Which (consumer category) SSDs have supercaps?

[Bill]

Since I learned about supercapacitors, which are to provide protection
from a power outage, I decided to look for that feature in my next ssd.
I am mostly looking at the ~500gb size and under $350 range. Evidently,
the Intel 320 I am using now has one, but maybe not the Intel 730 (I
couldn't tell).

Thanks,
Bill

 

[Paul]

Since I learned about supercapacitors, which are to provide protection
from a power outage, I decided to look for that feature in my next ssd.
I am mostly looking at the ~500gb size and under $350 range. Evidently,
the Intel 320 I am using now has one, but maybe not the Intel 730 (I
couldn't tell).

Thanks,
Bill

Some review sites, open up the SSD for a look. Maybe you'll get lucky,
and some new kit, the reviewer will mention that there is a supercap
present.

Paul

 

[Bill]

Some review sites, open up the SSD for a look. Maybe you'll get lucky,
and some new kit, the reviewer will mention that there is a supercap
present.

Paul

I have been flip-flopping between various ssd options like a fish. At
least I'm getting a little more discerning. I'm surprised this feature
doesn't appears in more "ssd showdowns/comparisons". At least where I
live, power outages are pretty routine, in season. Without an UPS, it
seems sort of foolish to buy an ssd without the feature.

 

[Paul]

I have been flip-flopping between various ssd options like a fish. At
least I'm getting a little more discerning. I'm surprised this feature
doesn't appears in more "ssd showdowns/comparisons". At least where I
live, power outages are pretty routine, in season. Without an UPS, it
seems sort of foolish to buy an ssd without the feature.

A UPS gives you better coverage for a desktop. You can even cable up the
control cable, the UPS sends a shutdown signal before battery is exhausted,
and in WinXP for example, there is a UPS agent to accept the signal
and shutdown the OS if you aren't there. I've had a UPS on this computer
for at least 11 years, and had to change out the battery after 10 years.
The new battery cost around $60-$70, and the price had dropped over time.

The battery in a laptop, is its own UPS. No need for the Supercap there.

*******

You can see some capacitor examples here. One of them appears to
be 0.09F or 90,000 uF. So they aren't the higher capacity ones.
But in the case of the single, metal jacketed one, I checked and
the capacity doesn't affect the price - the 1.0F cap costs $34
and a 0.09F one costs $34 too. The prices on the various cap types
is all over the place. Presumably one of the requirements, is
high G rating, so some of the ones that look like electrolytic
caps would be less preferred. That's so you can drop the SSD
on the floor and it doesn't immediately shatter.

http://www.anandtech.com/show/4366/computex-2011-sandforce-msata-drives-no-more-supercap

Whether they need a supercap, depends on how many outstanding operations
they can have queued inside. With AHCI, you can use tagged command
queuing. So that represents a small number of outstanding commands.
Some designs seem to have a cache RAM, but it isn't always clear
what they're caching in it. It could be a read-cache for example.
Or it could be used to hold the lookup table for blocks (remapped
for load leveling and for bad block management).

The scary part, would be the internal operations attempted, the
moving of data the drive does, when not being accessed. It's up
to the firmware to do that in a safe way, inside the device. Having
a Supercap and advanced power fail detection, just makes them
lazy.

The capacitor inside the ATX power supply, would be an excellent way
to manage a computer. It has enough energy stored in it, to last
for at least 16 milliseconds under full load. If the ATX power supply
had an advanced power fail signal to deliver to the OS, many of these
exposure cases could have been handled by other levels in the computer.
And a "flush" command could be sent by the OS, to the SSD. But modern
computers lack such a capability (there is no advanced power
fail signal on the main ATX cable).

Paul

 

[Bill]

A UPS gives you better coverage for a desktop. You can even cable up the
control cable, the UPS sends a shutdown signal before battery is
exhausted,
and in WinXP for example, there is a UPS agent to accept the signal
and shutdown the OS if you aren't there. I've had a UPS on this computer
for at least 11 years, and had to change out the battery after 10 years.
The new battery cost around $60-$70, and the price had dropped over time.

The battery in a laptop, is its own UPS. No need for the Supercap there.

