How much will a SSD help my boot times?

[taltamir]

Here is where I'm coming from, I'll use something that the average end user and consumer can understand and relate to: boot time.

Screw boot time, irrelevant and most insignificant effect of an SSD on your speed.

Someone has a old laptop that takes 30 mins to fully boot

No, it doesn't.

that is, 30 mins to reasonably expect to be able to do something as simple as right click My Computer and get a context menu in a reasonable amount of time, say, 1 second, rather than nothing happening due to disk IO. Or more simply put, the time for the HDD light to finally go out and stop being lit up brighter than the sun.

No it doesn't

They get a new computer with a SSD (not even a performance oriented one, just a bottom rung 1.8" SSD). Machine takes 30 seconds to fully boot. User about to have a heart attack it's such a revelation.

I come home to my oldest PC that is 7 years old with mechanical hard drives and boot up from POST to shell in <= 15 seconds and HDD light goes DEAD the instant the desktop appears. It's DONE. Granted I'm not running anti virus and 100 running processes on startup and so on, but point remains, my example user has never experienced less than 60 seconds boot until SSD and attributes performance to the SSD, yet I had it 7 years ago long before SSDs. Both a combination of environment optimization and hardware yes, but primarily possible because they were the highest performing drives on the planet at the time. At the same time I was using this, the example user had... I don't even want to know, what drives were the cheap ass OEMs using at the time, 40 MB/s Maxtor drives with seek time slower than my current optical drives? (Quantum Bigfoot epitomizes OEMs cheaping out on the most important component in a PC...)

And now that person with with SSD tries to install a superslow to install game. It takes 10 minutes. You on the other hand would take 50 minutes to install the same game (real times differences between my caviar blue WS6400AAKS dual platter and my intel SSD). Because environmental optimization (no AV, no startup programs, etc) doesn't help at all with that task.
Now he plays a video game prone to texture popping, his textures load in a fraction of a second while yours take 4-5 seconds. Now he plays a game that autosaves and "stutters" whenever it does so on a spindle disk but doesn't stutter on his SSD. Now he installs a windows update that takes a fraction of a second to install but many seconds on yours. Now he opens word or excel and it opens much much faster.

Get it yet?

at least until the cells fry themselves

After a year and a half of use I calculated my intel's write amplification, total written, and life remaining... 32 years left at this rate.

See REAL WORLD example, not whitepaper numbers and benchmarks.

No, BS example. You pulled the numbers out of your behind, they are meaningless, and they deal with a meaningless issue and ignore the real benefits

 

[exdeath]

No, it doesn't.


No it doesn't

Oh yes it does, and I'm not even fucking joking. You have not used these PCs I'm refering to. I'd take a video for you and show you but cameras are not allowed. PGP, anti virus, SCCM + wuauclt checking manifest and running their cycles, network policy scripts doing a dozen things of their own, etc. On the desktop workstations with Raptors, takes less than a minute. On the notebooks with SSDs, takes less than a minute. On certain models of laptops with cheap drives? 20-30 min bootups...

Installing Lotus Notes on a Raptor workstation or SSD? Takes about 5 minutes. On one of these 30 min booting machines? I'm browsing the web on my phone while staring at a "30 seconds remaining" dialog that hasn't changed in... 15 minutes? Even when fully defragged, there are certain systems that are just pigs compared to either the SSD machines OR Raptor machines (namely anything that doesn't have a SSD or Raptor/Cheetah class HDD).

I get the same story from everybody, how they boot their computer in the morning and go use the bathroom, get a cup of coffee, etc, come back, and it's still not finished, etc.

And as for textures popping, it's called RAM, helps with that with HDDs so that's not specifically an SSD thing either. It only needs to load once if you have the RAM for it.

You should have used an example like compiling a large project or zipping a million files or running a full virus scan or something that is completely real time disk IO limited and at it's heart involves processing files on the disk and not just loading things into RAM. That is where SSDs *really* pwn serious ass.

 

[taltamir]

Oh yes it does, and I'm not even fucking joking. You have not used these PCs I'm refering to. I'd take a video for you and show you but cameras are not allowed.

I get the same story from everybody, how they boot their computer in the morning and go use the bathroom, get a cup of coffee, etc, come back, and it's still not finished, etc.

