• AMD Confirms Radeon Driver Bug That May Accelerate SSD Wear, With Fix E

    From MummyChunk@3:633/10 to All on Monday, November 09, 2026 14:15:49
    Subject: AMD Confirms Radeon Driver Bug That May Accelerate SSD Wear, With Fix Expected in Adrenalin 26.11.1

    AMD Confirms Radeon Driver Bug That May Accelerate SSD Wear, With Fix Expected in Adrenalin 26.11.1

    AMD has acknowledged a software defect affecting some Radeon graphics systems that can generate continuous background disk activity, potentially increasing unnecessary writes to solid-state drives even when a computer is sitting idle. The problem has been traced to a component of AMD's Adrenalin graphics software, with users reporting sustained activity ranging from approximately 0.3MB/s to as much as 5MB/s. If those figures represent actual storage writes maintained over extended periods, the additional workload could consume a substantial portion of an SSD's rated write endurance.

    The issue involves AMD External Events Utility, a Windows service responsible for various display-related functions, including monitor detection and graphics software shortcuts. Its associated process, atieclxx.exe, has been observed generating unusually high amounts of input/output activity on affected systems. Some users have reported that their SSD health indicators declined faster than expected, including one account describing a reduction of approximately one percentage point per month over a 16-month period.

    AMD representatives have confirmed that the company is investigating the problem and have identified an apparent interaction between the External Events Utility and Microsoft's PCIe filter driver. A correction has reportedly already been submitted by AMD's driver development team, with the company now targeting Adrenalin Edition 26.11.1 for its release. The update is expected later in November, although no exact availability date has been announced.

    There is, however, an important distinction between the existence of the software bug and the extent of the damage it may cause. While abnormal background activity has been documented and AMD has acknowledged the underlying issue, the reported reductions in SSD lifespan have not been independently established as a direct consequence of the driver. Furthermore, some Windows performance counters can report I/O operations that do not translate directly into physical writes to NAND flash memory.

    The available evidence therefore establishes a legitimate driver problem that warrants investigation, but it does not demonstrate that every affected computer is experiencing the maximum theoretical SSD wear described in some reports.

    The Problem Originates in AMD External Events Utility

    AMD External Events Utility is installed alongside Radeon graphics software and normally operates quietly in the background. Its responsibilities include responding to changes in connected monitors, handling certain display events and supporting functionality associated with AMD's graphics control software.

    The service uses processes including atieclxx.exe, known as the AMD External Events Client Module, and atiesrxx.exe, the corresponding service component. Under normal circumstances, these processes should consume relatively few system resources while performing their designated tasks.

    Reports from affected users suggest that a software interaction can instead cause the client process to repeatedly perform operations that generate persistent I/O activity. This behavior may continue even when no applications are running that would ordinarily require significant disk access.

    Because the amount of activity can be relatively modest at any individual moment, the problem may go unnoticed during everyday use. A background process generating several hundred kilobytes per second is unlikely to attract attention on a modern NVMe SSD capable of transferring several gigabytes per second. The computer may remain responsive, games may perform normally, and there may be no obvious warning that something unusual is happening.

    The concern arises from the persistence of the activity rather than its immediate impact on performance.

    AMD representative AMD_Vik discussed the issue in a Reddit troubleshooting thread, explaining that the behavior appeared to involve the External Events Utility interacting with Microsoft's PCIe filter driver. This component operates within the broader Windows hardware and driver infrastructure, where software communicates with devices connected through the PCI Express bus.

    That explanation suggests the problem is more complicated than an application simply creating an oversized diagnostic log. It appears to involve repeated operations within the graphics driver and operating system environment, although AMD has not published a complete technical root-cause analysis.

    AMD_Vik also confirmed that the company had reproduced the issue internally and that a colleague had submitted a proposed correction. The representative described the change as relatively low risk but explained that it had to be coordinated with AMD's established driver release schedule.

    The defect is therefore not merely an unverified complaint circulating through online forums. AMD has acknowledged the underlying behavior, although important questions remain about the precise mechanism and its consequences for physical storage devices.

    How Much Data Could the Bug Write to an SSD?

    The most alarming claims concern the cumulative amount of data that could be written if the reported activity continues uninterrupted.

    Users have reported different rates, with several observations falling between 0.3MB/s and 1.5MB/s. One particularly concerning account described activity reaching approximately 5MB/s while the computer was otherwise idle.

    At first glance, even 5MB/s appears insignificant compared with the performance of a modern SSD. However, sustained activity can accumulate into much larger totals over days and weeks.

    Assuming decimal megabytes and continuous activity for 24 hours, the theoretical figures are:

    0.3MB/s: Approximately 26GB per day, or 0.78TB over 30 days.

    0.7MB/s: Approximately 60GB per day, or 1.81TB over 30 days.

    1.5MB/s: Approximately 130GB per day, or 3.89TB over 30 days.

    5MB/s: Approximately 432GB per day, or 12.96TB over 30 days.

    These calculations illustrate why the issue has attracted attention. At the highest reported rate, a computer left running continuously could generate nearly 13TB of reported activity in a single month.

    However, those totals are mathematical projections, not measurements from a controlled month-long experiment. They assume the reported rate remains constant every second, that the computer operates continuously and, most importantly, that the activity corresponds to data actually written to storage.

