Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!

 


Forgot your password?
Close
typodupeerror
×

Seven Mobile ATA Hard Drives Compared 125

Posted by Zonk from the compare-and-win dept.
AnInkle writes "Though hard drives are allegedly the fastest advancing high-tech product, most laptop manufacturers persist in saving a buck by outfitting their units with a low-end, low-cache, low-capacity, low-spindle-speed HDD. The Tech Report takes a different angle from other mobile hard drive reviews by including one of those maligned 4,200 RPM, 2MB cache models in their roundup of 2.5" hard drives, which includes 'a 160 GB perpendicular monster and a couple of 7,200-RPM speed demons.' The results are clear that most of us would see a tremendous boost in performance by upgrading this one component."
This discussion has been archived. No new comments can be posted.

Seven Mobile ATA Hard Drives Compared More

Seven Mobile ATA Hard Drives Compared

Comments Filter:

  • suprise :( (Score:5, Funny)

    by TheRealMindChild ( 743925 ) on Friday April 28, 2006 @11:45AM (#15221396) Homepage Journal
    tremendous boost in performance by upgrading this one component

    If you think THATs suprising, imagine my face when I found out that FLAMMABLE and INFLAMMABLE mean the SAME THING.

  • Quite true indeed (Score:4, Interesting)

    by jawtheshark ( 198669 ) * <slashdot@nosPAm.jawtheshark.com> on Friday April 28, 2006 @11:46AM (#15221407) Homepage Journal
    I use a real old laptop every day. (Just because I'm too cheap to buy a new one) I got it for 100€ from my former employer. It's a P-III 600MHz with 256Meg RAM. I added 256Meg that I had lying around and that already boosted performance.

    However, one day the included 6Gig harddisk with a really low speed (Must have been a 4200RPM, but could be less) and I bought a new 5400RPM 80Gig harddisk . That was pretty much the upgrade that gave me most speed. That, and I could finally install more than one OS and keep the machine usable ;-)

    Fast harddisks do matter.... Even if I tought that it was one of the least important things in the overall speed of the machine.

    • Well, I just upgraded from a 3 year old 4200 rpm drive to a 100G 7200 rpm TravelStar 7K100, and I went ahead and did some tests while I was at it:

      Copy a big directory: 2m18s -> 1m17s (Wow!)
      Boot up: 54s -> 47s (Ok...)
      Compile some code: 58s -> 56s (Meh.)

      In short, it depends on what you're doing. Unfortunately the things I wait for the most, didn't speed up much. Overall, going from 4200rpm to 7200rpm is an OK upgrade but I think overrated. Power consumption of the laptop as a whole only increa

      • Wouldn't it speed up your laptop immensely and reduce power usage if you upgraded to enough RAM to have no swap file? About 1 Gig should be enough in most cases. How much power battery life/ speed could be expected from a configuration like this?
        • I agree swap is pretty useless, and have often run systems without it, with no ill effects. But there are other programs that hit the disk. In linux, it's not that easy to stop all accesses to the disk. You have daemons like syslog that access every minute or so. There is (or used to be) some special module or patch that would defer writes to memory. This allows the disk to sleep more. I don't bother with that anymore because in my experience the runtime only increased 10 minutes or so. IME the drive
        • the ram boost from 512mb to 1gb helped my machine a lot since now i have more space to cache the files from the 4200rpm hdd :D

          however, the latency tests on the benchmark were not sufficient, they should really cripple and fragment the hdd before they make the benchmarks. there is the point where the 7200rpm drive with 20% less seek time really hits in.

          with continious data like the "default test setup", the speed difference isn't really worth to mention.

          if my ubuntu runs, it's all great and fast. if it works

  • Hard Drives are the slowest advancing components (Score:3, Informative)

    by Anonymous Coward on Friday April 28, 2006 @11:47AM (#15221421)
    While for instance Moore's law still holds the hard drives are not developing as fast. Ca. 10% performance improvement per year in the previous years is a good estimate. That means the hard drives are actually the slowest advancing components. In modern higher end PC they are the slowest link and in many applications the most horrible bottle neck.

