POST A COMMENT

69 Comments

Back to Article

  • Dug - Monday, October 29, 2012 - link

    If Samsung continues to produce the 830 as a value drive compared to the 840, I think Intel will have trouble selling these. I don't see any advantage. Reply
  • Kristian Vättö - Monday, October 29, 2012 - link

    Samsung is phasing out the SSD 830 by the end of November, so it won't be available much longer. Reply
  • mfenn - Monday, October 29, 2012 - link

    Nooooooooo....

    I want my cheap fast drives damnit!
    Reply
  • DanNeely - Monday, October 29, 2012 - link

    They're only cheap now to get rid of them; once they're out of the channel people can't point to the old model discount for why the new one is too expensive.

    If you want a bigger SSD, now's the time to buy the 830 you've been lusting over.
    Reply
  • BrianDustin - Tuesday, October 30, 2012 - link

    Love my job, since I've been bringing in $5600… I sit at home, music playing while I work in front of my new iMac that I got now that I'm making it online(Click on menu Home)
    http://goo.gl/ntYAX
    Reply
  • JohnUSA - Tuesday, October 30, 2012 - link

    You can shove your iMac in your rear where the sun does not shine. Reply
  • jwilliams4200 - Thursday, November 01, 2012 - link

    It is too bad the comments software does not have a "report spam" button. Reply
  • JonnyDough - Friday, November 23, 2012 - link

    AGREED. :( MODERATOR HELP! Reply
  • hrga - Thursday, November 01, 2012 - link

    @BrianDustin SPAMMER Reply
  • Wolfpup - Monday, October 29, 2012 - link

    Samsung's drivers are flaky if hit hard, and of course less reliable. These Intel drives with Sandforce controllers aren't as good as Intel controllers, but still...Micron/Crucial and Intel are the best choices. Reply
  • Bull Dog - Monday, October 29, 2012 - link

    Do you have any hard data to back up this these assertions? Or are they just your unfounded opinions?

    I ask because the SSD 830 has a terrific reputation for being reliable. It has even demonstrated an exceptionally impressive lifespan over at XtremeSystems Forums.
    Reply
  • kkwst2 - Monday, October 29, 2012 - link

    Yeah, where's your data on that? I'm not aware of any data that suggests the 830 is less reliable in any way than Crucial or Intel. And my perception from experience and reviews are that Samsung has a significantly better reliability record than Sandforce. All vendors have had firmware issues to some degree, and Intel has had some pretty significant ones. The 8 MB issue on the 320 that caused data loss comes to mind. Reply
  • Samus - Monday, October 29, 2012 - link

    Honestly the first Intel SSD I've come across dead was an SSD520 120GB, just wont detect by the system, clearly a Sandforce controller problem.

    I'd never seen an Intel SSD fail until now that (August 2012) and Intel has made a big mistake joining the likes of OCZ quality with its Sandforce marriage.
    Reply
  • JonnyDough - Friday, November 23, 2012 - link

    You must forget about all the issues that plague hard drives. Reply
  • MichaelD - Tuesday, October 30, 2012 - link

    To what "Samsung drivers" are you referring? You don't need to load any drivers for this or any other SSD (at least in Windows 7, 8 and Server 2012). The driver is provided by Microsoft and has a date of 2006.

    And I'll note that Samsung SSDs are not plagued by the infamous "Sandforce controller bug." And that Samsung makes the controller the NAND and everything else in the SSD.

    So how again, is it less reliable than other SSDs?
    Reply
  • centosfan - Wednesday, October 31, 2012 - link

    I don't think you know what you are talking about. I have two Samsung 830's and use them everyday and push they quite hard. Haven't had 1 single issue. I haven't heard of any problems with them either. Reply
  • hrga - Thursday, November 01, 2012 - link

    It all depends about the market Samsung 830 128GB is here 170USD while 256GB version is 400USD so they aint affordable at all. While Intel 520 series 240GB costs 350USD and 120GB 180USD respectively Reply
  • seapeople - Friday, November 02, 2012 - link

    So $170 and 128GB is less affordable than $180 and 120GB? You live in strange world. Reply
  • djshortsleeve - Monday, October 29, 2012 - link

    I dont see the need for all these various models. Make a value drive and a high end one. You either buy cheapest or best. Reply
  • Kristian Vättö - Monday, October 29, 2012 - link

