your humble author did the same test on his own Intel 11th gen CPU. I ran Cinebench R23 and with my PC “as is” and got a score of 11,158. Next, I downloaded the utility and clicked “reset counters” and ran it again. My score with the counters reset was 12,163; which is 8.6 percent uplift. That said, I’ve had this system for roughly a year now and it’s never felt slow or unresponsive. It features an Intel Core i7-11700KF, 32GB of DDR4, and a PCIe 3.0 NVME SSD.

When you fire up the utility, which can be downloaded here, you’ll see the status of the “IA32_FIXED_CTR_CTRL” register on Intel CPUs. Here’s how to interpret the number you see, copied from TechPowerUp:

• Not Used – 0x000: The three fixed function counters are stopped. None of the counters are presently being used.
• Defender – 0x222: All three fixed function counters are programmed to mode 2. This is the value that Windows Defender sets these counters to when it is using them.
• Normal – 0x330: Two counters are programmed to mode 3. One counter is programmed to mode 0 and is not being used. This is normal. Most monitoring programs that use these counters will program the counter control register to this value.
• Warning – 0x332: This is shown when two counters are being used normally by monitoring software while the third counter has been set to mode 2, likely by Windows Defender. This is a warning that two different programs might be fighting over control of the shared counters. You might see the counter control register constantly changing between 0x222 and 0x332. This is what you will see when running HWiNFO if Windows Defender is trying to use the IA32_FIXED function counters at the same time.

If you use the utility and click “reset counters,” it will resolve the issue. Defender will not try to change it back for the duration of that session. If you reboot, you will need to check it again. As far actual fixes go, you can always disable Windows Defender’s real-time notification system, but that’s not recommended. However, if you want to do it anyway, here’s how you do it. On Windows Pro OSes, go to the Local Group Policy Editor (gpedit.exe). Next navigate to “Computer Configuration / Administrative Templates / Windows Components / Microsoft Defender Antivirus / Real-time Protection.” Here you can enable “Turn off real-time protection.”

If you’re on Windows Home, you will need to edit the registry. Navigate to Computer\HKEY_LOCAL_MACHINE\SOFTWARE\Policies\Microsoft\Windows Defender\Real-Time Protection. Next, if you do not see a value called DisableRealtimeMonitoring, right click and create a new DWORD value. Name this DWORD value DisableRealtimeMonitoring and set this to a value of 1.

If you want to leave Defender alone, you can also run ThrottleStop.  It has a feature called “Windows Defender Boost.” Enabling it it activates one of the programmable timers. Windows Defender will notice this and cease trying to access them until the system is restarted.

The software’s author is curious to see if more people are experiencing this issue. Hopefully, he writes, if enough people complain about it, Microsoft will fix Defender, permanently.

Now Read:

• Microsoft Testing ‘Unsupported Hardware’ Watermark on Windows 11 Desktop
• Microsoft Kicks Unsupported PCs From Windows 11 Testing Program
• Microsoft Windows 11: It’s Windows, But Elevener

Report: Samsung Plans Mid-Range Foldable Phones by 2024

Just a few years ago, foldables were a rarity in the smartphone ecosystem, but they’ve quickly become an annual component of Samsung’s release cycle. They’re still fantastically expensive devices, though. As we wait on the Galaxy Z Fold4 and Flip4, a new report points to a more reasonably priced future for Samsung’s foldables. The company is allegedly working on mid-range foldables that will be part of its A-series lineup, which could push foldable sales to new heights. 

If you wanted to pick up a foldable right now, you’d have to spend at least a thousand dollars, ignoring any short-term sales. Samsung’s first foldable retailed for a whopping $2,000, and the updated Z Fold3 isn’t much cheaper at $1,800. The smaller Z Flip3 costs $1,000, a sizeable reduction from the $1,500 asking price for the first Z Flip. Samsung has managed to bring these prices down modestly even as component prices have increased, but moving foldables into the mid-range is going to take some time. 