    There isn't a good solution available either. RAIDs can get expensive, flash and similars can be fast but there are problems with interfaces (quality, selections, ..), the fastest things are expensive or on development cycle anyways... The state of hard drives and their performance is simply put pathetic and will be at least for the next a few years.
    • When I had a 100 MHz PC, I had a 1 GB HDD. Now I have a 400 GB HDD and I don't have a 40 GHz processor. What are you talking about?
      • I've had my powerbook for almost 30 months. It came with an 80 GB hard drive, which at the time, was the largest laptop drive available, and had been for quite a while...at least a few months.

        The largest laptop drive currently available is 160 GB, and that has only been available for a few weeks.

        So, in roughly 3 years, the capacity of 2.5" hard drives has doubled. (A far cry from Moore's law.) Processor speeds haven't kept up with Moore's law either, but they have certainly more than doubled in the

      • Re:What? (Score:3, Interesting)

        by Glonoinha ( 587375 )
        Throughput. Actually I occasionally chart the growth in CPU speed vs the size of hard drives and there is an amazing difference (with drive space growing a LOT faster than CPU speed) - but on a purely bandwidth based observation, narrowing it down to laptop hard drives only : drives haven't really kept up.

        Six years ago most laptop drives were 3600 rpm. Now most laptop drives are 4200 rpm (yes there are plenty of 5400 and 7200 rpm drives out there, but I'd wager to say that most people don't have them.) T
        • Due to higher density disks, the throughput had advanced quite a lot faster than just the rotating speed. So what you are saying is - to be honest - a load of bunk.
          • Actually what I was saying is that laptop drive speeds (throughput) haven't seen the massive gains that CPUs have over the last six years.

            Not magic 'burst' throughput (which is akin to measuring horsepower at the piston rings) but real throughput.

            Open two command line windows, and on each one start a copy of a zillion small / medium size files from some directory to another directory on the same drive (your MP3 directory would be perfect.) Two threads just doing concurrent reads and writes of uncached data
      • HDs were exploding in size in the late 90s becaues of the replacement of coil read-heads by thing film heads based on quantium magnetic resitance effects (sounds better than just GMR heads).
        That ended quite a while ago.

        We were stuck at 250GB for _years_.

        Pertenticular recording could give another boost the next years, maybe an order of magnitude, yes, but the last 5 years HD were very slow in progress.
    • I hold that floopy drives are the slowest advancing technology. they started out at 360 KB some time in the eighties, and are currently at 1.44 MB.

      [\dry wit]
    • In my day, a hard drive was going across the country in a car with kids in the back whining, "Are we there yet?" We didn't have it easy like you whipper-snappers with your SATA and your RAID! Hell, Raid was for killing bugs.
    • It seems no one here is able to stick to one interpretation for Moores Law.
      The original law related to increased transistor count for money per year. Here we have some people changing it to Mhz increase per year. Then they want to compare it to hard disk transfer speed increase per year while others are saying storage capacity (or density) per disk per year is more appropriate.

      No wonder there is so little agreement.

  • Power consumption (Score:4, Interesting)

    by Junta ( 36770 ) on Friday April 28, 2006 @11:48AM (#15221433)
    I see in their published specs that the 7200 RPM drives run at least 0.4 to 0.6 Watts higher. This may not seem like much, but right now my laptop is sucking about 17 watts of power, and that means about 2.4 to 3.5 percent higher power consumption.

    Still not much, but a factor to consider.

    • power is rate * time (Score:4, Insightful)

      by LunaticTippy ( 872397 ) on Friday April 28, 2006 @12:08PM (#15221634)
      If the computer runs faster you may use less power. If it takes 10 minutes to check email and slashdot with the old drive and 9 minutes with the new, you've just saved 10% time spending 3.5% higher power rate. That's about 7% less power consumed.
      • No, since the disk, while in use, would be on all the time. It does not go from spinning to non-spinning within the second or something like that. Maybe there is a small difference because of the head moving or not moving, but that's about it.
        • Let me explain again. You need to check your email and slashdot, ok?

