Forget Flash: Resistive RAM Crams 1TB Onto Tiny Chip

nk497 writes "Flash memory could soon be a thing of the past, according to U.S. startup Crossbar, which claims it's close to bringing resistive RAM (RRAM) to the market. Crossbar is touting impressive specs for the RRAM technology, promising 20 times the write performance at a fraction of the power consumption and size of the current best-in-class NAND flash modules — and squeezing terabytes of storage capacity onto a single chip the size of a postage stamp. The company also claims its technology can retain data for up to 20 years, compared with the standard one to three years with NAND flash."

Comments

Will we finally get a replacement for hard disks?

[Anonymous Coward]

I want to get rid of the last moving parts:
* hard disk
* keyboard
* mouse buttons

I want to believe

[musmax]

I really do.

Re:I want to believe

[muon-catalyzed]

Here is the official release note from Crossbar Inc.
http://www.crossbar-inc.com/events/press-releases/crossbar-emerges-from-stealth-mode.html

They have a production sample ready for the SOC integration, that looks all nice, but does the production scale well? Could this actually is be price comparable or cheaper then the NAND flash?

Re:I want to believe

[Anonymous Coward]

I will only believe it when I can buy it.

but what about the cloud?

[alen]

why do i need this if there is the cloud to keep my data? why carry 1TB on my iphone when i can just pay at&t more money for more data to stream my music and netflix?

Re:but what about the cloud?

[Anonymous Coward]

Because the streaming service providers need to store the data somewhere?

Re:but what about the cloud?

[QilessQi]

Because you might want to listen to your music -- or watch your favorite movies -- even when you're in locations/situations where you can't easily/cheaply stream your data. Such as on an airplane, or deep inside a building where there's no reception.

This is not to say that the data wouldn't live in the cloud. Think of your portable device as simply containing a cache, which is loaded on demand. The bigger the cache, the better.

Re:but what about the cloud?

[Anonymous Coward]

woosh

Re:but what about the cloud?

[Anonymous Coward]

1TB out to be enough for everybody...

Re:but what about the cloud?

[Anonymous Coward]

woosh

i assume that is the sound of a rainstorm in the cloud with all your data falling out of it?

Re:but what about the cloud?

[Anonymous Coward]

WOOOSH

Re:but what about the cloud?

[Anonymous Coward]

woosh

i assume that is the sound of a rainstorm in the cloud with all your data falling out of it?

In the police state of Amerika it sounds more like vrroooomm as it gets sucked thru "KIRBY" or some such NSA codename.

Re:but what about the cloud?

[jellomizer]

Because technology alternates back and forth on this stuff.
Early Computers did all the work. You had a computer it did all the work for you and only you.
Then we got the mainframe One Big computer with terminals so people use the terminals and the data gets saved on the Mainframe(s).
Then came the PC, we started to move off the mainframe and ran programs directly on our computer again.
Once broad band became cheap enough and popular services began to move to the cloud as on the average your data was safer there, and easier for the company to manage the software.
So with cheap and a lot of data we could go back to more of a Personal computing role again. Probably keeping the strong points from the past and making using computing a little more different.

Say you now have a Netflix app that will in the background download what it expects you to watch. Then if you want to watch it it is available even if it is offline.
Or your system will host an archive of your data in cases your networks speed is too slow or are offline.

Will their be trade offs you bet. But this type of stuff cycles around. CPU+Storage+Networking+Price Fluctuate over time. So the popular solution will change base on the systems strong points.

Desktops for average Joe Web Browser user, is starting to get out of fashion, and going to Phones and Tablets (I am not touting death to any technology here). But to get the optimal conveniences we are trading off Slower CPU, and Storage to get small form factor at a good price. So many apps are popular on the cloud. Because the servers have the Big CPUs and storage and will just send output to the low end Tablets. Now these tablets are getting faster and more storage so people will want to run more apps on them, thus more apps will be created.

Re:but what about the cloud?

[DavidHumus]

You might need it if you use your computer as something other than a toy or care about security or performance.

Re:but what about the cloud?

[Anonymous Coward]

Clearly, the concept of sarcasm is lost on you.

Re:but what about the cloud?

[Anonymous Coward]

Thank you for your loyal subscription. To better enhance your service we are increasing the rate per month from 8.99 to 12.99. This will position us in the future to deliver even more exciting things.

Or ....

Thank you for your loyal years of service with our company. It is with a heavy heart in 3 months we will be shutting our doors. Please use a search engine and fine a service similar to ours. Again thank you for your years of loyalty.

Re:but what about the cloud?

[Anonymous Coward]

Woosh anyone?

Re:but what about the cloud?

[Anonymous Coward]

Because you might want to listen to your music -- or watch your favorite movies -- even when you're in locations/situations where you can't easily/cheaply stream your data. Such as on an airplane, or deep inside a building where there's no reception.

This is not to say that the data wouldn't live in the cloud. Think of your portable device as simply containing a cache, which is loaded on demand. The bigger the cache, the better.

Thank you for participating in our Slashdot Sarcasm Calibration program. By replying to this post, you are ensuring that you and others are able to more accurately identify and appropriately respond to sarcastic posts and enjoy a more fulfilling sarcasm experience on Slashdot. Be sure to recalibrate for sarcasm on a regular basis.

Thank you from all of us here at the Slashdot Sarcasm Calibration Team.

Re:but what about the cloud?

[Belial6]

Say you now have a Netflix app that will in the background download what it expects you to watch. Then if you want to watch it it is available even if it is offline.

Very likely. Some of the streaming music services already do this. Google Play Music and MOG are two I am familiar with.

Re:but what about the cloud?

[Anonymous Coward]

Say you now have a Netflix app that will in the background download what it expects you to watch. Then if you want to watch it it is available even if it is offline. .

Last I checked, Netflix doesn't have porn. If you know of an app that really knows what I watch, please let me know...

Re:but what about the cloud?

[Anonymous Coward]

Do you think the cloud is made of magic pixie farts? Its just a network of computers, and it has to store data somehow.

Re:but what about the cloud?

[metaforest]

Early Computers did all the work. You had a computer it did all the work for you and only you.

Early computers did exactly one task. Programming consisted of rewiring the device for a new task. Previous task was discarded. You owned/ built the computer and it only worked for you.

This evolved into turing complete machines that were still application specific, but more flexible. Batch processing begins.

The rest of your evolution chain I mostly agree with.

Re:but what about the cloud?

[Anonymous Coward]

depending on the cloud to store and do computation for your data is silly and the stupidest idea ever its like back in 1950's where the all continent had only one super computer to do calculations for every one, people buy personal computers so they can save and compute their data personally, why dont people get the idea it is simple i need my own personal ram and 1 terabyte is just about what i need, my own hard drive to store my stuff and y own processor to play my games and do my stuff, its bullshit talking about the cloud because it just shows that the industry is lazy and cant figure out a way to make pcs faster so it relys on the cloud fuck that.
, take the cloud to president roosevelt we dont need it nobody needs it.

Re:but what about the cloud?

[Anonymous Coward]

I'm assuming you've never been without cell coverage, traveled more than 50 miles from home, and live in an area with 4G reception. The cloud is not an option for most people. That's like renting storage for furniture that won't fit in your house.

Re:Will we finally get a replacement for hard disk

[h4rr4r]

The first one we already do. SSDs are great.

The latter two, what would you replace them with? A trackpad could work I guess, but I can't see a replacement for a keyboard. Speaking is way slower than typing, typing on a touch screen is an error prone suckfest, and those are pretty much the only options right now.

Number of re-writes?

[Conspiracy_Of_Doves]

Do the memory points wear out after a certain number of re-writes?

Re:Number of re-writes?

[Anonymous Coward]

If this storage is made with memristors, then the amount of re-writes should be infinite.
In fact it may one day replace main memory (yes, main memory which is persistent). Memristors will also become components which can be used as part of logic circuits in processors. Memristors are also very interesting in Neural nets, which will become much smaller.

Re:Number of re-writes?

[ShanghaiBill]

Do the memory points wear out after a certain number of re-writes?

Yes, according to Wikipedia, but new developments have been increasing the material endurance while decreasing the power consumption (less power == less harmful heat).

Re:Number of re-writes?

[suutar]

I'm not seeing an overall benefit to persistent main memory. It could pretty much unify 'sleep' and 'suspend', but the security holes resulting from bad cleanup of memory data (or no cleanup, from all the code that assumes ram is volatile) seem like a much worse deal. Do you see benefits I'm missing?

Re:Number of re-writes?

[Anonymous Coward]

No refresh, thus having 64 GB in a mobile phone would be ok. With DRAM I expect the power cost of the refresh alone would be close to blowing the power budget.

Re:Number of re-writes?

[tibit]

Last I checked, pretty much no code assumes that RAM is volatile. Even the volatile keyword in C doesn't mean that if you store something there, it'll magically disappear, just that it may get modified by "other things" (other threads of execution, other processes, peripherals, ...).

I think that you've made up an imaginary problem for yourself. Just think about it. When an operating system maps a fresh regular memory page for a process, it will be zeroed sometime before the process reads it. Or didn't you know that in C/C++, all non-automatic, non-heap variables are always initialized/default constructed, respectively? In C running on a modern OS it can get wasteful, since first the OS zeroes the pages, then usually the C runtime library zeroes them again.

The biggest benefits include immunity to power outages (with cooperation from the hypervisor layer) and ease of diagnostics on systems that "die" and are in need of such diagnosis (say a car ECU after a crash). Saturn V's engine controllers used non-volatile core memory and IIRC it came handy more than once.

Re:Number of re-writes?

[suutar]

Oh, no, i didn't mean code assumes that values are going to disappear magically while in use. But code does assume that memory gets flushed by a power cycle. Or is there code I'm not aware of that zeroes RAM pages during shutdown processes? Reading info from RAM that's kept cold to extend data life is already a usable (if difficult) attack; pulling the plug on a box and booting off a USB stick to read data from non-volatile RAM would be much easier.

Re:Number of re-writes?

[mattack2]

Last I checked, pretty much no code assumes that RAM is volatile.

It doesn't assume that RAM is volatile, but isn't the fact that it is volatile PART of the reason "did you try turning it off and on again?" works so often?

If a program/OS has a memory corruption issue, it seems like there are at least _some_ cases where it would be harder to temporarily work around the issue if the memory is permanent.

(Devil's Advocating myself -- yes, in another way, this would be good, hopefully helping to flush out/get these types of problems fixed... But I mean mostly after something is in a customer's hands.)

Re:Number of re-writes?

[Belial6]

This is easy enough to fix. A small battery and a circuit that gets configured from the OS to indicate if the memory gets wiped on power loss.

Re:Number of re-writes?

[AmiMoJo]

Most modern operating systems do clear RAM when it is allocated to prevent programs accessing what was previously stored there. Of course there is the "cold boot" attack, but I don't see how this makes is any worse really. Anyone worried about that already has their OS clear RAM on shutdown and/or the BIOS so a full RAM test (which overwrites every byte) on boot.

Re:Number of re-writes?

