Flash is not green

[Steve_Barker]

Whilst not being a fan of Flash, I can now state that its not green as on of my objections:

http://www.theregister.co.uk/2008/12/02/gr...browsing_study/

Apparently, flash adverts use more energy to load than the page you want to see, and continue using energy after the download is complete.

1. Might have to close some of the 40 to 50 tabs I typically have open.

2. Flash blocker add on to Firefox, allows optional opening of flash you want to see when you want to see it.

[Simon]

Now that is interesting. Nice find!

[brettbh]

RFC 793 Section 2.10.

To say that Flash is not green is really rather silly. No, it isn't green, but then neither is watching TV, surfing the internet, playing video games or staying up so late at night that you need to switch on the lights. The problem lies not with Flash, but with the way that some developers/webmasters choose to use Flash. In addition to energy concerns, there are also security and performace concerns (ever opened a number of tabs and found that your browser suddenly becomes about as speedy as molasses?).

A more appropriate headline may have been, "Study spanks irresponsible webmasters and developers for abuses of power." Hmmm. But then the Flash ad at the top of El Reg's story may have made them appear to be somewhat hypocritical :-)

Whatever, at the end of day using NoScript is good way to reduce energy consumption, enhance security, speed up your web browsing - and, if enough people do it, spanking developers/websmaters into becoming more responsible.

[Steve_Barker]

RFC 793 Section 2.10.

To say that Flash is not green is really rather silly. No, it isn't green, but then neither is watching TV, surfing the internet, playing video games or staying up so late at night that you need to switch on the lights.

It can be argued that any human activity is not green, even things like recycling (third option on the hierarchy of waste management). However, is useful relatively, although sometimes its important to appreciate the parameters of the comparison.

[mountainhiker]

It can be argued that any human activity is not green, even things like recycling (third option on the hierarchy of waste management). However, is useful relatively, although sometimes its important to appreciate the parameters of the comparison.

Most of the time "flash" is used to make up for lack of ability to design an attractive site and the hope is people will be attracted to the flash and not notice the lack of content. It does use more resources....

[brettbh]

It can be argued that any human activity is not green, even things like recycling (third option on the hierarchy of waste management). However, is useful relatively, although sometimes its important to appreciate the parameters of the comparison.

My point was simply that it is wrong to point a finger at Adobe as El Reg did ("Study spanks Adobe Flash for abuses of power"). Flash is no less green than any other form of technology; the problem is how it's used.

[mountainhiker]

My point was simply that it is wrong to point a finger at Adobe as El Reg did ("Study spanks Adobe Flash for abuses of power"). Flash is no less green than any other form of technology; the problem is how it's used.

the problem is how it's used

Yes!

[EnviroGadget]

There are ways you can make a PC greener, regardless of what you're doing.

1) Upgrade the RAM to the highest the motherboard can support. This will reduce paging by the Operating System (i.e. disk accesses).

2) Use a memory card (solid state, SSD) rather than a conventional hard drive. The energy consumption difference is rather impressive.

Or you could get an 8W PC like this:

http://www.envirogadget.com/solar-powered/...st-8w-of-power/

Dan

[brettbh]

>>Upgrade the RAM to the highest the motherboard can support. This will reduce paging by the Operating System (i.e. disk accesses).<<

Sorry, but that's bad advice, Dan. To determine whether a system may benefit from more RAM, it's necessary to establish whether or not the system is currently performing paging operations and the extent of those operations. Should the system not be performing paging operations (or paging minimally) then adding RAM will make absolutely no difference to performance and will increase - not decrease - energy consumption (as RAM uses energy). Similarly, adding more RAM than is necessary to eliminate (or almost eliminate) paging would also be counterproductive.

In fact, adding RAM may not reduce energy consumption at all - even on paging systems. Theoretically, reducing or eliminating paging operations may slightly (very slightly) reduce energy consumption, but the saving would likely be completely offset (or more than offset) by the energy used by the additional RAM. There are a number of factors which would affect the outcome including the extent of the paging, the amount of energy used by the HDD or SSD when paging/not paging (which will vary), the amount of energy used by the RAM (which will vary too) and the extent to which the system is used. I suspect that, in the majority of cases, a small increase in consumption would be more likely than a decrease.

I have never seen any research related to this matter, but my gut reaction is that adding RAM is not an effective way to reduce energy consumption and that people would be better advised to spend their money on other energy efficiency measures. That said, should you know of any evidence pointing to the contrary, I'd certainly be interested to see it!

>>Use a memory card (solid state, SSD) rather than a conventional hard drive. The energy consumption difference is rather impressive.<<

Hmmm.

[brettbh]

I have never seen any research related to this matter, but my gut reaction is that adding RAM is not an effective way to reduce energy consumption and that people would be better advised to spend their money on other energy efficiency measures. That said, should you know of any evidence pointing to the contrary, I'd certainly be interested to see it!