    The final assumption has not been established across all affected systems.

    Some users have reported significantly lower activity, while others have found no abnormal behavior whatsoever. One Radeon owner described a process accumulating approximately 19.5GB of I/O writes during 12 hours of operation, substantially below the maximum theoretical figure.

    Another user reported different results on two Radeon systems: one computer showed almost no activity from the process, while another continuously accumulated approximately 30MB per minute.

    These differences indicate that the problem is not universal and may depend on a combination of driver versions, graphics hardware, operating system configuration and other installed components.

    AMD has not published a comprehensive list of affected configurations or identified the conditions necessary to reproduce the problem reliably on every system.

    High Windows I/O Counters Do Not Necessarily Mean Equivalent NAND Wear

    One of the most important technical qualifications concerns how Windows measures application activity.

    Task Manager can display process-level I/O counters, including the number of bytes associated with write operations. However, these counters do not necessarily provide a direct measurement of data physically programmed into an SSD's NAND flash memory.

    Windows processes interact with numerous resources through its I/O subsystem, including files, devices and other operating system components. Certain operations can be counted at the process level without producing an equivalent amount of physical storage activity.

    In addition, Windows uses caching and other mechanisms that can change when and how data is committed to storage. An application's reported write activity therefore should not automatically be treated as identical to the SSD's total host writes, much less the amount of data programmed internally into NAND.

    This distinction has emerged in discussions among Radeon owners attempting to investigate the problem.

    One user examined the affected process using Microsoft's Process Monitor utility and observed repeated access to system information, device enumeration and file metadata operations. Although the process accumulated I/O activity, that particular investigation did not identify corresponding physical file writes.

    Another user examining a different configuration reported sustained disk writes visible in Windows Resource Monitor and found that stopping AMD External Events Utility eliminated the activity.

    These observations are not necessarily contradictory. They may reflect different manifestations of the same bug, different hardware and software configurations, or limitations in the monitoring methods being used.

    The important point is that an increasing I/O counter alone does not prove that the SSD is experiencing the same amount of NAND wear.

    A more reliable investigation would combine process-level monitoring with the drive's own health information, including its reported total host writes. Comparing those measurements over a controlled period can help determine whether the abnormal activity is actually contributing to increased storage writes.

    Even then, attributing every additional write to one process requires care because Windows, antivirus software, background updates and other applications also generate storage activity.

    Consequently, claims that the bug has already destroyed a particular percentage of an SSD's lifespan should be treated cautiously unless supported by detailed measurements that establish a causal connection.

    Why Unnecessary Writes Could Shorten SSD Lifespan

    Unlike conventional hard drives, which store information magnetically on rotating platters, SSDs use NAND flash memory. Flash cells have a finite endurance because programming and erasing data gradually changes their electrical characteristics.

    Modern SSD controllers manage this limitation through wear leveling, overprovisioning, error correction and other techniques intended to distribute activity and maintain reliable operation over time.

    Manufacturers commonly specify drive endurance using a rating called terabytes written, or TBW. This describes the amount of host data a drive is designed to accommodate under defined operating conditions, often as part of its warranty specification.

    For example, an SSD rated for 300TBW is designed around a substantially different total write workload from a drive rated for 600TBW. Capacity, NAND technology, controller design and the manufacturer's endurance targets all influence those figures.

    If a genuine software defect adds several terabytes of unnecessary writes every month, it could consume an appreciable portion of the rated endurance of a lower-capacity consumer SSD.

    The potential impact is especially relevant to inexpensive drives using QLC NAND, which stores four bits of information in each memory cell. QLC technology offers high storage density and can reduce manufacturing costs, but it generally has lower raw program-and-erase endurance than comparable TLC NAND, which stores three bits per cell.

    That does not mean every QLC drive is inherently fragile. Controller quality, spare capacity, error correction and firmware behavior can make a substantial difference, while many modern consumer SSDs are designed to withstand years of normal use.

    Nor should the TBW rating be interpreted as a countdown to inevitable failure. Reaching the manufacturer's rated endurance does not mean a drive will immediately stop working, and SSDs can sometimes operate well beyond their advertised write limits.

    Nevertheless, unnecessary writes are undesirable because they consume storage resources without providing a useful benefit.

    Another complication is write amplification. An SSD may need to perform internal NAND operations beyond the amount of data originally submitted by the host, depending on its internal organization and workload.

    This means that even an accurate measurement of host writes does not directly reveal the complete internal wear experienced by the flash memory.

    The real-world consequence of AMD's bug therefore depends on whether the reported activity reaches the drive, how much data is actually written and how the SSD handles that workload.

    At present, those factors have not been measured consistently enough to establish a general estimate of reduced lifespan.

    Which Radeon Drivers and Windows Systems Are Affected?

    Reports have emerged from users running several versions of AMD Software: Adrenalin Edition, including 26.10.1 and 26.10.2.

    The latter was released on October 30 and addressed several unrelated stability problems, including game crashes, driver timeouts and issues involving graphics overlays. However, its official release notes do not list a correction for the External Events Utility problem.

    Some users have traced similar behavior to earlier driver releases, and one report suggested that the issue may have been present since Adrenalin 26.7.1. Other accounts describe problems occurring after Windows updates, raising the possibility that changes to the operating system or PCIe driver environment contribute to triggering the defect.

