Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
So you need to give some credit and explanation of WHY SSD has been adopted so quickly...
Actually, your second point demonstrates that I don't need to give credit/explanation...
As far as HDD being replaced eventually, I sure hope people realize that's for everything. SSD's will be replaced with whatever surpasses that. And then again, and again and again. All technology has a shelf life.
This is true, and why it's always both amusing and disconcerting to see anyone claiming that xzy technology, be it tape drives, floppy drives, CDs, DVDs, HDs, or whatever, is "here to stay". The assumption that it won't happen because the new tech isn't good enough right now to completely replace the previous tech is always absurd. In the case of SSDs vs HDDs, SSDs are at the beginning of their life cycle while HDDs are nearing the end of theirs. As I said before, it's just a matter of time.

----------

(I've given up all hope of ever being able to upgrade Apple's paper-thin anorexic laptops in the future)
I suppose you want to be able to upgrade the Apple watch too?

It's time you accept the fact that the "upgrades" you like to do are themselves obsolete.
 
The thing some of the dogmatic comments on whether there's a future for rotating hard drive media here seem to be missing is the price/storage curve.

Remember, for example, that a lot of servers traditionally used 15K RPM 2.5" HDDs for much of their storage, and a 600GB drive (the highest capacity available) from Seagate in that category currently runs $500. The price/performance ratio is already skewed in favor of SSDs in that space unless you need extreme write endurance.

Currently, since 256/512GB SSDs have reached relatively reasonable price points, a significant number of users simply don't need more, because they just don't have that much data. I've started defaulting all internal storage at work to 256GB SSDs, and only two or three workstations out of twenty will need external bulk storage to compensate for the limit. Our server, shared by the entire workforce, only has 4TB of RAIDed data, so if I wanted to bump it to SSD I could without a horrifically high expense, although for that volume HDD still has the advantage. I have a 20TB RAID box at home for cheap bulk storage, but my 512GB SSD holds nearly everything I need, and all of my regularly-accessed data, apart from movie archive, is around 1TB, so it's not going to be too long from the looks of it before I'll be able to cost effectively switch my main working array from 3 RAID4'd 2.5" HDDs to SSDs.

In any case, though, if you look at the future, here's the issue: rotating storage media areal density has hit a wall the last few years. Increases in max capacity in 3.5" HDDs have been almost entirely due to more platters, not more data per platter. Prices per GB have fallen, but there's a hard floor created by the material cost of the chunk of aluminum, the motor, the platters, and the write head and PCB, so it will only go down so far.

The sole exception is the just-released 8TB SMR drive from Seagate, but if you read up on it, it has horrifically bad random write performance due to the way SMR writes data. This might be addressable with different firmware in future drives, but there's no proof of that so far.

Basically, storage of any kind follows a curve, but you're going to eventually hit some kind of leveling off due to physical constraints. They say the only people who think exponential growth can go on forever are economists and the insane, after all.

Hard drives, at least currently, appear to be getting to the mature, leveled-off part of the curve. SSDs do not appear to be there yet.

Now, it's possible that better application of SMR or some other new development will put HDDs back on the path for continued growth for a while. And it's possible SSDs will hit their mature leveling-off state sooner rather than later. We won't know either for sure until a few years passes. But just looking at the last two or three years of history in terms of max capacity and price per GB of storage, it sure looks like HDDs are getting mature and SSDs are still in the growth phase.

The question is if and whether the SSD curve is steep enough that it'll eventually just pass HDDs and make them largely obsolete, or whether they'll level off, too, making HDDs a continued viable bulk storage technology.

Even if so, though, a huge percentage of average users simply do not and will not need more than a couple hundred GB of storage, so they have no personal need to buy an HDD, whether their stuff in the cloud uses them or not.
 
Last edited:
The question is if and whether the SSD curve is steep enough that it'll eventually just pass HDDs and make them largely obsolete, or whether they'll level off, too, making HDDs a continued viable bulk storage technology.

Even if so, though, a huge percentage of average users simply do not and will not need more than a couple hundred GB of storage, so they have no personal need to buy an HDD, whether their stuff in the cloud uses them or not.

