The Joy of the Flash Drive

An anonymous reader writes "A post to the C|Net site covers the numerous benefits of flash drives, such as speed, temperature, and battery consumption. The perk author Michael Kanellos is most fond of? The distinct lack of noise. 'The notebook I'm testing--a Dell Latitude D830 with a 64GB flash hard drive from Samsung--hasn't emitted a sound in three days. Flash drives, which store data in NAND flash memory, don't require motors or spinning platters. Thus, there are no whirring mechanical noises. Compare that with my T42 ThinkPad. It sounds like a guinea pig got trapped inside, particularly during the start-up phase. Vzoooot. Cronk, cronk, cronk. Zip, zip. (Pause.) Gurlagurlagurla...zweeee. '"

Comments

I like it.

[Wellington Grey]

It sounds like a guinea pig got trapped inside, particularly during the start-up phase. Vzoooot. Cronk, cronk, cronk. Zip, zip. (Pause.) Gurlagurlagurla...zweeee

I like the hard drive noises. Lets be honest here, they are soft clicks and chirps, not chainsaw noises. It gives me a non-visual feel of what the computer's up to.

-Grey

Re:I like it.

[Anonymous Coward]

Me too. But we have to admit that the same function could be fulfilled by an LED or something else that could be activated or disabled, instead of constant noise pollution regardless of the user's wishes.

Use the actual read/write line (heh, on a PATA interface, but you get my drift -- the Flash chips each have read/write lines, and an OR of those lines will work in the context of SATA) to activate the LED, and emphatically not something controllable by drivers or client-side software, and you've got my buy-in.

Funny story about blinkenlights... or crunchensoundz, as the case may be.

One of the things I initially disliked when migrating my gaming rig to XP (versus 98SE, yeah, I held out that long. The 98SE system listened on no ports, so I completely slept through the whole string of uPNP and DCOM/RPC exploits without so much as a scratch) was that the OS was always fucking around with the disk, even if not swapping. My rigs have always had enough RAM such that 9x would rarely, if ever, swap under normal usage, and I'd been used to years of total quiescence when reading long Slashdot threads. The machine's totally idle, right? Anyways, when I started migrating, it annoyed me that the XP box was always poking around WBEM\wherever, $MFT (by definition!), and so on. I'm looking at how much swap you're using, and it's not changing, so stop that. This box doesn't need to be writing anywhere. What if the power goes out at the exact moment that... journaling or not, this is just a silly design. (I'd never lost data on 9x/FAT32 due to power failures or crashes, but that's because the system was either quiescent on powerfailure, or I waited until the system reached quiescence before hard-booting, and I manually ran Scandisk from DOS mode to make sure I'd cleaned up the cruft... so with a track record like that, can you blame me for not trusting NTFS? Ironically, in the years since migration, I've lost data under XP/NTFS once, which is still one more time than I lost data under 9x.)

Which is a long way of saying that I like hearing the hard drive crunching away in the background. If my drive starts crunching when I'm browsing the web, and it's not about the same time of day that Windows Update typically phones home, the first thing I'm doing is sliding to the nearest open window and running Russinovich's old FILEMON.EXE to see WTF's going on this time. 99.99% of the time it's just been some other Windows process, or some phone-home crap from Adobe or Steam. But once, the 0.001% case paid off. I got bit by one of the "virus via ad banners on reputable sites" events (serves me right for not blocking the provider on sight) a couple of years ago, and the only reason I found out about it was because the hard drive makes a noise when it seeks.

Re:I like it.

[wfberg]

You really notice the lack of noises when you're logged on to some system remotely using a GUI. Double click something, and if there's no immediate reaction (or hourglass..) the system seems unresponsive. If you were sitting at the machine, you'd hear buzzing and whirring as *something* happens in the background. If I were marketing nifty thinnish-client solutions, I'd make sure there's always some sort of activity indicator (CPU, disk, ..) on the screen, so the system seems as responsive as a local client.

Re:I like it.

[Gordonjcp]

I've got a circa-1986 DEC PDP11/73 with a couple of 40M hard disks that are surprisingly quiet - in fact, thanks to the proliferation of tiny high-speed fans (as opposed to the 4" low-revving fans the PDP11 has) my PC is louder than the whole PDP11 rack. Except, of course, when you fire up the RL02s.

