8 Packs 4ghz 32m Challenge win full Haswell 4770K Bundle.

PiKe said:
Ivy/Sandy users might Haswell not bother now...
Click to expand...

I have participated for the challenge, I knew I wasn't going to win. Just hoping 8pack posts so pointers as to which values should be looked at first to improve the superpi times, as his opening screen failed to show memtweakit screens, so we could see what he was adjusting.
 
Duff-Man said:
1.85V ?! Do you have a deathwish, 1Deep? :D 1.7v is the most I dare run through my sticks.

Very nice improvement from the extra voltage though... Almost tempts me into a suicide shot myself :D

From googling I *think* that I have Samsung IC chips in my XMS3 modules... They certainly run well at fairly tight timings (8-10-10-21 at 1070Mhz with 1.68v), but they don't really seem to respond so well to voltage. Perhaps if I pushed them up to 1.8v they would squeeze up to 1200Mhz at CAS9? Or don't Samsung ICs like the extra volts so much?

Are any of you guys "actively" cooling your RAM? I have a 12" desk-fan available that I'm tempted to shove into the side of my case :p




Also - any idea why CPU-z is reporting my RAM as being on a 1:8 multiplier? It should be set as a 10.66 multiplier (for 2133Mhz at stock BCLK)... Can CPU-Z not handle non-integer values of the multiplier?
Click to expand...

Hehe, my pc is watercooled and well ventilated so thought 'hmmmm wonder what more voltage would do.....' I was pleasantly surprised that I was able to drop the cas down by one notch.

I have no idea on samsung based memory, I'm sure they like more voltage, just make sure you keep them cool, put a fan near them to make sure.

Must be a cpu-z bug I think.
 
PiKe sad but true...... WE cant get anymore out of our cpus.

My room mate got 3570k i could go for record there with my samsung grens but no point tbh :/
 
Best I could get out of my 2600k on a MSI P67A-GD65 is ~8m 17s.

2600k8m17s.jpg

Never really tried doing this before, fun and frustrating especially when you feel you should be able to get a little bit more from it but the mobo just says no.
 
Out of interest (as it's more likely to tempt me into trying than not being able to win a prize is) what sort of framerate increase are people seeing in games (and which games) with their tweaked RAM timings?

Would I see faster video encoding times (i.e. using Handbrake) and if so what sort of %?
 
Updated Leaderboard/cleanup

1.Duff-Man : 7m 37.330s - i7 4770k "???" 2154Mhz 8-10-10-21-96 1T

2.shadow Boxer : 7m 42.947s - i7 4770k "???" 2212Mhz 10-10-10-21-177 1T

3.PCZ : 7m 53.141s - i7 3770k "Team Group" 2624Mhz 10-12-12-18 1T

4.Raikiri : 7m 57.285s - i5 2500k "Kingston" 1909Mhz 10-11-11-32 1T

5.1Deep : 7m 57.930s - i5 3570k "G.Skill" 2624Mhz 10-13-11-28 1T

6.Sir SpankalotUK : 7m 59.687s - i5 3570k "Samsung" 2284Mhz 10-12-11-25 1T

7.wazza300 : 8m 00.689s - i7 3770k "Samsung" 2209Mhz 9-10-10-19 1T

8.vonex : 8m 01.459s - i7 3770k "???" 2210Mhz 9-10-10-19 1T

9.Lord_ZED : 8m 04.657s - i5 2500k "Samsung" 2208Mhz 9-10-10-21 1T

10.Kaktus69 : 8m 05.598s

11.arknor : 8m 09.729s - i5 4670k "Kingston" 1596.4Mhz 9-9-9-27-208 1T

12.horrorwood : 8m 09.877s - i7 3770k "???" 2401Mhz 11-12-11-30 1T

13.PCZ : 8m 09.672s - i5 2500k "TeamGroup" 2208Mhz 8-10-10-16 1T

14.settler : 8m 09.997s - i5 3570k "???" 2133Mhz 9-10-10-21 1T

15.Mik3 : 8m 13.959s - i5 3570k "OCZ" 1907Mhz 10-10-10-30 2T

16.wazza300 : 8m 15.332s - i7 2600k - "Samsung" 2210Mhz 9-10-10-19 1T

17.Kaapstad : 8m 16.923s - i7 3930k "Corsair" 2484.2Mhz 9-11-11-25-96 1T

18.Monkeh : 8m 17.063s - i7 2600k "???" 2154Mhz 9-11-11-19 1T

19.MonsterMoshi : 8m 18.681s - i7 3770k

20.pgi947 : 8m 18.842s - i7 3820 "G.Skill" 2416.8Mhz 10-12-12-21-88 1T

21.whyscotty : 8m 20.940s - i7 3930k "Corsair" 2486.4Mhz 10-11-11-25-96 1T

22.Tonester0011 : 8m 21.785s - i7 3930k - "TeamGroup" 2356.8Mhz 9-11-11-25-96 1T

23.Subliminal Aura : 8m 23.580s - i7 2600 "???" 1412.8Mhz 9-9-9-24 1T

24.Rossi~ : 8m 24.106s - i7 3930k - "Samsung" 2152Mhz 9-10-10-21-88 1T

25.PureJaz : 8m 26.908s - i7 2600k - "???"

