Neoseeker : : : : Intel Core 2 Duo E4300 Review and Overclocking Analysis

Intel Core 2 Duo E4300 Review and Overclocking Analysis - PAGE 10

- Thursday, February 22nd, 2007 Like Share


next: Overclocking »




Get updates when we publish new articles

Comments

Sort by date: ascending descending
0 thumbs!
^
ati_guy Feb 23, 07
This article was an interesting read although I'm a little confused at the conclusions you drew.

You think Intel has a sure fire winner on its hands but your benchmarks show a dead heat against a less expensive competitor.

I'm NOT trying to start the old AMD vs. Intel fanboy argument and I'm only referring to the benchmarks and scores you present.

Now I love to OC stuff just as much as the next guy but it really does decrease the life expectancy of the part(s). It has also been my experience that most people don't actually OC their stuff.

So I read the article showing a tie (I counted and each proc won 10 non-OC benchmarks) between the two CPU factions in performance wins but one of them costs more than the other.

You hit the nail on the head picking the E4300 as the OC winner but I'm concerned about the "stability" you claim of that OC. You tested it for 20-30 hours but when will a user start to see the benefit fall off? No chip wins a benchmark if it's dead.

Outside of overclocking the only thing the E4300 can claim is that it costs more than an AMD part that performs just as well.

If you want to make this a clear winner I think you should run the thing OC'd until it dies while simultaneously running the AMD part without OC. Once you figure out how long the Intel part lasts you can figure out how many you have to buy over three years.

Because THAT is how you determine whether or not the E4300 is stable and worth the time to OC it and replace it after it dies in order to beat a part that costs less.

Methinks you may find that you would be money ahead if you spent the money upfront and bought a better CPU to begin with.
0 thumbs!
^
Anonymous Feb 24, 07
Nobody banks on having a constantly OCed chip last for 10 years. Eventually, they will all burn out notably faster than ones at stock voltages. If you are going to be burning up chips fast enough where your statements are a concern, then you're not doing it right. A rule of thumb is that if your chip dies from being over-volted, realistically it's time for an upgrade anyways. If you are in it for a very long haul...then don't overclock at all.

Props to you, bhenning, for that tasty 100% OC...on air!!

PM sent.
0 thumbs!
^
yxcvbnm Feb 25, 07
What motherboard did you use for the e4300 tests? I didn't find anything about that in the article.
0 thumbs!
^
Anonymous Feb 25, 07
It's in the set-up list. link
0 thumbs!
^
yxcvbnm Feb 25, 07
quote Anonymous1
It's in the set-up list. link
There are 5 different motherboards that support Core 2 Duo in the list and it says nowhere, which of rthem was used for the e4300.
0 thumbs!
^
Anonymous Feb 25, 07
Whoops, I misread your question. I thought you meant motherboards in general. Judging from the benchmark graphs, it would seem that the e4300 was used in the P5B...but that board isn't even listed in the set-up page.
0 thumbs!
^
Redemption Feb 26, 07
quote Anonymous1
Props to you, bhenning, for that tasty 100% OC...on air!!
The 100% overclock was not on air. I believe Bill will have the details, but with Air we were only able to hit 3.4GHz or so (Apparently beating out Anand and some others ). For the 100% overclock we had to resort to using active watercooling (by active I mean with something more than just a passive radiator and fan setup).

To address ati_guy's comment about lifetime, I'd say that if you were overclocking your CPU with extensive overvolting and running the CPU at very high temperatures then you'd definitely shorten the lifetime of the chip. However, in our case I don't think we hit very high temps relative to the overclock ratio. I think most people who overclock properly and safely can expect their life expectancy to be well over the obsolescence period of the chip.
0 thumbs!
^
bhenning Feb 26, 07
Hi,

I do believe the AMD is a far better value for office/general usage; but for media encoding the e4300 at stock speeds noticably beats the similarly priced x2 4200+, and it even beats the x2 5000+ at Doom 3, Halo, Jedi Knight,

The E4300 even beats an FX62 at stock speeds for Call of Duty!

The X2 4200+ does win at Comanche and UT2004 by a small margin.

As for the overclocked tests, the AMD's cannot touch the overclocking potential of the Core 2 Duo's.

Personally, I can hardly wait for the K8L cores so that AMD gets back into the race.

FYI, most of my systems are AMD based.


quote ati_guy
This article was an interesting read although I'm a little confused at the conclusions you drew.

You think Intel has a sure fire winner on its hands but your benchmarks show a dead heat against a less expensive competitor.

I'm NOT trying to start the old AMD vs. Intel fanboy argument and I'm only referring to the benchmarks and scores you present.

Now I love to OC stuff just as much as the next guy but it really does decrease the life expectancy of the part(s). It has also been my experience that most people don't actually OC their stuff.

So I read the article showing a tie (I counted and each proc won 10 non-OC benchmarks) between the two CPU factions in performance wins but one of them costs more than the other.

You hit the nail on the head picking the E4300 as the OC winner but I'm concerned about the "stability" you claim of that OC. You tested it for 20-30 hours but when will a user start to see the benefit fall off? No chip wins a benchmark if it's dead.