Oh yes. Here we have a 10M removable disk pack about the size of a kitchen bin lid, driven by something the size of a washing machine motor, with the heads mounted on a pint glass-sized voice coil positioner. The drives aren't that noisy when spinning up (well, one is but that's because a motor bearing is a bit dry - some servicing needed). When you actually get them going, they make a satisfyingly chunky "gweep thock gweep gweep thock gweeeeeep ka-thunk" noise. If you don't have the sound-deadening rubber feet screwed down, but just leave the rack standing on its solid castors, the noise is conducted into the floor and is loud enough to upset the downstairs neighbours.

Re:I like it.

[phoenix321]

HDD activity indicators are great when logging into a machine you're not directly in front of, be it a remote desktop or KVM switched. Especially swap trashing or scheduled virus scans can slow down the entire system with barely visible symptoms in cpu utilization in taskmgr or top. They leave a remote operator with only faint clues on why the machine is so damn slow right now, as the CPU load is negligible and caused only by processes that run all the time anyway.

It's a boon when you do support on a client machine of unknown horsepower, a rotting Windows installation or fragmented filesystem. You remotely started a program, say Outlook, a typical offender, five minutes ago and you don't see any operational window yet. System load for OUTLOOK.EXE is almost nil. How do you tell if it has crashed or is just starving for HD access without looing at the HDD light?

Re:I like it.

[CAIMLAS]

The problem is that LEDs are visual, whereas the sound the disk makes is, well, auditory. This poses several problems:

1) Auditory sampling rate is hundreds of times faster than visual sampling rate in humans (if you could make such a comparison to recording mediums).
2) Auditory signals can be picked up imperceptibly/subconsciously, without conscious effort. A LED has to be actively looked at to be alerted to any possible problems.
3) Most disk activity LEDs do not appear to flicker at the same frequency as the disk is making noise. In order to accurately represent it, I think you would need, at least, two or three different colors being represented within a single LED. And said LED would need to blink substantially faster than it does currently.
4) A disk's noise is not just an auditory indicator; it is the actual sound the disk makes. Each disk is different and, once a person is familiar with the sound of a given disk, is able to predict exactly what the disk is doing at any point in time. That'd not be easy to do with a LED.

Hey, maybe I'm off my rocker and I'm the only one who will listen to a disk -carefully- (and with a stethoscope in a loud room) to help determine the physical health of a drive. It's saved me more than once - and the hard drive LEDs would've been a very poor substitution.

Re:I like it.

[phoenix321]

What if I don't want to paint all my LED-bearing indicators with an ugly black stripe? That may be great at night but an abomination by nasty by daylight. And then again, even laptop manufacturers well-known for not following the abominable blue-lights-and-phony-silver design craze (Lenovo...) is going for blue lit power buttons right now. I can't tell you how disappointed I am, but the sharpie/masking tape solution obviously won't work on a power button.

And there are equipment manufactures out there that put a diode of epic blue-laser-proportions beneath every damn button. I certainly remember an offensive DVD player at a friend's house that severely distracted from watching the TV screen with no less than five bright blue lights, one of them strobing all the time. Each movie looked like the "Battling Seizure Robots" unless someone put a DVD case in front of it. And even then the whole cabinet was flashing wildly by scattered light from these diodes...

The design of this DVD player made me believe there are manufacturers in East Asia that really try to take over the West, literally, I swear. The design of this unit was hideously perfect, second only to a nuclear blast in underlying brainpower and evil beauty:

- all important buttons were glassy transparent with the laser diode beneath, shining directly into your eyes when the DVD player is placed below the TV
- the currently active function BLINKS incessantly. And yes, STOP is considered a function :)
- all function symbols were printed ON the button and the buttons were otherwise identical. The printing was done from behind and they were not arranged in a logical manner, so you would have no tactile or logical clue after covering them with a Sharpie.
- the front plate was recessed at each button's location with each button having a T-shaped cross section, making it next to impossible to paint all light emitting plastic.
- covering them with masking tape was prevented because these buttons were also sticking out a few millimetres from the unit, emitting light to their sides.
- putting a DVD case in front was prevented by knobs and design "features" sticking out from the front plate, so a gap one centimeter wide was always there, allowing the Seizure Robot's lasers to emanate from the sides. Even when placed *behind* the couch *and* blocked by a DVD case it was enough to light up the room in seizure-friendly blue strobes.