I have seen malware infected PCs that were practically "frozen" all the time (the main OS wasn't responding due to too much work elsewhere)... 100&#37; network and CPU usage. And once reinstalled I was told how amazingly faster they are.
But normal programs will not take 30 whole minutes. A cup of coffee doesn't take 30 minutes.
Besides I put two "no it doesn't" there. One for the time claim, and one for the description on how I/O affects performance.

And besides you are sidestepping the issue. You ignored the meat of my post. With boot you can eliminate all startup programs (which might not be an option), not use an AV, and otherwise strip your computer for a fast boot. This does not prove that SSDs are not awesome. This just proves MY OWN POINT that boot time is a stupid thing to look at when measuring performance. And besides, people might actually WANT an AV and specific startup programs. One of the cool things with my SSD is that I went from stripped down startup to having all those neat programs that do things that I want. But no, you are making me follow your tangential discussion.

What is important about SSDs is actual use.
Actual usage, installing and running games and programs, manipulating files, etc... that is amazingly faster. That is why people rant and rave. Ok, some might specifically rant and rave about huge boot time improvements but doesn't make it less legitimate an upgrade. You were dissing SSDs, dissing those who rant and rave, looking down your nose at the "average Compaq user", saying that raptors and cheetas are "as good", saying that anyone who rants and raves about an SSD has never used a raptor, and now saying anyone who rants and raves about SSD has never used a clean OS boot...
Please correct your statements. A raptor and a cheeta cannot compare to an SSD, end of discussion. And people who rant and rave on SSDs tend to be very knowledge and personally upgraded from a raptor or a cheeta (Who do you think can afford SSD and care enough to buy one?) and actually use an SSD for the same reasons! (I remember choosing raptor over faster sequential drives due to its impressive random performance).

 

[exdeath]

K, another common task for you:

You're a user copying a folder with a 20 GB Notes replica in it or moving half a dozen bluray video ISO images:

200+ MB/sec Raptor RAID

200+ MB/sec single SSD

some random 40 MB/sec generic drive

You're telling me the progress bar on the SSD will be just as much faster over the Raptors as the crappy drive.

I'm telling you the improved speed of the SSD over the crappy drive isn't amazing if you had the Raptors 10 years prior.

Yeah it's a specific scenario (linear transfer) that doesn't cater to SSDs strengths to build your case, but for me it's a real world thing that I encounter daily.

I never said they were as good or better, I said that if you actually had real hard drives before SSDs came out, it's not QUITE as amazing as if you replaced your crappy HDD with a SSD. Going from $80,000/yr to $100,000/yr is not QUITE as amazing as going from $20,000/yr to $100,000/yr.

The other example I gave in my very very first post on this topic, LCD monitors. If you had a F500R FD Trinitron, you basically had a LCD (better actually) quality razor image a decade before LCD desktop monitors became mainstream. LCD is only amazing and breathtaking if in 1999 you had a $99 15" CRT @ 60 hrz and then got a LCD. In other words people who were wow'ed about LCD and are glad to throw away their CRT just finally found out what they were missing all those years not having a REAL CRT.

Installing a game? Yeah my SSDs can beat out my Raptors by a minute or two, but BOTH are going to beat an average HDD based system by what, 20-30 mins?

The word I'd like you to study is "relative".

 

[taltamir]

You're a user copying a folder with a 20 GB Notes replica in it or moving half a dozen bluray video ISO images:

200+ MB/sec Raptor RAID

200+ MB/sec single SSD

some random 40 MB/sec generic drive

Random generic drive is 80 to 120MB/s
and why compare raptor raid to single SSD and not SSD Raid?

You're telling me the progress bar on the SSD will be just as much faster over the Raptors as the crappy drive.

I'm telling you the improved speed of the SSD over the crappy drive isn't amazing if you had the Raptors 10 years prior.

No this is not what I am telling you. This example is a case of a purely sequential transfer. Sequential transfer is 2x than previous drives. On your daily use you will come across many cases where random transfers count, I listed about half a dozen already.

I said that if you actually had real hard drives before SSDs came out

You continue to focus on the sequential speed (where SSD is 2x) and ignore the random speed (averages over 100x improvement. G2 vs Raptor is 86x improvement) to back your ridiculous claims that what we had before was "as good" or even comparable. It wasn't, there is absolutely no comparison, we saw a 2 orders of magnitude improvement

 

[exdeath]

Random generic drive is 80 to 120MB/s

Bullshit, a brand spanking new current production HP 8540P with a 320 GB 7200 RPM "generic average drive" will not come CLOSE to 80-120 MB/sec I promise. They are slow as dog shit rolling uphill on a humid summer day.

