• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

RAM power consumption

Joined
Mar 21, 2021
Messages
5,762 (3.64/day)
Location
Colorado, U.S.A.
System Name CyberPowerPC ET8070
Processor Intel Core i5-10400F
Motherboard Gigabyte B460M DS3H AC-Y1
Memory 2 x Crucial Ballistix 8GB DDR4-3000
Video Card(s) MSI Nvidia GeForce GTX 1660 Super
Storage Boot: Intel OPTANE SSD P1600X Series 118GB M.2 PCIE
Display(s) Dell P2416D (2560 x 1440)
Power Supply EVGA 500W1 (modified to have two bridge rectifiers)
Software Windows 11 Home
I have a laptop that runs Windows 11 well with 4GB (DDR3L), but am thinking to upgrade it to 8GB (DDR3L); but will this significantly increase power and so reduce battery time?
 
I have a laptop that runs Windows 11 well with 4GB (DDR3L), but am thinking to upgrade it to 8GB (DDR3L); but will this significantly increase power and so reduce battery time?
Are both SODIMM's populated?
 
Just one slot, so the 4GB gets replaced.
 
Just one slot, so the 4GB gets replaced.
It's very unlikely to be higher power consumption as the number of devices on the DIMM wouldn't have changed and you'd still be reading or writing only 64 bits per cycle.
 
Generally higher capacity DIMMs use a more advanced manufacturing and design, so the voltages and power consumption are similar.

There are some cases where they draw slightly more power, but you're talking about maybe one watt, if that, so I wouldn't worry about it.
 
From what I can tell, the 8GB version has twice the number of chips; but I think you are right, not worth worrying over.
 

Attachments

  • 4GB.jpg
    4GB.jpg
    133.8 KB · Views: 73
  • 8GB.jpg
    8GB.jpg
    14.3 KB · Views: 78
It shouldn't really end up with higher power usage no matter the density of the DDR3L module it only uses 1.35v it cannot go higher and I doubt you can oc it in a laptop.
 
If the number of chips is doubling, you might also be switching from single to dual rank SODIMMS. You could then memory speed test, and try to answer the long argued debate: 1R or 2R, which is faster? (answer: it depends).
 
The really big move for me was replacing the 500GB hard drive with a 256GB solid state drive I had around.
 

Attachments

  • 15-bs020wm.jpg
    15-bs020wm.jpg
    17.2 KB · Views: 67
If the number of chips is doubling, you might also be switching from single to dual rank SODIMMS. You could then memory speed test, and try to answer the long argued debate: 1R or 2R, which is faster? (answer: it depends).
Under load, dual ranked should be faster. For simple tests where you test the latency of one read, both should be close, but any memory bound workload should be better with more ranks.
 
Since you are replacing one stick with another, the power consumption will be similar. It will differ with a tiny margin. But it will not have any significant impact on battery life. Replacing that hdd with a ssd reduced power consumption alot more than whatever small increase a larger ram stick will cause anyways.
 
Under load, dual ranked should be faster. For simple tests where you test the latency of one read, both should be close, but any memory bound workload should be better with more ranks.
Shrek has, and we are talking about a) single SODIMM laptop, not b) multi stick multi socket with dual/quad channel interleaved memory controller. Your comment is true for b). My comment is true for a). And QED, we restarted the long argued debate.
 
but will this significantly increase power and so reduce battery time?
No.
the number of devices on the DIMM wouldn't have changed
Even if you would be adding a 2nd stick, DDR3 uses about 3 watts of power - when taxed. That is not enough to significantly impact consumption or battery run time. But the fact you are replacing one stick for another, that impact is even less, and likely negligible.
 
Even if you would be adding a 2nd stick, DDR3 uses about 3 watts of power - when taxed. That is not enough to significantly impact consumption or battery run time. But the fact you are replacing one stick for another, that impact is even less, and likely negligible.
That's an important point. The power consumption figures on the datasheets are typically the worst case.

Shrek has, and we are talking about a) single SODIMM laptop, not b) multi stick multi socket with dual/quad channel interleaved memory controller. Your comment is true for b). My comment is true for a). And QED, we restarted the long argued debate.
I'm also talking about a single DIMM, but let's not derail this thread with a debate about ranks.
 
No.

Even if you would be adding a 2nd stick, DDR3 uses about 3 watts of power - when taxed. That is not enough to significantly impact consumption or battery run time. But the fact you are replacing one stick for another, that impact is even less, and likely negligible.

And DDR3L even less

I think you are right and I may be overthinking this and should save my worries for bigger stuff (like replacing the hard drive with a SSD)
 
I think you are right and I may be overthinking this and should save my worries for bigger stuff (like replacing the hard drive with a SSD)
And of course, SSDs tend to consume less power than HDs so any additional power consumed by more RAM will quickly be negated by the lower power consuming SSD.
 
Power consumption would be less concerning to me than HP being HP on this product. There can be a fairly wide range of targets and expected behaviors that a company like HP lock down to very safe levels on production and OEM parts.

Should be fine. Good you are a bit worried and will probably notice any erratic changes. BIOS and hard enforced limits in capability would probably box in anything pushing the limits anyways.
 
Given that it's HP I'd have a hard time believing it'd do 5 hours, unless you turn the brightness to basically zero & force the lowest CPU clocks on power saving mode!
 
Just found out that the battery is 31 Wh

if so the consumption is around 3W, so 1 Watt here or there makes a big difference.
No it doesn't. First, 31Wh does NOT mean that battery is operating at that capacity 100% of the time. That is just the max. And the odds your CPU/GPU, RAM, drives, motherboard, NIC and screen will be demanding maximum power at the same instant for extended periods are next to none.

Even with RAM intensive tasks, most of the time the RAM is just sitting there at idle. And even when actively reading and writing data, that pales in comparison to the power demands of the CPU/GPU and especially the monitor. Hard drives consume much more. Even SSDs consume significantly more than that RAM.

I am reminded of an insightful statement a wise man once said - I suggest you heed it. He said,
I may be overthinking this and should save my worries for bigger stuff
:p ;)
 
No it doesn't. First, 31Wh does NOT mean that battery is operating at that capacity 100% of the time. That is just the max.

Wh is an energy, not a power; 1 Wh = 3600 J

But I agree that a 10 hr run time does seem like an exaggeration.
 
Last edited:
Wh is an energy, not a power
Come on Andy! That's a nitpicking nit!

True, energy and power are not the same thing but energy (in electronics) is measured in watts so they are very much intertwined. Regardless, any difference does not change the context or validity of my last post.
 
Come on Andy! That's a nitpicking nit!

True, energy and power are not the same thing but energy (in electronics) is measured in watts so they are very much intertwined. Regardless, any difference does not change the context or validity of my last post.
Wh is a measure of capacity, not usage...

100 Wh can power a 100 W device for one hour.

No it doesn't. First, 31Wh does NOT mean that battery is operating at that capacity 100% of the time. That is just the max. And the odds your CPU/GPU, RAM, drives, motherboard, NIC and screen will be demanding maximum power at the same instant for extended periods are next to none.
31 Wh means when the battery is fully charged it holds 31 Wh.

It has nothing to do with whether the device can consume 31 W, which it may or may not.
 
Back
Top