Printable View
No offence, but the only reason you can get away with a comment like this is because you work for OCZ, you're highly respected and have always provided the community with quality information. However, I don't see why you would not share information that would help people not to kill their memory. That would benifit both the user and the manufacturer, isn't it? :)Quote:
Originally Posted by Tony
By the way, please don't take this the wrong way, I'm not trying to offend you in any way.
NDA's Massman :)
The PSC part is better Eva, lower voltage and less issues i hear.Quote:
Originally Posted by eva2000
.. I still stand ground... I do not think one can say ONE thing is the reason for all the deaths.... Exactly as that Saaya seems to agree on...? Strange thing is, that sure isnt what being said in first post of the thread...
Thats why I call BS, and honestly thinks the thread is misleading....
Will it work calling Luken and invite for coffee? :rofl:Quote:
Originally Posted by RyderOCZ
Breaking NDA by posting detailed information is one thing, giving hints that helps to prevent people from killing their ram is a whole other issue.Quote:
Originally Posted by RyderOCZ
I agree that many of the memory secrets cannot be shared due to NDA, but I don't agree that you can't help by giving hints on what settings might cause these memory sticks to fail.
I think you read more into Tony's statement than is actually there.Quote:
Originally Posted by massman
None of us would withhold that info unless it was strictly forbidden for some reason, which under normal circumstances wouldn't be anyway.
Well said Mr RyderQuote:
Originally Posted by RyderOCZ
thanks for beeing as clear as you can without spilling all the beans :toast:Quote:
Originally Posted by Tony
i think this statement was totally uneccessaryQuote:
Originally Posted by Tony
how is this helping?
if you cant share it, why would you even post about it?
m.beier, ive said it dozen of times... there is no way to say xx and yy and zz combined will kill your memory. there is absolutely no way!
even if you would kill hundreds of chips under whatever conditions and then come up with a formula of x and y and z etc and all the parameters that would, for a given configuration, kill the memory. there would be memory falling out of that formula and you would be proven wrong.
there IS no safe way to say what will kill/damage the memory and what will not.
the point of this thread is to give people as much information as we can, so we can at least have an idea of what configuration and parameters CAN cause problems. this doesnt mean there is enough info in this thread you can use to be 100% sure your mem wont fail. thats not the point...
as soon as 1 person realizes hes on the edge and changes the settings and safes his mem that way, then this thread was worth it!
and if you ask me, there is enough info in here to prevent a lot of people from damaging or killing their mem. and the biggest gain might not be preventing mem from dieing, but people at least knowing they are entering a mindfield with certain voltages, and dont end up with dead mem without ever knowing they were risking it to die.
saaya, i work with partners on all sorts of issues, nothing to do with ddr3 dying etc...this is what i was meaning. So much happens but I just can't tell you as I can't....im probably so deep under NDA's and working agreements it would take a week just to dig my self out the 5h1t.
You guys know i will help keep your ram alive but I was talking about a MUCH bigger picture...MUCH MUCH bigger LOL
calm down guys please.
well tony, i know what you mean but then DO keep it to yourself...
you cant really wonder why people arent too happy when you tell them
"I know so much cool stuff you guys dont know, but oh, i cant tell you ;)"
i mean what kinda replies to you expect to that?
the context was not that at all, read my post again. I was quoting the forum world Vs the real world and what goes on, nothing more, nothing less.
Jeez you guys are way to critical, maybe I should not help in future
Tony, on behalf of myself, no one other then myself....Quote:
Originally Posted by Tony
I admire your work with BIOS, I do remember Oskar Wu and your BIOS' back in the days of nF4...
However, thats one thing, satisfying the needs, and really really great, I truely respect that!
However, this isnt only you, but I'd really like all the company reps to get out of the xtreme sections.... I really hate the fact that theres a thread in the "XTREME" section regarding safety.... However, I have mentioned, and hope it will be taken into consideration... That I think a general and then the xtreme should be working parallel, two divided sections, and with strong restrictions on the xtreme one.... Because frankly, I find it really really annoying to read all this safety stuff in a section called xtreme... I think its ment for extraordinary performance, people not caring much for lifespan of equipment, people with too much bloody money, or blood on their theeth....
So my personal oppinion is; "yes", I do think you should stick out of the section, along with all other representatives that has to be concerned about warranty etc....
