If chkdsk has replaced bad clusters, there is almost certainly nothing wrong with the hard drive itself. A bad cluster is simply the DFS ignoring it for use because of (probably) data corruption within those clusters. Remember, clusters are not sectors.

Presumably, chkdsk saved the data from the bad clusters in e.g. file001, file002 etc. If so, the disk can clearly be read. You can open those files and see if you recognise where the data comes from, which may help you understand what happened. Chkdsk will have also made the data space that those clusters occupied re-available to the DFS for normal use i.e. your available space does not continuously reduce through getting bad clusters as long as chkdsk is finding them.

Most common causes are power loss, or otherwise terminating a program in the middle of a file operation, which results in the DFS not 'understanding' the contents of the clusters and losing track of where they fit into the great scheme of things. Because of that, those clusters are not allowed to be overwritten by normal file operations.

Of course always back up your data but if the drive is only one year old, unless it was second hand when you bought it, or it is some obscure Chinese crap, you cannot have exceeded its working life no matter how much use you have given it.

Catastrophic failure is entirely different, is not foreseeable, and is really, really annoying... 

well, the actual problem was bad blocks, which, as I understand it, is a physical problem.
chkdsk was reallocating bad clusters which I'm assuming were on the bad blocks (which is the same as bad sectors, which is not the same as bad clusters... at least in my limited understanding).

anyway, being paranoid about loosing data these days, i've already gotten a replacement. i just don't have enough experience with this kind of thing to make a better guess on the matter. 

If a hard disk gives me any read errors i just turf it. Not worth any messing around imho. 

wow, cloning the drive took forever; must have been almost half a day.

everything seems to work correctly now and i can get back to quake stuffs. ^_^ 

The 2500K is a monster of a processor (good taste in CPUs BTW), but it is predominantly designed FOR overclocking

You are confused. Intel doesn't design CPUs for 0,001% of their usebase. 

I've always wondered what you have to do to do that.

I've got a 10y/o drive at work with all our admin on it, and i wanna replace the drive by cloning it onto a new one. 

You are wrong. About the 2500K and 2600K and the new 3###K's not being for overclocking.

I mean have you read some info before crapping on my 'claim', which is public knowledge?

You can buy a core i5 2500 (no K here), and a 2500K, the only difference between the two is that the K one is designed to be overclocked and about $5. You can't change the multiplier on a non-K CPU, but you can on a 'K' one.

I'm not even going to find some links to back my 'claim' up, but good troll BTW. Look at my rant ^ (!) 

http://www.techspot.com/community/topics/what-is-the-difference-between-i5-2500-and-i5-2500k.164512/

http://www.mmo-champion.com/threads/1088969-i5-2500-VS-i5-2500K

http://www.overclock.net/t/912968/i5-2500-vs-i5-2500k

And this is unrelated but HA! what a noob:

http://forums.overclockers.co.uk/showthread.php?p=21930437 

The K options have an unlocked multiplier; the non-K have it locked. But you can change the BCLK on a non-K, so actually, both are "overclockable". So which was designed for what or are they both the same design, with the K versions just having some extra like a heated front windscreen or low profile tyres? Other than that?

Still, think I'll go Ivy Bridge and drop the graphics card (tee, hee, hee). 

That is true, but it's been proven that because of the way that Intel have designed their architecture, and due to the fact that the FSB is synched with the PCI bus directly, you cannot achieve a stable FSB overclock of more than about 5%. But if you use a 'K' series CPU and a motherboard chipset that supports overclocking in this way, overclocking via the multiplier will achieve results of stable 50% overclocks. Which is a factor of 10 times more than using the FSB.

As per usual, Intel are just 'disabling' the functionality for overclocking in their non-K CPUs rather than adding the functionality into the K series ones.

But the fact remains that the ONLY benefit (ahem) you will get from a K series CPU is the ability to overclock by up to 50% using the multiplier. So the affinity between the two means that Intel have released a range of CPUs which are designed specifically for the overclocker. QED.

Also, the onboard GPU is actually not that bad.
I have an i3 2330 in my laptop with HD3000 graphics, and I can actually run Skyrim on it. With super-low settings. It runs UE3 pretty well, Fallout New Vegas fine, even on med-hi. 

So the affinity between the two means that Intel have released a range of CPUs which are designed specifically for the overclocker. QED.

