News - SanDisk to unleash 240GB SSD with 40,000vrpm

[twobells]

Complete and utter bollocks indeed.

I prefer to measure the speed of my SSDs in smoked mackerels. This is a much better way because they are rich in protein and high in Omega 3 oils which let the data slip through faster.

Before long, the entire industry will be measuring in SM/sec, so watch out for the benchmark.

[Embargoed 30/03/09] Sisandra are proud to introduce the SiSoftSandra 2009 SM Version by July 2009.
Working with Sandisk since day one on their smoked mackeral standard they hope to create the new benchmark in hard drive solutions.
According to our latest figures nine out of ten cats are shown to prefer it.

[TooNice]

I was going to shoot the idea down, but after reading the article, I don't think it is necessarily as bad as it seems.

Hell, for most people sequential read/write figures would be enough - only SMEs and up will really care that much about anything else.

While it is probably true that many enthusiasts put a lot of emphasis (some even a sole emphasis) on sequential read/write performance, I don't think that it should be the case. Not that I would complain about improved sequential read/write performance any more than I would complain about more higher clockspeed in a CPU (all else being equal), but there is far more to the average desktop use than just transferring a large files between two HD, video encoding, and the odd applications where high sequential read/write performance yields the most benefits.

And I find burst speed even less usefulness as a benchmark for determining a drive's performance for desktop use. Access time is good to know, but it is still difficult to make sense of how much improvement you'll gain for each reduction.

Performance of random read/write is at least as important as sequential read/write (I would argue more), at least on OS/application drives, yet those tend to be often overlooked.

So I don't think it is the wrong idea to have a benchmark that takes this into account. What I do have an issue though, is that Sandisk came out with this metrics (using their own weighting), as opposed to a more independent sources. This could undermine the credibility of the benchmark suit.

And I am also willing to go along and say that 'vRPM' is conceptually daft given that not all 7200RPM performs alike (though the differences *might* be relatively insignificant in contrast to SSD, for certain tasks). I would have preferred it if they presented it more like a 3DMarks score than a number followed by 'vRPM'. But I guess that vRPM is more marketable than just a plain number on the box.