2160, Q737Axxx
OCCT and Prime almost 6 hours stable @ 3.5Ghz ~1.52V on air. FSB wall @ ~470 on air
http://img299.imageshack.us/img299/2...00primeab6.jpg
Printable View
2160, Q737Axxx
OCCT and Prime almost 6 hours stable @ 3.5Ghz ~1.52V on air. FSB wall @ ~470 on air
http://img299.imageshack.us/img299/2...00primeab6.jpg
Yeah I am finding the same. I have an Arctic Freezer 7 cooling mine with stock fan, and when I run fan at around 60-70% I can just about live with the noise, but load temps hit the high 60s at times and despite being very stable in Orthos, the machine reboots I assume due to high temps. I dont have any software setting that tells it to reboot anywhere if it hits high temps so assume it reboots as loses stability at those temps.
Ramp the fan up to like 85% though where it becomes noticeably audible, and the load temps always stay under 60 really. Im looking to either lap, remove IHS, and/or change my cooling to a TRUE or scythe with a couple of silent running 120mm fans, and possibly a duct to to the top of my case for fresh air. If load temps still hover so high, its back down to 3gig for me. What is interesting is I barely need much vcore to run stable at 3.2ghz either. I have vcore only at 1.34v which is actually around 1.3v under load on my board due to vdroop.
Who has removed the e21XX IHS yet? I gather they are not soldered on so come off easily. Is it the same as the Athlon 64 IHS removal pretty much? Also, what mounting issues does one have once it is removed, with regard to heatsinks? I read somehwere that it might be more prone to moving around being lose not clamped in on the socket mount. Is a shim required? What cooler works well? etc.
Reboot eh? I experienced the same...
I'm running my 2180 on an XBX2 board which was known stable to 412 FSB using a 6600. Beyond 412 FSB, it was mysterious reboot time running Orthos on that CPU (P35 board fixed that :cool: ). So I figured that the XBX2 would make a great board for this 10x multi 2180, but maybe not. I pushed up to 1.4 volts into the 2180 to attempt 3.2 Ghz, along with variations of MCH/FSB voltages known good for much higher speeds, but ran into the Orthos reboot issues. Interestingly enough it was Orthos stable if not for the random reboots. I believe that higher than 310 FSB or so is a limitation for my CPU. The 2180 runs Orthos stable at 3.0 very well at 1.2875 vcore (lowest my board will allow), so it's staying there. Temps peak out to maybe very low 60C's under small FFT's at that speed with a higher ambient room temp where the computer is running, so I'm happy with this setup. All in all, I suspect that there may be some other limitations to these cpu's beyond thermal transfer through the IHS that can cause issues with stable running.
Instructions here
http://www.xtremesystems.org/forums/...5&postcount=15
Just got my E2140, checked the stepping via Google and came up with nothing, any ideas?
SLA93 MALAY
Q734A241
How do you tell if its M0 or L2?
I'm hitting an FSB wall between 375 and 380 on my E2140 M0. Anyways--my Abit IP35-E won't give any more than 1.2V at load--regardless of set voltage in BIOS. I set it anywhere from 1.24V to about 1.4V... no change under load (read with Abit EQ and CPU-Z). I can't stability test at 3GHz (FSB375) as I have WinVista HP 64. Anyways--any stress-test agents around that work on Vista 64?
Anyone else notice that 945/G/GC chips don't like 200FSB cpus?
Mine (rig in sig) is stable only till ~285; whereas similar chipsets with 266FSB
cpus go up until ~350. My mobo even supports 333FSB, but with this cpu, it
just gives nasty artifacts and restarts or even locks up in worse cases.
Anyone tried a BSEL266 mod with positive results?
Anyways--I have some results finally. I eventually hit a wall at 3GHz--I could probably get another 8-32mhz out of it, but I figure what's the point (<1% boost) now, especially after the board's apparent stability issues. Here are my settings as they stand.
CPU & Chipset
FSB: 375 (1500MHz rated)
Multiplier: 8 X
Clock Speed: 3000MHz
PCI Express Frequency: 100MHz
CPU Core: 1.3575V (set in BIOS, 1.240V at idle and 1.296V-1.304V loaded) [+0.12V]
CPU VTT: 1.2375V [+0.375V]
ICH (Southbridge): 1.55V [+0.05V]
MCH (Northbridge): 1.290V [+0.04V]
CPU GTLREF: 63.0%
RAM
CPU/DRAM Ratio: 1:1.00 (DDR2 750)
Timings: 5-5-5-16 (Subs on Auto) 2T
Voltage: 1.80V
Here's my setup
Abit IP35-E Rev1.00, BIOS "14" (m630a14 for IP35/IP35-E)
Intel Pentium Dual-Core E2140 (Q712A647) M0
Scythe Infinity SCINF-1000 over the silicon thermal goop. I'll be cleaning the heatsink and processor within the week and replacing with AC MX-2.