          Old rig: takes 10 minutes. You turn off your rig.
          New rig: takes 9 minutes. You turns off yer rig.

          Voila, less power!.

          It's true that if you sit there using it constantly you will have less total run time, but my point is that you will get more done in less time (same total power use if you run the bat flat) with a faster drive.

          I guess there are some people that only care about total run time. They'll just poke around until their battery d

      • Re:power is rate * time (Score:1, Informative)

        by Anonymous Coward
        The thread heading is plain wrong. Do not confuse Power and Energy. Rather, with electricity-related units added for illustration:
        • Power [W] = Energy rate = Energy [W s] / Time [s]
        • Power [W] = Voltage [V] * Current [A]
        • Energy [W s] = Power [W] * Time [s] = Voltage [V] * Charge [A s]

        If the computer runs faster you may use less power.

        Which one consumes less energy depends on both the power and the total time (which, granted, was the point of the post).

      • yeah, but I'd probably just use that extra minute to surf slashdot
    • Yeah, that's the trade off. Do you want higher performance or lower power consumption? I would hazard a guess, and say that the majority of laptop buyers want lower power consumption. Those that want performance can upgrade at the time of purchase or sometime in the future.

      Most of the time, my 5400 rpm drive is fine on my laptop. When I want performance (say for video editing), I'm most likely to be somewhere where I can plug in to the wall, and use a higher performance, higher capacity firewire drive for m
  • If you had enough experience with computers, or especially if you lost one or more hard drives with precious data, that hard drive stack on the first page will shorten your life with a couple of years.

    It's enough for someone to fire up a match for me to panick and go check if it's not the HDD. Works every time.

  • It's Simple... (Score:5, Insightful)

    by MudButt ( 853616 ) on Friday April 28, 2006 @11:50AM (#15221446)
    most laptop manufacturers persist in saving a buck by outfitting their units with a low-end, low-cache, low-capacity, low-spindle-speed HDD.

    Well, they're saving an average of 111 bucks in these examples. The "low-end" model is about 35% of the cost of the others (on average).

    Now look at the performance differences. WorldBench is clocking the more expensive drives as only 30% faster (on average) than the "low-end" drive.

    My own conclusion: yes, you're getting a performance boost if you pay more... But it's definately not a 1 to 1 ratio. In fact, for the money, the "low-end" drive is the best solution. So... Why do "most laptop manufacturers persist in saving a buck (or 111 bucks)? Because it's a better choice for the average consumer! Believe me... If Company A started selling only expensive drives, their market would go niche (like Alienware), and most people would purchase a "lower-end" machine.

    • Re:It's Simple... (Score:5, Interesting)

      by nfarrell ( 127850 ) on Friday April 28, 2006 @12:21PM (#15221735)
      True, the other drives are ~$100 more expensive. If all you got was a 30% speed increase, I'd agree with you.

      But these drives are not just faster, they're also higher capacity. An ipod holds more than these low-end drives, and anyone who wants their laptop to be their MP3 player will happily spend $100 extra for ~80Gb more space.

      Yep, some people will buy the cheapest thing without looking at what they're missing out on. But it wouldn't be hard to market a lappy as "NEXT GENERATION: MASSIVE 100GB DISK DRIVE". But what would I know, I'm not in marketing.
      • But these drives are not just faster, they're also higher capacity.

        Point well taken... Overall, you're probably getting a 1:1 increase for your money. (Or maybe you're getting more bang for the buck by spending more when you consider all things). But I think that people like you and me can see the value in the high end drives. That's a given.

        But my next door neighbor, for example, asked me for advice on buying a notebook that would let him surf/email/type in his living room wirelessly. I recommende
    • What is your time worth? What if I sold you a laptop, and I gave you the hard drive for free, but it only held 8 kilobytes, and had a read/write speed of, say, 12 bits per hour. The performance to cost ratio is INFINITELY better than any other laptop on the market! Would you take it?
      • What if I sold you a laptop, and I gave you the hard drive for free, but it only held 8 kilobytes,

        What if my Grandma had balls? She's be my Grandpa. What does this have to do with anything?

