Slashdot Mirror

On which planet?

According to my own recollections, which match this historical list, in 1993, low end was 40MB, high-end was ~500MB.

Most of the installs I did at the time were in the 120MB range. Anything over 200MB was rare.

That's 26 times the price. Sure SSDs are getting cheaper every day but so are hard drives. I am sure they will get so close that the price gap becomes less important than all the other features which separate them. Some time after that, SSDs may even become cheaper, or both SSDs and hard drives will be supplanted by some other technology. It just won't happen right away.

Going by this website:
http://lab-notes.blogspot.com/2007/05/historical-storage-prices-raw-data.html
and harddrive prices verified here: http://www.mattscomputertrends.com/harddiskdata.html

I took the sweetspot price of both HDD and flash for 2002:

A 128mb flash module cost $147 making it $1176 per GB.
A 60GB harddrive cost $275 making it $4.58 per GB.

$X/GB, the flash price was 256.8 times the price of hard drive. So it dropped an entire order of a magnitude since then in 7 years. Assuming everything goes along as in the past, that another magnitude of an order drop would occur by 2016/2017. That would leave flash 2.5x that of a 3.5" HDD. Of course, that would make it a comparable price for a higher-capacity 2.5" notebook, which is the size of all SSDs as well.

I don't think the wait will be until 2020. There will always be desktop computers and servers. But the growth market are small devices now. Unlike a decade ago, a notebook can be the primary machine for a great many people without much performance penalty on normal tasks, and that's what many people are buying in lieu of a desktop machine. The previous decade saw digital cameras, GPS systems, and iPods really grow the flash market, iPhone/smartphones are adding cellphones to that list. Who knows what will be next?

Really depends on what they mean by replacement.

I would argue that SSD's have already replaced magnetic drives for situations where speed is the most important factor, although there arn't really too many situations like that (and many of those can use ramdisk).

For desktops, laptops, netbooks and workstations a few GB are enough and the increase speed is much more useful. I have a 60gb OCZ Vertex SSD on my desktop. Spae wise its fine and I'm dual booting. I do have a 2ndry drive but its only used for games.

Most people won't likely need 14TB in 2020. I have 4TB currently and am using around 3TB, but I'm not most people. Obviously data centres will need bulk storage too. And 14TB would probably last me until 2020.

Another issue is that it's going to get harder and harder to cram bits magnetically (the superparamagnetic effect), I believe they are already running into some problems although there are alternative solutions such as a flat square bed that moves rather than the magnetic disk with arm for example. Flash on the other hand can just stack the chips and will likly follow Moore's law which won't be likely to run into problems until after 2020.

We could see a large migration to cloud storage for home users (obviously the backend will still be magnetic). People will stream videos rather than save them for example.

The price halving time for NAND Flash is 1.4 years, Kryder's Law says something similar: http://www.mattscomputertrends.com/Kryder's.html

http://www.mattscomputertrends.com/flashdiskcomparo.html

^ this guy disagree's, saying the transition will come as early as 2013-2014 (five years from march 2008) for 2.5" drives.

I have MFM and RLL cards still, and a 386 somewhere that will use them. Obviously that's not the case for most people, but I'm not frail or old, just a geek. ;-)
However, you're comparing comparatively small run, early hardware with modern hardware. MFM controllers were complicated for the day, (S)ATA controllers are relatively simple.

Also, the (S)ATA specs currently use 48 bit LBA, allowing for 128 PB per drive.
Assuming the HD capacity doubles every 3 years(about right based on http://www.mattscomputertrends.com/harddrives.html) starting from 2TB today, the largest HDs will be 64TB in 15 years. Coincidently, if we take a 256GB SSD and apply Moore's law, we get 32TB in 15 years. Both give a factor of 2000-4000 smaller then SATA interface allows.

In 15 years, I'm willing to bet you can buy an add in card for $30 or less, inflation adjusted, that will let you read a SATA II disk from today, if that functionality isn't still built into the motherboard of most computers.

Let's make some wild predictions based on recent price trends. (Trends found here). Over the last few years, flash memory has been increasing in GB/$ at a rate of 185% per year. Meanwhile, hard drives have slowed to only 42% improvement per year.

