Thinkpad X300 With SSD Performance Evaluation

Ninjakicks writes "Hard drives are typically one of the more significant performance bottlenecks in any system today. An evaluation of Lenovo's new ultra portable Thinkpad X300 notebook shows a fast solid state hard drive can substantially improve the performance of a system. This is especially true of a low-end, low power processor and integrated graphics, in addition to reducing overall power consumption. Despite its 1.2GHz CPU the Thinkpad X300 is actually able to outperform some desktop replacement notebooks equipped with dual 7200RPM hard drives in RAID 0 in productivity benchmarks, and in data transfers. Interesting results, especially considering the X300's ultra portable form factor."

Ummm...

[What Would NPH Do (1274934)]

Despite its 1.2GHz CPU the Thinkpad X300 is actually able to outperform some desktop replacement notebooks equipped with dual 7200RPM hard drives in RAID 0 in productivity benchmarks, and in data transfers.

Sure it's data transfer performance was impressive but in all but 1 of the performance benchmarks it was last place.

Re:Ummm...

[What Would NPH Do (1274934)]

Though I guess I should add that even when it was in last place, the number it's pushing are rather impressive.

Re:Ummm...

[Concerned Onlooker (473481)]

"the number it's pushing are rather impressive"

Not nearly as impressive as being modded +5 Interesting and then being modded +4 Interesting for a reply to your own post that basically negates it.

Re:Ummm...

[Koiu Lpoi (632570)]

Battery life. It absolutely smokes the other three systems, and while it is in last place, it's almost tied for 3rd. It's an impressive machine. In my opinion, though, not worth the $3258.00 price tag.

Re:

[by Anonymous Coward]

If they really wanted to show the performance improvement of SSD vs HDD, the least they could have done was run the tests using the X300 with its SSD drive replaced with a 5400 and 7200 RPM HDD even though neither is an available option.

-1 Troll

[fred fleenblat (463628)]

microsoft introduced readyboost just in time!

Interesting idea for older notebooks

[mcsqueak (1043736)]

The article summary gave me an interesting idea. I have an old 1.5 GHz Pentium M notebook I was going to clean up and give to my folks. I'm wondering if replacing the existing HDD with a SSD would improve performance for it. It's a little old and clunky now, obviously.

Re:Interesting idea for older notebooks

[jackharrer (972403)]

I did it with my Media Centre PC. Old Compaq Presario 900. I bought dirt cheap CompactFlash to IDE 44 converter and put it instead of HDD. Mythbuntu start in half of the time, even that throughput of CF is almost the same as HDD. Best of all that SSD cost me £15 for 4GB. Straight from eBay.

Do it, it works brilliant.

Re:Interesting idea for older notebooks

[Jaime2 (824950)]

I did the same with my MythTV front end. I bought an adapter and a cheap 8GB CF from newegg. Mine boots reasonably fast (about the same as my other front end with an old 120GB drive), but it took about 20 hours to install Fedora Core 8 and run a software update. Every once in a while it freezes on live TV playback, and I think it is some sort of delay writing to the flash drive. My other front end has no such problems. The one with the CF has better specs than the one with the hard drive - much faster processor, more memory, better video card.

Re:Interesting idea for older notebooks

[jackharrer (972403)]

Interesting. I had no such problems. My Mythbuntu install went nice and smooth. I haven't tested it really well as I'm doing some contract away from home. But my wife hasn't mentioned any problems.

Oh, my one is Frontend and Backend on one machine + Samba shares on server mapped to folders through fstab. Also added noatime to fstab and got rid of swap whatsoever, just to save space on CF.

Exceptional Battery Life

[zedlander (1271502)]

Check out the comparison on the next page [hothardware.com]. The Thinkpad got almost 3 times the battery life of the Dell, coming in at close to 4 hours.

Re:

[Hemogoblin (982564)]

That's a silly comparison:

(1) The Thinkpad is a ultraportable notepad with a 13" display
(2) The Dell XPS 1730 has a 17" display, dual videocards, dual harddrives, and 2.5x the cpu clock speed.

No-wonder the XPS gaming laptop had a shorter battery life.

Re:

[Shadow-isoHunt (1014539)]

I just wanna point out that if you're looking for battery life, modern laptops are not where it's at. I've got an old dell latitude c610, 1.2ghz pentium 3, 1gb of pc133, ATI Radeon mobility m6. Using both bays with 66whr batteries, I get about 13 hours of battery life. I've never actually managed to run it down with the LCD closed.

Yes, it'll run linux, it actually dual boots.

Re:Exceptional Battery Life

[What Would NPH Do (1274934)]

Why do high-end laptops necessarily get less battery life?

Because the higher end hardware consumes more power. The newer XPS laptops have things like dual graphics cards in them via SLI. Do you honestly think that's going to use less power than something using a lower end integrated graphics card?

Why can't things be "turned down"?

Why would you turn things down when you're buying the laptop purely for performance?

Speedstep technology existed for a reason.

Yeah, but when you're caring about performance you wouldn't be using it.