    That's essentially what Intel offers. SSD 335 is the value drive, whereas the successor of SSD 520 will be the high-end one. SSD 335 and 330 are basically the same and 335 will replace the 330 sooner than later (I'm hearing Q1'13 for the other capacities). Reply
  • MichaelD - Tuesday, October 30, 2012 - link

    I agree! Take Corsair for example; they've got like a hundred (sic) different SKU's. "Force" "Subforce" "Battle" "Skeedaddle" and "Primo" versions of SSDs...marketing FUD at it's finest. Only the .5% of SSD buyers (like AT readers) will actually look at specs and decide. The other 99.5% will just buy whatever box has the "fastest looking" cover art. Reply
  • zanon - Monday, October 29, 2012 - link

    Now they just get giggle-stomped across the board by Samsung. I hope Intel decides to be competitive again someday, but in the mean time it's hard to see any reason to bother. Reply
  • MadMan007 - Monday, October 29, 2012 - link

    Read past the sythetics to real-world tests. It is at least competitive in most cases, and all these drives are stupid fast anyway. Reply
  • jeffbui - Monday, October 29, 2012 - link

    Why is there such a discrepancy between the power consumption figures given by the manufacturer vs what you're getting from your testing? (Samsung mostly) Other websites are getting completely different power usage figures as well. Reply
  • DanNeely - Monday, October 29, 2012 - link

    Comments on one of AT's other recent SSD articles claimed this is because the power consumption test is being done using an external enclosure that never lets the drive drop into it's lowest power states. I didn't see any official comment on it. Reply
  • Kristian Vättö - Monday, October 29, 2012 - link

    Some manufacturers such as Samsung report their power numbers with DIPM/HIPM (Device/Host Initiated Link Power Management) enabled, which can lower the power consumption significantly. DIPM/HIPM are not enabled on desktop by default and I'm not sure if all laptops have them enabled either.

    We have ran tests with DIPM/HIPM enabled and gotten results similar to what manufacturers report, but so far we have kept on publishing numbers with DIPM/HIPM disabled. We will probably add DIPM/HIPM numbers once we redo our SSD testing methodology.
    Reply
  • DanNeely - Monday, October 29, 2012 - link

    If available, would the feature be called DIPM/HIPM in our bios's; or is it likely to be obfuscated to something else?

    Also, why is it often disabled by default? Is there a penalty related to enabling it?
    Reply
  • Kristian Vättö - Monday, October 29, 2012 - link

    Here are instructions for enabling DIPM/HIPM:

    http://www.sevenforums.com/tutorials/177819-ahci-l...

    In desktops it's not as important because you aren't running off of a battery and the power that SSDs/HDs use is so little anyway that it won't affect your power bill. I'm not sure why it's disabled, though, because I havent heard of any concrete issues caused by it.
    Reply
  • Per Hansson - Monday, October 29, 2012 - link

    Actually I'm not sure that DIPM/HIPM is the whole reason.
    I mentioned it previously and it for sure can have a dramatic difference.

    But just as important is the measuring equipment used for the power consumption.
    A cheap DMM only measures in very slow intervals, you would get very different results when measuring using a $100 Fluke vs a $1000 Fluke vs a Scope with really high bandwidth.
    It's because the SSD changes power levels many hundred times per second and this is too fast for a regular DMM so you just get some of the data points, not enough to make a reliable averge...
    Reply
  • MrSpadge - Monday, October 29, 2012 - link

    Wouldn't the slower / cheaper DMM measure over longer intervals (that's why it's slow in the first place) and hence automatically average over some fluctuation? Reply
  • Per Hansson - Tuesday, October 30, 2012 - link

    No, it does not work like that.
    A slow DMM might take a reading every second.
    An example, in seconds:
    1: 2w
    2: 2w
    3: 2w
    Average=2w

    A fast DMM might take readings every 100ms:
    1: 2w
    2: 0.5w
    3: 2w
    4: 0.5w
    Average=1w

    As you see a DMM does not take a continous reading, it takes readings at points in time and averages those...

