POST A COMMENT

40 Comments

Back to Article

  • repoman27 - Saturday, February 15, 2014 - link

    I believe PMR is a TLA for "Perpendicular Magnetic Recording" not "ParaMagnetic Recording".

    That video is the nerdiest thing I have ever seen though... Amazing!
    Reply
  • dananski - Saturday, February 15, 2014 - link

    Nerdiest thing ever? You've obviously not seen "A Capella Science - Bohemian Gravity". Reply
  • Serialcrusher17 - Saturday, February 15, 2014 - link

    My vendor already seems to have the HGST He6.. 7 platter design.. SATA or SAS ~$850 Reply
  • jjj - Saturday, February 15, 2014 - link

    Toshiba said they will be sampling in february.
    And ofc Segate said they'll ship 6TB, 6 platter enterprise drives early in the second quarter.
    But i wouldn't expect anyone to be in a hurry to ship in consumer,they need to control the pricing decline to maximize profits and with only 3 players it's not very likely for anyone to try to gain share.Going forward,new tech requires investments so that's another aspect that will slow them down,not because they can't do it but because they have no reason to .The limitations are less about the tech and more about financial aspects.The lack of competition is unlikely to change, might even get worse since WD has been forced by China regulators to run Hitachi as a separate company but very soon they'll try to get rid of that.So expect price/GB to decline very slowly and capacity to go up only to allow them to stay in line with demand at an ideal cost.

    PS:Since some replies might start bashing the HDD makers for the lack of competition i got to mention that it's far from ideal but on the other hand others have way higher margins. Seagate and WD have gross margins bellow 30% while Intel for example has more than twice that so things are far worse in many other places (the much higher margins are partially offset by the fact that Intel does invest more, in %, in R&D). So go ahead and decry the situation in the HDD market but don't forget that it's sometimes much worse elsewhere.
    Reply
  • Brutalizer - Saturday, February 15, 2014 - link

    What competition? There is an oligopoly right now, that is the reason the hard disk prices dont fall. Before the Thailand floodings, there were four vendors. The fourth bought the third vendor, and the prices stopped falling. The Thailand floodings were just an excuse to uphold prices. During the floodings, MORE disks were shipped than the year before - so there were no shortage. The reason these 5TB and 6TB disks arrive, is because they are afraid of SSD disks. Without the SSD vendors, we would still be paying up through our nose for 1-2TB disks. Read about market manipulation here:
    http://news.softpedia.com/news/Enjoying-Market-Man...

    Without SSD competition, the HDD vendors would still be screwing us today. There are soon 1TB SSD disks, we see 480GB SSD disks. What is 3-4TB HDDs in comparison to 1TB SSD disk? Nothing. In the HDD labs, I bet they have 8-10TB disks ready to be released when they feel threatened by SSD disks. Today we should all be using 6-7TB disks for the price of a normal 3TB disks - but there is an oligopoly. :(
    Reply
  • name99 - Saturday, February 15, 2014 - link

    Prices have fallen and fallen substantially. To claim otherwise is ridiculous.
    I'm no fan of the HD companies --- see my large post below --- but this is precisely the problem. The prices HAVE fallen so much that there is no money in the business to do research.
    Reply
  • Dentons - Saturday, February 15, 2014 - link

    The poster is clearly referring to the past 2.5 years. In that time, hard drive prices have not fallen. That you suggest otherwise is ridiculous.

    In mid 2011, ostensibly due to the Thai floods, prices greatly increased. As others have noted, the flood excuse didn't begin to hold up. The industry shipped more drives in 2011 than they had in 2010. By Q1 of 2012, all of the impacted factories were either reopened or relocated, yet prices remained high.

    Since mid 2011, prices have fallen somewhat, but have still not reached the 2011 rates. A full retail priced 2TB drive in 2011 was cheaper than today's retail pricing. Such pricing trends are nearly unheard of in the technology industry.