    AMD has not confirmed a single version as the starting point, and the evidence does not establish that every release after a particular date is affected.

    Reports cover both Windows 10 and Windows 11, along with several generations of Radeon hardware.

    The issue has also been observed on systems with different processor and chipset configurations, making it inappropriate to assume that it is limited exclusively to computers using AMD Ryzen CPUs.

    This is relevant because AMD's graphics software supports Radeon cards installed in systems built around either AMD or Intel processors.

    There is also uncertainty about whether integrated Radeon graphics solutions are affected in the same way as dedicated graphics cards. Some Ryzen processors use graphics components supported by related Adrenalin packages, but the current reports do not provide sufficient evidence to establish the prevalence of the issue across those systems.

    Importantly, independent testing has not reproduced the problem universally.

    PC Gamer reported that one of its staff members using AMD graphics hardware was unable to reproduce the excessive activity. Reddit users have similarly described systems running recent driver releases without unusual disk behavior.

    These findings suggest that Radeon owners should check whether their own computers exhibit the problem rather than assuming that every installation of AMD's graphics software is damaging an SSD.

    How to Check Whether Your Computer Is Experiencing the Problem

    Windows includes diagnostic tools that can help identify abnormal background activity without requiring additional software.

    The most useful starting point is Resource Monitor, which provides more detailed information about disk activity than the standard Task Manager overview.

    To open it, press the Windows key and R simultaneously, enter resmon and press Enter. Select the Disk tab and examine the processes listed under Disk Activity or Processes with Disk Activity.

    Look for atieclxx.exe, identified as AMD External Events Client Module, and check whether it is generating persistent write activity when the computer is otherwise idle.

    It may also be useful to examine atiesrxx.exe, the related AMD service process.

    The relevant behavior is sustained activity rather than an occasional spike. Background services normally access files and system resources from time to time, and brief activity immediately after starting Windows does not establish that the system is affected.

    For a more complete investigation, Task Manager can display process I/O counters. Open the Details tab, right-click the column headings, choose Select columns and enable the I/O write counters.

    These measurements can reveal whether a process is repeatedly generating operations, but the limitations discussed earlier still apply. High I/O totals alone do not establish equivalent physical SSD writes.

    Users who want to examine actual storage endurance should consult the health-monitoring software supplied by their SSD manufacturer or a reputable utility capable of reading the drive's SMART information.

    Useful indicators include total host writes, percentage used or remaining lifetime estimates, depending on the information supported by the drive.

    Recording those values and comparing them after a period of ordinary operation can reveal whether total storage writes are increasing unusually quickly.

    A meaningful comparison should take account of other applications and background activity. A system downloading games, installing updates, synchronizing cloud files or recording gameplay may legitimately write substantial amounts of data.

    If abnormal activity is observed, noting the graphics driver version, Windows version, GPU model and whether Resizable BAR is enabled can also help when submitting a bug report.

    AMD provides a Bug Report Tool for reporting driver problems, and detailed information about reproducible conditions may help its engineers identify configurations that are still affected after future updates.

    Temporary Workarounds Are Available but Have Drawbacks

    Until AMD releases a corrected driver, affected users have identified several potential workarounds.

    The simplest involves stopping AMD External Events Utility through Windows Services. This can be done by pressing Win + R, entering services.msc and locating the service in the list.

    Opening its properties allows users to stop the service and change its startup behavior, preventing the associated process from running automatically.

    Some users report that doing so immediately eliminates the abnormal activity.

    However, the service performs legitimate graphics-related functions, and disabling it may interfere with AMD Adrenalin shortcuts, display detection behavior or other features that depend on the service.

    The effect on FreeSync appears to vary between configurations. Some users report that adaptive synchronization continues working normally, while others have observed changes in frame pacing or problems involving particular frame-generation applications.

    It would therefore be inaccurate to state that disabling the service always breaks FreeSync, just as it would be misleading to promise that the change has no disadvantages.

    Another workaround involves disabling Resizable BAR in the motherboard's BIOS or UEFI settings.

    Resizable BAR allows a processor to access larger portions of a graphics card's video memory through the PCI Express interface. AMD markets related functionality under its Smart Access Memory branding on compatible platforms.

    The feature can improve performance in certain games, although the magnitude of the benefit varies substantially according to the game and hardware configuration.

    Disabling it has reportedly stopped the abnormal disk activity on some affected computers. Other users have found that the change makes no difference, indicating that it is not a universal solution.

    Because modifying BIOS settings can affect hardware functionality and performance, users should document their original configuration before making changes.

    A third possibility is temporarily returning to an older graphics driver that does not exhibit the problem on the affected system. Some users have reported success with earlier Adrenalin releases, but there is no universally confirmed safe version for every hardware configuration.

    Rolling back also means giving up subsequent bug fixes, security improvements, game optimizations and support for newer software.

    More aggressive suggestions have circulated online, including replacing driver files manually or deleting AMD components from protected Windows directories.

    These methods are not appropriate as general recommendations. Modifying driver files outside the normal installation process can introduce instability, compromise system integrity or complicate future updates.

    For most users, monitoring the system and waiting for AMD's official fix is preferable to making unnecessary changes, particularly when no abnormal disk activity has been detected.