[tibit]

The application code doesn't assume any such thing, because it's an assumption that is irrelevant to anything most application code could care about. It's the operating system that provides a certain execution environment that the applications are coded to. Just as an application programmer doesn't have to care (if they don't want to) about system going to sleep or being hibernated, you wouldn't care about the sleep or hibernation being implemented with nonvolatile main storage. The execution environment seen by the applications would be the same.

In every virtualized environment out there, the functionality of suspension and resumption of a VM's snapshot (including the implicit current state's snapshot) is equivalent to persistence across power cycles. Pray tell again, how is it a problem, because so far you're really insisting there is a problem, but you just won't share with anyone exactly how is it a problem. You're trying to make an argument, by all means, do make it, don't keep us guessing.

Code that's serious about security will keep at least the sensitive keys, if not the cleartext as well, only in the CPU's caches, and this requires some cooperation from the OS obviously, but it's not something that people usually bother to implement in spite of at least x86 and amd64 architectures letting you do so. The RAM gets entirely out of the picture if you assume that you're running on bare metal. In a virtual machine it might be much harder.

Re:Number of re-writes?

[tibit]

It doesn't assume that RAM is volatile, but isn't the fact that it is volatile PART of the reason "did you try turning it off and on again?" works so often?

Nope. The reason is that the CPU starts execution in a certain well defined state and at a certain physical address, where you better had some BIOS or monitor code to get you going. The memory contents are not assumed to be anything in particular and do not affect what a power cycle does.

The legacy PC BIOS, and presumably some other BIOSes, do place a magic word to in memory to differentiate between a cold and warm boot, but that's not that big of a deal if you had nonvolatile memory - remember, the whole thing would need to be designed to support it, BIOS included, so let's not be silly by pretending you take an off-the-shelf PC and just substitute the DIMMs with nonvolatile ones. It doesn't work that way, for crying out loud.

If a program/OS has a memory corruption issue, it seems like there are at least _some_ cases where it would be harder to temporarily work around the issue if the memory is permanent.

Given that every page that's not backed by a file is handed to the applications after being zeroes by the OS, and the OS itself obviously is supposed to initialize its own variables before use, I just don't see how this is of any concern to anyone. It's an imaginary problem.

Just so that you know, RAM contents do persist between power cycles in everyday hardware, including whatever you use to read this post. The RAM contents decay exponentially, with the exponent depending on temperature in such a way that lower temperature increases the half-life of memory contents. It doesn't affect anything, really, since a power cycle explicitly wipes all this memory. Of course if you're after cryptographic keys and such, the fact that main memory is nonvolatile does make a difference, but that's because most code that holds sensitive keys is not written to keep the keys solely in CPU caches.

Re:Number of re-writes?

[Cederic]

Well, it means I can still read the last bytes held in the memory of the mainframe computer I have hanging on my wall.

http://www.nzeldes.com/HOC/images/CoreMemory03.jpg

Amazing

[Anonymous Coward]

All he needs is a bunch of venture capital, and 12 to 18 months to make it happen!

Marketing literature is "news" now?

[Anonymous Coward]

Sorry, but I'd really like some confirmation from someone WITHOUT a financial stake in this before I'm ready to believe this really does cure cancer and save puppies.

Re:Will we finally get a replacement for hard disk

[fuzzyfuzzyfungus]

I don't understand why you would even want to replace them. Mechanical switches are the best thing available in terms of providing input with tactile feedback. The only reason to use anything else is cost or space constraints.

Re:Will we finally get a replacement for hard disk

[Anonymous Coward]

I want to get rid of the last moving parts:
* hard disk
* keyboard
* mouse buttons

What about fans?

I'll belive it when I'm holding it in my hand.

[Anonymous Coward]

Promises promises. It took decades for solid state devices to start to eclipse magnetic storage, as it's just starting to do now. Rom. EEPROM. Battery backed SRAM. Linear flash. AND Flash. SLC NAND. MLC NAND. - I've been reading about mram since I was in middle school. There was a Scientific American article about it. - 20 years ago. (Does anyone still read SA anymore? It seems to have gone to crap. I remember reading about the original research in multi layer optical disks that led to DVDs almost a decade before DVD came to market. Nothing like that anymore)

Anyway, I'll believe it when they have a working product. There's been a lot of pretenders claiming they'll overthrow flash in a matter of a few short years, but so far none have been able to scale up to the density/cost/production volume that flash has achieved.

Re:I'll belive it when I'm holding it in my hand.

[networkBoy]

Between Battery Backed SRAM and Linear (NOR) Flash you forgot: Bubble Memory.

Re:I'll belive it when I'm holding it in my hand.

[CanHasDIY]

Bubble Memory.

Is that the kind that sells address space to bits at sub-prime rates until the whole thing collapses in on itself?

Re:I'll belive it when I'm holding it in my hand.

[mlts]

It will only hit market when there is some new DRM standard applied to it, similar to SD cards with 20% of their capacity set aside for encrypted stuff.

Re:I'll belive it when I'm holding it in my hand.

[Anonymous Coward]

Scientific American sucks now. Try New Scientist.

Re:I'll belive it when I'm holding it in my hand.

[Anonymous Coward]

Anyway, I'll believe it when they have a working product. There's been a lot of pretenders claiming they'll overthrow flash in a matter of a few short years, but so far none have been able to scale up to the density/cost/production volume that flash has achieved.

Fair enough except MRAM != RRAM . Flash is bound to die a painful death sooner or later. The sooner it does the better off we all are.

Re:I'll belive it when I'm holding it in my hand.

[ShanghaiBill]

It took decades for solid state devices to start to eclipse magnetic storage

That was not a fair contest. Because of the development of GMR, HDDs increased in capacity far faster than the Moore's Law rate of solid state. But GMR is now pretty much played out, which should give other technologies a chance to finally catch up. The future of storage is unlikely to be a repeat of the last two decades.

Re:I'll belive it when I'm holding it in my hand.

[SuricouRaven]

Flash works quite well. Not ideal, but good enough. Faster than spinning discs, no significant seek time, durable enough to last the expected life of a laptop now.

Unless you mean the other flash. In which case, death is too good for it.

Re:I'll belive it when I'm holding it in my hand.

[TheRaven64]

Flash has a number of problems. It's byte addressable, but only block erasable, which means that filesystems that deal with it directly are complex and for the rest you need complex logic in hardware to make it appear to be a block device and it then has performance characteristics that are hard to predict and optimise around. Each generation reduces the per-cell reliability at the expense of capacity, complicating wear levelling and requiring more complex controllers.

Flash sucks as a technology for persistent storage. The only reason people tolerate it is that spinning magnetic platters suck even more.

Re:I'll belive it when I'm holding it in my hand.

[VanessaE]

Naw, it's where you keep all those photos of your and your children playing with variously-sized rings and mixtures of soap, water, and glycerine. :-)

Re:Will we finally get a replacement for hard disk

[TheRealMindChild]

You MAY be able to convince me that I could use something eventually that replaces what the mouse currently does. However, you will never get rid of the keyboard. I can type faster than I can talk by a multitude, sometimes even faster than I can think

Re:Will we finally get a replacement for hard disk

[Anonymous Coward]

Just think of the possibilities!

"Ok Glass, search for some hot sweaty gay porn"

Captcha: approval

Hype reserved until I can buy it from Amazon

[Danathar]

Cool announcement.

But...

Given how often we hear researchers exclaiming they've invented the next "Greatest thing (TM)", I'll reserve judgement until I can purchase what comes out of their research.

I'd bet given the patent landscape at the moment that no matter what they have they will be sued for infringement by somebody. It's the way of things today.

Re:Hype reserved until I can buy it from Amazon

[UnknowingFool]

Don't forget pricing. The technology claims to be implemented using current fab technology. That would help keep the price down. But it's $1000 per 1TB, it might be relegated to enterprise usage initially.

Re: Hype reserved until I can buy it from Amazon

[O('_')O_Bush]

Maybe, but 1$/GB is nearly as cheap as the cheapest SSDs, and cheaper than many high end SSDs.

Re: Hype reserved until I can buy it from Amazon

[UnknowingFool]

True unless they only offer 1TB sizes or higher. While it is possible to manufacture them in smaller sizes, it may be cost prohibitive/unwieldy to do so at the packaging level.

Re:Hype reserved until I can buy it from Amazon

[Anonymous Coward]

Amen to that!

If it works, great, and our industry has made another significant advance.

However... remember bubble memory? How about RD-RAM? Then there was that company that made multi-layer CD-ROMs, claiming up to 300 layers? I could go on and on about technologies that were announced and later failed at various stages of the product pipeline.

Re: Hype reserved until I can buy it from Amazon

[stoatwblr]

$1/Gb production cost is NOT the same as $1/Gb on the shop floor. To compete with flash, RRAM will need to be down arounsd 10c/Gb at rollout.

Conspicuously silent on a major question

[Anonymous Coward]

But does it blend?

Re:Conspicuously silent on a major question

[Anonymous Coward]

Mram dust! Don't breathe this.

Dictation versus typing

[sjbe]

Speaking is way slower than typing

Dictation can be quicker than typing in the right circumstances. My wife dictates reports all the time for her job precisely because it is quicker and she is a very good typist. Dictation is however a learned skill (like typing) that takes some practice to become proficient. You have to be able to form complete coherent sentences prior to speaking.

Re:Dictation versus typing

[h4rr4r]

I have some trouble believing this. Does she sound like the micromachines guy?

The issue I have with believing this is not the forming complete sentences before speaking, but the simple fact that speech is slow and error prone. Even two humans from the same region will ask each other to repeat things from time to time. That only even works because there is so much error correction going on. With a machine it only gets worse. Then fixing those mistakes is even slower than fixing a typographical one.

Re:Dictation versus typing

[centipedes.in.my.vag]

I'll add my voice to the dictation working camp. My new phone has a decent VTT texting feature. Occasionally I will need to repeat myself, but it's miles faster than typing on the whole. Voice recognition is getting to a functional point.

Re:Dictation versus typing

[h4rr4r]

Faster than typing on a phone sure, but how is it faster than typing on a proper keyboard? I cannot speak as quickly as I can type.

Re:Dictation versus typing

[DarkVader]

If that's actually true, you're an extreme outlier. Most people can speak much faster than they can type, and that's true for even very fast typists.

Re:Dictation versus typing

[epine]

Then fixing those mistakes is even slower than fixing a typographical one.

If you're writing for the New Yorker, fixing mistakes takes weeks. But then they get into not just whether the noun itself should be in the possessive form, but whether your sentence should require a noun in the possessive form in the first place.

The kind of mistakes you're talking about are specifically keyboard level mistakes. Spelling, orthography, and missing or duplicated words.

It's a tremendous cognitive burden to type at 120 words per minutes of something you're composing on the fly while also getting all the minor details of spelling, punctuation and orthography correct (not to mention getting your homonyms correct which I can usually do at speed if there are two isolated main forms to resolve, but not for palette/pallet/palate or muddles like Seine/sine/sign/sing/singe/singer/signer/seignior/Seigneur/senior/Seniour/senorita where finger habits start seeing double).