According to ET News, Samsung’s mobile experience division has focused its efforts on making mid-range foldables happen by 2024. The pricing is vague right now, but the report says they are targeting under 1 million won, which works out to about $770 at current exchange rates. That’s not a huge change for the flip phone-style foldables, but the report could be speculating about the pricing for a Fold-based device that becomes a small tablet when opened. 

These cheaper foldable devices will lose some premium features to hit that price, but the report doesn’t specify. Odds are, we’ll get cheaper camera modules and simpler designs. For example, current foldables have incredibly intricate hinge designs that clear dust from behind the screen. I would also expect the external cover displays to get smaller or lower resolution. Extras like S Pen stylus support and water resistance could also fall by the wayside. 

Analysts expect foldable shipments to increase from about 10 million in 2021 to 15 million. That’s healthy growth, but the overall market share of foldables is still sitting in the single digits. Samsung believes that adding foldables to its budget-oriented A-series could boost that to double digits. It could also help Samsung solidify its place as the premier manufacturer of foldable phones, a form factor that will most likely become common in the next decade. Even Apple is rumored to have foldable plans, but this time, Samsung has a big lead.

Now read:

• Samsung Caught Cheating Customers Again, But On TVs This Time
• Samsung Promises App Throttling Controls in Future Galaxy S22 Update
• Samsung Partners with iFixit to Offer DIY Smartphone Repair Kits

Construction Begins on Largest Direct-Air Capture Plant

(Photo: Climeworks) A Swiss startup has broken ground on the world’s largest direct-air carbon dioxide capture plant. 

Climeworks, a technology-based climate solutions company, specializes in building facilities that capture carbon dioxide out of the air and then recycle or dispose of it. Mammoth, its gargantuan new facility, will capture about 36,000 tons of carbon dioxide each year once fully operational. That’s nine times as much carbon dioxide as its previous plant Orca, which took the “world’s largest” title when it opened almost a year ago. 

Each of Climeworks’ plants consist of stacked modular carbon-capture machines. These machines begin by using a fan to draw air inward. Inside, a “selective” filter material captures carbon dioxide until it’s full, after which the machine is closed off. Climeworks operators turn up the interior temperature to between 80 and 100 degrees Celsius (176 – 212 degrees Fahrenheit), which releases the concentrated carbon dioxide for collection. 

The collected carbon dioxide then has two paths: upcycling or permanent storage. Climeworks says most of its upcycled carbon dioxide enters the circular economy, where it’s used for aviation fuel or carbonated beverages. While this helps to create a closed carbon cycle by slowing the increase of new carbon dioxide in the atmosphere, it doesn’t help to decrease the total amount of carbon dioxide—something Earth desperately needs. This means some captured carbon dioxide needs to be geologically stored, most often using porous and permeable reservoir rock.

A 3D rendering of Mammoth, which should be completed within 18-24 months. (Image: Climeworks)

Of course, all direct-air capture plants require energy to run in the first place. Mammoth and Orca are situated within the ON Power Geothermal Park in Hellisheiði, Iceland, so both are able to power functions using geothermal energy and waste energy. Climeworks says the rest of its facilities also use renewable energy, but hasn’t specified exactly what kind. 

Mammoth will join the ranks of about 20 other direct-air capture plants around the world. (Fifteen of these are Climeworks’.) As the startup scales, it hopes it will reach a collective multi-megaton capacity by 2030 and gigaton capacity by 2050. The massive plant is expected to be completed in 18-24 months. 

Direct-air capture plants make up just one climate solution out of several necessary ones. Scientists around the world estimate that 10 gigatons of carbon dioxide would need to be removed from the atmosphere each year in order to achieve US and global emissions reduction targets by 2050. Reaching such a high capacity would require a lot of Mammoth facilities. But, as they say, Rome wasn’t built in a day—and neither will be tens of thousands of direct-air capture facilities. 