In the context of laptops, it's already obsolete. But there's nothing preventing the use of HDDs as external storage devices with a different space/performance ratio, in a NAS for example.
 
Actually, your second point demonstrates that I don't need to give credit/explanation...

This is true, and why it's always both amusing and disconcerting to see anyone claiming that xzy technology, be it tape drives, floppy drives, CDs, DVDs, HDs, or whatever, is "here to stay". The assumption that it won't happen because the new tech isn't good enough right now to completely replace the previous tech is always absurd. In the case of SSDs vs HDDs, SSDs are at the beginning of their life cycle while HDDs are nearing the end of theirs. As I said before, it's just a matter of time.

----------


I suppose you want to be able to upgrade the Apple watch too?

It's time you accept the fact that the "upgrades" you like to do are themselves obsolete.

The way you talk down to people is disconcerting.

Most of us here would interpret "here to stay" to mean, here to stay for the foreseeable future. No matter how highly you think of yourself as a futurist, none of us can actually predict what next great thing in technology will supplant another.

I will 2nd the opinion that HDD are here to stay for quite a while.
Since its invention in 1956 it might have changed from refrigerator size down to today's common 2.5" spinners, but the basic concept has remained unchanged in nearly 60 years.

Will it still be used in 60 years from now ? Probably not, but neither will SSDs.
To argue lifecycle you'd have to make assumptions about the future. They've only been around in practical form a few years, unless you count EEPROMs from the 80s.

Maybe you hold stock in an SSD manufacturer, but like so many other transitional media storage, SSDs current advantage is just speed as a secondary storage medium, but at a 10x cost to HDDs.

Longevity is questionable for many reasons, ESDs, EMPs and other environmental influences. Each storage medium has advantages & disadvantages. Femtosecond lasers etching dots into glass silica or capturing DNA in silica glass spheres are being touted as potentially "million year" storage mediums:
https://www.ethz.ch/en/news-and-events/eth-news/news/2015/02/data-storage-for-eternity.html
 
The thing some of the dogmatic comments on whether there's a future for rotating hard drive media here seem to be missing is the price/storage curve.

So you proceed to make some dogmatic comments yourself :rolleyes:

Our server, shared by the entire workforce, only has 4GB of RAIDed data

Really? 4GB RAIDed ? That much ? :)

Even if so, though, a huge percentage of average users simply do not and will not need more than a couple hundred GB of storage, so they have no personal need to buy an HDD, whether their stuff in the cloud uses them or not.

Reminds me of Bill Gates' infamous 640K quote.
Most photographers & videographers here will disagree with that statement.
Furthermore, "the average users" is making a huge assumption.

Many people I know already store terabytes worth of personal media at home.
 
Maybe annotating this with timelines would be a useful exercise so we know what positions we're arguing from. I'd be surprised to see global HDD unit shipments fall 25% in the next 5 years.
Magnetic hard drives are a dying product. They will exist as a niche product sort of like film. The company that was Kodak recently agreed to continue making film for a few people in Hollywood, the fact that they did this doesn't mean that film is alive. It is effectively dead, the same thing will happen to magnetic disks, they will eventually wither and die while a few Luddites try to keep the industry alive.
I'm going to skip responding to stuff that doesn't look productive and start by responding to your conclusion. If I've failed to respond to something you consider a particularly insightful statement, let me know.

It looks like you have managed to substitute your personal experience for a global market trend. Did you look at the link I included? Shipments are continuing to grow, which I find to be a strange definition of "dying". Maybe in the nihilist sense of a 30 year old man is dying because, in the end, aren't we all? SSDs aren't long for this world either. That sounds profound, and visionary, but not terribly useful.

SSDs have brought us new capabilities in storage. They're wonderful for many things. HDDs will eventually fade away, along with magnetic cores and punch cards-- but we're not even seeing the decline of unit shipments yet, let alone a collapse. So I don't think we can say when HDDs will die away any more reliably than we can say what technology will eventually replace them en masse.