A thick dark woolen blanket finally put an end to the Blofeld's plans for world domination and his Seizure Robot when the unit thankfully died from a sudden case of severe overheating some months later.

Re:I like it.

[MttJocy]

Human beings are invariably tactile beings, who rely very heavily on our senses so that when we take some action in our environment our brain relies on sensory data to mediate that action, something our technology also needs to replicate in order to suit the beings whom it is intended to be used by.

This is one of the reasons why a completely flat keyboard like some of those roll up models tend to be a pain to type on when using a normal keyboard you get both the physical sensation of the key depression itself and the click sound, both of which allow the brain to deduce that the key depression was successful sure you could look at the screen but there is also the learned behavior aspect the reinforcement mechanism we have learned is the physical sensation and the click change that behavior from the expected and many people are not going to like it very much. Sure people can learn new feedback mechanisms but for the most part people are generally averse to investing energy relearning things unless there is a very compelling reason.

Another example is why many digital cameras replicate the click sound of older cameras again it is a feedback mechanism yes I know some people don't like this one however that is why they have an option to turn it off if you prefer not to have it.

The beep sound of an ATM terminal keypad is probably another example (although I personally hate the tone) because those hard wearing don't give much in the way of physical feedback on their own their movement is far too small to be really noticeable.

One could go though examples all day but the end result is that some people are going to find it somewhat of a transition to loose a form of feedback by the design of these devices, no doubt someone will come up with a software method to appease these users exactly like they did with the digital camera basically (play a sound file to replicate the expected feedback). I think with computers it is also an issue to have a non-visual indicator because there is typically already enough visual information on the screen already, extra visual objects are a less desirable solution there is only a certain amount of visual indicators the mind can take in at one time but it is possible to take in data from other sensory sources at the same time more easily than relying on passing everything through one input channel.

Re:I like it.

[dotancohen]

The consistent noise you are referring to is called white noise. That's what the fan makes, which is why "the fan is loud" does not mean "the fan is distracting". There are those who deliberately turn on white noise devices as a sleep aid, in fact.

Re:I like it.

[Peet42]

Do you want to know something really scary...?

Did you ever wonder why the full stop was made of brass on an otherwise plastic daisywheel? It's because somebody came up with a program that let you use microspacing mode to dump bitmap graphics on a Brother daisywheel using just the full stop.

Eek.

Re:Flash drive longevity?

[gradedcheese]

Technically, they don't really become unreadable, there's just an uncorrectable bit flip or two (out of say, 128KB) and that block gets marked "bad" and then it's not used anymore. Whatever data it contained is still there though, and you could read it if you wanted to. That said, on an SSD there is an onboard controller that abstracts away the Flash itself, so I suppose that it might not provide any interface to reading "bad" blocks, other than that there's really nothing stopping you.

Building your own Solid State Drive

[badzilla]

I saw this link via The Inquirer - how to build your own from a bunch of RAIDed CF cards.

Assemble a SSD disk for less than 75 Euro
http://www.guru3d.com/article/memory/506

DIY Compact flash in RAID good for 133MB/s

[distantbody]

...There is a pci card available that will take four CF cards and RAID-0 'em into a single drive. I was going to get it myself, but I slightly resented the poky pci bus at 133MB/s. In the future if they made one with 8 CF slots and put it onto a pci-e bus, I could then use 8 40MB/s CF cards in RAID-0 to make a single flash drive with 320MB/s on tap. That's a sweet-sweet prospect, but as yet they haven't made such a product.

Re:Flash drives sure have come a long way

[Sheepy]

dK'Tronics released a silicon disc for the Amstrad CPC which could be used as either a memory expansion or as a solid state drive.

Re:Bad Experiences with Flash

[RAMMS EIN]

This is not about wear, as far as I know. The flash cards work fine. It's just that Linux starts giving I/O errors after some time. Reboot the system and all is fine again. I think there is a software limit somewhere.

Re:One Major Disadvantage, however...

[Fweeky]

Flash media (NAND-gate type) is fundamentally slower than hard drives for sustained serial write behaviour And, er, random writes. You're lucky to get 20/sec because every little write ends up in a read-modify-write of a 4MB block or so.

The upcoming solution to this seems to be to turn random writes into serial ones; presumably buffering up writes in battery backed up memory.

