If you just shelled out some pretty pennies for the a high-speed, low-power
SSD, Tom's Hardware may have stumbled onto some findings that won't sit well. According to a rigorous benchmarking session, they discovered that not only do the drives not save you battery power... they eat
more of it. How is this possible you ask? Well mechanical drives only hit peak drainage when the actuator has to move the heads, whereas SSDs use full power whenever they're in use, so the end result is actually a diminished efficiency. What that means is that the hype over "green" drives may be just that: pure hype. On the other hand, you're still getting a drive with no moving parts, which at least makes your data more secure, if not more eco-friendly.
[Via
Eeeuser; Thanks, Tony]
posted Jul 1st 2008 4:32PM
The graphic there doesn't seem to support that idea. Seems all the SSD drives in the graphic eat less power than that standard HDD. Maybe a different graphic should have been picked here?
Look again. The higher number does not mean better when you are looking at power consumption.
I know, all the ssd's seem to consume less.
and according to the graphic, when they are idle they take 1/2 the power, which contradicts the post that "whereas SSDs use full power whenever they're in use"
wth engadget... this article makes no sense.
I understand that: 2 of the drives are lower in idle 2 are higher - unfortunately for the standard drive: all drives are lower than it. The top 2 drives pretty much solidly kick that drive's butt.
While it may seem like that, if people read the article, they would know the HDD peaks at the state, while the SSD is constantly in that state.
Nevermind, Jason just pointed something out I didn't even see.
So now I'm wondering too.
The graphic is indeed misleading, the article explains it much better. The jist is that while the standard drive can do worse, with average use you don't gain much battery life saving, if at all. Also interesting is that the SanDisk does indeed consume much less power, but is lower performing in other benchmarks.
The graph appears to be wrong.
The article does indeed show that for at least the SSDs and HDDs they tested, the SSDs consumer more power on average.
However, I do NOT think that this pattern will hold into the future. Manufacturers have been finding ways to optimize and lower the power usage of mobile harddrives for over a decade, and power-saving features are now very mature.
On the other hand, SSDs are an incredibly young technology, and there are not even basic power saving measures integrated yet. I'm sure as the technology quickly matures, manufacturers will develop a range of power saving techniques and optimizations that will allow SSDs to take the power usage crown as well as having incredible performance.
Timothy Sottek @ Jul 1st 2008 9:37AM
Tom's Hardware just did a recent comparison between SSD power consumption and traditional platter drives.
If you're going to put that in a laptop beware: SSDs drain batteries FASTER than platter drives!
BobTurbo @ Jul 1st 2008 9:54AM
"SSDs drain batteries FASTER than platter drives!"
Correction: The SSDs tested using TH's methodology did not perform as well as the platter HDD.
Timothy Sottek @ Jul 1st 2008 10:08AM
Correction, part deux:
"However, we have discovered that the power savings aren’t there: in fact, battery runtimes actually decrease if you use a flash SSD."
http://www.tomshardware.com/reviews/ssd-hdd-battery,1955.html
BobTurbo @ Jul 1st 2008 10:19AM
Nope, I don't think you understand yet. Let me come back tomorrow and see if anything has changed.
(AND FINALLY)
Jeff @ Jul 1st 2008 12:36PM
I'll be less vague than Bob. The TH article was, as usual, significantly flawed. The benchmark they used to test battery life restarts itself after each completion, doing so until the battery is dead. However, the article did not report how many times the benchmark was able to run on the SSD vs. the mechanical HD.
The reason this is a critical flaw is that, as their own graphs show, the performance of the SSD was substantially better than the other HD. With a slower drive, the CPU spends a larger proportion of time in idle mode, as the system is blocking on IO, waiting for the slow drive to return data.
With the faster SSD, the CPU spends less time idling. It is using more power, but also doing more work. If TH had reported the number of times the benchmark ran on each machine, we would have seen the SSD machine run the benchmark many times more than the mechanical drive. Their conclusions are flat-out wrong.
The whole exercise was very silly to begin with. An SSD draws less power in use than a mechanical drive does while idle. Of course it will improve, or at least not effect, battery life. The article you reference is just another sad example of the decade-long slump Tom's Hardware has faced.
Good findings...This is very important, thanks for sharing!
