Mushkin uses a stacked daughter board to achieve 480GB. This is usually ok in a motherboard, but not in a z-height limited mobile device. While it's still electrically mSATA, it is technically outside the mSATA specifications which limit z-height (if I recall correctly).
Yup, the 768GB SSD in rMBP has twelve 64GB (8x8GB) packages.
However, SandForce has much stricter restrictions, the SF-2281/2 can only access up to 64 NAND dies (that's up to 512GB with 8GB dies). Samsung could build a 1TB drive if they wanted to, they just don't see the market for it (yet).
On a slightly related note, the iPad (late 2012) 16 GB, has a single NAND chip, as could be seen in the teardowns, and there doesn't seem to be any room for more. Do you perhaps know, or care to speculate, how the higher capacity ones are configured, especially the new 128 GB version?
The 128GB model is most likely using new 128Gb (16GB) NAND dies to enable the higher capacity. You can only stack up to eight dies in a single package, so a higher capacity die was needed before you could go over 64GB (8x 16GB is 128GB).
At least Samsung and IMFT have 128Gb dies in production but they most likely weren't available in volume when the iPad 4 was initially launched, hence the delay.
I was hoping for a comparison with the m4/C400 mSATA 256GB, since it seems to differ a bit in performance from the 128GB model, and would be better suited as a "direct competitor" to the 240GB 525 for the same of comparison IMHO.
Too bad you didn't include it in the charts, Anand.
Since IOPS consistency improves significantly when setting aside 25% spare area, what is the practical effect in real world? Has this been documented using the AT Storage Bench? Under default conditions, the 840 Pro dominates the top of the charts, but does it still retain the crown after being "stroked"? Just curious...
The prospect of the S3700 technology in a consumer drive has appeal except that the S3700 uses too much power. Is Intel's approach inherently inefficient or is it reasonable that Intel can tune the technology differently for the consumer market to achieve competitive efficiency?
" Why does Intel continue to use a third party SATA controller in many of its flagship drives? Although I once believed this was an issue with Intel having issues on the design front, I now believe that a big part of it has to do with the Intel SSD group being more resource constrained than other groups within the company. "
This seems strangely short sighted. How is flash controlled on mobile devices? Obviously performance is substantially lower. It's not clear to me how that lowering is split between - cheaper, lower-end flash - only one rank (or whatever flash their call their equivalent), ie limited parallelism - a dumb controller.
However there doesn't seem to be any aspect of the problem that is inherently power limited. Which implies that if Intel wanted a way to make their perpetually lagging Atom SOCs a little more noteworthy, one way to do so would be to work on them having flash support that was substantially faster than what's available on ARM today.
Sandforce controllers are high performance, have no DRAM need, and allow both SF standard and custom firmware. Until SF drops the ball on performance or support I would look for more SSDs to be based on the SF design, not less. Enterprise is different ball game where ASPs/margins are much higher so custom controller might make sense (volumes are much lower though). If other 3rd party controllers mature, I expect them to gain market share as well.
We’ve updated our terms. By continuing to use the site and/or by logging into your account, you agree to the Site’s updated Terms of Use and Privacy Policy.
21 Comments
Back to Article
vcorem - Wednesday, January 30, 2013 - link
http://m.tomshardware.com/news/Mushkin-480GB-mSATA...IanCutress - Wednesday, January 30, 2013 - link
Mushkin uses a stacked daughter board to achieve 480GB. This is usually ok in a motherboard, but not in a z-height limited mobile device. While it's still electrically mSATA, it is technically outside the mSATA specifications which limit z-height (if I recall correctly).lukarak - Wednesday, January 30, 2013 - link
What is the chip density in the 768 GB SSD board in a rMBP? Does it have 12?Kristian Vättö - Wednesday, January 30, 2013 - link
Yup, the 768GB SSD in rMBP has twelve 64GB (8x8GB) packages.However, SandForce has much stricter restrictions, the SF-2281/2 can only access up to 64 NAND dies (that's up to 512GB with 8GB dies). Samsung could build a 1TB drive if they wanted to, they just don't see the market for it (yet).
lukarak - Wednesday, January 30, 2013 - link
Thanks for the answer.On a slightly related note, the iPad (late 2012) 16 GB, has a single NAND chip, as could be seen in the teardowns, and there doesn't seem to be any room for more. Do you perhaps know, or care to speculate, how the higher capacity ones are configured, especially the new 128 GB version?
