Cell-SHOCKS ag. DDR3-2000 7-6-5-18 1T Prime_Pi Stable

[VictorWang]

Cellshocks released their DDR3 product.
PC3-14400 8-7-6-21 1G*2 Dualchannel kit.
rated 1.7v~1.9v.
sexy black box and blue heatspreader.

memory pictures:
http://vic.expreview.com/attachment/1188107508_0.jpg

more photos:
http://vic.expreview.com/attachment/1188107508_1.jpg
http://vic.expreview.com/attachment/1188107508_2.jpg
http://vic.expreview.com/attachment/1188107508_3.jpg
http://vic.expreview.com/attachment/1188107508_4.jpg
http://vic.expreview.com/attachment/1188107561_0.jpg
http://vic.expreview.com/attachment/1188107561_1.jpg
http://vic.expreview.com/attachment/1188107561_2.jpg
http://vic.expreview.com/attachment/1188107561_3.jpg


testing rig:
Intel E6850es E0 @ Ultra120
Asus Blitz Extreme, bios=0801
Cellshocks PC3-14400 8-7-6-21 1G*2 @ 8015 2000rpm ram-cooling fan
HIS HD2900XT 512m @ Koolance Watercooling
Topower 2000w Power Supply
Ambient=23c

testing software:
pi_1m
sisoft mem bandwidth
everest mem read test.
[prime95 blend test + Spi_32m] Sync = 1.85G+ mem usage.


testing results:

1) 1.64v 1800 8-7-6-21 stable
http://vic.expreview.com/attachment/1188107704_0.jpg

2) 1.84v 1800 7-6-5-18 stable
http://vic.expreview.com/attachment/1188107704_1.jpg

3) 2v 1900 7-6-5-18 stable
http://vic.expreview.com/attachment/1188107704_2.jpg

4) 2v 2000 8-7-6-21 1T stable
http://vic.expreview.com/attachment/1188107704_3.jpg

5) 2.14v 2000 7-6-5-18 1T stable
so we can see that, it's not about the MCH, but the ram itself can or cannot stable at high-mem-clk :D
http://vic.expreview.com/attachment/1188107704_4.jpg

6) more volt. 2.28v 2080 7-6-5-18 1T.
awesome bandwidth at only 3.64Ghz. :up:
http://vic.expreview.com/attachment/1188107755_1.jpg
http://vic.expreview.com/attachment/1188107755_0.jpg



it's late today, gonna have some sleep, will go on tomorrow for CL6, CL9 test :D

[Oliver]

Looks really good champ ;).

Keep pushing it :D.

[PanteraGSTK]

These modules seem to be better at lower cas than the corsair modules. Can't wait to see the higher cas testing.

[OCFX]

Very nice Victor :)

1.7v~1.9v...omg imagine paying $600+ for a memory kit and kill it with 2.0v with no warranty support :D

and would be perfect black sp with this sexy blue hs

[eva2000]

wow the speed binning has begun it seems :)

[VictorWang]

2nd day, show some CL5, CL6 test :D
---------------------------------------------------------------
2.24v 1800 6-6-5-18 1T stable, that's quite amazing !
http://vic.expreview.com/attachment/1188149932_0.jpg

same vdimm, 1800 6-5-5-18 1T pi_32m also benchable.
http://vic.expreview.com/attachment/1188149932_1.jpg

max CL6 memclk arround ddr3-1850

------------------------------------------------------------------
2.26v, 1545 5-5-5-15 1T test, awesome speed and bandwidth !!!
5-5-4-15 also benchable, but a little bit slower than 5-5-5, dont know why.
http://vic.expreview.com/attachment/1188149932_2.jpg
http://vic.expreview.com/attachment/1188149932_3.jpg

max CL=5 memclk arround ddr3-1580

-------------------------------------------------------------------
cpuz validation:
1) 1545 5-5-4-15
http://vic.expreview.com/attachment/1188149932_4.jpg

2) 2120 7-6-5-18
http://vic.expreview.com/attachment/1188149959_0.jpg

3) 2145 8-8-8-27
http://vic.expreview.com/attachment/1188149959_1.jpg

--------------------------------------------------------------------
CL=9/10 dont scale,
and i've tried flash Cellshocks spd to Supertalent(origin Micron's SPD), but no performance gain.
so DDR3 is still hard to have a conclusion which is good or not.
some is good for CL7, but some is good for higher clk, really really strange.

[VictorWang]

have to say that Cellshocks join the ddr3 battle with a great shock !
dont know if this is the world's first ddr3-2000 7-6-5 STABLE running :D

[Oliver]

Looks really good, these sticks are really impressive :D.

Great job champ ;).

[PanteraGSTK]

Awesome.

[Achill3uS]

AMAZING :D thx to share Victor :toast: you could sleep now :D

[jugeen]

:shocked: fast and furious! amazing scores! omfg! i need that!

[euonline]

Mega Shock :eek2:

[SeLecT]

Very nice results...!

:D

[Gautam]

Wow, really really super DDR3 clockin Vic. Totally blows everything else outta the water. :up:

[nut311]

how much price Mr.Vic?

[VictorWang]

Wow, really really super DDR3 clockin Vic. Totally blows everything else outta the water. :up:

:D i'm really excited with this pair, CL5/6/7 are soooooooooo cool :D


to nut311:
seems the kit will not > 500$ :)

[nut311]

good cell rock shock

[Infa]

They are allready in stock at MemoryC for $ 510.30

Cheapest High-end DDR3 so far...for compare..
the OCZ 1600mhz costs over 700...so its a fair deal!

[stone_cold_Jimi]

Just bought some from MemoryC for my winter build. So it'll lie in a drawer and "mature" for 3 months. :D

Thanks for the heads-up Victor. :up:

[Infa]

OMG :) how can you do that hehe...
*put it in a drawer* gee's :D

i practically allready put in my things in my comp before i even get home hehe... why buy it now anyway? if you wait the price will drop and so on, and the other brands will release perhaps better DDR3?

allthough its leaning for these babys for me too with X38 ;)

[stone_cold_Jimi]

OMG :) how can you do that hehe...
*put it in a drawer* gee's :D


The trouble is, the early sticks in a classy series are often the real stars. 3 months down the line either you can't get 'em or they've changed the IC to a lesser version. I paid a lot for G. Skill PC2-8000 4-4-4-5 early on and they've been champion. Then you couldn't get 'em.... then they changed the chips....

[Yoxxy]

Awesome results, awesome kit.

Nice cpu :)

Cheers!

[LexLuther]

Really good! Excellent Victor!

[saaya]

wow :slobber:
wow.....

And they clock well with cas7 and 8 too... :eek:

First of all, for those who dont know yet, i work for cellshock and do something similar to bachus who works for corsair.
*waves at bachus* hey michal :D

Despite the 1800 Label this is a 666 ES kit though, like i said via email.
THats why its dressed in sexy blue :D
I guess the seal and sticker caused some confusion, our packaging lady put 1800 stickers and labels on the blue kits and not the 666 ENGINEERING SAMPLE ones.

Blue = 666 ES
Black = 876/766

We decided to go for 1600s and 1800s but I wanted to make some cas6 mem as well since cas6 at high speeds runs really fast.

Cas6 needs high volts and we want to be careful and wait how micron ddr3 chips cope with high volts before we release memory that has a higher default voltage. Who is better at finding out if a mem can take high volts under stressfull conditions than victor and some other enthusiasts, thats why I convinced my colleague to make a few 666 ES pairs :D

The interesting thing now is that this kit clocks well with Cas7 and 8 too.
I honestly didnt test them for 766 and 876 hmmmm
Actually they seem to clock better than the 1800s awardfabrik and Kamel got.
This is really surprising, maybe kamel and awardfabrik still havent maxed out the mem and are still tweaking?

Our 1800s should do better than this kit with cas8 and cas7, but those results are crazy...

now im curious if all 666 ES kits we sent out run well with cas7 and 8 too, heh those 666 kits are truly evil :devil: :rocker:, we should put our red heatsreaders on them i guess haha :D

need to look at my notes and test the blue kits we still have here :D

[Kamel]

..."All Animals are equal, but some animals are more equal than the other..."

George Orwell ( Animal Farm )




Same happens for memory sticks...:rofl: :rofl:

[saaya]

yeah and in this case in the end the pig (ddr3) turns out to be "more equal" and not the race hors or fox like youd expect. Didnt have time to look at the other blue kits yet, too much to do, damn :D

[Lastviking]

I wonder how much vdimm they will take before they die :rolleyes: , to bad the ddr3 sticks are not so cheap.

[saaya]

Lastviking! Hey man! how are you doing? :D

Well using more volts than the sticks like doesnt make sense, you only get worse results. The more tight the timings the higher the sweet spot, but even for 665 the highest sweet spots are 2.25v. Well im using a p5k3 so i cant use more than 2.25v... :D
So i guess with cas665 the max that makes sense for some sticks is probably 2.35v, for cas5 maybe up to 2.45v but i dont think cas5 is worth it and cas6 and 7 make most sense. so the max voltage that makes sense to use is 2.3v and i doubt it will die at that voltage, unless it runs hot of course...

You have a lot of experience with mem, have you ever experienced something like this before? I never had any mem that had a sweet spot for voltage and didnt like more. if it ran too hot it oced worse or gave errors, sure, but thats not the case here.

[Yoxxy]

I am glad that you have said these were 1800C6 ES sticks. I and I am sure others would have been disappointed to find out the retails did not have the great results of Victors (consistently that is).

Thanks for clearing things up. Also >2.15v seems to lose a lot of gains. I think there might be on-die regulation finally on Micron chips. Just something to look at.

Also 2.45v through a 78nm circuit is going to annihilate it very quickly. The max voltage you want to use is around 2.3v. I predict >4 Hours @ 2.45v is going to kill your sticks.

First one to hook a pot up on the board will be able to tell though.

[saaya]

Thanks Yoxxy :)

I hope i wasnt too late, Shaminos flight was on Tuesday so me and Fr3ak showed him around in Nuernberg and Frankfurt, was real fun :D
I arrived in the office on wednesday and was burried in work and emails and then posted on thursday as soon as i realized victor thought this was 1800 mem.

Im really sorry if anybody bought the 1800s thinking itll clock like this.
I did some quick testing and some of the 1800 kits actually get very close to this on a p5k3, so they might clock the same on a blitz.
but the average 1800s clock worse, especially with cas6 and 5.
Kam and booblemagnifique received random 1800s, so their results are more representing.
The blue kits are just a preview of our (hopefully soon, looks at colleague :D) upcoming 665 mem.

[dinos22]

excellent clocks

surprised a bit with seemingly slow Pi times but i assume you didn't really give it a hard bench :)

cellshocks did well with DDR2, hopefully retail chips are like this...looks promising for sure :up:

[saaya]

the retail 665 should clock like this :D

[dinos22]

the retail 665 should clock like this :D

nice :)
you work for cellshock??

[saaya]

Yupp :)
Very nice people, if you ever come to germany lmk and ill show you around :D
The fabs are really cool.

[[XC] Aerosupra]

Yupp :)

good to hear ... always been a CellSHOCK fan :up:
and these 666 are nasty - CSH must have some godlike ICs on them
wow ... the CAS5 performance is outrageous!

