You will have to up voltages, most likely. I would leave it at standard until you get stable at 3.6ghz.
I use 425 x 8.5 = 3612mhz @1.23v
I use 425 x 8.5 = 3612mhz @1.23v
- Dec 25, 2007
- 1,243
- 3
- 76
I'm trying that right now, Turbo caused a BIOS reset/no POST again so I set it back to standard.
I've put loadline back at auto, MCH at 0.05 ("default") all other volts normal and upped DDR voltage to +0.1v. I tried 0.05v on DDR and it booted but failed Orthos again. I'm hoping +0.1v solves it.
I've put loadline back at auto, MCH at 0.05 ("default") all other volts normal and upped DDR voltage to +0.1v. I tried 0.05v on DDR and it booted but failed Orthos again. I'm hoping +0.1v solves it.
Originally posted by: TheJian
http://www.intel.com/support/p...ssors/sb/cs-020033.htm (warranty info here!)
At this point I say RMA if you can, tell them you couldn't get it to work properly in your board/setup...
Then buy a 3yr box.
excaliberpc.com $229 in stock
Allstarshop.com $239
J/K
I'm wondering if your PSU just sucks. If you have a fry's nearby you could pick up a monster PSU just to test and RMA it the next week. Get creativeOr keep it since it solved your problem
I love that guys nick...LOL.
I'd still like to hear what happens at 1.3v. Or even 1.28+. 1.22 is really low. Hell 1.3v is pretty low.
http://processorfinder.intel.c...tails.aspx?sSpec=SLAPL
SLAPL is probably on your chip. Which means it's safe up to 1.3625v
I can't resist...You're an ass.
(I kid, I kid)Seriously though I doubt Intel gives serious cause for concern over their RMA rates.
There is the "ethical" question but this has got to rank right up there with the "ethical" question of anyone that drives 1mph over the posted speed limit. How dare you break the law! Think of the children!
RMA/warranty policies of the kind that Intel and AMD have are more structured to intentionally conform to outstanding potential liability calculations that GAAP requires.
If they left the door open for a stated policy on the books that 100% of their CPU's could potentially be returned if every customer overclocks and burns them out then they'd have to account for this liability.
As it stands they can arbitrarily limit just how much of a replacement rate they are willing to tolerate by turning a blind eye to RMA's on badly over-volted CPU's.
Can they tell when a CPU has been badly overvolted? Yes. Are they going to refuse an RMA just because it's clear that the chip was abused? No, not unless the abuses are so rampant and widespread that the replacements begin to cost them more than the goodwill they want to generate in the community in general.
Remember that an RMA costs the manufacturer to the tune of their replacement costs, not wholesale or retail costs. Replacing that RMA'ed chip might set Intel back $25 in true dollar costs. They'd prefer not to lose $1 let alone $50, but they aren't going to spurn customers just to avoid a $50 charge. No business survives too long doing that...not even Walmart has the cohonys to close their returns dept and tell their customers to stuff it.
- Dec 25, 2007
- 1,243
- 3
- 76
Well, the decision on whether Intel wants to warranty an overvolted CPU ends up being whether Intel feels like it or not.
Btw, quick question. Using large FFTs in Orthos also stresses the CPU, right? There's no disadvantage to using large FFTs over small ones if you're trying to stress both CPU and memory at the same time?
ATM, I'm running Orthos at
Type: Custom... Min: 768 Max: 32768 InPlace: No Mem: 3072 Time: 10
CPU: 3600MHz FSB: 400MHz [400MHz x 9.0 est.]
CPU: 3600MHz FSB: 400MHz [400MHz x 9.0 est.]
Running for 14 minutes so far, I'm going to leave it up while I head to work. Hopefully it stays active and keeps running when I get home.
Btw, quick question. Using large FFTs in Orthos also stresses the CPU, right? There's no disadvantage to using large FFTs over small ones if you're trying to stress both CPU and memory at the same time?
ATM, I'm running Orthos at
Type: Custom... Min: 768 Max: 32768 InPlace: No Mem: 3072 Time: 10
CPU: 3600MHz FSB: 400MHz [400MHz x 9.0 est.]
CPU: 3600MHz FSB: 400MHz [400MHz x 9.0 est.]
Running for 14 minutes so far, I'm going to leave it up while I head to work. Hopefully it stays active and keeps running when I get home.