Based on these trends, here is the estimated cost of 10 TB using either technology:

July 2009: Platter = $750, Flash = $28,125

July 2010: Platter = $528, Flash = $9,868

July 2014: Platter= $130, Flash = $150

July 2019: Platter= $23, Flash = $0.80

July 2024: Platter= $4, Flash = $0.004

In July 2024, a 10 PB flash drive would cost $42! Of course, we can't assume these trends will continue, but it seems a good bet that we won't be worrying about the size of our mp3 collections. The traditional hard drive may only have five years of competitive life remaining.

Let's make some wild predictions based on recent price trends. (Trends found here). Over the last few years, flash memory has been increasing in GB/$ at a rate of 185% per year. Meanwhile, hard drives have slowed to only 42% improvement per year.

Based on these trends, here is the estimated cost of 10 TB using either technology:

July 2009: Platter = $750, Flash = $28,125

July 2010: Platter = $528, Flash = $9,868

July 2014: Platter= $130, Flash = $150

July 2019: Platter= $23, Flash = $0.80

July 2024: Platter= $4, Flash = $0.004

In July 2024, a 10 PB flash drive would cost $42! Of course, we can't assume these trends will continue, but it seems a good bet that we won't be worrying about the size of our mp3 collections. The traditional hard drive may only have five years of competitive life remaining.

Hard disk capacity grows in spurts though. In 2001 we had 100GB drives max or so, by 2003 that was 300GB and by 2005 it was 500GB.

But from 2005 to today, we've only gone from 500GB to 1.5TB

http://www.mattscomputertrends.com/harddrives.html

This sort of backs up my observations there... things have kind of petered out since 2005.. with a steady but slow growth compared to before.

What makes sense to analyze is not just the hard disk prices per se, but how the price sweet spot evolves in specs, as well as the sweet spot price trends. There's an interesting article about this here

I have been studying a variation on this for a while and the answer is yes.
Hard drvie growth has slowed down, or more specifically, hard drive price improvement has slowed down.
  You can see on the 1st chart on my page that the last 5 years have been a marked decrease over the previous decade:
http://www.mattscomputertrends.com/harddrives.html

Interestingly, in just the last 4 months it has speeded up dramatically. Using my standard data sources there has been an 80% price improvement in the last 4 months. Thats about the same as the last 2 years worth of growth. I think this is due to the emergence of serious solid state drives. Right now drive manufacturuers only have 4 other drive manufacturing competitors to worry about, but they will be facing some tough competition if any old electronics company in Asia can mount some chips on a board and become competition. The only solution is to maximise their competitive advantage, which for hard disks is cheap space and lots of it.

Anyone can check the spot price for flash anytime by looking at this site:

http://www.dramexchange.com/

Scroll down to the flash section.
SLC is the good stuff used in the big fast SSD's you get from people like Apple.
MLC is the slower, less long lasting, stuff commonly used in thumb drives.
$2.08 for a Gigabyte in MLC
$6.70 for a Gigabyte of SLC

If you want to know the long term price improvement rate for flash, you can join that site for $1000 a year or if you want the cheap version, I've been tracking retail flash (MLC) prices for 9 years at my site here:
http://www.mattscomputertrends.com/flashmemory.html

Spot on. Making a hard disk for a competitive price is hard. Thats why there are only a handful of hard disk manufacturers. Making a circuit board with some chips on it can be done by hundreds of companies all over the world. I cant think of any reason to buy a Seagate SSD over any one of the other hundreds of competitors, especially when they all have the same electronics inside.

Separately, it's nice to know that analysts agree with research I've done that it's only 4 years before SSD surpasses HD, at least in 2.5 inch drives. I've been comparing the relative price improvement of hard disk prices to flash and its pretty easy to estimate a crossover point.

