Comments on: Horrible analysis claims flash drives consume more power

By: Technology for Mortals » Tomshardware redeems themselves with new flash versus hard drive energy efficiency tests

Mon, 20 Jul 2009 04:36:20 +0000

[...] time Tomshardware posted an article on SSD Flash drives versus HDD Hard Drive power consumption, I gave them a well deserved “F”. They had botched the analysis badly and drew incorrect hypothesis because they [...]

By: George Ou

George Ou — Thu, 10 Jul 2008 17:14:11 +0000

Tomshardware was saying that SSDs only knew how to consume more power on average to affect battery life; not some instantaneous power jump. Your theory on instantaneous jump is not consistent with any kind of IO logic I’ve seen or the jumps are so small that no equipment measures them to the kind of extreme you’re talking about. I’ve seen very fine-grain meters showing variation in power consumption but not with the kind of instantaneous max/min you’re talking about.

By: Wes Felter

Wes Felter — Thu, 10 Jul 2008 17:09:30 +0000

"Tomshardware would have you believe that an SSD would consume 0.5 watts in idle and immediately jump to 2.9 watts as soon as you throw any work at it. So they’re claiming that an SSD pushing out data at 1 MB/sec would consume just as much power as it would at 100 MB/sec."

You’re both right here. I am willing to believe that these SSDs have only two power states, but during low load the drive is rapidly switching between these states. So for example, at 100MB/s the drive might spend 100% of its time active while at 1MB/s the drive would be active 1% of the time and idle 99% of the time. If you used a fast power meter you would indeed see the drive rapidly switching between 0.5W and 2.9W, but if you average the power consumption it would be somewhere in between.

By: George Ou

George Ou — Wed, 09 Jul 2008 03:31:52 +0000

http://blogs.zdnet.com/Ou/?p=906

Remember the 2.4 GHz problems I had. If you tried to use Wi-Fi at the same time, you’d kill the mouse. 2.4 GHz allows devices to have longer range because of the amount of transmit power you can use. 27 MHz doesn’t give you nearly as much transmit power so your distance must be short, but you won’t face any problems.

Personally I wish they’d use something like DECT 1.8 or 1.9 GHz for wireless mice and keyboards because the data rates are very low. I have a DECT phone already and the channels are only 1.7 MHz wide as oppose to 20 MHz wide used by Wi-Fi. This allows much better channel separation because you get 5 1.7 MHz channels in a 10 MHz block.

George Ou

By: nucrash

nucrash — Wed, 09 Jul 2008 03:25:57 +0000

http://www.tomshardware.com/reviews/ultimate-mouse-hunt,1114-10.html

One of my friends blasted THG for supporting the idea of a gaming mouse and how they would always claim the latest and greatest from Logitech had always seemed to eliminate the lag problem associated with their cordless models only to have their next model accomplish the same thing. This on the 2.4 GHz wireless spectrum, because, I mean who uses 2.4 GHz for transmission. Surely not cordless phones, or wireless access points, or blue tooth connections, or IP phones, or the list goes on and on my friends.

THG needs to be put out to the pasture, or better yet, shot and put on some one’s plate.

Some one please tell me that Maximum PC hasn’t turned the same way.

By: russki

russki — Tue, 08 Jul 2008 04:48:56 +0000

It used to be one of the best hardware sites on the ‘net, but sadly it turned for the worse about 3-4-5 years ago. Recent articles are poorly researched, badly executed, and almost-poorly written.

I also had a problem with their composite measure. If you want to measure one variable, let’s measure the variable. If you are trying to say there might be a trade-off, sure, there might be a trade off. But in the principle of isolating variable for testing, it seem counter intuitive to then multiply it by something else.

By: vic from manila

vic from manila — Mon, 07 Jul 2008 23:37:19 +0000

good job guys.. TH should post a correction and apology

By: George Ou

George Ou — Mon, 07 Jul 2008 18:18:16 +0000

Tomshardware came to the conclusion SSDs were less energy efficient based on bad analysis and no testing. I showed the problem in that analysis. My conclsion is that SSDs are more efficient, but the difference at present time didn’t justify the costs. That conclusion may change as prices for SSD keep dropping rapidly. I have no idea how you would read that my conclusion was the same or that they "did a good job".

George Ou

By: Jeff

Jeff — Mon, 07 Jul 2008 18:15:02 +0000

"But the analysis and conclusions drawn by Tomshardware are not to be taken seriously."

I thought this was a given…

By: Douglas

Douglas — Mon, 07 Jul 2008 18:14:48 +0000

So the result is the same but you disagree with how they did it? Seems to me they did good job of detailing how people use their computer and the battery life.