Maybe I'm wrong and we're just a year or two away from the utter collapse of the hard drive market, but I'm not seeing any evidence of it and this sounds like the kind of seismic change that would be well telegraphed. If Google, Apple, or Amazon thought they were going to replace their spinny disks with flash in the next few years, they'd be hollering far and wide to ensure they had the supply.

(The reasons for keeping film are unrelated. In the analog domain, the medium affects the information it holds so has value beyond its storage capacity. Art versus archive.)
Beyond that there is one factor that you and deviant don't seem to consider and that is the cost of a hard drive crash.
Of course, if you're a corporation, there's a good chance you have your backups on tape-- which also hasn't died 60 years on:
http://www.sony.net/SonyInfo/News/Press/201404/14-044E/

Seriously, read the link this time.

What is the differential cost of protecting against a hard drive crash versus an SSD or SD system failure?
Not as rapidly as affordable SSD technology. This is in a nut shell the basis of the argument that hard drives are dying.
As this is the crux of the argument, it would have been a good place to insert data... I'd really like to see some support for this argument beyond "SSDs are awesome" and "everybody knows that next year is going to be more awesome than last year"
As the capacity of flash technology increases the need to push out to rotating media diminishes significantly. Beyond that you miss the alternative solid state solutions that can replace those external rotating solutions. SD cards for example are often fast enough to serve as a secondary store.

As far a data centers go, SD's can be justified simply based on the power they save. Of course not all data centers are the same but your idea that the could is entirely rotating media needs to be reexamined.
As the cost of rotating media and high bandwidth communications falls, the trend has clearly been to limit the amount of expensive flash storage in our devices and rely on a redundant secondary bulk storage, often as a service. It started as syncing portable devices to desktop devices and is now evolving to syncing everything to the cloud-- which is entirely driven by rotating media.

Are you aware of a major data center that has replaced magnetic media with banks of SD cards? How does this play with your reliability argument earlier?
 
Last edited:
SSDs have brought us new capabilities in storage. They're wonderful for many things. HDDs will eventually fade away, along with magnetic cores and punch cards-- but we're not even seeing the decline of unit shipments yet, let alone a collapse. So I don't think we can say when HDDs will die away any more reliably than we can say what technology will eventually replace them en masse.
Many large companies have been very successful right up until they failed. The 3.5" floppy drive was required on any decent personal computer, right up until it wasn't. Same with the DVD drive. Why do you think hard drives, or even SSDs at some later date, will be different?
Maybe I'm wrong and we're just a year or two away from the utter collapse of the hard drive market, but I'm not seeing any evidence of it and this sounds like the kind of seismic change that would be well telegraphed.
It is well telegraphed. Like it or not, the IT departments are not where the real change comes from these days. Consumers drive the market, and consumers aren't interested in hard drives. Eventually, IT will have to follow suit as the economics will change because of consumer demand.
Are you aware of a major data center that has replaced magnetic media with banks of SD cards?
Are you aware of any data center that has no SSDs? Change on this scale always starts slowly. First you have no SSDs, then some, then the economics change and the trickle becomes a torrent.
 
Many large companies have been very successful right up until they failed. The 3.5" floppy drive was required on any decent personal computer, right up until it wasn't. Same with the DVD drive. Why do you think hard drives, or even SSDs at some later date, will be different?

It is well telegraphed. Like it or not, the IT departments are not where the real change comes from these days. Consumers drive the market, and consumers aren't interested in hard drives. Eventually, IT will have to follow suit as the economics will change because of consumer demand.

Are you aware of any data center that has no SSDs? Change on this scale always starts slowly. First you have no SSDs, then some, then the economics change and the trickle becomes a torrent.
You are inadvertently or intentionally misreading what I'm saying if you think I'm arguing that spinny drives are never going away.

Wizard was arguing for SD cards in data centers, the comment was not about SSDs.
 
take even full hd. you'd have to be filthy rich to work with uncompressed full hd videos on ssds. a lot of people make full hd videos. look at youtube. cmon now

And yeah, unfortunately i realize it's like that. That's why apple PRO software is disappearing, everything is dumbed down etc etc. Well, that's how businesses work. It sells, so no complains.