If you're not trying to go the orthographic last mile (while neglecting the stylistic last mile) dictation is hugely faster than typing to capture the gist. With dictation, you also get a useful side channel on your emotional inflection and the pacing of your word flow. It's not the same cleaned up and transcribed.

With dictation, one is free to swoop around and really think and make connections and shift and shape and reorganize. If you sit down at a keyboard in that state, you might as well open Mind Manager and type with your mouse.

Back to the actual subject, this is a typical worthless (and breathless) press release. He's sounding the "invest now, or forever be left behind" klaxon. They might be close, or not so close, or we might never see this.

Sure, Seagate could have told people back in the 1980 that they were targeting 1 TB/platter with their fancy magnetic recording technology. But really, with where they were at at the time, there was no connection to where we're at now. It wasn't a better investment in 1980 because we hit 1 TB/platter now. So these "could do" numbers are often exceptionally worthless, even when true.

Re:Dictation versus typing

[afidel]

I doubt it, plenty of people can type 80-100 wpm but there are very few people that can talk that fast (micromachine guy is an extreme outlier)

Re:Dictation versus typing

[TheRaven64]

Most people can't dictate at anything like their normal talking speed, however.

Re:Dictation versus typing

[tibit]

not to mention getting your homonyms correct which I can usually do at speed if there are two isolated main forms to resolve, but not for palette/pallet/palate or muddles like Seine/sine/sign/sing/singe/singer/signer/seignior/Seigneur/senior/Seniour/senorita where finger habits start seeing double

That is very interesting - I never imagined that anyone who knows the spelling of those various homophonic forms would have any issues typing them correctly each and every time, save for random typos. People are complex beasts :)

Re:Dictation versus typing

[flargleblarg]

It really depends what you're typing. I guarantee you there is not a single person on the planet who can speak C or C++ code as fast as it can be typed, even by a novice typist. If you disagree, well, then consider Perl code instead. Case closed.

Re:Will we finally get a replacement for hard disk

[Anonymous Coward]

Remove the user, that's the biggest moving part in a system!

Re:Will we finally get a replacement for hard disk

[Anonymous Coward]

Two words:

Peltier chips.

Re:Will we finally get a replacement for hard disk

[Anonymous Coward]

Direct brain interface.

Re:Will we finally get a replacement for hard disk

[Anonymous Coward]

* fans, do not forget about fans...

I'm confused

[squiggleslash]

Genuinely: what's this about regular old Flash being unable to store data for more than a year or three? Have I seriously misunderstood or is this a real problem I've been extremely lucky to avoid thus far?

Re:I'm confused

[h4rr4r]

I believe they mean unpowered and on the shelf. That would allow this new stuff to be used in place of backup tapes.

Re:I'm confused

[networkBoy]

unpowered the flash cells will leak electrons off their floating gates (powered too if the device doesn't do some sort of maintenance cycle). with as few as 100 electrons making the difference on a cell...

Re:I'm confused

[Alioth]

But all the flash devices I've used have a retention period of at least 20 years (disclaimer: I'm thinking of flash ROMs and CPLDs and SPI flash for FPGAs, but the way they store bits is the same as a USB flash drive). I've never seen any as short as 1-3 years.

Re:I'm confused

[squiggleslash]

OK, to clarify, are we talking about the same technology used for, say, SD Cards? And if so, is there a serious risk that an SD card left in a box or on a shelf for an extended (say, half a decade) period of time will actually lose some or all of its contents?

Re:I'm confused

[AmiMoJo]

They are not cheap NAND flash though. That's why they cost so much, relatively speaking. An Atmel Dataflash 1MB chip costs about $1, which is $1000/GB.

Re:I'm confused

[marcosdumay]

Yes, and yes. I have no idea if it half a decade, or two decades, but that's the timeframe.

HDs also lose their data, but that takes a lot longer.

Re:I'm confused

[AdamHaun]

Genuinely: what's this about regular old Flash being unable to store data for more than a year or three? Have I seriously misunderstood or is this a real problem I've been extremely lucky to avoid thus far?

I only know about embedded NOR flash, but in that case the rated lifetime is after the max number of write/erase cycles with storage under worst-case conditions on the worst units to come out of the fab. Note that commodity NAND flash is heavily dependent on ECC, so the spec number might not reflect the true lifetime of the bits themselves. At reasonable temperatures and usage patterns with a more typical unit, the data will likely last much longer.

But again, I haven't seen anyone's internal NAND reliability data, so take this with a grain of salt, and always back up your data.

Re:I'm confused

[Anonymous Coward]

>But all the flash devices I've used have a retention period of at least 20 years

Nope, 10 years is the longest claim I've seen for consumer grade flash. And I wouldn't care to take bets on that.

>I'm thinking of flash ROMs and CPLDs

You're seriously comparing SLC flash optimized for reliability versus high-density, consumer grade TLC flash?

Re:I'm confused

[Anonymous Coward]

NOR flash is usually used for code and has much better retention than NAND flash which has much lower retention.

Re:I'm confused

[jones_supa]

Yes, that is true. That's why I still save my long-term archived stuff on optical media.

Re:I'm confused

[PRMan]

I found an 8 MB(?!?) SD card from my first camera the other day and I could still read the pictures. I'm pretty sure that SD card has not been read in about 10 years, but I was still able to read it. And I can still read EVERY CD that doesn't show obvious physical damage, and they were written with one of the first 2X CD writers in about 1996.

Re:I'm confused

[tibit]

It's not just a serious risk, you can pretty much count on it. SD cards are not meant for permanent data storage when they are not powered. When powered, the built-in controller does patrol reads and reclaims pages that go weak before you lose any data. An unpowered SD card is really meant to be used to move data between things, not to store it for any prolonged periods of time. It's simply not designed for that.

Re:I'm confused

[unixisc]

>But all the flash devices I've used have a retention period of at least 20 years

Nope, 10 years is the longest claim I've seen for consumer grade flash. And I wouldn't care to take bets on that.

Microchip's NOR flash products - originally from SST, which they acquired - uses a split-gate cell which guarantees 100 years of data retention. The 10 years that you've seen (from the likes of Numonyx/Micron/Intel and Spansion/AMD) are from stacked gate cells.

>I'm thinking of flash ROMs and CPLDs

You're seriously comparing SLC flash optimized for reliability versus high-density, consumer grade TLC flash?

SLC flash wasn't optimized for reliability or anything - it just happened to be the first flash that came out - and you're probably thinking about NOR flash here. NOR flash was what first debuted as serial E2Proms and later as flash. NAND flash was the high density, consumer grade flash that followed, as a recordable media. MLC flash - both NOR and NAND - followed much later.

But NAND flash is just 2-4 times higher in density than NOR. So once your flash hits a certain process node, it will be just as cheap to have SSD devices out of NOR flash, and then, one won't need to worry about ECC or any of that other stuff - everything would then be code grade.

Re:I'm confused

[networkBoy]

NOR will always cost more than NAND, about 4 times more.
NAND is 4 times denser, thus you fit four times as much memory per square mm of die space.

Re:I'm confused

[networkBoy]

That old card likely used SLC, NOR, and large gates, each of which contributes to data retention (and higher cost per bit).

Re:I'm confused

[Anonymous Coward]

Dyes in optical media fade. you'll probably find all your long-term archived stuff isn't.

HDDs, tapes and printed paper are the only real options for longterm archiving (And esoteric things like electron beam writing on gold).

Digital replication is better, and with the continuous growth of storage density and comensurate drop in cost, it on average only costs root 2 as much to keep all your data online and replicated as it does to only store recent data online.

Re:I'm confused

[unixisc]

Yeah, 4 times, but as your die keeps shrinking and reliability issues become more critical, vendors will look for a sweet spot b/w reliable & cheap, and that is where NOR could come into play. Only thing inhibiting them would be their write speeds - something like the page programming that used to be there in E2PROMs may make a comeback to alleviate that problem.

Re:I'm confused

[Anonymous Coward]

HDDs do take longer, unless exposed to strong magnetic fields. However I've seen mechanical failures with hard drives left unpowered for extended periods of time -- as little as 2-3 years. I'm being told that any of the moving parts can fail after extended periods of inactivity. This doesn't mean the data is lost but it's quite expensive to recover it (I didn't care about it enough to pay for recovery).

Re:I'm confused

[Anonymous Coward]

Optical media is VERY unreliable for long term data storage. Unless you use m-disc which hasn't been around for long and is quite a bit more expensive.

Re:I'm confused

[stoatwblr]

Those devices have LARGE flash cells compared with flash cards or SSD. Losing a few hundred electrons doesn't matter. Losing a few hundred electrons at 22nm means the difference between 00/01/10/11

But...

[DoofusOfDeath]

is it agile???

Re:But...

[Anonymous Coward]

No, but it is scrRAM.

Re:Will we finally get a replacement for hard disk

[somersault]

Does your phone have a fan? How about your tablet? How about Ultrabooks..?

CEO information

[GoNINzo]

Looks like the CEO has 3 patents, one for portable storage, one for non-volatile memory, and one for a memory controller. http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO2&Sect2=HITOFF&u=%2Fnetahtml%2FPTO%2Fsearch-adv.htm&r=0&p=1&f=S&l=50&Query=IN%2FMINASSIAN-GEORGE&d=PTXT So who knows, could be legit.

Re:Will we finally get a replacement for hard disk

[TheLink]

What about fans?

I thank you all!

*bows*.

the size of a postage stamp

[TerraFrost]

Micro SD cards are already quite a bit smaller than postage stamps lol.

Re:the size of a postage stamp

[niftydude]

Yeah but you need 16 of the 64GB ones to make 1TB

Re:the size of a postage stamp

[CanHasDIY]

Micro SD cards are already quite a bit smaller than postage stamps lol.

Are there any with a terabyte of storage space?

Re:the size of a postage stamp

[Anonymous Coward]

micro sd cards are also not 1tb capacity devices. Largest one I have is 8gb, I'm sure you could get 16, maybe even 32gb into that form factor, but over 1000gb? so how many 32gb micro sds can you fit in the same volume as these new RRAM devices?

Re:the size of a postage stamp

[UnknowingFool]

Can they hold 1TB of data for 20 years without power?

Re:the size of a postage stamp

[Anonymous Coward]

And this is a whole lot faster than crap slow SD Cards.

Re:Will we finally get a replacement for hard disk

[mark-t]

You exaggerate. You must think in order to direct your fingers to type the keys that they do... this requires a type of deliberate intent that could, at least in theory, be picked up by devices designed to scan brain activity, and if the issues regarding understanding how the brain waves correspond to what, exactly, is being thought about can ever be worked out, such an apparatus could probably improve your throughput by more than an order of magnitude. The amount of time it takes for signals to reach your fingers from your brain alone is staggeringly slow compared to the speed of electronics.

Re:Will we finally get a replacement for hard disk

[Lawrence_Bird]

Have you had a SSD fail? Not so great

Re:Will we finally get a replacement for hard disk

[h4rr4r]

My ultrabook sure does.
Most of them do, just very well hidden. The Macs for example vent right infront of the screen hinge. The Dell ultrabooks vent right out the back.

Touch screen fanboys

[tepples]

Mechanical switches are the best thing available in terms of providing input with tactile feedback.

