Low voltage ECO memory vs. regular

Ecological issues around computing. This is an experimental forum.

Moderators: Ralf Hutter, Lawrence Lee

Post Reply
qpc
Posts: 21
Joined: Tue Nov 03, 2009 11:49 am
Location: Canada

Low voltage ECO memory vs. regular

Post by qpc » Sun Jan 10, 2010 10:00 pm

I see that low voltage memory is not as popular as regular. At least on some online stores, based on the number of reviews.

I refer to low voltage as 1.35V and regular as 1.5V or 1.65V

Is there any reason why would one go with higher voltage? My choice would be 1.35V, but wonder about any back draws.

Thanks

MtnHermit
Posts: 99
Joined: Fri Jan 01, 2010 9:25 am
Location: Colorado

Post by MtnHermit » Sun Jan 10, 2010 10:52 pm

I know not but I just ordered a 4GB (2 x 2GB) kit of DDR3 1.35V for my new low power build. I didn't even know 1.35V existed until I saw a G.Skill ad about a week ago. Here's hoping. Will post some power numbers in a couple of weeks.

Klusu
Posts: 196
Joined: Thu Jul 24, 2008 10:57 am
Location: Riga

Post by Klusu » Mon Jan 11, 2010 1:21 am

Low voltage is not necessarily lower power. Anyway, the difference is small and hard to measure. Many "regular" RAMs would work at lower voltage.

PaulShapiro
Posts: 24
Joined: Sat Jun 07, 2003 9:30 am
Location: New York City

Post by PaulShapiro » Mon Jan 11, 2010 3:55 am

I bought 8GB of the G.Skill ECO 1.35v ram for a recent computer build. The memory defaults to running at 1.5v. The 1.35v setting is contained in the XMP extended memory profile, which both lowers the voltage and raises the memory speed. I did not see any noticeable change in system power consumption from the Kill-A-Watt power meter. Power fluctuates even at "idle" depending on what the system is doing, so I wouldn't have been able to see a few watts either way. To use the XMP memory profile your motherboard needs to include XMP support in the bios, and in my case it had to specifically be enabled. Alternatively, I could have manually set the RAM speed and voltage. Maybe keeping the same speed and lowering the voltage would have reduced power consumption.

MtnHermit
Posts: 99
Joined: Fri Jan 01, 2010 9:25 am
Location: Colorado

Post by MtnHermit » Mon Jan 11, 2010 5:39 am

Klusu wrote:Low voltage is not necessarily lower power. Anyway, the difference is small and hard to measure. Many "regular" RAMs would work at lower voltage.
All other variables being equal, lower voltage WILL net lower power. But, as you said, the difference will be very small.

In my case the 1.35V RAM was about the same price as regular 1.5-1.65V DDR3, so why not. I expect I'll have to manually set the 1.35V in BIOS, no big deal. This is somewhat like undervolting or overclocking, no guarantee just that margins exist in all these components and seeking those margins can have rewards.

Vilx-
Posts: 17
Joined: Mon Jan 11, 2010 5:24 am
Location: Latvia

Post by Vilx- » Mon Jan 11, 2010 5:56 am

I somehow doubt that 0.15V will make a noticeable difference in your carbon footprint, even in long term. If you want to spare power, start with the power-hungry devices like CPU, VGA or HDD.

So I'd go with the one that has the better stats/benchmarks/price, and not worry about the voltage.

OK, maybe the lower voltage means better overclocking capabilities (no idea, I'm no overclocker myself), but in that case you're not talking ECO in the first place. :)

qpc
Posts: 21
Joined: Tue Nov 03, 2009 11:49 am
Location: Canada

Post by qpc » Mon Jan 11, 2010 8:46 am

Nice replies. Thanks.

My idea about low voltage was based on i7-850 and its recommended 1.5V

I thought, OK, with 1.35V I'll be really on safe side and let my CPU be relaxed. No overclocking.

That's where I'm coming from originally.

Post Reply