    That this anomaly has lasted over 2.5 years clearly suggests tacit price fixing, perhaps even active collusion among the few remaining hard drive manufacturers. Tacit price fixing is often not actionable, but collusion is. For most of the past year, have been rumors of government price fixing investigations into the hard driver industry.

    Hard drive prices still show no signs reverting to natural price trends. Lack of price drops over such long periods of time only happen there is overwhelming demand, supply shortages, or price fixing. Since first two can easily be ruled out, the last possibility, price fixing, is the most likely.

    Were there still a major 4th player in the market, available sizes would very likely be larger and prices would almost certainly be lower. There isn't, so customers are beholden to stagnant products, at stagnant prices. Absent a price fixing action by the governments, this awful status quo seems destined to persist.
    Reply
  • Solandri - Sunday, February 16, 2014 - link

    It wasn't price fixing and it wasn't collusion. The HDD industry was simply too competitive. Margins prior to the flooding were razor thin, unhealthily so (I heard about 5%, with some HDD companies going below 3% - this is the reason IBM bailed out of the industry). They had the slimmest margins in the computer industry (aside from memory when there's a glut).

    The lack of significant price drop in the last 2.5 years is simply the margins increasing to healthy levels. Manufacturing costs dropped, but retail prices did not, thus increasing the margins.

    As for the consolidation in the HDD industry, unfortunately it happened in the middle of a recession and at a time when SSDs began making significant inroads into the desktop market. Consequently I think we ended up over-consolidating. 4 manufacturers was probably a healthy number, but the economic conditions at the time dropped it down to 3. (And technically we consolidated all the way down to just 2.5 HDD manufacturers. Toshiba only made 2.5" HDDs. One of the conditions placed on WD for acquiring Hitachi Storage was that it sell off some of its 3.5" HDD manufacturing to Toshiba, to make it a full 3 HDD manufacturers.)
    Reply
  • sonicmerlin - Saturday, February 22, 2014 - link

    What dropped it down to 3 was incompetent and business worshipping "regulators". Reply
  • name99 - Sunday, February 16, 2014 - link

    OK, let's look at the graph.
    Seagate 4TB internal drive.
    http://camelcamelcamel.com/Seagate-Desktop-3-5-Inc...

    WD external 4TB drive:
    http://camelcamelcamel.com/Drive-Security-Local-Ba...

    Travelstar 1TB 2.5" drive.
    http://camelcamelcamel.com/HGST-Travelstar-2-5-Inc...

    They all follow the same pattern.
    For a more aggregate view look at
    http://www.extremetech.com/computing/153879-storag...
    Or
    http://www.zdnet.com/storage-in-2014-an-overview-7...

    Everything shows the same pattern. Prices fell relentlessly. Floods in Thailand led to a bounce in prices. THEN around early 2013, we reverted to pattern, with prices again falling. The Backblaze price history of 4TB drives is particularly clear on this point because it's not confused by the Thai floods.
    Reply
  • Dentons - Monday, February 17, 2014 - link

    That you avoid using 2TB drives in your example is telling. Doing so would prove just how inaccurate your contention is. 2TB was perhaps the most common size, both then and now. 4TB drives were not available in 2011, or barely so.

    There has been price fixing in the hard drive market. There is no question of this. The only question is whether it has been tacit price fixing or active collusion.

    It's difficult to understand why anyone would give Seagate and WD a pass on this, unless perhaps they work for one of those companies,
    Reply
  • Fallen Kell - Monday, February 17, 2014 - link

    I have to say you are picking and choosing to get those results. As Dentons pointed out, you did not look at drives that were on the market prior to the floods:

    http://camelcamelcamel.com/Western-Digital-Caviar-...

    http://camelcamelcamel.com/HGST-Deskstar-3-5-Inch-...