I really am curious about the higher VDIMM survival of the DDR3 Microns

[dinos22]

Yupp :)
Very nice people, if you ever come to germany lmk and ill show you around :D
The fabs are really cool.

nice man gotta hook us up with some of that lovely ram :p j/k

hey what city is the fab in??


you really should have called those ram sticks 666:devil: :D

[stone_cold_Jimi]

The ones I just received are the 876 then. Still, I see elsewhere that they do:

"CellShock DDR3 @ 1928 with 7-6-5-16 (Micron D9GTR) and 2,2 V"

Which is worth playing with. :)

[indiecom]

@Saaya : If i buy from memoryC, can i find chip like this..??

Thanks

[STEvil]

Lastviking! Hey man! how are you doing? :D

Well using more volts than the sticks like doesnt make sense, you only get worse results. The more tight the timings the higher the sweet spot, but even for 665 the highest sweet spots are 2.25v. Well im using a p5k3 so i cant use more than 2.25v... :D
So i guess with cas665 the max that makes sense for some sticks is probably 2.35v, for cas5 maybe up to 2.45v but i dont think cas5 is worth it and cas6 and 7 make most sense. so the max voltage that makes sense to use is 2.3v and i doubt it will die at that voltage, unless it runs hot of course...

You have a lot of experience with mem, have you ever experienced something like this before? I never had any mem that had a sweet spot for voltage and didnt like more. if it ran too hot it oced worse or gave errors, sure, but thats not the case here.

Quite likely voltage/current regulation of the board (vtt circuit in particular, the regulator chips are often running horribly hot to begin with).

Yoxxy - On-chip voltage/current regulation isnt really feasible, dont think we'll be seeing it in the dumbed down form that gets kicked around forums.

[eva2000]

Lastviking! Hey man! how are you doing? :D

Well using more volts than the sticks like doesnt make sense, you only get worse results. The more tight the timings the higher the sweet spot, but even for 665 the highest sweet spots are 2.25v. Well im using a p5k3 so i cant use more than 2.25v... :D
So i guess with cas665 the max that makes sense for some sticks is probably 2.35v, for cas5 maybe up to 2.45v but i dont think cas5 is worth it and cas6 and 7 make most sense. so the max voltage that makes sense to use is 2.3v and i doubt it will die at that voltage, unless it runs hot of course...

You have a lot of experience with mem, have you ever experienced something like this before? I never had any mem that had a sweet spot for voltage and didnt like more. if it ran too hot it oced worse or gave errors, sure, but thats not the case here.
yeah some ram don't like too much vdimm... some of my kingston and mushkin BH-5 babies liked specific max vdimm voltages.. could be board/memory controller related too.

[saaya]

nice man gotta hook us up with some of that lovely ram :p j/k

hey what city is the fab in??


you really should have called those ram sticks 666:devil: :D

That was the plan, 666 and red heatspreaders, but victor kept asking for blue hehehe then i asked around and everybody liked blue...
Maybe we will use red ones for the final 666 :D

The fab is close to Karlsruhe, south to Frankfurt :)

The ones I just received are the 876 then. Still, I see elsewhere that they do:

"CellShock DDR3 @ 1928 with 7-6-5-16 (Micron D9GTR) and 2,2 V"

Which is worth playing with. :)
I checked around 10kits and found 2 1800 kits that do 2ghz 765 with 2.05v :)
orthos stable for 10mins, so it might not be 100% stable but pretty stable.

@Saaya : If i buy from memoryC, can i find chip like this..??
Thanks
You mean chips as in D9GTR? Our Sticks come with D9GTR and Z9HWQ.
Dont know why micron still labels some chips as Z9...
I checked and D9GTR and Z9HWQ clock about the same, so it seems to be the same chips just with different marking.

If you mean if you will get sticks like victors when buying the 1800 from memoryc... most likely not. there is a small chance... but if you really want 665 sticks like victors then wait a few weeks ;)

Quite likely voltage/current regulation of the board (vtt circuit in particular, the regulator chips are often running horribly hot to begin with).hmmmm thats what i would have thought if there wasnt such a big variety between the chips. i think its the chips, not the board, i think... :D

Yoxxy - On-chip voltage/current regulation isnt really feasible, dont think we'll be seeing it in the dumbed down form that gets kicked around forums.
i was also thinking that micron might have added a hard fuse that as soon as the current reaches a certain value starts to leak current to ground before it reaches the logic to prevent the chip from dieing.
But then again, why would micron even care... they dont get any RMAs with dead chips on them, do they?
Unless crucial and large partners ask for an overcurrent protection i dont see why micron would even care... and microns biggest partners are oems, and their customers surely wont overvolt their mem to the point that it fries...

yeah some ram don't like too much vdimm... some of my kingston and mushkin BH-5 babies liked specific max vdimm voltages.. could be board/memory controller related too.hmmmm well but think about this, if you have a few sticks that can do 1900 with a certain voltage in sc, some of them will do 1900 dc with the same voltage as well. i dont know what you think of this but to me this looks like the memory controller isnt limiting. otherwise the difference between sc and dc would be around 50mhz. then again, there are sticks that reach amazing speeds in sc, but in dc theres a wall no matter what timings, no matter what voltage.
one of the p5k3 boards i have here, one from the very first batch that was shipped, only does ~1825 dc max...

like victor said, its hard to come to conclusions about why ddr3 works like it does :D
well its still very early and jedec still hasnt really finalized ddr3 specs.
But the fun ddr3 brings will surely be enough motivation for us to figure out why it works like it does :D

[dinos22]

That was the plan, 666 and red heatspreaders, but victor kept asking for blue hehehe then i asked around and everybody liked blue...
Maybe we will use red ones for the final 666 :D

The fab is close to Karlsruhe, south to Frankfurt :)


666 red FTW :D :clap:

[STEvil]

ship with heatspreaders that change color depending on temps.

[dinos22]

ship with heatspreaders that change color depending on temps.

ROFL nah i think Crucial should get into that blingy stuff ROFL

seriously who really cares what colour they are

i mean WTF do you guys actually look inside your PC and thing wow that looks nice :ROTF: :D

[saaya]

well if id mod my case with lights n all then yeah, id want the colors to match... but besides that... i personally dont care about the color.
Still i wouldnt buy pink memory for example, even if its locked in my case and i never see it :lol:

[stone_cold_Jimi]

Still i wouldnt buy pink memory for example, even if its locked in my case and i never see it :lol:

Can I have the pink 666 sticks you don't want?

[Yoxxy]

Quite likely voltage/current regulation of the board (vtt circuit in particular, the regulator chips are often running horribly hot to begin with).

Yoxxy - On-chip voltage/current regulation isnt really feasible, dont think we'll be seeing it in the dumbed down form that gets kicked around forums.


On die voltage regulation is already on Elpida, Infeon/Anfeon, and Hynix DDR2. It is also on Elpida DDR3. It is pretty easy to add.

Maybe someone from Micron Texas will enlighten us.

[noobzed]

in daily use, do you see something change in using 2 ghz ddr ?

[saaya]

Can I have the pink 666 sticks you don't want? hehehe, nice try :D

On die voltage regulation is already on Elpida, Infeon/Anfeon, and Hynix DDR2. It is also on Elpida DDR3. It is pretty easy to add.

Maybe someone from Micron Texas will enlighten us.
Are you sure?
Well the Micron D9HNL chips sure look like they have an overvolt protection built in... but it seems to be broken as they stop scaling above 1.8v already.

But I dont think DDR3 has overvolt protections... Elpida DDR2 and DDR3 stop scaling with higher volts, but they dont produce errors. Micron DDR3 errors and even refuses to boot if the voltage is too high... so if they have an overvolt protection then its not working properly id say.
Plus Micron can take 2.3v+ and Elpida 2.1v+...

So IF elpida and Micron DDR3 ics feature an overvolt protection then its not working properly. If it would work dont you think Micron and Elpida would want to have it kicking in way before 2.1v for chips that have a default voltage of 1.5?

If Elpida has an overvolt protection for DDR2, then it kicks in at 2.1v.
So for DDR3 they would definately want to lower it to less than 2.1v...

Also dont forget, when lowering the latencies on Micron and Elpida DDR3 the sticks suddenly like higher volts than they do at higher latencies.
How would you want to build an overvolt protection that works like this?
Possible? sure! but not easy to implement, and if you screw it up your chips only run half the speed they would if that screwed up overvolt protection wouldnt be there.

in daily use, do you see something change in using 2 ghz ddr ?
Yeah, the reflection of a big grin on the monitor :D

[erwinz]

hey saaya..

whats a safe voltage for DDR3 for 24/7 use?? in overclock mode of course.. :D

[pinto]

My kit is nearly as good as victor's kit

first tests:

1.9v 890 76518 24/24 stable

http://img256.imageshack.us/img256/4459/pi324ghzddr3ad7.th.jpg (http://img256.imageshack.us/my.php?image=pi324ghzddr3ad7.jpg)

2.15v 940 796518 fsb limited by my QX6800

http://img440.imageshack.us/img440/5220/pi3214400940yl8.th.jpg (http://img440.imageshack.us/my.php?image=pi3214400940yl8.jpg)


2.2v 890 66518

http://img212.imageshack.us/img212/1495/pi3214400890665181rn2.th.jpg (http://img212.imageshack.us/my.php?image=pi3214400890665181rn2.jpg)

Really happy with this kit, P5K3 is a very good board too!

[saaya]

hey saaya..

whats a safe voltage for DDR3 for 24/7 use?? in overclock mode of course.. :D

1.9v :P

but if you keep it cool i think 2.25v is fine as well.
I think! im not sure... im still testing :D
Ive run mem at 2.25v with a fan on it for a few weeks now and didnt notice any degradation whatsoever.

Every mem is different, some probabaly get damaged before others, so id recommend you to bump the voltage bit by bit and see if you notice errors or degradation. once you do reduce the voltage and check if the mem is fine at that voltage.

[saaya]

My kit is nearly as good as victor's kit

first tests:

1.9v 890 76518 24/24 stable

http://img256.imageshack.us/img256/4459/pi324ghzddr3ad7.th.jpg (http://img256.imageshack.us/my.php?image=pi324ghzddr3ad7.jpg)

2.15v 940 796518 fsb limited by my QX6800

http://img440.imageshack.us/img440/5220/pi3214400940yl8.th.jpg (http://img440.imageshack.us/my.php?image=pi3214400940yl8.jpg)


2.2v 890 66518

http://img212.imageshack.us/img212/1495/pi3214400890665181rn2.th.jpg (http://img212.imageshack.us/my.php?image=pi3214400890665181rn2.jpg)

Really happy with this kit, P5K3 is a very good board too!
Nice! yeah i really like the p5k3 too :D
no bugs at all, and it was rockstable and everything worked from the first bios release.

[Fr3ak]

I am also testing the 1800 8-7-6-21-kit with black heatspreaders right now and my first impression is: Ram is good, but it's hard to reveal it's full potential, cos of crappy strap settings and limiting mb or CPU.

I still cannot run 900MHz stable, cos I need a very high VMCH to even boot into windows on the Blitz Extreme. 850 MHz runs fine and stable. Will have to find out tomorrow what is holding me back. I tested the board and CPU with up to 525MHz stable, so FSB-wise there shouldn't be a problem. Seems like the MCH is stressed a lot the higher the memory runs, so that might be a limiting factor. Maybe more voltage will help. I will find out :)

[pinto]

I think that your Blitz is limiting you. Did you try all available bios?

[erwinz]

1.9v :P

but if you keep it cool i think 2.25v is fine as well.
I think! im not sure... im still testing :D
Ive run mem at 2.25v with a fan on it for a few weeks now and didnt notice any degradation whatsoever.

