36 Comments(s). 1 Pages(s). Showing page 1. [ 1 ]

   #32. Posted at 07:04 PM on Apr 17th 2008, Edited at 07:04 PM on Apr 17th 2008 Edit   Reply

Seeing as how I entered the PC enthusiast scene in 1997/1998, it is still a very odd feeling to find multicore Xeon processors selling at $300 and being reviewed for overclocking potential in a casual gaming station.
collapse

   #34. Posted at 11:51 AM on Apr 18th 2008 Edit   Reply

Opteron 165 represent! Yes, I beat Dissonance by 100mhz (2.5ghz)! Haha.

It's too bad the Xeon didn't overclock more, but great article.
collapse

   #2. Posted at 01:29 AM on Apr 16th 2008 Edit   Reply

Is it just me, or are the images stretched?
collapse

   #8. Posted at 09:40 AM on Apr 16th 2008 Edit   Reply

retail E8x00 chips are coming with a similarly slimline HSF guys. which of course is promptly thrown in the parts bin and a Scythe Infinity is installed in its place. :P
collapse

   #1. Posted at 01:22 AM on Apr 16th 2008 Edit   Reply

My, the .epee... speed freaks will eat this up
collapse

   #18. Posted at 01:15 PM on Apr 16th 2008 Edit   Reply

I have my x3350 running at 3.6ghz right now, and I hit the same wall at 425ish bus speed. This is because of the VTT and GTL settings need to be adjusted, bumping the VTT to 1.3v @ 95-95-108 GTL's I was able to do 450*8 with low voltage on the chip (1.3vCore), but my NB is in the 1.55range. These chips love high speeds, and do it at low voltage, but you have to use a nice board with VTT and GTL adjustments to be able to push them. I would try to get to 475*8, but the VTT absolute max from intel datasheet is 1.45v for 45nm's, so I may just stay where I am.

The duals do not have this problem, most of the time you can leave vtt at 1.1, but since the quads use the fsb to talk between the 2 sets of dual cores, the GTL's are very critical if you want a high fsb. Also, I am running a DFI x48, and as it turns out, if you want to overclock one of these chips the P35's handle the quads slightly better.
collapse

   #12. Posted at 10:49 AM on Apr 16th 2008 Edit   Reply

I am afraid your Xeon CPU cooler theory leaks water.

Purchased today new shipment in Australia Boxed quad-core Desktop Q9450 SLAWR.

Inside exact same low profile cooler.

Maybe it has more to do with 45nm power consumption generally which doesn't require the large cooler.
collapse

   #19. Posted at 02:03 PM on Apr 16th 2008 Edit   Reply

Thanks for running showing the power consumption and temperature. While I have always overclocked my processors a little bit, This past year (while I was in school) I've forced my Opteron 170 (2Ghz) to run at 1Ghz. I just don't need that much power for writing essays and surfing the 'net.

My next processor purchase will most likely be bought on the power consumption/performance concept instead of the usual price/performance.

Seeing the difference in power consumption is very nice to see.
collapse

   #22. Posted at 08:42 PM on Apr 16th 2008, Edited at 08:46 PM on Apr 16th 2008 Edit   Reply

Toldya in the recent 'deals' post that the multi would hold this thing back. I'd be as willing to bet it was a mobo limitation or setting as much as the CPU itself holding back the oc.
collapse

   #4. Posted at 02:28 AM on Apr 16th 2008 Edit   Reply

Your overclocking results now have me all paranoid as well. I took the Q9450 to 3.2GHz/1600FSB with no effort, but haven't been able to go higher due to memory limits (P35 = memory runs at least 1/2 of FSB speed, so 1600=DDR2-800).

Now I'm starting to wonder if there's anything more to be had. Buying 1066 to replace my 800 and finding it was a waste would be painful.
collapse

   #15. Posted at 11:33 AM on Apr 16th 2008 Edit   Reply

Greetings!

Regarding temperatures, the value that matters is the Core temperature, not the CPU temperature.

Think about it, ambient temperature would have to be 5-10ºC for the idle to be reported as 20ºC.
collapse

   #11. Posted at 10:42 AM on Apr 16th 2008 Edit   Reply

This was a cool little article. Next time can we see a similar article overclocking the nearest equivalent Opteron?
collapse

   #10. Posted at 10:16 AM on Apr 16th 2008 Edit   Reply

i love these kinds of quick-hit pieces.

can i request another one-pager, where you attempt undervolting the same proc and show the results from that? :)
collapse

   #6. Posted at 04:16 AM on Apr 16th 2008, Edited at 04:17 AM on Apr 16th 2008 Edit   Reply

Intel marks every processor with its unique operating voltage. According to their processor finder, the Q9300's VID range is 0.85 V - 1.3625 V.

http://processorfinder.intel.com/details.aspx?sSpec=SLAWE

Sadly, the SLAWF X3320's details don't seem to have been added in. When they are, expect to find it at

http://processorfinder.intel.com/details.aspx?sSpec=SLAWF

(The sSpec is as per the markings on the processor pictured in the article.)

I would be very surprised if there wasn't a large degree of overlap between their VID ranges.
collapse

   #5. Posted at 04:14 AM on Apr 16th 2008 Edit   Reply

So what are the chances of someday hitting 4.21jiggawatts on a Q9450? New Bios? Chipset? This one is sad because the X3110 sounds SOO much more fun to play with.
collapse

   #3. Posted at 02:22 AM on Apr 16th 2008 Edit   Reply

FWIW, I got that HSF but minus the copper core with my E8400. I doubt it could have done much.

Also, I haven't had a 6MB quad to check against, but judging by how much the 12MB on my Q9450 helps versus the 8MB on my Q6600, I'd strongly encourage everyone to drop a few dollars more and get a less-FSB-starved quad.
collapse
36 Comments(s). 1 Pages(s). Showing page 1. [ 1 ]
 
Name/Password: / Remember
Reply to:
[click to clear]

[RED] [GREEN]
[BOLD]
[ITALIC] [STRIKE]
[UNDERLINE]

Notice: All posts should abide by the rules, please.
Note: Ctrl-Enter submits the post. (In IE)
DThread keys: Click on a reply to position the blue bar. 'A'/'Z' move it up/down.
Jazztags: (they MUST be closed)
    r{ red }r     g{ green }g     /[ italic ]/     *[ bold ]*
    _[ underline ]_     -[ strike ]-     s[ sample ]s     o[ spoiler ]o  q[ (QUOTE) ]q