    Those show a pretty clear result that prices have been stagnant since the flood.
    Reply
  • sonicmerlin - Saturday, February 22, 2014 - link

    Lol wow 1 TB and 4 TB. Did you really think no one would call you out? Reply
  • zodiacfml - Sunday, February 16, 2014 - link

    less profit? i highly doubt it. they could have been making more money these days due to selling to enterprises and data centers. there is only less incentive for them to provide the consumer market with larger drives. Reply
  • fteoath64 - Sunday, February 16, 2014 - link

    " they could have been making more money these days due to selling to enterprises and data centers". Too right. A captive market of repeating customers for sure. Look at the growth of data centres everywhere. They all rely primarily on HDD as their storage mediums. in fact, many are using consumer drives so they can replicate plenty of storage in order to handle data redundancy issue.
    I hope Toshiba keep making 2.5 inch drives of higher capacity and power saving ones for laptops. Also 1 inch drives of low power for future tablets. Due to the fixed lifespan of those drives, there is continued business so it can be sustainable for a top grade manufacturer serious about storage. The existing ones are just squeezing every cent from existing process without innovating much as SSDs takes over the low-capacity storage devices.
    Reply
  • Gigaplex - Saturday, February 15, 2014 - link

    Soon? We've had 1TB SSDs for quite a while now. Reply
  • name99 - Saturday, February 15, 2014 - link

    "I must confess that at this point in the HDD cycle we might have 6TB 3.5” drives in the consumer market, but the recent focus on SSD consistency and a combination of overcoming physical limitations via new methods might be causes for the delay."

    I don't think this is a correct analysis. The companies (and individuals) working on better SSDs are not the HD companies.

    I think a better explanation is that commoditization has destroyed the HD business. Competition has sent the price of drives so low that there is no profit left over to fund the R&D necessary to improve them. For a while IBM did the heavy R&D necessary (GMR, PRML, etc) to keep things going, but we've hit the limits of those techniques and no-one with deep pockets is around to do more than minor tweaks of existing tech that allow a reduction in price by a few cents.

    This is basically the same process that killed the PC industry. (Intel kept up it's R&D, so CPUs kept improving; MS, for reasons unrelated to commoditization, screwed up its part of the R&D/improvement process; and the vendors, from Dell to HP to Lenovo, did nothing to improve their product over five years. Every "improvement" they did make was geared towards a cheaper device, not a better one.)

    This is not COMPLETELY the vendors fault (PC or HD). Markets are not as perfect as your libertarian friends would have you believe, and one place they fail is precisely this sort of situation. In the absence of a benevolent more-or-less monopolist; once ATT, once IBM, currently (but maybe not for much longer) Intel, currently Apple and Google; competition to survive today means that fundamental R&D which benefits the entire industry does not get done.

    The situation becomes unsustainable when commoditization is complete, when all I care about is buying a PC (not a Dell or an HP) or a drive (not a Seagate or a WD).
    Apple has avoided it so far because people actually care that they get a Mac or an iPhone and are willing to pay substantially more for that.
    Likewise Google so far delivers a superior enough product that people care that they are getting Google search not generic internet search.
    Intel are headed for danger because there are few people who care that they actually have "Intel Inside", what they care about is "fast CPU, low power" inside and Intel is safe only as long as they deliver that more than the ARM competition...

    But the HD vendors ARE to blame for the fact that every last one of them acted like a complete and utter retard in the face of SSDs. Rather than viewing their product as "persistent storage" and asking how they could use flash to improve that product, they insisted that what they were selling was "spinning metal" and that nothing could ever compete with that.
    Even now, as they deliver too little too late, they insist on acting like idiots. For example WD sells, in the Black2, a product that is PERFECT for Macheads because it works so well with Apple's Fusion technology. And Macheads are the one segment of the market that are actually willing to spend money. How does WD exploit this opportunity? By conspicuously telling Apple fans "fsck you and all your lovely money, we're going to pretend it's still 1998 and sell a product that deliberately won't work well on your computers, even though all it would take to fix this is to provide a small app that automatically exposed our two drives and ran Core Storage Fusion on them."