Every mem is different, some probabaly get damaged before others, so id recommend you to bump the voltage bit by bit and see if you notice errors or degradation. once you do reduce the voltage and check if the mem is fine at that voltage.




thanks.. :)

hehehe I always use fan over my memories.. :) Cellshocks have very good memories.. :D

[saaya]

I am also testing the 1800 8-7-6-21-kit with black heatspreaders right now and my first impression is: Ram is good, but it's hard to reveal it's full potential, cos of crappy strap settings and limiting mb or CPU.

I still cannot run 900MHz stable, cos I need a very high VMCH to even boot into windows on the Blitz Extreme. 850 MHz runs fine and stable. Will have to find out tomorrow what is holding me back. I tested the board and CPU with up to 525MHz stable, so FSB-wise there shouldn't be a problem. Seems like the MCH is stressed a lot the higher the memory runs, so that might be a limiting factor. Maybe more voltage will help. I will find out :)

Hey Olli,
well it depends on what you consider a very high mch voltage :D
Im running the highest option, 1.7v on the p5k3 boards we have here for weeks without any problems. I always have a fan lying half on the mem half on the nb, so all is cooled well.

The relation between chipset voltage and the max ddr3 clockspeed is a bit odd. you might think that once the chipset hit the max speed it can do with a certain chipset voltage thats it, but if you increase vdimm you actually get higher results with the same chipset voltage. Weird eh? :D
So first i thought damn this mem needs a lot of vdimm to clock high, but actually it needs a lot less vdimm when you bump up the chipset voltage :D

Here is a graph i made that might be helpfull :)
EDIT: keep in mind that this graph not only shows what this boards mch can do, but a combination of what the board AND mem can do.
With another set it might have been stable at 1900 with lower vdimm for example.

http://img502.imageshack.us/img502/6288/fr3akcq4.png

I think that your Blitz is limiting you. Did you try all available bios?
yupp, but i think its a matter of the settings and not the board being limited to less than 1800 memspeed.
I hope! :D

In the end thats why asus is asking intel to create x48.
asus cant speed bin the chipsets in trays, and since their blitz boards use a different pcb they can bin the boards once the chipset is soldered on.
so the way the ddr3 mem controllers vary in their max speed asus and others probably have issues with high end ddr3 boards catching a bad nb and only reaching 1800 memspeed.

thanks.. :)
Your welcome :D
I just hope you wont kill your mem and then blame me hehehe
hehehe I always use fan over my memories.. :) Cellshocks have very good memories.. :D
Thank you :toast:

[saaya]

here are all settings im using fr3ak (and whoever else has problems with his ddr3) :toast:

advanced:
fsb 450(+)
dram frequency 1800(+)

cas 8
ras to cas 7
ras precharge 6
ras active 21

cpu voltage 1.55
dram voltage 1.9v
fsb termination voltage 1.5v
northbridge voltage 1.7v

everything else left at auto

please keep in mind that to reach 1800mhz memory speed you need an fsb of 450+ which some quadcores might have problems with, and which overclocks the cpu, so you might have to lower the cpu multiplier.

[jugeen]

now i got my new toys :) test coming soon.... ;)

[jugeen]

first screenshot :) http://img514.imageshack.us/img514/5763/cs18008762120070904io4.jpg

[eva2000]

I am also testing the 1800 8-7-6-21-kit with black heatspreaders right now and my first impression is: Ram is good, but it's hard to reveal it's full potential, cos of crappy strap settings and limiting mb or CPU.

I still cannot run 900MHz stable, cos I need a very high VMCH to even boot into windows on the Blitz Extreme. 850 MHz runs fine and stable. Will have to find out tomorrow what is holding me back. I tested the board and CPU with up to 525MHz stable, so FSB-wise there shouldn't be a problem. Seems like the MCH is stressed a lot the higher the memory runs, so that might be a limiting factor. Maybe more voltage will help. I will find out :)
when you loosen ram timings can you get it more stable at lower NB/vmch volts ?

[VictorWang]

oh, good to see Retail Cellshocks almost same performance.
u did good job saaya :D

[jugeen]

my first experince is a cooler over the kit is strongly recommended. if i test 1.9 vdimm W/o fan, memtest86+ 1.7 the kit make errors.

now i am testing the kit @ 1800 7-8-6-21 1.7 vdimm and seems to be ok after 1 memtest pass. i hope my kit is not defect.

[saaya]

oh, good to see Retail Cellshocks almost same performance.
u did good job saaya :D
thanks :toast:

my first experince is a cooler over the kit is strongly recommended. if i test 1.9 vdimm W/o fan, memtest86+ 1.7 the kit make errors.

now i am testing the kit @ 1800 7-8-6-21 1.7 vdimm and seems to be ok after 1 memtest pass. i hope my kit is not defect.

Yes! please use cooling on the mem!
You mean 7-6-5-18 timings right?
If it errors at 1.9v then it could be that those sticks dont like high vdimm.
Start with low vdimm and see how far you can go, only then bump up vdimm.

[jugeen]

sry, the correct timigs are 8-7-6-20.

thanks :toast:



Yes! please use cooling on the mem!
You mean 7-6-5-18 timings right?
If it errors at 1.9v then it could be that those sticks dont like high vdimm.
Start with low vdimm and see how far you can go, only then bump up vdimm.

[indiecom]

oh, good to see Retail Cellshocks almost same performance.
u did good job saaya :D


Do u try Cellshock with black heatspreader like this :

http://www.memoryc.com/products/description/2GB_CellShock_DDR3_PC3_14400_1800MHz-8_7_6_21-Dual_Channel_kit/index.html


Thanks

[jugeen]

Do u try Cellshock with black heatspreader like this :

http://www.memoryc.com/products/description/2GB_CellShock_DDR3_PC3_14400_1800MHz-8_7_6_21-Dual_Channel_kit/index.html


Thanks

he has blue. we have black.

[indiecom]

he has blue. we have black.

Ss, please.....thanks

[Fr3ak]

I still have trouble to get 32m stable at above 900MHz, but I blame the board for that. It seems like the NB is stressed a lot with the memory running that high. I can run 32m stable with memory at 750MHz and a FSB of 525MHz with 1.4VMCH. To boot at 450MHz with RAM running DDR3 1800 I already need about 1.6V on the MCH to even boot into windows.

I didn't have much time so far, so I tried a couple of different settings with 1M to get a slight idea what is possible and what is not.
I will do more testing with 32m over the weekend, here is what I have so far:

Edit: All tests were done with 2,24 V Vdimm. I know I am a jerk :) Will try with less voltage tonight. I have a strong feeling it is the board limiting me and not the RAM, so the settings below will most likely run with less Vdimm too.

[hipro5]

Ha, ha, ha.....I like the LAST one with the Cas 5 - 4 - 4 you Freak.....:D :up:

[v_parrello]

I still have trouble to get 32m stable at above 900MHz, but I blame the board for that. It seems like the NB is stressed a lot with the memory running that high. I can run 32m stable with memory at 750MHz and a FSB of 525MHz with 1.4VMCH. To boot at 450MHz with RAM running DDR3 1800 I already need about 1.6V on the MCH to even boot into windows.

I didn't have much time so far, so I tried a couple of different settings with 1M to get a slight idea what is possible and what is not.
I will do more testing with 32m over the weekend, here is what I have so far:

Edit: All tests were done with 2,24 V Vdimm. I know I am a jerk :) Will try with less voltage tonight. I have a strong feeling it is the board limiting me and not the RAM, so the settings below will most likely run with less Vdimm too.
Fr3ak,
which motherboard are you using?
If you want to go above 900 you have to give volts to nortbridge, with these DDR3 RAM the memory controller integrated into northbridge is very stressed.

[Fr3ak]

Ha, ha, ha.....I like the LAST one with the Cas 5 - 4 - 4 you Freak.....:D :up:

hehe :) 5-4-3 didn't work. 5-4-4 was as low as I could go with 750 MHz.

Maybe better timings work with less clockspeed.

Fr3ak,
which motherboard are you using?
If you want to go above 900 you have to give volts to nortbridge, with these DDR3 RAM the memory controller integrated into northbridge is very stressed.

I am using a Asus Blitz Extreme. I already figured out that more VMCH helps a lot. Unfortunately I am really busy right now and only have a few hours in the evening to bench. But by that time I am quite exhausted and not really in the mood to use a computer any longer, which is why my progress is rather slow. I am planning to bench a lot more at the weekend :) The memory surely is nice, but the mainboard is giving me a hard time :) I don't think Asus engineers didn't think of running 900MHz on the memory when they designed the board. Same for Intel and the chipset :)

[LardArse]

u workaholic!! :P u need to fix your bed first!
But you're doing justice to the good stuff from saaya da man :)

[v_parrello]

hehe :) 5-4-3 didn't work. 5-4-4 was as low as I could go with 750 MHz.

Maybe better timings work with less clockspeed.



I am using a Asus Blitz Extreme. I already figured out that more VMCH helps a lot. Unfortunately I am really busy right now and only have a few hours in the evening to bench. But by that time I am quite exhausted and not really in the mood to use a computer any longer, which is why my progress is rather slow. I am planning to bench a lot more at the weekend :) The memory surely is nice, but the mainboard is giving me a hard time :) I don't think Asus engineers didn't think of running 900MHz on the memory when they designed the board. Same for Intel and the chipset :)

Officially (as stated in all offcial DDR3 motherboard specs) all DDR3 P35 motherboards support RAM with DDR3-1333 as maximum frequency (because default FSB speed is 333 MHz and with maximum multiplier FSB:RAM=1:2 there is DDR3-1333 frequency for memory).

But, with overclock, if you want to go above DDR3-1800-1900 with DDR3 RAM you need a CPU with very high FSB wall (500-520 MHz) and you have to try to raise the North Bridge Voltage, PLL voltage, and FSB Termination Voltage (obviously also vcore and vmem).

I am not so sure about the facts that Asus and Intel engineers didn't think of running 900MHz on the memory when they designed the boards, infact there are motherboards that are able to run memory above DDR3-2000 MHz (like P5K3 and Blitz Extreme). Obviously we are talking about benching sessions and not for daily use (but for daily use you could run for sure at DDR3-1600 or with some overvoltage at DDR3-1800).

I agree about the fact that with these memories it could be better to have a chipset with default FSB of 400 MHz and similar processors able to run at 400 MHz as default.

[saaya]

I still have trouble to get 32m stable at above 900MHz, but I blame the board for that. It seems like the NB is stressed a lot with the memory running that high. I can run 32m stable with memory at 750MHz and a FSB of 525MHz with 1.4VMCH. To boot at 450MHz with RAM running DDR3 1800 I already need about 1.6V on the MCH to even boot into windows.