Unfortunately, the article comes to miserably faulty conclusions, and nobody seems to have noticed it.
One thing everybody needs to note is this: the test that Tom's ran was designed to run a program to completion over and over again. All that Tom's came up with, thanks to that test, is that the computer with the SSD ran out of power faster. What they SHOULD have noticed, and what any sane and logical tester would have done, is counted the number of times the program RAN. A system with an SSD will run the program a significantly higher number of times because of its increased speed and lack of moving parts. With the higher number of runs, the CPU STAYS MORE ACTIVE, sucking more power out of the system and thus causing it to power down earlier. The only reason the HDD system died later is because the CPU idles waiting for it to retrieve data.
The SSD may have chewed up your battery faster, but it did, comparatively, run through that program a BUNCH more times. I am willing to bet the work per watt was much better out of the SSD system than the HDD - and once again, Tom's Hardware's scientific testing system proves itself anything but.
This is just like when any new technology comes out. People wonder why we need it and the old technology works just fine. Case in point, every memory advance we've ever made. DDR2 had faster speeds but higher latency so many people put it off as a failed technology without waiting for it to grow.
By the time ANYONE commenting here will be able to buy a reasonably priced high capacity SSD I'm sure they will be more efficient!
I was just on about this in the post about the 905/904 whatever version EEE that they were talking about.
SSDs are either on or off so low power or full power
HDDs are either low power or varied power according to use.
A low voltage SSD with a high usage voltage will use more power than a low voltage HDD with a moderate usage voltage.
The "on or off, low power or full power" might hold some ground... but even so, SSD's typically use less power full load than low load/spin hard drives... This is a TH review that I would not make references to.
There are fundamental flaws with this argument. Tom's Hardware conducted this review with flawed testing. It says "Read The Article" - I did that, and came to the same conclusion as many who commented on the original article itself.
The geist of it is: The experiment was done measuring a repetition of tasks through a period of time. However if you did the task 3 times faster than your competitor, you'd have to divide your power usage per period of time by three. Otherwise the results don't make sense.
This is beside the other obvious mistakes like "If a SSD is consuming a maximum 1W of power, then there is no possible way it can drain a fixed number of Wh from a battery faster than another drive that draws a minimum of 1.1W of power."
I think the majority of articles from Tom's Hardware is essentially good - this particular article drips of sensationalist news one normally associate with "technical writers" from your local rag.
Ha-Ha!
Well... that sucks.
Looks like the SanDisk and MemoRight are using less power than the Hitachi disc drive... I'm correct in assuming the Hitachi is indeed a dis drive, yes? Its 200Gb.
we got scammed again by tech capitalist!
PWN3D C1T1Z3N5!!
I suggest they give the technology time and it will eventually be better than regular HD's..... I believe they run quieter, boot faster and have a cooler temp... so those are the benefits as of right now.
And lighter, right?
Actually, temperature is related to power consumption because that's where the all the heat comes from - so a lower power HDD would be cooler than a SSD that takes more average power. But, add mechanical durability to the list of SSD advantages.
@morcheeba
That isn't true the wattage would determine the max heat but where all that heat comes from is in fact lost energy. Of curse the more power you run htrough something the more heat is put out becuase the more energy si lost. But since SSDs and HDs are vastly different technology SSDs do not lose as much energy as an HD so it can in fact be cooler.
exactly.
i have a laptop with an SSD drive, and also the EXACT same configuration with a 5400rpm HDD.
the fan that turns on to cool the WASTE heat generated by the HDD will eat the battery faster than the HDD or SSD itself.
HDD have moving parts which turn electical energy into motion based energy.
This is a very inefficient process and thus a large amount of the electrical energy is turned into heat.
since the ssd does not have moving parts, it is not facing the same electrical to physical energy transference and therefore is actually more energy efficient when taken as a whole system.
Well, then... My attention is now shifted even further away from the Eee series.
I replaced the 1.8" 4200 rpm ATA drive that came with my Latitude X1 ultra-portable by a 32 GB SSD, and it's much snappier, cooler, and got a significant (10 to 20%) increase in battery life. So I would take this article with a grain of salt.
I'm not sure I can trust someone I can't find.