        You can't take an analogy to an extreme and then expect it to still hold true.

        Do you think taxes are too high? Well what if we disbanded the IRS and nobody ever had to pay taxes again? No more fire dept, police, schools, courts, military, etc. Would you like that? No? Then okay... You should agree with higher t
        • You can't take an analogy to an extreme and then expect it to still hold true.

          Ok, well let me strip my comment of the sarcasm so you can follow it.

          Your metric sucks.

          If an average consumer uses a laptop for 3 years, and the performance hit due to a low end HD is on the order of 10 minutes of productivity per day on 300 days out of the year, thats a loss of 150 productive hours. Assuming the consumer's time is worth more than $1.35/hour, they are better off with a faster hard drive. Thats before we
          • Ok, well let me strip my comment of the sarcasm so you can follow it.

            Me is able to follow much. Not need simple words.

            Your metric sucks.

            I don't even use the metric system! WTF are you talking about!

            If an average consumer uses a laptop for 3 years, and the performance hit due to a low end HD is on the order of 10 minutes of productivity per day on 300 days out of the year, thats a loss of 150 productive hours. Assuming the consumer's time is worth more than $1.35/hour, they are better off with a fa
    • ...If Company A started selling only expensive drives, their market would go niche...

      I never thought of it that way before, but consider:

      I can name a couple of PC brands like Alienware and Voodoo [voodoopc.com], but I can't really think of anyone who is "known" for making similarly high-end laptops.

      I realize that's because most people need them for mundane, business-related tasks, but with so many (myself included) using laptops at work as desktop replacements, you'd think the specs would matter more.

      Our company just rol

      • but I can't really think of anyone who is "known" for making similarly high-end laptops.

        I see your point. Maybe the market's ready for a high-end notebook alternative...
        • Oddly, they advertise just such a beast on the Voodoo website (ANTI-DISCLAIMER: Don't work for Voodoo, don't own any of their products and don't even have any experience with them whatsoever).

          However, I didn't even know, until I saw it, that Voodoo made laptops.

    • However, it should be noted that dollar for dollar, assuming you have already gone to 512mb ram, nothing else will boost your speed better than investing in the hard drive. Many people will invest the same amount in a 200 mhz faster processor before investing in a hard drive, though, which is a blatant mistake, and likewise a jump to 1gig of ram is typically more expensive, and a mistake for most users.
    • Consumers dont know nor care about hard drive speed.

      We are geeks so we know but when they look for a new laptop they look for processor speed and memory and wifi connections. Thats it.

      Go look in ads for circuit city and Best Buy? Not even the graphics cards are listed as consumers dont care.

      So if Joe sees laptop A: for $1411 or laptop B: for $1300 with the same specs which one do you think he will go home with?

      Same argument for winmodems and the consumers picked the winmodem machines because they were $100
    • Now try the same argument on speed of the processors and you will find that the market should by and large never use any higher end CPU's.

      The speed increase from a low end to high end is rarely more than 30% but the price is often 2-3X higher.

      By the way combining a high end CPU with a low end Diskdrive is pretty much maximum folly but this is the way they are configured. Resason: People do not know anything about the disks but "know" they "need" the faster Intel chip to make TehIntanet go faster

    • That's worldbench, one test out of the dozens they ran. If you go through all the metrics, some of those drives win by over 300% over the 4200rpm drive. To combat the price difference, I recently purchased a 60GB Travelstar 7K100 for $100, and it beat the Barracuda, a freaking desktop drive, in some of the tests. The drive seriously makes my old Dell Inspiron seem like a new computer in every way.

      So yes, I'd say these tests prove the difference is worth the price premium.