*******

You can see some capacitor examples here. One of them appears to
be 0.09F or 90,000 uF. So they aren't the higher capacity ones.
But in the case of the single, metal jacketed one, I checked and
the capacity doesn't affect the price - the 1.0F cap costs $34
and a 0.09F one costs $34 too. The prices on the various cap types
is all over the place. Presumably one of the requirements, is
high G rating, so some of the ones that look like electrolytic
caps would be less preferred. That's so you can drop the SSD
on the floor and it doesn't immediately shatter.

http://www.anandtech.com/show/4366/computex-2011-sandforce-msata-drives-no-more-supercap


Whether they need a supercap, depends on how many outstanding operations
they can have queued inside. With AHCI, you can use tagged command
queuing. So that represents a small number of outstanding commands.
Some designs seem to have a cache RAM, but it isn't always clear
what they're caching in it. It could be a read-cache for example.
Or it could be used to hold the lookup table for blocks (remapped
for load leveling and for bad block management).

The scary part, would be the internal operations attempted, the
moving of data the drive does, when not being accessed. It's up
to the firmware to do that in a safe way, inside the device. Having
a Supercap and advanced power fail detection, just makes them
lazy.

The capacitor inside the ATX power supply, would be an excellent way
to manage a computer. It has enough energy stored in it, to last
for at least 16 milliseconds under full load. If the ATX power supply
had an advanced power fail signal to deliver to the OS, many of these
exposure cases could have been handled by other levels in the computer.
And a "flush" command could be sent by the OS, to the SSD. But modern
computers lack such a capability (there is no advanced power
fail signal on the main ATX cable).

Paul

Your answer shows a lot of insight. Thank you for taking the trouble to
post it.

Maybe the power supply manufacturers would consider that they were
"asking for a headache"
if they implemented the feature you describe. All of the sudden, people
would be trying to hold them liable. ha ha.

Bill

 

[Yousuf Khan]

I have been flip-flopping between various ssd options like a fish. At
least I'm getting a little more discerning. I'm surprised this feature
doesn't appears in more "ssd showdowns/comparisons". At least where I
live, power outages are pretty routine, in season. Without an UPS, it
seems sort of foolish to buy an ssd without the feature.

I don't see much reason to protect an SSD. Its storage medium is
flash-based, so it's not really going to be affected by power surges
that much.

Yousuf Khan

 

[Bill]

I don't see much reason to protect an SSD. Its storage medium is
flash-based, so it's not really going to be affected by power surges
that much.

Yousuf Khan

It might it a write operation is interrupted. "That much" doesn't
really make sense in this context. I've had 2 modems that were killed
by power surges..lol

 

[Yousuf Khan]

It might it a write operation is interrupted.

Modern file systems like NTFS are journal-based, meaning that they have
lots of recovery possibilities. The worst that can happen is that you
might lose a couple of seconds of updates.

"That much" doesn't
really make sense in this context. I've had 2 modems that were killed
by power surges..lol

Modems are of course different, as they are directly connected to
electrical lines, such as phone lines or cable lines. SSD's are behind
the computer's PSU, and as well as whatever surge protectors that you
installed in front of them.

Yousuf Khan

 

[Flasherly]

I don't see much reason to protect an SSD. Its storage medium is
flash-based, so it's not really going to be affected by power surges
that much.

Unless supercaps are simply within technology's "feasibility window"
-- a sort of tenuous road pragmaticism and America's signature
contribution to civilization has hoed.

Priced any new pickups, lately? Seen the extended Japanese families
in America's new showrooms? That bottom window sticker figure that
reads, "opportunism for all elitists the end of the rainbow or bust."

-
An Indian newspaper's counter to American newswire reception of a
$2000/US Indian produced car: "We haven't that problem here [in India]
with obesity."

An Indian court judgment for a continued and effective $1US
prescription, Indian technology reverse engineered, for capital
measures commiserate to a $10,000US medicinal, AMA pharmaceutically
dispensed pill: 'It would be simply be too vast of an understatement
to attempt to encompass what innate and inalienable right are to 5/6th
of population in a country [such as India], illiterate and without a
formally nationalized corporate WWW connectivity plan."