    So to summarize:
    - drives suck because their no pot of money subsidizing their improvement
    - this is 80% commoditization and 20% stupidity of the management at the drive companies.
    Reply
  • purerice - Saturday, February 15, 2014 - link

    Nobody can deny that you have done your research but your argument has a problem. For the longest time Apple insisted that "Mac" was not a "PC". PC vendors did try to differentiate themselves ("Dude, you're getting a Dell").
    HDD vendors to some extent do too, at least in my market. Yet flaming WD as an example for saying this Toshiba announcement is meaningless does not compute with me.

    Then you attack libertarian thinking because you don't see more innovation in the HDD market. Really? 9 years ago I got a 250GB external drive for $200 that was wall-powered and weighed 1 pound. 4 years ago I got a 750GB external drive for $100, bus powered, weighing 0.6lb.
    Last year I got a 1TB external drive for $90, bus powered, weighing 0.5lb.
    This year for $90 you can get a 1.5TB external drive weighing 0.4lb.
    So in 1 year there is a 50% increase in capacity and 20% reduction in weight at the same price point.
    Does that not suggest the market is providing constant improvement in customer value?
    Would not commoditization if it were happening, also mean improvement in customer value?
    Reply
  • name99 - Saturday, February 15, 2014 - link

    I'm not interested in religious wars. I'll just comment on two things.
    (a) A Mac is NOT a PC, in the sense that a Mac is not a member of the PC commodity. That's precisely my point. Apple HAS avoided becoming a commodity. The PC vendors have tried to avoid that fate, but have been unsuccessful. There WAS a time when PC brands were to some extent differentiated --- Compaq in particular really meant something as brand. But not for the past five years.
    The PC vendors, like the HD vendors, are aware of the problem. But they have both waited too long to try to deal with it, and simply don't have the resources for meaningful differentiation. The difference between PC vendors is in what crap they ship pre-installed. The difference between HD vendors is likewise in useless crap they ship on the drive. These are NOT differentiators that persuade a buyer to pay more for a product; unlike the differentiation that Apple has achieved.

    (b) You would have to be deluded to claim that the rate of innovation in HDs today is the same as it was.
    2007 1st 1TB HD
    2009 1st 2TB HD
    2010 1st 3TB HD
    2011 1st 4TB HD
    2012 nothing
    2013 nothing
    2014 An extremely expensive 6GB HD and a (who knows the price) 5TB HD. (Yeah, yeah, the 6GB was announced in 2013. Let me know when it ships...)

    To me this is a dramatic slowing down. We've gone from essentially exponential growth in capacity (doubling in size every two years or so) to sub-linear growth in capacity.

    We've seen slow improvements in the 2.5TB form factor over that time. While 3.5 capacity has largely stood still, 2.5TB has grown from about 750GB to 2TB. But 3.5 strikes me as the canary in the coal mine. That's where the research will appear first. Instead all we have is that standing still while 2.5" repurposes 3.5" ideas from a few years ago.
    Reply
  • MrSpadge - Sunday, February 16, 2014 - link

    There are technical reasons for the slow-down. We're used to rapid progress in HDD technology because there were quite some low-hanging fruit to be had. Look at what they're doing now: shingled recording, helium filling, laser assisted recording or putting more platters into the same form factor than ever before. Any of these is significantly more complex and often expensive (for them to research and produce) than previous HDD technologies. Rest assured they're not using this just for fun - if they could continue business as ussual (but slower) they'd just do so and maximize profits. Reply
  • Flunk - Saturday, February 15, 2014 - link

    I think the bigger issue right now is that software companies haven't come up with a compelling reason to need all that storage space. People would buy larger hard drives if they had any need of them. Reply
  • Kevin G - Saturday, February 15, 2014 - link

    There are plenty of reasons for larger drives but they tend to be either enterprise scale (virtual machines) or content (HD video).

    For the majority of consumers, the 1 TB class of hard drives is more than enough for applications, their personal data and local content. Of course there are end user exceptions but scaling to 4 TB drives decreases that niche and then various RAID implementations resolves the rest.

    Streaming and cloud storage (which does merit large capacity drives on the backend) has eased the pressure for larger local storage. Stepping down to a 250 GB or 500 GB class SSD is sufficient for most consumers in terms of capacity, not to mention the massive performance benefits.