Sorry didn't mean the PSU brand/model sucks, just that you may have a bad one. Maybe a voltage is fluctuating to much and dropping too low. which is kind of why I was trying to get you to aim a bit higher. I understand wanting to run low, but do it later after you find what works. You could be spending time all day on something that vcore could quickly fix, which is the first thing people would tell you to raise in your case. That's why I always set mem to .1 over whatever the default is for my modules (I only buy 1.8v anyway, or 1.8-1.9 knowing they are just fudging for voltage fluctuation on diff boards), and cpu's at whatever the going stable rate seems for what I'm trying to hit in mhz.
Sorry no info on orthos. I'm always using prime for that stuff.
If your chip dies at 1.3v this week I'll send you my e4300...LOL If you just bought it though you should have plenty of warranty to feel fine at 1.35v even. I'm using stock fan on my 4300 at almost 1.36v (it's 1.35xx), but it does show in pc health a bit lower. It's cool, but another click and I noticed heat starts to jump. Default out of the box was 1.31xx if memory serves. I'm running at 3ghz though, not 1.8ghz. If you really don't want to raise it, does the chip even run fine (24hrs prime95, or orthos I guess) at stock 3ghz? I read the first post again, and it seems if you're having this many problems we need to look at stock and fixing the reboots. A PSU from frys can test that. Get one you could live with or just a powerful one that you know you're returning after test We're trying to get something stable that is already known to have problems at stock so far. If it is the PSU we would just be exacerbating the issue. With a decent run at stock, but reboots, and worse problems overclocked that make no sense with your parts (except cpu volts being low) there is only one thing left most likely. The PSU.
We're running out of options.
Sorry no info on orthos. I'm always using prime for that stuff.
If your chip dies at 1.3v this week I'll send you my e4300...LOL If you just bought it though you should have plenty of warranty to feel fine at 1.35v even. I'm using stock fan on my 4300 at almost 1.36v (it's 1.35xx), but it does show in pc health a bit lower. It's cool, but another click and I noticed heat starts to jump. Default out of the box was 1.31xx if memory serves. I'm running at 3ghz though, not 1.8ghz. If you really don't want to raise it, does the chip even run fine (24hrs prime95, or orthos I guess) at stock 3ghz? I read the first post again, and it seems if you're having this many problems we need to look at stock and fixing the reboots. A PSU from frys can test that. Get one you could live with or just a powerful one that you know you're returning after test
We're trying to get something stable that is already known to have problems at stock so far. If it is the PSU we would just be exacerbating the issue. With a decent run at stock, but reboots, and worse problems overclocked that make no sense with your parts (except cpu volts being low) there is only one thing left most likely. The PSU.We're running out of options.
- Dec 25, 2007
- 1,243
- 3
- 76
I believe I said it was completely stable at stock everything. I haven't encountered a reboot in a while though, like I said it's very random. (Though having said that, I expect it to reboot on me in a few minutes just to spite me.)
CPU runs stable at 3.6GHz as well as long as I run Orthos using small FFTs (Ran that overnight with no problems)
If I stress memory with Memtes86+ or HCI memtest, it turns up no errors either.
I see problems ONLY when running Orthos with Large FFTs.
For now I'm not seeing anything wrong but time will tell.
CPU runs stable at 3.6GHz as well as long as I run Orthos using small FFTs (Ran that overnight with no problems)
If I stress memory with Memtes86+ or HCI memtest, it turns up no errors either.
I see problems ONLY when running Orthos with Large FFTs.
For now I'm not seeing anything wrong but time will tell.
You can probably up voltage and still be under 1.3v and stable. I have read orthos is buggy and since then i use Prime95.
What does Everest say your 12v rail voltage is?
What does Everest say your 12v rail voltage is?
LOL@Idontcare
I'll take that as a compliment coming from someone with that nick...
You're correct about the warranty stuff. But on a boxed chips. I'm not sure how Intel tracks it, it's been so long since I've had a dead chip, I have no idea if they have a heatsink policy. AMD has a heatsink code that matches your specific cpu AFAIK. You actually have to mail it in with your cpu. And if the sticker is gone they tell you buzz off. Would they let it ride if it was gone? I don't know. If someone switches I always tell them keep the heatsink/fan around in the box for later.
I've heard (but not experienced) that AMD freaks over leaving grease OTHER than theirs. I have had a few of theirs RMA'd (partly because until core 2 that's all I used or sold for the most part during A64's reign) and always cleaned it beloved patriot & span before sending it back. No trace of ANY grease I have an AM2 X2 3800+ that needs to go back here now, so I'll have another round of action from AMD. I've never been hasseled though from either. Never had a customer get hassled from either in 8yrs I had the pc business.
