POST A COMMENT

32 Comments

Back to Article

  • jwilliams4200 - Saturday, February 25, 2012 - link

    This article is misleading.

    Most Sandforce drives have flash set aside for both overprovisioning (OP) and redundant array of independent silicon elements (RAISE).

    For example, take a SF2281 SSD with 128GiB of flash on the circuit board (usually advertised as "120GB SSD"). Typically, 8GiB of the flash is used for RAISE. That leaves 120GiB of flash for storage and OP.

    Since the SSD is advertised as "120GB", it must be capable of storing at least 120,000,000,000 bytes. Since there is still 120GiB = 120*(1024^3) = 128,849,018,880 bytes of flash left after subtracting 8GiB for RAISE, that leaves 128,849,018,880 - 120,000,000,000 = 8,849,018,880 bytes available for OP.

    How much OP is that? 128,849,018,880 / 120,000,000,000 = (1024^3) / (1000^3) = 7.37%

    In other words, most "120GB" Sandforce SSDs have both 7.37% OP and 8GiB of flash used for RAISE.

    But this Adata drive advertises 128GB of storage capacity. If that is true (and not just a misrepresentation like OCZ did last year), then Adata must have dropped either OP or RAISE.

    I find it very unlikely that they would drop all overprovisioning, since that would seriously hurt performance and longevity.

    So, most likely what they have done is dropped RAISE. That is not especially surprising, since Intel also dropped RAISE in their 520 SSD.

    But this misleading article makes no mention of RAISE at all. And cites a source, thessdreview, which gets this completely wrong, talking about "0% OP".

    I am disappointed in the quality of the journalism here from anandtech. I would have expected better.
    Reply
  • jwilliams4200 - Saturday, February 25, 2012 - link

    By the way, for those who are not familiar with RAISE, it is similar to RAID, in that a parity block is set aside for each X number of data blocks.

    For example, with a typical Sandforce SSD with 128GiB of flash on board, 8GiB of flash will be devoted to RAISE. That means that there are 15 data blocks and 1 parity block (similar to a RAID 5 with 15 data disks and 1 parity disk).

    The reason this article is misleading is because the overprovisioning (OP) and the RAISE in a typical "120GB" Sandforce SSD are each given approximately 8GiB of flash (to be more accurate, 8GiB for RAISE, 8.24GiB for OP).

    In the case of Adata, they have increased the usable storage capacity from 120GB to 128GB, meaning they must have dropped either RAISE or OP to get that extra ~8GB usable.

    Since it is extremely unlikely that they dropped the overprovisioning, the conclusion is that they dropped the RAISE. This makes sense, because RAISE only functions to increase the reliability of the SSD if a large block of flash dies, so dropping RAISE only decreases the reliability of the SSD. Eliminating overprovisioning, in contrast, would drastically reduce both the performance and the longevity of the SSD, and no rational designer would do such a thing.

    It may be that the author of this article has confused overprovisioning with RAISE, and that is why he has written such a misleading article.

    Bottom line is that the Adata SSDs probably have the same overprovisioning as other SSDs (about 7%) but that they have dropped RAISE and so are not protected from failure when a large chunk of flash memory fails.
    Reply
  • JarredWalton - Saturday, February 25, 2012 - link

    Or ADATA may have decreased RAISE and OP to ~4% instead of 7-8%. They haven't exactly stated what they are doing, merely that their drives are now 128GB instead of 120GB. Reply
  • jwilliams4200 - Saturday, February 25, 2012 - link

    No, that is not supported by the SF2281 controller. To use RAISE, you must devote at least one whole die to RAISE, which would be 8GiB in most cases. Reply
  • JarredWalton - Saturday, February 25, 2012 - link

    True. I suppose they could either eliminate RAISE or eliminate OP, and I'm with you in guessing they sacrificed potential reliability (RAISE) rather than OP. If they kept RAISE and ditched OP, they'd actually have a 129GB (128.8GB) SSD. :p Reply
  • Kristian Vättö - Saturday, February 25, 2012 - link

    Intel 520 Series is still sold as a 120GB drive, though, and has the same usable capacity as OCZ Vertex 3 (which has RAISE enabled). Shouldn't Intel be a 128GB drive then, or does Intel use more space for OP? This isn't making much sense now.