    AMD Targets Adrenalin 26.11.1 for the Official Fix

    AMD has indicated that its planned correction should be included in Adrenalin Edition 26.11.1, the company's next expected major graphics software update.

    An AMD representative first acknowledged the reports in October, confirming that the company was investigating the issue with the intention of resolving it in a future driver release.

    Subsequent discussions provided more specific information. AMD_Vik explained that a colleague working on the kernel-mode driver had submitted a fix and that the proposed change appeared to involve relatively low implementation risk.

    However, the correction missed the development deadline for Adrenalin 26.10.2, which explains why the October 30 update does not address the problem.

    The representative later indicated that version 26.11.1 should include the necessary change.

    As of November 9, AMD has not published a confirmed release date for that version, and the current official release notes do not establish that the fix has already been distributed.

    The reported plan should therefore be treated as an expected correction rather than a completed resolution.

    Once the updated driver becomes available, affected users should examine AMD's official release notes and install the appropriate package for their graphics hardware.

    It will also be important to verify whether the update eliminates the activity on systems that previously reproduced the problem. A driver release advertised as containing a fix does not necessarily guarantee that every unusual I/O pattern observed by users has the same underlying cause.

    A useful follow-up test would involve comparing process activity and SSD host-write measurements before and after installing the corrected driver.

    That could help determine not only whether the excessive background operations have stopped, but also whether those operations were contributing materially to physical storage writes.

    For now, AMD's acknowledgment and the reported progress toward a fix are encouraging, although the final effectiveness of the correction remains to be demonstrated.

    A Driver Reliability Issue That Deserves Attention Without Unnecessary Alarm

    The Radeon External Events Utility problem illustrates how a seemingly minor defect in background software can raise concerns well beyond the component's intended purpose.

    Graphics drivers are normally judged by gaming performance, stability, compatibility and the features they provide. However, they also contain services and supporting components that interact continuously with Windows and other system hardware.

    A problem in one of those components can have consequences unrelated to graphics rendering, including unnecessary system resource usage, background activity and potentially additional storage wear.

    That makes AMD's investigation important even if the eventual impact on SSD longevity proves smaller than the most alarming estimates suggest.

    The current evidence supports several conclusions. Some Radeon systems exhibit abnormal I/O activity associated with AMD External Events Utility, AMD has reproduced and acknowledged the underlying problem, and a correction is expected in Adrenalin 26.11.1.

    What has not been established is that every reported byte of process activity represents a physical NAND write, or that the driver is directly responsible for the full SSD health reductions described by individual users.

    Those distinctions matter because they determine whether the problem is primarily one of inefficient background processing or a more serious issue involving sustained storage endurance consumption.

    For Radeon owners, the immediate response should be proportionate to the evidence. Checking Resource Monitor and the SSD's health information is sensible, particularly on systems that remain powered on for long periods. Disabling services or changing BIOS settings is less advisable unless the computer actually exhibits the abnormal behavior and the consequences of the workaround are understood.

    The pending driver correction should resolve the underlying software defect without requiring users to sacrifice graphics functionality.

    The larger lesson concerns transparency in system software. Background processes are expected to operate efficiently, and users generally have little reason to monitor the activity of individual driver components. When a defect creates persistent operations that ordinary users cannot easily explain, reliable diagnostics and timely vendor communication become essential.

    AMD has acknowledged the problem and identified a likely technical cause. The next important step is delivering the fix, documenting the affected configurations and establishing whether the abnormal activity translated into meaningful physical SSD writes.

    Until those questions are answered, the issue warrants attention and careful monitoring, but not the assumption that every Radeon-equipped computer is silently destroying its storage.


    --
    Via JLA Forums web gateway for alt.comp.hardware.pc-homebuilt

    --- PyGate Linux v1.5.19
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From Ant@3:633/10 to All on Wednesday, November 11, 2026 02:35:50
    Subject: Re: AMD Confirms Radeon Driver Bug That May Accelerate SSD Wear, With Fix Expected in Adrenalin 26.11.1

    No link?

    MummyChunk <mummycullen@gmail-dot-com.no-spam.invalid> wrote:
    AMD Confirms Radeon Driver Bug That May Accelerate SSD Wear, With Fix Expected in Adrenalin 26.11.1

    AMD has acknowledged a software defect affecting some Radeon graphics systems that can generate continuous background disk activity, potentially increasing unnecessary writes to solid-state drives even when a computer is sitting idle. The problem has been traced to a component of AMD's Adrenalin graphics software, with users reporting sustained activity ranging from approximately 0.3MB/s to as much as 5MB/s. If those figures represent actual storage writes maintained over extended periods, the additional workload could consume a substantial portion of an SSD's rated write endurance.

    The issue involves AMD External Events Utility, a Windows service responsible for various display-related functions, including monitor detection and graphics software shortcuts. Its associated process, atieclxx.exe, has been observed generating unusually high amounts of input/output activity on affected systems. Some users have reported that their SSD health indicators declined faster than expected, including one account describing a reduction of approximately one percentage point per month over a 16-month period.

    AMD representatives have confirmed that the company is investigating the problem and have identified an apparent interaction between the External Events Utility and Microsoft's PCIe filter driver. A correction has reportedly already been submitted by AMD's driver development team, with the company now targeting Adrenalin Edition 26.11.1 for its release. The update is expected later in November, although no exact availability date has been announced.

