[MemoryReaction] G.Skill F2-8000CL5D-4GBPQ - Low in cost, high in speed!

[Sparda]

this shop rated the OCZ up to 2.2v: http://www.hpm-computer.de/product_i...-Platinum.html, if they use the same chip, then u can use the G.Skills also up to 2.2v?

maybe not realy....maybe they use differrent pcb ? it may give some impact to the chip.

[CERO]

hmm I PMed G.Skill Technician, and I asked if u use active cooling fe.e 2x80mm blowing directly on them if I can use up to 2.15v , he said no probleme, did anybody gain some better clocks when passed 2.1v?

[Cronos]

There is little to none improvement over 2.1V for this memory.

By the way, there is a trick to increase maximum stable clock , especially for 8GB configuration, which may or may not work for given set(s) and motherboard combination. Worked for me.

First, determine the maximum stable clock for each module in single channel configuration. Use memtest86+ to test stability,
10+ passes of test 5 and 5+ passes of test 4.

Then place all modules according the rule : best inserted in the sockets closest to MCH, worst -farthest. Works best for 4 dimm setup, but may also somewhat improve overclocking for two dimms.
For two dimms, they should be placed in the farthest pair of sockets (black on ASUS boards.)

[bkraptor]

For two dimms, they should be placed in the farthest pair of sockets (black on ASUS boards.)

Is there a scientific explanation for this or is it just based on personal experience?

[CERO]

A Asus technician stated that the Black slots on the P5K-line were better for overclocking idd. And Also confirmed by myself, possible because of a higher drive strenght

[Cronos]

After re-arranging the dimms according above rule, i was able to pass 200+ passes of Linpack64 with two different sets of kits (each set is 8GB) at 1000MHz/2.05V real 5-5-5-15 performance level 7.
I struggled for this over three weeks.

I am testing one pair of kits (8GB) now at 1000MHz 1.8V BIOS/1.85V real on P5K-Pro, 5-5-5-15 performance level 7.
So far so good. Incredible!

[TEDY]

someone told me that this are nothing but 6400 oced to work as 8000 ?

[Cronos]

someone told me that this are nothing but 6400 oced to work as 8000 ?

Yes, thats true. This also can be said about any other memory faster than DDR2-800, as 800 is maximum standardized DDR2 speed for now.

[EnJoY]

someone told me that this are nothing but 6400 oced to work as 8000 ?

Yea, no.

[BailA]

I have done some tests with my modules. 555Mhz is unstable no matter how many voltage you throw at them, so for 24/7 now I use the speed and timmings listed in my sig. 5Mhz drop allows to low the voltage 0.2v! Now rock stable at 1100Mhz 5-5-3-13 1.96v

EDIT: Damn guys, this is awsome. 5 hours Orthos stable at 1.88 volts! Just insane for a 2x2GB kit.

Have a valid setting for try this Frequency?
Thank's!

[BailA]

Edit

[CERO]

I tried to get my Gskill kit up and running @ 1081 5-5-5-15 tried up to 2.1v and I cant get it stable with HCI memtest. The board I am using is Asus Maximus Formula, for the TRCF( or sumtin) I cant choose anything higher then 42, but it reads on auto as :52. I really have no clue about subtimings, I think maybe thats the culprit? The subtimings are as follow:
On Auto subtming @ bios:
CAS# Latency : 5
RAS# to CAS# Delay : 5
RAS# Precharge : 5
RAS# ActivateTime : 15
RAS# to RAS# Delay : 3
Row Refresh Cycle Time : 52
Write Recovery Time : 6
Read to Precharge Time : 3

Read to Write Delay (S/D) :8
Write to Read Delay (S) : 3
Write to Read Delay (D) : 5
Read to Read Delay (S) : 4
Read to Read Delay (D) : 6
Write to Write Delay (S) : 4
Write to Write Delay (D) : 6

and here a sshot of memset:http://img218.imageshack.us/my.php?i...tgskillno4.jpg

Are my subtmings to tight?, as this is what I get when everything is set on auto.....if not I ques my kit just sucks.....

I just want to run 1081 mhz 5-5-5-15/12 with "low volts" , as I browse through your thread this should be easy attainable by many peeps already with 1.90-2.00v.... and I test it up to 2.1 still no go ANy help appreciated.

[EnJoY]

Try increasing TRRD to 5 or 6 (give both a shot), and also try increasing TRTP to 5. Try them both separately and play around with them.

[CERO]

Try increasing TRRD to 5 or 6 (give both a shot), and also try increasing TRTP to 5. Try them both separately and play around with them.

Hey enJoy, are those subtimings correct to start with, I just want to know if my kit is "normal kit" and not a flaky 1:X I wil deffo try that Thnx

[Cronos]

This memory wears down fast even at low 2.0V.

I have one pair of kits ( 8GB ) which was good enough to pass over 400 iterations in Linpack at 1000MHz without single error.

Retesting lately, no changes, several errors popped out during same 400 iterations. Have to reduce the speed. I fear in several
weeks of intense use i will have to further reduce the frequency.

[tomb18]

I got 11 kits to test, this is good memory for the price.