You do realize that there are other Pro uses for Macs besides editing uncompressed HD/4K video, right? I use mine (15" rMBP) for development purposes and the 500GB SSD is more than plenty for everything I toss at it, including a few games, photos, and even video.

I'm not disagreeing that if you need a ton of storage that a HDD isn't a great solution. But that doesn't mean Apple should keep building HDDs in their machines. Let that stuff live in USB/Thunderbolt/NAS enclosures instead.
 
The sole exception is the just-released 8TB SMR drive from Seagate, but if you read up on it, it has horrifically bad random write performance due to the way SMR writes data. This might be addressable with different firmware in future drives, but there's no proof of that so far.

Basically, storage of any kind follows a curve, but you're going to eventually hit some kind of leveling off due to physical constraints. They say the only people who think exponential growth can go on forever are economists and the insane, after all.

Hard drives, at least currently, appear to be getting to the mature, leveled-off part of the curve. SSDs do not appear to be there yet.

Now, it's possible that better application of SMR or some other new development will put HDDs back on the path for continued growth for a while. And it's possible SSDs will hit their mature leveling-off state sooner rather than later. We won't know either for sure until a few years passes. But just looking at the last two or three years of history in terms of max capacity and price per GB of storage, it sure looks like HDDs are getting mature and SSDs are still in the growth phase.

The question is if and whether the SSD curve is steep enough that it'll eventually just pass HDDs and make them largely obsolete, or whether they'll level off, too, making HDDs a continued viable bulk storage technology.

Even if so, though, a huge percentage of average users simply do not and will not need more than a couple hundred GB of storage, so they have no personal need to buy an HDD, whether their stuff in the cloud uses them or not.

SMR will get better. It's in its infancy at this point. But at the same time, it's not quite clear how much better it can get. If I was you, I'd take a look at two technologies. HAMR and Bit Pattern media. There's always the possibility of using these two together... not that I would know ;)

As for you last paragraph, I have to greatly disagree. You're data usage will continue to increase greatly. Music, movies, and the ever increasing size of the photos you take with your smartphone. If you took your whole data storage into consideration, you'd be quite amazed at what you have... and it's only going to grow more.

Edit: I find it interesting that people who argue my points haven't touched on the fact that demand is growing and will exceed manufacturing capabilities in a few years. SSD's are not the solution for that. HDD are.
 
Last edited:
Not even close. HDDs are for storage, SSDs are for the system. Is it THAT hard to understand? I hate to read this stuff over and over again. HDDs are CHEAP. You couldn't possibly afford a 50TB NAS storage filled with SSDs.

Around the time where it cost alot to get a 128 gb ssd maybe, but nowadays SSDs have moved to either full or partial storage.
 
Nope. And that 10TB SSD will cost 10 times more than even 8TB HDD.
I have SSDs in all of my machines. It's not like i'm ANTISSD or something. It's just that HDDs are here to stay. I have a movie collection that would cost 3000$ to store on SSDs. That would be the dumbest thing to do.

Are you kidding me? You say SSD's aren't for storage and your examples are UNCOMPRESSED STORAGE OF 1080P VIDEOS (On youtube or otherwise)???!!!

1080p (And 4k) uncompressed, or slightly compressed, are EASILY the toughest thing for HDDs! Even with todays 10-50 gb games! Games give you hours of entertainment while 1-3 hour movies take up around 10-40 gb!

Do you not even see how absurd your argument is?? It's like saying nobody can afford food because of how much one fat dude eats every day! It's like saying nobody can afford gas because of how much truckers spend on it driving goods down the road!

You are using THE MOST EXTREME FORM OF DATA STORAGE AND CALLING IT NORMAL. NOT EVEN PEOPLE WHO WORK GENERALLY KEEP THAT MUCH DATA.

For most people today, ssds can provide either full or partial storage of data. You are trying to act like ssds are 60 gb like several years ago.
 
Are you kidding me? You say SSD's aren't for storage and your examples are UNCOMPRESSED STORAGE OF 1080P VIDEOS (On youtube or otherwise)???!!!