    In short, the consumer demand for more storage simply isn't there except in very specific niches.
    Reply
  • cjb110 - Monday, February 17, 2014 - link

    I'd agree, larger drives have become less of a requirement as people's storage either moves into the 'cloud', and as things like NAS becomes more prevalent and cheaper. A single 5/6tb drive is a shed load of data on a single point of failure, and shouldn't be shipping on any consumer based device (unless its the aforementioned NAS device).
    Possibly when 4k content becomes more common, there will be more cases for larger drives for consumers.
    Reply
  • flyingpants1 - Tuesday, February 18, 2014 - link

    The WD Black2 is alright. It's $250 on Amazon, compared to Crucial 480GB SSD for $260 and 960GB for $450. I think on a laptop you'd usually just get the 480GB SSD and buy a 1TB external. Reply
  • ruzveh - Saturday, February 15, 2014 - link

    I want to ask Anandtech and HDD manufacturer one simple question. Why cant internal HDD go beyond the usual size of 3.5" to more inorder to launch 10 to 100TB HDD? I wouldnt mind buying one. Its better then buying 10 pc of 2TB or more HDD. Are you listening? Reply
  • brshoemak - Saturday, February 15, 2014 - link

    They can and have before (look up Quantum Bigfoot - the scariest sounding drive ever) but at such a high density level you will most likely run into issues where the heads and arms travel are the limiting factor in terms of accuracy. I think there are some physics involved that make it not viable from a cost/performance standpoint. Reply
  • Kevin G - Saturday, February 15, 2014 - link

    I've personally owned 5.25" full heigh hard drives in the past and seen a 8" unit. With the emphasis on mobile technology and increasing density killed off the 5.25" drives in the mid-90's (the shift to 3.5" hard drives started in the late 80's). In the early 2000's the rise of the laptop saw a further shift from 3.5" to 2.5" as focal point for hard drive manufactures. 3.5" has become form factor for external storage, the incredibly shrinking desktop space and enterprise servers who need storage capacity.

    That of course doesn't prevent a HD manufacturer from bucking the trend and releasing a 5.25" today, other than it being a very niche product. Going back to the mid-90's 5.25" drives, they had several issues. The motors to spin the platters consumed more power and energy efficiency has become a major factor in computing today. Latencies of 5.25" hard drives were typically lower than their 3.5" counter parts (spindle speeds played a factor as a high RPM 5.25" could have a lower latency than a low RPM 3.5" drive). Then there was the noise factor. 5.25" drives were loud, very LOUD. Imagine living in an apartment and you're backing up to data to a coupe of bare 5.25" and you get a knock on the door. Police have been called to answer a noise compliant about a small motor running inside the residence.* I'd have to look at the data sheets but I've been under the impress that 5.25" drives were more susceptible to vibration and had lower reliability rates.

    20 years of technological improvements could negate/reduce many of the disadvantages. However, at current 1 TB/3.5" platter densities the raw capacity gains would be roughly 2.4 TB/5.25" platter. Other ideas like increasing the number of platters. An eight platter 5.25" drive would equate to roughly a 19 TB drive. Costs increases wouldn't be linear but under that false assumption such a drive would retail for roughly $1,000 USD. Honestly, I'd be more comfortable running five 4 TB drives in RAID5/RAIDZ than the single hypothetical 19 TB unit.

    *Second hand story but having owned some 5.25" units, I find this believable if they were particularly loud, high RPM units.
    Reply
  • MrSpadge - Sunday, February 16, 2014 - link

    I like this idea, too, but there are quite a few points against this:

    - at the same rpm (angular velocity) the outer regions of the platters move much faster in a 5.25" drive compared to 3.5" -> vibration and platter flattering become much worse
    -> you need to lower rpm and probably areal density (sicne with more flattering the read/write head can't get as close to the platter)

    - at larger diameters access times become worse, es the heads have to travel further
    - together with lower rpm this makes for a slow drive (just like the Bigfoot was)