  • Faster Harddrive? (Score:2, Insightful)

    by stonefoz ( 901011 )
    When I fire up my laptop, its because I'm not at a outlet, or am moving somewhere. The harddrive is the largest battery pull on my old rig. I'm not looking for fast but smart. They should have large caches and 4200 rpm or maybe even lower. An old toshiba from the 80's I've had, used to run for a few hours with power management. I haven't seen that again from these new 'mobile super computers' that is flooding stores. SATA is one of the smart ideas, but they'll all be chasing the benchmark crown for speed.
    • I used to get 6 hours on mine (centrino 1.7, dell latitude D505). Of cause, battery quality goes down pretty quickly, 12-18months in now I must be, and will get around 4 hours.

    • Compact Flash drives just got large enough to hold your basic operating system, web-surfing, IM'ing, productivity tool set and office suite, cheap enough to actually go for it (4G Ultras recently dropped below the $150 mark.) Granted, four gigs isn't enough for leet warez/mp3z/gamez monkeys, but it is enough to keep someone productive if they want to be productive on the go. At 9MB/s read/write they aren't as fast as the fastest drives out there, but ... I wonder what kind of current they draw under use -
    • I have a Dell Inspiron 9300 that I just upgraded to the Hitachi 7200rpm drive in the review. The speed is very nice. You might be content with a slower drive (like I was), but I could never go back to a 5400rpm or lower drive. With a utility from hitachi to put the drive in low power mode, I still get 4-5 hours of battery life and the drive still performs much better than a 5400rpm drive. When I'm plugged in, it automatically goes into full performance mode, not parking the heads or spinning down and it's e

  • Other Benefits (Score:1, Interesting)

    by iPodUser ( 879598 )
    I have personally witnessed many mobile PCs speed up and get quieter after having the HDD swapped for a higher spec unit. In addition to the increased performance in read/write speed, many of the newer drives offer another benefit to mobile users: Lower power consumption. Upgrading an older drive to a new faster drive could increase battery life.
    • Yeah, my theory on that is that if a drive takes 2 seconds instead of 4 seconds to do what it needs to do, then even if it takes more power to operate, it's still going to be a net decrease in the amount of power that an individual operation takes.

      I don't know if it actually works that way, but it's my theory and I like it.
  • Notebook hard drived (2.5 inch for factor) cost about three times as much as desktop drives, because they almost never seem to go on sale. I just had to buy a replacement for my notebook, and an 80GB/5400rpm/8M 2.5 inch drive cost me about $100. I can frequently get a 250GB/7200rpm/16M 3.5 inch drive on sale for a half to a third that price these days. With a $/GB ratio that high, cost is a significant factor in adopting higher capacitry drives on mobile platforms. A secondary issue is heat - some noteb
  • Sun opteron servers are shipping with Segate Savvio 2.5" 10K serial attach SCSI drives. Astonishingly expensive, and they suck about 8 watts each. Burn your thighs today . . .

    http://www.google.com/search?ie=UTF-8&oe=UTF-8&q=S eagate+ST973401SS&btnG=Google+Search [google.com]

  • I want reliabity (Score:3, Interesting)

    by hey ( 83763 ) on Friday April 28, 2006 @12:04PM (#15221597) Journal
    I want a raid in a harddrive form factor. So I can just plug it in like a new hardrive but if one disk fails it can still live.
    • I have to agree. Personally I don't buy 7200 rpm drives for my desktop, I would rather have the cooler temperature, better battery life that the slower drives have. Laptops tend to run hot IME anyway, why add more heat by using 7200 rpm drives. I don't recall anybody questioning the relationship between heat and lifespan of hard disks. As for cache, I can't see any excuse other than cost. As for laptops and raid, unless 1.8" disks gain the reliablitiy of a 2.5" disk, I think we are stuck with slower driv
      • For a long time Dell made drive carriers that were the same size as their removable CD drives (back when you could swap out a CD or Floppy on your laptop) - you could put another laptop drive in the carrier and run two drives in the same laptop. I don't believe there was hardware support for RAID, but I'm sure you could do it in software.