1080p (And 4k) uncompressed, or slightly compressed, are EASILY the toughest thing for HDDs! Even with todays 10-50 gb games! Games give you hours of entertainment while 1-3 hour movies take up around 10-40 gb!

Do you not even see how absurd your argument is?? It's like saying nobody can afford food because of how much one fat dude eats every day! It's like saying nobody can afford gas because of how much truckers spend on it driving goods down the road!

You are using THE MOST EXTREME FORM OF DATA STORAGE AND CALLING IT NORMAL. NOT EVEN PEOPLE WHO WORK GENERALLY KEEP THAT MUCH DATA.

For most people today, ssds can provide either full or partial storage of data. You are trying to act like ssds are 60 gb like several years ago.

I remember the same argument about CRTs when LCDs first came out.

Bahh! LCD color accuracy are color temperatures are horrible, they are expensive and low resolution! REAL Desktop publishers will NEVER use LCD for production work!
 
but in this case they didn't? they just implemented existing techniques

What people here have a problem with is that "pushing the envelope" not necessarily means "innovating".

They're always among the first to adopt a new tech or discard the old one. Some call that innovation for some reason

----------

Having a collection is not about where watching it. It's about "having" it. As in "i can watch it whenever i want, in the highest quality there is"
But i realize you tried to make appear yourself as some kind of an "elitist" who watches movies in theaters. That little sarcastic note was received, but i don't really care. For all i care you could watch your movies on a iphone 4. Having a collection has absolutely nothing to do with watching/not watching movies in theaters. So go brag somewhere else :p

Also, you said what, 300gb? You must have those 97 movies in 1400MB Dvdrips or something, lol.

That means you're a hoarder.

If you were doing that with real thing, someone would come and take you to a mental institution.

But seriously, your argument is completely invalid. You can make any irrational choice and claim some computer technology isn't going to cut it.

i.e. "5k iMac? Lol. I Can't watch 97 movies simultaneously on full resolution on that!" "It's not about watching a movie, its about being able to watch 97 movies at the same time"

see?

In all reality, broadband has more or less changed the way people do stuff. They download a movie (legal or not), watch it, delete it. Or they just stream it directly.

And when 4K comes, you're going to need a lot more space... And there is a limit of how dense can you make a hard-drive.


oh well
 
I have no idea why deviant's statement was even the least bit controversial... He was responding to the idea that HDDs are dying out. They aren't.

http://www.forbes.com/sites/tomcoughlin/2014/02/10/hdd-decline-expected-to-slow-and-reverse/

Show me someone with an SSD Time Machine and I'll show you someone with more money than sense.

SSDs are great for pulling in small amounts of data quickly, but they are incredibly expensive for bulk storage. If accessing a particular block of data is a significant fraction of system time, SSD is the right place for it. If it's only accessed occasionally, HDD is the right place.

Flash is getting cheaper, but our appetite for data is growing rapidly as well.

The trend we're seeing is to put SSDs in portable devices, and push bulk storage out to secondary rotating media or, more commonly now, the cloud. The cloud is entirely driven by rotating media.

So no, HDDs aren't dying. They're just less visible to most people.

I agree. I've got two 256GB SSDs running game in RAID 0 for my main system drive, but I've got over 20TB of external HDD full of movies/TV shows. While I agree that SSD prices will come down, they will do it in conjunction with HDD capacities increasing. Unless there is some disruptive technological breakthrough in digital information storage, HDD aren't going anywhere soon.
 
And yeah, unfortunately i realize it's like that. That's why apple PRO software is disappearing, everything is dumbed down etc etc. Well, that's how businesses work. It sells, so no complains.


Do you even use any of the Apple PRO software?
FCPX can do more, plays better with external audio (the dreaded audio sync drift, heard about that?)

Logic Pro X has *one* arguably necessary and unprofessional function removed, and added billions compared to LP9.


Yeah. but you still have to store everything you've done SOMEWHERE after that, right? ;)

No actually, once you publish, you can discard it.

