turpit

Splendid
Feb 12, 2006
6,373
0
25,780
Excellent FACTUAL data from a reliable poster.

Thank you and well done.

I read your blog, I just try not to spoil it with my banal rethoric posts
 

Slobogob

Distinguished
Aug 10, 2006
1,431
0
19,280
A great thread. Thanks for putting it up. :!:

Regarding the Temp Graphs: You should choose different colors for the bars or put descriptors underneath them. :idea:
 

gOJDO

Distinguished
Mar 16, 2006
2,309
1
19,780
Very good read. :p It would be nice if you can post some benchmarks of real-life applications.
Anyway I don't get, what is the difference between the 3060 & E6600.
You can draw the charts with MS Excel
 
G

Guest

Guest
Sweet thanks Ninja, nicely done and comprehensive. In fact that post was so good that we forgive the delay between the reception of the chip and the posting of the data
:lol:

Thanks man, you bring a lot of value to these forums!
 

gOJDO

Distinguished
Mar 16, 2006
2,309
1
19,780
Other than a slight voltage drop for the 3060, they're exactly the same. Even the same CPUID. The 3060 is just a E6600 binned for server use.
Thanks. Than there is no need for benchmarking. I wonder how OC-able are the 3060 CPUs compared to E6600.
 

GrimReaperGuy

Distinguished
Aug 22, 2006
63
0
18,630
You should see a reasonably significant advantage regarding maximum overclocks for the 3060 compared to the E6600, the Xeon QC process is quite grueling. That said, I tend to find myself constrained more by the maximum FSB of the motherboard then by the processor itself, considering the meagre 9x multiplier.
Ninja, what board are you running this on? I'm wondering if it would unlock up with the latest ASUS BIOSes.
 
I tested using a Intel Bad Axe 975, and the 965 Asus P5B SE SLI and the P5W32 Deluxe. So far, I know it only works with the 975 and 965 chipsets. It could possibly work with 955, or 945, but I wouldn't waste my time trying.
 

apt403

Distinguished
Oct 14, 2006
2,923
0
20,780
i read your blog entry too, and you recommend the 3060 but what about the 3040? do you think it would have a similar edge over the E6300 or should i just spend the extra $20 on an E6400? ill be OC-ing for the first time with this processor.
 
I'm not sure about the Allendales, but since you read the conclusion in my blog, and you said you plan to OC, I must ask, do you need the 2 - 5C drop in temperature? Is it worth the $20 extra to you? Are you going to need every volt you can squeeze out of the processor? If you answer yes to these questions then I recommend the 3xxx Xeons to you. If of course you have a 965 or 975 chipset board.
 

F1N3ST

Distinguished
Oct 2, 2006
139
0
18,680
Glad to see this finally :). Seems that the 3060 has more OC potential, maybe a Xeon Extreme Edition would be sweet lol.
 

Dante_Jose_Cuervo

Distinguished
May 9, 2006
867
0
18,990
Yeah, I've always liked the DP xeons but now this is really awesome. Kind of like the Opertron 165 idea. Server quality on a desktop, me likee. Anyways, my mom needs a new comp and I was considering using one of these. What do you guys think?
 
If you can answer yes to any of the questions that I asked in the earlier post then yes. If not stick with the E6xxx series as you'll see little gains in performance till you OC.
@F1n3st
The chances of us getting a X3080 are slim to none. Keep wishing though. You know who'll be the first to test it when it comes out. :wink:
 

RichPLS

Champion
loadoccc1.jpg


Looking good!!! This seems like it would be a coveted part for overclockers, as long as the pricing remains within 10 to $20...

Is the temp drop with the Xeon due entirely to more efficient and higher binning? Same cooler comes with each?
 
With the exception of the box and the engraving on the IHS, I couldn't tell the Xeon apart from the Core 2. Exact same heat sink also. I know nothing of the process used to make these higher efficiency chips though. :(

@Wusy
What you did there is exactly how Intel bins them through voltage reduction, except they do it with expensive machine on a bare die. :Wink
Make your own Xeons? A DIY for overclockers perhaps? :p
 

Doughbuy

Distinguished
Jul 25, 2006
2,079
0
19,780
As with all Xeons, it is basically a standard desktop version CPU with added symmetric multiprocessing (SMP) capabilities.

From this article: http://www.tomshardware.com/2006/10/26/intel_woodcrest_and_amd_opteron_battle_head_to_head/page2.html

Which makes me wonder if the xeon really does run cooler and require a bit less voltage. I believe it does, but I could also believe that Ninja just got a better e3060 then he got a e6600...

Anyone have any proof i.e. linkies? I really don't keep up with the server market so any info is appreciated.
 
The funny thing is, that since this is a Conroe core, Intel turned the MP capabilities off for the 3xxx series. Only the 5xxx and 7xxx have the MP turned on. As gor you hypothesis, I really don't care how its proved (edit: okay, I do care) but I'd like to know exactly why the Xeon performed better than the E6600. If time permits, I'll test the 3060 against a second E6600.
 

avarice

Distinguished
May 10, 2006
633
0
19,160
May I suggest a better scale for the graphs - you have the high end of 2600 for the E6800 CPU at the top line which is at 2150 (or so). Anyway - since the scales are not correct it is hard for the graphic to point out what it needs to.

Of course since the first graph shows such a small delta between the first two processors (the delta is 5 instead of 4 which you state), I can understand the need to show the difference and to adjust the scale for this.

Ok - I am being critical of minute details. Very good data all around - well done.


Cheers.