Account Options>

  1. Sign in
i5 2500k overclocking attempt 2 (now with documentation!)
 Share
The version of the browser you are using is no longer supported. Please upgrade to a supported browser.Dismiss

 
£
%
123
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
|
 
Still loading...
ABCDEFGHIJKLMNOPQRST
1
Stock + XMP4,5 GHZ try 1 + XMP4,8 GHZ try 1 + XMP4,5 GHZ try 2 + XMP4,5 GHZ try 3 + XMP4,5 GHZ try 4 + XMP4,5 GHZ try 5 + XMP4,5 GHZ try 6 + XMP4,5 GHZ try 7 + XMP4,5 GHZ try 8 + XMP
2
CPU Ratio setting (max ratio)auto454845454545454545
3
Internal PLL overvoltagedisableddisableddisableddisableddisableddisableddisableddisableddisableddisabled
4
Intel speedstep technologyenableddisableddisableddisableddisableddisableddisableddisableddisableddisabled
5
Intel turbo boost technologyenabled
6
turbo boost power limitauto
7
short duration limit
8
long duration limit
9
long duration maintained (sec)
10
additional turbo voltageauto
11
Core current limit150200200200200200200200200200
12
BCLK100100100100100100.2100100.1100.1100.1
13
Spread spectrumautoautoautoautoautoautoautoautoautoauto
14
15
RAMXMPXMPXMPXMPXMPXMPXMPXMPXMPXMP
16
DRAM frequencyDDR3-1600
17
CAS# latency (tCL)9
18
RAS# to CAS# delay (tRCD)9
19
Row precharge time (tRO)9
20
RAS# active time (tRAS)24
21
Command rate1N
22
Write recovery time (tWR)12
23
Refresh cycle time (tRFC)128
24
RAS to RAS delay (tRRD)6
25
Write to read delay (tWTR)6
26
Read to precharge (tRTP)6
27
Four activity windows (tFAW)30
28
Memory fast bootauto
29
memory power down modeauto
30
ODT WR (CHA)auto
31
ODT NOM (CHA)auto
32
ODT WR (CHB)auto
33
ODT NOM (CHB)auto
34
35
Power saving modedisableddisableddisableddisableddisableddisableddisableddisableddisableddisabled
36
CPU core voltageautofixed 1.350fixed 1.310fixed 1.310fixed 1.310fixed 1.260fixed 1.230fixed 1.240fixed 1.250fixed 1.260
37
CPU load line calibrationautolevel 1level 1level 2level 1level 1level 1level 1level 1level 1
38
DRAM voltageautoautoautoautoautoautoautoautoautoauto
39
CPU PLL voltageautoautoautoautoautoautoautoautoautoauto
40
VTT voltageautoautoautoautoautoautoautoautoautoauto
41
VCCSA voltageautoautoautoautoautoautoautoautoautoauto
42
43
44
Startupgoodgood, bus speed at 99,8, multiplier 16x (45x under load)good, bus speed at 99,8. muliplier 16x (48x under load)good, bus speed at 99,8, multiplier 16x (45x under load)Idle voltage at 1.298bus speed of 100.2 now :(
idle voltage: 1.248
bus back to 99.8...
idle voltage: 1.216
bus speed 100.1 :S
idle voltage: 1.224
idle voltage: 2.40
45
Burntest (10x 2gb run)stableAll cores 100%, 63.3-64 Gflops. cpu usage dips between some of the runs, temps dip some aswell.
Core #0 stays about 7C cooler than the rest.
Voltage changing between 1.352 and 1.360
result: stable
all @ 100% 65.5 to 68 Gflops.
Voltage=1.312 or 1.320 (alternating a bit)
Temps too high, but continuing for good measure.
Result: stable (hot)
Voltage drops to 1.208 or 1.216 under the load.
Result: stable
Load voltage at 1.312 or 1.320
Result: stable
Oddly Gflop numbers are down to 62.2 to 63. They should not drop when Ghz is the same, should they?
Nevermind, they got back to near 64, might have been some background task taking CPU time.
Result: stable
Result: stableResult: stable
46
max temps (degrees Celsius)46-5164-7365-7555-6263-7160-6860-6761-67
47
Prime95 about 15 minutesstableAll workers start without problem. (earlier attemps saw worker 1 fail right away)
multiplier stays at 45x (earlier attempts saw it drop inexplicably while prime95 was running)
Voltage troughout: 1.352
Core #0 around 7C lower than highest, #3 4C lower.
All workers start
After 1 minutes, worker 1 (core #0) fails
Resulting sum is wrong. Other workers keep running. Temps in general drop 6-10C
Result: FAIL
Worker #1 failed on startup, rounding error.
Result: FAIL
Despite a constant 100% load, temps seem to go up and down quite a bit, I have seen the current highest as 65 several times even though it tops out at 70
Result: stable
load voltage: 1.264 or 1.272
Result: stable
Load voltage: 1.232 or 1.240
Worker #1 (core #0) failed after 4 minutes, rounding error.
Result: FAIL
Load voltage: 1.240 or 1.248
Result: stable
Load voltage: 1.248 or 1.256
Result: stable
48
max temps46-5163-7264-71-62-7059-6658-6459-65
49
50
furmark + prime95 max temps52-58no point at this voltagerounding error after 5 minutes
51
notesRoom temp about 17C starting the tests.
Monitor software: CPU-Z 64bit 1.59 and Core Temp 1.0 RC3
CPU-Z seems to fail detecting the CPU the first time, but when restarted is always fine
Multiplier = 16x at idle despite speedstep and power saving being off
Loadline cal. lvl 1 chosen based on earlier results
Voltage = 1,31V because I do not accept temps over 72C in such a cold room anyway.
going back to 4,5Ghz. which seems more realistic to obtain around the 70C mark.
Trying Loadline calibration level 2 to see if it is enough, this was not a success!
Dropping voltage to 3.10V. will do finer tuning when point of instability is reached.
Ideally I want to see max temp on CPU-test of about 70C, since temp will rise come summer and with the GPU running.

Next attempt will again feature loadline calibration level 1.
At least this gives me some voltage numbers I know will be too low
I would say the room temperature is probably up 1-2C since i started

For future tests I will run prime95 first, then intelburntest, since prime95 seems to be the one showing off problems.
trying BCLK of 100.2 just to get even 100 in CPU-Z
Temperatures are starting to match my goal.
going up a full 0.010 to give a chance that it will be stable for hours.
I believe in this configuration, letting furmark and prime95 run for at least an hour.
20mins in, CPU max 66-73 GPU max 63

after 48 minutes, Worker #1 (core #0) has a rounding error.
CPU max: 67-74
GPU max: 63
temperatures are good, but core #0 seems less overclockable than the rest.
Aparently, furmark has been maxing out the core that is no longer running prime95.
Room temperature probably up a bit from 1½ hours of prime and furmark.
Getting close probably, any 4.5ghz result at below 1.3V would be above average from what I have seen.
The plan: leave computer priming at 1.260V tomorrow.
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
Loading...
 
 
 
Sheet1