Re:FYI

[Mike1024]

I wonder how Dell and Apple and everyone else can provide 64 gig SSD options for their notebooks for less than 1000 dollars. None of the brands had any info on the specs of the drives easily locatable, and I am worried these are the low end SSDs that are much much slower Well, one option is economies of scale - Apple could get 1,000 SSDs at $1,000 (total $1,000,000) but it's unlikely 'Rocketdisk' has that much spare cash to spend. Rocketdisk might keep 5 in stock at $2,000 (total $10,000). Also, if SSD supplies are limited Apple and Dell and IBM might be buying up the entire supply - big contracts tend to get preferential treatment compared to small contracts, for obvious reasons.

Fortunately, you don't have to worry about not knowing the performance of these SSDs because there are reviews aplenty comparing the macbook with and without the SSD. Here is one such review. Here's the summary: a bunch of benchmark bars showing the macbook air SSD outperform the macbook air sans-SSD; but being outperformed by the macbook and macbook pro without SSDs.

The Good:

        * No more entire machine slowdowns! (well, most of the time...)
        * Speedy boot, disk read, and build times

The Bad:

        * The moderate gains in everyday use aren't worth $1,300 And now you know!

Re:Flash drive longevity?

[monkeyfork]

At a seminar I attended recently, a rep from a Japanese flash manufacturer was all giddy at the prospects (and official rumor) that M$ was going to do for the flash drive what was done for the winmodem, ie extract the intelligence out of the peripheral and make the OS do it. M$ doing the wear leveling algorithm, what a comforting thought. And I'm sure it will be accomplished in a way that make dual boot even easier.

Re:Flash drives sure have come a long way

[xZgf6xHx2uhoAj9D]

As has been debunked I'm sure hundreds of times before on Slashdot, I will debunk this falsehood once again. The number of writes is not a limiting factor in the life cycle of a modern flash drive. It was a limiting factor 10-ish years ago, but it is not now. Even if you swapped to a modern flash drive constantly at maximum throughput and the drive was completely full except for one block (the worst possible scenario for a flash drive), it would conservatively last for at least 60 years (if I remember right). Maybe modern drives are now into the hundreds of years.

I'm not saying a flash drive will last 60 years. I'm saying the number of writes is not the limiting factor. Wear levelling and block moving (the nice thing about flash is you can relocate a rather static block of memory somewhere else, since it's all random-access) algorithms are to the point where the number of writes is not a limiting factor. Once again, write and swap and rewrite as much as you like to a modern flash drive. The number of writes is not a limiting factor.

I'm starting to think early flash drives will have the same effect of rechargeable batteries. When people think of rechargeable batteries, they too often think of the ones from yesteryear with a limited number of recharges and that awful "charge memory". So too it seems with flash drives. Even otherwise informed people seem to think that writing to a flash drive some billions of times will somehow have a deleterious effect. It won't.

Re:Flash drive longevity?

[TClevenger]

Depending on how the drive is used. If you have a flash drive 95% full and unchanging, won't the other 1% get killed pretty quickly, even with wear levelling?

Good Vibrations

[thatblackguy]

I too like the sound and the vibration that you can feel from outside the case. I can tell which drive it's accessing, if there's something wrong or it's not doing anything at all by just placing my hand on the case. Plus the expresson on people's when I touch the case and say something like "it's accessing d: drive" is priceless. It also sounds like the dude's drive is on its last legs.

Please consult a real dictionary

[Lucid_Loki]

From a real dictionary (oed): 'The Concise Oxford Dictionary of English Etymology perk(s) sl. abbr. of PERQUISITE(S)). XIX.' Where sl. means slang. Sure a slang word can be used in this context and I even pointed out that it was an widely used spelling. I stand by the assertion that 'perq' is the correct abbreviation. From your own link: Main Entry: 3perk Function: noun Date: 1824 : perquisite --usually used in plural Welcome to the world, where the American is not always correct simply because it is the most popular.

My wierd setup - mp3 player attached to laptop

[saigon_from_europe]

My old laptop had quite typical HD, but still quite noisy if you work at night while all background house noises are gone. So if I wanted to listen mp3, instead to listen if from the HD, I would attach my mp3 player to an USB slot, and then plugged headphones to the laptop. I could use the mp3 player directly, but this way I did not have to worry about batteries. (And Winamp is a better mechanism to control your playlists than player's internal software.)