I didn't have much time so far, so I tried a couple of different settings with 1M to get a slight idea what is possible and what is not.
I will do more testing with 32m over the weekend, here is what I have so far:

Edit: All tests were done with 2,24 V Vdimm. I know I am a jerk :) Will try with less voltage tonight. I have a strong feeling it is the board limiting me and not the RAM, so the settings below will most likely run with less Vdimm too.

the 544 is great :D
looks like nobody ran 544 with ddr3 before, so this is the WR, highest 544 ;)
congratulations!!! :toast: :D

its quite high too, i didnt even think 555 could get that high we need to bump up trcd to 6 as well to get decent speeds with cas6. well at least that was my experience. victors 555 really caught me by surprise and this is even better :D

I don't think Asus engineers didn't think of running 900MHz on the memory when they designed the board. Same for Intel and the chipset :)
thats probabaly true :lol:
if it helps, i checked on one of the p5k3s and 1800+ boots with 1.4v chipset voltage, havent tested for stability but i could bench in windows.

so it looks like your board is a dud...
is it an ES board? maybe an early sample with an early p35 nb?
Several early ddr3 boards couldnt get above 1500 memspeed and im pretty sure its the chipset. P35 first launched for ddr2 boards, so maybe gigabyte and others used those early chips for their ddr3 boards as well and then noticed that those early nbs integrated ddr3 mem controller sucks.
asus seems to have used the latest p35 nbs even for their early ddr3 boards, maybe they noticed this or they always keep their warehouses tidy and empty and hence didnt have older stock.

anyways, i hope intel is going to bin p35 as well and not just x38.
x38 is just too expensive and from what ive seen and heard so far not worth it compared to p35.

u workaholic!! :P u need to fix your bed first!

Hell yeah, no wonder you are tired all the time when you sleep in that bed :lol:
For those who dont know, i kinda broke fr3aks bed ^^
Now gorillakos, before you even start to b!tch at fr3ak, i tried to seduce him but he wouldnt give in, he loves you too much... :D

But you're doing justice to the good stuff from saaya da man :):bows:
Just too bad the board sucks... hope you get a better board soon olli!

if you want to go above DDR3-1800-1900 with DDR3 RAM you need a CPU with very high FSB wall (500-520 MHz) and you have to try to raise the North Bridge Voltage, PLL voltage, and FSB Termination Voltage (obviously also vcore and vmem).
im not so sure about the pll voltage, does it really help you?
fsb wise or memory wise or cpu wise?
I bumped it up everytime i maxed out something but it never helped, so i left it at auto and i can reach 530fsb and 2ghz+ memspeed.
I guess it has to do with maxing out the cpu speed?
Thats the only thing i havent done yet :D

I am not so sure about the facts that Asus and Intel engineers didn't think of running 900MHz on the memory when they designed the boards, infact there are motherboards that are able to run memory above DDR3-2000 MHz (like P5K3 and Blitz Extreme). Obviously we are talking about benching sessions and not for daily use (but for daily use you could run for sure at DDR3-1600 or with some overvoltage at DDR3-1800).well the blitz is gen2 of p35 ddr3, so already gen3 of p35. and i think the secret of the asus p5k3 boards clocking better than other ddr3 boards when they just came out was that they used the latest nbs and others used older stock.

Wasnt the first people who hit 2ghz with ddr3 asus engineers?
So yeah, they did know about it, but i think that was only weeks before the launch... i dont know but i think micron ddr3 arrived pretty late so asus probabaly didnt use it to test their boards. afaik there is no other ddr3 that can reach 2ghz, not even close, so i dont think asus knew about this and planned those high speeds when designing the p5k... but in the end only asus and micron will know :D

I agree about the fact that with these memories it could be better to have a chipset with default FSB of 400 MHz and similar processors able to run at 400 MHz as default.
yeah! not only this, but we need way more dividers or an un link mode like on 680! I really hope intel is working on this or nvidia will get a ddr3 chipset out soon... wait, i take that back, i hope nvidia releases a ddr3 chipset AS SOON AS THEY MANAGED TO GET RID OF ALL THE BUGS IN IT :P

And wouldnt it be cool if intel would release an Xtreme Edition cpu with a default fsb of 666?
Now THAT would rock! no more looking for high fsb capable cpus.
im sure there are cpus that can handle 666fsb and im sure intel could bin them...

sorry for the long post ^^

[GoriLLakoS]

Thanks god Fr3ak is loyal...:D

At first i thought that the distance is big...and that's gonna destroy everything...

But i think it is more serious than i first thought...

xaaxxa

[saaya]

you think its serious??
so you dont know about the baby yet? :eek:
Fr3ak you have to tell him!
He has a right to see his son!

[hipro5]

you think its serious??
so you dont know about the baby yet? :eek:
Fr3ak you have to tell him!
He has a right to see his son!


HA, HA, HA.........You flaken Freak.......!.......:ROTF: :rofl:

[v_parrello]

....
im not so sure about the pll voltage, does it really help you?
fsb wise or memory wise or cpu wise?
I bumped it up everytime i maxed out something but it never helped, so i left it at auto and i can reach 530fsb and 2ghz+ memspeed.
I guess it has to do with maxing out the cpu speed?
Thats the only thing i havent done yet :D
.....


PLL voltage is referred to VCCPLL or PLL VCC with nominal value 1.50v+/-5%, and it is input voltage applied to CPU across pin 23 (as intel datasheets stated for Conroe). VCCPLL provides isolated power for internal processor FSB PLLs.

A phase-lock, or phase-locked, loop (PLL) is an electronic control system that generates a signal that is locked to the phase of an input or "reference" signal.

This means internal processor FSB PLLs are circuit responsibles to lock the phase of external FSB (the FSB of the external bus between northbridge and cpu) with internal processor bus.
In very simple words these circuits are responsible to sincronyze the external FSB with internal processor bus where flow all signals outuput from FSB to internal processor bus (command and data) and viceversa.

Now some processors have better internal FSB PLLs than another one, and this mean that that first rpocessor has higher FSB wall. If you want to help to raise CPU FSB wall you have to raise pll voltage because in this way internal processor FSB PLLs become more stable.

[Fr3ak]

u workaholic!! :P u need to fix your bed first!
But you're doing justice to the good stuff from saaya da man :)

Hehe, I know about that bed, it's getting worse every day :P
Thank you, but without you I wouldn't have been able to do the memory justice. Thanks a lot for giving that board to me. IF you ever need it again, just tell me and I will send it back to you.

the 544 is great :D
looks like nobody ran 544 with ddr3 before, so this is the WR, highest 544 ;)
congratulations!!! :toast: :D

its quite high too, i didnt even think 555 could get that high we need to bump up trcd to 6 as well to get decent speeds with cas6. well at least that was my experience. victors 555 really caught me by surprise and this is even better :D

hehe thanks, it was a really quick attempt to see if the sticks run at those timings at all. I am pretty sure that there is still room for improvement.


so it looks like your board is a dud...
is it an ES board? maybe an early sample with an early p35 nb?
Several early ddr3 boards couldnt get above 1500 memspeed and im pretty sure its the chipset. P35 first launched for ddr2 boards, so maybe gigabyte and others used those early chips for their ddr3 boards as well and then noticed that those early nbs integrated ddr3 mem controller sucks.
asus seems to have used the latest p35 nbs even for their early ddr3 boards, maybe they noticed this or they always keep their warehouses tidy and empty and hence didnt have older stock.


I have no idea if the board is a ES or not, but it looks retail to me. Maybe Shamino can bring some light into this discussion about the origin of this board.


Hell yeah, no wonder you are tired all the time when you sleep in that bed :lol:
For those who dont know, i kinda broke fr3aks bed ^^
Now gorillakos, before you even start to b!tch at fr3ak, i tried to seduce him but he wouldnt give in, he loves you too much... :D

:bows:
Just too bad the board sucks... hope you get a better board soon olli!


hehe I know the bad is a mess. I will try to fix it soon, that is for sure. But I can still sleep surprisingly well :)


im not so sure about the pll voltage, does it really help you?
fsb wise or memory wise or cpu wise?
I bumped it up everytime i maxed out something but it never helped, so i left it at auto and i can reach 530fsb and 2ghz+ memspeed.
I guess it has to do with maxing out the cpu speed?
Thats the only thing i havent done yet :D

well the blitz is gen2 of p35 ddr3, so already gen3 of p35. and i think the secret of the asus p5k3 boards clocking better than other ddr3 boards when they just came out was that they used the latest nbs and others used older stock.

Wasnt the first people who hit 2ghz with ddr3 asus engineers?
So yeah, they did know about it, but i think that was only weeks before the launch... i dont know but i think micron ddr3 arrived pretty late so asus probabaly didnt use it to test their boards. afaik there is no other ddr3 that can reach 2ghz, not even close, so i dont think asus knew about this and planned those high speeds when designing the p5k... but in the end only asus and micron will know :D


yeah! not only this, but we need way more dividers or an un link mode like on 680! I really hope intel is working on this or nvidia will get a ddr3 chipset out soon... wait, i take that back, i hope nvidia releases a ddr3 chipset AS SOON AS THEY MANAGED TO GET RID OF ALL THE BUGS IN IT :P

And wouldnt it be cool if intel would release an Xtreme Edition cpu with a default fsb of 666?
Now THAT would rock! no more looking for high fsb capable cpus.
im sure there are cpus that can handle 666fsb and im sure intel could bin them...

sorry for the long post ^^

PLL voltage usually helps you to clock the CPU's FSB better when the CPU hits a wall. With most CPUs I have tested I need about +0.1V above 450 MHz FSB. FSB-Termination voltage usually helps to clock the board higher, FSB-wise. I didnt have to use more than +0.15V to hit 500 MHz FSB so far.

The 666FSB CPU sound good, but I dont think its gonna come true any time soon. 400MHz will first hit the server market, then desktop CPUs. 666 is still a very long way to go. And even Oskar had a hard time hitting 600 MHz FSB on a retail board, so to think 666 MHz FSB on a retail board will come true son, is basically impossible if you ask me :)

But 666MHz FSB and 6-6-6 timings looks "damn" good :devil:

Thanks god Fr3ak is loyal...:D

At first i thought that the distance is big...and that's gonna destroy everything...

But i think it is more serious than i first thought...

xaaxxa

How could I ever betray you? ^_^

HA, HA, HA.........You flaken Freak.......!.......:ROTF: :rofl:

hehe, just make fun on my expense lol :)

PLL voltage is referred to VCCPLL or PLL VCC with nominal value 1.50v+/-5%, and it is input voltage applied to CPU across pin 23 (as intel datasheets stated for Conroe). VCCPLL provides isolated power for internal processor FSB PLLs.

A phase-lock, or phase-locked, loop (PLL) is an electronic control system that generates a signal that is locked to the phase of an input or "reference" signal.

This means internal processor FSB PLLs are circuit responsibles to lock the phase of external FSB (the FSB of the external bus between northbridge and cpu) with internal processor bus.
In very simple words these circuits are responsible to sincronyze the external FSB with internal processor bus where flow all signals outuput from FSB to internal processor bus (command and data) and viceversa.

Now some processors have better internal FSB PLLs than another one, and this mean that that first rpocessor has higher FSB wall. If you want to help to raise CPU FSB wall you have to raise pll voltage because in this way internal processor FSB PLLs become more stable.

Very nice explanation! :up: =)

Edit: I had half an hour to play with the sticks :) 1 GHz is a no go even with 10-10-10-30. I managed to improve the low latency clock a bit, but not by much. And it seems like I really need 2.24V to boot with cl5 and 750MHz.
I managed to get this 1m stable, but the screenshot is corrupt :/
http://valid.x86-secret.com/show_oc.php?id=239364

Some real stability tests are planned for the weekend :)

[saaya]

vincenzo, so does increasing the cpu pll voltage help to get a high fsb stable or does it help to boot and bench at a higher fsb?
It didnt help me to boot and bench at a higher fsb at all, so i left it at auto.
FSB voltage bumped to 1.5v did help though and is needed to reach 530+ with the 6600 in this system.

what cpu are you using?


hehe I know the bad is a mess. I will try to fix it soon, that is for sure. But I can still sleep surprisingly well :)
dreaming of gorillakos strong arms holding you?
what does he have that i dont have? :,(

:lol:

PLL voltage usually helps you to clock the CPU's FSB better when the CPU hits a wall. With most CPUs I have tested I need about +0.1V above 450 MHz FSB. FSB-Termination voltage usually helps to clock the board higher, FSB-wise. I didnt have to use more than +0.15V to hit 500 MHz FSB so far.so fsb vtt or vfsb as its sometimes called is more of a help to get the chipset stable at higher fsbs and to be able to boot at higher fsbs, correct?

well first i could only get around 450fsb with this 6600 but bumping the vcore to 1.5v solved that and i got to 480 iirc, bumping vfsb gave me a headroom to reach 530, above that things got unstable. i tried raising cpu pll voltage but this didnt help me to get 535 stable either and i didnt want to try more vcore with the crfappy hs i was using back then.

