Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Why Are Higher Capacity SSDs Faster on New MBPs
- Thread starter smithdr
- Start date
- Sort by reaction score
That's how flash storage works, larger capacities and larger chips can move data quicker. Same with the older models too, the 3.3 GB/sec read for the 13" MBP is only on the 2 TB model.
Basically because there are more memory chips on the larger drive, the controller can access more of them at the same time in parallel. This is why you see the faster read and write speeds.
I'll preface this by saying I'm not exactly sure what makes up Apple's SSDs these days but I'm m going to assume that they work pretty much the same as everyone else's and let someone correct me of it's not correct.
the important thing to know is the two most common types of chips that store data for an SSD and that is SLC and TLC.
Basically SLC stores one bit of data per cell and TLC stores three. Using TLC let's you get to higher capacities with less chips (because each cell stores more) but it is slower than SLC.
What most SSD makers do these days is reserve a portion of a TLC drive to behave as though it is SLC. This let's them have high storage capacities whole still retaining the fast speed of an SLC drive.
Because cells that are acting like SLC store less data, each drive has to decide how much they want to dedicate to a SLC cache. For a 256GB drive, it could be 8-12GB. If you run a drive read/write operation that is larger than the SLC cache you will take a performance hit. If you fill the drive up the limit, the cells will start acting like TLC again and you will take a performance hit.
on a 1TB drive, the cache can be up to 100GB. This is usually more than enough for most drive operations but as you fill the disk up the smaller the cache becomes. If you store 500GB maybe your SLC cache is only 60GB now.
This varies upon each manufacturer and I'm just throwing out generic numbers but hopefully it conveys the point.
It doesn't have anything to do with the number of chips or writing in parallel - it's usually is your SLC cache large enough to handle the operations being sent to the drive. Larger drives = larger caches.
the important thing to know is the two most common types of chips that store data for an SSD and that is SLC and TLC.
Basically SLC stores one bit of data per cell and TLC stores three. Using TLC let's you get to higher capacities with less chips (because each cell stores more) but it is slower than SLC.
What most SSD makers do these days is reserve a portion of a TLC drive to behave as though it is SLC. This let's them have high storage capacities whole still retaining the fast speed of an SLC drive.
Because cells that are acting like SLC store less data, each drive has to decide how much they want to dedicate to a SLC cache. For a 256GB drive, it could be 8-12GB. If you run a drive read/write operation that is larger than the SLC cache you will take a performance hit. If you fill the drive up the limit, the cells will start acting like TLC again and you will take a performance hit.
on a 1TB drive, the cache can be up to 100GB. This is usually more than enough for most drive operations but as you fill the disk up the smaller the cache becomes. If you store 500GB maybe your SLC cache is only 60GB now.
This varies upon each manufacturer and I'm just throwing out generic numbers but hopefully it conveys the point.
It doesn't have anything to do with the number of chips or writing in parallel - it's usually is your SLC cache large enough to handle the operations being sent to the drive. Larger drives = larger caches.
Denser too. But the bandwidth tops out with the 4TB version. That's when the PCIe bus becomes saturated.
Yeah, one of the reasons I picked 1TB over the lesser 512GB. 2TB is too expensive on the other hand.