Outside of overclocking the only thing the E4300 can claim is that it costs more than an AMD part that performs just as well.

If you want to make this a clear winner I think you should run the thing OC'd until it dies while simultaneously running the AMD part without OC. Once you figure out how long the Intel part lasts you can figure out how many you have to buy over three years.

Because THAT is how you determine whether or not the E4300 is stable and worth the time to OC it and replace it after it dies in order to beat a part that costs less.

Methinks you may find that you would be money ahead if you spent the money upfront and bought a better CPU to begin with.
0 thumbs!
^
bhenning Feb 26, 07
Sorry, I guess the distinction got lost when I added the other motherboards to the list.

I used the Asus P5B-E, which is currently my favorite board for overclocking, although the P5W DH comes close.

quote yxcvbnm
What motherboard did you use for the e4300 tests? I didn't find anything about that in the article.
0 thumbs!
^
bhenning Feb 26, 07
Redemption is right, the 100% OC was not on air - it was a chilled water cooled rig that we will be modding for even better results - currently we are limited by the water block; as it is not a readily available commercial water cooling we have not published about it YET.

We were able to attain a stable 3.4GHz just with a Noctua 12, and even got 3.2GHz at stock Vcore with the stock heatsink (this was stated on the last page)

Somehow the P5B-E line got lost during editing, thanks for catching that, I've added it back.

quote Redemption
quote Anonymous1
Props to you, bhenning, for that tasty 100% OC...on air!!
The 100% overclock was not on air. I believe Bill will have the details, but with Air we were only able to hit 3.4GHz or so (Apparently beating out Anand and some others ). For the 100% overclock we had to resort to using active watercooling (by active I mean with something more than just a passive radiator and fan setup).

To address ati_guy's comment about lifetime, I'd say that if you were overclocking your CPU with extensive overvolting and running the CPU at very high temperatures then you'd definitely shorten the lifetime of the chip. However, in our case I don't think we hit very high temps relative to the overclock ratio. I think most people who overclock properly and safely can expect their life expectancy to be well over the obsolescence period of the chip.
0 thumbs!
^
Anonymous Feb 26, 07
My mistake, I thought you hit 3.6ghz with the Noctua. Still, props for the 100%OC regardless.

Chilled water cooler? Can you elaborate? How are you chilling the water, TEC?
0 thumbs!
^
bhenning Feb 26, 07
If you were really lucky and got a better e4300, one that could hit 3.6GHz at around 1.45Vcore, a Noctua would be enough.

You are correct; a peltier element is used to chill the reservoir, I can get it down to 4'C without worrying about condensation too much. Fans cool a heat sink on the hot side of the peltier

Currently I'm using a cheap Zalman waterblock which is not as afficient as I'd like; I hope to get some DangerDan and other better waterblocks soon; then watch out

However even with the current waterblock I am not temperature limited, I am Vcore limited.

quote Anonymous1
My mistake, I thought you hit 3.6ghz with the Noctua. Still, props for the 100%OC regardless.

Chilled water cooler? Can you elaborate? How are you chilling the water, TEC?


This message was edited by bhenning on Feb 26 2007.
0 thumbs!
^
Anonymous Feb 27, 07
It takes that special board to hit that serious amount of vcore^_^
With the 6-23 BIOS for my board I can hit 1.9v. (1.55 x 123%)
The 7042bta has the 123% disabled because too many people were misusing it and apparently causing some serious voltage fluctuations...or something bad of that nature.

Are you using sealed elements? Or are you just hoping for the best?

Would you be able to supply me with pics of your TEC modded reservoir? After debating with myself over the efficiency of such a set-up, I've been giving it second thoughts.

What are your thoughts on a dual TEC plate set-up that I have described in my above linked post?
0 thumbs!
^
bhenning Feb 28, 07
I may try to go further on the p5w dh, as I recall that get get to 2v

I am using a medical laser cooler... we have not started modding it yet, as currently the waterblock and vcore are the biggest limitations.

For my home rig, I am considering getting a copper tank, a 365W peltier (with a 40A/15VDC supply to drive it), and either a car or fridge radiator to cool the hot side of the peltier; using a 50/50 water/anti-freeze mix. I could regulate the temperature by pulse width modulating the power to the peltier; I'd have to, to avoid condensation. Such a rig would keep my processor cool, and my den warm

quote Anonymous1
It takes that special board to hit that serious amount of vcore^_^
With the 6-23 BIOS for my board I can hit 1.9v. (1.55 x 123%)
The 7042bta has the 123% disabled because too many people were misusing it and apparently causing some serious voltage fluctuations...or something bad of that nature.

Are you using sealed elements? Or are you just hoping for the best?

Would you be able to supply me with pics of your TEC modded reservoir? After debating with myself over the efficiency of such a set-up, I've been giving it second thoughts.

What are your thoughts on a dual TEC plate set-up that I have described in my above linked post?
^
Sponsored
Sort by date: ascending descending
Add your comment:
Name *:  Members, please LOGIN
Email:  We use this to display your Gravatar.

Sign in with
Comment *:
(0.1035/d/web2)