 

[Bill]

Unless supercaps are simply within technology's "feasibility window"
-- a sort of tenuous road pragmaticism and America's signature
contribution to civilization has hoed.

I can't tell whether you agree with Mr. Khan, or disagree with him.

Priced any new pickups, lately? Seen the extended Japanese families
in America's new showrooms? That bottom window sticker figure that
reads, "opportunism for all elitists the end of the rainbow or bust."

-
An Indian newspaper's counter to American newswire reception of a
$2000/US Indian produced car: "We haven't that problem here [in India]
with obesity."

An Indian court judgment for a continued and effective $1US
prescription, Indian technology reverse engineered, for capital
measures commiserate to a $10,000US medicinal, AMA pharmaceutically
dispensed pill: 'It would be simply be too vast of an understatement
to attempt to encompass what innate and inalienable right are to 5/6th
of population in a country [such as India], illiterate and without a
formally nationalized corporate WWW connectivity plan."

 

[Flasherly]

I can't tell whether you agree with Mr. Khan, or disagree with him.

Dialectically, a contingent to premise without warrant if based within
economic means for sole advancement as status quo, and not as much by
means foisted for technological over-engineered builds, often blandly
accepted as a popular [mis]conception and derivative, that new and
improved per force irrevocably connotes what is better.

When you pay, say, $1000-2000US for a remote sensor system to trigger
a low-tire air pressure vehicle alarm -- now nationally mandated by
law for production standards across the automotive industry -- would
you think your best interest is safely served to agree, or, would you
rather disagree on principle that you are of otherwise sound and sane
mind [not to drive the national highways on metal rims without a
tire]?

Are you looking particularly forward to driverless cars controlled
through nationally instituted grids of radio/satellite telemetry for
standardized routes, of fixed consequence and determinate rules by
which a methodology of pure logic best attains that end?

 

[Yousuf Khan]

I can't tell whether you agree with Mr. Khan, or disagree with him.

Trust me, no one really knows if he's talking about the subject at hand,
or he's off on a tangent somewhere.

Yousuf Khan

 

[Paul]

Modern file systems like NTFS are journal-based, meaning that they have
lots of recovery possibilities. The worst that can happen is that you
might lose a couple of seconds of updates.

Imagine the SSD is doing active wear leveling, and the operation
has nothing to do with external commands. Perhaps some
internal state information gets corrupted. Then the SSD refuses
to power on and answer queries, on the next power cycle.

Some of the early SSDs, had a tendency to do that. They would
be working during one session, and refuse to answer any probes
on the next session. And no utility was provided to revive them.

The same thing has happened to hard drives. Drives that were
"here today gone tomorrow", and the event was traced to a firmware
bug. More than one Seagate has suffered from this. At least
one of them, the hard drive could be revived using a special
serial cable.

http://www.msfn.org/board/topic/128807-the-solution-for-seagate-720011-hdds/

Paul

 

[Flasherly]

Trust me, no one really knows if he's talking about the subject at hand,
or he's off on a tangent somewhere.

Matter of correct interpretation -- that if supercaps are to be
affordable incentives to technological advancement among resources,
byproducts to competitively selling modernity, then we've are the
living proof of its facets and means for adaptive purposes.

Computers always cost less to provide for more over a broader time to
average for that cozy benefit.

Because I simply wouldn't lend as much credo to disallow for what
added importance supercaps may have to lend is perfectly fine as
conditional estimation of that worth.

I really couldn't care about power surges, apart from being intimate
with them in my particular location;- nor journaled filesystems and
ancillary power system conditioners, as I haven't either the luxury or
inclination, as the case may be, to indulge them.

There being nothing more tangential that what I simply choose not to
closed my mind to -- should supercap usages be a part of a future that
will cost me nothing.

A difference between calling me cheap shot should I choose to take
freely from may be and store and effectively given, and making them,
when labeling their use irrelevant, which I really should properly
identify for a fallacious understatement at this juncture and
inconclusive of supercap potentials largely not deployed or field
tested for validity from a broader [accepted marketing] sense.