You can have a look at my data (charts) and conclusions here. http://www.mattscomputertrends.com/flashdiskcomparo.html

I've studied the price improvement trends for flash memory and, unfortunately, it isnt quite the 4 fold yearly improvement you are suggesting. For the last 5 years it averages something about 2.6 fold.
Supprisisngly it doesnt make much difference to your numbers over a relatively short period like 3 1/2 years:
today $100 for about 26 Gig (using your starting point of $30 for 8 GB)
1 year from now - 68 Gig
2 years from now - 175 Gig
3 year - 450 Gig
4 years - 1.2 Tera

I have also studied hard drives in the same way and they will "only" be about a $50 for a Terabyte in 4 years if trends of the last 5 years continue.

http://www.mattscomputertrends.com/harddrives.html
http://www.mattscomputertrends.com/flashmemory.html

I've studied the price improvement trends for flash memory and, unfortunately, it isnt quite the 4 fold yearly improvement you are suggesting. For the last 5 years it averages something about 2.6 fold.
Supprisisngly it doesnt make much difference to your numbers over a relatively short period like 3 1/2 years:
today $100 for about 26 Gig (using your starting point of $30 for 8 GB)
1 year from now - 68 Gig
2 years from now - 175 Gig
3 year - 450 Gig
4 years - 1.2 Tera

I have also studied hard drives in the same way and they will "only" be about a $50 for a Terabyte in 4 years if trends of the last 5 years continue.

http://www.mattscomputertrends.com/harddrives.html
http://www.mattscomputertrends.com/flashmemory.html

"(right now it has been dropping linearly with density, vs. HDD's which have tended to drop price/GB exponentially)."

Well thats not right.
Flash prices/GB have been dropping dropping dramatically faster than disk for the last five years.
I've sudied it.

http://www.mattscomputertrends.com/flashdiskcomparo.html

I agree with your sentiment about what you want in a laptop.

I think your predictions are a little off though.
By 2015, if trends of the last 5 years continue, flash drives will have displaced 3.5 inch drives for storage per dollar meaning laptops will not have to make any sacrifices in the area of available storage. I think you got the drive size about right.
I looked at Windows software bloat and MS has never used more than 10% of the "sweet spot" hard drive at the time of release. I wrote about it here: http://www.mattscomputertrends.com/softwarebloat.html

The grand parent is pretty much spot on.
I have been charting the annual price improvement of flash memory and hard disks. I have data for flash going back 9 years and for hard drives, 16 years.
About 5 years ago, hard drives stopped improving at their traditional 100% a year rate and dropped back to about 40% a year. Flash on the other hand has averaged about 150% improvement a year for the 9 year period and over the last five years its gone up to 165%.
Flash actually seems to be speeding up where disk seems to have hit a slump.
If you look at where prices for disk and flash are now and do simple extrapolation you get pretty much what the grand parent said:
1 inch drives - not competitive, ceased production.
1.8 inch drives, 12 months and they are gone. Seagate gets about 20 to 25% of it's revenue from these babies.
2.5 inch drives - 4 years. In 4 years laptops will represent 66% of the market and none of them will come with hard disks.
3.5 inch drives - 6 years. Its all over baby.

Here is my data and charts:
http://www.mattscomputertrends.com/flashmemory.html
http://www.mattscomputertrends.com/flashmemory.html
and the two compared:
http://www.mattscomputertrends.com/flashdiskcomparo.html

Flash doesn't seem to have followed Moores law in its price improvement for the last 9 years. I doubt this is about to change in the next few years.

The grand parent is pretty much spot on.
I have been charting the annual price improvement of flash memory and hard disks. I have data for flash going back 9 years and for hard drives, 16 years.
About 5 years ago, hard drives stopped improving at their traditional 100% a year rate and dropped back to about 40% a year. Flash on the other hand has averaged about 150% improvement a year for the 9 year period and over the last five years its gone up to 165%.
Flash actually seems to be speeding up where disk seems to have hit a slump.
If you look at where prices for disk and flash are now and do simple extrapolation you get pretty much what the grand parent said:
1 inch drives - not competitive, ceased production.
1.8 inch drives, 12 months and they are gone. Seagate gets about 20 to 25% of it's revenue from these babies.
2.5 inch drives - 4 years. In 4 years laptops will represent 66% of the market and none of them will come with hard disks.
3.5 inch drives - 6 years. Its all over baby.

Here is my data and charts:
http://www.mattscomputertrends.com/flashmemory.html
http://www.mattscomputertrends.com/flashmemory.html
and the two compared:
http://www.mattscomputertrends.com/flashdiskcomparo.html

Flash doesn't seem to have followed Moores law in its price improvement for the last 9 years. I doubt this is about to change in the next few years.