        I don't know if the current generation Dell laptops have removable bay CD drives, as it has been a while since I bought a new laptop.
    • I want a raid in a harddrive form factor. So I can just plug it in like a new hardrive but if one disk fails it can still live.

      How would this work? two drive motors? A better option would be removable stepper motors, If one of your motors went(platter or heads) it could be swapped out without compromising the protective atmosphere in the casing. this would drasticly reduce your spenditure on harddrives
      ..I wonder why the manafacturers don't do this...

    • Hah (Score:2)

      by Junta ( 36770 )
      I have two hard drives in my Thinkpad. Have to go without internal CD drive, but still interesting. Particularly with external USB CD drives (even ones that draw all the power they need via USB).
    • Dear hard drive manufacturers,
      Instead of adding more and more capacity to hard drives, I'd like to have a reeeeeeeally fast drive with about 5GB. All else can remain the same: cables, plugs, size, electricity bill, etc.. Those 5GB will be more than enough for my operating system and most used applications and the *useful* things I do with my computer will get done faster!!! I don't mind keeping the MP3s and pr0n on a "normal" drive or in a server away from my desktop.

      thank you

  • Warning: hyperbole detected (Score:3, Insightful)

    by thesp ( 307649 ) on Friday April 28, 2006 @12:05PM (#15221605)
    Since when was 10-30% overall from worst to best performer regarded as tremendous? The impact of the disk subsystem is around 30% on daily tasks, and around 70-90% on disk-intensive tasks. So we're looking at a ' tremendous performance increase' of around 10% to 25% in the best case, only achievable by owning the worst performer, and thence upgrading to the best available technology.
    • Boosting over all speed by up to a quarter is not a tremendous advance? Heck, I'd consider a 10% raise in my paycheck to be pretty tremendous. If my paycheck went up by a quarter, I'd blow a gasket.

      I feel the same about computing. If I could boost my laptop's overall performance by 10% by making a single change, I consider that pretty impressive.
    • This [techreport.com] sure is the strangest 10-30% I've ever seen.

      Yeah. Some tests were only won by a few percentage points. Others absolutely obliterated the 4200rpm drive by over 300%. Overall effect? Who knows... depends on how you use it. Either way, I'd say it's worth the price difference.
  • From the article photos it seems they were primarily interested in comparing their width, height and depth, as well as trying to build various models using them as freakishly expensive LEGO blocks.

    What I wonder is if after the experience, at least one of them is behaving odd or making funny noises on startup.

  • Face it (Score:2, Insightful)

    by DoctorDyna ( 828525 )
    Face it, mobile technology trails regular stuff. It's just a symptom of technology.

    While it is nice to have fancy shmancy specd laptops to tote around, you can only put faster (read: more power / heat) devices in a laptop to a certain extent. There is a curve that follows along with an opposite one, which refers to efficiency / portability and the other to power / speed.

    The other end of this discussion that I've not seen discussed yet is being mobile also presents real dangers to physical disks. Perhaps

  • Can we lose the troll writeups? (Score:4, Insightful)

    by Karma Farmer ( 595141 ) on Friday April 28, 2006 @12:33PM (#15221829)
    most laptop manufacturers persist in saving a buck by outfitting their units with a low-end, low-cache, low-capacity, low-spindle-speed HDD

    That's because rational consumers 'persist' in saving a buck by buying the least expensive thing they think will fill their needs.

    Most people buying PCs have absolutely no idea how to compare one computer to another. Even most Jeff K's understand nothing beyond screen dimensions and clock speed (and I've worked with enough IT people toto understand that Jeff K is the rule, not the exception). Of course, even the bottom of the line $650 Dell XPS comes with a 7200 RPM 8MB Cache HD, so I'm not sure what kind of poor sucker is still getting the 4200 RPM dog described in the article.
    • The poor suckers who buy an IBook, that's who. I still don't see why Apple has such a tendency to skip on relatively cheap upgrades such as extra RAM and a faster HD, even on mid-range models.
      • The poor suckers who buy an IBook, that's who. I still don't see why Apple has such a tendency to skip on relatively cheap upgrades such as extra RAM and a faster HD, even on mid-range models.

