backtop

Print E-mail del.icio.us 110 comment(s) - last by mindless1.. on Jul 26 at 7:17 PM
SanDisk CEO blames its SSD performance woes on Windows Vista

Solid state disk (SSD) news has been coming in fast during the past few weeks.  Most of the big revelations have been at the low-end of the SSD market with multi-level cell (MLC) based products, but single-level cell (SLC) based products have had their fair share of coverage as well.

Despite the hype surrounding the promising technology, SanDisk is placing blame on Windows Vista for not providing enough of a speed boost when using SSDs. SanDisk CEO Eli Harari went so far as to say that Vista is the reason why SanDisk is being left behind by competing solutions.

"We have very good internal controller technology, as you know...That said, I'd say that we are now behind because we did not fully understand, frankly, the limitations in the Vista environment," explained Harari. "As soon as you get into Vista applications in notebook and desktop, you start running into very demanding applications because Vista is not optimized for flash memory solid state disk."

"The next generation controllers need to basically compensate for Vista shortfalls," Harari added. "Unfortunately, (SSDs) performance in the Vista environment falls short of what the market really needs and that is why we need to develop the next generation, which we'll start sampling end of this year, early next year."

Those comments are quite a departure from SanDisk's own press release from mid-2007 regarding its SSD performance in Windows Vista:

SanDisk Corporation today announced that the Microsoft Windows Hardware Qualification Lab (WHQL) has certified the new SanDisk 1.8-inch UATA solid state drive (SSD), which is being used as a substitute for hard drives in selected laptop computers. The product also earned a high score on the Windows Experience Index, indicating that it is suited to a variety of Microsoft Windows Vista applications .

Additionally, the SanDisk UATA SSD scored 5.4 out of 5.9 when tested with the Windows Experience Index utility, a new feature in Windows Vista. The results indicate that the new Windows Vista operating system will run optimally when installed on the SanDisk SSD.

It is quite true that SanDisk's SSD are woefully subpar in performance when running Windows Vista. Numerous benchmarks from around the web have shown SanDisk SSDs getting outpaced by the competition.

In fact, it's not uncommon to see SanDisk SSDs rank last in testing in almost every benchmark and by a large margin -- even in Windows XP. Recent testing showed that MSI's Wind netbook was no faster with a SanDisk SATA 5000 SSD than with the standard 80GB HDD -- an Eee PC 1000h featuring similar specifications was significantly faster with a competing SSD from Samsung.

While Vista may be a performance inhibitor compared to Windows XP for SSDs, it appears that most new, current-generation SSDs are having no problems performing well with the operating system. The problem appears to be SanDisk's low reads and writes (67 MB/sec and 50 MB/sec respectively) compared to the competition (i.e., OCZ’s new Core Series SSDs which clock in at 120 to 143 MB/sec for reads and 80 to 93 MB/sec for writes).

Comments     Threshold


This article is over a month old, voting and posting comments is disabled
I also blame Vista...
By 67STANG on 7/22/2008 11:30:54 AM , Rating: 5
for the power going out at my office last week and for my dog running away.

Listen, I'm no fan of Vista by any means, but everyone blaming everything on Vista is getting a bit old.




RE: I also blame Vista...
By Radnor on 7/22/2008 11:39:05 AM , Rating: 2
Old as the Crysis Jokes.

Honestly, toms hardware already showed that a they don't really increase battery life. Only in certain environments. SSDs will take their sweet time to reach the Joe Consumer Environment.

I wouldn't touch one for several reasons. Price per GB is one, decay is another. Really, i do some files hashing and rendering on-the-fly. Still have several raid 0 going strong after 4 years. I doubt SSd would take that punishment. New technology yes, better technology ? we shall see.


RE: I also blame Vista...
By Lonyo on 7/22/2008 11:50:14 AM , Rating: 5
And other comments/breakdowns show THG's article to be nothing more than a complete load of crap.
No, I can't be bothered to link to the specific one I'm thinking of, but if you want to put your faith in THG, a site which showed an SSD with max power draw under load of lower than the mechanical HDD's idle power draw getting lower battery life, then please, go ahead and believe their numbers.


RE: I also blame Vista...
By DASQ on 7/22/2008 1:02:14 PM , Rating: 2
Funny how they've now deleted all the older power consumption graphs... and stubbornly refuse to change their stance from 'SSD power savings is a hoax!!!'


RE: I also blame Vista...
By livelouddiefast on 7/22/2008 1:38:09 PM , Rating: 5
didn't they post a partial retraction article?

new article- http://www.tomshardware.com/reviews/ssd-hard-drive... - summed up, some ssds are great with batteries. some suck.

old article- http://www.tomshardware.com/reviews/ssd-hdd-batter... - ssds are the battery devil


RE: I also blame Vista...
By jonmcc33 on 7/22/2008 2:55:46 PM , Rating: 5
Anyone that even visits THG is wasting their time.