One important thing for everyone to know,
neither Memtest86+ nor Prime95 aren't reliable memory tests, at least for modern chipsets + memory combinations.
Both can pass many hours without problems whereas Linpack fails in several iterations.

For those truly seeking rock solid stability, Linpack test is absolute must. Yes, it will cost some grief for many .

Watch out for the temps while testing - 64bit version is the most demanding stress tester for Core2 ever known, temps will be substantially higher vs TAT(all cores), much much higher vs Prime95 Small FFTs. Even on water,
so air coolers beware!

No Kidding! Tried it out on my water cooled quad and the CPU's hit 72C! Never got more than 62C on any other test.

[CERO]

Try increasing TRRD to 5 or 6 (give both a shot), and also try increasing TRTP to 5. Try them both separately and play around with them.

didnt helped, mine seems to be the worst around here? oO, I runned memtest86+ like 20 full passes on stock speed, no errors found tho.

[Cronos]

mc2k
1081 is a huge overclock for this memory, forget this! Few selected kits can do this stable (if any), and anyway, this overclock wont last long!

[CERO]

mc2k
1081 is a huge overclock for this memory, forget this! Few selected kits can do this stable (if any), and anyway, this overclock wont last long!

do u think 5-4-4-12 is possible @ 1000mhz?

[Cronos]

do u think 5-4-4-12 is possible @ 1000mhz?

Not for me.

This may also depend on the value of MCH performance level. This is the most important timing. First try to minimize performance level, then try to minimize primary memory timings. The latter usually have rather small effect, very often overestimated.

[Dainas]

well, I do 5-5-4-15 2T at my current clocks(which is actually 1070) if that helps. Just don't think these sticks like their timings tightened unless you drop the clocks like a rock.

@ Cronos, I've been running at these settings for over a month. Folding the whole time. Never though of it as memory intensive but I know from past computers if things start to go iffy ram wise, you will have bluescreens from a blue moon to all out galore depending how bad. Will do a memtest run and yell if anything pops up.

[tomb18]

oops

[tomb18]

You should be able to run 1000MHz 5-5-5-12 at 2.0V. All of my 8 kits
can do this, even in 8GB configuration.

No matter what, i can't make this memory run in 8GB configuration at 1000MHz
on P5K-Pro . Sooner or later, Linpack fails.

Synthetickiller
Intel C2Q does not work 24/7 stable near 500MHz FSB, forget about !
Max stable FSB is 450-470MHz maybe, but this will require some voltage tweaking.

As for those 500+ you may seen, they are screen-shot only runs, not 24/7 stable.
Set your FSB to 400, memory to 1000MHz, it will work happily.

This memory wears down fast even at low 2.0V.

I have one pair of kits ( 8GB ) which was good enough to pass over 400 iterations in Linpack at 1000MHz without single error.

Retesting lately, no changes, several errors popped out during same 400 iterations. Have to reduce the speed. I fear in several
weeks of intense use i will have to further reduce the frequency.

Cronos,

It seems that there are alot of conflicting comments from you concerning 8G of this memory.
I just ordered 8G for my nvidia 680i board and am interested in just getting stock speeds and timings. It seems though that you are having problems at 1000MHz, which is stock, or are you changing timings and getting the problems?
Any pointers such as voltage?

Thanks

[Cronos]

Cronos,

It seems that there are alot of conflicting comments from you concerning 8G of this memory.
I just ordered 8G for my nvidia 680i board and am interested in just getting stock speeds and timings. It seems though that you are having problems at 1000MHz, which is stock, or are you changing timings and getting the problems?
Any pointers such as voltage?

Thanks

Not alot really, this is all very logical.
Obviously, initially i was moderately impressed by this memory, but further more extensive testing shows it is not as good as i initially thought.
But remember , i have VERY strict stability requirement. Like passing 400 full iterations in Linpack 64bit (takes 2 days) with ZERO errors, this is quite an enormous task. I am sure no one else have ever tested this or any other memory like i do, so obviously my results are worse than anyones else.

I cant say much about 680i boards. It looks like people are getting higher overclocks on 680i than on P35.

[KTE]

Not sure about this RAM but 1200 5-5-5-15 can easily be stable at stock volts with many 8500 kits.

Not alot really, this is all very logical.
Obviously, initially i was moderately impressed by this memory, but further more extensive testing shows it is not as good as i initially thought.
But remember , i have VERY strict stability requirement. Like passing 400 full iterations in Linpack 64bit (takes 2 days) with ZERO errors, this is quite an enormous task. I am sure no one else have ever tested this or any other memory like i do, so obviously my results are worse than anyones else.

You're not sure, you're guessing and that's why you're very wrong there. Try a week to 12 days stability testing.
Nearly every firm I've been to has their IT department run far more rigorous testing than you are doing, the engineers at our work place do so too and that's over 52 branches just regional. LINPACK is long outdated now, LAPACK is the one to be using for modern core archs and test basis for workstations/servers and still, RAM MFGs don't advise nor test RAM using LINPACK because there is no need to. Memtest, Windows Memory Diagnostic, Prime95 suffice perfectly for testing desktop RAM stability and are the industries long proven methods used worldwide. I would stick with using them unless someone finds serious flaws with them or specifically needs to run server load applications on a desktop system.