Not sure what the change was but lets just say whatever they did in their choice of HDD from the outgoing 8530P to the new 8540P, the time for full PGP to complete went from like an hour to over 8 hours... with the same environment image so it's not a software problem, it's purely the drive.

and why compare raptor raid to single SSD and not SSD Raid?

Because I initially said (in other words) if you had an extreme high performance HDD subsystem already (eg: RAID0 configs with the fastest mechanical drives in the world), the change going to an SSD isn't as severe as when replacing a crappy HDD.

The average user going from a average crappy drive to an average SSD next time be buys a average budget/midrange new computer at Best Buy would be absolutely amazed with how much faster the SSD is over the old HDD, but would have been unaware of either drives such as the Cheetah and Raptor or this nifty thing called RAID that could have allowed him to have a similarly very quick and responsive system compared to his old one, all along, long before SSDs made having a very quick and responsive system somewhat mainstream.

I know you understand the point I've been making here, summerized by the bolded sentance above, you just want to have a pissing contest, I understand.

I'll give you that my original post appeared as though I was comparing the drives directly and arguing they were the same in literal numerical performance, but was poor writing on my part (or just brevity?), and my best attempt at making the point I've been making all night and which hasn't changed one bit is in bold above. It doesn't say "Raptor is faster or equal to SSD" as you claim I've been saying, it says that either a Raptor/Cheetah/RAID setup or a SSD will both produce a very fast and responsive PC while the crap hard drive that the SSD replaced will pale by lightyears compared to either. Worded differently, as I've been posting repeatedly, the improvement from SSD compared to the Raptor/Cheetah/RAID isn't quite the same order of magnitude of night and day difference vs. the case where all you've known your life before the advent of SSD was shitty mainstream hard drives.

'Night.

 

[taltamir]

Bullshit, a brand spanking new current production HP 8540P with a 320 GB 7200 RPM "generic average drive" will not come CLOSE to 80-120 MB/sec I promise. They are slow as dog shit rolling uphill on a humid summer day.

that isn't "random generic" that is "slowest I could find".
For gods sake even the green drives from WD, Seagate, and samsung way outperform your claims.

You are still focusing on the periphiary and completely ignoring the actual points.
Your argument is still total strawman.

The average user going from a average crappy drive to an average SSD next time be buys a average budget/midrange new computer at Best Buy would be absolutely amazed with how much faster the SSD is over the old HDD, but would have been unaware of either drives such as the Cheetah and Raptor or this nifty thing called RAID that could have allowed him to have a similarly very quick and responsive system compared to his old one, all along, long before SSDs made having a very quick and responsive system somewhat mainstream.

This statement is still a total lie.
The most impressive RAID0 (which does not improve random speed at all) of Raptors or Cheetas does not hold a candle to SSDs in random reads/writes. It has the same sequential speed and is two orders of magnitude slower in random speed.
The whole reason people used raptors in the first place was that they offered better random writes. I switched from a WD Caviar blue to a LOWER SEQUENTIAL SPEED raptor and was impressed with the improved performance brought by its lower seek time.

 

[exdeath]

This statement is still a total lie.
The most impressive RAID0 (which does not improve random speed at all) of Raptors or Cheetas does not hold a candle to SSDs in random reads/writes. It has the same sequential speed and is two orders of magnitude slower in random speed.
The whole reason people used raptors in the first place was that they offered better random writes. I switched from a WD Caviar blue to a LOWER SEQUENTIAL SPEED raptor and was impressed with the improved performance brought by its lower seek time.

JFC I'm not saying the RAID0 HDDs will beat a SSD in random reads/writes, I'm saying that the overall experience for the user with the whole package, as in the responsiveness of the PC in daily use, when fitted with various storage devices, looks like this:


average HDD <<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<< Raptors/Cheetahs <<<<<<< SSDs

YES THE SSDS ARE FASTER OK? I never fucking argued otherwise. Be that as it may, either SSD or Raptor/Cheetah class HDD in a PC for the typical end user are going to be FAR more enjoyable experience with EITHER one, than a typical system with a standard average cheap POS drive like the kind that come in nearly all box brand OEM builds.