    There is, however, an important distinction between the existence of the software bug and the extent of the damage it may cause. While abnormal background activity has been documented and AMD has acknowledged the underlying issue, the reported reductions in SSD lifespan have not been independently established as a direct consequence of the driver. Furthermore, some Windows performance counters can report I/O operations that do not translate directly into physical writes to NAND flash memory.

    The available evidence therefore establishes a legitimate driver problem that warrants investigation, but it does not demonstrate that every affected computer is experiencing the maximum theoretical SSD wear described in some reports.

    The Problem Originates in AMD External Events Utility

    AMD External Events Utility is installed alongside Radeon graphics software and normally operates quietly in the background. Its responsibilities include responding to changes in connected monitors, handling certain display events and supporting functionality associated with AMD's graphics control software.

    The service uses processes including atieclxx.exe, known as the AMD External Events Client Module, and atiesrxx.exe, the corresponding service component. Under normal circumstances, these processes should consume relatively few system resources while performing their designated tasks.

    Reports from affected users suggest that a software interaction can instead cause the client process to repeatedly perform operations that generate persistent I/O activity. This behavior may continue even when no applications are running that would ordinarily require significant disk access.

    Because the amount of activity can be relatively modest at any individual moment, the problem may go unnoticed during everyday use. A background process generating several hundred kilobytes per second is unlikely to attract attention on a modern NVMe SSD capable of transferring several gigabytes per second. The computer may remain responsive, games may perform normally, and there may be no obvious warning that something unusual is happening.

    The concern arises from the persistence of the activity rather than its immediate impact on performance.

    AMD representative AMD_Vik discussed the issue in a Reddit troubleshooting thread, explaining that the behavior appeared to involve the External Events Utility interacting with Microsoft's PCIe filter driver. This component operates within the broader Windows hardware and driver infrastructure, where software communicates with devices connected through the PCI Express bus.

    That explanation suggests the problem is more complicated than an application simply creating an oversized diagnostic log. It appears to involve repeated operations within the graphics driver and operating system environment, although AMD has not published a complete technical root-cause analysis.

    AMD_Vik also confirmed that the company had reproduced the issue internally and that a colleague had submitted a proposed correction. The representative described the change as relatively low risk but explained that it had to be coordinated with AMD's established driver release schedule.

    The defect is therefore not merely an unverified complaint circulating through online forums. AMD has acknowledged the underlying behavior, although important questions remain about the precise mechanism and its consequences for physical storage devices.

    How Much Data Could the Bug Write to an SSD?

    The most alarming claims concern the cumulative amount of data that could be written if the reported activity continues uninterrupted.

    Users have reported different rates, with several observations falling between 0.3MB/s and 1.5MB/s. One particularly concerning account described activity reaching approximately 5MB/s while the computer was otherwise idle.

    At first glance, even 5MB/s appears insignificant compared with the performance of a modern SSD. However, sustained activity can accumulate into much larger totals over days and weeks.

    Assuming decimal megabytes and continuous activity for 24 hours, the theoretical figures are:

    0.3MB/s: Approximately 26GB per day, or 0.78TB over 30 days.

    0.7MB/s: Approximately 60GB per day, or 1.81TB over 30 days.

    1.5MB/s: Approximately 130GB per day, or 3.89TB over 30 days.

    5MB/s: Approximately 432GB per day, or 12.96TB over 30 days.

    These calculations illustrate why the issue has attracted attention. At the highest reported rate, a computer left running continuously could generate nearly 13TB of reported activity in a single month.

    However, those totals are mathematical projections, not measurements from a controlled month-long experiment. They assume the reported rate remains constant every second, that the computer operates continuously and, most importantly, that the activity corresponds to data actually written to storage.

    The final assumption has not been established across all affected systems.

    Some users have reported significantly lower activity, while others have found no abnormal behavior whatsoever. One Radeon owner described a process accumulating approximately 19.5GB of I/O writes during 12 hours of operation, substantially below the maximum theoretical figure.

    Another user reported different results on two Radeon systems: one computer showed almost no activity from the process, while another continuously accumulated approximately 30MB per minute.

    These differences indicate that the problem is not universal and may depend on a combination of driver versions, graphics hardware, operating system configuration and other installed components.

    AMD has not published a comprehensive list of affected configurations or identified the conditions necessary to reproduce the problem reliably on every system.

    High Windows I/O Counters Do Not Necessarily Mean Equivalent NAND Wear

    One of the most important technical qualifications concerns how Windows measures application activity.

    Task Manager can display process-level I/O counters, including the number of bytes associated with write operations. However, these counters do not necessarily provide a direct measurement of data physically programmed into an SSD's NAND flash memory.

    Windows processes interact with numerous resources through its I/O subsystem, including files, devices and other operating system components. Certain operations can be counted at the process level without producing an equivalent amount of physical storage activity.

    In addition, Windows uses caching and other mechanisms that can change when and how data is committed to storage. An application's reported write activity therefore should not automatically be treated as identical to the SSD's total host writes, much less the amount of data programmed internally into NAND.

    This distinction has emerged in discussions among Radeon owners attempting to investigate the problem.