    To be honest, I had no reason doubt TheSSDReview's explanation because it was logical, especially because they had been in contact with LSI (aka SandForce). If it had been their own analysis, then I would have taken it with a grain of salt and most likely confirm it before posting this. I think they are correct because disabling RAISE didn't turn Intel 520 Series into a 128GB drive, so "0% OP" would explain that.

    I have emailed LSI and ADATA to ask for some clarification but it may take a few days now that it's weekend. We'll see...
    Reply
  • jwilliams4200 - Saturday, February 25, 2012 - link

    I don't know for certain that Intel disabled RAISE on the 520, but there is one persuasive piece of evidence that makes me think that they did:

    The UBER spec on the Intel 520 is 10^-16.

    http://www.intel.com/content/www/us/en/solid-state...

    However, Sandforce specifies that RAISE will improve the UBER up to 10^-29

    http://www.sandforce.com/index.php?id=174&pare...

    I think Intel disabled RAISE on the 520 and increased the overprovisioning in order to increase performance and longevity. If I am correct, the 120GB Intel 520 uses about 14% OP and no RAISE.
    Reply
  • Kristian Vättö - Sunday, February 26, 2012 - link

    Tom's Hardware is also saying that RAISE is disabled on 520 Series

    http://www.tomshardware.com/reviews/ssd-520-sandfo...

    I also asked this on the email so hopefully we will get a clarification.

    I still think there must be something with the OP as well. Most 60GB drives have RAISE disabled because they consist of only eight 8GiB dies. Okay, there is just enough space to market the drive as 60GB if 8GiB is used for RAISE but it doesn't sound right that only ~130MB would be used for OP, especially because we are talking about the old firmware which requires ~7% OP. Hence I think something must have been done to the OP in order for ADATA to achieve a usable capacity of 64GB.
    Reply
  • Kamil_FPC - Saturday, February 25, 2012 - link

    Intel didn't dropped RAISE, why do you think they do this? SSD 520 is a standard SandForce based drive, with ~14% of OP, some of it is probably used to support RAISE.

    I don't know the details behind AData drives and new LSI SandForce firmware, but I suspect it still supports RAISE since it is one of core function in this controller, but they are not using the whole NAND, they don't have to. The difference betwen GB and GiB is enough for OP and RAISE.

    We will know for sure when Anand put his hands on AData products :)
    Reply
  • jwilliams4200 - Saturday, February 25, 2012 - link

    The SF2281 controller has some restrictions on RAISE, specifically, it requires at least one die of flash be devoted to RAISE. Since SSDs with 128GiB on the circuit board generally use 8GiB or larger dies, they must devote no less than 8GiB to RAISE.

    So it is unlikely that Adata is using 4GiB for RAISE and 4GiB for OP in their "128GB" SSD.
    Reply
  • TheSSDReview - Saturday, February 25, 2012 - link

    Don't jump in often but thought I might on this one occasion.

    The article is exactly correct as it is in our original which was the result of conversation directly with LSI SandForce.

    The typical SSD has 128GB capacity, this being actually incorrect in the total byte calculation thanks to the original formula set out by HDDs some time ago. 1GB is not 1GB, but rather 1024 bytes which we all know. It is still described as 1GB however.

    Traditionally, a 128GB SSD utilized 7% of available RAM which brought it down to 120GB...or so we thought. In all actuality, the total OP was that 7% PLUS the extra 7.37 that you mentioned above. This has always been there but not accounted for in end calculations. Now, when the original 7% is not used (8GB) we are left with the actual unaccounted for space for OP, this being 7.37% which is used.

    The new drives are technically, 64/128/256/512GB drives as are those advertised by all non-SF SSDs... Its the background space that many are unaware of that has always been used for GC and other activities.

    As confusing as this may seem, it is no different than the same capacities listed by Crucial, Samsung and others.

    Thanks ahead
    Les@The SSD Review
    Reply
  • TheSSDReview - Saturday, February 25, 2012 - link

    Oooops if that could be corrected.... 'memory' and NOT 'RAM'. okok dawning bulletproof vest! Reply
  • jwilliams4200 - Saturday, February 25, 2012 - link

    Actually, thessdreview article, as with so many articles on that site, is full of mistakes and misleading information.