I'll take that as a compliment coming from someone with that nick...
You're correct about the warranty stuff. But on a boxed chips. I'm not sure how Intel tracks it, it's been so long since I've had a dead chip, I have no idea if they have a heatsink policy. AMD has a heatsink code that matches your specific cpu AFAIK. You actually have to mail it in with your cpu. And if the sticker is gone they tell you buzz off.
Would they let it ride if it was gone? I don't know. If someone switches I always tell them keep the heatsink/fan around in the box for later.I've heard (but not experienced) that AMD freaks over leaving grease OTHER than theirs. I have had a few of theirs RMA'd (partly because until core 2 that's all I used or sold for the most part during A64's reign) and always cleaned it beloved patriot & span before sending it back. No trace of ANY grease
I have an AM2 X2 3800+ that needs to go back here now, so I'll have another round of action from AMD. I've never been hasseled though from either. Never had a customer get hassled from either in 8yrs I had the pc business.You could be right. I always use prime because it's been around forever and used by all the major sites. Memtest is pretty dang reliable too AFAIK, and again used by everyone. I've also never seen it crash unless I had something wrong. I mean once I'm stable I'm good until I get tired of seeing it run
DarkRogue,
You just mentioned the reboots at stock in your first post.
"I'm running Orthos to do some stress testing (only that for now as I don't have too much time to tinker with it.) At STOCK speeds (333x9 1.225v) Orthos runs for hours on end perfectly fine with Blend (until the PC reboots, may be unrelated.) "
We could just be dealing with a bad PSU. I keep running that statement around in my head thinking it easily could be the root of all of this since it affected a PREVIOUS pc build. The only thing in common there would be PSU/vid (didn't notice, maybe you said you bought the vid for this build). Of these two parts the 90% likely problem would be PSU. If it's not related to other parts the PSU is probably dipping too low every now and then and the pc thinks you hit reset button and reboots.
You might try one module then the other. Just to see if they make it alone, or one just craps out over the other that makes it fine. Also if you could, what are the voltages showing for the 3,5,12v in the bios (pc health page)? Everest should be able to tell you while @ load so you can see if it's dipping quite a bit (if memory serves..haven't used it in a while). Here's a quick search at google, another with the same issue:
http://www.compuforums.org/gen...dom-reboot-issues.html
PSU solved it.
What vid card we dealing with here?
Originally posted by: Idontcare
You ever notice people's screen names rarely seem to reflect their true character? I use to think it was the opposite, people used their screen names to reflect how they really view themselves...but it seems to be the opposite on so many occasions.
Even my own screen name falls into this observation.
I think that the persons' screenname is probably who they would like to be, and not who they really are. In your case, I would guess that you would like not to care much about what other people think, but you obviously do based on your high post count. I know that I do many things that I wish I didn't do, but I still do because I can't help myself.
Ever notice how people assume what someone's name stands for? Because it is OC it must stand for overclock, correct? It couldnt possibly be that I live in Orange County, CA.
Anyway, the OP saying "Hi, I'd like to overclock to 3.6ghz, but Id like to keep stock Vcore, because I hear other people do it" is silly. Yea, i'd like my car to go faster without supercharging it.
Actually my nick was never intended to relate to my disposition.
It literally was a response to the question "what would you like your nick to be?" to which my response was "oh, I don't care what it is"...meaning I thought I'd mull it over and come up with something later.
Kinda like habitually naming intentionally temporary files "deleteme.xyz" so that later you can quickly do a system search for "deleteme.*" and delete all your created temp files in one sweep.
At any rate I got what I asked for and by the time the "damage" was done I realized I couldn't change a nick once an account is made...so I've just "lived" with it ever since.
BTW, you can't really call 1600 posts in 5.5 years a high postcount, can you? (note 9/18/2002) But I do agree I have been posting a lot more lately.
That's true, but I know that you have been posting a lot lately. Of course so have I. I actually realize that is a bad habit of mine as I will post some mean things if I post too often, so I try not to post many comments.
Your symptom suggest that the CPU is running okay but the RAM is not doing too well. Anyways, turn off any BIOS overclocking stuff, turn off power throtthling as well during your testing. Try this, lower FSB, but up the RAM:FSB divider to raise the RAM to the advertised speed. Then run small vs blend. At this point your CPU should run below the E8400's stock speed so it will be out of the picture just your RAM is the culprit. If you still got trouble with Blend, try Large FFT, if both no good but pass small FFT. You probably got RAM issues. Although exactly what issues need more info to tell. But pretty much your RAM is screwing your up.