Until the fanboys come out and claim that tactile feedback is overrated, that platformers, fighters, and other video game genres whose control relies on tactile feedback are outdated genres, and that video game developers should just accept this and design their games around the lack of tactile feedback. Every Slashdot discussion of the handheld gaming market seems to bring out fanboys on both sides: fanboys for the mechanical switches in PlayStation Vita or Nintendo 3DS and fanboys for the flat sheet of glass on smartphones.

Re:Touch screen fanboys

[Zmobie]

Interesting this comes up. Over the weekend at Quakecon John Carmack addressed the evolution of controllers and his thoughts on the subject in his keynote. My favorite was when he was mocking Apple having the one button mouse and then one upped that by saying the kinect was "like a 0-button mouse."

My main argument would probably be, when you have a physical key/button with tactile feedback you can much more easily ensure an intentional action on the part of the user. Whereas touchscreens you are much more prone to fat fingering the wrong key (although this happens sometimes on keyboards too to be fair, just not nearly as often in my experience). A promising tech I remember reading about I think on slashdot was the touchscreen that could bend in such a way as to give tactile feedback at any point of the screen for a variety of sizes and shapes. Arguments could still be made to have solid keys/buttons still, but brings the two a little closer together at least.

Re:Touch screen fanboys

[Anonymous Coward]

Interesting this comes up. Over the weekend at Quakecon John Carmack addressed the evolution of controllers and his thoughts on the subject in his keynote. My favorite was when he was mocking Apple having the one button mouse and then one upped that by saying the kinect was "like a 0-button mouse."

Which just shows how out of date his jabs are. Apple mice haven't had buttons at all for years. They use multi touch gestures and detect among other things: left click, right click, one finger scroll, and two and 3 finger swipes. And before that they were two button mice with a scroll ball (2-d scrolling nit juts a 1-d wheel), and supported a squeeze action in addition to the standard left/right click). The single button mouse died around the same time CRTs were removed from iMacs.

By comparison the standard 2 button and wheel mouse that comes with most off the rack PCs, is fairly quaint.

Re:Touch screen fanboys

[pipatron]

Then there's actually computer users, clutching the real, full-size keyboad in fear while glancing on Ubuntu's and Windows' rapid and rampant move from sanity to the insane "everything is a touch screen" interface, which no computer user ever asked for but seems oh-so-trendy if you're a marketing drone who mostly use the computer for playing games on facebook or following blogs.

That's probably the longest sentence I've ever written, btw.

Re:Touch screen fanboys

[Score Whore]

Apple's Magic Mouse is a one button mouse. Literally. The entire top surface depresses a mechanical switch. The fact that it's also a multi touch surface doesn't change that.

Re:Touch screen fanboys

[davester666]

Actually, both the magic mouse and the magic trackpad have buttons, they are just not as clearly visible to the user as 'regular' mice.

The body of the magic mouse is actually split in two, and when you press down to do either a left or right click, the upper half of the mouse moves and depresses a physical button that you can feel and hear.

The magic trackpad has two buttons which are located on the underside of the pad, disguised as the front two rubber pads that hold the trackpad in it's place. Again, pressing on either the left or right front of the trackpad results in a physical button being depressed that you can feel and hear.

Re:Touch screen fanboys

[Anonymous Coward]

The notches and the feel of the longer keys on a keyboard also makes sure you know where your hand is. On a touchscreen keyboard, your hand could easily be shifted an entire key and you wouldn't realize it until you start typing nonsense. Additionally, if your hand somehow did get shifted on a tactile keyboard, you can fix it without looking at the keyboard based purely on the feel; this is impossible with a touchpad keyboard, you need to look.

Re:Touch screen fanboys

[gl4ss]

look man, if you can't press the two buttons at the same time independent of each other then they aren't two buttons. that the whole mouse is a button(and that's what it is despite you claiming otherwise) is just a crapp design choice. they just wanted to sell a nice rendering.

carmack was totally spot on. remember that sweet big xbox mecha controller and game? oh they made a sequel to that game - with fucking kinect! and it's totally like carmack said: 0 button mouse time.

those apple mice are horrible to use...they're actually so horrible to use that apple makes a killing selling touchpads for desktop pc's - they suck slightly less, even for photoshop! next up, "ipad is actually a 3000 key keyboard".

Re:Touch screen fanboys

[Anonymous Coward]

I have never seen anyone who preferred touchscreen controls over physical buttons for those types of games. I think you are mistaking people who simply make do with what they have as fanboys. Let's face it, gamepad style controls are not going to ever become popular on smartphones or tablets for many reasons, among them size, weight, cost and dubious added value. Smartphone games are generally played to pass time, not because people really get into them as they would a PC or console game.

Re:Touch screen fanboys

[Anonymous Coward]

I find the magic mouse great, and don't know anyone who thinks they're "horrible". Sure you can't "click" both buttons at once, but there's nothing stopping you from left-clicking and right-tapping or swiping at the same time. There are a lot of stock gestures in OS X that differentiate between taps and clicks, things that are impossible with a shitty old mechanical mouse.

They sell trackpads because the size allows the user to do even more gestures than the mouse.

Re:Touch screen fanboys

[Belial6]

The sad thing is that touch screens are actually a really good idea, and should have made it into general computing years ago. Unfortunatly, the way the get implemented is to try and replace a mouse, or worse yet a keyboard. Trying to replace a mouse with a touch screen makes about as much sense as trying to replace a keyboard with a mouse. Touch screens are a third input device. Not a replacement for the current two.

Re:Touch screen fanboys

[Anonymous Coward]

You're an idiot. The fact that Apple's Magic Mouse has only one physical button is irrelevant. It feels and behaves exactly as if there were two physical buttons — and the software/OS doesn't know the difference. Everything works transparently. For all practical purposes — which is what matters — it is a two-button mouse.

Re:Touch screen fanboys

[Anonymous Coward]

look man, if you can't press the two buttons at the same time independent of each other then they aren't two buttons.

Why on God's Green Earth would you want to do that? What software requires that?! I've never seen that.

Re:Touch screen fanboys

[Score Whore]

Really? Then do this: Click the left button and hold it down. Now click the right button and hold it down. Release the left button. Then release the right button.

I have one of these, it came with my iMac. I have a MacBook Pro. I bought the magic touch pad for the iMac. I have some fucking idea what I'm talking about.

Re:Touch screen fanboys

[LordWabbit2]

Clearly you don't like a clean screen. The last thing I want is fingerprints all over my monitor. I am constantly wiping down my tablet. When work pig's walk up to me eating taco's and then insist on jabbing at my screen like retards instead of saying "line 256" I want to beat them to death with my chair. Touch screens are a fvck up, until we stop secreting oil they will always be a fvck up. I need to see what's on my screen, not misread an 8 as a 6 because of crap on my screen.

Re:Touch screen fanboys

[Cederic]

Games, regularly. Other software, less frequent.

Re:Touch screen fanboys

[The Mighty Buzzard]

If a physical switch is in proper working order, it has an error rate of zero. Touch-sensitive surfaces? Not even remotely close. That will never change until every user is able to provide input with zero deviation every single time, which is not humanly possible without feedback.

Re:Touch screen fanboys

[Zmobie]

Yea, see you should probably have looked at more context if you planned to insult what he was talking about. Even without hearing the rest of the keynote, the Kinect comment illustrates exactly the problem. Mechanical buttons are still far superior for user actions than trying to virtualize the buttons via touchscreen or whatever bullshit apple is trying to pull. The Kinect is the same idea as their stupid magic mouse only a bit worse, where you have to make some exact action gesture combination as opposed to a mechanical button "one trick pony" that has two states: pressed or not pressed. That mouse does not have a binary state because of the software interpretation layer trying to translate it to a binary state when it has more cases than that.

That isn't to say that these could become better controls, and Carmack himself even said with better accuracy, less latency, and taking fuzzy cases into account this could become fairly viable, but right now it is still a damn joke.

Re:Touch screen fanboys

[Belial6]

You are a fringe case. Having 'pigs' walk up to your monitor eating tacos and jabbing at your screen is not normal. It is likely that you elicit that specific behavior somehow. I offer every single ATM as evidence of my point. While the screens tend to be fairly scratched up, I have yet to see one with food on it. My guess is that you haven't either. These are touch screens that will see many orders of magnitude more people using them than your monitor ever will, and will have be seen by the users as being owned by a faceless corporation instead of someone they might want to avoid upsetting.

The only conclusion that can be made based off your statement and the empirical evidence is that either you are massively exaggerating the problem. You are over sensitive. Or, people are trying to piss you off specifically.

Re:Touch screen fanboys

[omfgnosis]

It's used in Minesweeper (at least the old versions that were worth a damn) to clear the tiles surrounding your cursor. That is the only use I can think of.

(You can also use a "middle" or third button, but the finger travel can slow you down depending on your muscle memory.)

Re:Will we finally get a replacement for hard disk

[Anonymous Coward]

More like
"Glass, check forecast for rainstorm"
'[OK, searching for cast of gay porn]'

Re:Will we finally get a replacement for hard disk

[h4rr4r]

How is that different than any other storage device failing?

This is why backups exist.

Re:Will we finally get a replacement for hard disk

[gfxguy]

And speaking an email, text, or slashdot comment is one thing; speaking C++ or formulas in a spreadsheet is something entirely different. Moreover, can you imagine a team of programmers in an open style office all talking code to their computers?

Re:Will we finally get a replacement for hard disk

[Impy the Impiuos Imp]

> I want to get rid of the last moving parts

That's what she said!

Re:Will we finally get a replacement for hard disk

[h4rr4r]

I am not suggesting you should. I am typing this on a keyboard likely older than many other posters. My company now has IT employees younger than this keyboard.

Re:Will we finally get a replacement for hard disk

[NoImNotNineVolt]

I can type faster than I can talk by a multitude, sometimes even faster than I can think

A quick look through YouTube's comments section reveals that you're not alone.

Re:Will we finally get a replacement for hard disk

[Charliemopps]

You mean a tablet? Cause those are all over the place.

Re:Will we finally get a replacement for hard disk

[Anonymous Coward]

I have fallen asleep during the middle of an online conversation - the results (yes, I kept typing) are truly fascinating, yet will not be revealed here lol

Chrome for Android misbehaves in low memory

[tepples]

Chrome on my tablet also likes to reload any browser tab that isn't frontmost, causing me to lose scroll position, state of collapsed and expanded divs, and (worst of all) text entered into a form. And sometimes it can't load the page again until I get to the next Wi-Fi hotspot, as it has expired from cache. I've read that Chrome does this because it purges the DOM for all tabs that are not currently visible when Android notifies Chrome of memory pressure. Strangely, Chrome on my netbook, which has about the same amount of RAM and the same kernel (Linux), manages without having to purge pages.

Now how is this related to the fact that my tablet lacks a fan? Some things commonly cooled by a fan are hard disks, RAM, and especially the CPU. Now there are time-memory tradeoffs in keeping a DOM open. The browser could somehow compress the DOM for pages in the background, but it doesn't because that would take more time. It could swap the DOM to storage, but tablets have far smaller storage because far smaller storage runs fanless.

one to three years?