Kristian Vättö - Wednesday, January 30, 2013 - link
The 128GB model is most likely using new 128Gb (16GB) NAND dies to enable the higher capacity. You can only stack up to eight dies in a single package, so a higher capacity die was needed before you could go over 64GB (8x 16GB is 128GB).At least Samsung and IMFT have 128Gb dies in production but they most likely weren't available in volume when the iPad 4 was initially launched, hence the delay.
lukarak - Wednesday, January 30, 2013 - link
Makes sense, thanks for the insight.SAMSAMHA - Wednesday, January 30, 2013 - link
hi, AnandI am curious what board are you using to test this since none of the desktop board with mSATA that I know has SATA 6Gbps interface.
SodaAnt - Wednesday, January 30, 2013 - link
There are msata to sata conversion boards that work fine because msata and sata are electrically the same.Meaker10 - Wednesday, January 30, 2013 - link
The msi gt60 can ship with a pair of msata gen3 slots configurable in raid.philipma1957 - Wednesday, January 30, 2013 - link
meaker10 have you used a msi gt60?critical_ - Wednesday, January 30, 2013 - link
I'm using Startech's SAT2MSAT25. It is a "passthrough" design so it'll work at 6Gbps is your controller supports it.HyperJuni - Wednesday, January 30, 2013 - link
I was hoping for a comparison with the m4/C400 mSATA 256GB, since it seems to differ a bit in performance from the 128GB model, and would be better suited as a "direct competitor" to the 240GB 525 for the same of comparison IMHO.Too bad you didn't include it in the charts, Anand.
nathanddrews - Wednesday, January 30, 2013 - link
Since IOPS consistency improves significantly when setting aside 25% spare area, what is the practical effect in real world? Has this been documented using the AT Storage Bench? Under default conditions, the 840 Pro dominates the top of the charts, but does it still retain the crown after being "stroked"? Just curious...SanX - Wednesday, January 30, 2013 - link
Is it burning hot for its size? Will it fry your eggs?SanX - Wednesday, January 30, 2013 - link
damn cellphone spell-correction typo and lack of edit option like on cheap websites: ***incompressible***dealcorn - Wednesday, January 30, 2013 - link
The prospect of the S3700 technology in a consumer drive has appeal except that the S3700 uses too much power. Is Intel's approach inherently inefficient or is it reasonable that Intel can tune the technology differently for the consumer market to achieve competitive efficiency?name99 - Wednesday, January 30, 2013 - link
"Why does Intel continue to use a third party SATA controller in many of its flagship drives? Although I once believed this was an issue with Intel having issues on the design front, I now believe that a big part of it has to do with the Intel SSD group being more resource constrained than other groups within the company.
"
This seems strangely short sighted. How is flash controlled on mobile devices?
Obviously performance is substantially lower. It's not clear to me how that lowering is split between
- cheaper, lower-end flash
- only one rank (or whatever flash their call their equivalent), ie limited parallelism
- a dumb controller.
However there doesn't seem to be any aspect of the problem that is inherently power limited.
Which implies that if Intel wanted a way to make their perpetually lagging Atom SOCs a little more noteworthy, one way to do so would be to work on them having flash support that was substantially faster than what's available on ARM today.
emvonline - Thursday, January 31, 2013 - link
Sandforce controllers are high performance, have no DRAM need, and allow both SF standard and custom firmware. Until SF drops the ball on performance or support I would look for more SSDs to be based on the SF design, not less. Enterprise is different ball game where ASPs/margins are much higher so custom controller might make sense (volumes are much lower though). If other 3rd party controllers mature, I expect them to gain market share as well.damnintel - Wednesday, March 13, 2013 - link
heyyyy check this out damnintel dot comdamnintel - Wednesday, March 13, 2013 - link
heyyyy check this out damnintel dot com