I agree. I've got two 256GB SSDs running game in RAID 0 for my main system drive, but I've got over 20TB of external HDD full of movies/TV shows. While I agree that SSD prices will come down, they will do it in conjunction with HDD capacities increasing. Unless there is some disruptive technological breakthrough in digital information storage, HDD aren't going anywhere soon.

HDDs aren't dying out, but for an average consumer, they're gonna pretty soon.

Hoarding 20TB of movies/tv shows is something I won't go into... but technology is changing. HDDs will be used for archiving, replacing Magnetic Tape.

There's a physical limit of how dense can you make a HDD platter...
 
These new standards are always what screw over custom PC users. New Macs come out that are faster in some way that they can't match with an upgrade (short of a full motherboard/SSD replacement). As it is, a MacBook Air has a faster SSD than the average 1337 gaming PC :cool:

----------

I agree. I've got two 256GB SSDs running game in RAID 0 for my main system drive, but I've got over 20TB of external HDD full of movies/TV shows. While I agree that SSD prices will come down, they will do it in conjunction with HDD capacities increasing. Unless there is some disruptive technological breakthrough in digital information storage, HDD aren't going anywhere soon.

Completely true. There was an article on MR about a recent breakthrough in SSD tech that would allow for 10TB SSDs, but I don't know if that means they'll get any cheaper.
 
I have gone completely SSD and not looked back.

Was the up front cost expensive? Yes.
Was it worth it? Absolutely!
Would it be nice to have even more space? Yes, but that isn't a new thing with the invention of the SSD.
 
So you proceed to make some dogmatic comments yourself :rolleyes:
I made absolutely no dogmatic comments. Every comment I made was either speculative or based on specific information and in no way judgmental. The only nontechnical comment I made was that for some significant percentage of computer users, they simply don't have enough data to need more than the 200-500GB that an SSD can provide at a reasonable price point--something evidenced by the fact that Apple doesn't even make laptops with an HDD option anymore, as well as ample real-world evidence. Not that everybody has modest storage needs, but many people fall into that category.

Otherwise, I just pointed out that the increase in areal density curve of rotating media has flattened out considerably in the last several years. There are lots of exotic technologies that could bump it up, of course--HAMR, MAMR, TDMR--but the only one that's currently feasible on a commercial scale is SMR, which as implemented thus far isn't viable as a consumer storage technology.

SSDs, at least so far, appear to be on a sharper part of the $/GB curve, based entirely on real-world products available.

This doesn't mean that HDDs aren't a competitive technology for bulk storage now--they are. It doesn't mean that they might not continue to be so. But at least based on current trends, the decreasing $/GB curve for SSDs vs HDDs has them on course to intersect eventually, at which point HDDs will have little if any advantage past write endurance in very-high-throughput environments.

SSDs might flatten out suddenly, or one of the upcoming HDD technologies might cause a big spike in HDD $/GB performance. It's really impossible to say at this point.

Which was my whole point--it's dogmatic to claim certainty one way or the other based on where we currently stand. The eventual results will be determined entirely by how the $/GB curves look over the next 5 or 10 years.

Most photographers & videographers here will disagree with that statement.
Furthermore, "the average users" is making a huge assumption.

Many people I know already store terabytes worth of personal media at home.
Yes, I have a cumulative 30TB of rotating media storage at home. Many videographers and pro photographers have even larger data storage needs than that.

Currently, for me and them, rotating media is far more cost effective as bulk storage. You'd be silly to argue that, and I'm not--otherwise I wouldn't have boxes full of HDDs myself. Bulk-storage cloud companies obviously also take huge advantage of cheap $/GB storage, as do other organizations with "big data" needs.

But assuming that I, you, and a professional photographer are broadly representative of the entire user base of personal computers is a little silly. Many people do not need 30TB of storage at home. As a percentage, I can't tell you what fraction of the hundreds of millions of computer users in the world need 200GB or less, and what fraction need more than 1TB, but I'm pretty certain that there is a significant percentage of users in that <200GB bin. Whether it's 40%, 60%, or 90%, I can't say, but it's more than a few.