Either I didn't understand a word you just wrote or you supported what Mike said: It's the same CPU sans the lock. The only difference is that the overclockable CPUs have passed more quality tests in the factory. 

...that's the point. Were they designed FOR overclockers or were they designed to allow overclocking? I would suggest that in use, far more will not be user-overclocked than will be. Of course, system-overclocking is another matter.

By the way; God Save the Queen - it's pissing down. 

Lots of people will buy K series CPUs and not overclock them. Which IMO is dumb. And the customers that do that are silly. Because they would be just as well buying a non-K CPU. FFS.

I mean who the F would buy a higher performing CPU just to under-use it? Get real. By the time a non-K CPU has burned out, it will be totally obsolete. People just don't think like that.

They don't. 

When buying a greenhouse, buy the biggest you can afford because once you start using it, you are going to fill it no matter how big it is.

But if you are only growing tomatoes for your summer salads, just put a couple of grow-bags on the patio. After all, you can only eat so many tomatoes.

Besides which, you can buy tomatoes at your local supermarket, and you don't have to water them everyday, or hoe the weeds, or squish the bugs, or watch out for bottom-end rot. Kind o'like sorta thing...

It's still pissing down and she's still smiling and waving. Go girl. 

i got a non-k 2500 specifically because i did not plan to overclock. :P

as for drive cloning, i used the one that WD gives for free, Acronis True Image WD Edition...

It seems like it is only usable if you have a WD hdd so I don't know if it will work otherwise.

there are other free alternatives though that i remember seeing on google while searching.

Something to keep in mind is that the cloning process takes the machine out of commission for the duration of the clone. it can't seem to do it in the normal windows environment and has to reboot to run it's own software.
This makes sense though, because you wouldn't want someone adding or removing files on the disk while you're trying to make a complete copy of it. 

ricky: clonezilla 

will work too. 

http://www.anandtech.com/bench/CPU/2
I was looking at that site. Was wondering what cpu benchmark would best describe what to expect from vis.exe performance? 

is an i7-3770k with 16 GB ram and a gtx 680 enough for the next 2 yrs or so? 

I do all of my map compiling on a Pentium 4 2G. I also use it for music, photography archiving, surfing, and spreadsheets. I did not believe that would be the case when I bought it all those years ago.

I also have an i7 laptop running Windows7 @ 64bit, which I use for 'proper' work such as spreadsheets (2M+ cells) and photomanipulation.

No point in buying the i7 K version unless you are definitely going to overclock - last time I looked the plain version was cheaper.

Apart from that, yes it will easily last for two years provided that you do not look at all the releases in the meantime and convince yourself that you must have the latest otherwise the sky will fall in. 

If you get a good PSU and Mobo you can do a SLI after that lol.

Mike is right about the overclocking - unless you are going to overclock there is no need to get a 'K' series CPU. Haswell is half a year away too. Not sure if that will use the same RAM and socket as the Sandy Bridge and Ivy Bridge 1155 stuff.

680 is a beast, but if it's value for money you are looking for then look at the 660Ti or 670.
The performance goes down, but the VFM goes up. 

yeah looking to get a system that will be overclocked to 4.3 Ghz.

Mike, dont need the latest etc was just trying to make sure I can play some of the newer games at decent settings over the next 2 yrs. Doesnt have to be at max settings or be every game, but would be nice not to worry about it if I want to get some titles.

Ricky, yeah might go with 670, it is about $150 cheaper in otherwise the same system. 

16gm ram? Shit a brick. I think you'll be able to continue playing games for a lot longer than 2 years with that spec. I usually upgrade once every 4-5yrs. 

I have read a few articles on old systems with new video cards. They play games well. Seems the video card slot is what will obsolete a system build. If a new slot spec comes out. 2-3 years out a video card update is all that you should need. Get a 4 memory slot board so if later you want to do 32 gigs memory you can.

BTW anyone use a SSD drive?

Another thing get THE BEST power supply you can. Every computer failure I've had was a failed power supply. Either random shutdown or out right shorts and blown capacitors. 

Excellent things! I have two at work. Both machines boot to the desktop in <30secs. Having the OS and core applications installed on the SSD makes for a much smoother experience. Everything loads almost instantly.

For my home rig I am waiting for the 480-512GB ones to drop a little more, so I can happilyu instal all my steam games on it, and not worry too much about the downloads file getting full.

And yes - a good PSU is worth the money. Corsair, Antec or OCZ, Cooler Master or XFX.