[ZorinLynx]

I have SD cards and USB thumb drives far older than 1-3 years and can still read the ancient data that was on them just fine.

Where did this "1-3 years for NAND flash" figure come from? It's a bit concerning.

Re:one to three years?

[AmiMoJo]

Some really cheap types of NAND used in SSDs do have that limitation, but the controller just re-writes the data periodically to refresh it. Of course if you leave it on the shelf for 3 years you might be screwed.

In fact HDDs do it too, or at least used to. My old Seagate drives used to periodically start ticking as the drive did a surface scan and read/rewrite cycle, during which bad blocks would also be remapped. All my newer drives are in a NAS now so I wouldn't notice if they still did it.

Re:one to three years?

[UnknowingFool]

I think that has to deal with guarantees and mean lifetime. Some HDs have last 20+ years but none of the manufacturers will ever guarantee that. NAND flash will leak over time and 3 years without any power is the max that they can be assured data will be held. Some will last longer but the manufacturers will never guarantee it.

Re:one to three years?

[Anonymous Coward]

Has it been powered for any of that time? Have you checked it for bit errors? Does it use error correction? How many such flash chips have you checked after that time? Its a probabilistic failure mode for individual bits. Also, flash from 3 years ago generally is bigger and longer lasting that the current stuff.

Re:one to three years?

[gman003]

That's the standard warranty period on SSDs. SSDs often get much more written to them (dozens of gigabytes per day, in some cases), so they can run through their limited endurance much faster, even though they generally use more-durable MLC instead of TLC. With care, they can last much longer, but if you're using them as swap or something, it can run out after a year or two.

Re:one to three years?

[Anonymous Coward]

Do you compare the data to something else to ensure what you've read is what you think you had ?

Re:one to three years?

[ZorinLynx]

They were full of mostly images, so corruption would be obvious.

Re:one to three years?

[Anonymous Coward]

Be careful with SD/USB drives. The memory on them is known to be VERY unreliable, like orders of magnitude less reliable than typical storage devices ( HDD's, desktop SSD's, etc).

The guys from anandtech have mentioned this several times in their podcasts. I'm very cautious to backup my data from cameras, etc immediately now after taking them.

Good news everyone!

[Gravis Zero]

a few notes:
- RRAM (aka ReRAM) is memsistor based RAM
- super simple design
- requires less power (lower voltage too) than FLASH and racetrack memory.
- 10ns switching (faster than DDR some DDR RAM)
- 1 trillion write operations according to US startup Crossbar
- possibly scaled down to 2nm (when they invent the manufacturing process)

so if this really works out, it may be a replacement for RAM and FLASH memory in lots of stuff. i'm not sure if this includes computers but at the very least, it could be used to retain data on RAM sticks (hopefully directly on them) when you turn off your PC.

Re:Good news everyone!

[Anonymous Coward]

Memristor RAM are heavily developed by various companies, especially HP which owns tons of patents. This is very interesting technology and potencially can replace both flash and dynamic RAM thanks to simple design (and high densities) with very good speeds (potentially faster than recent dynamic RAM). If this startup did not come with some break-through technology I would be sceptical that they can compete with the big guys - I am expecting first memristor based chips starting to replace flash not sooner than in 5-10 years.

Re:Good news everyone!

[unixisc]

What is it made up of? Silicon, or a different material? How abundant is that material, when compared to Silicon? This sounds too good to be true - non volatile, very high switching times, lower voltage & power, et al. So what are its drawbacks, other than potentially being more expensive, at least initially?

big deal, I have unlimited memory

[swschrad]

write-only memory has an infinite density.

Re:big deal, I have unlimited memory

[Anonymous Coward]

Yeah, but personally what impresses me is the bandwidth and latency!

Re:big deal, I have unlimited memory

[Anonymous Coward]

I've been running "dd if=/dev/zero of=/dev/null" for 3 years straight and I don't know what is more impressive, that the input just keeps on coming or that the output can handle it all.

Re:big deal, I have unlimited memory

[Anonymous Coward]

write-only memory has an infinite density.

Yes, the WOM rules!
Link to the official data sheet: WOM Datasheet

Re:big deal, I have unlimited memory

[Anonymous Coward]

write-only memory has an infinite density.

And the datasheet is awesome: http://www.ganssle.com/misc/wom1.pdf

Re:big deal, I have unlimited memory

[Anonymous Coward]

It really doesn't, at least if it's made, *ahem* properly. Just because you can't read it doesn't mean it isn't stored.

Re:big deal, I have unlimited memory

[Anonymous Coward]

a.k.a /dev/null

Re:big deal, I have unlimited memory

[Anonymous Coward]

infinate storage on /dev/null

Re:Will we finally get a replacement for hard disk

[ShanghaiBill]

Have you had a SSD fail?

No, I have not. Which is why I think the claim in the summary that NAND flash will fail in "the standard one to three years" is BS.

In 15 years. Maybe.

[Reliable Windmill]

I can't wait until this is available in 15 years or never, whichever is sooner, as is usually the case with all these amazing breakthroughs we read about every day on the Internet.

Re:In 15 years. Maybe.

[Anonymous Coward]

I can't wait until this is available in 15 years or never, whichever is sooner

15 years will be sooner than never, I can assure you of that.

Re:In 15 years. Maybe.

[Belial6]

Dude, did you miss the last 30 years of computer advancement? Get this, you can fit a computer in your house now.

Re:Will we finally get a replacement for hard disk

[Anonymous Coward]

Unless you have a super huge heatsink or a regular sized one with a fan that Peltier isn't going to do you much good.

Re:Will we finally get a replacement for hard disk

[Anonymous Coward]

With (at least in the previous generation) the controller of SSD it is a lot worse.
With harddisks the are first soft faults, later there are hard faults, but often most data can be retrieved from a disk.
With SSD, from one second to the next the whole thing will not work anymore, you can't even read from it anymore, nothing.

Talking is faster than typing (for most people)

[sjbe]

how is it faster than typing on a proper keyboard? I cannot speak as quickly as I can type.

You probably can and do speak significantly quicker than you type unless you have some sort of speak impediment. Most people can comfortably speak at around 150 words per minute which is far faster than most can type. Dictating however does take some practice so you quite likely would be slower at first until you get comfortable dictating.

Re:Talking is faster than typing (for most people)

[armanox]

Well, I have a speech impediment, while not to the degree that it was when I was younger, I have a stuttering problem, especially when nervous, upset or angry. That would be horrible for dictation (now if only I was a good speller I might be ok).

Re:Talking is faster than typing (for most people)

[dj245]

how is it faster than typing on a proper keyboard? I cannot speak as quickly as I can type.

You probably can and do speak significantly quicker than you type unless you have some sort of speak impediment. Most people can comfortably speak at around 150 words per minute which is far faster than most can type. Dictating however does take some practice so you quite likely would be slower at first until you get comfortable dictating.

If everyone dictated, the noise in most offices would be unbearable.

Hell, if ONE person dictates all the time, it is unbearable.

Re:Talking is faster than typing (for most people)

[michael_cain]

Yep. Back when Gates was still running Microsoft and on his voice recognition bandwagon, one of my coworkers often remarked, "You can tell how long since he worked in a cubicle farm!"

Re:Talking is faster than typing (for most people)

[Anonymous Coward]

Most people can comfortably speak at around 150 words per minute which is far faster than most can type.

Indeed, most people are mediocre typists at best. Yet, use a keyboard a lot and soon you will be able to easily type much faster than you can speak.

Re:Talking is faster than typing (for most people)

[msobkow]

However, I do not speak "code". Do you have any idea how long it would take to say "public int f( int x ) { return( x * 2 ); }". Let's see:

Public int letter-f left-paren int x right-paren left-brace return left-paren x star two right-paren semicolon right-brace

Yeah, that would work *real* well. *LMAO*

Pair with SATA Express (NVMe or PCIe) and WHAM!

[Anonymous Coward]

If and when there is an actual product, move over emerging SATA Express and PCIe drives... And yay, I'll now have plenty of storage that's fast enough to capture videos on my future 4K phone. Of course I'll need a new quantum computer so I can manipulate my 4K videos in real time, and render a 2 hour UHD video in under 10 minutes. Ah, pipe dreams... Gotta love them!

RAM data retention

[swb]

So not only will they sell new computers without a Windows install disc, they won't even install it on a disk drive, it will be preinstalled in RAM and all you have to do is turn it on.

Although it is kind of an interesting idea to consider a computer where there is no distinction between mass storage and RAM, where RAM is rewritable but permanent.

You could even leave programs in a running state but just stop executing them on the CPU. You could install new software in an already-running and configured state (how's that for a backup?).

Re:RAM data retention

[Anonymous Coward]

The problem I see with all that is that a design like that doesn't work very well with current operating systems and programs. How many even supports the XIP (eXecute In Place) feature of PCMCIA?
I feel that the support will be very gradual or very hackish.

Re:RAM data retention

[Anonymous Coward]

Wow, preinstalled in RAM (functioning like ROM)... just ike a Commodore 64 (or Apple II, etc). What a technological advancement!

Re:RAM data retention

[Anonymous Coward]

Yeah, no more CDs, DVDs, or hard drives! Programs could just come in cartridges which could be plugged into a hardware console!

Re:RAM data retention

[jeffb (2.718)]

Although it is kind of an interesting idea to consider a computer where there is no distinction between mass storage and RAM, where RAM is rewritable but permanent.

That is an interesting idea. Maybe you could call it Multiple-Capability Storage, or Multics for short.

Re:RAM data retention

[CCarrot]

So not only will they sell new computers without a Windows install disc, they won't even install it on a disk drive, it will be preinstalled in RAM and all you have to do is turn it on.

Although it is kind of an interesting idea to consider a computer where there is no distinction between mass storage and RAM, where RAM is rewritable but permanent.

You could even leave programs in a running state but just stop executing them on the CPU. You could install new software in an already-running and configured state (how's that for a backup?).

Hmmm...makes me wonder what types of new and exciting malware and viruses will be making the rounds...

Re:RAM data retention

[Rogue Haggis Landing]

So not only will they sell new computers without a Windows install disc, they won't even install it on a disk drive, it will be preinstalled in RAM and all you have to do is turn it on.

Someone else has noted that the Commodore 64 and Apple II and other computers of that vintage had this feature. A computer as late as the Mac Classic (released in 1990) also had this feature. You could boot the system software off of a hard drive or a floppy, but without these it would boot from a ROM to System 6.0.3. It was really great -- it went from off to a working desktop almost instantly. A return to something like that would be very welcome.

Re:RAM data retention

[Terwin]

How do you 'Reboot' windows if not even unplugging the machine can re-set your state?

This will cause every windows Help-Desk script to be re-written, and I can only hope that the new script does not change to 'go out and buy a new machine'.

Re:RAM data retention

[swb]

Well, pre-installed ROMs aren't exactly the same thing. Apple ][s would run a flavor of BASIC from ROM, but software was executed from RAM. It was also immutable, so any variables or data had to be stored in RAM and were subject to loss with power. DOS required booting from media, although I'm sure somebody figured out a way to pack DOS into maybe INTBASIC and burn to an EEPROM that would be the active boot ROM in slot 0.