    - there's no infrastructure for producing such platters any more -> higher prices due to limited volume

    - the motor would have to be significantly stronger and idle power consumption would be significantly higher than 3.5" HDDs (power per capacity might not be too bad, though)

    - redundancy: if you run more smaller drives it doesn't hurt as much if one of them fails (you could argue that this won't matter of you ned pretty much infinite disk space)
    - big super-expensive drives won't sell in the same quantity as the current disks -> economy of scale won't work as well to drive price down
    Reply
  • cjb110 - Monday, February 17, 2014 - link

    Really? You want that much data on something that can die with a complete loss of everything that was on it? Reply
  • Concillian - Wednesday, February 19, 2014 - link

    Tooling for making / processing larger platters no longer exists. As much as anyone wants it, it's not going to happen. Reply
  • Gnarr - Saturday, February 15, 2014 - link

    HGST already has a 6TB drive.
    http://www.hgst.com/hard-drives/enterprise-hard-dr...
    Reply
  • flyingpants1 - Tuesday, February 18, 2014 - link

    It says it's the world's first hermetically sealed drive, because they've replaced the air with helium. Reply
  • BroderLund - Saturday, February 15, 2014 - link

    LaCie, which I belive are owned by Seagate, already have 5TB drive for sale. In stock in some shops:

    http://www.amazon.com/Lacie-9000465-USB3-Thunderbo...

    http://www.bhphotovideo.com/c/product/1024100-REG/...
    Reply
  • name99 - Saturday, February 15, 2014 - link

    I don't think this is genuinely for sale yet. My guess is that, in spite of what Amazon and the other shop say, if you actually order you'll be waiting for a while. La Cie, for example, advertise but say that it's out of stock (not that it's ever been in stock).
    It will ship the same time Seagate's shingled 5TB drives ship. Which is presumably some time soon this year, but who knows...
    Reply
  • BroderLund - Wednesday, February 19, 2014 - link

    May be, but as far as I have seen the stock on Amazon has droped from 19 (few days ago) to 5 (when I'm writhing this).

    Fake maybe, but I dough it. Still waiting for the drives to be avalible near me, Norway. The shops in Norway are expecting shipment in early March. Time will tell.
    Reply
  • Alientech - Sunday, April 06, 2014 - link

    Seagate's already shipping shingled 3TB drives. They are painfully slow and can not even be used as backup drives. When you are tying to copy 1TB of data and it says 1 day left to finish, that is just too long. It used to take like 5 hours before which itself was slow. I wonder how reliable the new disk is compared to the old one. Using this shingles should not affaet the write speeds this much. But they seem to be go in a U type bit density in a zone so the speeds vary it each zone itself unlike before where each zone was the same speed and the speeds slowed down the inner you got to the center. If WD eants to take market share all they need to do is use some other techonology that does not slow down the drive so much during writes. But heating would also slow down writes since it has to heat the path before it can write to it. Reply
  • Jalek99 - Sunday, February 16, 2014 - link

    Toshiba? *looks at case of dead standard and AV-specific drives* No, just no. Reply
  • ethebubbeth - Sunday, February 16, 2014 - link

    I just spent a ton redoing my personal NAS with 24x 3tb drives. People have the need for the storage. Reply
  • Navvie - Sunday, February 16, 2014 - link

    That first sentence is a mess. I had to read it several times to understand the intended meaning. Reply
  • zimanodenea - Wednesday, February 19, 2014 - link

    All this talk of why on earth would people want to so much data on a single drive. Personally, I like a single large capacity hard drive as my backup destination. At the moment I use a 3TB drive to back up all the data from multiple computers. As soon as an affordable 6TB drive arrives I'll be able to use my current 3TB backup drive in my computer and have space to backup everything to my 6TB drive.

    I'm aware that I could buy an external enclosure (fine until you damage the USB3 connector) which has 2 x 3TB drives or buy a NAS. I like being able to just remove my single drive from my system (is in an internal caddy) and store in another room somewhere.
    Reply

Log in

Don't have an account? Sign up now