    One user examined the affected process using Microsoft's Process Monitor utility and observed repeated access to system information, device enumeration and file metadata operations. Although the process accumulated I/O activity, that particular investigation did not identify corresponding physical file writes.

    Another user examining a different configuration reported sustained disk writes visible in Windows Resource Monitor and found that stopping AMD External Events Utility eliminated the activity.

    These observations are not necessarily contradictory. They may reflect different manifestations of the same bug, different hardware and software configurations, or limitations in the monitoring methods being used.

    The important point is that an increasing I/O counter alone does not prove that the SSD is experiencing the same amount of NAND wear.

    A more reliable investigation would combine process-level monitoring with the drive's own health information, including its reported total host writes. Comparing those measurements over a controlled period can help determine whether the abnormal activity is actually contributing to increased storage writes.

    Even then, attributing every additional write to one process requires care because Windows, antivirus software, background updates and other applications also generate storage activity.

    Consequently, claims that the bug has already destroyed a particular percentage of an SSD's lifespan should be treated cautiously unless supported by detailed measurements that establish a causal connection.

    Why Unnecessary Writes Could Shorten SSD Lifespan

    Unlike conventional hard drives, which store information magnetically on rotating platters, SSDs use NAND flash memory. Flash cells have a finite endurance because programming and erasing data gradually changes their electrical characteristics.

    Modern SSD controllers manage this limitation through wear leveling, overprovisioning, error correction and other techniques intended to distribute activity and maintain reliable operation over time.

    Manufacturers commonly specify drive endurance using a rating called terabytes written, or TBW. This describes the amount of host data a drive is designed to accommodate under defined operating conditions, often as part of its warranty specification.

    For example, an SSD rated for 300TBW is designed around a substantially different total write workload from a drive rated for 600TBW. Capacity, NAND technology, controller design and the manufacturer's endurance targets all influence those figures.

    If a genuine software defect adds several terabytes of unnecessary writes every month, it could consume an appreciable portion of the rated endurance of a lower-capacity consumer SSD.

    The potential impact is especially relevant to inexpensive drives using QLC NAND, which stores four bits of information in each memory cell. QLC technology offers high storage density and can reduce manufacturing costs, but it generally has lower raw program-and-erase endurance than comparable TLC NAND, which stores three bits per cell.

    That does not mean every QLC drive is inherently fragile. Controller quality, spare capacity, error correction and firmware behavior can make a substantial difference, while many modern consumer SSDs are designed to withstand years of normal use.

    Nor should the TBW rating be interpreted as a countdown to inevitable failure. Reaching the manufacturer's rated endurance does not mean a drive will immediately stop working, and SSDs can sometimes operate well beyond their advertised write limits.

    Nevertheless, unnecessary writes are undesirable because they consume storage resources without providing a useful benefit.

    Another complication is write amplification. An SSD may need to perform internal NAND operations beyond the amount of data originally submitted by the host, depending on its internal organization and workload.

    This means that even an accurate measurement of host writes does not directly reveal the complete internal wear experienced by the flash memory.

    The real-world consequence of AMD's bug therefore depends on whether the reported activity reaches the drive, how much data is actually written and how the SSD handles that workload.

    At present, those factors have not been measured consistently enough to establish a general estimate of reduced lifespan.

    Which Radeon Drivers and Windows Systems Are Affected?

    Reports have emerged from users running several versions of AMD Software: Adrenalin Edition, including 26.10.1 and 26.10.2.

    The latter was released on October 30 and addressed several unrelated stability problems, including game crashes, driver timeouts and issues involving graphics overlays. However, its official release notes do not list a correction for the External Events Utility problem.

    Some users have traced similar behavior to earlier driver releases, and one report suggested that the issue may have been present since Adrenalin 26.7.1. Other accounts describe problems occurring after Windows updates, raising the possibility that changes to the operating system or PCIe driver environment contribute to triggering the defect.

    AMD has not confirmed a single version as the starting point, and the evidence does not establish that every release after a particular date is affected.

    Reports cover both Windows 10 and Windows 11, along with several generations of Radeon hardware.

    The issue has also been observed on systems with different processor and chipset configurations, making it inappropriate to assume that it is limited exclusively to computers using AMD Ryzen CPUs.

    This is relevant because AMD's graphics software supports Radeon cards installed in systems built around either AMD or Intel processors.

    There is also uncertainty about whether integrated Radeon graphics solutions are affected in the same way as dedicated graphics cards. Some Ryzen processors use graphics components supported by related Adrenalin packages, but the current reports do not provide sufficient evidence to establish the prevalence of the issue across those systems.

    Importantly, independent testing has not reproduced the problem universally.

    PC Gamer reported that one of its staff members using AMD graphics hardware was unable to reproduce the excessive activity. Reddit users have similarly described systems running recent driver releases without unusual disk behavior.

    These findings suggest that Radeon owners should check whether their own computers exhibit the problem rather than assuming that every installation of AMD's graphics software is damaging an SSD.

    How to Check Whether Your Computer Is Experiencing the Problem

    Windows includes diagnostic tools that can help identify abnormal background activity without requiring additional software.

    The most useful starting point is Resource Monitor, which provides more detailed information about disk activity than the standard Task Manager overview.

    To open it, press the Windows key and R simultaneously, enter resmon and press Enter. Select the Disk tab and examine the processes listed under Disk Activity or Processes with Disk Activity.