The 666FSB CPU sound good, but I dont think its gonna come true any time soon. 400MHz will first hit the server market, then desktop CPUs. 666 is still a very long way to go. And even Oskar had a hard time hitting 600 MHz FSB on a retail board, so to think 666 MHz FSB on a retail board will come true son, is basically impossible if you ask me :)but it is possible, and yes you would basically need a custom motherboard but hey, what else is quadfx and intels v8?
they could pick boards that can hit 600+fsb and cpus that can hit 600+fsb.
IF k10 beats 45nm (very unlikely unfortunately) then intel might actually do this... but yeah, its more of a nice thing to dream about :D

But 666MHz FSB and 6-6-6 timings looks "damn" good :devil: :slobber:

Very nice explanation! :up: =)
jupp, thanks vincenzo! :toast:

Edit: I had half an hour to play with the sticks :) 1 GHz is a no go even with 10-10-10-30. I managed to improve the low latency clock a bit, but not by much. And it seems like I really need 2.24V to boot with cl5 and 750MHz.
I managed to get this 1m stable, but the screenshot is corrupt :/
http://valid.x86-secret.com/show_oc.php?id=239364

Some real stability tests are planned for the weekend :)
have fun then :D

[boblemagnifique]

Few Test Superpi (Comparative in preparation) :)

I test soon with my P5K3 deluxe too :)

32mo = OS 2003 32bits optimised by learn :clap:

http://boblemagnifique36.free.fr/Overclocking/Memoires/40gp1mo.jpg

http://boblemagnifique36.free.fr/Overclocking/Memoires/40gp32mo.jpg

[v_parrello]

vincenzo, so does increasing the cpu pll voltage help to get a high fsb stable or does it help to boot and bench at a higher fsb?
It didnt help me to boot and bench at a higher fsb at all, so i left it at auto.
FSB voltage bumped to 1.5v did help though and is needed to reach 530+ with the 6600 in this system.

what cpu are you using?

.....

Sascha,
you can not make a miracle but increasing cpu pll voltage could help you to to get a high fsb.

I used a E6420 that is able to do 555 FB with air cooling (it is very lucky CPU). High cpu pll voltage help me to raise of 30-50 MHz the FSB.

[Kamel]

Saaya...e6600 seems not to be the perfect cpu for p35, a 6600 i've tested, 560 and over on 965... stops @530 on p35!

don't ask me why..maybe new bioses are not perfect for the older cpu steps..

Kam

[boblemagnifique]

Saaya...e6600 seems not to be the perfect cpu for p35, a 6600 i've tested, 560 and over on 965... stops @530 on p35!

don't ask me why..maybe new bioses are not perfect for the older cpu steps..

Kam

There is a problem with P35 and the DDR3, I have cpus which block has 570/575mhz on my extreme Blitz or the P5K3 deluxe (then which makes 590mhz on my P5K or the DFI P35 in DDR2)

It's stange problem :cool:

[eva2000]

yeah it's strange

my E6600 L709A958 on Blitz Extreme 485FSB max on Blitz Formula 530FSB max both on water heh.. nearly every cpu i put in my Blitz Extreme maxes 485FSB even ones that do 500FSB on Blitz Formula/P5K Deluxe heh

[Kamel]

6420 over 540 easily.... 6850 540 easily... and i'm waiting for an E6700 with 580/590 liquid cooled and 610 under LN2...

Kam

i think could be a problem with bios cpu microcode programming... or compatibility..

[boblemagnifique]

Since I flash the bios P5K3 and Blitz I am with the max 575/580mhz whereas I arrived has 590mhz in watercooling with my E6600 before :(

A small Everest towards 4ghz :

http://boblemagnifique36.free.fr/Overclocking/Memoires/bpddr3.JPG

[saaya]

Sascha,
you can not make a miracle but increasing cpu pll voltage could help you to to get a high fsb.

I used a E6420 that is able to do 555 FB with air cooling (it is very lucky CPU). High cpu pll voltage help me to raise of 30-50 MHz the FSB.
hmmm ill try again to bump pll voltage and see if it helps, thanks man :)

Saaya...e6600 seems not to be the perfect cpu for p35, a 6600 i've tested, 560 and over on 965... stops @530 on p35!

don't ask me why..maybe new bioses are not perfect for the older cpu steps..

Kam
There is a problem with P35 and the DDR3, I have cpus which block has 570/575mhz on my extreme Blitz or the P5K3 deluxe (then which makes 590mhz on my P5K or the DFI P35 in DDR2)

It's stange problem :cool:
yeah it's strange

my E6600 L709A958 on Blitz Extreme 485FSB max on Blitz Formula 530FSB max both on water heh.. nearly every cpu i put in my Blitz Extreme maxes 485FSB even ones that do 500FSB on Blitz Formula/P5K Deluxe heh

yeah i noticed the same, but i didnt think it was p35 or p35 ddr3 related... every board i put a cpu in has a different max fsb. even good boards that get a higher fsb out of cpus than other boards sometimes dont like a cpu and i cant reach the same fsb as on other "worse" fsb clocking boards.

we are 4 people now experiencing this...
interesting to see that there seems to be some corelation between this and p35 and maybe even p35 ddr3 compared to p35 ddr2.
thx for sharing your findings :toast:

6420 over 540 easily.... 6850 540 easily... and i'm waiting for an E6700 with 580/590 liquid cooled and 610 under LN2...

Kam

i think could be a problem with bios cpu microcode programming... or compatibility..
well lets hope its a bios thing and has nothing to do with board design or the ddr3 mem controller limiting the fsb somehow.

[Walkerak]

What a nice try~~~~~~:up:

[saaya]

What a nice try~~~~~~:up:
what do you mean? :confused:

[eva2000]

yeah i noticed the same, but i didnt think it was p35 or p35 ddr3 related... every board i put a cpu in has a different max fsb. even good boards that get a higher fsb out of cpus than other boards sometimes dont like a cpu and i cant reach the same fsb as on other "worse" fsb clocking boards.

we are 4 people now experiencing this...
interesting to see that there seems to be some corelation between this and p35 and maybe even p35 ddr3 compared to p35 ddr2.
thx for sharing your findings :toast:
yeah i asked for a 2nd Blitz Extreme to be shipped so i could compare to see if it's my board that is FSB limited.. fedex docs all done but i think they might be waiting to send me X38 too :D

Crosses fingers that X38 can give cpu's a nice max FSB boost over P35.. i hope :D

[Fr3ak]

Your last sentence wasn't true for the testing I have done today :( No FSB boost...

[Revv23]

someone get hipro two set of these and some LN2 stat!!

[x3ro]

Nice clocking m8, Im getting some ddr3 myself and I was wondering what's the best price/performance ddr3 mobo?

Regards

[saaya]

someone get hipro two set of these and some LN2 stat!!
he already has a set and im sure we dont need to worry about his ln2 supply :D

hmmm does x38 at least give a perf boost at the same fsb and mem clock then?

[eva2000]

Your last sentence wasn't true for the testing I have done today :( No FSB boost...
doh :(

[hipro5]

he already has a set and im sure we dont need to worry about his ln2 supply :D

.....where - where?.... I don't have a set of them yet.... :(

[VictorWang]

.....where - where?.... I don't have a set of them yet.... :(

'coz u have toooooooo much toys , haha :ROTF:

[saaya]

.....where - where?.... I don't have a set of them yet.... :(
you still didnt get it???
ill check

EDIT: We sent it out Monday last week, so... 9 days/ 7 work days ago.
I know the mail in greece is very slow, but could it be this slow?
I hope it didnt get lost...

[Johnny Bravo]

I hope it gets redirected to me!

[raju]

I hope it gets redirected to me!

after I've burned it in....:)

regards
Raja

[Fr3ak]

I finally tested the Cellchocks on a decent board. My Blitz Extreme is somewhat not really working with high memory speed for whatever reason. I used a P5K3 Deluxe this time and got way better results :)

I will post pics later, don't have the time right now. I already posted in a German forum, but the screenshots should speak for themselves. The voltage used is written above the pictures.
Link: http://extreme.pcgameshardware.de/showthread.php?t=244

[eva2000]

I finally tested the Cellchocks on a decent board. My Blitz Extreme is somewhat not really working with high memory speed for whatever reason. I used a P5K3 Deluxe this time and got way better results :)

I will post pics later, don't have the time right now. I already posted in a German forum, but the screenshots should speak for themselves. The voltage used is written above the pictures.
Link: http://extreme.pcgameshardware.de/showthread.php?t=244
Interesting might have to unpack my P5K3 Deluxe and swap out my Blitz Extreme then hehe

How much of a difference on P5K3 Deluxe vs Blitz Extreme ?

[stone_cold_Jimi]

I finally tested the Cellchocks on a decent board. My Blitz Extreme is somewhat not really working with high memory speed for whatever reason. I used a P5K3 Deluxe this time and got way better results :)

Nice! Are they the blue or the black? With the P5K3 my blacks are good for 800 7-6-5-18 1T at 1.65v (black slots). Could you see what freq you get with same voltage and timings? It will be a few days more before I can do more extensive tests, so I'm curious :)

Edit: hmmm, just ran a 32m with 860 7-6-5-18 1T at 1.75v. These are looking teh sexy.

[Fr3ak]

George, the P5K3 made the hell of a change for me :)
With the Extreme I needed a vdimm of 2.24V and a MCH voltage of 1.7V to be 1M stable at 900 MHz 8-7-6-21-2T. With the P5K3 I needed 1.4V VMCH and 1.7V to be dual 32M stable at 900 MHz and the same timings.

I also have the black sticks (well, I had ;)). I used the black slots, cos on P35 Asus boards (the DDR2 ones too) I was able to clock the memory higher using the black slots.

[eva2000]

Interesting indeed wonder if extreme has tighter MCH latencies than P5K3 hence the difference in MCH voltage needed ?

[saaya]

i think that particular blitz was crap :D
it was probably an older chipset and hence didnt clock very well...
if you ever get the time to remove that heatpipe madness and spot the nb revision let me know olli :D

btw, did you use the same cpu on both boards?