- Dec 25, 2007
- 1,243
- 3
- 76
Bah, this is frustrating.
Sorry for the long absence, I've been tinkering with voltages all night after work.
I'll respond to the relevant posts I missed in order.
Xpoc, I do not have Everest, but BIOS reports my 12v rail at 12.63 or something. I took that bit of advice on Orthos being buggy and am running two instances of Prim95 now, and while it does run a little longer, it still fails.
Jian, I just set up an RMA with Newegg for my PSU, I need to eliminate these variables, it's just pissing me off way too much now. I am using an eVGA 8800GT at stock speeds currently. My Accelero S1 Rev 2 arrived yesterday so I'm going to attach that and OC the video card a bit after I get the damn system stable.
As far as voltages go, CPU-Z reports my Vcore at 1.232 - 1.248 when I set my Vcore in BIOS to 1.225-1.250. At Vcore 1.300 in BIOS, CPU-Z is reporting 1.296.
Speedfan reports my +3.3v as 3.28v, +5v as 4.95v, +12v at 0.06v (idle atm), -12v at -16.48v (???), -5v at -8.48v, +5v at 5.38v (Don't know why there are two), Vbat at 3.20v.
Oddly, it also says my Vcores are vastly different for some reason. Vcore1 is 1.30v, Vcore2 is 1.97v.
As far as my being silly at wanting to OC at stock volts, I wouldn't know whether I can do it or not if I don't try now will I?
nyker, I'm not entirely sure if it is the RAM now either. If small FFT's only stress the CPU, then my CPU is rock solid even at 1.225v. Memtest86+ and HCI Memtest can also run forever and not find anything wrong with my memory now. The only error I've seen so far is the failures in Prime/Orthos when running large FFTs (I set it up to use 1024K - 32768K FFT's now so I can bypass it testing all those smaller FFTs and get a fail, if any, sooner.)
My initial bootup was with the CPU at stock 333x9 1.225v and the BIOS "auto" configured my memory for me (picked the 5:6 divider iirc.) That ran everything fine.
Right now I'm on the verge of going Hulk on this thing. I've set DDR voltage to +0.15 (don't know about +0.20 because BIOS reports my memory voltage at 1.968 or something.. +0.15 is already 2.118), MCH to +0.100, and Vcore at 1.300 (1.350 if it fails again.) If it fails after that, I don't know anymore. For now, I guess I need to wait for a new PSU from Newegg RMA and see what happens.
Sorry for the long absence, I've been tinkering with voltages all night after work.
I'll respond to the relevant posts I missed in order.
Xpoc, I do not have Everest, but BIOS reports my 12v rail at 12.63 or something. I took that bit of advice on Orthos being buggy and am running two instances of Prim95 now, and while it does run a little longer, it still fails.
Jian, I just set up an RMA with Newegg for my PSU, I need to eliminate these variables, it's just pissing me off way too much now. I am using an eVGA 8800GT at stock speeds currently. My Accelero S1 Rev 2 arrived yesterday so I'm going to attach that and OC the video card a bit after I get the damn system stable.
As far as voltages go, CPU-Z reports my Vcore at 1.232 - 1.248 when I set my Vcore in BIOS to 1.225-1.250. At Vcore 1.300 in BIOS, CPU-Z is reporting 1.296.
Speedfan reports my +3.3v as 3.28v, +5v as 4.95v, +12v at 0.06v (idle atm), -12v at -16.48v (???), -5v at -8.48v, +5v at 5.38v (Don't know why there are two), Vbat at 3.20v.
Oddly, it also says my Vcores are vastly different for some reason. Vcore1 is 1.30v, Vcore2 is 1.97v.
As far as my being silly at wanting to OC at stock volts, I wouldn't know whether I can do it or not if I don't try now will I?
nyker, I'm not entirely sure if it is the RAM now either. If small FFT's only stress the CPU, then my CPU is rock solid even at 1.225v. Memtest86+ and HCI Memtest can also run forever and not find anything wrong with my memory now. The only error I've seen so far is the failures in Prime/Orthos when running large FFTs (I set it up to use 1024K - 32768K FFT's now so I can bypass it testing all those smaller FFTs and get a fail, if any, sooner.)
My initial bootup was with the CPU at stock 333x9 1.225v and the BIOS "auto" configured my memory for me (picked the 5:6 divider iirc.) That ran everything fine.