Now Read:

• Scientists Achieve Photosynthesis Without Sunlight
• Hawaii Startup Launches World’s First Ocean-Assisted Carbon Removal Plant
• Corn-Based Ethanol Is Actually Worse For The Environment Than Gasoline, Study Finds

Astronauts Suffer Significant, Permanent Bone Density Loss in Space

For all the time humanity has existed, every single person has spent their lives under normal Earth gravity — 9.806 meters per second squared. And then, a few decades ago human beings started spending days, then weeks, and then months in freefall. We’re only now beginning to understand what living without gravity does to the body, but it’s not good. Kinesiologists Leigh Gabel and Steven Boyd from the University of Calgary have published a new study that examines the bones of astronauts, finding that extended time in space causes degradation in bones that never fully reverses. If our future is in space, this could be a major problem. 

The pair does not mince words in the study, which was published in the journal Scientific Reports. They call the effects of weightlessness on astronaut skeletons “profound.” Even after a year back on Earth, bone density does not fully recover. Essentially, being in space for even a few months causes your bones to age at an increased rate — as much as 10 years per flight. 

This study focused on people who had been on long-duration space missions, which means three months or longer, in this case. In partnership with NASA’s Johnson Space Center in Houston, Gabel and Boyd examined 17 astronauts (14 men and three women) both before and after a 6-12 month excursion in space. 

Only one of the astronauts in the study saw bone density return to normal after an extended time back on Earth. For the others, most of the density loss was concentrated in weight-bearing bones of the lower extremities, but the arms did return to normal after about a year. The team analyzed the mineral content and density of bones, allowing them to determine how the “failure load” had changed. On average, the failure load of the astronauts’ tibias before spaceflight was 10,579 newtons, but that dropped to 495 newtons after the mission. They partially recovered as time went on, but after a year were still down an average of 152 newtons. 

Russian cosmonaut Anatoly Ivanishin exercises on the ISS in 2012.

The researchers also note that those who were in space longer suffered greater drops in bone strength. Astronauts who were on the International Space Station for longer than six months recovered less of their bone density after returning. That suggests that long-term missions, like the ones planned for the Artemis Program, could have serious impacts on astronaut health down the road. It also raises more questions about how humans would fare long-term in an environment where there is less gravity than Earth, for example on the Moon or Mars. 

It may be possible to mitigate the effects of weightlessness to some degree. Astronauts undertake a rigorous routine of exercise under simulated gravity during missions to prevent muscle wasting. The researchers used data on exercise routines in the study, and they say added focus on the legs (deadlifts, for example) could reduce density loss. However, we may be years from fully understanding how the body responds to microgravity.

Now Read:

• NASA Awards Next-Gen Spacesuit Contracts
• Spending Time in Space May Change Astronauts’ Brains Forever
• NASA Rolls SLS Back for Repairs, Likely Delaying Artemis 1 Launch

Valve Cuts Steam Deck Performance, Doesn’t Disclose It

(Photo: Petar Vukobrat/Unsplash) Valve has quietly changed the Steam Deck’s specs, unbeknownst to most who have put in an order in recent weeks. 

Steam Decks ordered at the time of the console’s release came with either a 256 or 512GB SSD, which were connected using an x4 PCIe connection. SSDs in more recently ordered consoles, however, are connected using just two lanes, according to the German tech blog HardwareLuxx. PC Gamer spotted the blog post Wednesday.

Cutting the drive’s bandwidth in this way could hypothetically result in slower performance. Worse, Valve didn’t disclose the change in a company statement, or in an email to customers, leaving new Steam Deck owners to find out on their own whether they received a console with altered specs. (PC Gamer’s report includes instructions for those who want to risk a peek.) The only “notice” Valve put out around the time of the change was an edit to the console’s webpage. According to the Wayback Machine, the edit was made at the end of May. 

The Steam Deck’s specs page contains quiet edits to its SSD info.