You can take this article with a grain of salt, but I'm sure as hell not going to take your comment for anything more than that. No offense, but Tom's Hardware benchmarking is a little more rigorous, scientific, and all inclusive than your one scenario. Also, it is possible to get more battery life out of an SSD as the article says, but it won't happen for the average user. So you may be 100% right, but their article proves how that's far from true for everyone.
Hey Reader, check the comments of the TH article and you'll see he's right. The 'scientific' testing was anything but. The easiest test to perform and understand would be to watch watch a ripped DVD over and over until the laptop died. Not write testing but enough reading to debunk this hoax article.
Just like everything else (such as say, Nuclear Power), I and the rest of the American Public refuse to change our opinion just because new facts have come to light!
Nuclear Power is not safe! Not at all and should not ever be allowed to be built ever again!
DDT will kill all the birds!
SSDs save battery power!
well the sandisk SSD sure seems to be more efficient.. so im confused
I think this confirms what we already knew, don't buy a crappy mtron SSD.
Pfft...
Again I'm a little confued, according to that graph 50% of SSD's have lower power draws then a standard HDD in Idle and load. and the rest beat the HDD at load but not at idle. So basically the graph shows that SSD's have a lower power draw... which doesn't make much sense when you're stating that SSDs drain more power, when more then 75% of the time they draw less.
The test was flawed. The ssd completed the programs for the test faster and more efficiently because the cpu didn't have to wait long for a response but used up the power faster because of that. The Hdd took longer and the cpu had to wait more and idle using less power thus the hdd ran longer but took longer to preform the tests.
Uhhmm the graph shows the opposite, it shows that all SSD's perform better then a conventional HDD under load, and most at idle.
Read the article. It explains it much better than that graph does. Overall, the SSD's they tested drew more power in real-world applications than standard HDD's, mainly because they didn't have effective power saving features which result in them running at peak power nearly all of the time. A standard HDD runs at or near idle power consumption the vast majority of the time. While at peak consumption a HDD uses more power than a SSD, when you put them both in a real computer and test them the SSD draws more on average.
Too lazy to read the article... but dont regular HDDs spend a lot of time spinning even when not accessing data, or spinning up to just write or read a tiny file, whereas the flash memory would be at essentially zero power unless it was actually accessing the memory?
You'd have to look at average power draw over say 30 minutes of typical usage to get a real sense of which system was more efficient.
If you're too lazy to read the article, then why should anyone put in the energy to answer your question? You'll probably be too lazy to read the reply, too.
Read the article, dummy. Don't expect people to just spoonfeed it to you.
Your data is more secure? Nonsense. Even SSD's are guaranteed to fail eventually. Which means that you still have to make regular backups.. just like with a regular hard drive, or any storage device for that matter. So there goes that advantage out the window too.
So what have we got? They're just as secure, more power hungry, only marginally faster, and way more expensive than regular HDD's. Why on earth should I want an SSD then?
In related news Apple announces the new HDD iPhone with improved battery life.
That's the question I've been asking since SSDs starting coming into vogue. I see the value of SSDs in specific applications where it's necessary, but I don't see why we necessarily *should* make an industry-wide transition to them. It's just not necessary.
Guys, the yellow bar represent PEAK power consumption. A HDD only uses peak power when the actuators move the read heads. A HDD can consume close to idle power when reading data and not moving it's heads.
A SSD on the other hand has two modes. Idle and In use. While in use, the SSD consumes the maximum power.
I think people believe that when you are reading or writing data from HDDs then the actuators are moving the heads and you have peak power and when you are reading or writing data from SDDs then it is "in use" and is using peak power, but the SDDs are "in use" until the power time out hits - not when data isn't being sent to and from.
Everyone who hasn't read the article just read this page instead:
http://www.tomshardware.com/reviews/ssd-hdd-battery,1955-13.html
Much better graph which shows actual mobilemark bettery runtime results. The graph engadget used is just the specs for power consumption for the drives tested, which is nearly meaningless if you read the article (it explains in detail why).
That's the runtime value weighted by it's performance number.
What the graph actually means is a whole lot of nothing.
This graph is slightly better, but I and others question the methodology used here.
http://www.tomshardware.com/reviews/ssd-hdd-battery,1955-12.html
What we need is a battery-loading benchmark which is representative of REAL-WORLD USE. For example: boot up the machine with maximum charge at a controlled temperature and sit it on 5-6 auto-refreshing controlled content web pages. Reduce all the other power outlays to the minimum - Vista power saver mode, turn down the brightness, but maybe keep the wireless network connection just to keep it close to a real use-case.