The grand parent is pretty much spot on.
I have been charting the annual price improvement of flash memory and hard disks. I have data for flash going back 9 years and for hard drives, 16 years.
About 5 years ago, hard drives stopped improving at their traditional 100% a year rate and dropped back to about 40% a year. Flash on the other hand has averaged about 150% improvement a year for the 9 year period and over the last five years its gone up to 165%.
Flash actually seems to be speeding up where disk seems to have hit a slump.
If you look at where prices for disk and flash are now and do simple extrapolation you get pretty much what the grand parent said:
1 inch drives - not competitive, ceased production.
1.8 inch drives, 12 months and they are gone. Seagate gets about 20 to 25% of it's revenue from these babies.
2.5 inch drives - 4 years. In 4 years laptops will represent 66% of the market and none of them will come with hard disks.
3.5 inch drives - 6 years. Its all over baby.

Here is my data and charts:
http://www.mattscomputertrends.com/flashmemory.html
http://www.mattscomputertrends.com/flashmemory.html
and the two compared:
http://www.mattscomputertrends.com/flashdiskcomparo.html

Flash doesn't seem to have followed Moores law in its price improvement for the last 9 years. I doubt this is about to change in the next few years.

You can take solice in the fact that you saved some money over a flash based iPod of the same capacity. 160 gig of 1.8 inch disk is still a bit cheaper than 160 of flash storage. But not for long.
I've been studying the price improvement of hard disks and flash for a few years now. If both of them keep improving at the same rate they have for the last 5 years then 1.8 inch drives wont be viable in 12 months from now. The flash will be cheaper.

Graphs and data for this prediction here: http://www.mattscomputertrends.com/flashdiskcomparo.html

In a way, you have bought a peice of computing history, the last of the disk based MP3 players.

Yes, the price improvement of flash is awesome.
I've been studying this and if the price improvement rate of flash stays about the same as it has for the last 5 years (and hard disk does the same) it will only be 4 years before every laptop has a flash drive.

Charts and data here: http://www.mattscomputertrends.com/flashdiskcomparo.html

Yes I agree. My reserach shows that, in relation to price, the annual improvement over three years for flash comes in at 109% whereas for hard disks over the same period the figure is only 35%

http://www.mattscomputertrends.com/flashmemory.html

This means, if the two trends continue over time, it will actually become hard to justify buying a hard disk instead of flash, especially the smaller ones the cost a lot more per gig.

I studied this with hard disks and my data showed that the "sweet spot" hard disk (the drive with the best bang for buck) has actually steadily decreaased over time.
You can see the chart at the bottom of this page:
http://www.mattscomputertrends.com/harddrives.html
look in the Annual Sweet Spot Price Trends section.

Basically, my data disagrees with you. The average drive is getting cheaper.

I have studied the price improvements going back 16 years.
The long term average space improvement per dollar spent, year on year is about 100%. That is, you'd get twice as much space for your dollar each year. That was until about 4 years ago. These days you can only expect a 50% year on year improvement.

See my data and analysis here:
http://www.mattscomputertrends.com/harddrives.html

40 Meg drive had a huge 10 year life span from about 1984 to 1994.
So 13 to 23 years ago.

Given that many Slashdoters had their first computer at 10 years old, its entirely posible for you to be only 23 which for many people isnt old at all.

If you ever want to put a date on a hard drive you can use my page here:
http://www.mattscomputertrends.com/harddiskdata.html

My website on historical hard disk pricing shows that 10GB HDs were only sale roughly between 1998 and 2001. given the maximum extremes this puts the poster current age range between 15 to 20.

http://www.mattscomputertrends.com/harddiskdata.ht ml

This page is great for when you want to date a hard disk or when a certain size disk first became available.

What you say is generally true but I did find a counterexample.

The wikipedia entry on Kryder's Law, which is just Moore's law for hard disks was an example of a technical article older than 6 months, which should not have been controversial. It turned out to have some serious problems, like there never was any such thing as Kryder's law until Wikipedia invented it.