Based entirely on my own observations of workgroups I admin, of the 10 office/CAD users in a manufacturing business, only 1 stores enough data to need an HDD, and even the shared server is under 1TB. Of an office of 22 energy researchers, 3 store enough data that an HDD becomes a cost-effective necessity; the rest are all under half a TB (and yes, I typo'd 4TB for the shared storage). Of the 7 friends and family members I do IT for, myself included, 3 have large bulk storage needs.

That's a small sample, of course, and I'd love to see an actual graph of used-storage-space in a statistically significant survey, but in any case the point stands that for some substantial number of users, they just don't need multi-TB bulk storage now or for the foreseeable future. Apple certainly seems to think so, or they would still be shipping a laptop with internal HDD storage.

Whether they will 10 years from now or will continue to rely mostly or entirely on the cloud, and where the HDD and SSD $/GB ratios will stand at that point--not to mention whether SSD $/GB will grow faster than average data storage needs--is pretty much just a guess at this point. The current trends appear to be in favor of SSDs in the long term, and there are plenty of very viable exclusive use cases already, but it's impossible to say for sure.
 
As for you last paragraph, I have to greatly disagree. You're data usage will continue to increase greatly. Music, movies, and the ever increasing size of the photos you take with your smartphone..
I'm skeptical about that for the "average" user. While in the cloud these things may continue to be stored on rotating media until and unless SSDs become cheaper per TB than HDDs, if anything the trend seems to be toward less local media storage.

Music? Increasingly, as evidenced by actual market share, streamed. Videos? Streamed. Even personal photographs are migrating toward the cloud in general.

And my point is more generally, if the "average" user needs (these are completely fabricated numbers, of course) 20GB of storage on their smartphone and 150GB of storage on their laptop today, there is some % factor by which their storage needs will increase every year due to taking more photos, recording more videos, and fatter apps. But it's not at all clear to me that the $/GB curve for SSDs (HDDs as well, of course) isn't decreasing more sharply than those needs are increasing.

That is, if my mom has only 170GB of data today (so an SSD is cost effective), and her needs are for 10% more storage every year, but you can get 20% more GB of SSD per $ every year, then she will never reach a point at which she starts needing the $/GB advantage of an HDD, no matter how huge or cheap HDDs get.

Thus far, in my direct personal experience supporting a few dozen users across a couple of small organizations, SSDs pretty rapidly passed the point at which it was cheap enough to put all of a majority of users' data on them, and the increased need for storage over time isn't nearly as rapid as the decrease in $/GB of SSDs.

There are, of course, those minority (in the world at large it may be a substantial minority--not sure) who do have greater storage needs (I myself am in that category, with tens of TB of storage at home), who are better served by HDDs at this point. And may continue to be. Time, and whether HAMR, MAMR, TDMR, and SMR successfully ramp up to commercial and/or consumer viability faster than exotic technologies that push the $/GB proposition of SSDs down do, will tell.
 
Last edited:
Not sure what you are all arguing about here. The SSD is going to kill HDD for internal (e.g. OS) storage, but is not going to make HDD disappear from the face of earth. There are still lots and lots of computers (especially outside Apple ecosystem) that come with HDD internal storage. That's a shame, and it ruins the user's experience big time, as it really "buries" the machine's performance.

On a data center/cloud category, both technologies are being used at the same time. The SSDs are being used for operating system installation and a few more crucial (aka more often used) files, like system datafiles for a database etc. HDDs (usually at 10k rpm or even 7.2k) are for everything else (e.g. main storage). This is the picture and is not going to change relatively soon.
 
Show me a company that works with SSD only storage clusters. I'll wait. Google has big-ass datacenters, Google is rich. One would think they MUST have SSD storage. Right? rrrrright........... Don't talk about stuff you don't know.


Storage Engineer here. Can confirm use of Xtreme IO, Pure, v9000 all SSD enterprise arrays. They do inline dedup and compression some can offer microsecond latencies. Traditional HDD still has its use for video and filservers and archive. SSD for database and other high IO low latency applications.
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.