Hardware Samsung Spinpoint F3 1TB Review

Can one tell me how this 'short stroking' works and how to do it and when you should do it? Sounds interesting. Always looking for best performance.

I've got a 1TB Caviar Black, and two Seagate 250GB's I can use.

 

My gamble of getting three of these pre-review for my new Core i7 machine paid off!

 

I just ordered an F3 today, hopefully I get it before the weekend.

 

Thanks for the review guys - nice.

I'm with talladega, in that all this talk of 'short-stroking' is news to me - does it really make that much difference? (ie, 'real world' difference, not just shaving a few milliseconds off a benchmark result.)

If it does make a significant and tangible impact, then for around the £55+ these cost I could quite happily lose 700(ish)GB on one of these for a much faster boot/master drive.

That's assuming of course Samsung have/release the software.

As with Cupboard, I remain sceptical that just formatting 200GB would do it automatically, (how would you 'know' it's on the outer-rim?). I don't like making assumptions, especially if I were to risk ending up with a smaller AND slower drive...

 

...it's been possible to assign a small partition to the "outside" and use it as OS partition or even better "cachefile".
though it does take some tinkering (which I just can't be ar*sed to to anymore)
Any tools that to this automatically?

 

hmmz instead of sacrificing 700GB of data I rather buy a fast small SSD for OS boot. Store cachefile on F3...

 

But look at this logically: at 6p/GB, losing 700GB "loses" you £42, but you can't spend £42 on an SSD.

Even 300GB at £55 costs just 18.3p per GB, which is over 10x less than an SSD.

 

Exactly what I was thinking - as the cheapest halfway decent SSD is still a fair way over the £250+ mark for 128GB. (Or roughly 5x the price for 1/2 the capacity, which is 10x more expensive as you say).

I appreciate a short-stroked F3 still won't be approaching SSD performance in certain instances, but am intrigued to know what kind of improvement it would make.

Like overclocking a CPU for faster performance, but at the expensive of a probably shortened lifespan - I really quite like the idea of 'overclocking' my HDD for faster performance, but at the expense of losing some of its capacity...

 

(When I say 1/2 the capacity, I mean 1/2 the capacity of a short-stroked HDD - in case I wasn't being clear...)

 

This is interesting.

Would bit-tech be prepared to run a test feature on stroking please?

 

We do daily

 

lololol +1 for bindi

 

I installed my F3 last night and I can honestly say it's the dogs bollards. I copied 36gb of mp3's from my ageing 160gb Maxtor and it only took 15 minutes!

 

That copy would have probably been limited by your 160GB's read speed as well..

 

Well I've removed the Maxtor so all is well.

I have a question though, is it worth ditching my OS drives, 2 x 74GB Raptors (Striped) and replace them with another F3 or maybe 2!

 

Yep - RAID does bugger all for real world (game loads, boot times) and in those regards the F3 will dump all over an aging 74GB raptor.

 

Strangely, Speedfan doesn't show any temperature readings either....

 

Anyone running linux will find that hddtemp works, also.

 

how about this short stroking?! no bindi, not the kind you are into, but the kind we are asking about.

 

I have one of these (as of yesterday) and a Seagate 500GB single platter 7200.12. Both are subject to duff info in both Speedfan and HDTune. HD Tune says the Seagate is 11 degrees and it's been used for 8000 odd hours and the Samsung info is just hyphens. Might just be a case of the developers of these programs, updating them.

The write result I am getting is about the same as the Seagate 500GB (which is quicker than the 1TB Seagate 7200.12 for some reason) with the burst speed being more and the access being quicker and the read being slower but not by much. However both are 500GB per platter.