Since I originally pointed out the error, the article has been updated. You can read about what was wrong with it at http://www.mattscomputertrends.com/Kryder's.html

"Why? Flash goes down in price per capacity linearly while HDD's go down in price per capacity nearly exponentially."

Wrong!

This page charts the annual improvement of price per capacity of hard disks (amongst other things): http://www.mattscomputertrends.com/harddrives.html

This page does the same thing for flash: http://www.mattscomputertrends.com/flashmemory.htm l

Here is a key quote: "The improvement rate for flash for the last three years comes in at 109% a year whereas for hard disks over the same period the figure is only 35%."

"Why? Flash goes down in price per capacity linearly while HDD's go down in price per capacity nearly exponentially."

Wrong!

This page charts the annual improvement of price per capacity of hard disks (amongst other things): http://www.mattscomputertrends.com/harddrives.html

This page does the same thing for flash: http://www.mattscomputertrends.com/flashmemory.htm l

Here is a key quote: "The improvement rate for flash for the last three years comes in at 109% a year whereas for hard disks over the same period the figure is only 35%."

My research shows the 85% average trend hasn't held for the last 3 years:
http://www.mattscomputertrends.com/harddrives.html
It has been closer to 30%.
It also shows that the trend seems to be slowing down.

You are right. Paying double per gig doesnt seem like a rediculous premium for the drive.
It suprised me but your data doesnt show anything unusual. New large drives have always been around twice the megsper $ figure of the drive in the sweet spot.
Well, acording to my research anyway:
http://www.mattscomputertrends.com/harddiskdata.ht ml

So true.
One bad thing is that the growth of large drives seems to have slowed down dramtically in the last few years and as a consequence the improvment in bang per buck of "normal" drives has also slowed down.

I've been studying this for a while now. You can see the trend for youself at my site, http://www.mattscomputertrends.com/harddrives.html

Sorry for a second reply but I just noticed this:
"We'll see how Moore's Law pans out... but there is a limit, eventually, with our data (although not for a while) and our ability to fill drives."

Moores law for hard disks is called Kryders law and Kryders Law, is already broken.

Yes 60MB disks were shiped when 750 MB drives were new to the market. It happened in 1994. Actually, it seems 1.4 Gig drives were shiping concurrently with 40 MB drives

Source, my site, here: http://www.mattscomputertrends.com/harddiskdata.ht ml

Witty or sensational headlines don't just deceive search engines.
Human readers can get fooled just as easily. Heres an example:

I was doing research to show that Kryder's Law (a kind of super Moore's Law for hard disks that says bit densities have increased factor of 1000 in 10.5 years meaning a doubling every 13 months) is no longer being achieved by hard drive manufacturers. Instead I discovered that Kryders Law was just a creation of Wikipedia's overenthusiastic editors that misinterpreted a single Scientific American headline. Wikipedia editors accidentally invented the "law", and it isn't even correct.

You can read about it at my site here: http://www.mattscomputertrends.com/Kryder's.html

The search engines are dong us all a favor getting rid of this problem.

Can you provide the link for the Economist article. This is an area of interest for me.
My own research shows the opposite is happening. Flash is charging hard after disk and the rate it is catching up is accelerating.
http://www.mattscomputertrends.com/flashvsharddisk .html
I am due to update this years figures but a quick analysis shows the trend is continuing.

Ram has worsened, relative to hard disks in the last 14 years. Thats because hard disks have improved about 105% each year in the megs per dollar figure but RAM has only improved about 70% each year. This has meant RAM has fallen way behind.

Figures come from my research available on my site:
http://www.mattscomputertrends.com/

I have looked into the performance trends of flash vs hard disk and I suspect, in a just few years, that flash will catch up to the point that there is no meaningful difference in the sustained read and write speeds between the two. This means when the $ per meg is the same, there will be no logical reason to chose disk over flash, including in a hybrid setup. Could be a few years yet for desktops, 11 years for 3.5" IDE drives according to my study. Probably a lot sooner for 2.5" drives.

I've linked to it elsewhere here but here is the link to my study again:
http://www.mattscomputertrends.com/flashvsharddisk .html

I did a study which estimates that flash will surpass 3.5 inch IDEs in every price by 2017.
Read about it here:
http://www.mattscomputertrends.com/flashvsharddisk .html

Today, you can buy a 256MB USB Watch without much trouble at all. I dont see any issue with these novelty USB drives being available in reasonable sizes at the time of pruchase.

