Will going from fx-62 to 4850e draw less power?

All about them.

Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee

Post Reply
deathspared
Posts: 17
Joined: Fri Jun 13, 2008 10:11 pm
Location: san jose, ca

Will going from fx-62 to 4850e draw less power?

Post by deathspared » Wed Jun 25, 2008 12:06 pm

At first glance, it would be an obvious yes right? But after looking at Tom's Hardware comparison of AMD chips on Cool n Quiet mode, it is only a difference of about 3 watts.

http://www.tomshardware.com/reviews/amd ... 925-7.html

I run my pc about 17 hours a day and am drawing 100w. I'd like to get my wattage down even more.

Seeing as though my htpc is 99% in cool n quiet mode, is there any other benefit to me switching to the 4850e from my power hungry 64 fx-62 that I am currently using?



Thanks!

Zack

thejamppa
Posts: 3142
Joined: Mon Feb 26, 2007 9:20 am
Location: Missing in Finnish wilderness, howling to moon with wolf brethren and walking with brother bears
Contact:

Post by thejamppa » Wed Jun 25, 2008 2:32 pm

What other hardware you have? Changing motherboard might do greater impact but considering that FX-series is 89W+ and 4850 is mere 45W in load. Power cutdown is power cutdown. Still 100W's Idle is pretty respectable achievement.

I changed my hardwere into my current main rig hardware and I am using HD 3200 IGP instead discrete vga card. I am completely satisfied on my IGP and I manage to get 30W's off my system consumption. ( I need to get Kill-A-Watt somewhere though )

deathspared
Posts: 17
Joined: Fri Jun 13, 2008 10:11 pm
Location: san jose, ca

Post by deathspared » Wed Jun 25, 2008 2:41 pm

I think my point is that it appears the cool n quiet is so efficient that the watt difference between the fx and 4850e is very small when enabled. I am just wondering if there is some other benefit i would see.

I guess one thing is that it would probably run cooler even though the wattage is negligible. Or am I wrong?

~El~Jefe~
Friend of SPCR
Posts: 2887
Joined: Mon Feb 28, 2005 4:21 pm
Location: New York City zzzz
Contact:

Post by ~El~Jefe~ » Fri Jun 27, 2008 6:30 am

yes

deathspared
Posts: 17
Joined: Fri Jun 13, 2008 10:11 pm
Location: san jose, ca

Post by deathspared » Fri Jun 27, 2008 7:48 am

yes? How insightful.

Based on Tom's hardware's consensus, the power difference is very small in cool n quiet.

Why do you say yes?

jaganath
Posts: 5085
Joined: Tue Sep 20, 2005 6:55 am
Location: UK

Post by jaganath » Fri Jun 27, 2008 9:38 am

I guess one thing is that it would probably run cooler even though the wattage is negligible. Or am I wrong?
if there is no (or small) power difference, there will be no (or small) temp difference (assuming same cooling solution). power=heat.

would you mind listing the components in your HTPC? 100W is actually a fairly high idle, many on here have <50W idle now, esp. w/ AMD.

austinbike
Posts: 192
Joined: Sun Apr 08, 2007 5:09 pm

Post by austinbike » Sat Jun 28, 2008 4:42 am

Here is a comparison. I have a Gigabyte 690-based board (GA-MA69GM-S2H) with 2GB of memory, a 2.5" notebook HD, a DVD-RW, 2 120mm fans and an Antec earth watts power supply (~500W).

The draw at idle for an athlon 4800 dual core is ~42W. At load the draw is in the 60's if I remember correctly.

I recently dropped an Opteron 1356 in it and I am seeing ~60W at idle and up to 120W at load.

So I generally believe that Tom's numbers are fairly representative.

The idle numbers may be somewhat misleading however, because under load the procs may not behave the same. There is a minimum amount of power that is has to pull to power up all of its transistors. The lower power procs *probably* do a better job of scaling the power up as the load grows.

They are all the same silicon, but the low power procs have been selected (binned out) because of their electrical capabilities, allowing them to be more efficient.