    Look for atieclxx.exe, identified as AMD External Events Client Module, and check whether it is generating persistent write activity when the computer is otherwise idle.

    It may also be useful to examine atiesrxx.exe, the related AMD service process.

    The relevant behavior is sustained activity rather than an occasional spike. Background services normally access files and system resources from time to time, and brief activity immediately after starting Windows does not establish that the system is affected.

    For a more complete investigation, Task Manager can display process I/O counters. Open the Details tab, right-click the column headings, choose Select columns and enable the I/O write counters.

    These measurements can reveal whether a process is repeatedly generating operations, but the limitations discussed earlier still apply. High I/O totals alone do not establish equivalent physical SSD writes.

    Users who want to examine actual storage endurance should consult the health-monitoring software supplied by their SSD manufacturer or a reputable utility capable of reading the drive's SMART information.

    Useful indicators include total host writes, percentage used or remaining lifetime estimates, depending on the information supported by the drive.

    Recording those values and comparing them after a period of ordinary operation can reveal whether total storage writes are increasing unusually quickly.

    A meaningful comparison should take account of other applications and background activity. A system downloading games, installing updates, synchronizing cloud files or recording gameplay may legitimately write substantial amounts of data.

    If abnormal activity is observed, noting the graphics driver version, Windows version, GPU model and whether Resizable BAR is enabled can also help when submitting a bug report.

    AMD provides a Bug Report Tool for reporting driver problems, and detailed information about reproducible conditions may help its engineers identify configurations that are still affected after future updates.

    Temporary Workarounds Are Available but Have Drawbacks

    Until AMD releases a corrected driver, affected users have identified several potential workarounds.

    The simplest involves stopping AMD External Events Utility through Windows Services. This can be done by pressing Win + R, entering services.msc and locating the service in the list.

    Opening its properties allows users to stop the service and change its startup behavior, preventing the associated process from running automatically.

    Some users report that doing so immediately eliminates the abnormal activity.

    However, the service performs legitimate graphics-related functions, and disabling it may interfere with AMD Adrenalin shortcuts, display detection behavior or other features that depend on the service.

    The effect on FreeSync appears to vary between configurations. Some users report that adaptive synchronization continues working normally, while others have observed changes in frame pacing or problems involving particular frame-generation applications.

    It would therefore be inaccurate to state that disabling the service always breaks FreeSync, just as it would be misleading to promise that the change has no disadvantages.

    Another workaround involves disabling Resizable BAR in the motherboard's BIOS or UEFI settings.

    Resizable BAR allows a processor to access larger portions of a graphics card's video memory through the PCI Express interface. AMD markets related functionality under its Smart Access Memory branding on compatible platforms.

    The feature can improve performance in certain games, although the magnitude of the benefit varies substantially according to the game and hardware configuration.

    Disabling it has reportedly stopped the abnormal disk activity on some affected computers. Other users have found that the change makes no difference, indicating that it is not a universal solution.

    Because modifying BIOS settings can affect hardware functionality and performance, users should document their original configuration before making changes.

    A third possibility is temporarily returning to an older graphics driver that does not exhibit the problem on the affected system. Some users have reported success with earlier Adrenalin releases, but there is no universally confirmed safe version for every hardware configuration.

    Rolling back also means giving up subsequent bug fixes, security improvements, game optimizations and support for newer software.

    More aggressive suggestions have circulated online, including replacing driver files manually or deleting AMD components from protected Windows directories.

    These methods are not appropriate as general recommendations. Modifying driver files outside the normal installation process can introduce instability, compromise system integrity or complicate future updates.

    For most users, monitoring the system and waiting for AMD's official fix is preferable to making unnecessary changes, particularly when no abnormal disk activity has been detected.

    AMD Targets Adrenalin 26.11.1 for the Official Fix

    AMD has indicated that its planned correction should be included in Adrenalin Edition 26.11.1, the company's next expected major graphics software update.

    An AMD representative first acknowledged the reports in October, confirming that the company was investigating the issue with the intention of resolving it in a future driver release.

    Subsequent discussions provided more specific information. AMD_Vik explained that a colleague working on the kernel-mode driver had submitted a fix and that the proposed change appeared to involve relatively low implementation risk.

    However, the correction missed the development deadline for Adrenalin 26.10.2, which explains why the October 30 update does not address the problem.

    The representative later indicated that version 26.11.1 should include the necessary change.

    As of November 9, AMD has not published a confirmed release date for that version, and the current official release notes do not establish that the fix has already been distributed.

    The reported plan should therefore be treated as an expected correction rather than a completed resolution.

    Once the updated driver becomes available, affected users should examine AMD's official release notes and install the appropriate package for their graphics hardware.

    It will also be important to verify whether the update eliminates the activity on systems that previously reproduced the problem. A driver release advertised as containing a fix does not necessarily guarantee that every unusual I/O pattern observed by users has the same underlying cause.

    A useful follow-up test would involve comparing process activity and SSD host-write measurements before and after installing the corrected driver.

    That could help determine not only whether the excessive background operations have stopped, but also whether those operations were contributing materially to physical storage writes.

    For now, AMD's acknowledgment and the reported progress toward a fix are encouraging, although the final effectiveness of the correction remains to be demonstrated.