The real problem is that the sweet spot for USB drives have acheived an improvment about 80% each year. The means your USB watch seems out of date in only 2 years.

Betting on Moores law is a smart idea. MicroSoft has done it for more than a decade and won every time.

The Megabytes per $ for flash has averaged about 130% improvment each year over the last 5 years. There is no reason to think this year will be any different.

If you are saying that you can only afford a 128 MB flash drive right now, then it will only be about 10 months before you can afford a 256 MB one.

"There are 4 considerations when looking at a storage medium"
I think power usage will also be a big consideration going foreward.

"Hard drives have probably increased in capacity faster over the long term"
I know that you have said found my site http://www.mattscomputertrends.com/ but I am not sure if you read the capacity trend info. The hard disk capacity trend is about 104% each year but flash comes in at 130%. Feel free to dispute the data, though.

" I'd predict this to happen in about 2013."
I know you retracted your paragraph 4 but I am pretty impressed that some one could come up with such a close guess to mine on what I assume is just off the top your head calculations.

"One minor point I'd like to get back to is the low number of writes allowed to flash RAM devices."
Wear leveling algorithms just move the high traffic data (like swap files) to lesser used areas of the flash as needed, evening out the wear across the whole flash storage device over time. These are yet to implemented in major operating systems but will probably be quickly implemented when flash as primary storage becomes viable. Wikipedia has good references about these file systems.

"However, I actually expect hard drive capacities to slow their rate of increase, so it could happen earlier." AND "Hard drives are doubling in capacity at a rate of probably more than 18 months." I think these two statements show you have noticed something that I thought noone else had discovered. Kryder's law is dead. Kryders law is moores law for hard disk capacities. It says they should double every 13 months. The other way of saying this is a 1000 fold increase in capacity every 10.5 years. The second way evens out short term bumps and dips. In March when I am due to do the 2006 figures on my site, there will be solid evidence that the trends are way behind Kryders law. I'll do a story about it and submit it here (where it will be rejected, haha).

Actually, this is predicted this to happen in 11 years from now, not in 2006.

http://www.mattscomputertrends.com/flashvsharddisk .html

They didn't just make up the 11 year figure either. The prediction is based on price trends from the last few years.

The article also explains why performance and maximum write issues will not be an problem by then.

I think you got it right.

A 1 Gig flash card for my camera could hold about 600 images. Considering 100,000 writes is usually given as a minimum failure point that means the total number of images written can easily be in the billions. If I took 1 photo every second it would still take 31 years (!) to get to a billion. (31 years = 978 264 705 seconds, ask Google).

With the application of wear leveling algorithms even swap space applications are viable.
My site http://www.mattscomputertrends.com/flashvsharddisk .html points out that a theoretical example of a flash drive of 200GB with a write speed of 40 Megabytes per second and doing some basic calculations shows that it could be written to continuously for just over 15 years before every block passed the 100 000 write mark. The equivalent of todays 200GB drive some 15 years ago was the 210MB disk. There are not many machines running today with 210MB hard drives, let alone dong the kind of work that requires continuous writing to the disk.

The write limits of flash are basically at a point where they are not worth worrying about.

Very perceptive.

I did a small study that shows the rate of improvement in capacities is slowing down. My study was on the "sweet spot", the drive size with the most megs per dollar but the trend should be across the board.

It really seems like Hard Disk technology has hit a wall.

http://www.mattscomputertrends.com/harddrives.html - the second chart is the one that applies here.

Actually, I say they new hard disks will be obsolete in just 11 years.

Read about it here: http://www.mattscomputertrends.com/flashvsharddisk .html

The gist of it is that right now your dollar buys about 130 times more hard disk space than flash memory. In almost every year, you can buy more space for your dollar than you could last year. This improvement for hard disks in the last two years was measured at 44% per annum. The annualised improvement for flash storage over the same period was measured at 118%. By simply extrapolating these figures into the future until the megs per dollar figure for flash beats that of hard disks gives the date of 2017 or in just 11 years time.
 
The rest of it covers why performace shouldnt be an issue is 11 years time.