Toms didn't do that - they ran a benchmark designed to "keep the notebook busy" and surprise, surprise the SSD's died faster under load - disproportionately to the power they consumed (perhaps NOT disproportionately to the work achieved in that time... but Tom's doesn't tell us how much "work" was achieved relative to the drive's performance... they don't even mention the power-saving settings.
Check out the test setup:
http://www.tomshardware.com/reviews/ssd-hdd-battery,1955-7.html
What's missing here? Oh... Vista SP1 would be a MUCH better choice than a server-based OS.
ALSO... Tom's failed to include the SAMSUNG MCBQE32G5MPP-0VA00 a new drive which is nearly as fast as the mtron and MUCH more power efficient if the sales literature is to be believed.... oh and they're awesome cheap (compared to mtron).
I think I'll skip the SSD until a later gen simply because you're whoring a website. :P
I love it when a review validtates my hardware purchasing choice. Yay 7K200!
Hard drive tech is much more mature than SSD tech is...if this is still the case in five years, I'll be surprised.
Haha, they just added "JUST READ THE ARTICLE" :O
Using Caps! So BOLD!
I always though engadget posts articles and forgets them, never thought they revisit it to see what their users are saying.
How else would they know when they've the spelling errors?
"On the other hand, you're still getting a drive with no moving parts, which at least makes your data more secure, if not more eco-friendly."
How does "no moving parts" = "eco-friendly"? My monitor has no moving parts and it's not eco-friendly ("but you shoulda seen the moving parts version!")
I think they mean "even though it may not be eco-friendly, ATLEAST it has no moving parts"
Might just be bad wording
This conclusion is biased! I cannot believe anyone would take this seriously! Shame on the people who did made the conclusion! I could take the Sandisk SSD and compare it to four random/crap HDDs in the market. Therefore, I can conclude that SSDs have much better performance and drains much less power.
In order to draw ANY conclusion, you need to test the population, not a sample of 1 versus a sample of 4. Stats 101.
The one-liner conclusion may be biased, but the article does prove the misconception about SSDs being more energy efficient all-around. Your results will naturally vary depending on what drives you test, but the point is that you can't measure energy consumption by idle-power alone -- you have to measure performance-per-watt, and frankly no one has done that until now.
The article doesn't prove anything. The test was flawed. Check the comments of the article.
their tests don't say what they seem to think they say. their test (questionably) shows that the standard hard drive is more efficient under load. that is, when running intensive applications like photoshop, you'll be able to get more done with the same amount of battery with a conventional hard drive (though it might take you longer to do it).
unless you're the type of person to run these sort of applications 100% of the time that your laptop is on and on top of that you're doing this while on battery power, this really doesn't mean very much to you during typical usage.
The Crucial and the Mtron at the end of the list has higher idle power usage, and Joshua already mentioned the gist of it: standard hard drives consume less power because they only hit peak when at load, and more often than not they are at idle.
However, the highest-performing drive on that chart is the Memoright, and at a glance, Tom's Hardware says that nearly all the SSD drives drain more battery than the Hitachi (Except the Sandisk, I think. That drive performs awfully, though). This might be a discrepancy, but note that they used a particular benchmark to simulate heavy laptop use. I suppose the Memoright and the most of the other SSDs hit peak power usage while the Hitachi alternated between idle and load (Going back to what Joshua said).
Theoretically, if you use the laptop for light web browsing, then maybe the SSDs would win out in battery life. That's kind of a sad situation, though. I wouldn't buy an SSD for that.
Ultimately, you'll currently have to swallow the extra battery consumption if you get one of these high-performance SSDs.
Perhaps they'll find a way to fix this in the future.
Actually this article is VERY misleading. Please read the comments on the associated article. Not only is it very poorly written, but their testing methods are not sounds at all.