In other words if it doesn't have an SSD, then at least have Raptors/Cheetahs since they are the next best and closest thing, and if they don't have any of the above, I don't want to fucking touch it because I have better things to do than watch progress bars and solid HDD access lights all day.

In terms of the user experience, there is a thing called diminishing returns. DIMINISHING RETURNS, very important concept here, got it? Going from the fastest hard drive on the planet to a faster still SSD is not going to have anywhere near the impact on the user experience in a basic home/office use PC as going from a slow mainstream drive to either of the above. How is the above bolded statement "a lie" now? Nothing in there says HDD are faster than SSD, it says the fastest HDD or an SSD will both produce very quick PCs, while anything else will make a slow PC.

You mean to tell me that because we have SSD now that suddenly RAID0 Cheetahs wouldn't be considered a quick and responsive system to the average user who has only ever known that his shitty $300 PC takes 2 minutes of crunching sounds to launch anything?

Whether it's the SSD, or the brief split second sound burst of ripping burlap followed by immediate silence (because it's done that fast) from a pair of Raptors flying through shit faster than you can blink an eye, both are going to be absolute heaven for someone only used to shitty drives that sound like someone dragging a dog leash through wet gravel draggin ass for 20 seconds to complete the same task. They MIGHT notice the SSD is slightly faster stil while browsing the web and checking their email, but in real world practice both are going to be challenging the user to launch solitaire, word, IE, and Lotus Notes faster than they can pop up vs putting the user to sleep and doing figure 8s with the mouse always waiting on IO.

 

[exdeath]

The average user going from a average crappy drive to an average SSD next time be buys a average budget/midrange new computer at Best Buy would be absolutely amazed with how much faster the SSD is over the old HDD, but would have been unaware of either drives such as the Cheetah and Raptor or this nifty thing called RAID that could have allowed him to have a similarly very quick and responsive system compared to his old one, all along, long before SSDs made having a very quick and responsive system somewhat mainstream.

Total lie? Really? :hmm:

Hmm I click through a bunch of things really fast, so fast you can't even see what the dialog boxes were that flashed by because I'm really good with a certain piece of software and extremely fast and accurate with a keyboard/mouse like a FPS vet:

-SSD system? No problem, computer is faster than me and can keep up seamlessly. My mechanical input speed on the input devices is the limiting factor.

-Raptor/Cheetah system? No problem, computer is faster than me and can keep up. Every once in a while a minor split second hitch accompanied by 1/30th of a second burst of ripping burlap sound, but otherwise shit still pops up faster than I can mechanically input commands on the keyboard and mouse. If there is a 1/30th second delay loading the dialog box, it will still pop up right under my mouse as I'm moving the already in flight mouse cursor to the button I want, so it doesn't affect input speed at all.

-Normal average HDD system? Somewhere in the chain I right click on something and go to click properties... but suddenly I figuratively slam into a brick wall as I go to hit the button... it's not there yet, and I proceed to do circles with the mouse for 1-2 seconds where I anticipate the button will finally be in a moment, with my finger hovering over the mouse button the instant it comes up, while I hear something that sounds like somebody crumpling up paper into a ball and the little context menu loads 1-2 seconds after I commanded it... sometimes it takes a little longer, I'll take my hand off the mouse and start tapping on the desk getting impatient... ok there it is where was I now?

Which of these two user experiences are similar?

 

[DominionSeraph]

taltamir, you use Firefox, don't you.

 

[taltamir]

average HDD <<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<< <<<< Raptors/Cheetahs <<<<<<< SSDs

average HDD < Raptors < Cheetahs <<<<<<<<<<<<<< SSDs

And btw, your own argument was of a stripped down boot environment! No AV, no background programs, etc! That is not even related to HDD/SSD speed.

taltamir, you use Firefox, don't you.

I use IE8, Firefox, and Chrome. All at the same time, for different things.

 

[DominionSeraph]

I use IE8, Firefox, and Chrome. All at the same time, for different things.

Damn. Combining the two slowest browsers with the horrible memory management of using three browsers. No wonder you need a SSD.

Also, 15k SAS does not beat a Velociraptor in desktop usage patterns. Hell, a newer 7200RPM drive can beat an older 15k drive even when the STR is similar.

http://www.storagereview.com/php/be...&numDrives=1&devID_0=368&devID_1=345&devCnt=2

Desktop hard drives have firmware optimized for desktop access patterns, which isn't a random access pattern.