Re:Exceptional Battery Life

[qbwiz (87077)]

This is true to an extent, but there are trade-offs that can't be made when using the computer. You may be able to turn off sections of a chip, but not nearly with the level of detail that you can by not adding transistors at the beginning. Trying to convert a fast, out-of-order CPU with many pipeline stages into a slower, in-order CPU with fewer pipeline stages at runtime would be effectively impossible. Additionally, the way that the chips are manufactured affects their speed: for example, fast transistors generally have higher leakage currents, so you have to compromise between high clock speeds and low static power.

Re:

[Gat0r30y (957941)]

As soon as everyone who buys a computer is willing to put and extra 1000 dollars to get an SSD instead of an HDD.
That or the price of flash starts dropping (right now it has been dropping linearly with density, vs. HDD's which have tended to drop price/GB exponentially).

Re:

[by Anonymous Coward]

I can't believe hard drive manufacturers aren't aware that the devices they built their businesses on are headed for the museum right next to buggy whips and engine cranks. So when are we going to see that big move to solid state storage? Less weight, less heat, less power, no moving parts...what's not to like?

Less space. Higher cost. Shorter life (though that one may have been solved and I just don't know about it.)

Re:SSD Write times suck, wear issue still there

[by Anonymous Coward]

Most of the newer models have wear-leveling and that reduces it greatly. But it's still an issue.

Do you have anything at all to back this up?

Then you have the huge issue with write times. Many reviews show real-world speeds of 3-4 times SLOWER then a typical 2.5" 5400 RPM HDD.

You may think that isn't much, but it can be. Things like moving files around, compiling software (Gentoo :), or just using swap space, will show huge hits in performance.

Well first of all moving a file (on the same device) is irrelevant, I assume you mean copying it. Yes SSDs have slower write speeds and that is an issue, 3-4 times slower is an exaggeration though (and the rotational speed of the drive has very little relevance to its write speed unless the drive is nearly full and heavily fragmented - which of course it isn't in any common benchmarks). Swap space is the only thing this may become an issue, but then again you're springing the extra $1k for an SSD in your laptop you've probably also paid the extra $50 for 2GB of memory, making swapping a rare event.

Re:SSD Write times suck, wear issue still there

[TooMuchToDo (882796)]

But for write few, read many data warehousing tasks, SSDs are an enourmous benefit. Think about Google, where the filesystem is optimized for reading due to large files being created and read from all the time for search results (yet the files aren't constantly rewritten). Or think about Netflix needing a huge video library to serve movies over the web. The movie content isn't changing, so it would make sense to have huge libraries of SSDs that save power by not spinning, get written to once with a block of movies at a time, and get read from all the time from customer devices.

SSDs have their place now. And they're only going to get more popular as the price comes down.

Re:SSD Write times suck, wear issue still there

[penguinstorm (575341)]

Just like thumbnail drives, they will "wear out" with use.

My understanding is that the technology being used in "SSD-Hard Drives" is quite a bit different than the tech being used in the average cheap thumb drive such as my 4GB one.

By the same token, the tech being used in the iPod Touch is quite a bit different, which is how it can offer 32GB of flash storage for ~CDN$500 while a 64GB SSD upgrade for a MacBook Air is CDN$1,400.

So if you can back your statements up with some evidence, knock yourself out. Otherwise...I think the issue isn't nearly as real as you seem to suggest it is.

Re:

[bjourne (1034822)]

Not true, write speed isn't all that important. The reason why hard disk drives are such huge bottle necks is because reading data is a synchronous operation. When you read a file, you do so because you need to do something with its data. Right now, not some time in the future. So your program has to wait (block) until the hard disk has finished reading all data. Depending on how far the disk head has to seek, the wait may take a huge amount of time.

To put it in perspective: when the CPU accesses a registe

Re:

[swillden (191260)]

Just like thumbnail drives, they will "wear out" with use. Most of the newer models have wear-leveling and that reduces it greatly. But it's still an issue.

No, it isn't. Partly because of increases in the number of write cycles they can support, but mostly because of size increases and wear leveling.

Consider a 64GB device with a write cycle limit of 100,000. Assuming constant rewriting of all of the data, you'd have to write 6.4 petabytes of data to wear it out. Assuming you could deliver sustained writes at 22 MB/s (150x), it would take 6,400,000,000 / 22 = 290,909,090 seconds, which is over nine years of continuous, max data rate writing.

In practice

Re:Lenovo Hardware is Unreliable Junk

[StarHeart (27290)]

My office uses Thinkpads exclusively. I would say maybe the failure has gone up some since Lenovo took over, but depot times have always been fast. I just turned in a hard drive RMA today, and I will probably have it with advanced replacement by Monday.

They also seem to be having sales all the time these days. Which means prices have come down.

Re:

[treeves (963993)]

I have a Thinkpad T43 that had to have its main board replaced. I sent it out, they got it the next day, replaced the board and sent it back the same day, so I was without it for only about 48hours. I kinda wished it hadn't had that three year warranty, though. Then I could've gotten a new one instead of just fixing it.

Re:Lenovo Hardware is Unreliable Junk

[amirulbahr (1216502)]

This is FUD. I can see why you posted as AC.

AFAIK Lenovo bought IBM PC Division in its entirety. In other words the ThinkPads are still being made by the same entity.

In our experience, maybe things have changed in terms of design choices on the newer models, but the service level and DOA rate has not changed all that much at all. In some territories support is still being outsourced by Lenovo to IBM.