So far, tests performed by Kotaku have failed to find any real differences in performance, such as with load times or FPS. This must be a relief for Valve, which scrambled to explain the switch as soon as it became publicly known. The console’s design team followed up with PC Gamer by saying users won’t notice a difference except in “extremely uncommon cases.” According to them, the change was essential if Valve wanted to grab hold of the sticky supply chain, which (as we’re all exhaustingly familiar with by now) has been wrecking electronics manufacturing for a couple years. The new SSD option enables Valve to double its weekly Steam Deck shipments—a welcome development for those who ordered theirs a bit late in the game. 

“Our team has tested both components extensively, and determined that there is no impact to performance between the two models,” Steam Deck designer Lawrence Yang told PC Gamer.

Still, it isn’t unreasonable for customers to wish Valve had clearly conveyed the change earlier on. Switching up the console’s hardware doesn’t make up nearly as much of the problem as the lack of communication does. That being said, plenty of gaming—er, testing—will be required to determine whether Valve’s dismissal of potential performance issues is well-warranted. 

Now Read:

• Valve Designer Warns Against Dangerous Steam Deck Mod
• iFixit Begins Selling Replacement Parts For Almost The Entire Steam Deck
• Judge Denies Valve’s Request to End Antitrust Case

Samsung Becomes First Foundry to Begin Production at 3nm

(Photo: Samsung) Last week, rumors surfaced that Samsung would soon claim it had beaten TSMC to the 3nm punch. Yesterday, Samsung confirmed the news with a celebratory press release. It is the first global foundry to begin 3nm silicon production. It’s also the first foundry to move beyond FinFET to gate-all-around (GAA) transistors, which is a major achievement. We should note that nowhere in Samsung’s press release does it say it’s begun high volume manufacturing (HVM). Therefore, it’s unclear how many wafer’s it’s capable of producing at this time.

The news marks the culmination of a multi-year effort by Samsung to hit this crucial milestone without serious delays. Samsung first announced its GAA design in 2019, calling it MBCFET. That stands for Multi-Bridge Channel Field Effect Transistor, which is a nanosheet design. This is different than Intel’s RibbonFET process, which uses thin nanowires. TSMC will also be moving to nanosheets, but not until  2nm. For 3nm it’s sticking with FinFET via a customizable design named FinFlex. TSMC is expected to begin 3nm production sometime in late 2022.

(Image: Samsung)

The announcement from Samsung offered some numbers to quantify the benefits of the move to GAA. Compared to 5nm, its 3nm process allows for a 45 percent reduction in power consumption, 23 percent more performance, and a 16 percent reduction in area. The company’s second generation 3nm process will take things even further. It’s promising a 50 percent reduction in power, 30 percent more performance, and a reduction in area of 35 percent. Samsung has said previously its second generation design will arrive approximately one year after its predecessor. Its first generation 3nm nanosheet design is focused on “high performance, low power computing” the company stated. It will then move to focusing on chips for mobile applications.

Samsung’s nanosheet design will allow it to precisely fine-tune the characteristics of the transistors. By tweaking the width of the sheets, it can customize the power and performance curve with more precision than what FinFET allows. For more power/performance it can use wider sheets, or narrower sheets to improve efficiency.

So far Samsung hasn’t announced any official customers or products for its 3nm process. It’s also not clear when it will begin high volume manufacturing, and what kind of yields it’s achieved. Still, it’s a shot across the bow of its main rival TSMC. As noted by Anandtech, Samsung has been losing customers to TSMC lately. In May Qualcomm announced it was moving to TSMC for its Snapdragon 8+ Gen 1 SoC. Obviously, Samsung would like to reverse that trend, and reaching 3nm is a feather in its cap along those efforts, if it can get yields up. When TSMC begins 3nm production later this year, it’ll be interesting to see how its flexible FinFET design stacks up against Samsung’s GAAFET. Intel isn’t expected to leave the FinFET era until 2024 with its 20A “Angstrom” process. Like TSMC’s, its Intel 3 node will still use FinFET.

Now Read:

• TSMC Founder Says US Silicon Fab Expansion Plans ‘an Exercise in Futility’
• TSMC Says it Will Move to Nanosheet Transistors at 2nm
• Intel 4 is a Major Step on Intel’s Path Back to Semiconductor Dominance