    An SSD drive might actually change power levels much more frequently, like every millisencond (consider their performance, how long does it take to write 4KB of data as an example?)
    Reply
  • hrga - Thursday, November 01, 2012 - link

    dont think SSD even try to write such a small amount of data as 4kB every milisecond considering how large buffers usually have 128GB LPDDR2. So thes kind of small writes occur in bursts when they accumulate every 15-30s (at least hope so as this was case with hard drives) That ofc depends on firmware and values in it. Reply
  • Per Hansson - Thursday, November 01, 2012 - link

    That makes no difference, I sincerely hope that no drive waits 15 > 30 seconds to write data to disk because that is just a recipe for data loss in case of power failure or BSOD.
    I also hope no drive uses a 128GB write cache. (Intel's in house controller keeps no user data in cache as an example, but I digress)

    Even if the drive waits a minute before it writes the 4KB of data you must still have a DMM capable of catching that write, which is completed in less than a millisecond.
    Otherwise the increased power consumption during the disk write will be completely missed by the DMM
    Reply
  • Mr Alpha - Monday, October 29, 2012 - link

    Wouldn't it make more sense to the idle power consumption on a platform that supports DPIM? The idle power usage is mostly a matter on mobile devices, and it is on those you get DPIM support. Reply
  • sheh - Monday, October 29, 2012 - link

    The text says total writes were 1.2TB, (+3.8TB=) 5TB, and 37.8TB. The screenshots show "host writes" at 1.51TB, 2.11TB, and 3.90TB? Reply
  • sheh - Monday, October 29, 2012 - link

    And why the odd power on hours counts? Reply
  • Kristian Vättö - Monday, October 29, 2012 - link

    You are mixing host writes with the actual NAND writes. Host writes are the data that the host (e.g. an operating system) sends to the SSD controller to write. NAND writes show much is written to the NAND.

    When the SSD is pushed to a corner like I did, you will end up having more NAND writes than host writes because of read-modify-write (i.e. all user-accessible LBAs are already full, so the controller must read the block to a cache, modify the data and rewrite the block). Basically, your host may be telling the controller to write 4KB but the controller ends up writing 2048MB (that's the block size).
    Reply
  • extide - Monday, October 29, 2012 - link

    Block size is 2048KB* Reply
  • sheh - Monday, October 29, 2012 - link

    So the write amplification in the end was x9.7?

    Are NAND writes also reported by SMART?

    And with the messed up power on count, how can you know the rest of the SMART data is reliable?
    Reply
  • Kristian Vättö - Tuesday, October 30, 2012 - link

    Yes, write amplification was around 9.7x in the end. That makes sense because the drive becomes more and more fragmented the more you write to it.

    As you can see in the screenshots, the SMAT value F9 corresponds to NAND writes. Most manufacturers don't report this data, though.

    We just have to assume that the values are correct. Otherwise we could doubt every single test result we get, which would make reviewing impossible. The data makes sense so at least it's not screaming that something is off, and from what I have read, we aren't the only site who noticed weird endurance behavior.
    Reply
  • sheh - Tuesday, October 30, 2012 - link

    Thanks for the clarification. Reply
  • jwilliams4200 - Thursday, November 01, 2012 - link

    I think the most accurate figure is the write amplification during the 4KiB QD32 test.

    In your first table, E1 = 1.51 TiB (Intel seems to have the same bug as Windows, labeling TB when they mean TiB)

    And F9 = 1208 GiB (I am assuming it is GiB not GB, since Intel usually seems to use the binary power units)

    Then in the last table, E1 = 3.90 TiB and F9 = 37791 GiB.

    Then WA = (37791 - 1208) / (3.90 - 1.51) / 1024 = 14.95

    So WA is about 15 for the QD32 4KiB random writes.
    Reply
  • jwilliams4200 - Monday, October 29, 2012 - link

    What is the exact wording from Intel for the $184 price?

    The article says MSRP, but in the past, Intel has often quoted their distributor price for 1000 units. In other words, how much newegg would pay Intel if they ordered 1000 units.

    So I am wondering whether $184 is really the MSRP, or whether it is the 1000 unit price (in which case the street price would be higher than $184)
    Reply
  • Kristian Vättö - Monday, October 29, 2012 - link

    Intel said the $184 is the RCP/MSRP (Recommended Customer Price/Manufacturer's Suggested Retail Price). Reply
  • jwilliams4200 - Monday, October 29, 2012 - link

    Thank you for the clarification. So apparently newegg is selling for significantly more than the RCP (currently $210). Reply
  • Kristian Vättö - Monday, October 29, 2012 - link

    Thanks for the heads up, I updated the article with the NewEgg pricing info. I guess this once again proves that MSRPs are totally meaningless. Reply
  • MadMan007 - Monday, October 29, 2012 - link

    Or it proves that Newegg prices a bit high on newly launched tech. Maybe Newegg shouldn't be used as the sole basis for pricing any more. Reply
  • jwilliams4200 - Monday, October 29, 2012 - link

    No, it is not just newegg. The cheapest price (from a major retailer) that I see for the 335 now is $195 from buy.com. Most of the others want more than $200 including shipping. Reply
  • meloz - Monday, October 29, 2012 - link

    SSDs continue to be a minefield for consumers. It is pathetic that this SSD wore out as quickly as it did. Such a fundamental issue should have been caught in Intel's internal testing before the drive was dispatched for reviewing, specially when the drive started with MWI of 92. No one at Intel thought this was odd?