Right now I'm on the verge of going Hulk on this thing. I've set DDR voltage to +0.15 (don't know about +0.20 because BIOS reports my memory voltage at 1.968 or something.. +0.15 is already 2.118), MCH to +0.100, and Vcore at 1.300 (1.350 if it fails again.) If it fails after that, I don't know anymore. For now, I guess I need to wait for a new PSU from Newegg RMA and see what happens.
I don't have any big video cards, but I do have several Seasonic PSUs, several 330W and one 380W, none have ever caused any problems. I would qustion this although stranger things have happened.
As for the CPU, I don't have any Intels, but my AMDs will all run drastically undervolted and still overclock a little at the same time. For example on my Biostar TA690g board, I run the 3600x2 CPU at only 1.1V BIOS settings, slightly overclocked @ 2GHz, never tried it farther, been running this way for a year. I even know my Seasonic 330W is a little low on the 12V (~11.77?). My SuperTalent 667 RAM is running default speed, default 1.8V, maybe tigher timings, don't know because I'm at work.
As for the CPU, I don't have any Intels, but my AMDs will all run drastically undervolted and still overclock a little at the same time. For example on my Biostar TA690g board, I run the 3600x2 CPU at only 1.1V BIOS settings, slightly overclocked @ 2GHz, never tried it farther, been running this way for a year. I even know my Seasonic 330W is a little low on the 12V (~11.77?). My SuperTalent 667 RAM is running default speed, default 1.8V, maybe tigher timings, don't know because I'm at work.
Originally posted by: Idontcare
Originally posted by: Owls
My screenname is extremely indicative of my overclocking prowess
I don't care!
... or do I?
I see what you mean
I totally agree with running as low as you can as soon as you get stable (no it's not silly, it lasts longer
). Good call on the RMA methinks. IF you have a chance to switch, up it to the 430/500. They're only 15/20 more respectively. I know you're not pushing 380w but you're only testing cpu/mem. Once the video kicks in you're getting close enough for my comfort to want more probably at 3.6ghz. Check here and hit the power consumption page, they used a 8800GTX but that's close to our cards (8800GT):http://www.xbitlabs.com/articl...y/core2quad-q9300.html
Admittedly worse case probably but I like overkill in the psu. I've seen too many die a 6month death because of being pushed a bit (and capping overclocking because of lines dipping ever so often anc crashing pc). They may rate a psu at 380 but I'd knock 25% off that (they're only 80-85% efficient, and as heat goes up that goes down). The only PSU I know of that rates honestly is PC Power and Cooling. They rate their psu's for the worst case scenario on the planet 24/7 and still warranty for 5 or 7 years! I love them
Another PSU option:
http://www.newegg.com/Product/...x?Item=N82E16817194011
I keep selling those, they haven't come back yet. I love the 42 amps 12v. Your model has 34 total. Quite a difference where us overclocking power hungry graphic card freaks live. The +12v is most important.
IF that +12v line is correct VERY good call on the RMA...ROFL (.06v? Sheesh). It should say 12v +/- .5 or so. When you get the PSU back you should put a load on it and check volts. I.E. Check volts in windows, then fire up 3dmark (any version, maxxed out on quality, resolution etc) and fire up prime95 to get the rest hot. Then check volts again and see how far they dipped while at load running these two apps. Or substitute any other graphically intense game etc you can find. Very weird volts on that PSU though (funky vcores too ...LOL), so check it ASAP when you get the new one. I wouldn't go 1.35v. IN fact I'd wait on a PSU. Too many things rotten with this setup with most evidence pointing at the PSU right now.
I'd advise against going hulk on anything over $50. ROFLMAO.
I mean at $300-500 you wake up in the morning (assuming the pc drove you to drink, leaving you uninhibited and hulking your pc) and wish you had your money back...:Q and wonder who the hell did that to your PC? :laugh: Now go have 2 drinks, not 12...
- Dec 25, 2007
- 1,243
- 3
- 76
Okay, sorry for the long delay in posting, I was away for the weekend at a convention in Boston.
Just got back and decided to try to narrow down my voltages a bit.
As of right now I *appear* to be stable at 3.6GHz with 1.25000 Vcore in BIOS.
I say appear because I haven't been running Prime for too long (Currently 2 hours) but I have set it to run at 4096-32768K FFTs to try to "force" it to error. The default maximum FFT size for a blend is 4096K so I hope that being able to clear that for 2 hours and going is a sign that it's stable under normal circumstances. I'll maybe re-run a normal Blend later tonight. As for other voltages, I've set everything back to normal so the ONLY voltage change is vcore from 1.22500 -> 1.25000. Any lower than 1.25000 failed prime with the aforementioned settings relatively quickly (Couldn't clear 10 passes with 4096K FFTs.)