Transcend JetRAM 1x2GB DDR2 800 5-5-5-16 2T
Diamond Radeon HD 3850 (RV670) 256MB GDDR3
Thermaltake Purepower W0100RU(iirc) 500W
Western Digital Caviar SE16 WD3200AAKS (with OS and primary storage) SATA
Western Digital Caviar WD800BB (with Linux space, swap files, and secondary storage) IDE
Lite-On 20x DVD Burner w/ Lightscribe, SATA
XClio 188A (right now), to be replaced by an XClio Propeller (as I can't fit the side panel on with the 188A, due to the massive size of the Infinity)
Microsoft Windows Vista Home Premium 64-bit
I don't have much for stress-testing right now. The PC I'm on right now has no working USB ports. I can't get 3DMark, PCMark, or even updated Catalyst drivers on there--as I had an initial copy on another computer that included CPU-Z, PC Wizard 08, SuperPi (XS mod), Apophysis, and that's about it. So, here are my results. Stand by for edit.
CPU-Z Validation
3.00GHz (375*8, load)
2.25GHz (375*6, idle)
2.93GHz (366*8)
2.00GHz (333*6, idle. I missed the validation for 2.66GHz, apparently)
Screenshots
SuperPi 1M
21.750s (3.00GHz, DDR2 750)
21.875s (3.00GHz, DDR2 750)
22.438s (2.93GHz, DDR2 733)
24.734s (2.67GHz, DDR2 667)
SuperPi 32M
19m38.750s (3.00GHz, DDR2 750)
Load Testing
Dual 32M 01 (max observed temperature--41C on CPU)
Completion #1 (23m40.796s)
Completion #2 (25m37.630s)
Idle Temperature (a minute or two after-the-fact, 25C)
Dual instances of Apophysis (40C max)
Temperature update:
Ambient/Idle/Load:
18C/25C/40C
22C/30C/42C
I hit 3.0ghz finally today on my L2 E2140 ive had for so long. I awoke this morning to find that our heater had broken, on the coldest day of the winter so far, 20 degrees farenheight outside, so i decided, hey, lets see how high i can push it.
I was working on a e2140 tonight. It had/has an FSB wall @ 415 wont be stable but it will boot and get into windows. @ 413~ FSB it will be rock solid with only 1.35v but even with 1.47v its not stable @ 415.
So tonight we thought we would try the BSEL 1066 mod. I must admit it was harder than i expected but was very pleased with the result. So we stick it in his Abit ip35-E and on bootup bios shows 8x204=1630 so im thinking damn didnt work. Checked it out in CPU-Z and it was 266! i was so pleased.
Then we tried to get it to get past the FSB wall. All i can say is with this chip going to 1066 doesn't do anything to the FSB wall. It seems to boot higher but again not stable. Does anyone know if there has been a successful attempt at increasing the FSB wall by using the BSEL mod? Ive read it worked with e4X00 series.
This is the maximum FSB of mine 2160 under a custom SS on Abit IP 35Pro...
Just a quick run at 8 x 400 1.39Vcore (maybe lower works too but not the goal here), temps are way high under orthos 60-61°C on both cores, but at least it did break 3ghz
http://users.pandora.be/OAP/E2140.JPG
100 % OC on air. AC Freezer 7 Pro, Gigabyte P35-DS3L, G.Skill F2-CL4D-2GBHK (4-4-3-5 2T @ 800mhz, 2.1V)
1.5 VCore in BIOS, +0.1V MCH, +0.1V FSB
http://img177.imageshack.us/img177/8495/cpuzrd1.png
Sub 20s SuperPI 1M, for a $75 CPU and $20 cooler? nice.
http://img177.imageshack.us/img177/4536/superpihz9.png
Unfortunately not Orthos stable for more than 20 minutes. Max long-term stable is 3100mhz @ 1.45V
http://img20.imageshack.us/img20/1588/orthosaa2.png
FSB Wall max bootable is about 400mhz.
2180 @ 3.0Ghz with 1.465v in bios shows as 1.49v in Uguru. 100% stable. will do 3.2Ghz but not stable with any voltage under 1.5v. fsb wall at 375. Its funny how some seem to have temp issues. I'm running a tuniq tower and my idel temps are 18 to 21c and load is 34 to 36c.
my best pi 1m score is 18.
What kills me is my buddies 6750/crossfire setup at same clock speeds will eat mine alive in FEAR benchmark (Peak fps part only), almost like i bottleneck my crossfire setup with mine. 45nm here I come.:up:
http://img84.imageshack.us/img84/6810/superpinj1.th.jpg
E2160, 3.2 GHz, L2 rev. :rolleyes:
1.5 or 1.55V as I can recall.
A bit old score, will try better when my new watercooling is done :up:
I guess I got a lucky chip or something, I've got an E2140 M0 from newegg (retail) that is so far, OCCT stable at 3.2 with 1.285v on a P5b with a cooler that works about as well as the stock one. I got it for a customer build but with some of the results I'm seeing I may just hang on to this little beasty and buy him another!
for who runs their E2160's OC'd 24/7 whats your core voltage? and what temps you getting?
Damn sorry for the above post....I'm loaded..... I'll try it again..
For those who are running 24/7 oc on their E2140 & E2160 What is your vcore at? and what temps you getting?
I got a E2160 a few days ago that im tweakin on... currently at 3.3ghz 1.48vcore 50-51C` temp (Ultra 120x w/120mm S-flex)
I got much more tweakin to do... What would be the max vcore you would run 24/7 on air ?
I have an e2140 @ 3.2ghz 400x8 @ ~1.32V stable in orthos tested for 6hrs+