[eva2000]

Yeah same cpu on both.. blitz extreme chipset numbers

http://i4memory.com/reviewimages/motherboards/asus/Blitz/extreme/nbcooling/photos/BlitzExtreme_Cooling_17.JPG

[saaya]

sl... a...9r? do i need glasses or does that pic need a zoom? ^^

[eva2000]

Le82p35
Sla9r
L715a337

:)

[BenchZowner]

LE82P35
SLA9R
L715A337 ( if my eyes are capturing the pic correctly :D ) [ LOL ] { think I'm gonna get those boards voltmodded tomorrow! :D }

Edit: Thanks George, you confirmed my eyesight, so volt mods here I come :p:

[saaya]

thanks george, and thanks bill, looks like your eye surgery was well worth it :lol:

im leaving to yet another ln2 party for the weekend ^^
im taking a p5k3 with me and will check the stepping this weekend.
i compared 3 boards today and yesterday, interesting...

theres a difference of 50mhz between the best and the worst in regards to max clockspeed. same cpu, same bios, same settings.
theres something else that confused me.

remember the old board i always mentioned that only did 1825 max dc stable?
its one of the first retail p5k3... well now it does 1900+...
i switched the cpu and wham, suddenly it runs much better.

the weird thing is, before it was running with a 6600 that could run a high fsb.
now its running a 6750... im sure i checked the fsb stability of the 6600 by using the same fsb or higher with a different mem divider. and it was stable... so i assumed that fsb was stable... but for some reason the 6600 was not stable at the same fsb when i used the 1:2 mem divider.

and the difference between the dividers must be huge, because with the other divider i could run an fsb of 500 easily and it was stable.
with the 1:2 divider i was limited to 1825, which is 455fsb.

could this be correct?
a max fsb difference between memory dividers?
and not a small one but a whooping 50mhz?
im puzzled... ill see if i can reproduce this with the 6600.
or could it be some registers changing once you set a high mem clock?
so once you set a memspeed of 1800+ the chipset/board relaxes some timings or changes something automatically to make a higher fsb stable?
and for some reason this only helps with 1333fsb cpus and not at all or not so well with 1066cpus?
reminds me of the l12 bridge thing from socket a :D

people could push their nf2 boards to higher fsbs with the bridge mods... its just weird this time that the memory controller is limiting that much.
but in general thats very characteristically for p35 ddr3 in my experience so far, the chipset is very limited by the ddr3 mem controller.
we can reach a notably lower max fsb with p35 ddr3 compared to p35 ddr2, which i think its the ddr3 mem controller limiting.

maybe we should compare the registers with p35 ddr3 and p35 ddr3, maybe on a hybrid board ideally so only the registers important for ddr2 and ddr3 operations change. that way we could maybe find a trick to get the same high fsbs on p35 with ddr3 as with ddr2...

[eva2000]

LE82P35
SLA9R
L715A337 ( if my eyes are capturing the pic correctly :D ) [ LOL ] { think I'm gonna get those boards voltmodded tomorrow! :D }

Edit: Thanks George, you confirmed my eyesight, so volt mods here I come :p: LOL... good to hear your eye sight is okay :D

thanks george, and thanks bill, looks like the surgery was well worth it :lol:

im leaving to yet another ln2 party for the weekend ^^
im taking a p5k3 with me and will check the stepping this weekend.
i compared 3 boards today and yesterday, interesting...

theres a difference of 50mhz between the best and the worst in regards to max clockspeed. same cpu, same bios, same settings.
theres something else that confused me.

remember the old board i always mentioned that only did 1825 max dc stable?
its one of the first retail p5k3... well now it does 1900+...
i switched the cpu and wham, suddenly it runs much better.

the weird thing is, before it was running with a 6600 that could run a high fsb.
now its running a 6750... im sure i checked the fsb stability of the 6600 by using the same fsb or higher with a different mem divider. and it was stable... so i assumed that fsb was stable... but for some reason the 6600 was not stable at the same fsb when i used the 1:2 mem divider.

and the difference between the dividers must be huge, because with the other divider i could run an fsb of 500 easily and it was stable.
with the 1:2 divider i was limited to 1825, which is 455fsb.

could this be correct?
a max fsb difference between memory dividers?
and not a small one but a whooping 50mhz?
im puzzled... ill see if i can reproduce this with the 6600.

Yeah i suspected the divider might play apart in it on an ealier blitz extreme bios i couldn't get stable using 3:5 divider at known stable cpu clock speed... swapped to 1:2 divider and was stable.

Guess also depends on the NB strap set as some straps offer the same divider.

[BenchZowner]

theres a difference of 50mhz between the best and the worst in regards to max clockspeed. same cpu, same bios, same settings.
theres something else that confused me.

remember the old board i always mentioned that only did 1825 max dc stable?
its one of the first retail p5k3... well now it does 1900+...
i switched the cpu and wham, suddenly it runs much better.

the weird thing is, before it was running with a 6600 that could run a high fsb.
now its running a 6750... im sure i checked the fsb stability of the 6600 by using the same fsb or higher with a different mem divider. and it was stable... so i assumed that fsb was stable... but for some reason the 6600 was not stable at the same fsb when i used the 1:2 mem divider.

and the difference between the dividers must be huge, because with the other divider i could run an fsb of 500 easily and it was stable.
with the 1:2 divider i was limited to 1825, which is 455fsb.

could this be correct?
a max fsb difference between memory dividers?
and not a small one but a whooping 50mhz?
im puzzled... ill see if i can reproduce this with the 6600.

Well, a difference between dividers ( max clock speed dependency on divider ) is there ( for sure ).
But I don't remember spotting ( well, didn't tried hard :p: almost at all ) any difference in mem clocking with different CPUs.
But I think it's time to try some CPUs with the same board + settings to see if I can reproduce your findings ;) [ the more you test, the more things you need to test :D ]

[Kamel]

Saaya, i told u p35 was not so good with E6600, better with new stepping CPU such as 6420, 6750, 6850....

So now it's confirmed... i don't know why but E6600 is not loved by P35.. maybe some bios matter in recognizing the cpu step...who knows?

Kam

[Fr3ak]

I used a E6600ES on the Blitz and a E6750ES on the P5K3. I wonder if that might have to do with it? I tested both CPUs for max stable FSB. The E6600 does 525 MHz and the E6750 510 MHz.

[saaya]

Yeah i suspected the divider might play apart in it on an ealier blitz extreme bios i couldn't get stable using 3:5 divider at known stable cpu clock speed... swapped to 1:2 divider and was stable.

Guess also depends on the NB strap set as some straps offer the same divider.
I might have overlooked the strap setting and that could have caused the difference... :doh:

Well, a difference between dividers ( max clock speed dependency on divider ) is there ( for sure ).
But I don't remember spotting ( well, didn't tried hard :p: almost at all ) any difference in mem clocking with different CPUs.
But I think it's time to try some CPUs with the same board + settings to see if I can reproduce your findings ;) [ the more you test, the more things you need to test :D ] i think it was the strap, not 100% sure tho...

Saaya, i told u p35 was not so good with E6600, better with new stepping CPU such as 6420, 6750, 6850....

So now it's confirmed... i don't know why but E6600 is not loved by P35.. maybe some bios matter in recognizing the cpu step...who knows?
Kamyou had similar problems right?
did you try different straps for the 6600?

I used a E6600ES on the Blitz and a E6750ES on the P5K3. I wonder if that might have to do with it? I tested both CPUs for max stable FSB. The E6600 does 525 MHz and the E6750 510 MHz.
i hope i didnt kill that blitz man, im really sorry if i did :(
i keep thinking about it and maybe one of the smd caps on the nb package ripped off when the nb heatsink came off? other than that... no clue...

when i first booted it the fans were spinning fast, then rebooted and fans were slow, so the board still seems to be working somewhat as it apparently went to defaults after the initial boot didnt work.

does anybody know what a debug card error code of D4 "uncompressed data fault" or something like that means? Are there any tricks on how to reflash the bios? i booted with the mobo cd in the dvd drive and hoped it would load the bios but it didnt even start to read the sata dvd drive...

Btw, the stepping of the NB on that blitz board is

LE82P35
SLA9R
L717A231

same as your eva, right?

EDIT:
L715A337
vs
L717A231

15A vs 17A and 337 vs 231 does anybody know how to read intel chipset steppings? :D
I think the latter is the datecode since freaks board is older than yours eva, at least i think so...

[eva2000]

i think it's like cpus L715 and L717 = week 15 and 17, 2007 ???

I've swapped to P5K3 Deluxe and much better FSB/MEM clocker than my Blitz Extreme!

[Fr3ak]

So its no coincidence. A modded P5K3 would be better for benching I guess than the Extreme. Apart from the voltage options the Extreme doesnt offer any more features. The crosslink chip is, well, not that great too.

How much better is the P5K3 over your Extreme, George?

[hipro5]

It's ALL about the chipset P35 - they are like the CPUs - Best, better, medium, bad - ......My second Extreme does a bit worse than my first one as for getting OVER 1000MHz the rams.....:(

[eva2000]

For my E6600 L709A958 difference in max FSB

Blitz Extreme = 485FSB max
P5K3 Deluxe = 530FSB max and 535FSB bootable but hangs in memtest86.

[Fr3ak]

Same voltages being used?

Thats quite a big difference.... Almost like using a quadcore instead of a dual.

[saaya]

wow, thats a huge difference!
so eva, your board is older than freaks then... do you have it for a long time already? you must have a bad blitz too then...

hipro, did you check what nb stepping your p35 is?
eva, can you check on your p5k3?

its strange that there seem to be way more blitz boards with bad nbs than p5k3s... or maybe thats just our perception since most p5k3 users dont push their boards that hard while EVERY blitz owner pushes the board to the max? :D

i have several p5k3 boards here, some with made in china pcb and some with made in taiwan pcb. ill check if theres a trend of one being better than the other.

[boblemagnifique]

Le82p35
Sla9r
L715a337

:)

LE82P35
SLA9R
L716A390

For my Blitz Extreme :)

[saaya]

ehhh ca va BOB! :D :toast:
good to have you on this thread as well hehehe
thanks for the info

how well does your blitz clock?
did you test a p5k3 so far?
thanks a lot for the dump for roger btw :toast:

olli, what was the max of your blitz again?

eva2000:
LE82P35
SlA9R
L715A337
490 FSB max/1960 DDR max?

boblemagnifique:
LE82P35
SLA9R
L716A390

Fr3ak:
LE82P35
SLA9R
L717A231
435 FSB max/1750 DDR max?

[eva2000]

Same voltages being used?

Thats quite a big difference.... Almost like using a quadcore instead of a dual.Yeah same voltages as well as the optimal voltages to reach max FSB... the P5K3 Deluxe is clearly better for FSB

wow, thats a huge difference!
so eva, your board is older than freaks then... do you have it for a long time already? you must have a bad blitz too then...

hipro, did you check what nb stepping your p35 is?
eva, can you check on your p5k3?

its strange that there seem to be way more blitz boards with bad nbs than p5k3s... or maybe thats just our perception since most p5k3 users dont push their boards that hard while EVERY blitz owner pushes the board to the max? :D

i have several p5k3 boards here, some with made in china pcb and some with made in taiwan pcb. ill check if theres a trend of one being better than the other.
I've had the blitz formula/extreme sent together since July and p5k dx/p5k3 dx sent together since May so I would assume P5K3 DX might have an even older P35 core.. can't check yet as just set up the rig and all screwed into motherboard tray and all heh

All Asus boards are made in China

[boblemagnifique]

ehhh ca va BOB! :D :toast:
good to have you on this thread as well hehehe
thanks for the info

how well does your blitz clock?
did you test a p5k3 so far?
thanks a lot for the dump for roger btw :toast:

olli, what was the max of your blitz again?

eva2000:
LE82P35
SlA9R
L715A337
490 FSB max/1960 DDR max?

boblemagnifique:
LE82P35
SLA9R
L716A390

Fr3ak:
LE82P35
SLA9R
L717A231
435 FSB max/1750 DDR max?

