Jump to content

marcuri

Members
  • Content Count

    27
  • Joined

  • Last visited

About marcuri

  • Rank
    New Member
  1. ah my hopes are dashed. do you happen to know what voltages they used or have any posts specifically about it? i couldn't find anything in the search.. seems like its hard to find old relevent threads like that.
  2. you sure about that? hm... interesting. i heard about the better preformance stuff but thought that it was only noticable over gigabit due to it being more like a PCIE lan card cause its onboard and intergrated with the chipset. i guess i could be wrong.. i never ran any benchmarks... however are you sure the cpu usage is down and not up over the marvel?!?
  3. see my post here: http://www.dfi-street.com/forum/showthread.php?t=34891 In short i personally don't think burning in is what you're looking to do. Theres simply no shortcut to breaking in a processor except for time. If you want, just simply be comfortable running it at a *relatively* higher voltage (than what u'd deem as your personal max.. as long as its not over the max recommended by people for your particular processor)... as long as its stable for about a month and then lower the voltage down as far as you can after one month of use without changing any other settings and seeing how much you can reduce it with it still being stable. Keep re-evaluating every 1-3 months and lowering it accordingly. I think a processor is like a car or a baseball glove. Only thing thats going to break it in is time and use. The safest way is to use it for about two months without overclocking at all. Then attempt to overclock it as I believe it will net better results at lower voltages once the processor has been broken in with use (not simply high amounts of heat and voltage). Maybe you could replace months with weeks... for me it worked after a few months.. i didn't try any sooner than a couple of months Please note this is something thats based simply on my own observations with my own computer setup. I think its a really reasonable and overlooked idea/theory with processors. Who knows how right it is.. but logically i think it makes sense. At least wait a good week after the computer has been running 24/7 or as close to that as possible. A few good cooling and heating cycles would also help get the thermal paste transfering heat optimally... so maybe a prime 95 stress test for an hour followed by an hour of the computer being off about 3-4 times during that 1 week before overclocking would help with temps when you finally do overclock. Time for breaking in a processor is something people don't take seriously enough... and while i've never heard anyone suggest this.. or any of the above really how i've stated it (thats not to say it hasn't been said) i think if ou look ate my idea/theory it makes alot of sense. Hopefully if you follow my advice you'll have a processor that lasts longer and has a better intial overclock. Remember, every couple of months try to lower the voltages and get it stable at a lower voltage if you can. it'll help temps and increase the processor life. eventually you will reach a point where you can't go any lower.. but that probably won't happen until approx 2-3 adjustments over 6-8 months. my guess is you can get it down a good entire volt over that period. likewise theres also the possiblity of increasing your overclock with the same voltage... but i don't recommend that... you'll probably end up with a processor that will die MUCH MUCH sooner. Good luck. hope this helps
  4. this is something i think alot of people tend to forget to do... and then wonder why their processors don't last as long as they should. What i find most interesting is the idea of burn in. I had my memory recently running at stock settings with stock voltage, BH-5 chips.. which is 3.1 (mushkin) and after about 2-3 months i decided to see if i could lower it without problems. Well easily, without having anything overclocked... i was able to lower it to 3.0 and keep all settings the same. Clearly processors and other components tend to do better when "broken in", similar to how a baseball glove is new, compared to months of use. Or a car engine is new, compared to a couple thousand miles on it. Every 2-3 months re-evaluate your voltages... i think thats a great tip for anyone whos overclocking, especially if they recently did it.
  5. geez what an oc... o well. mine right now is water cooled.. and even then the water cooling system has alittle bit of "gunk" around the block (i cleaned it out but alittle is still there . that being said.. after posting i looked around alittle bit. average i've seen is about 2.4 without increasing voltage, and 2.6 with increasing voltage. its still prime stable and its been going 2+ hours now. gonna wait til it does 8 hours and then i'm gonna do a few more different tests. it might not be a record breaker even for water cooling but it sure seems like its up there.... and i think i can easily squeak it out up to 2.9 if i wanted to increase the volts to 1.6ish.. maybe 3.0ghz if i tried getting it close to 1.7volts...... but i don't think i'd want to run that constantly... maybe just to see if i could. Was that 3.4 ghz overclock prime stable? ;p (my guess: no). -blisk