Pointed out is the fact that SSD's are faster, meaning the computer doesn't sit around waiting on the hard drive as much. The test they used for battery life runs over and over and over as fast as possible and was not focused at stressing the hard drive in particular. This means less time was spent waiting on the hard drive for the SSD and MORE ITERATIONS were completed. This means the CPU, for instance, was in use a LOT more on the SSD. And an active CPU will drain power a lot faster than an active hard drive. The entire test was flawed in execution and the presentation shows it. Yes, their graphs are confusing, because only one of the MANY they throw at you even relates to the battery life. And that one given isn't even an accurate representation of the load on the system.
A far better idea would've been to normalize the result not only by runtime but by how many iterations were completed in each case. Or perhaps use a test that has a constant run time, like DVD playback. Seriously, it's all in the first few pages of the comments to the article if they haven't been deleted.
It just seems to show that some SSDs have awful power consumption. The Sandisk beats the hard drive for both idle and load and thus can't use more energy. And in my view it can only get better.
One question I've never actually seen asked, let alone answered: is there *really* a problem with HD reliability in laptops - or is this a perception thing? Most of the arguments for SSDs are: speed (iffy), low power (well - there goes that one), durability (the debate on real MTBF is still on) and ruggedness (technically, SSD should win this one).
However, there's a tendency to confuse 'potential' with 'reality'. If a product has an MTBF of 100,000 hrs and another with an MTBF of 500,000 hrs.. but it's typical use lifespan is 50,000 hrs - then the two products are essentially *identical* for this application, even though one is technically 5x 'better'.
The average lifespan of a laptop is about 2-3 years but Western Digital warranties their laptop drives for 5 years. One assumes they don't expect a significant number of failures in that time frame (or they'd shorten the warranty). Modern HDs also include advanced shock protection including accelerometers to park the drive head within 2/10th of a second to avoid head/disk collision.
Then there's cost. Yes, SSDs are getting cheaper - but then so are HDs. 320GB for < $200 still trumps 64GB < $250. For the price of 128GB SSD, you can get TWO 320GB HDs and use one just a s straight backup for the internal drive.
So, sorry - the arguments for SSD really just don't stack up. It's neat technology - and yeah, if play pingpong with your laptop (or are a fan of software tricks that do things with the accelerometer, and so constantly slap or shake your laptop, increasing the risk of impact), then SSD is for you.
For the rest of us, HDs just make more sense.
My laptop got about 30 minutes extra battery life because of my ssd, so that's crap...
That's only true if you like to drain your battery unnecessarily by leaving it idle.
I hope you guys refer people to a proper SSD website like dvnation.com
I don't care about their tests, but my battery life increased by 30 minutes from 4:30 to 5:00 when running on lowest performance, 1/7 brightness, and wireless w/ Bluetooth both on; ran a game server to ensure processor was being used and sdd was being accessed. downloaded around 50 GB of files to make sure wireless internet was being used, not just powered on.
I'm happy with this increase; allows me to get through a day without finding a charger or socket.
So far no one has mentioned my favorite part of the article:
"Could Tom’s Hardware be Wrong?
No, our results are definitely correct."
You know I keep seeing this being mentioned and the ting that strikes me as missing in the article at TH is that they do not take into account the increased efficiency of the computer when using SSD all they say is well if we have the same computer run with a hard drive it runs longs, BUT they forget to talk about the fact that when the laptop runs with the SSD it runs 2 - 3 times faster in many of the tests, ergo the CPU uses more power, the North bridge uses More power, the RAM uses more power, the South bridge uses more power and lo and behold the graphics card uses more power all because it can manage to do more calculations. So until they actually stick a Amp meter around the power cords to the hard drive and the SSD drive to see how much power they each pull through the tests I will not believe the crap that is in that article.
Wow, I haven't read Tom's Hardware since about 2000... looks like I haven't missed much.
yes
@Jinnai -
I was convinced by your argument and was going to add on to it until i read the Mobilemark2007 whitepaper. The modules ran during the battery life test DID level the amount of work done per unit of time under both SSD and HDD systems.
There are 3 modules TH could have used to test this (DVD, Reader, and Productivity). The DVD module will run the same video (work) per unit time. The same is true for the Reader module, which simulates 1 page turn every 2 minutes. For the Productivity module:
“The Productivity module is somewhat more complex. One cycle of this workload is completed every two hours. There is a fixed amount of work that takes place within that period of time, with the balance of the time taken by fixed-length and variable-length user delays distributed throughout the workload to simulate user “away” or “think” time. The variable-length pauses ensure that a faster system and a slower system will do the same amount of work in a given amount of time. This two-hour cycle is repeated until the battery is depleted.”