26.Mrnob : 8m 34.000s - i7 3820

27.Tekgun : 8m 34.926s - i7 2600k "G.Skill" 1617.6Mhz 8-8-8-24 1T

28.Skeng : 8m 35.084s - i5 2500k "???" "????Mhz 11-11-11-36 2T

29.dlknight : 8m 42.239s - i5 2500k "Samsung" 2133MHz 10-11-11-26

30.Hicks : 8m 46.158s - i7 3770k "Avexir" 1601.4Mhz 9-9-9-24 2T

31.Something Else : 8m 57.499s - i5 2500k

32.Duff-Man : 9m 03.910s - i7 930 "Corsair" 1830Mhz 8-8-8-21 2T

33.Kaapstad : 9m 07.889s - i7 something "Corsair" 2013.8Mhz 7-8-7-20-88 1T

34.Mick-1965 : 9m 12.911s - i5 760 "Cruicial" 2240Mhz 9-9-9-27-181 1T

35.Hotwired : 9m 14.129s - i5 750 "???" 1701Mhz 6-7-6-18-94 1T

36.shadow Boxer : 9m 14.414s - i7 920 "OCZ" 1601Mhz 8-8-8-24-59 1T

37.the1gooner : 9m 23.790s - i5 760 "Kingston" 2020Mhz 10-10-10-31-162 1T

38.Nightst@lk3r : 9m 24.471s - i5 760 "Corsair" 2000Mhz 10-10-10-21-74 1T

39.Gzero : 9m 30.481s - i5 750 "Kingston" 1603.4Mhz 9-9-9-24-208 1T

40.bru : 9m 38.733s - i7 920 "???" 1528Mhz 8-9-9-28-124 2T

41.Kei : 17m 01.141s - X4 955 "Kingston" 1680Mhz 8-9-8-18-22 2T

42.dfour : 23m 51.868s - FX-8350 "???" 1672Mhz 9-10-10-24-33

43.evohicks : 24m 09.666s - FX-8350
 
Last edited:
Kei I think you've got mine and raikiris score mixed up, I would love to be that quick though :-P

Edit: Nice :)
 
Last edited:
Hard to believe this result is allowed in the list.

4.Raikiri : 7m 57.285s - i5 2500k "Kingston" 1909Mhz 10-11-11-32 1T
 
Fixed, i also had you in twice.

I've also uploaded a few shots showing the difference various timing adjustments made to my score too. It seems that tRAS and tRC contribute quite heavily to the time. Command rate doesn't actually seem to make a significant difference.

17m 05.547s - 8-9-9-24-30 2T
17m 04.593s - 9-9-9-22-24 1T
17m 03.312s - 8-9-8-22-27 2T
17m 01.141s - 8-9-8-18-24 2T
 
PCZ said:
Hard to believe this result is allowed in the list.

4.Raikiri : 7m 57.285s - i5 2500k "Kingston" 1909Mhz 10-11-11-32 1T
Click to expand...

... It does seem a little incongruous I'll grant you that. 10s faster than any Sandy Bridge score on overclock.net, and they have guys pulling 07:42 on Ivy (and 7:20-odd on Haswell etc).
 
It is a strange anomaly, wonder if turbo was accidentally left on?

I'm not accusing anyone of cheating, there's no point, a haswell system will win this anyway.
 
Kei said:
Command rate doesn't actually seem to make a significant difference.
Click to expand...

Really?!

On x58 it took 5s off my score, and on Haswell ~3s. Seems to be the biggest single change I've made (more so than dropping CAS9 -> CAS8).
 
Last edited:
Duff-Man said:
Really?!

On x58 it took 5s off my score, and on Haswell ~3s. Seems to be the biggest single change I've made (more so than dropping CAS9 -> CAS8).
Click to expand...
Probably got a lot to do with the differences in the architectures. (just look at the chasm between amd & intel) It needs some more time put into it to work out the best compromise. Problem is, at 17m a go plus tweaking, it consumes time fast.
 
You must log in or register to reply here.
Back
Top Bottom