    Looks like we will have to wait another 18 months for SSDs to become truly reliable. But I thought the same 18 months ago. Ugh. Only the Samsung 840 inspires some sort of confidence.

    I shudder to think how these manufacturers will cope with newer NAND in future which will have even lesser write-erase cycle life.

    Reading about all these laughable "oops" by SSD manufacturers makes you *really* appreciate the job Segate, WD and Samsung do with their platter drives.
    Reply
  • jeffrey - Monday, October 29, 2012 - link

    After Anandtech's 840 Pro review sample died, I'm not super inspired with confidence about the plain 840. Reply
  • meloz - Tuesday, October 30, 2012 - link

    >After Anandtech's 840 Pro review sample died, I'm not super inspired with confidence about the plain 840.

    Oops, I had almost forgotten about that, thanks for the reminder.

    Damn, all SSD manufacturers suck in their own ways.
    Reply
  • jwilliams4200 - Thursday, November 01, 2012 - link

    I'm disappointed that there hasn't been any more information on that 840 Pro that died.

    Anand should really post some more details. Like what it was doing just before it died, the symptoms of how it failed, whether the SMART parameters could still be read, etc.

    Also, Anand should be hounding Samsung to get back to him about it, if they haven't already. The 840 Pro is apparently shipping on Nov 6. If Samsung has not been able to diagnose the problem and report back by then, it looks bad for Samsung.
    Reply
  • Kristian Vättö - Thursday, November 01, 2012 - link

    Anand was filling the drive with sequential data (preconditioning it for our enterprise tests) and it just died in the middle of the run. After that it was no longer recognized in BIOS, not even when connected using USB to SATA adapter.

    As far as I know, Samsung has not gotten back to us about it yet but let me ask Anand and see if he knows more.
    Reply
  • jwilliams4200 - Thursday, November 01, 2012 - link

    Thank you for the additional information.

    I am highly interested in what Samsung has to say about the failure. It seems to me that anandtech should be able to put some pressure on Samsung to give them a thorough failure analysis in a timely manner, or else anandtech will report that Samsung was unable to explain the failure and that looks bad for Samsung.
    Reply
  • mmonnin03 - Tuesday, October 30, 2012 - link

    The flash on SSDs arent going to get more reliable. ECC basically scales exponentially as the process dimensions keep shrinking. As the lines get closer and closer, the number of electrons holding the charge becomes harder and harder to measure. Each cell is 2 bits, so 4 different amounts of electrons need to be measured. Errors occur more frequently and get fixed. And that's not going to make the NAND any faster by going smaller. SSD speed/reliability improvements will/have come at the controller level. If you truly want a reliable SSD, go 34nm SLC. Its still being produced. Reply
  • jeffrey - Monday, October 29, 2012 - link

    Between the Samsung 840 Pro dying during testing and now the endurance issue testing Intel's 335, I believe OCZ should get some more credit for their Vertex 4.

    Everyone wants to mention Intel, Micron/Crucial, and Samsung for SSDs, but the Vertex 4 deserves to be there too.
    Reply
  • josephjpeters - Monday, October 29, 2012 - link

    Endurance will play a larger role in differentiating future SSD's as the industry continues to move to smaller NAND geometries.

    I'm interested in seeing OCZ's Vector which will use 20nm MLC NAND. It'll be a big test for OCZ to see how their endurance technology stacks up against the competition (or lack thereof in the consumer space).
    Reply
  • Samus - Monday, October 29, 2012 - link

    I've had my Intel SSD320 160GB for about two years, use it daily, and still have only written 10.6TB to it and the MWI is still 100%.

    Either way you look at it, this drive will last the average user a decade easily, even with less than 1000 P/E
    Reply
  • IntelUser2000 - Tuesday, October 30, 2012 - link

    This is false: "Based on the data I gathered, the MWI would hit 0 after around 250TB of NAND writes, which translates to less than 1,000 P/E cycles."

    There is a forum that exclusively tests SSD endurance and Intel drives last far past the MWI of 0. In fact, after it reached zero, it started counting up. I remember the original X25-M lasting until the second MWI is significantly greater than 25(it could be 50 I don't remember).