I don't recall any of my early Macs (starting with the Mac Plus) bootable to Mac OS in ROM, although large parts of what we'd now call system libraries (Mac OS Toolbox) were in ROM but were commonly relocated to RAM with patches and upgrades. I still think they required booting from an OS boot media and I don't think Finder was in ROM.

There was a ROM monitor, but I don't think it did much besides let you view memory locations and possibly issue a "goto $hex" type command to continue execution of whatever was addressable.

Re:RAM data retention

[swb]

I can only guess, but I would imagine that a system like this would partition its storage into a "system" area and have a power-on sequence that would run an "initialization" sequence that would blank the system area and then re-load it if desired.

Re:RAM data retention

[CannonballHead]

Isn't a clean reboot primarily a software thing? If you have to turn the power on and off, you're resetting it, not rebooting it.

And anyways, even if power on/off doesn't "reset the RAM" (clear it, whatever word you want to use), presumably you'd be able to tell the controller to do that. I can't "reset" my hard drive state by turning the power off, but I can format it (either just replacing the filesystem info or by actually rewriting it)

Re:RAM data retention

[BitZtream]

Congratulations, you've just described how apps work on Palm Pilots (the ones before the phone).

And lots of other platforms too for that matter, but Palm was the only one I could think of recently. Its not a new concept. Its horrible from a developer perspective however. Developers suck. Restarting helps deal with that suckage by clearing the slate.

Re:RAM data retention

[Areyoukiddingme]

I had the same question the last time memristors were brought up on Slashdot.

Short answer: Linux supports execute in place, and has for some time.

And really, what more do you need to know? It's the Year of the Linux Palmtop, after all.

Re:RAM data retention

[Rogue Haggis Landing]

I don't recall any of my early Macs (starting with the Mac Plus) bootable to Mac OS in ROM, although large parts of what we'd now call system libraries (Mac OS Toolbox) were in ROM but were commonly relocated to RAM with patches and upgrades. I still think they required booting from an OS boot media and I don't think Finder was in ROM.

The bootable ROM was something exclusive to the Mac Classic. From Low End Mac:

A feature unique to the Classic is the ability to boot from ROM by holding down command-option-x-o at startup. The ROM Disk is called "Boot Disk" and is 357 KB in size. The ROM Disk uses Finder 6.1.x and System 6.0.3 -- this combination is specifically designed for the Classic. The only control panels are General, Brightness, and Startup Disk. MacsBug and AppleShare Prep are also part of the System, which loads into 294 KB of the Classic's RAM. Because this is in ROM, there is no way to add anything to the ROM Disk.

It was a neat little feature. The machine was underpowered for its time, but it had this one thing that nothing else did. (And sometimes I still miss the compact form factor).

Re:RAM data retention

[gbjbaanb]

Although it is kind of an interesting idea to consider a computer where there is no distinction between mass storage and RAM, where RAM is rewritable but permanent.

don't give them ideas, Windows already takes up many, many gigabytes of storage, and .NET apps take up many many many megabytes of memory... once you tell them they have a terabyte of RAM to use, they will do so! And then we'll all have to go and buy another memory stick just to run 2 of their apps simultaneously. :-)

Re:RAM data retention

[Anonymous Coward]

This approach was actually used in the original, "classic" Palm OS. Applications are stored in the battery backed RAM or ROM and are executed in-place.

Re:RAM data retention

[kermidge]

Yes.

Re:RAM data retention

[Anonymous Coward]

The glorious future of batch processing is upon us!

Re:RAM data retention

[Anonymous Coward]

This sounds like core memory. Everything old is new again.

My favorite old-timer story is about the time the power went out when I was in the machine room compiling a Fortran program on the DG mini. Dark, sound of the disks spinning down, people clunking around for flashlights. Power came back on, the machine came back up, and my compile finished without error. This was back in the 80's, guys. No battery backup, no UPS, but memory was non-volatile, and some OS's knew how to restart after a power fail. (This was RDOS, I think. Kernel had an entry point called "jehov" commented "Create the world.")

Cheap semiconductor RAM ended core memory, but now it might be coming back.

Re:RAM data retention

[msobkow]

Yes, but a reboot forces a complete OS reset including all memory and variables.

Persistent memory would require that people write specific routines to reset the variables, and that's as fraught with risk of errors and omissions as C/C++ memory allocators are.

Somehow, somewhere, you need to have an "initial state" memory image for a program that can be reloaded in order to keep resets simple and relatively error free. Whether that's in a RAM disk or on a physical disk makes no difference -- the key is that the OS has to be able to completely terminate a process and reload the known state.

There is also the issue of inter-program and inter-library dependencies that get resolved during the load process. This idea of having a complete image of an initial state for a program would presume that there is no need for a linker, which would imply there is no way to update libraries to apply security fixes, for example.

Re:RAM data retention

[Gravis Zero]

or we could go with a name that doesnt suck like Memdisk.

Re:RAM data retention

[dublin]

I'm old enough to have seen this done in moving a running program from one machine to another. The computers involved were old (even then) DEC PDPs (I don't remember whether they were 8s or 11s), and the program was running on a 8KB (IIRC) magnetic core memory card - literally, a thousands of little tiny magnetic toruses strung on what seemed like billions of hairlike laquered wires.

It went like this:
1 - Program running on computer one
2 - Halt computer one (stop program counter), pull out memory card, carry across building to other computer. (The bits are magnetic, big, and need no power to persist for ages - if you wanted to store the running program on a shelf for a while here, that would have worked, too)
3 - Replace core memory card from computer 2 with the one from computer 1
4 - Resume program counter, and running program continues exactly where it left off in left computer 1! (Much faster resume than Windows or even OSX!)

Re:Will we finally get a replacement for hard disk

[TheRaven64]

Not yet, although I've only got a small number of machines with SSDs as yet. Over the last 15 years, however, I have had a lot of hard drives fail, some with no warning at all.

Yes talking is faster than typing

[sjbe]

I have some trouble believing this. Does she sound like the micromachines guy?

Listen to physicians dictate some time, particularly pathologists or radiologists who tend to do a lot of it. Yes, they often do speak very quickly - 200+ words per minute is not rare. Once they get good at it, they can easily rattle off a report in far less time than they could possibly type it no matter how quickly they type. Transcriptionists normally have to slow the recording down while typing to get what is being said. It's very easy to talk faster than anyone can type. Takes some practice to do so in a useful manner though. Helps too if there is some consistency in what is being said - like if you have to produce a consistent type of report. Lawyers and doctors very often use dictation systems to good effect and they do it 100% because it saves time, even for good typists.

The issue I have with believing this is not the forming complete sentences before speaking, but the simple fact that speech is slow and error prone.

Not once you are used to dictating. With a proper dictation system you can easily start, stop and record over what you've already done if you make an error. (People make a lot of errors typing too) Typically you get the report sent back to you for review and correction after transcription. However even including review it is still usually faster than typing it yourself.

Re:Yes talking is faster than typing

[h4rr4r]

If you have another human doing the typing then it is not really faster either. Maybe cheaper, but surely not faster. Since you have to wait for them to do that.

Re:Yes talking is faster than typing

[TheRaven64]

If there's sufficient duplication that you don't need to think very much about the report, then you may be even faster with either a template system or an adaptive input device. Editors that autocomplete sentences from a corpus of existing documents can reduce sentences to a dozen keystrokes, which is even faster than most people can speak.

Re:Yes talking is faster than typing

[sjbe]

If you have another human doing the typing then it is not really faster either.

You can have a human or a program doing the transcription. I've done both. You mostly need a human if the transcription has a lot of specialized words. Otherwise transcription software can often do an adequate (though rarely spectacular) job.

Maybe cheaper, but surely not faster. Since you have to wait for them to do that.

You're thinking of faster for just one report. Nobody dictates when they just have one document to create. When you have a huge pile of reports to generate dictation quickly becomes much faster. Trust me that plenty of companies have done the math on this. It's really a form of automation - higher upfront costs for improved productivity. Most hospitals employ a lot of transcriptionists precisely because they same money by saving time for high priced doctors. Dictation allows you to get more reports written in the same amount of time. You rarely just have one transcriptionist if you use a service and turnaround is typically a few minutes to a few hours. If you need something immediately there sometimes are cases where typing will get a report out sooner (at the expense of all reports that follow). I've watched people use transcription systems for years. Dictation doesn't make sense for a lot of cases but when it does make sense it is a big

Re:Yes talking is faster than typing

[h4rr4r]

I did not mean to suggest it had no use cases, just that it was not going to replace keyboards in general.

The transcriptionists surely use keyboards.

Re:Yes talking is faster than typing

[sjbe]

I did not mean to suggest it had no use cases, just that it was not going to replace keyboards in general.

Quite so. Keyboards are going to remain the dominant form of data entry for the foreseeable future. Dictation is for high volume data entry by very busy and/or very expensive people. It doesn't make much sense for most random data entry where you typically don't know in advance what you are going to say. Also dictation is hard to use in public places and takes a fair bit of training to use effectively.

Re:Yes talking is faster than typing

[sjbe]

If there's sufficient duplication that you don't need to think very much about the report, then you may be even faster with either a template system or an adaptive input device

Most doctors I've seen use autocomplete heavily in addition to dictation since certain reports or sections of them are more or less boilerplate. Plus the transcription software often has autocomplete word suggestions for technical words to speed things along further.

Re:Will we finally get a replacement for hard disk

[Anonymous Coward]

I've seen a keyboard that projects the keys onto your desk so there are no moving parts there.

Now this

[vikingpower]

is a technology that makes me think "Where can I get this ?! Now !??! "

Re:Will we finally get a replacement for hard disk

[Reverand Dave]

I feel sorry for people that will never know the joy of typing on an IBM model M mechanical keyboard. Composing anything with that awesome "click click click" just made it seem so much better.

Re:Will we finally get a replacement for hard disk

[Anonymous Coward]

My company just threw out a sh*t load of them (no, I wasn't allowed to take them home). :(

Re:Will we finally get a replacement for hard disk

[Anonymous Coward]

When SSD prices are on par with standard Hard Drives, then you've got something to troll about.

Re:Will we finally get a replacement for hard disk

[Anonymous Coward]

Obligatory cartoon.

At The Limit

[LoRdTAW]

"20 times the write performance"

I wish we could actually use that performance instead of being hamstrung by the limits of SATA 6gb. Even with today's flash memory we have hit the limits of SATA 6gb (around 600MB/sec). Can we please get cheap bootable PCIe x4/x8 cards instead of SATA. And stop making PCIe cards that are nothing more than SATA RAID + SATA SSD's. Design an ASIC that looks like an ATA or SCSI controller and directly talks to the memory and PCIe bus. If a 1 terabyte PCI card which has at least 2GB sec read speed for around $300 came out I would buy it immediately on impulse. I want to jump into a game and not even realize its loading. I want my programs to simply pop up. I want to forget that there is a difference between main memory and storage speeds.

