CrystalDiskInfo Guide: How to Read Your Drive's Health Status
CrystalDiskInfo (often searched as “Crystal Disk Info”) is the free Windows tool most people reach for when they want to know whether a hard drive or SSD is healthy. It reads the drive’s built-in S.M.A.R.T. data and sums it up as Good, Caution, or Bad — but the color is only the start. This CrystalDiskInfo guide shows you how to get it safely, what each status means, which numbers in the attribute table actually predict failure, and how to run a CrystalDiskInfo SSD health check.
If your drive is clicking, grinding, or beeping, don’t install anything — power it down. A drive with a mechanical fault can get worse every minute it runs. See our clicking drive guide.
How to download CrystalDiskInfo safely
CrystalDiskInfo is developed by Crystal Dew World. Get it from the official site, crystalmark.info, and follow the download links from there.
- Watch out for lookalike sites. Because it’s popular, many third-party “download” sites rehost CrystalDiskInfo — some wrap it in their own installers with bundled adware. Ads at the top of search results can point to these too. Check the address bar before you download.
- Pick the Standard Edition. The other editions (such as Shizuku Edition) work the same, just with themed artwork.
- Installer or portable ZIP — either works. The portable version runs without installing.
CrystalDiskInfo is read-only and safe: it reads the drive’s health log and doesn’t write to your files or repair anything. (Leave the AAM/APM settings under the advanced menus alone — those change drive behavior and aren’t needed for a health check.)
Reading the CrystalDiskInfo main screen
Open the program and each detected drive appears as a button across the top, labeled with its status, temperature, and drive letter. Click one to see its details.
| Health Status | Color | What it means |
|---|---|---|
| Good | Blue | No SMART warnings. For SSDs, a percentage (e.g. “Good 92%”) estimates remaining life. |
| Caution | Yellow | A warning attribute is non-zero or an SSD is low on life. Back up now. |
| Bad | Red | The drive has crossed its own failure threshold. Back up immediately and replace it. |
| Unknown | Gray | CrystalDiskInfo can’t read SMART data — common with USB enclosures (see below). |
The panel on the right shows useful context: Power On Hours, Power On Count, and for many drives Total Host Writes. Below it is the attribute table — that’s where the real information is.
Tip: raw values are shown in hexadecimal by default, so 0A means 10. To switch to normal numbers, go to Function → Advanced Feature → Raw Values and choose the decimal option.
The SMART attributes that matter
Most attributes are vendor-specific and safe to ignore. These are the ones worth reading, especially the Raw Values column:
| ID | Attribute | What to look for |
|---|---|---|
| 05 | Reallocated Sectors Count | Bad sectors already swapped for spares. Any growth is a warning. Details. |
| C5 (197) | Current Pending Sector Count | Sectors the drive can’t read right now. Data there is at risk. Details. |
| C6 (198) | Uncorrectable Sector Count | Sectors that failed even after retries. Any non-zero value is serious. |
| BB (187) | Reported Uncorrectable Errors | Read errors error correction couldn’t fix. |
| C7 (199) | UltraDMA CRC Error Count | Data corrupted in transit — usually a bad cable, port, or enclosure, not the drive itself. |
| 09 | Power-On Hours | Total running time. Context, not a fault by itself. |
| C2 (194) | Temperature | Current temperature. Check the maker’s rated range; sustained heat shortens drive life. |
By default, CrystalDiskInfo flags Caution when 05, C5, or C6 is above zero. A single reallocated sector on an older drive isn’t an emergency, but pending or uncorrectable sectors, or any count that rises between checks, mean the drive is degrading.
A rising C7 count is the one “warning” that’s often fixable: replace the SATA cable or try a different port or enclosure, then watch whether it keeps climbing.
CrystalDiskInfo SSD health check
SSDs don’t have sectors that wear like a hard drive’s platters — they wear by writes. CrystalDiskInfo shows a health percentage for most SSDs, based on what the drive reports about its own endurance. The attributes to read depend on the interface.
SATA SSDs (names vary by manufacturer):
- Wear Leveling Count / Media Wearout Indicator / SSD Life Left — how much rated endurance remains.
- Reallocated Sectors / Reallocated NAND Blocks — retired flash blocks.
- Total Host Writes — compare to the drive’s rated TBW (terabytes written) on its spec sheet.
NVMe SSDs use a standard set of fields:
- Percentage Used — the drive’s estimate of rated endurance consumed. It can exceed 100%; the drive doesn’t stop working at that point, but it’s beyond its rated life.
- Available Spare and Available Spare Threshold — reserve blocks remaining. Spare falling toward or below the threshold is serious.
- Media and Data Integrity Errors — errors the drive couldn’t correct. This should be 0. Any non-zero value means data integrity problems have already occurred.
- Critical Warning — should be 0.
- Data Units Written — total writes, comparable to TBW.
Note that SSDs often fail suddenly — controller or firmware failures can happen with a “Good 99%” health reading. Health status is a wear gauge, not a guarantee. See SSD not detected if one has already disappeared.
“Health Status: Unknown” on external drives
If an external drive shows Unknown or doesn’t appear at all, the drive may be fine — the problem is often its USB bridge, the small circuit that translates USB to SATA or NVMe. Many bridges pass SMART data through and CrystalDiskInfo supports a wide range of them, but some simply don’t. Options:
- Try a different enclosure or USB-to-SATA adapter (only for drives with a standard SATA connector — many portable drives have the USB port built onto the drive’s own board).
- Run the drive maker’s own utility, which may support its own enclosures.
- Connect an internal drive directly to a SATA port.
If an external drive isn’t detected anywhere, see external hard drive not showing up.
Set up alerts so you hear about problems early
CrystalDiskInfo can watch your drives in the background:
- Function → Resident keeps it running in the system tray.
- Function → Startup launches it when Windows starts.
- Function → Alert Features turns on alert sounds, Windows event log entries, and email notifications when a drive’s status changes.
The main value is catching a change — a new pending sector, a jump in reallocations — rather than a single snapshot.
What to do on Caution or Bad
- Back up first. Copy your most important files to a different drive before anything else. Don’t run defragmenters, long surface scans, or “repair” tools — they add hours of stress to a weakening drive.
- Note the attribute values and check again in a few days. Rising counts mean a countdown.
- Replace the drive. Our guide to SMART status bad walks through cloning and replacement.
- If files won’t copy, the drive freezes, or it starts making noise, stop. Every extra hour of struggling reads lowers the odds of a full recovery.
The bottom line
Download CrystalDiskInfo only from crystalmark.info, then look past the Good/Caution/Bad label to the raw values. For hard drives, watch reallocated, pending, and uncorrectable sectors; for SSDs, watch percentage used, available spare, and media errors. Turn on resident mode and alerts so you see changes as they happen, and back up the moment anything turns yellow.
If CrystalDiskInfo flagged your drive and it’s now failing to read, our hard drive recovery service recovers failing hard drives and SSDs for one flat $1,200 price for single devices. Shipping is free and prepaid, and you pay nothing unless the recovery succeeds.