Does the 4850e really use less power than a 4800+[?]
Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee
Does the 4850e really use less power than a 4800+[?]
I've read some speculation that they may be specially binned. That would imply that the worse chips are being put into the mainline models. Can anyone point me to an undervolted par for par measurement?
It would have to affect mainly loaded performance, not idle, wouldn't it?
It would have to affect mainly loaded performance, not idle, wouldn't it?
Yes. 11 dollars over one year of 24/7 use at full loads 8 hours of the day. I think this makes it worth it, considering its only 15 dollars extra and the bonus right off the bat is the lower thermals.
Source is Tom's Hardware btw, don't remember the specific article off the top of my head, but I posted it earlier.
Edit: Sorry, didn't read clearly enough, just woke up. But yeah, I'm sure they just improved the overall manufacturing process and are skimming the top. Should be able to obtain better performance at lower voltages I'm just assuming. Also interested...
Source is Tom's Hardware btw, don't remember the specific article off the top of my head, but I posted it earlier.
Edit: Sorry, didn't read clearly enough, just woke up. But yeah, I'm sure they just improved the overall manufacturing process and are skimming the top. Should be able to obtain better performance at lower voltages I'm just assuming. Also interested...
Yes.
http://www.tomshardware.com/reviews/amd ... ,1925.html
I compares a ton (35) of AMD CPUs and their power consumption.
http://www.tomshardware.com/reviews/amd ... ,1925.html
I compares a ton (35) of AMD CPUs and their power consumption.
Re: Does the 4850e really use less power than a 4800+
I'll just note also, binning includes power usage. They most likely are binned, with the lowest leakage processors being used for the Energy Efficient series. Even processors with the same exactly clock speed can have huge differences in energy usage. Also obviously, the voltage is lower.astrayan wrote:I've read some speculation that they may be specially binned. That would imply that the worse chips are being put into the mainline models. Can anyone point me to an undervolted par for par measurement?
It would have to affect mainly loaded performance, not idle, wouldn't it?
Probably yes, but not worth getting a new 45W 4850e if you already had a 4800+. No processor is running at full load all the time; instead, processor stays at idle most of the time. Plus, 4800+ doesn't consume 65W at full loading, it certainly draws lower than rated 65W.
Btw, if I were to purchase a new system rather than upgrade existing AM2 processor, I would get a 4850e without thinking. Just my 2 cents.
Btw, if I were to purchase a new system rather than upgrade existing AM2 processor, I would get a 4850e without thinking. Just my 2 cents.
How big can the difference really be?
I've had a X2 4800+ undervolted to 2.4 GHz, at 1.125v. When fully loaded, AC power consumption increased by 32W, which means the CPU DC power consumption increased at most 25W.
With a 10W idle DC load, this means the undervolted CPU is consuming 35W DC at load.
The 45W CPU models run at 1.25v for comparison.
I've had a X2 4800+ undervolted to 2.4 GHz, at 1.125v. When fully loaded, AC power consumption increased by 32W, which means the CPU DC power consumption increased at most 25W.
With a 10W idle DC load, this means the undervolted CPU is consuming 35W DC at load.
The 45W CPU models run at 1.25v for comparison.