G.Skill GBPK DDR2 800 @ 1000 Mhz 2-3-3-3

[hersounds]

settings E4300 @ 3.0 Ghz ( 24/7) with G.Skill F2 PC2-6400 GBPK 4-4-4-12 @ DDR2 1000 Mhz 2-3-3-3 WITH 2.3 Vdimm

CPU - Z



SUPER PI 1M



SUPER PI 32M



EVEREST 4.10 TESTS

MEMORY READ



MEMORY WRITE



MEMORY COPY



MEMORY LATENCY



CPU QUEEN



CPU PHOTOWORXX



CPU ZLIB



CPU AES



FPU JULIA





FPU MANDEL





FPU SINJULIA






SISOFT SANDRA 2008



MEMORY BANDWIDTH



MEMORY LATENCY



the rams are settings in UNLINKED,and a little increment of vcore on the NB
( northbridge) for stability.

----- EDIT -----------

NO BIOS BUG on asus striker extreme ( I use 1102 STRIKER EXTREME 100% stable bios)


some diferent settings and see how afect in the sistem the changes.


4-4-4-12 DDR 1000 @ 2.3 Vdimm

18.734s




4-4-4-8 DDR 1000 @ 2.3 Vdimm

18.656s




4-4-4-4 DDR 1000 @ 2.3 Vdimm

18.641s




3-3-3-3 DDR 1000 @ 2.3 Vdimm

18.562S



2-3-3-3 DDR 1000 @ 2.3 Vdimm

18.500s



3.60 Ghz DDR 1000 @ 2.3 Vdimm 2-3-3-3

15.703s




3.75 Ghz DDR 1000 @ 2.3 Vdimm 2-3-3-3

15.282s

[maxine]

Plase remove heat spreaders and verify chip MOSEL or PROMOS and model.
Thanx.

[hersounds]

this time with themes disables on services.msc

15.140s



i waiting for g.skill answer,i'm not remove the heat spreaders ( for now)

[BNDMOD]

Hersounds!! really nice.... you know... tell us if you can which chip that kit has

[CyberDruid]

When and where were the G Skill purchased (me wants)?

[maxine]

This kit arrives to me for testing....

Paired with an Abit FP-IN9(650i) + E2140 can´t reach 2.1V+. ICs are PROMOS and more than 2.1V is a blue screen of death or hang.

It do 940Mhz at 4-4-4 at 2.1V,
740 at 3-3-3 at 2.1
and 950 at 5-5-5 at 2.1.

Conclusion: AVERAGE low cost kit.

So absolute no way for cas 2, so i´m sure you have the golden one kit, or your motherboard is bugged.

[dinos22]

i didn't think cas2 is possible with DDR2

but what the heck

have you got a 4MB L2 cache CPU to test at 3.6GHz to see how your efficiency fairs to others

[hipro5]

Good one BUT with Super-Pi 1M you just CAN'T SEE any differencies easy.....
For testing it......UNPLUG the one DDR2 module from it's slot and leave ONLY ONE module on......SINGLE CHANNEL...... Make your Super-Pi runs once more with ONLY ONE DDR2 module pluged and you will see that you will get ~same numbers.....

Try something else to compare.....A 32M for example.....

[The0men]

At only 2.3v I highly doubt cas2 at 500mghz is possible. Also for a quick reference, your everest latency is WAY to high. I dont mean to be sceptical but at cas4 mine's far lower than that at 500mghz, about 44.6ns

Have a look at the voltages required to run DDR2 at 500mghz cas 3, the lowest is like 2.8v

The bios bug not affect 680i boards at CAS3, but has anyone actually bothered testing cas2?

[hersounds]

the SPD are reprogramated in this rams, with thaiphoon burner

validation

http://valid.x86-secret.com/show_oc.php?id=244992

[maxine]

Her, please post subtimings with memset for checking with my kit.
And some BIOS photos help too...

Thanks.

[hersounds]

Good one BUT with Super-Pi 1M you just CAN'T SEE any differencies easy.....
For testing it......UNPLUG the one DDR2 module from it's slot and leave ONLY ONE module on......SINGLE CHANNEL...... Make your Super-Pi runs once more with ONLY ONE DDR2 module pluged and you will see that you will get ~same numbers.....

Try something else to compare.....A 32M for example.....

thankx hipro i trying that you say.the performance is not to much,but is better,than 3-3-3-3 that i try before. i keep on testing

[celemine1Gig]

Any news on that? Or did you find out that the results were just bugged?

[BenchZowner]

CAS2 @ DDR2-1000 with the current chips = impossible.
The times/bw in the benchmarks shown ain't good at all, and surely doesn't match what you'd get with real tCL 2 @ DDR2-1000.

Bug ftw

[celemine1Gig]

That's why I posted the second question.

At least it looked nice on the screens.

[The0men]

Just another clown thinking he has some world record, yet super PI times equal about the same as 400mghz cas5 bog stock... i dare say its just using cas5

It must be another joke?

[hersounds]

Just another clown thinking he has some world record, yet super PI times equal about the same as 400mghz cas5 bog stock... i dare say its just using cas5

It must be another joke?

hey be carefull with your words,anybody here is a clown,right? the rams are fine,i don't have time to tested,the eprom ( spd) are re rpogramated using taiphoon burner an d changing latencys ,using the HZ dumps,if you don't know shut the fu.k off! and don't post bull shiiit!

respect me and my respect to u.

[BenchZowner]

The0men, that wasn't nice of you.
You don't know whether a guy is doing something wrong on purpose or not.

@hersounds, even if you alter the SPD and program it for CAS Latency 2, it still won't work m8.

[BenchZowner]

For reference, here are some scores with a P35 motherboard ( Abit IP35 Pro ), garbaged Windows XP installation ( moved and used on 6 different mobos, with different chipsets ), untweaked sub-timings ( very relaxed on purpose ) and many background processes running.

_______________________
Tests Part I
Config: 333MHz FSB, 9x Multiplier, 2997MHz Core 2 Quad ( doesn't matter, it acts like a Duo or more precisely like a single core in SuperPi )
Super Talent T1000UX2G5 2GB Kit @ DDR2-999 5-5-5-15 2T

SuperPi 1M = 17.187s


Everest - Memory Copy = 6595MB/s


Everest - Memory Write = 6085MB/s


___________________________
Test 2
Configuration: 333MHz FSB, 9x Multiplier, 2997MHz Core 2 Quad
RAM @ DDR2-999 4-4-3-3 2T

SuperPi 1M = 17.109s


Everest - Memory Write = 6083MB/s


Everest - Memory Copy = 6927MB/s


Everest - Memory Latency = 57.1ns


--------------------------------------------

Notes...the latency depends more on the chipset in a comparison between P35 & 680i, this is on your side [ the advantage ].
The Memory Write bw is more dependent on the CPU Frequency + RAM configuration, than just RAM ( hence no difference between 5-5-5-15 & 4-4-3-3 ).

[hersounds]

This is what a need,i know that nforce 680i sli chipset,it's "slow",comparate with P35 in use memory,but this time,i see that timings was better,with your help have reference result,thank you so much benchzowner,i will use same settings qhen i test again.
cool.

[The0men]

@hersounds, even if you alter the SPD and program it for CAS Latency 2, it still won't work m8.

Maybe I should not have been so direct. But never the less. It's still never going to work. Hundreds of people have been there, and there was pretty much the exact same post a week or two ago. And as every one said back then. It just doesnt happen.

If he had scores/ times and voltages and even a half way beleivable frequency for CAS2 I wouldnt have been so brutally honest. But as we have already seen the exact same thing, its hard to see the light side of it??