  6. i too was overwelemed by the guide but it really isn't as hard as it looks. You learn by doing, and so long as you have a laptop or another computer to read it on (hell print it out if you don't) follow it step by step, go from one part to the next and back again if you need to look up stuff. they're right, if you don't know what you're doing you're NOT going to have a good time overclocking your processor. that being said heres a easy way, but I'M NOT RESPONSIBLE FOR YOUR LAZINESS IF YOU SCREW UP THAT PROCESSOR OF YOURS. YOU'LL LEARN SO MUCH MORE IF YOU READ AND ITS NOT THAT HARD. if you don't care about even reaching anywhere near what your processor can do.. simply increase the FSB mhz at approx 10mhz increases. as soon as windows doesn't boot or your network card is acting up, ect. increase the voltage of the dram. anything above a 200 mhz overclock might need an increase in the Vcore (processor) as well. remember, if its TCCD based memory don't increase the voltage as quickly or as much. if its BH-5 based memory you can bring those suckers up to 3.2volts no problem for the most part. BE CAREFUL. you really ought to read.... but i'm guessing your someone that has to learn the hard way. oh and you will eventually... maybe you'll get what you want from what i told you.. but maybe you'll also fry that processor of yours. maybe you won't even get the best preformance out of the chip. o well. you asked. you didn't want to learn. so there it is. good luck buddy you're going to need it.
  7. unfortunately the tracert doesn't rule out weither its a wan or a lan issue. try tracert on an ip address outside of the network, and then a tracert of a an ip inside your school's network.
  8. i've noticed a difference with the marbel over the nvidia in terms of CPU useage... so i prefer the marvel. i haven't don't directly benchmarks of one over the other. but i'm 95% certain the nvidia card uses more cpu.
  9. Hey i think i have a winner on my hands here.. whats the best overclock been for a venice 3000+ chip?!? This chip is nearly 3-4 months old but i only got around to overclocking it now. Right now i got the FSB at 351 on a 8.0 multi with a 3/4 ratio. That means i have a 2808 mhz chip right now. Memory is running at 3.2 volts (haven't tried above that) and Vcore (cpu) voltage is at 1.5 * 1.04% (think thats 1.56volts). Peak temp on prime (been going for 1 hour no problem now) is 104 F (40C). I haven't even reached my max RAM mhz (~240) or my CPU max mhz (just under 2.9ghz, but at nearly 1.53 * 110% (didn't try tweaking the voltage lower). So anyway whats the record for this chip. Seems like over a 1ghz overclock is definately pretty good... still haven't done all the stability tests so i'll get back to you all. -blisk
  10. well i got it up to 2.9ghz but not completely stable... so i'm at right now 2.79 ghz and prime stable after 8 hours of stress testing it. Its running at 1.55V. Is this still alittle high? i can always bump it down and get 2.75 or so at 1.50. I'm gonna keepmessing with the voltages so i can see what the least is that i need to get this running at a certain speed. Quick question... since supposedly higher voltage shortens the life cause of electron dissidence wouldn't that mean after a certain amount of time i could lower the voltage going to the processor and still have it stable at that speed? or am i completely wrong on what i think electron dissidence is.
  11. well i'm up to 1.550 Volts + 110% @ 2903 mhz stable with super pi and a few others... total volts is now alittle above 1.65 Volts going through that processor..... please lettme know if you think i'm likely to fry it.... appreciate it. ;p
  12. haha... yea i just saw that thread... unfortunately it doesn't tell you very much... its a bunch of things i've never heard of before... chips lasting only a few months at volts i thought were safe before (1.6-1.65 is recommended max by a few people on air.. and alittle more on water). But i really don't want to hurt the processor to the point where it won't last more than a couple of years. i know voltages will kill it eventually due to electon disidence (sp?) but i didn't think 4-5 months was it. O well... guess i'll run it slower for the time being until this is all resolved.
  13. I've gotten my 3000+ up to 2880 stable in windows (typing on it now) and its passed a Super PI test. Please note that this is with the ram on a slow divider so its not the limiting factor (following the ultimate guide posted here by someone)... i'm just testing to see how high the cpu can go, next up is the ram. BUT I NEED SOME ADVICE ON ITS SAFTEY... To get it this high voltage is at 1.525 * 110% (which MBM says is 1.63 volts). I'm watering cooling the system with a 2x120 fan and, 5 panaflow fans (4x120 +1x80mm) in a lian li v2000 case. Temps don't reach over 114F (MAX) when under load and case unforunately has reached a max of !!!129F!!!!, chipset is at 120F. Is this unsafe between the voltages and the temps? I know the CPU is under 50C which is good but I'm really concerned about the voltage and the case temp. I haven't verified that these temps are accurate as its all on motherboard diaodes but they seem relatively good. So do you think its safe? Thanks.
  14. what chips does that use? they run at high voltage so i'm assuming they get as hot as my redlines... am i more likely to see better results from the OCZ than the mushkin? also any TCCD recomendations or is that out of my price range if i want a good overclock with them?doing a quick look over at newegg some g.skill are within my price range however i don't know what would yield best price/o.c. potential... again my main issue is heat and to a lesser extent, but related, is noise. the pc will be on close to 24/7.
×
×
  • Create New...