Firetech
Friend of SPCR
Posts: 680
Joined: Fri Dec 16, 2005 4:50 pm
Location: Sydney, Australia
Contact:

4800+ > 4850e

Post by Firetech » Mon Jun 30, 2008 2:36 am

Has anyone here done a swap from 4800+ Brisbane to 4850e and noted a significant difference in heat output, cpu fan speed and therefore noise?
The tom's charts certainly show a difference but not a huge one.

The reason I ask is the large difference in purchase price here: $59 v $99.

NyteOwl
Posts: 536
Joined: Wed Aug 23, 2006 7:09 pm
Location: Nova Scotia, Canada

Post by NyteOwl » Wed Jul 09, 2008 2:16 pm

:icky you I can't find a retailer in Canada that has the Opteron 1356 available and I'm not sure a Phenom would work in a Budapest motherboard or not.

deathspared
Posts: 17
Joined: Fri Jun 13, 2008 10:11 pm
Location: san jose, ca

Post by deathspared » Wed Jul 09, 2008 2:20 pm

Well, now we have the definite answer.

I made the swap from fx-62 to 4850e and the power difference is about 10 watts on idle.

A noticeable difference but not as big as one might expect going from a 125w to 45w processor when using cool and quiet.

The temperature difference is about 9º cooler.

Based on my below hardware, how does a 90w idle sound? Not sleeping, but just sitting there.

Specs: This is what I am running:

Antec Solo

Modu82+ 425W

XIGMATEK|HDT-S1283

Gigabye 7600 Silent Pipe

SATA DVD R/W (IDE DVD R/W)

150GB Raptor / 1TB GP Drive

Dual analog tuner / HD Homerun

AMD 4850e

Asus M2n32 Motherboard

2GB Memory

jessekopelman
Posts: 1406
Joined: Tue Feb 13, 2007 7:28 pm
Location: USA

Post by jessekopelman » Wed Jul 09, 2008 6:53 pm

deathspared wrote: Specs: This is what I am running:

Antec Solo

Modu82+ 425W

XIGMATEK|HDT-S1283

Gigabye 7600 Silent Pipe

SATA DVD R/W (IDE DVD R/W)

150GB Raptor / 1TB GP Drive

Dual analog tuner / HD Homerun

AMD 4850e

Asus M2n32 Motherboard

2GB Memory
90W still seems kinda high. Is there really a need for a Raptor in this setup? Still, probably only costing you 5W beyond a more normal second drive or 10W beyond no second drive at all. I wonder what kind of draw that TV tuner has? Unless I'm missing something in my calculations, it is probably adding 15-20W AC to your total.

deathspared
Posts: 17
Joined: Fri Jun 13, 2008 10:11 pm
Location: san jose, ca

Post by deathspared » Thu Jul 10, 2008 9:52 am

After speaking with Hauppage, turns out the Hauppage 500 dual tuner uses 12 watts.

drees
Posts: 157
Joined: Thu Jul 13, 2006 10:59 pm

Post by drees » Fri Jul 11, 2008 10:08 am

deathspared wrote:After speaking with Hauppage, turns out the Hauppage 500 dual tuner uses 12 watts.
What sucks about the Hauppauge tuners I have (I have two PVR-250s) is that they use basically the same power whether or not you are using them, approximately 10w.

davidh44
Posts: 54
Joined: Wed Apr 16, 2008 8:51 pm
Location: USA

Post by davidh44 » Mon Jul 14, 2008 1:26 pm

Have you undervolted with CrystalCPUID? (Don't use RMClock for Brisbane CPUs...known stability problems when undervolting with dynamic p-states)

The combination of undervolting with dynamic p-states (basically a more aggressive Cool 'N Quiet) will give you the lowest power draw at both idle and load. With my 45w BE-2300, I'm undervolting, using dynamic p-states, AND overclocking by 40%. Best of both worlds...low idle power draw (0.825v, whereas CnQ default was 1.0v), higher performance when needed.

In my testing, voltage makes much more of a difference in power draw than the multiplier/speed of CPU.

Be sure to turn off Cool 'N Quiet if using CrystalCPUID, as the two will fight with each other.

And even though it is rated at 65W TDP, I have found the Intel E7200 45nm CPU to be as efficient (and even more efficient under load) than my AMD BE-2300 while trouncing the AMD in performance.

Post Reply