    I obviously cannot comment on what exactly was discussed in the LSI conversation, but clearly either you misunderstood what you were told, or the anonymous LSI rep was simply mistaken.

    The rest of your comment is nonsense. You seem to have confused GB and GiB, and so your statements make no sense whatsoever. The reason you think it may seem confusing, is because you have confused the matter greatly.
    Reply
  • TheSSDReview - Saturday, February 25, 2012 - link

    Not even worth the response but I am sure you will see it sooner or later... Thanks for the support! Reply
  • jwilliams4200 - Saturday, February 25, 2012 - link

    thessdreview has long been the absolute worst site for SSD journalism, with nearly every article chock full of mistakes and misleading information.

    There is definitely no "support" for thessdreview from me.

    I am only disappointed to see anandtech sinking towards the level of thessdreview. But this article was not written by Anand, so it is probably unfair of me to say that anandtech in general is sinking in quality. But Anand really needs to vet his staff writers more carefully.
    Reply
  • TheSSDReview - Saturday, February 25, 2012 - link

    In all fairness.. what you have tried to controvert above is this statement from the article:

    What is really happening behind the scenes of all SSDs is the difference between Billion Bytes (10^12 = 1,000,000,000) and Gigabytes (2^30 = 1,073,741,824). That delta is really 7.37% of the physical capacity.

    So when you see a drive at 120GB of 128GB of physical flash you are thinking it is 7% OP, but in reality it is more like 7% + 7.4% or 14.4%. Because the translation of Billion bytes and Gigabytes was made by HDD manufacturers some time ago, most users don’t “think” about that difference.

    Other than expressing a definite opinion, you haven't explained how this statement might be incorrect.
    Reply
  • jwilliams4200 - Saturday, February 25, 2012 - link

    I'm not sure why you feel it is my duty to correct your shoddy journalism, but in fact, I already did explain your error. You have confused gigabytes (GB) and gibibytes (GiB). Reply
  • JarredWalton - Saturday, February 25, 2012 - link

    No, he has simply explained that when we say an SSD has "128GB of flash on board and they use 8GB for OP", what we're really saying is "the SSD has 128GiB of flash on board and they only support 120GB of storage, thus giving you an extra 17,438,953,472 bytes of storage unaccounted for." This space is typically used for OP, but in the case of SF it can also be used for RAISE. How much of it is used for each? We're waiting for an answer from ADATA (see above comment by Kristian). Reply
  • jwilliams4200 - Saturday, February 25, 2012 - link

    Wrong. He wrote "Gigabytes (2^30 = 1,073,741,824)". That is incorrect.

    1 GigaByte = 1 GB = 1,000,000,000 Bytes.

    1 GibiByte = 1 GiB = 1024^3 Bytes = 1,073,741,824 Bytes

    Almost all SSDs in the 120-128GB (advertised) range have 128GiB of flash on board.
    Reply
  • JarredWalton - Saturday, February 25, 2012 - link

    FWIW, I think the powers of 2 SI standards are stupid either way. Gibi, Mibi, Kibi, etc. were all created after the fact because computer software coopted the Kilo, Mega, Giga prefixes. 99% of the time, people don't really care whether when you say "gigabytes" you mean "gibibytes". We all still say "I have 8GB of RAM" for example, when we really mean "I have 8GiB of RAM". So for you to jump down his throat over issues of GiB/GB when it's pretty damn clear what his intent is seems pointless. He understands what he's saying, and we all understand what he means, but you're on the Internet arguing over the misuse of Gigabyte. Reply
  • jwilliams4200 - Saturday, February 25, 2012 - link

    It is absolutely clear that he does NOT understand "what he's saying" because his article and comments make no sense.

    While I agree that in general usage it makes little difference whether someone uses GB and GiB correctly, this is certainly NOT general usage.

    In this case, GB and GiB are essential to expressing the situation clearly. You need only look at any discussion of overprovisioning and SSD capacity in order to see the confusion that comes from using GB and GiB incorrectly.
    Reply
  • TheSSDReview - Saturday, February 25, 2012 - link

    John...your sitting there on your antiquated XP driven computer and choking. You are doing as you always do which is bicker over small matters and missing the entire picture. Have you ever provided a constructive or productive response to anyone here or in the Forums? In your original statement you said...