At that point I won't have to worry about space limits on my SSD and eliminate the need for mechanical storage for non critical stuff like multimedia, backups, archives etc. That is how I do it today, one 256GB SSD for just my games, 1TB for boot/programs/VM's etc. I also use a 2TB eSATA drive for extra stuff when I ran out of room on my 1TB (too many experimental VM's). A high capacity SSD would allow me to stop juggling which games I have installed on my SSD. I mainly use steam so its not that big a deal but sucks when you want to dust off a game and wait for it to download.

Maybe in the future AMD or intel can provide Hypertransport or QPI connections to SSD's or like in that article a few months back, put the non-volatile memory on the main memory controller along with RAM. Then we can finally shed the need for mechanical disks.

Re:At The Limit

[SuperKendall]

The new Mac Pro attached SSD storage through PCIe, if you want to see it in action...

Re:At The Limit

[BitZtream]

Seriously?

http://www.apple.com/mac-pro/

http://lmgtfy.com/?q=pcie+ssd

Re:At The Limit

[LoRdTAW]

I know there are PCIe SSD's but they are either way too expensive or the cheaper ones are simply "naked" SATA SSD's tied together with a RAID controller and nearly as slow as SATA attached drives. The Intel 910 is a PCIe->SAS->SSD 400/800GB, 1-2GB sec read and costs $2000 for the 400GB and $4000 for the 800GB. Its fast compared to the OCZ RevoDrive but not worth the money no matter how you look at it from a consumer standpoint. They can certainly do better in terms of transfer speed, PCIe x8 Gen 3 has a bandwidth of nearly 8GB/sec.

Apples SSD is not worth including because as far as I can tell, the paper towel roll sized Mac Pro uses proprietary components. So its not using an off the shelf SSD that anyone can pop into their PC. Though, the controller might find its way onto standard consumer boards.

Re:At The Limit

[Courageous]

6gbit = 6,000,000,000 / 4,000 = 1.5M 4K IOPS.

Do you presently have a 1.5M IOPS SSD? I think you do not.

Bandwidth isn't everything, friend.

Re:At The Limit

[Anonymous Coward]

A 1TB PCIe SSD for $300?!?! AHAHAHA! Well, give it 2-3 years.

Of course, then you'll want a 2 TB drive for $300.

And that's what keeps the spinning disks in production...

Re:At The Limit

[Anonymous Coward]

A good SSD will do 600MB/s read which is 5gbit not including protocol overhead. Add on the 100MB/s writes and you've used up your 6gbit.

This is a known problem. The SATA standard people have basically said "fuck it, just use PCIe" and I'm not even kidding.

Re:At The Limit

[Anonymous Coward]

SATA express, aka SATA 3.2 will use PCIe

Re:At The Limit

[stoatwblr]

"Can we please get cheap bootable PCIe x4/x8 cards instead of SATA" Yes. The drive interface format is already available. It'll trickle down to consumer space within 12 months.

Re:Will we finally get a replacement for hard disk

[Zmobie]

What would make it more entertaining is when the computers "overhear" each other and now you have a logic block for two completely different problems being entered simultaneously. Sad part is, I have seen some people that probably wouldn't notice, would check the code in, and resolve the issue they were working on.

Re:Will we finally get a replacement for hard disk

[mlts]

SSD has its uses, but long term storage isn't one of them. Hard disks can be recovered, but a SSD, good luck... that data is gone.

Of course, the trick is to have some media for backups or just even snapshots, except that it seems that IT shops are embracing geographically separate SANs and async mirroring as opposed to archive-grade media. Virtually every shop I've worked with, no matter how big a SAN they get, if one doesn't watch out, it will get full overnight.

Since it appears that the pendulum finally is swinging away from "the cloud" as a means of long term storage, I'm wondering what the next few years will bring for backup/archive media. LTO-6 is the last tape holdout, and there doesn't seem to be much progress on archival grade optical (no, BD-R isn't enterprise grade.)

There are tons of improvements in the upper tiers, but the bottom slow/reliable/archival tiers are important too. I will not be surprised to see more data retention laws out being passed, not to mention the existing ones (Sarbanes-Oxley, FERPA, HIPAA) which require heaps of data to be stored for long periods of time, and storing stuff that likely will never be accessed on even tier 3 drives on a SAN is a waste of electricity.

Re:Will we finally get a replacement for hard disk

[h4rr4r]

So?
If the hard disk is faulty use backups. Don't trust that it is still working if you get any faults.

Re:Will we finally get a replacement for hard disk

[Gr8Apes]

Trackpad or touch screen. Same for the keyboard.

Re:Will we finally get a replacement for hard disk

[RenderSeven]

The latter two, what would you replace them with?

Silly question! Holographic projections with resistive force fields for feedback. Saw them on TV. Im pretty sure they use green zero-point energy somehow.

Re:Will we finally get a replacement for hard disk

[h4rr4r]

Troll about?
SSDs are far cheaper than hard drives were for most of my life. Speed costs.

Re:Will we finally get a replacement for hard disk

[krammit]

You ain't kidding. Just dug one out of a long forgotten closet yesterday. Got a sticker on the back that says it was built Nov. 1993. Looks like it just came out of the box and works like a champ. I am one happy camper.

Re:Will we finally get a replacement for hard disk

[Anonymous Coward]

Sigh. Read his post: he says SSDs don't give soft faults, they have just died on him in the past. The correct response is: you should always have backups.

Re:Will we finally get a replacement for hard disk

[Anonymous Coward]

Hanc marginis exiguitas non caperet

Re:Will we finally get a replacement for hard disk

[SuricouRaven]

Spun-down discs. Disconnected.

Use them just like tape media. You can even have a robot to pull them out of storage, or plug a small SATA controller and power device into the one you need to access.

Re:Will we finally get a replacement for hard disk

[Anonymous Coward]

SSD's suck ass. Why waste money replacing them every 12-18 months?

Re:Will we finally get a replacement for hard disk

[h4rr4r]

Yeah, but you also should not trust spinning rust that is starting to fail. In that case you always go to backups.

Re:Will we finally get a replacement for hard disk

[petermgreen]

I've never personally had one "fail" as such but i've seen mysterious corruption where files are no longer accessible or no longer contain what they should contain far more often than i've seen such things on hard drives.

My feeling is that the wear leveling algorithms are not handling some corner cases (such as powerloss) correctly.

Re:Will we finally get a replacement for hard disk

[KingMotley]

Virtually every shop I've worked with, no matter how big a SAN they get, if one doesn't watch out, it will get full overnight.

I fixed that long ago. I buy 10x what they expect to ever need, and then connect everyone to it over a shared 10Kbps link.

How much will it cost?

[deaf.seven]

This raises the question how much will it cost per GB?

It had to be said

[CCarrot]

Don't they know that resistance is futile?

Pfft. Hand in your geek cards at the door, please... :p

Re:Will we finally get a replacement for hard disk

[pipatron]

They have been getting some sort of a comeback recently, with Das Keyboard becoming mainstream and a lot of companies following them to grab a piece of the market. I wouldn't trade my Das Keyboard for a "normal" keyboard, although it would be interesting to try one of the newer, "silent" versions. Supposedly they have an almost identical tactile feel to them, at least when comparing the graphs.

Re:Will we finally get a replacement for hard disk

[armanox]

Not to mention are increasing rapidly in capacity.

Pair of Dimes Shifting

[carrier lost]

This is the thing that could cause people to ditch Facebook.

Just imagine an FB-like app on a mobile device that can store all your data, your friend's data and your family's data.

Who needs a cloud?

Re:Pair of Dimes Shifting

[Anonymous Coward]

Are you really that stupid?

Re:Pair of Dimes Shifting

[BitZtream]

... then people who want to keep it in sync?

The cloud isn't handy because it stores your data, thats just an implementation detail that puts 'them' in charge. The cloud is handy because it makes things easy to access from an unexpected location with minimal effort.

Re:Pair of Dimes Shifting

[carrier lost]

Why would it not synchronize all the time?

Is that too much to ask of a mobile device/connection?

Asking because I really don't know - don't have anything mobile other than an old flip-phone.

Re:Will we finally get a replacement for hard disk

[pipatron]

Most probably marketing bullshit from this company, indeed.

vaporware

[slashmydots]

There's still time for Micron or Samsung or someone else to buy it and kill it to prevent it from crushing their business model of slowly releasing marginally better stuff every month. Definitely still in vaporware territory.

Re:Will we finally get a replacement for hard disk

[Anonymous Coward]

*hands

Re:Will we finally get a replacement for hard disk

[quintus_horatius]

May I introduce you to the MacBook Wheel

Re:Will we finally get a replacement for hard disk

[pipatron]

Honestly, computers are so efficient these days that fans shouldn't be necessary. I guess it's some sort of evil spiral. "Oh, the computers are so fast anyway, we don't have to bother optimizing this, it's just unnecessary developer time". Then you get to a situation where people write applications in a framework running on javascript running in an operatingsystem that is running *in* a web browser which is running on some other type of virtual machine ad absurdum.

And it's slow, but who cares, do you have a slow computer? Just buy more power.

Yes, I'm old.

Sad, but.....

[spstrong]

What is a "postage stamp"?

Re:Sad, but.....

[Zaatxe]

What is a "postage stamp"?

It's a kind of fee authentication key for physical e-mails.

Re:Will we finally get a replacement for hard disk

[Anonymous Coward]

Yes, I have had several fail.

10 so far, out of 15+, in a 2-3 year spam. Thats a *much* higher failure rate than good old spinning disk.
I've had 2 Air SSD's fail (Toshiba, and Samsung), 4 OCZ in a row (i got my money back from newegg in the end, as i refused another replacement. they argued, but I won in the end). 1 x Intel (firmware related I suspect), a few random other brands.
In general the Mac firmware based ones (Toshiba, Samsung typically) seem to last better than other brands.
I've bought a bunch of Samsung 830's this year, and so far so good, although again, i trust them as far as I can throw them.

I tell clients use SSD for an OS drive, and a HDD for data (and make backups!).
For myself, I use a zfs raidz2 + ssd larc drive for network storage (videos/photos/documents/source etc), and treat the SSD in my laptops as imminent failure - anything important goes on the NAS. And the NAS backs up offsite daily to another duplicate at the office, so home data -> office, and office data -> home.

Re:Will we finally get a replacement for hard disk

[KingMotley]

Wait until you start using brain scanners and continue to "type" after you've fallen asleep. I cringe to think of how the conversation changes.

Re: Will we finally get a replacement for hard dis

[Anonymous Coward]

How so? It cools at the cost of electricity, why would you still need the same cooling capacity in a heat sink?

Sounds too good to be true

[nhat11]

Until this hits market, its all smoke and mirrors.

Re:Will we finally get a replacement for hard disk

[Anonymous Coward]

Didn't anyone catch the Kbps link? I'd imagine it would never get full at that rate :)

Re:Will we finally get a replacement for hard disk

[interkin3tic]

When a normal hard drive fails, you can take it apart and throw the discs like a Frisbee. You can't do that with an SSD. Well, you can, but it doesn't go as far.

Re: Will we finally get a replacement for hard dis

[Anonymous Coward]

You're assuming mechanical input is necessary. Brainwave input?

