Investigating Intel's Turbo Memory: Does it really work?
by Anand Lal Shimpi on June 19, 2007 3:39 PM EST- Posted in
- Laptops
Problem 3: Keeping that drive spun down
The final issue we had to address before we could even begin to look at how Turbo Memory impacted battery life was how to test the platform. Turbo Memory requires Windows Vista, as it piggybacks on two Vista technologies (ReadyDrive and ReadyBoost) but unfortunately most of our battery life benchmarks don't work under the OS.
We instead turned to some of our usual performance tests and with a few modifications, turned them into battery life measurements. By looping SYSMark 2007, WorldBench 6 and various other tests we were able to create meaningful test scenarios that ran the battery down in a repeatable manner, where Turbo Memory had a chance to work.
When testing Turbo Memory, Intel recommends reducing the HDD spin down timer to its minimum allowable setting, which is 1 minute under Vista. We opted for a more reasonable setting of 3 minutes, but also looked at results with the timer set to 1 minute to see if it impacted battery life at all. We found that the difference in battery life with Turbo Memory enabled between the timer set to 3 minutes vs. 1 minute was negligible at best, a couple of minutes was the largest difference we saw in our tests.
Facebook
Twitter
RSS
31 Comments
View All Comments
casket - Wednesday, June 20, 2007 - link
"If you add more memory to laptop, you use more power, emit more heat, etc"-- Using this logic... adding ReadyBoost (which is memory) would also use more power, emit more heat, etc...
The key here is that either readyboost or memory uses less power than a spinning hard drive. I would suspect you get the same power savings with more memory as well.
yyrkoon - Wednesday, June 20, 2007 - link
I dont know about anyone else, but I am starting to resent Intel using town names of the area I grew up in as a kid. You would think they could be a little more original.PrinceGaz - Wednesday, June 20, 2007 - link
I'd say that if it reboots every 6 minutes or so to re-run the test, it isn't worth a look and is totally useless as a notebook baterry-life benchmark as it in now ay reflects real-world usage, and all results using it should be discarded. Surely a better benchmark could be found than that. Unfortunately, removing the WorldBench results make Turbo Memory seem next to useless, which is understandable as it is likely to have been mainly the reduced HD activity when rebooting that the Turbo Memory was helping with.
JarredWalton - Wednesday, June 20, 2007 - link
And yet, the mocked up WorldBench 6 test shows a rather impressive 12% increase in battery life. It seems that the startup/shutdown process at the very least gets a decent benefit (in terms of battery life) from Turbo Memory. That indicates that the power savings from putting the hard drive to sleep are definitely tangible.redly1 - Wednesday, June 20, 2007 - link
I would like to buy this and test it out with my tablet PC. Anyone know where I can buy one of these mini-PCIe cards?skaaman - Monday, June 25, 2007 - link
Here is the part# NVCPEMWR001G110You can pick them up for under $35
BD2003 - Wednesday, June 20, 2007 - link
The problem I have with the article is that in general, they are still running benchmarks that do not reflect how an actual user interacts with their laptop, and do not really reflect the benefits that turbo memory/readydrive/readyboost would have.PCMark is supposed to give a number as to how fast your computer can run a barrage of application tests - but looping it over and over does not even come close to reflecting an actual usage pattern.
Now granted, they need a *repeatable* test to have numbers that are comparable, but that does not necessarily speak of the validity of the numbers.
For the average office laptop, you'll be running outlook, word, excel etc - the amount of data actually being loaded and saved is VERY small vs. large amounts from a benchmark, and in that very common scenario, the drive would rarely have to spin up, and the battery savings would probably be much closer to the ideal of 30 mins than what their benches showed.
I do agree with them on their final conclusion - 1gb is just not enough for more than basic office tasks. In order for this to really take off, to be able to cache an entire movie, they're going to need cache on the order of 4gb. Then I think it'll really make a difference battery/performance wise.
And they really, really need to fix the driver issues.
BikeDude - Wednesday, June 20, 2007 - link
If the movie is 4GB, a 1GB cache means (ideally) you will spin up the hard drive four times to load the next GB. I doubt you'll see much benefit from a 4GB cache in such a scenario.That said, the test didn't do any read ahead tests. All the descriptions so far seem to say the technology caches stuff already read. I.e. if streaming a movie from the hard disk there's nothing that will suck it all into a cache... (grrrr, this reminds me that my Hauppague TV tuner streams everything to a 7MB file which it then plays back -- works fine as long as I don't hit the same drive with heavy IO)
sorr - Wednesday, June 20, 2007 - link
i'd just use another Gigabyte of memory i.e, 2 GB in total and hybrid drive for now, then after 2~3 years just use the SSD when it comes down in price and goes up in capacitySilthDraeth - Wednesday, June 20, 2007 - link
Page 4 mentions Windows XP. I thought I read the article, but maybe I am missing something. I thought it was purely for Vista, but XP is mentioned several times.Please explain, because I am confused. Thanks.