.... Keep in mind, its my personal oppinion, nobody else but mine...
Apart from that, Im really pleased with the pictures of datasheet you have posted, I dont have that .pdf myself... Where can it be found...? Is it material send from micron to the memory companies, or?
in what context is "i got something cool but i wont share it ;)" helping?Quote:
Originally Posted by Tony
EDIT: please dont listen to m.beier, imo your last comment about the knowing so much etc wasnt necessary and its no secret you and me dont get along well, but you DO ppost useful information, so this might sound odd comming from me :D but please dont stop posting here.
m.beier, i see what you mean, but i dont agree.
exactly since this is a place dedicated to pushing hardware to the limits, imo its important to share where the limits are, so we know where the "ceiling" is and can "dance on the ceiling" without killing that much hardware. I would have made this thread as well if i wouldnt work for cellshock... i actually would have posted more infos, and i think the same can be said for tony.
if you prefer to find out for yourself at what point your hardware dies or gets damaged, then just ignore threads like this...
why do you keep posting here and in other threads telling everybody that you dont like the threads? just ignore them and make your own threads then... :stick:
:( Hm - so You can turn it to the left - turn it to the right, we all need DDR3 RAM capable of more volts without damage.....;)
Most of the today's available kits need 2.2 - 2.3 volts to show their whole performance but thats not safe anymore.:(
itll always be like this... if the max 24/7 stable voltage a ddr3 chip can take would be 3v then we would see the best results with it at 3.0-3.3v i think :D
unless they build the chip to not scale with high volts, and in this case we are stuck with chips that can take 3v but only scale up to 2v... so... either way, i dont think theres an easy design fix for this :D
sorry, i forget this thread , in Cas 7 1800 before i hit 2.3v for bench, my was run stable at 1800 7-6-5 2.00v, but after i run that voltage, more one week my ram was not stable 1800 7-6-5 2.00v i can try 1800 7-6-5 2.1v but is too high ....in 1000MHz i can reach boot now, not for windowsQuote:
Originally Posted by saaya
OMG...1.5v can killed DDR3:confused: bad psu or human error?Quote:
Originally Posted by M.Beier
1.9 is plenty so show killer numbers ;)Quote:
Originally Posted by camouflage
on 790i anyways heh
yeah but its not about fast, its about THE fastest, and for that you will always need to be at least half way in the " voltage minefield" :D
i was thinking about 24/7 here
actually ONLY in superpi do i enter voltage minefield frequently :D
:cool: Hm, I think it all depends on the quality of the chips. But sadly You'll never know before..........:( .
My first DDR3 kit was 1800 Cellshock and for benching I used 2.23v real.
Last Year I thought that's save....., I didn't kill them but I degraded them so that in the end there was no stable run at default speed & volts possible anymore.:down:
really? you degraded it with 2.23v? are you sure you didnt try more?
What board was that one? did you meassure vdimm with a dmm?
:rolleyes: Was on my Maximus Extreme - Vdimm measured with multimeter.Quote:
Originally Posted by saaya
2.23 was max. - not safe............:( - degration after this test:
http://www.xtremesystems.org/forums/...d.php?t=168298
:shrug:
and how did it degrade?
what was the max clock before and what is it now?
are you sure its the mem?
maybe the cpu or chipset degraded?
or with a new cpu or bios or whatever you can reach the same clocks?
I have a casualty to report:
Team Group 1600C7 - D9gtr - 7MB12 KFGB.
Benched up to 2.275V (mobo limit). Not much of a performance increase though, so mostly @ 2.15V. Always with a fan direct, ambient ~24C.
Pretty quick onset, only saw minimal degradation ( 925 - 900 MHz @ 2.15V for 3D)
One MAJOR caveat: Board is 790i, mcp/northbridge used for extended periods @ 1.55V (max).
put up with about 2 weeks of abuse.
anyone have an idea of cold bug temp for rams? :D I may need to make a 'safety' LN2 pot :cool:
sorry to hear about your loss :(
so now they degraded or are they dead?
do they still boot?
you only used them for 2 weeks? or only for 2 weeks with high vdimm?
-40 celsius works without a hitch, dont know about less
i know ddr1 and ddr2 dont like to boot at temperatures below -80 degrees or so... ask jmax, hes the king of frozen dram :D