        The iBook is the bottom-of-the-line model, not the mid-range model. The midrange PowerBook has a 5400 RPM drive, and the top-of-the-line MacBook has a 7200 RPM drive option.

        I don't see why consumers have such a tendency to skip on relatively cheap upgrades such as extra RAM and faster HD.
    • Dell inspiron B's which start $150 cheaper than that come with 5400 rpm hard drives. A lot of businesses and home users buy those. I'm pretty sure dell sells significantly more of those than xps laptops.
      http://www1.us.dell.com/content/products/features. aspx/featured_basnb?c=us&cs=19&l=en&s=dhs [dell.com]
    • That's because rational consumers 'persist' in saving a buck by buying the least expensive thing they think will fill their needs.

      The rational does not persist, rather it is a new way of thinking. It probably has something to do with the guilt of having to throw away a top-of-the-line-x-years-ago-that-cost-an-arm-and-a -leg-back-then equipment which is worth nothing today (actually in some cases it costs money to get rid of the equipment). In my case, I can't throw away, but I don't use for obvious reason
    • I don't understand what about the writeup is trolling? Too much blame on the manufacturer and not enough on the masses? One effective method of getting people's attention is to sound like you're taking the side of the masses, against "the corporation."

      The main goal of an article like this is to educate the consumer. Lots of us may not have known how much performance we sacrifice by buying a laptop with a crappy HDD. You wisely point out that most people have no idea how to intelligently compare these th

  • swapping is the bottleneck (Score:5, Informative)

    by jilles ( 20976 ) on Friday April 28, 2006 @12:58PM (#15222007) Homepage
    Swapping is the bottleneck. So remove the bottleneck. No seriously. The harddisk activity you are most likely to notice is memory swapping. Swapping can be disabled. Of course you run out of memory if you do that, so add more memory. I find that with 2GB no application ever complains of having not enough memory despite there being exactly 0MB of swap space. I run some pretty memory intensive stuff too. It turns out most of this stuff is designed to run well on systems with only 512-1024 MB (particularly games rarely use more, even if it is avaialable). That extra GB is cheaper than a new harddrive and if 2 is not enough make it 3 or 4. It's not like win32 processes can address more than 2GB anyway!

    At least under windows, memory swapping is implemented very stupidly. Basically the system will spend (your) time swapping even when there's plenty of memory available. I've observed it swapping applications to disk with over 75% memory available. This causes all sorts of noticable delays when you try to actually use your system (e.g. switching from application A to application B). With 2GB available, windows should run out of excuses to swap but it will still swap.

    Disabling swap space effectively stops this behavior. Especially on slow harddisks this means a huge performance improvement. Depending on your software you can do with much less memory. I've disabled swap space on machines with only 512MB which you are unlikely to exceed running just office type applications. In all cases that I did this the result was an immediate, noticable performance increase.

    In case you do run out of memory, you get an out of memory error. I find that closing applications usually is a good solution. Much better than windows continuously wasting my time with unnecessary UI blocking harddisk activity. Anyway, given the low cost of memory, I'm very intolerant towards having my time wasted due to the fact that there's not enough.

    • Re:swapping is the bottleneck (Score:1, Insightful)

      by Anonymous Coward
      Funny thing, I almost always end up doing this even on Linux laptops, and it seems to make the laptop a lot more stable. In this day and age of plentiful cheap RAM, OSes should default to swapping disabled and only prompt the user to enable it when the user first runs something that commits > 95% of memory. Swapping is an anachronism from the days of $100/M RAM, and most data-intensive apps do their own application-aware swapping anyway (photoshop, databases, etc.).