 

[taltamir]

Damn. Combining the two slowest browsers with the horrible memory management of using three browsers. No wonder you need a SSD.

I expected something such as that.
The majority of my browsing is done on Chrome due its unrivaled speed. But chrome has several severe limitations that google will not fix. For example it has no bookmarks sidebar. So when I browse my webcomics dir I use firefox with its sidebar. IE8 is used for interacting with my school's website on a daily basis, it is the only browser they support and as such I must resort to using it. Also whenever I go to government sites (that is quite a bit more rare) which only support IE8.

I have tried both firefox and chrome with my school's interactive student online site (the regular website is fine with chrome, its the one where i do tests, turn in homework, and get my lecture notes and grades that has the problem).

Kingdom of Loathing is designed for firefox and has issues with other browsers. But I don't really play that anymore.

As you see, I don't really have a choice here. And I actually use the zippity fast chrome whenever I can and for most things.

Regardless, I find the greatest contributer for browser speed to be the CPU. I feed those browsers TONS of RAM (look at my sig bub, I got 8GB) and as leaky as they are as long as they aren't eating it up it doesn't matter (And they aren't, I got excellent memory monitor always on as a desktop gadget).

the SSD is tremendously beneficial in a variety of manners, I even listed them. I don't know why you call into question my browser uses but that sort of passive aggressive behavior is not really conductive to a mature debate.

 

[Mr. Pedantic]

I think it is wrong to stick an SSD into a $350 computer. Seriously. It is a waste of flash.

 

[taltamir]

I think it is wrong to stick an SSD into a $350 computer. Seriously. It is a waste of flash.

I would agree, the OP's computer though was 350$ in 2006.

 

[Zap]

Which brings me to the real source of my griping: Why do OEMs ignore HDD performance in the mainstream segment and insist on using the slowest cheapest hard drives they can possibly find when it arguably is the #1 noticeable performance aspect to a non gaming office and web user?

That's because their target audience have been brainwashed into looking at megahertz and megabytes as indicators of performance, and all that putting a faster drive will do for them in the store is to make their computer more expensive than the one next to it.

 

[repoman0]

Have you people ever heard of standby? It works quite well these days, using less than a couple watts to have 2 second boot time and leave all your programs up and running. Makes for a much faster system overall, at least in Vista and 7 because they leave stuff in RAM so it's cached, works wonders with 8GB of RAM in my desktop.

 

[Mr. Pedantic]

I would agree, the OP's computer though was 350$ in 2006.

Oh God O_O

I revise my opinion. It is now not only a waste of flash, it is also a waste of space and a waste of electricity.

 

[Tom]

ssd is 100x faster than hard drive ?

Doing what exactly ? I don't mean running a benchmark, I mean real life usage. Since I never notice any slowdowns of apps or games with a hard drive, I'm having a hard time seeing how this 100x comes into play ?

Is it like loading a webpage file from cache in a millionth of a second instead of 10,000th of a second, or what ?

I understand the ssd advantage loading an app or booting, it's definetly not 100x faster at that, maybe 2x faster.

 

[taltamir]

ssd is 100x faster than hard drive ?

Doing what exactly ? I don't mean running a benchmark, I mean real life usage. Since I never notice any slowdowns of apps or games with a hard drive, I'm having a hard time seeing how this 100x comes into play ?

Is it like loading a webpage file from cache in a millionth of a second instead of 10,000th of a second, or what ?

I understand the ssd advantage loading an app or booting, it's definetly not 100x faster at that, maybe 2x faster.

2x and 100x are BOTH benchmarked speeds AND real life speeds.
This is because there are different types of speeds being discussed here. There is Sequential writes, sequential reads, random writes, and random reads. And to be honest there are also things in between but we don't bother measuring them.

This isn't just a "theoretical" "benchmark only" speed but real life improvement.

 

[Tom]

2x and 100x are BOTH benchmarked speeds AND real life speeds.
This is because there are different types of speeds being discussed here. There is Sequential writes, sequential reads, random writes, and random reads. And to be honest there are also things in between but we don't bother measuring them.

This isn't just a "theoretical" "benchmark only" speed but real life improvement.

Fine. How about answering my question ? 100x faster at what ? And what is a typical real world example ?