    A Driver Reliability Issue That Deserves Attention Without Unnecessary Alarm

    The Radeon External Events Utility problem illustrates how a seemingly minor defect in background software can raise concerns well beyond the component's intended purpose.

    Graphics drivers are normally judged by gaming performance, stability, compatibility and the features they provide. However, they also contain services and supporting components that interact continuously with Windows and other system hardware.

    A problem in one of those components can have consequences unrelated to graphics rendering, including unnecessary system resource usage, background activity and potentially additional storage wear.

    That makes AMD's investigation important even if the eventual impact on SSD longevity proves smaller than the most alarming estimates suggest.

    The current evidence supports several conclusions. Some Radeon systems exhibit abnormal I/O activity associated with AMD External Events Utility, AMD has reproduced and acknowledged the underlying problem, and a correction is expected in Adrenalin 26.11.1.

    What has not been established is that every reported byte of process activity represents a physical NAND write, or that the driver is directly responsible for the full SSD health reductions described by individual users.

    Those distinctions matter because they determine whether the problem is primarily one of inefficient background processing or a more serious issue involving sustained storage endurance consumption.

    For Radeon owners, the immediate response should be proportionate to the evidence. Checking Resource Monitor and the SSD's health information is sensible, particularly on systems that remain powered on for long periods. Disabling services or changing BIOS settings is less advisable unless the computer actually exhibits the abnormal behavior and the consequences of the workaround are understood.

    The pending driver correction should resolve the underlying software defect without requiring users to sacrifice graphics functionality.

    The larger lesson concerns transparency in system software. Background processes are expected to operate efficiently, and users generally have little reason to monitor the activity of individual driver components. When a defect creates persistent operations that ordinary users cannot easily explain, reliable diagnostics and timely vendor communication become essential.

    AMD has acknowledged the problem and identified a likely technical cause. The next important step is delivering the fix, documenting the affected configurations and establishing whether the abnormal activity translated into meaningful physical SSD writes.

    Until those questions are answered, the issue warrants attention and careful monitoring, but not the assumption that every Radeon-equipped computer is silently destroying its storage.



    --
    "God's voice thunders in marvelous ways; he does great things beyond our understanding. He says to the snow, 'Fall on the earth,' and to the rain shower, 'Be a mighty downpour.'" --Job 37:5-6. iPod's 25th BD after ?fox's 22nd BD.
    Note: A fixed width font (Courier, Monospace, etc.) is required to see this signature correctly.
    /\___/\ Ant(Dude) @ http://aqfl.net & http://antfarm.home.dhs.org.
    / /\ /\ \ Please nuke ANT if replying by e-mail.
    | |o o| |
    \ _ /
    ( )

    --- PyGate Linux v1.5.19
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From Paul@3:633/10 to All on Tuesday, November 10, 2026 22:31:11
    Subject: Re: AMD Confirms Radeon Driver Bug That May Accelerate SSD Wear, With Fix Expected in Adrenalin 26.11.1

    On Tue, 11/10/2026 9:35 PM, Ant wrote:

    MummyChunk <mummycullen@gmail-dot-com.no-spam.invalid> wrote:
    AMD Confirms Radeon Driver Bug That May Accelerate SSD Wear, With Fix Expected in Adrenalin 26.11.1
    No link?


    Punch a few of the keywords into Google, and you'll get "similar" articles.

    Even if you tried here, you're unlikely to get through to the web-to-usenet gateway
    the OP is using for these posts. It's quite likely the OP cannot even *see* your post,
    let alone reply to it. Maybe Mummy is connected via an RS232 modem port to make these posts
    (as in, Mummy is the site owner, and the site owner happens to be clueless).

    http://www.jlaforums.com/viewtopic.php?p=713271213#713271213

    You won't be able to get through the domain block on that,
    so really this situation is so much squirrel-shit, ignore and move on.

    Paul

    --- PyGate Linux v1.5.19
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From MummyChunk@3:633/10 to All on Wednesday, November 11, 2026 12:18:32
    Subject: Re: AMD Confirms Radeon Driver Bug That May Accelerate SSD Wear, With Fix Expected in Adrenalin 26.11.1

    Paul wrote:
    On Tue, 11/10/2026 9:35 PM, Ant wrote:



    MummyChunk <mummycullen> wrote:


    AMD Confirms Radeon Driver Bug That May Accelerate SSD Wear, With Fix Expected in Adrenalin 26.11.1


    No link?




    Punch a few of the keywords into Google, and you'll get "similar" articles.

    Even if you tried here, you're unlikely to get through to the web-to-usenet gateway
    the OP is using for these posts. It's quite likely the OP cannot even *see* your post,
    let alone reply to it. Maybe Mummy is connected via an RS232 modem port to make these posts
    (as in, Mummy is the site owner, and the site owner happens to be clueless).

    http://www.jlaforums.com/viewtopic.php?p=713271213#713271213

    You won't be able to get through the domain block on that,
    so really this situation is so much squirrel-sh*t, ignore and move on.

    Paul



    Gee, sounds like someone woke up on the wrong side of the bed today..

    ?? ??


    --
    Via JLA Forums web gateway for alt.comp.hardware.pc-homebuilt

    --- PyGate Linux v1.5.19
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)