      Don't even get me started on the supidi
    • Yeah, I often do something similar: I fix the amount of swap space to a set amount (e.g. 1gb). It seems that Windows wastes a lot of time resizing the swap file and I've seen noticeable reduction in thrashing when you stop it from doing this. A useful alternative when you don't have the large quantities of RAM to prohibit swapping completely.
      • Sometimes that can worsen performance as well. Windows will just swap more often if the swap size is smaller as it loves to use virtual memory and slow everything down. At least unix is smarter in that it only swaps if its needed.

  • The results are clear... ? (Score:3, Insightful)

    by FFFish ( 7567 ) on Friday April 28, 2006 @01:07PM (#15222070) Homepage
    Are they really? I'm on an iBook G4 w/768Mb of memory. I hardly ever shut the machine off; it does sleep mode perfectly, so there's never really a need to reboot. When I'm working, it's on only a few applications at a time; the second or two it requires to load a seldom-used application is so rare that I honestly can't see how a faster hard drive would provide me much benefit at all.

    In all honesty, the slowest thing about my computer is me. Even if an app were to load instantaneously, my brain is still gonna spend a few seconds getting its shit together to actually use the application, let alone do anything truly useful with it.
    • I agree. I most often use my laptop for playing DVDs (lotta help from the fixed drive there) and my wife uses it for browsing (which using firefox she can do while any remaining stuff loads). Please give me the slowest, lowest power, cheapest hard drive available.

    • So true, I ve myself the same problem. Do you know where i can upgrade this component?

      I m also looking for an extended waranty. The manufacturer (MOM Ltd.) is really cheap on this an I begin to see some degredation in performance and stability.

    • Consider that what you do with the computer will have an enormous impact on what you ask of the HDD subsystem. If you develop software at all, running a compiler on a complex project with hundreds of files is very I/O intensive. Improved hard drive performance can make a tangible difference to the compile time (and therefore development-test-debug cycle time) for many developers.
  • ...the fastest advancing high-tech product...

    One thing's for sure. It ain't Lithium-Ion battery life [wikipedia.org].

  • Pokey Hard Disks (Score:2, Interesting)

    by aarmenaa ( 712174 )
    Quite frankly, I feel that 30% understates how slow some laptops feel because of their slow hard drives. I don't really consider myself to be terribly impatient, but a fast hard drive is the best reason I can think of to use a desktop whenever possible. However much as these drives bother me, they do have thier merits. The 4200 RPM drives consume noticeably less power, and are much cheaper, though capacity still has more sway on the bottom line.

    There's a nice middle ground for laptops, I think. When my
  • There is plenty of trade offs to be made on hard-drives.

    Size, weight, power consumption, price, capacity, rotation speed, platter density, and ruggedness.

    For mobile harddrives, the number of platterns is verylimited by size, and platter size itself is limited.
    So they never get anywhere near the capacity of its desktop equivalent. However, less and smaller platters means also less power consumption so its still another point why its so in mobile computers.
    The platter density helps capacity, and bandwith and
  • I wish people would stop staring at the RPM when judging hard drive performance. It's equally informative as the MHz of a processor, just a single factor in a complex performance equation.

    As an anecdote, my 4200 RPM laptop drive is much faster than my 7200 RPM desktop drive, both using ATA 100 and disk-bound applications. I think one reason is that the laptop drive has more cache. But there are loads of other factors. Platter size is one interesting point where bigger doesn't always mean faster, especiall

  • I tend to go for the large capacity drives. Back in 2000 I picked up a 23gb behemoth for my powerbook. At the time it made it the fastest computer in the building where I worked, with more storage space than the fileservers. Bragging rights if nothing else. I always find myself saying "THIS will be the last word in hard drive space, I will never fill this". Then a year later I'm out of space and checking out the latest tech. I doubt this will ever change, and I would not be surprised in the least to h
  • Skimming the article one thing is very clear: the Seagate Momentus 5400.3 is by far the best drive for any use where noise and heat matter. It's results are very impressive.

    Note that it is the only perpendicular drive in the round-up and that the article says it is the most expensive.

Slashdot Top Deals

Beware of Programmers who carry screwdrivers. -- Leonard Brandwein

Close