Recommended Free Tools
Some links on this page are affiliate links: if you buy through them we may earn a commission, at no extra cost to you.
The AnandTech poll records substantial owner dissatisfaction with IBM’s 60GXP, 75GXP and 120GXP hard drives, including repeat failures—but its often-quoted 44–45% result is not a population-wide failure rate. It was a voluntary forum poll, counted people rather than consistently counted drives, and left “failure” open to interpretation. The historical evidence supports a serious 75GXP reliability controversy, but not the claim that every GXP generation failed at the same rate.
What the AnandTech poll asked
Opened on June 25, 2002, the AnandTech thread was titled “POLL: IBM HD OWNERS (60,75,120GXP owners in particular) – Have you had one or more drive(s) fail?” It asked IBM hard-drive owners whether they had experienced one or more failures, with particular attention to those three Deskstar families. Read the original poll and discussion.
| # | Preview | Product | Price | |
|---|---|---|---|---|
| 1 |
|
IBM HDS728040PLA320 40GB SATA DESKSTAR 7K80 7200r (Renewed) | $14.99 | Buy on Amazon |
| 2 |
|
IBM Deskstar 07N8134 20.4GB UDMA/100 7200RPM 2MB IDE Hard Drive | $24.95 | Buy on Amazon |
| 3 |
|
IBM DeskStar 20.5 GB EIDE Hard Drive | $74.00 | Buy on Amazon |
| 4 |
|
IBM Deskstar 07N9682 40GB UDMA/100 7200RPM 2MB IDE Hard Drive | $259.00 | Buy on Amazon |
| 5 |
|
IBM DeskStar 60GB EIDE Hard Drive | $142.21 | Buy on Amazon |
The distinction between a voter, a drive and a failure incident matters. One person with one failed drive and another with several failed drives could each cast one vote; a voter might also count a failed replacement as another incident. The thread therefore cannot be read as a clean tally of the share of all drives that failed.
What owners said happened
The posts are firsthand accounts, not audited service records. They show the range and intensity of experiences, rather than the odds that a randomly selected IBM drive would fail.
#1 Best Overall
Failures and repeat replacements
- One owner reported seven failures among eight IBM drives; another reported six RMAs among ten 75GXP drives. These are individual self-reports, not verified fleet statistics.
- Some participants said replacement drives failed too. Other owners described an exchange that went smoothly, while some complained of delays.
- One participant reported that a 40GB 120GXP failed in less than a month.
These accounts help explain why the poll attracted attention: some users were dealing with more than a single inconvenient failure. They do not establish whether repeat failures came from a product defect, a replacement batch, the user’s environment or another cause.
Warning signs and data problems
Owners described repeated clicking, scratching or crunching sounds, bad sectors, data corruption, boot-time detection problems, repeated operating-system trouble, and drives that degraded before becoming inaccessible. Some considered a drive effectively failed when it remained partly accessible but could no longer be trusted. One poster said a recovery utility retrieved about 90% of the data; that is one person’s outcome, not a general recovery expectation. The thread contains the reported symptoms and recovery anecdote.
Surviving drives and conflicting theories
Other participants reported 30GB 75GXPs or 60GXPs still working after years, including drives used continuously or in RAID configurations. Owners offered competing explanations for failures: heat, heavy workloads, RAID use, or operation in hot enclosures. Some with active cooling reported no failures; others reported problems despite cooling or prolonged hot operation. Those observations are correlations reported by owners, not controlled evidence that cooling or workload caused—or prevented—a failure. The second page includes additional owner reports and discussion.
Why 44–45% is not a drive failure rate
The thread author cited a result of 44%, then later 45%. That is the poll author’s running result among participants, not a verified estimate for IBM customers or for all drives sold. The author also acknowledged that voters owned different numbers of drives. The poll’s second page records the percentage and its limitations.
- Voluntary responses: Owners already worried about failures were especially likely to find and answer a failure poll. The people who replied were not a random sample.