However, I am still not convinced by TH's article. I anxiously await more articles that will either debunk this one or add to its credibility.
Actually, the white paper says it has modules that do load balance, but those weren't the ones used for some reason. From page 3 of the comments on the article itself I quote mastrom:
"Going to http://www.bapco.com/techdocs.html I found the white paper for Mobilemark 2007. Under paragraph 2.5.2 Battery life rating methodology I read: "The benchmark generates battery life ratings as its principal metric. The battery life rating in MobileMark 2007 is measured in minutes. This metric reflects the number of minutes the system can remain operational while executing a chosen module. Each module will produce a different battery life rating, reflecting differences in system loading."
Everything ok until now. Each module should accommodate different usage patterns. And it continues: "The battery life is established by recording the start time of the benchmark, then repeatedly performing the workload. When the remaining battery capacity has fallen to 7% the benchmark records a timestamp once per minute. Once the battery has been depleted and the computer plugged in and rebooted, the benchmark compares the “start” timestamp and last recorded (“end”) timestamp. The battery life rating is the number of minutes between these timestamps."
Did you notice the important detail? "REPEATEDLY performing the workload". Well my friend, this is the reason your conclusion is completely off base. It is proved and accepted the SSD drives are faster (on average) than conventional Hard Disk Drives. Even your graphs prove this fact. However when you test the battery life by repeating the same workload again and again you force the system to perform more cycles of the same workload when the SSD is used compared to the HDD because most tasks in the workload wait for the hard disk to finish a task to move to the next. So unless you tell us how many times the workload was repeated by each configuration you can't compare the battery life times...
That’s is why the DVD playback test is so popular in battery life tests. It makes sure the computer will perform the same work per x amount of time. So please explain to me why you didn't publish the results of the "DVD2007: Battery Life" and the "Reader 2007: Battery Life" modules which are part of the MobileMark2007 as I see at the paragraph "3.0 MobileMark 2007 Scoring Methodology" in the white paper."
Interesting, I read that part differently. When I read "repeatedly performing the workload," I assumed it was a balanced or time-leveled workload, given the section I posted above. I did not take it to mean that the system would perform the workload at maximum capacity and repeat an indefinite number of times. Hopefully someone will be able to clarify this.
I also wish that TH would post the scores for the 3 separate modules though.
I totally disagree with the Tom's Hardware findings. At LAPTOP, we ran a test with an Eee PC 1000H where we took its 5,400 rpm hard drive out and put in a Samsung SATA II 64GB SSD. The result? 20 more minutes of battery life.
http://blog.laptopmag.com/eee-pc-1000h-faster-with-ssd-upgrade
"then my work is done"
gee... that snide reply REALLY makes me want to visit your spam site now!
if ssd indeed trains ur juice faster, then why no one ever complained about it before, no body brought a ssd yet?
1) noone can deny the advantage of faster bootup/performance. This test does not account for increased performance, just endless cycling.
2) In the real world isn't there more idle time?
TH fails again.
Regardless of the power state, give them time, Id rather have a drive thats is just a wiiii bit more power hungry if it has no mechanical parts. But I disagree with the tests on its relavence to the SSD themselves, esp. Tom's tests. Hopefully better tests will spring up.
and hopefully Engadget will take away that yelling text on the image of the article. I hate being yelled at for no reason.
good thing i decided long ago to hold out for the 320 gb ssd..by then they should sort this out!
Interesting. But I'm not convinced this test represents a real-world scenario very well.
Mischa Lockton ALMOST got it right. As mentioned, SSDs are faster. ...This means they get more done in less time. So, let's presume Tom is right and eb else is wrong, that SSDs use slightly more power than a HDD. They still probably use less power to get the job done because they take less time to do the job.
If an SSD uses a little more juice but takes a little less time to get the job done it still saves battery life. Also remember that while the clock is running on the traditional HDD you're wasting CPU+Cell+WIFI, etc.
the power savings was never substantial enough to be a buying consideration for me. it's all about reliability and performance. ever tried using a laptop on a bumpy road? ...ever had a HDD crash with important data because airport security in Charlotte, NC dropped it? it sucks.