Yesss ca va et toi ? :D

For my Blitz , the First test (591x6 on water with my E6660) but now , the max is 575mhz (Flash bios :( )

572mhz stable : http://boblemagnifique36.free.fr/Overclocking/Memoires/40gp32mo.jpg

I test the Corsair 1800 c7 for the moment , i can benchs spi 1mo at 1890 c7/8/9 DDR3 max (and 1900/1910 with Cellshock 1800 c7/8/9)

I test with the Supertalent 1866 on single , the max screen is 1070mhz cas 9
(max benchs spi 1mo 1960/1980 on dual cas 8/9)

[saaya]

Yeah same voltages as well as the optimal voltages to reach max FSB... the P5K3 Deluxe is clearly better for FSBi dont think the p5k3 is generally better than the blitz, i think its just some p35 nbs sucking...

I've had the blitz formula/extreme sent together since July and p5k dx/p5k3 dx sent together since May so I would assume P5K3 DX might have an even older P35 core.. can't check yet as just set up the rig and all screwed into motherboard tray and all hehhmmm well im hoping the p35 nbs are getting better, so the newer your p35nb/board the lower the chance to get a really crappy board seems to be.
the worst boards ive seen so far were very early p35 boards, and we have several p5k3 boards here which do clock different from one another, but as i said, the difference is around 50mhz, not 300mhz like some old boards.
good boards can reach 2000 stable, and the worst i saw so far can only do around 1700 stable.

All Asus boards are made in China
they are assembled in china, but the pcbs are made in china and taiwan.
im scurious if they already solder the caps and cooils and chipset onto the boards when they assembled the pcbs or not. so maybe assembled in china means they put on the heatsinks in china? who knows... it looks to me like they already put on the caps when they build the pcb, since the made in china pcbs have different caps and parts than the made in taiwan pcbs... so its 2 different boards.
what boards do you have eva freak and bob? made in china or made in taiwan?

http://img228.imageshack.us/img228/5529/chinags1.jpg
http://img182.imageshack.us/img182/9365/taiwanxw5.jpg

Yesss ca va et toi ? :D J'en la grippe :D

For my Blitz , the First test (591x6 on water with my E6660) but now , the max is 575mhz (Flash bios :( )

572mhz stable : http://boblemagnifique36.free.fr/Overclocking/Memoires/40gp32mo.jpg
572fsb on water, that should prove it, the striker can do high fsbs, theres probabaly just a lot of fluctuation between p35 nbs.

I test the Corsair 1800 c7 for the moment , i can benchs spi 1mo at 1890 c7/8/9 DDR3 max (and 1900/1910 with Cellshock 1800 c7/8/9)wait, there is no difference between cas7 8 and 9 max speed? thats very weird...

I test with the Supertalent 1866 on single , the max screen is 1070mhz cas 9
(max benchs spi 1mo 1960/1980 on dual cas 8/9)
single channel... ok... and dualchannel?

whats the max stable dc speed with your blitz?

[Pedro Rocha]

i dont think the p5k3 is generally better than the blitz, i think its just some p35 nbs sucking...



I think you are right :D

Here is my quick results with Blitz Extreme and Cellshocks 1800

Air Cooled (TuniqTower)

http://pedrorocha.planetaclix.pt/BlitzExtremeCellshockDDR2010_7-6-6-18-1T_1200Sandra.jpg

http://pedrorocha.planetaclix.pt/BlitzExtremeE6850_air_spi.jpg
:D

VMem 2.24v, VMch 1.75, Vll 1.8v


Mini-cascade cooled (-75ºC bios)

http://pedrorocha.planetaclix.pt/BlitzExtreme5355.jpg

http://pedrorocha.planetaclix.pt/BlitzExtreme5355EverestRead.JPG

VMem 2.2v, VMch 1.95v, Vpll 1.90


My Blitz Extreme Max FSB stable the E6850:
- Air - 540Mhz
- Mini-cascade - 595Mhz (just a bit less than Formula that did 605)


The Cellshocks DDR3 black sticks cost me a fortune, but they are real good .. in the next day I will test with 2 stock 2900XT in CF to check how much can improve my favorite game...

BTW : nice posts on this thread :up:

[stone_cold_Jimi]

At last! P5K3, BIOS set vdimm 1.9v, vmch 1.55v and look at that poor CPU lol - that's the best "stable" FSB I can get at 8x, tho I can get 475x6. Great pair of sticks though, fulifilling a bit of that early potential :D

http://valid.x86-secret.com/show_oc.php?id=260706

[Infa]

these are my results...
it can go 2000 though, but i will adjust it better, and post results later..
but its Sweeeet mems :)

vDIMM 2.08v.

Http://hem.bredband.net/infan/cellshock/spi1m1868.JPG
Http://hem.bredband.net/infan/cellshock/spi32m1868.JPG
Http://hem.bredband.net/infan/cellshock/everestread1868.JPG
Http://hem.bredband.net/infan/cellshock/everestcl1868.JPG

And one 1M result with ddr2000, but it wouldnt go for 32m yet, need to adjust more.. 2.28v is this..

Http://hem.bredband.net/infan/cellshock/ddr2000.jpg

[stone_cold_Jimi]

Nice Infa :toast:

I just ran a 32m at the same settings as my 1m, no probs. If I could get a higher FSB, I'd like to try 2000 - maybe my Q6600 will get there.

[Infa]

Thank you :) finally im in the game, been waiting a while for my stuff :)

yeh, you gotta have atleast 500fsb for the ddr2000 to work though...
best luck! hope it will work for you, now i will try to get my 32m ddr2000 to work..

regards.

[stone_cold_Jimi]

A prettier picture (still can't get more FSB with this chip, hit a wall) -

[Infa]

Hmm, i just CANT get em to work ddr2000, nomather what i do, really..
i dont want to pump more than 2.3 either, but they should work with that atleast, not even cl 8, sure i got 2.06v now and typing this, and cl 8, and thats working better than more voltage, but its dead before first irretation at super pi, wont work! tried every setting...hmm.. annoying.

regards.

[dinos22]

guys just buy X38 chipsets and you will be able to get that extra little bit and even tighten the timings somewhat :D

[Infa]

well i got it, and it wont work, now i will push the Vdimm a bit higher to see if thats the prob, will go over 2.3 a while, hope they not will die on my, they were expensive, BUT I WANNA KNOW DAMMIT :P i want a 32m ddr2000, then im happy, i dont want it 24/7 couse my setting at 930 7-6-5-18 and 4.1ghz ARE FAST!, i just want that magic ddr2000 :P, any tip on a sucessful settinf for ddr2000?

[Infa]

Hmm, what can be wrong? Im at 2.32v now, still same not exact on round, aswell as in 2.24, it cant be the voltage?? i did test with memtest for a few mins and not a single error, but super pi, just goes to bed?

something cant handle this, the board? some setting ive missed?

[stone_cold_Jimi]

guys just buy X38 chipsets and you will be able to get that extra little bit and even tighten the timings somewhat :D

idd. My P5K3 seems to be the limiting factor - 462 FSB with my Q6600 as it turns out, pretty much the same FSB as the E2140. Infa m8, don't burn 'em, they don't like much over 2.1v afaik.

I wonder if the 0704 bios has a bit more for this.... says it's tweaked for more types of RAM. If it had a ratio above 1:2, I would like :)

[Infa]

Noo, ive come to the conclusion that i think its my bios, its just the third out for the p5e3, it just got the Damper enabled, sooo, the errors thats happening, it just dont feels like its the rams faults in some way.......i do put it atleast bios stable CL 7-7-7-20 2000mhz 2.18, and its rock solid in memtest, noo error, but still almost allways, BSOD when loading win, or BSOD in win, or mostly NOT EXACT IN ROUND, but still error free in memtest? whats up whit that? isnt memtest suppose to be a testprogram, so if it goes error free, then the comp will be stable?

ahh well, now im at 1827mhz, 4.1ghz and 7-6-5-17, and Transaction at 3, strap at 266, Twister at strong, and only at 1.96v, but att fsb 500 and ddr2000 it just goes crap?

ahhh i hope for the next bios might be a sunshine :) , now ive been trying for 2days in a row all eveing, and it wont goo, so i'll give it up until next release.

im satisfied anyhow hehe....goin such low Cl below rated is fair enough at this speed... seems p5k series got more mature bios, couse i dont see anyone hitting 2000 with p5e3 and cellshock yet?

atleast i got to play with all the settings in p5e3 deluxe :)
just gotta get this with AI clock twister, and exactly more what it does for the mem...and how.

regards.

[saaya]

thanks a lot pedro :toast:
that compliment from you really means a lot to me :)

Infa, dont use that much vdimm!
you dont nearly need that much for 2ghz cas8!
1.9-2v is enough to get 2ghz stable for most kits, if it doesnt run stable at that speed then its most likely the chipset limiting and not the memory.

Noo, ive come to the conclusion that i think its my bios, its just the third out for the p5e3, it just got the Damper enabled, sooo, the errors thats happening, it just dont feels like its the rams faults in some way.......i do put it atleast bios stable CL 7-7-7-20 2000mhz 2.18, and its rock solid in memtest, noo error, but still almost allways, BSOD when loading win, or BSOD in win, or mostly NOT EXACT IN ROUND, but still error free in memtest? whats up whit that? isnt memtest suppose to be a testprogram, so if it goes error free, then the comp will be stable?if memtest passes and windows doesnt then its most likely the chipset begging for more volts and not being able to keep up. what chipset voltage are you running?

can your cpu/board handle that high fsb?
Can your cpu handle the clocks?
tried a lower multi or memory divider?

it can also be the mem, but in most cases in my experience its the chipset. so switch the sticks around, try different slots and try stick1 closer to the cpu and then stick2, this can make a huge difference on some boards.

ahh well, now im at 1827mhz, 4.1ghz and 7-6-5-17, and Transaction at 3, strap at 266, Twister at strong, and only at 1.96v, but att fsb 500 and ddr2000 it just goes crap?

ahhh i hope for the next bios might be a sunshine :) , now ive been trying for 2days in a row all eveing, and it wont goo, so i'll give it up until next release.

im satisfied anyhow hehe....goin such low Cl below rated is fair enough at this speed... seems p5k series got more mature bios, couse i dont see anyone hitting 2000 with p5e3 and cellshock yet?

atleast i got to play with all the settings in p5e3 deluxe :)
just gotta get this with AI clock twister, and exactly more what it does for the mem...and how.

regards.hitting 2k is mostly limited by the chipset, not the memory in my experience. the same mem that can do only 1900 876 with 1.8v on one p5k3 we have here can do 1985 876 with 1.8v on another one... same settings... and its not onle one kit that behaves like that, all do so far.
Im pretty sure it all depends on the chipset.

about hitting 2k on the p5e3, ask pt1t :D
http://www.xtremesystems.org/forums/showthread.php?t=162931

[hipro5]

well i got it, and it wont work, now i will push the Vdimm a bit higher to see if thats the prob, will go over 2.3 a while, hope they not will die on my, they were expensive, BUT I WANNA KNOW DAMMIT :P i want a 32m ddr2000, then im happy, i dont want it 24/7 couse my setting at 930 7-6-5-18 and 4.1ghz ARE FAST!, i just want that magic ddr2000 :P, any tip on a sucessful settinf for ddr2000?