- Unequal ownership: A one-drive owner and an owner of ten drives could each contribute one vote. A voter percentage cannot be converted into a per-drive percentage without the number of drives and their outcomes.
- Different meanings of failure: Some accounts describe a dead mechanism; others describe bad sectors, corruption, clicking, or a drive that remained readable but seemed unsafe. Counting these together produces no consistent failure definition.
- Model uncertainty: Some posters were unsure whether their drive was a 60GXP or a 75GXP, limiting comparisons between models.
- Missing exposure information: The thread does not consistently provide manufacturing batch, power-on hours, temperature, ventilation, workload, firmware, installation details or a complete RMA history. It cannot isolate a product defect from environmental or ordinary mechanical causes.
- Survivorship: A working drive is not proof of long-term reliability, and owners without trouble may have had less reason to join a failure discussion.
The wider Deskstar controversy
The poll appeared amid an established public controversy, especially around the 75GXP. Contemporary StorageReview coverage described an unusually high volume of complaints about the 75GXP and a lesser, less clear wave around the 60GXP. It also cautioned that the 60GXP’s reputation could lead ordinary hard-drive problems to be attributed to that family. These are historical assessments of complaints, not laboratory failure-rate measurements. StorageReview’s 120GXP coverage discusses the earlier models and their reputation.
Contemporary reporting also covered a proposed or filed class-action lawsuit alleging a common 75GXP design defect. That was an allegation, not proof that a court established a universal defect. IBM was reported as claiming a failure rate of about 1–2% and describing it as normal for the industry; that figure is IBM’s reported position, not an independently verified measurement. Ars Technica reported on the lawsuit allegations; Computerworld reported IBM’s stated failure-rate position.
Rank #3
- IBM Deskstar 40GV 20 GB UDMA/100 IDE Hard Drive General Features: 20 GB formatted capacity
- UDMA/100 5400 RPM spindle speed 512 KB buffer 9.5 ms average seek time (read)
- 5.56 ms average latency IDE interface 3.5-inch form factor
- S.M.A.R.T. function (Self Monitoring and Analysis Reporting Technology)
- Drive Fitness Test (DFT) technology
Other contemporary coverage described complaints about 75GXP failures, replacement delays and replacement units that owners said failed again. These reports reinforce that dissatisfaction was not invented by the forum poll, but they do not establish the prevalence or cause of failures across all production. The Register’s 2001 coverage documents the complaints and replacement concerns.
60GXP, 75GXP and 120GXP should not be treated as one case
| Family | What the available historical evidence supports | What it does not establish |
|---|---|---|
| 75GXP | It was the focus of the strongest reliability complaints and the reported lawsuit allegations. | A proven failure percentage for every capacity or production batch, or that every unit was defective. |
| 60GXP | Owners reported failures, and contemporary coverage described a related but less severe controversy. | A rate comparable to the 75GXP or a confirmed common cause. |
| 120GXP | The poll includes failure anecdotes; contemporary discussion also centered on the operating-hours guidance. | A failure rate comparable to the 75GXP, based on this self-selected poll. |
The thread also contains uncertainty about exact model identification, and it does not provide systematic sample sizes by model or capacity. It cannot support a reliable ranking of individual capacity variants.
What the 333-hours-per-month controversy meant
Contemporary reporting on the 120GXP described IBM documentation as specifying approximately 333 power-on hours per month—about 11 hours a day averaged across a month—and framed that figure as a reason not to assume the drive was intended for continuous 24/7 use. MacRumors covered the reported guidance in March 2002.
Rank #4
- IBM Deskstar 07N9682 40 GB UDMA/100IDE Hard Drive General Features: 40 GB formatted capacity
- UDMA/100 7200 RPM spindle speed 2 MB buffer 8.5 ms average seek time E-IDE/ATAPI interface
- 3.5-inch form factor Regulatory Approvals: UL CE TUV MIC C-Tick
That figure should not be treated as a guaranteed failure threshold: the report does not show that exceeding it automatically caused a drive to fail. Nor does the available contemporary coverage settle the exact engineering or legal meaning of the wording across every manual or GXP model. It establishes that the operating-hours language became a public controversy, not that 333 hours was a precise boundary between safe and unsafe operation.