    They thought that after the Media Wearout Indicator reached 0, the drive would die. In fact, none of the drives did. NONE.

    Even 240TB is hell of a lot. My X25-M has 7.6TB written to it and I had it since the year the drive was announced. At this rate, I'll be 30 years older by the time it reaches that point. So its a needless worry about nothing.

    Contrary to Platter HDDs, which die off slowly and more and more data gets corrupted and gets slower and slower until you notice that the drive is dying. Less than 5 years for lot of people around me too.
    Reply
  • Kristian Vättö - Tuesday, October 30, 2012 - link

    I did not say the drive will die after the MWI hits 1. In fact, I said the opposite:

    "Even after it has hit 1, it's likely the drive can still withstand additional write/erase cycles thanks to MLC NAND typically behaving better than the worst-case estimates."

    The problem here isn't that 1,000 P/E cycles isn't enough for a consumer, but the fact that there seems to be a huge difference in endurance between 20nm MLC and 25nm MLC if our data is correct. Intel claimed that there is no difference, both are 3,000 P/E cycles, but our data contradicts with theirs. Given that the SSD 335 doesn't bring any immediate price cuts, you are getting a worse product for the same money compared to the SSD 330.

    It's of course possible that there is a simple firmware bug which reports wrong MWI or NAND writes, but at least so far Intel has not said anything to suggest that.
    Reply
  • MrSpadge - Tuesday, October 30, 2012 - link

    "It is always impressive to think about performance going up with subsequent NAND generations"

    Well.. it might be impressive, but nevertheless it's a false assumption (if NAND generations are defined by lithography - newer interfaces could be implemented on the same lithography just as well).
    Reply
  • Kristian Vättö - Tuesday, October 30, 2012 - link

    My point was that overall performance has gone up even though the NAND is slower. Reply
  • mmonnin03 - Tuesday, October 30, 2012 - link

    As for the comment about stocks of 25nm flash... IMFT/Micron are still making a significant amount of 25nm NAND as well as some 34nm chips. Reply
  • CristianM - Wednesday, October 31, 2012 - link

    you wrote on average 1000 times the capacity of the disc but that does not translate in write cycles because there is the nasty write amplification which seems to be about 3 for your testcase, so there you have it. But for an average user writing 10GB per day it would take many years to wear (68) or about 7 years for writing 100G which seems unlikely for a home user Reply
  • CristianM - Wednesday, October 31, 2012 - link

    My bad, the write amplification is taken into account, so just divide by 10 :D and the lifespan drops to 7 years for medium use and 1 year for a mad torrent user :)) Reply
  • bankster66 - Friday, November 02, 2012 - link

    Trim needs:
    series 7 chipset (Z77 Z78) - NOT Z68
    11.5+ Intel RST OROM in bios
    11.6+ Intel RST driver pak
    Preferably Win 8
    "RAID falls under the T10 specifications. These specs require the UNMAP command to be issued to TRIM the drive. No current Microsoft operating systems supports the UNMAP command so regardless of driver or OROM RAID arrays will not be Trimmed. Win8 will be the first MS operating system with an API in place to support the UNMAP command "

    There is no way in hell you can get trim functioning with your setup, all prereqs are wrong
    Reply
  • Per Hansson - Saturday, November 03, 2012 - link

    I don't know what you are on to make an accusation like that.
    TRIM is supported by Windows 7
    It requires the AHCI drivers in Windows to be used, or any other driver that specifically mentions that it supports TRIM.

    What you are referencing is TRIM support with 2x harddrives in a RAID-0 array, and that has a long list of requirements, but how that is relevant to this Intel 335 single drive review is beyond me.
    http://www.anandtech.com/show/6161/intel-brings-tr...
    Reply
  • Bruno2 - Tuesday, March 12, 2013 - link

    I noticed the following firmware update for the SSD Series 335. Perhaps this was the reason for the endurance test problems.

    November 2012
    3.1.2
    This release adds a firmware update for the Intel® SSD 335 Series fixing the following issue :

    - Incorrect reporting of the E9h media wear indicator value
    Reply
  • Bruno2 - Tuesday, March 12, 2013 - link

    FYI, I noticed the following firmware update for the SSD Series 335.

    November 2012
    3.1.2
    This release adds a firmware update for the Intel® SSD 335 Series fixing the following issue :

    - Incorrect reporting of the E9h media wear indicator value
    Reply

Log in

Don't have an account? Sign up now