    Since the SSD is advertised as "120GB", it must be capable of storing at least 120,000,000,000 bytes. Since there is still 120GiB = 120*(1024^3) = 128,849,018,880 bytes of flash left after subtracting 8GiB for RAISE, that leaves 128,849,018,880 - 120,000,000,000 = 8,849,018,880 bytes available for OP.

    So...lets follow what YOU wrote and I will ask two questions...

    In a typical 128GB (thats gigabyte) SF SSD, how many GB are used for OP and then, how many GB are used for RAISE? Lets keep them distinctly separate shall we? And wait a minute... lets not use GiB or try to break it all down as to confuse you any further?

    Reply
  • Beenthere - Saturday, February 25, 2012 - link

    So now there will be cheaper versions of defective SSD drives instead of the premium priced defective SSDs. Is that suppose to be good? ;) Reply
  • ckryan - Saturday, February 25, 2012 - link

    So SF used to need one a fixed amount of capacity for "SF Stuff" and now they don't. On a 25nm Sync equipped 2281 that would be 8GiB. For RAISE primarily, and like the 520, that's getting dropped. If Intel hadn't used that saved space for OP, you'd get 119GB usable capacity on a 120GB 520. They've chosen to increase OP, which is wise. SF is probably leaving that up to it's partners, but anyone (end user) can OP a drive. So in reality, you're getting a little more capacity and a little more flexibility. Assuming the prices are the same, you're getting more available capacity for the same money.

    Personally, I don't care about available user space. I care about how much NAND is on the drive. I do believe RAISE isn't as awesome as it's cracked up to be, and I feel Intel's decision not to use it as a vindication of my beliefs. But with respect to OP -- I can do that my damn self.
    Reply
  • mepenete - Sunday, February 26, 2012 - link

    Uhoh, it's hittin Engadget levels of comment battles.

    On the plus side... the more SSDs, the merrier.
    Reply
  • macuser2134 - Sunday, February 26, 2012 - link

    What a load of bickering. Lets face it - most people care only about the formatted capacity of a drive. And hopefully have a bit left over. Because that's how we use disk space in the real world.

    And the best way to find out the disk space is so simple. It just requires getting hold of the drive(s) your interested in. Then format it HFS+, NTFS, ext3, whichever you want. Not everyone can afford to go out buying SSDs just for this purpose. However there are forums where people who have already purchased the drive are happy to help out with reporting such information. For not-yet-released drives its rather more a case to get hold of / wait for a review sample.

    This all sounds so simple I honestly don't know why I'm bothering to explain it. Whatever operating system - they can all report the number of formatted space available in total BYTES. And you're pretty much beyond any kind of ambiguity since BYTES can never be confused. Divide by 1024*1024*1024 and theres your wholesome no-skimping Gigabytes. If you wish to compare to the true capacity of NAND dies on the SSD then that should be pretty easy. However NAND packages often are in bits - so just divide by 8. Noting that cells in a (presumably non-ECC) die aught to be laid out in a square and powers of 2^ = 1024*x. That's a teeny bit presumptuous so check the Manufacturer's data sheet if in doubt. Should be pretty clearly documented.
    Reply
  • TheSSDReview - Sunday, February 26, 2012 - link

    I have to say that I agree totally and apologize for my role in the above. Earlier today, I commented on my thoughts on another site and they are, quite frankly, that this is a very SMART business move for LSI SandForce which will serve them well. As much as this site is revered for its technical aspect, the truth simply is that the typical computer user will never be able to tell the difference in performance, regardless of what SSD they buy, "in typical computer use".

    In fact, I thinks even the boss here will agree that, in typical use it is next to impossible for ANYONE to tell the difference simply because the access speeds of all SSDs are so similar and, although they are responsible for the very visible upgrade when comparing a HDD to a SSD, they are also responsible for each SSD running very similarly visibly DURING TYPICAL USE.