Beware that NOT ALL P35 can get them ABOVE 2000MHz......It's chipset related and not rams..... ;)
One of my Blitz Extreme could get mine at 2070MHz and the other one with the EXACT SAME config/voltages, couldn't take them over 1960MHz.... ;)

[saaya]

Hi George,

so you made the same experience...
Did more vmch help?
Did you think about changing the nb already? :D
If anybody can do it its you :D
Our guys here told me they can remove the chipset easily, but soldering back is something they dont even want to try :D

i wonder if a p35 could work on an x38 mainboard.... and if it would still be 16x 16x :D
But i dont think so... if intel wouldnt lock it then im sure some asian mb mfg would already have made a p35 where we can unlock something to have 16x 16x

[stone_cold_Jimi]

Hey saaya, have you tried the 0704 BIOS on the P5K3 yet? I want to know if it helps der WunderCellShocks :)

[Infa]

Saaya, thanks for you answer, Im feeling a little better now that ur saying this :)

Yes my cpu can handle 510fsb in another MB.
Dunno if the board can handle it, but its an all new x38, it should really take 500fsb atleast..i think, but you'll never know...

i've tried max 1.91 vNB, didnt help, sweetspot for an error free memtest at CL7 2000mhz, was 2.18v.. but it was a no goo, i really cant get it to even do a Super-pi 1MB, load windows i can get it to do, but it crashes easy..

no i dont use that high vDimm , just wanna try if it was that, but now i know that it isnt, maybe i got a bad board? gaah..cant afford to buy a new just to take a chance that its better..

no i havent tried lower, couse im looking for a high cpu speed, aswell as high mem, so im not interested if it goes ddr2000 at 3.6, when i can use 4.1ghz :)

Im running them in the black slots, tried the orange, but that was noticible worse.. so Dual channel in the black ones i do..
Maybe its the bios? perhaps i should try to work it with an older bios, 301 or 402.. i really hope its the bios...

so 9x456 ddr 1827 7-6-5-16 goes fine at 1.96-2.00v... ahh well... maybe i can sell it to someone and get another one....its not broken so i cant return it..
really a bummer, so close :)

and another thing, i dont have That express thing on my boad, not even physically ON the board... thought all p5e3 deluxe came with that? really strange, its a Rev 1.03G board.

regards.

[saaya]

Hey saaya, have you tried the 0704 BIOS on the P5K3 yet? I want to know if it helps der WunderCellShocks :)
I got the 0704 beta 2 months ago or so, i didnt have time to use it but passed it on to 2 people and both said they didnt see a difference.

I dont think new bios versions will bring a lot of improvements, all they will do is probably fix any issues that they find, which is fair enough, its not like the p5k3 performs or ocs bad :D

The problems really do depend on the chipset im 99% sure about that.
I just talked to a reviewer who tried 4 or 5 boards, p35 and x38 from different manufacturers and the max memclocks are all over the place. some nbs clock well some dont, it seems to be the same for p35 as for x38, only that x38 for some reason needs slightly less vmch to get high mem clocks and doesnt boot with more than 1.55v chipset voltage more or less... I think intel bound vmch to some other voltage for the x38 chipset so it scales better with less volts but doesnt like high volts.

Saaya, thanks for you answer, Im feeling a little better now that ur saying this :)

Yes my cpu can handle 510fsb in another MB.
Dunno if the board can handle it, but its an all new x38, it should really take 500fsb atleast..i think, but you'll never know...well... i suspect x38 and p35 to be identical in that regard as well, with p35 the ddr3 boards seem to clock fsbs around 20mhz worse than their ddr2 counterparts. So i suspect x38 will not clock that well fsb wise, but not much worse, 20mhz maybe 30mhz in a bad case.

i've tried max 1.91 vNB, didnt help, sweetspot for an error free memtest at CL7 2000mhz, was 2.18v.. but it was a no goo, i really cant get it to even do a Super-pi 1MB, load windows i can get it to do, but it crashes easy..sounds like a bad nb to me... :(
the chipset is cooled well and all?

no i dont use that high vDimm , just wanna try if it was that, but now i know that it isnt, maybe i got a bad board? gaah..cant afford to buy a new just to take a chance that its better..sucks... well i recommend everybody the same thing, if you want to go ddr3 buy 3 boards, keep the best, and sell the other two on forums or ebay for less than the retail price. you will lose some money on each board but in the end youll have a really good board and buying 3 p5k3s and selling 2 with a loss of 50 euros each means you spent the same or less as buying a maximus formula.
and if you dont run xfire theres really no need at all to buy an x38 board anyways, at least thats my opinion. memory wise there is 0 difference.

no i havent tried lower, couse im looking for a high cpu speed, aswell as high mem, so im not interested if it goes ddr2000 at 3.6, when i can use 4.1ghz :)
try less and work your way up slowly.
thats the best way to find out whats holding you back as well, if you notice that more vdimm just doesnt help then its pretty clear its the chipset.

Im running them in the black slots, tried the orange, but that was noticible worse.. so Dual channel in the black ones i do..
Maybe its the bios? perhaps i should try to work it with an older bios, 301 or 402.. i really hope its the bios...dont think that makes a difference... : /

so 9x456 ddr 1827 7-6-5-16 goes fine at 1.96-2.00v... ahh well... maybe i can sell it to someone and get another one....its not broken so i cant return it..
really a bummer, so close :)1827... thats the max you can get stable in dc? damn... thats a bad board you got there... make sure its not the cpu, play with the fsb straps or better yet try a 1333 cpu if a friend of you has one and see if the limit is the same.

and another thing, i dont have That express thing on my boad, not even physically ON the board... thought all p5e3 deluxe came with that? really strange, its a Rev 1.03G board.
regards.oh you mean that pciE controller chip?
i think only the maximus and blitz have it. the blitz to create artificial 8x8x and the maximus to create artificial 16x8x8x for tripple xfire and sli.
you have 16x16 atm, so unless you plan to run tripple xifre as soon as its out theres no need for that chip anyways :)

[Infa]

Hmm, you say that X38 dont like high volts, actually i havent tried 1.55, to see if it is that im using to HIGH vMCH? If it is like that with the dimm, why not NB?

No, i can get it stable higher, but as i said :) im looking for high cpu speed..so im allways using the 1:2 Divider which is the highest divider..

And i am using a 6850...so its a 1333fsb cpu :)
i will try lower volts in the MB settings overall and see how it works.

I got a fan blowing on the mems aswell as NB, it never rises above 35c, think thats low enough :)

And no i didnt mean pci-e , i ment the "splashtop" thing, on the board there are supposed to be a "usb" between the Pci-e..i dont got that, its not even in the bios either.

Tried all Straps, 200 no boot, and on every other one, it boots and behaves the same, 266,333,400,auto.

No, more vdimm helps to a certain point..then it doesnt help anymore, and like i said before, maybe im using to high vMCH? i never go below 1.60.

i WILL get it to work..hehe..the damn board aint gonna win before ive tried ALL.

[saaya]

Hmm, you say that X38 dont like high volts, actually i havent tried 1.55, to see if it is that im using to HIGH vMCH? If it is like that with the dimm, why not NB?well i dont know what the default nb voltage for x38 is, for p35 its 1.25v afaik, on x38 gigabyte +0.375 seems to bee too much and +0.300v seems to work. and on asus boards a few people said the sweet spot is between 1.5 and 1.55v.

And i am using a 6850...so its a 1333fsb cpu :)
i will try lower volts in the MB settings overall and see how it works.ahh nevermind then :p:

I got a fan blowing on the mems aswell as NB, it never rises above 35c, think thats low enough :)good!

And no i didnt mean pci-e , i ment the "splashtop" thing, on the board there are supposed to be a "usb" between the Pci-e..i dont got that, its not even in the bios either.no idea what you mean... :confused:

more vdimm helps to a certain point..then it doesnt help anymore, and like i said before, maybe im using to high vMCH? i never go below 1.60.try less....

i WILL get it to work..hehe..the damn board aint gonna win before ive tried ALL.well i just hope you can do that without killing any hardware, you shouldnt increase volts more and more and more without even knowing if those high volts are necessary or not :P Your went nuts with vdimm and running 1.6v chipset from the start is overkill as well, you should start low and work your way up to see what volts you need to bump up and how much :)

[Infa]

yep, will try when i get home today :) *cross my fingers*
the RAM's can so easy handle 2000, (Very good German Qualtity :)) and i really want it to work.

and yes, i mean the splashtop, thats the Linux Enviroment that is suppose to come with p5e3 deluxe that everyone was talking about, i dont have it for some reason? maybe a early revision? or a late, where they have removed it?

I got a taste for cellshock now, and will by it more times, i like the solid Heatspreaders that are put on them, it just feels quality and that u guys are good at this, not atleast the top performance that you have deliviered for some time now :) , appriciate that you are here as a rep, and a big helper nomather what :) thanks!

yeah, the rams are almost as cold as the air blowing on them..ddr3 are cold compared to ddr2.. even with 2v.

yeah im a fan of high volts, ive allways thhought higher were nescessery to push it, so im used to start high hehe, i killed my striker that way i guess, but the board sucked anyway ;) but that was before, now the technology has improved i guess... :)

Regards.

[saaya]

yep, will try when i get home today :) *cross my fingers*
the RAM's can so easy handle 2000, (Very good German Qualtity :)) and i really want it to work.

and yes, i mean the splashtop, thats the Linux Enviroment that is suppose to come with p5e3 deluxe that everyone was talking about, i dont have it for some reason? maybe a early revision? or a late, where they have removed it?ah... good questions... too bad it doesnt seem to come with all boards... it sounded really interesting... ask asus support?

I got a taste for cellshock now, and will by it more times, i like the solid Heatspreaders that are put on them, it just feels quality and that u guys are good at this, not atleast the top performance that you have deliviered for some time now :) , appriciate that you are here as a rep, and a big helper nomather what :) thanks!your welcome :toast:
I regret that i dont have more time to spend here... atm i dont have internet at home which really annoys me :D

yeah, the rams are almost as cold as the air blowing on them..ddr3 are cold compared to ddr2.. even with 2v.yeah i never understood why some ocz and i think it was mushkin? reps said ddr3 would be much hotter than ddr2. maybe the samsung ddr3 they had initially ran very hot. I really gotta say, this micron ddr3 memory never stops to amaze me, awesome job micron guys! :toast:

it clocks incredibly high, can run almost any cas latency, performs very well, runs cool and doesnt seem to die all the sudden like D9GMH/D9GKX (not yet, fingers crossed! :D)

[stone_cold_Jimi]

Forgot to say congrats Pedro, that is an FSB to die for m8 - awesome figures. Tweak those sticks some more please! :up:

[Infa]

now ive tried lower NB volts, and it behaves exactly the same, wich i then have figured out that my board just cant do ddr2000 stable...that sucks.

1900 cl7 it can do without a glitch, but not 1960.. so i have a crappy board and im not satisfied with that...hmmm..and i have to buy a new one before i sell my old one if thats the case...ahh well..

When are the maximus Extreme due to the market?

regards.

[metro.cl]

Victor all settings are boot from BIOS?

[Pt1t]

Victor all settings are boot from BIOS?

Me it s stable like him but i have to boot @ 470 ;) . I dont know why he can boot @ 500mhz ...

[NapalmV5]

@ 420$ its very tempting

[saaya]

infa, your board isnt "bad" if it can run "only" 1900 dc stable or thereabouts, but yeah, there are def better boards/nbs out there... as i said most boards i have here can do 1925-1975 more or less... so they are all close to 2ghz but no way to get it stable so far... gotta try a vmch vmod :D

[boblemagnifique]

LE82P35
SLA9R
L716A390

For my Blitz Extreme :)


My new Blitz Extreme :

LE82P35
SLA9R
L719A510

[vanovich]

nice stuff pedro