What the poll can—and cannot—tell us
| Question | Careful reading of the evidence |
|---|---|
| Did IBM GXP owners report failures? | Yes. The thread contains numerous failure accounts. |
| Did some owners report repeated failures or RMAs? | Yes, as individual owner reports. |
| Did some drives survive for years? | Yes, according to owners in the thread. |
| Did 44–45% of all IBM drives fail? | No. That was the poll author’s reported voter result, not a population-wide drive rate. |
| Does the poll prove heat or workload caused failures? | No. It contains uncontrolled owner observations. |
| Were all three GXP families equally unreliable? | The available evidence does not establish that; the strongest historical controversy concerned the 75GXP. |
In short, the poll is useful evidence of how many enthusiasts felt affected and what failure looked like in use. It cannot compare IBM’s failure rate with competitors, identify a batch-level defect, or calculate the chance that a particular surviving drive will fail.
Windows Errors? Fix Them Before They Spread
Repair common Windows errors and clear accumulated junk for a smoother, more stable PC - no reinstall needed.Free scan · no reinstallOutdated Drivers Are Slowing You Down
One free scan finds every outdated or missing driver and matches the right update for your exact hardware.Free scan · exact hardware matchWhat to do with a surviving IBM GXP drive today
A surviving unit may be valuable for period-correct hardware testing or archival work, but its survival so far does not make it dependable modern storage. Do not make it the only copy of important data. If it contains material worth keeping, preserve the data before experimenting with the drive.
Best Value
- 60 GB of storage
- UltraDMA 100 interface
- Disk spins at 7,200 RPMs
- 8.5 millisecond seek time
- 3 year warranty
- Identify the exact model and capacity. Confirm whether the label says 60GXP, 75GXP, 120GXP or another Deskstar generation; do not infer the family from a vague recollection.
- Check SMART data if available. A legacy interface or adapter may not expose all drive health information. A clean report cannot guarantee future reliability.
- Image the drive before stressing it. If it is readable, make a complete image before running tests or attempting repairs. Keep at least two independent copies of valuable data elsewhere.
- Keep it cool and mechanically secure. These are sensible precautions for a mechanical drive, not a proven cure for the historical GXP problems.
- Use it only where failure is acceptable. For a retro-computing project, keep working data and backups on other storage; avoid relying on the GXP as primary storage.
If it clicks or becomes intermittently readable
Treat unusual clicking, repeated seek-reset behavior or sudden detection problems as warnings. Stop normal use and avoid repeated power cycles. Do not run write-repair utilities or destructive diagnostics on the only copy of valuable data. If the drive is still readable, clone it with a tool designed to handle read errors, then work from the clone. When the data is irreplaceable and the mechanism appears unstable, a professional recovery service is safer than repeated DIY attempts.
If it has bad sectors or was part of RAID 0
Bad sectors on an aged drive are reason to migrate data, not evidence by themselves that every drive in the line was defective. A successful format or scan does not make the disk trustworthy again. In RAID 0, one member’s failure can make the array’s data unavailable; the thread includes RAID-related owner reports, but does not show that RAID itself caused the failures. The poll discussion includes RAID and workload anecdotes.
Replacement drives and warranty context
Forum participants described IBM replacing failed drives, sometimes with refurbished or replacement units; their accounts varied from quick exchanges to long waits and repeat failures. They do not establish the exact warranty terms for every original Deskstar sale. IBM’s current warranty pages say that coverage and service levels vary by product and country, and that replacement parts may be used if they are in good working order and functionally equivalent. Those current general policies are not evidence of the consumer Deskstar warranty terms in 2000–2002. IBM’s current warranty overview explains that warranty terms vary.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
The transfer of IBM’s storage business to Hitachi forms part of the period’s corporate context and was discussed by forum participants, but it does not prove anything about the GXP failure rate. The poll is a record of consumer confidence and owner experience, not a technical postmortem.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