    So, in the end, this new SSD will now afford the consumer with the same performance that they see from all SandForce drives (remember typical user scenario), yet the capacity has just put them beside the tech moguls like earlier Intels, Samsung and Crucial.

    Performance difference for the typical user = None.
    Capacity increase for the typical user = Yup
    Business decision For LSI SandForce = A+

    IMHO, regardless of past experience or our decision to go with one SSD company or another, this was a very smart business decision for LSI SandForce which will result in the one thing that every company vies for...sales.

    As for the above, I won't be commenting further as I have received several e-mails regarding the history of JW on other sites for which he is now banned 'apparently'. Again apologies and I hope to have contributed somewhat which was my original intent.
    Reply
  • jwilliams4200 - Tuesday, February 28, 2012 - link

    It would be better if you would apologize for printing such a misleading article on your site, or better yet, if you would strive for accuracy in all future articles that you post on your site, rather than continuing to post articles that are chock full of mistakes and misleading information. Reply
  • magnetar - Tuesday, March 06, 2012 - link

    I've read the articles at TheSSDReview that jwilliams is referring to ("ADATA First to Release New Higher Capacity SF SSD Line", and "LSI Releases Code to Manufactures..."), and while the portions of those articles he is taking issue with are ambiguous and unclear, I can see his point. Unfortunately, IMO he did not express it clearly enough for those who have not read those articles.

    To paraphrase the article, the claim was made that the loss of OP by using the full capacity of the NAND available will be made up for by the difference between counting the NAND in GiB and GB. To quote the article, which was also quoted in a post from TheSSDReview rep:

    "What is really happening behind the scenes of all SSDs is the difference between Billion Bytes (10^9 = 1,000,000,000) and Gigabytes (2^30 = 1,073,741,824). That delta is really 7.37% of the physical capacity.

    So when you see a drive at 120GB of 128GB of physical flash you are thinking it is 7% OP, but in reality it is more like 7% + 7.4% or 14.4%. Because the translation of Billion bytes and Gigabytes was made by HDD manufacturers some time ago, most users don’t “think” about that difference."

    The problem with this that jwilliams and I have, is that the difference in GiB and GB does not translate into MORE NAND available. HDD and SSD capacity is given in GB by manufactures, and GiB by an OS, like Windows. As we know, that translation is a (apparent) reduction, a 128GB drive becomes ~119GiB, and a 120GB becomes ~111GiB. Actually, nothing reduced (or increased) at all, only the units we are counting in changed the digits that we see.

    Regardless, it is not correct to say there is actually 14.4% of OP NAND available. Those few sentences I quoted are prefaced by the author discussing the OP situation with LSI, so the casual reader may assume that the LSI rep was saying that much OP existed. If that were true, then we have been misinformed about NAND chip capacity and operation by many publications, including this one, which I extremely doubt.

    NAND chips are manufactured in very specific capacities, and a SSD would need an extra NAND chip to have, for example, both a 128GB capacity and ~7% of OP, or a 120GB capacity and ~14% OP.

    Some SSD manufactures choose to OP their SSDs by design, and others do not. Neither is a crime, but the later puts more of the responsibility for the SSDs performance into the hands of the user.
    Reply
  • magnetar - Tuesday, March 06, 2012 - link

    Uh-oh, upon further review, I am wrong about the extra OP space available in NAND chips. Indeed there is extra space available in a NAND chip, which are rated in GB, when their actual capacity is in GiB.

    The numbers quoted in TheSSDReview are correct, there is 7.37% extra space in a NAND chips, independent of the typical 7% OP amount that may be provided.

    My apologies to TheSSDReview and others I may have maligned. At least I learned something.
    Reply
  • TheSSDReview - Friday, March 09, 2012 - link

    Wow.... Thank you magnetar as it takes a big person to return with that last comment. It is nice to see that someone does provide a bit of support where another, without cause, does his best to reduce a report without understanding what was stated right off.

    At CeBIT in Germany right now and elected to get this in quickly.
    Reply
  • connor4312 - Sunday, February 26, 2012 - link

    It'd be nice if the SSD companies take a month or two off from trying to cram as much storage as possible into a 2.5" drive, and instead refine their processes to make SSDs cheaper and more reliable. Reply

Log in

Don't have an account? Sign up now