Re:Will we finally get a replacement for hard disk

[TheRaven64]

How are you measuring price? Cost per GB? Then they're more expensive. Cost per IOPS? Then they're a lot cheaper. Cost per Watt during active use? Still a lot cheaper. Cost for the amount of storage that you need? Then they may or may not be more expensive: a 40GB SSD costs about the same as a 40GB hard disk, but a 1TB SSD costs a lot more than a 1TB hard disk. If you only need 16GB then the SSD option can be a lot cheaper.

I wouldn't replace the hard disks in my NAS with SSDs, but there's no way I'd go back to using hard disks in my laptop or in the rack-mounted machines I use for big jobs - the productivity hit would be too great, and my time is worth a lot more than the cost of the SSD.

Re:Will we finally get a replacement for hard disk

[TheLink]

Yeah it'll be a big problem if you continue to control stuff with your mind after you fall asleep.

Fall asleep while wearing your iThink and watching a movie, and dream of sending That Email to your boss/customer? Wake up and find out it's sent...

Normally most people get somewhat immobilized when they sleep. But even then some sleepwalk.

Re: Will we finally get a replacement for hard dis

[TheRaven64]

A Peltier device cools one end and heats the other. You need to dissipate the heat on the other end. If you don't, then the power required to maintain the temperature gradient increases and eventually the heat leaks back. Their main advantage is that they provide a good way of shifting the heat from the small CPU to a big heatsink. They don't magically make heat go away, they just move it.

Re:Will we finally get a replacement for hard disk

[mlts]

On a side note, a few years back, I was working on designing a tape robot that would be able to use bare 2.5" drives.

It had almost zero time to load (once it was plugged into a reader slot and spun up.)

It also had various uses it could work as, be it a VTL, the disks presented as tape libraries, spanning disks where data that wasn't used would be moved to platters, then demounted.

The problem was the robotics. Only one company was able to make the robots that could reliably grip, move, and ungrip the drives, and they were asking $10,000 per unit for starters.

Eventually, the project got shuttled aside, but having a silo that was able to use disks without any special enclosure required would have been nice for the enterprise (IMHO, of course.)

Re:Will we finally get a replacement for hard disk

[tibit]

When you start seeing those advertised by the Reynolds Corporation, you'll know that big heatsinks it's going to be :)

Re: Will we finally get a replacement for hard dis

[ceoyoyo]

Bah, you and your silly click click. Clearly you've never experienced the joy of pressing a key and hearing a chunk of steel slam into another chunk of steel with a bit of paper between.

Having experienced both these things and a lot of what's followed, I prefer the chicklet style keyboards with short travel and low resistance. My ideal keyboard would probably be some kind of finger tracker that was good enough to recognise finger movements as I tapped lightly on my desk.

Re:Will we finally get a replacement for hard disk

[kevingolding2001]

What about fans? Are they not a 'moving part'?

Hmm!

[eyenot]

Perhaps this is why HP has failed to produce their Memristor on time. They, too, had made mention of using something called a "crossbar" schematic to compress the whole thing.

Re:Will we finally get a replacement for hard disk

[SuricouRaven]

Maybe you're doing it backwards.

Leave the drives where they are. Have the robot move the reader around. Just a little circuit board with a power supply and an interface between SATA and something (ATA over ethernet?) able to drive a long enough cable. You'd need to align it quite precisely as SATA+POWER isn't self-centering, but there'd be no need to deal with grippers.

Re:Will we finally get a replacement for hard disk

[Iniamyen]

I've had mine for 3 years, still runs great. Anecdowned!

Re: Will we finally get a replacement for hard dis

[viperidaenz]

The power consumed by the peltier goes somewhere. It turns into additional heat that must be dissipated.
More heat sinking required.

Re:Will we finally get a replacement for hard disk

[Kichigai Mentat]

Obligatory reference to the Newton.

Re: Will we finally get a replacement for hard dis

[chromeronin799]

Surface rt with touchy keyboard?

Re:Will we finally get a replacement for hard disk

[mlts]

That is true, but for a lot of things, the ability to read/write more than a drive at a time comes into handy. For example, when setting up a batch of media that is going offsite in the weekly Iron Maiden pickup, or if the silo is partitioned so two different backup servers can access independently.

This would need multiple readers, although just having one coupler that handles all the I/O might be useful for a small array that backs up a critical server in a rack.

Re:Will we finally get a replacement for hard disk

[Anonymous Coward]

Captcha: approval

Nobody gives a fuck about your goddamned captcha.

Re:Will we finally get a replacement for hard disk

[Anonymous Coward]

Pick a page in a random book, time how long it takes to read it out loud, then time how long it takes you to retype it.

A very fast typist might get within striking distance of or slightly past parity, but there's no way you type "a multitude" faster than speech.

Re: Will we finally get a replacement for hard dis

[Anonymous Coward]

Hud and bci. Duh.

Re:Will we finally get a replacement for hard disk

[hairyfeet]

SSDs suffer from the hot/crazy scale and if anything they have gotten worse with each shrink and unlike HDDs the majority of SSDs failures I have seen had ZERO WARNING, it literally went from working yesterday to "flip a switch and its gone" and unlike HDDs which you can often get data off of a dying one with SSDs you don't get dying, you get perfect and then paperweight. This is why I only recommend them when the person is gonna be keeping everything in the cloud or is gonna be religious about backups, otherwise its just too risky. Hell you can't even do a trick similar to the old "swap platters to get one last shot at the data" bit because the controllers have encryption so if you managed to get the chips off all you'd get is gibberish.

And keyboards is why I think the desktop/laptop will never die, it only takes typing a long email on a tablet or talking into one and having to fix a bazillion errors for people to see the value of the traditional keyboard and mouse. There is a reason why the keyboard and mouse have lasted this long folks, they have tried everything from touchscreens to voice command since the late 80s and nobody has come up with an input method that works better than a good old keyboard, not even close.

Re: Will we finally get a replacement for hard dis

[Belial6]

Have you tried using a tablet as a keyboard for touch typing? You may be the exception, but for most people it pretty well sucks. The little nubs on the F and J keys, the indentations of all the keys themselves, the fell of the edge of the key letting you know how far off dead center your click is, and the feel of the space bar under your thumbs are all generally use as non-visual cues to let you know exactly where on the keyboard your hands are placed.

I assume that I am not alone in the fact that I could be blind folded, sat at a desk, and then proceed to type with a very high accuracy rate.

By the way, This sounds like what you are looking for.

Re:Will we finally get a replacement for hard disk

[Black LED]

Dr. Stuart Ashen did a great review of the Newton which shows off just how absolutely horrible it was.

Re:Will we finally get a replacement for hard disk

[TheRealMindChild]

I absolutely beg to differ

Re:Will we finally get a replacement for hard disk

[Anonymous Coward]

Unicomp. Old IBM mechanical buckling spring keyboards in the modern world.

They bought the design and business and still make them. Not the nice heavy, indestructible, steel housings but the same keys and they have USB and the various modern buttons you won't find on the classic models.

I have one at home and love it.

Re:Will we finally get a replacement for hard disk

[Anonymous Coward]

We have tried them several times where I work. I have yet to see one last 6 months.

Re: Will we finally get a replacement for hard dis

[ceoyoyo]

I have. I don't mind it that much, although it's not as good as a keyboard.

You're misunderstanding though. I specifically said "some kind of finger tracker that was good enough to recognise finger movements." Not finger placement, finger movement. A decent typist certainly doesn't feel his way to each key. He uses "those little nubs" to make sure his hands stay in the right place on a physical keyboard, then his brain knows how far to move his fingers to hit the right keys. The requirement to have your hands stay in the right place is purely a limitation imposed by physical keys or trackers that look for position instead of relative motion. A motion-based keyboard would be perfectly happy even if your hands drifted all over the place.

The thing you linked to from Think Geek relies on position, not motion.

Re:Will we finally get a replacement for hard disk

[Anonymous Coward]

* keyboard
* mouse buttons

BMI would be really awsome! ( http://en.wikipedia.org/wiki/Brain%E2%80%93computer_interface )

Then we could get rid of the screen and speakers/headphones too

Re:Will we finally get a replacement for hard disk

[jnork]

I feel sorry for people that will never know the joy of typing on an ASR-33 teletype. Composing anything with that awesome "KACHUNK KACHUNK KACHUNK"...

...erm...

...oh, um, on second thought, forget I said that.

Hah hah, no, seriously. You're right, the old IBM keyboards were AWESOME. Great tactile feedback, impervious to damage, and in a pinch you could use it as a blunt weapon. Against a bear. And when you were finished bludgeoning the bear into submission, hose off the blood and fur and go back to typing.

Pasting in X without a three button mouse

[reluctantjoiner]

It was (is?) a common setting in older Linux window managers so a person with a two button mouse could paste into a window by 'chording' the left and right mouse button, as if they had a three button mouse. Chording may have worked for Englebart, whose mouse was a sturdy mechanical thing. With a plastic mouse it works about as well as you think.

Re:Will we finally get a replacement for hard disk

[mark-t]

I would expect that the "iThink" would simply not function while you were asleep... and it is possible even with today's technology to differentiate the brain signature of a person who is sleeping from when they are awake.

Tactus

[tepples]

A promising tech I remember reading about I think on slashdot was the touchscreen that could bend in such a way as to give tactile feedback at any point of the screen for a variety of sizes and shapes.

You're thinking of Tactus (CNET, CNET video, and Wired video), which provides enough of a raised surface to give the user a reference point for his thumbs. But good luck finding enough smartphones that support Tactus functionality to make developing an application that relies on Tactus financially feasible.

Re:Tactus

[Zmobie]

Ah yes, that is the one. Hopefully some proprietary device development might start focusing on specific functionality to give it some traction, though I still kind of think mechanical buttons are going to win out for the time being. NVidia actually had some Shield demos out at Quakecon this year, maybe something like that with more touchscreen focus could help them out (Nintendo, I am looking at you with the damn DS/3DS already...).

Re:Will we finally get a replacement for hard disk

[TheLink]

Looks like it's possible: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2737577/?tool=pmcentrez

Multiple people dictating

[sjbe]

If everyone dictated, the noise in most offices would be unbearable.

That's not actually true. I've been in pathology labs where multiple doctors are dictating simultaneously. For the most part it works just fine. It's no noisier or more distracting than having a number of phone conversations going on which happens in lots of offices without problems.

Hell, if ONE person dictates all the time, it is unbearable.

That just tells me you've never really been around people dictating. It's not any more distracting than listening to one end of a phone call. If that amount of noise bothers you then you're going to have a very hard time in most offices.

PIRACY!!!!

[Jason Levine]

So you've got 1TB on a chip the size of a postage stamp? Who's to say that you didn't put a movie's DVD rip on that and mail it to a friend of yours? PIRACY!!! We've got to outlaw all mail since it could possibly be used to pirate our valuable intellectual property! Think of the children (of the overpaid content company executives)!

- This message brought to you by the MPAA and RIAA

Re:Will we finally get a replacement for hard disk

[volmtech]

Those of us with severe carpal tunnel syndrome can not. Even two finger typing can't be sustained as the pain level rises.