In modern solid-state storage architecture, the question “Why does a 1TB SSD show only around 930GB of usable space?” resurfaces again and again.
For professionals working with NAND engineering, controller firmware, or system integration, this discrepancy is neither a flaw nor a trick — it’s the result of established standards, physical characteristics of flash memory, and the storage management logic that keeps modern SSDs fast and reliable.
To understand this gap, we need to look at storage capacity from an engineering perspective rather than a consumer-facing one.
1. Capacity Is Not a Single Definition: Standards Shape What “1TB” Means
On SSD packaging, “1TB” follows the SI decimal standard.
But once the drive is mounted, the operating system interprets the same number using binary units.
Two different measurement systems drive the difference:
| Systeem | Definition | Scale |
|---|---|---|
| Decimal (SI) | 1TB = 10¹² bytes | Used by all storage manufacturers |
| Binary (IEC) | 1TiB = 2⁴⁰ bytes | Used by operating systems |
When the OS recalculates the “1TB” decimal capacity into binary, the result becomes:
➡ ~0.93TiB (about 930GB usable)
This isn’t a loss—it’s the direct consequence of two different capacity standards that have co-existed since the early days of consumer storage.
2. NAND Flash Architecture Requires Space You Never See
Flash memory isn’t a flat, linear medium. Every SSD relies on its controller, firmware, and Flash Translation Layer (FTL) to function properly.
To manage NAND’s behavior, SSDs must withhold a portion of capacity from user access.
Hidden space is required for several critical mechanisms:
• Bad Block Management
All NAND comes with factory bad blocks, and additional ones appear over time.
Reserved blocks ensure that defective areas never hold user data.
• Wear Leveling
NAND cells can only endure a limited number of program/erase cycles.
The firmware needs free blocks to rotate data and prevent premature wear.
• Garbage Collection (GC)
Flash memory must erase before writing, and erasing happens per block.
Extra buffer space allows GC to operate efficiently, especially under load.
• Write Amplification Control
More over-provisioned space directly reduces write amplification, increasing both sustained performance and lifespan.
Consumer SSDs typically reserve 7% to 28% of their capacity.
Enterprise SSDs reserve significantly more.
This internal buffer is invisible to users — yet essential for performance and reliability.
3. Your File System Also Takes Space — Often More Than Expected
Even once you see “930GB,” that number is not fully usable.
Every file system maintains metadata structures, logs, indexes, and allocation tables.
Voorbeelden:
-
NTFS: MFT, journaling, USN change logs
-
ext4: superblocks, inode tables, journals
-
APFS: multi-container architecture, snapshots
Depending on the platform, these structures can take hundreds of MB or several GB.
So the final usable capacity always becomes smaller than the raw partition.
4. The Capacity Gap Reflects Engineering Choices — Not Loss
When we combine:
-
decimal vs binary measurement
-
over-provisioning
-
file system overhead
…the “930GB from 1TB” result falls perfectly within standard engineering expectations.
More importantly, this reserved space directly contributes to:
-
sustained write speeds
-
consistent performance under heavy load
-
lower write amplification
-
longer NAND life
-
better thermal behavior
-
reduced fragmentation
-
higher data integrity
In other words:
Modern SSDs give up a portion of their capacity to protect performance and longevity.
That difference isn’t a drawback — it’s a design choice to help drives behave predictably and reliably across diverse workloads.