 

[Bill]

Trust me, no one really knows if he's talking about the subject at hand,
or he's off on a tangent somewhere.

Matter of correct interpretation -- that if supercaps are to be
affordable incentives to technological advancement among resources,
byproducts to competitively selling modernity, then we've are the
living proof of its facets and means for adaptive purposes.

Computers always cost less to provide for more over a broader time to
average for that cozy benefit.

Because I simply wouldn't lend as much credo to disallow for what
added importance supercaps may have to lend is perfectly fine as
conditional estimation of that worth.

I really couldn't care about power surges, apart from being intimate
with them in my particular location;- nor journaled filesystems and
ancillary power system conditioners, as I haven't either the luxury or
inclination, as the case may be, to indulge them.

There being nothing more tangential that what I simply choose not to
closed my mind to -- should supercap usages be a part of a future that
will cost me nothing.

A difference between calling me cheap shot should I choose to take
freely from may be and store and effectively given, and making them,
when labeling their use irrelevant, which I really should properly
identify for a fallacious understatement at this juncture and
inconclusive of supercap potentials largely not deployed or field
tested for validity from a broader [accepted marketing] sense.

I may have a small sample size, but Intel ssds haven't let me down yet! : )

 

[Flasherly]

I may have a small sample size, but Intel ssds haven't let me down yet! : )

Of course not. Intel and Samsung are two of the highest regarded SSD
players.

 

[Yousuf Khan]

Of course not. Intel and Samsung are two of the highest regarded SSD
players.

I have one Corsair SSD, and one Adata SSD, neither of them have let me
down either. Though I did have to play with their firmware a bit.

Yousuf Khan

 

[Bill]

I have one Corsair SSD, and one Adata SSD, neither of them have let me
down either. Though I did have to play with their firmware a bit.

I had to update the firmware of the older Intel X25-M when I added an
ssd from the 320 Series. They threatened that I "might lose all of my
data", but I didn't.
I did the backup anyway, of course.

 

[Flasherly]

I have one Corsair SSD, and one Adata SSD, neither of them have let me
down either. Though I did have to play with their firmware a bit.

Believe it was an Adata, I was looking at, used the same controller as
a Crucial MX100 I bought. I do however have two Samsungs, and would
have bought Intel if the opportunity had availed itself. Intel R&D
speaks for itself, as would Samsung's present position for total SSD
sales.

Nothing particularly substantiative to say why with this latest
Crucial SSD, other than shaving off $40 less than a comparable drive
from Samsung, I don't as yet feel comfortable about the Crucial
technically being in the same class as a Samsung.

Perhaps later if I develop a simpler system for a boot arbitrator for
indiscriminately interchanging my present partition structures between
the Samsung and Crucial. I've already had once to rebuild damaged
MBRs from an overzealous assumption of impunity when manipulating
partitions, as well incurring regular streaming errors on the Crucial
with binary sector backups of XP, I never had when using the Samsung
for XP, or binary streams that continue to take well in the Samsung's
present capacity as a Win7 drive.

Maybe. There's certainly a wide buffer zone for maybe's among
particulars and all I don't understand or by force of habit am duly
careful about with computers;- As well all, thankfully, we've moved
along and advanced past, in yesterday's pre-MS platforms, such as
DESQview. I still remember 9-chipped banks, 8 or so rows of banks,
and blowing a single memory module from particularly nasty programs
and crash/reboots that would leave one, sometimes two dead chips, for
hunting them down;- Windows 95 was neither a piece of cake, for a
early generations of MB architecture and 386s. Wasn't until around
NT, early Win98 technology, at least for me, that hardware seemed to
have advanced to a point past a course on par for "dealing [rolling]
with it," in the regular bugworks of commonplace computer hardware.

 

[Yousuf Khan]

Believe it was an Adata, I was looking at, used the same controller as
a Crucial MX100 I bought. I do however have two Samsungs, and would
have bought Intel if the opportunity had availed itself. Intel R&D
speaks for itself, as would Samsung's present position for total SSD
sales.

Well, Intel's latest SSD's are using the standard Sandforce chipset like
almost everybody else.

Yousuf Khan