E4300, low power CPU for desktop
Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee
E4300, low power CPU for desktop
I've just seen a review of the new Core 2 Duo model, the E4300 (9 x 200 MHz). It's not in English, but you can see that it draws 15 W less than a E6300 (7 x 266 MHz) in their tests. That's quite much for a small 66 Mhz difference, but remember that the mobo also draws less power because of the lowered FSB.
One of the reasons why the difference is so high could be because of the new L2 revision. We'll see if Intel have improved EIST as well. The power consumption in idle should be pretty low anyway since it runs at 1200 MHz and not 1600 MHz with EIST.
One of the reasons why the difference is so high could be because of the new L2 revision. We'll see if Intel have improved EIST as well. The power consumption in idle should be pretty low anyway since it runs at 1200 MHz and not 1600 MHz with EIST.
-
- Posts: 271
- Joined: Sun Oct 09, 2005 8:35 pm
- Location: CA
-
- Posts: 286
- Joined: Sat Dec 09, 2006 9:45 pm
- Location: Montréal, Canada
So the lowest multiplier is still 6 and you get 4 speed steppings? Hum... and it'll be socket 775 compatible with the cpus that support the E6x00 core2? Man... now this is tempting. Depending on price. I don't know which system to sell now, my A64 x2 3800+ or the core2....
Ah but wait! I hope intel didn't lock the lower VIDs. If it bottoms out around 1.1V, i'm out.
Ah but wait! I hope intel didn't lock the lower VIDs. If it bottoms out around 1.1V, i'm out.
-
- Posts: 271
- Joined: Sun Oct 09, 2005 8:35 pm
- Location: CA
Daily Tech article from Dec 19.
-
- Posts: 286
- Joined: Sat Dec 09, 2006 9:45 pm
- Location: Montréal, Canada
I just read the article a few hours ago, and I don't remember reading that (though I did read it pretty quickly). Everything I've read so far on the E4300 seems to indicate it does have Speedstep.
Here's an X-bit Labs article that shows the only thing it's lacking is Intel VT (which was also mentioned in the anandtech article).
Here's an X-bit Labs article that shows the only thing it's lacking is Intel VT (which was also mentioned in the anandtech article).
-
- Posts: 286
- Joined: Sat Dec 09, 2006 9:45 pm
- Location: Montréal, Canada
Intel Core 2 Duo E4300: Affordable and Highly Overclockable
http://www.anandtech.com/cpuchipsets/sh ... spx?i=2903
Running my E6300 @ 1.35v @ 7x 466 = 3262 Mhz...
E4300 should be easier to overclock due to 9x instead of 7x (e6300) and 8x (6400)...
http://www.anandtech.com/cpuchipsets/sh ... spx?i=2903
Running my E6300 @ 1.35v @ 7x 466 = 3262 Mhz...
E4300 should be easier to overclock due to 9x instead of 7x (e6300) and 8x (6400)...
-
- Posts: 286
- Joined: Sat Dec 09, 2006 9:45 pm
- Location: Montréal, Canada
I'm not sure i understand how this thing will be EASIER to overclock. Tell me if i'm wrong:
Right now i have a very overclock unfriendly P5B-VM and doesn't overclock anywhere over 285MHz. If that limit is inherent to the motherboard (because i hear the e6300 goes way way over that) then with a higher multiplier, i don't have to raise the FSB as much to overclock?
Right now my e6300 maxes at 1995MHz, then i can expect the e4300 to go 2565MHz right?
Right now i have a very overclock unfriendly P5B-VM and doesn't overclock anywhere over 285MHz. If that limit is inherent to the motherboard (because i hear the e6300 goes way way over that) then with a higher multiplier, i don't have to raise the FSB as much to overclock?
Right now my e6300 maxes at 1995MHz, then i can expect the e4300 to go 2565MHz right?
The higher the multiplier, the higher the OC will be @ the same FSBSpare Tire wrote:I'm not sure i understand how this thing will be EASIER to overclock. Tell me if i'm wrong:
Right now i have a very overclock unfriendly P5B-VM and doesn't overclock anywhere over 285MHz. If that limit is inherent to the motherboard (because i hear the e6300 goes way way over that) then with a higher multiplier, i don't have to raise the FSB as much to overclock?
Right now my e6300 maxes at 1995MHz, then i can expect the e4300 to go 2565MHz right?
Yes.
-
- Posts: 271
- Joined: Sun Oct 09, 2005 8:35 pm
- Location: CA
I'm more interested in undervolting, and it's SpeedStep capability. I just want decent performance without a lot of noise and thermal problems. This seems like a good chip on which to build a decent quiet desktop, WITHOUT having to resort to Merom and using odd-ball chipsets to get your laptop CPU recognized and similar contortions. I like the idea of the Merom, but the very limited choice of motherboards is a real problem.
IMO, overclocking is more at home over at the HARDOCP forums where they are all about lots of fans and painting flames down the side of the case.
IMO, overclocking is more at home over at the HARDOCP forums where they are all about lots of fans and painting flames down the side of the case.
-
- Posts: 286
- Joined: Sat Dec 09, 2006 9:45 pm
- Location: Montréal, Canada
I don't know about the e4300 but the e6300 doesn't undervolt anywhere below 1.162V, that is the VID doesn't go below. Unless you resort to pin modding, there's wasted potential there. So the only sensible thing to do is to OVERCLOCK IT until you use every volt to it's full potential, overclock it till it hits the stability limit of the 1.162V. Why would anyone leave it at stock speed when the undervolt just doesn't do it any justice.
I do not completely agree with you. It is true that voltage is the main factor in power consumption, but frequency affects too. In fact, it is considered that power consumption is proportional to V * V * f. Although it is smaller, power consumption is also reduced with lower frequency.Spare Tire wrote:I don't know about the e4300 but the e6300 doesn't undervolt anywhere below 1.162V, that is the VID doesn't go below. Unless you resort to pin modding, there's wasted potential there. So the only sensible thing to do is to OVERCLOCK IT until you use every volt to it's full potential, overclock it till it hits the stability limit of the 1.162V. Why would anyone leave it at stock speed when the undervolt just doesn't do it any justice.
So I would use the lowest voltage possible at the lowest frequency possible (within a reasonable limit) for idle state. In idle state you usually don't need extra performance. And I would use the highest possible frequency at the maximun safe voltage for load state.
-
- Posts: 286
- Joined: Sat Dec 09, 2006 9:45 pm
- Location: Montréal, Canada
21c at the moment.jaganath wrote:I gotta say, those are fantastic temps; what's the ambient temp in Sydney right now? Low 20s?
Temps seem almost too good to be true don't they. I've taken the overclock off and it's 40c full load.
Maybe the temp monitor utilities haven't caught up yet? But it does "seem" to be cooler than my opty.
kogi
Sounds like a great chip for silence addicts, lower power consumption and highly overclockable.
http://www.anandtech.com/cpuchipsets/sh ... i=2903&p=7
http://www.anandtech.com/cpuchipsets/sh ... i=2903&p=7