Like I said, if a hard drive takes 10000th/sec to do something and an ssd takes a millionth/sec, that's 100x faster, but it doesn't matter.

Something that takes 1 second with hard drive and ssd takes 100th/sec, will seem faster, but doesn't save much time.

What are the tangible real world benefits, apart from boot and app loading ?

 

[taltamir]

Like I said, if a hard drive takes 10000th/sec to do something and an ssd takes a millionth/sec, that's 100x faster, but it doesn't matter.

I did answer your questions, plural, you had MANY.
As I already said ~100x faster in random writes & random reads.

Your supposition that a spindle drive takes 10,000th of a second is ridiculously false. It typically takes 12ms, that is 1/83 of a second. the SSD takes ~100th. 12ms is noticeable; especially when you need to access MANY such files every second.
http://www.anandtech.com/show/2808/4
Compare random read speed of 4k and you get:
WD VelociRaptor: 0.68MB/s
Intel X25-M G2: 58.5MB/s

As for real life examples, any writing or reading of a 4k file / file segment. Every program with a log file (any instant messenger, every time you receive or send a message; peerblock; coretemp; windows installer; fiddler; etc), indexing, thumbnails, various browser functions, installing software. This is why you defrag a spindle HDD. The problem is really that you have so many separate programs generating I/O at once. You have logs being updated with highly inefficient tiny writes from a variety of programs at once. Your IM, various windows background processes, pagefile access. Those things all happen at once and every time one happens your HDD must finish one before it can get to others (although NCQ helps a LOT with that, I have been using it for years with my spindle drives, very useful technology that isn't getting enough credit).

 

[Tom]

I did answer your questions, plural, you had MANY.
As I already said ~100x faster in random writes & random reads.

Your supposition that a spindle drive takes 10,000th of a second is ridiculously false. It typically takes 12ms, that is 1/83 of a second. the SSD takes ~100th. 12ms is noticeable; especially when you need to access MANY such files every second.
http://www.anandtech.com/show/2808/4
Compare random read speed of 4k and you get:
WD VelociRaptor: 0.68MB/s
Intel X25-M G2: 58.5MB/s

As for real life examples, any writing or reading of a 4k file / file segment. Every program with a log file (any instant messenger, every time you receive or send a message; peerblock; coretemp; windows installer; fiddler; etc), indexing, thumbnails, various browser functions, installing software. This is why you defrag a spindle HDD. The problem is really that you have so many separate programs generating I/O at once. You have logs being updated with highly inefficient tiny writes from a variety of programs at once. Your IM, various windows background processes, pagefile access. Those things all happen at once and every time one happens your HDD must finish one before it can get to others (although NCQ helps a LOT with that, I have been using it for years with my spindle drives, very useful technology that isn't getting enough credit).

The example you give is about 10x faster, not 100x faster. Give me a real world example of 100x faster.

Show me something, not a benchmark, that takes 1 second with an SSD and takes 100 seconds with a hard drive.

Thanks for the description of small file access. However, my systems with hard drives do not exhibit any delays in normal use from all the I/O you describe, so I fail to see how an SSD is going to make any observable difference ?

I can see it would matter for disk intensive things like antivirus scanning or maybe searches if there's no index, but in the real world there's no need to do a full antivirus scan at a time that interrupts productivity.

 

[Zap]

taltamir, maybe you should give up. Let Tom enjoy his hard drives while we enjoy our SSDs.

Tom, 10x or 100x, the takeaway from this whole discussion is that SSDs are faster than HDDs. Sometimes not too noticeable, but sometimes extremely noticeable. Whether you want to pay more for more performance, as always, is up to you.

 

[COPOHawk]

I have to agree with Zap & Taltamir. I own a number of Intel and Crucial SSDs. I had seen the benchmarks, but since I already owned a number of Raptor HDDs...I was skeptical. Until I bought my 1st one back in Dec, 2009.

Tom...maybe you are one of the select few that only uses your computer only for internet browsing (under 3 tabs) with no AV software or other processes running in the background, with a fresh copy of Windows every 3 months. If so...then you won't notice a difference.

Otherwise, you will Every one of the 12 SSDs I have installed for customers in the last year has been a home run. About 8 were new computers and 4 were retrofit...I couldn't pry the SSD's from my customer's hands if I tried.

Cost is a factor...and if you don't want to spend the money, then I understand. However, the performance difference is significant...especially when heavily multi-tasking.