I was interested to find out whether there is any validity to this and so asked an Environmental Technologist at Microsoft for an opinion. Here's her reply:

I checked around with my colleagues and we haven’t heard of this as a solid benefit or seen any whitepapers ... below are a few of the comments:

“This is sort-of more server oriented, but you can look at the graphs which show the power consumption of various system components. I suspect increasing RAM (now the single largest drain in servers) – even on a desktop machine – is likely to offset any possible gains from spinning down a single disk.

However, I could be wrong. I’d be interested to hear if you receive other responses.

“Power In, Dollars Out: How to Stem the Flow in the Data Centerâ€

http://www.microsoft.com/whdc/system/pnppw...wr_ITAdmin.mspx

“

------------------------------------------------------------------------------------------------------------------

“A typical SATA drive uses about 9W whether its doing anything or not (i.e. its spun up), so I think the answer is no, adding RAM won’t help reduce energy use (quite the opposite).â€

[mountainhiker]

I was interested to find out whether there is any validity to this and so asked an Environmental Technologist at Microsoft for an opinion. Here's her reply:

I checked around with my colleagues and we haven’t heard of this as a solid benefit or seen any whitepapers ... below are a few of the comments:

“This is sort-of more server oriented, but you can look at the graphs which show the power consumption of various system components. I suspect increasing RAM (now the single largest drain in servers) – even on a desktop machine – is likely to offset any possible gains from spinning down a single disk.

However, I could be wrong. I’d be interested to hear if you receive other responses.

“Power In, Dollars Out: How to Stem the Flow in the Data Centerâ€

http://www.microsoft.com/whdc/system/pnppw...wr_ITAdmin.mspx

“

------------------------------------------------------------------------------------------------------------------

“A typical SATA drive uses about 9W whether its doing anything or not (i.e. its spun up), so I think the answer is no, adding RAM won’t help reduce energy use (quite the opposite).â€

[mountainhiker]

“A typical SATA drive uses about 9W whether its doing anything or not (i.e. its spun up), so I think the answer is no, adding RAM won’t help reduce energy use (quite the opposite).â€

increasing RAM (now the single largest drain in servers) – even on a desktop machine – is likely to offset any possible gains from spinning down a single disk

Yep, this is on target with results from big web servers, if you are truly putting pages in RAM it is sucking up power and you may still beat on the disk...... so the same will apply to a desktop....

[brettbh]

Yup, this thread really illustrates the problems associated with obtaining information from ad-supported websites. The more content a website has, the greater their ad revenue will be and so quality takes second place to quantity. And, if you ask a blogger to write a green tip each day at $20 a pop, you can bet your bottom dollar that he'll find something to write about each day - and, more often that not, he’ll do that by reworking (mis)information from other websites. Here’s a recent example which appeared on numerous Windows websites, trickling down from blogger to blogger to blogger with each being paid for the submission and/or collecting the ad revenue:

When you have a computer with a recent model CPU, chances are it's a dual-core CPU. Both Intel & AMD have been producing dual core CPU's for a few years now. By default, Windows Vista will only use a single core during boot-up. You can easily change this from the System Configuration utility:

1) In Vista's Start Search type msconfig & hit the [Enter] key on your keyboard

2) Once System Configuration has started, select the Boot tab and click the Advanced Options button

3) Check the Number of processors check box, and choose 2 for the number of processors (Figure)

4) Click OK twice.

In my testing this reduced my boot-up time around 5 seconds.

http://www.helpwithwindows.com/WindowsVist..._Boot_Time.html

“In my testing ...†Haha! Caught in lie! You didn’t really test this at all, did you, Mr Blogger? Had you done so, you would have discovered that this setting did not decrease your boot speed at all! Why? Because the tip is BS! By default, Vista does not use only one core of a dual or quad core proc during boot, it actually uses all of them. The number of procs option in MSCONFIG exists so that users can change the default behaviour (set Vista not to use all cores) during diagnostics/troubleshooting.

But back to Dan's RAM tip. I’d guess that he obtained it from some commercial website which had copied it from some other commercial website that had posted it without checking it’s accuracy simply in order to obtain the ad bucks.

And that’s how misinformation is spread. And misinformation is why people end up buying things like bamboo USB sticks instead of doing things that may really help them become more environmentally-friendly!

I'm not being critical of websites which are supported by ad revenue, but people do need to remember that the primary goal of many such sites is not to help people help the environment, but to make bucks for the owner of the website.

[Simon]

"In my testing ..." Haha! Caught in lie! You didn't really test this at all, did you, Mr Blogger?

I wouldnt say that "around 5 seconds" is a difference... <_<

[brettbh]

I wouldnt say that "around 5 seconds" is a difference...

It'd save more than 350 hours each year in an environment with 1,000 desktops. Or, rather, it would if it worked!

[mountainhiker]

It'd save more than 350 hours each year in an environment with 1,000 desktops. Or, rather, it would if it worked!

5 seconds of hardcore processor time is a difference.....

[brettbh]

5 seconds of hardcore processor time is a difference.....

Huh?