RE: I also blame Vista...
By Clauzii on 7/22/2008 6:20:32 PM , Rating: 2
Not if You are in a bad mood. I can go there and leave ROFLed up :)


RE: I also blame Vista...
By B3an on 7/22/2008 11:39:24 AM , Rating: 5
Exactly. Other SSD makers can make well performing SSD's. SanDisk just have a poor product that cannot compete so are blaming Vista.


RE: I also blame Vista...
By Souka on 7/22/2008 2:15:35 PM , Rating: 3
just too funny... "SanDisk SSDs rank last in testing in almost every benchmark and by a large margin -- even in Windows XP"

So let's blame XP too....

*sigh*


RE: I also blame Vista...
By jonmcc33 on 7/22/2008 2:57:47 PM , Rating: 4
Just do like Jason Mick and blame Microsoft!


RE: I also blame Vista...
By Flunk on 7/22/2008 5:53:24 PM , Rating: 2
Then they have the problem explaining away their poor Linux performance as well. I think it's clear where the problem is.


RE: I also blame Vista...
By FITCamaro on 7/22/2008 12:21:02 PM , Rating: 5
Beat me to it.

"Its not our fault our products suck compared to the competition, its Vista's fault."

That said probably OSes do need to be optimized to deal with SSDs. But when other SSDs are performing better than yours, I don't see how you can blame the OS. You have to make your product work with the Microsoft's OS, not the other way around.


RE: I also blame Vista...
By Lerianis on 7/22/2008 2:57:04 PM , Rating: 1
What kind of optimization would they have to do, in all honesty? Unless you are doing extreme amounts of downloading off the internet (as I do), there should be nothing that changes on a hard drive that frequently (unless you are stupid and haven't disabled the swap/page file).

That is the biggest bit of advice I give to my friends who are looking at SSD's: DISABLE THE PAGE FILE AND UPGRADE YOUR MEMORY!
Because once you do that, the benefits of an SSD go up extremely.


RE: I also blame Vista...
By ChronoReverse on 7/22/2008 4:18:46 PM , Rating: 4
That's pretty worthless advice right there.

Disabling the Page File does not in fact stop paging to the hard drive nor is there any benefit to doing to if you had enough RAM in the first place.


RE: I also blame Vista...
By jketzetera on 7/22/2008 9:29:15 PM , Rating: 2
quote:
That's pretty worthless advice right there.


If you are going to categorically dismiss someones advice you better have a good factual grounds for it, which you do not have.

quote:
Disabling the Page File does not in fact stop paging to the hard drive


That is incorrect. If you disable the Page File, the OS will not page memory to disk. It will create a "virtual" page file in RAM though.

quote:
nor is there any benefit to doing to if you had enough RAM in the first place.


That is also incorrect. In a situation where your RAM is significantly larger than your usual memory usage, disabling the page file will enable you to instantly switch between minimized applications, something that you are unable to do with a page file even if you have lots of unused memory.


RE: I also blame Vista...
By mikefarinha on 7/23/2008 12:38:05 PM , Rating: 2
quote:
That is also incorrect. In a situation where your RAM is significantly larger than your usual memory usage, disabling the page file will enable you to instantly switch between minimized applications, something that you are unable to do with a page file even if you have lots of unused memory.


This used to be a big deal back in the pre WinXP days, however Microsoft made the swapping and virtual memory efficient enough that there is no performance loss when using a swap file, even if you have 2GB+ of RAM.

This is simply internet lore that, for some reason, refuses to die. I have fruitlessly looked for performance comparisons to prove me wrong. I'm tired of this tidbit of 'advice' that keeps popping up.

Please, prove me wrong!


RE: I also blame Vista...
By mindless1 on 7/23/2008 3:28:22 PM , Rating: 2
You are clearly wrong. No matter how efficient it is or ever will be, it is still something additional being done, taking a time period, and resources.

Perhaps what you meant was that in some kinds of usage, the different can't be perceived by the user, and in a limited artificial benchmark scenario which only tests one application AFTER the paging activity, you won't see the difference because it wasn't something the benchmark was designed, or able, to test.

I don't understand how you can't see this, that accessing a page file takes longer than not accessing it, no matter how fast it does. SSD and mechanical hard drives are much slower than main memory, this shouldn't need to be written. The main issue is only that you have to have enough physical memory to allocate if the page file is diabled, in which case there is no sense at all in keeping it enabled except that someone with dynamic use of a PC may not be able to fully estimate what that peak memory usage would be ahead of time.

Be tired, rest a bit, then get up the energy to intelligently think about what the system is doing and how you are trying to measure it. I am not claiming a typical usage pattern will have a large performance increase but technically, there is an increase w/o paging to disk and it can be significant depending on what is being paged out, especially other application code or data because it is a multi-tasking OS environment.


RE: I also blame Vista...
By ChronoReverse on 7/23/2008 6:30:31 PM , Rating: 2
The whole point is that with a modern operating system, data doesn't get paged for no reason.

The only time things are paged is if the RAM is needed or if the data is specifically marked to be paged out.

In the first case, if paging isn't performed, the program you're trying to run that needs the memory simply fails.

In the latter case, it's intended behavior. Thus, if the program is doing a lot of paging despite RAM being available, the program itself is culprit and not Windows.