I'm a bit bummed that I couldn't get 3.6GHz on "stock" voltage like some other lucky people, but I also am wondering what "stock" means. Some people claim they have achieved 3.6GHz on "stock" only to reveal that their vcore is set to auto in BIOS, and is thus assigning ~1.3v or so. What voltages are you guys running this E8400 on at 3.6GHz? Is 1.25000v higher than or lower than average? Or is it just about right in the average butter zone?
In any case, another question I have is, would it be pointless for me to try raising the other voltages (MCH/FSB/DRAM) just to lower the Vcore a notch or two? I'd like to lower it a bit more if at all possible but I don't see it as very likely, and really don't want to waste any more time testing right now.
I'm still going ahead with the RMA on my PSU, hopefully a new one solves the rebooting problem, but I don't have a spare so I want to clear as much stuff as I can first. I wonder if it could be possible to lower the Vcore a bit with a new PSU.
Just got back and decided to try to narrow down my voltages a bit.
As of right now I *appear* to be stable at 3.6GHz with 1.25000 Vcore in BIOS.
I say appear because I haven't been running Prime for too long (Currently 2 hours) but I have set it to run at 4096-32768K FFTs to try to "force" it to error. The default maximum FFT size for a blend is 4096K so I hope that being able to clear that for 2 hours and going is a sign that it's stable under normal circumstances. I'll maybe re-run a normal Blend later tonight. As for other voltages, I've set everything back to normal so the ONLY voltage change is vcore from 1.22500 -> 1.25000. Any lower than 1.25000 failed prime with the aforementioned settings relatively quickly (Couldn't clear 10 passes with 4096K FFTs.)
I'm a bit bummed that I couldn't get 3.6GHz on "stock" voltage like some other lucky people, but I also am wondering what "stock" means. Some people claim they have achieved 3.6GHz on "stock" only to reveal that their vcore is set to auto in BIOS, and is thus assigning ~1.3v or so. What voltages are you guys running this E8400 on at 3.6GHz? Is 1.25000v higher than or lower than average? Or is it just about right in the average butter zone?
In any case, another question I have is, would it be pointless for me to try raising the other voltages (MCH/FSB/DRAM) just to lower the Vcore a notch or two? I'd like to lower it a bit more if at all possible but I don't see it as very likely, and really don't want to waste any more time testing right now.
I'm still going ahead with the RMA on my PSU, hopefully a new one solves the rebooting problem, but I don't have a spare so I want to clear as much stuff as I can first. I wonder if it could be possible to lower the Vcore a bit with a new PSU.
Very good point. People run on Auto which Automatically bumps their voltages when they OC.
It wouldn't be bad to up the DDR voltage because most RAM can run around 2.1V. I'm not a huge fan of raising the FSB or MCH voltages. Just stay below 1.38V on your CPU and you should be fine.
You may be able to lower it on the new PSU but it wouldn't be much. ONLY if your new PSU has TIGHTER tolerances on the 12v lead (most of the power from cpu/gpu all comes from 12v). If it varies MORE than your current one then you'll actually have to raise it. That's whats so good about PC Power and Cooling. They have 3%, and on their best PSU's it's 1%. This allows you to KNOW there will be no Vdroop. The line stays at whatever you put it at PERIOD. I love them. You can change from 1 drive to 4, overclock the crap out of your cpu and still see no change. They are just that good. The 750 silencer will drop 2% but still far above the 3% they advertise. The 1%/1.5% series just rock.
Anyway back to you. Don't be bummed, that's just avg. And yes some people say stock, but their stock is NOT the same as your stock (and we're talking 3ghz not your 3.6!) or anyone else's for that matter. The worst stock on E8400 could be up to 1.36v @ 3ghz (that guy isn't getting to 4ghz+ without some serious cooling). On a E3110 it's only 1.225v. Which is why I'd rather have E3110. Even one more bump on yours would still be pretty damn good (whats that like 1.2525? or something?). Anything below 1.275 is great in my mind at 3.6ghz. Lets just put it that way The problem comes when your heat starts shooting up because of ANY bump higher. You are reaching diminishing returns then. Like if you prove its still unstable and your next volt bump causes the cpu to go up 25c yet makes it stable, I'd have to question whether I really wanted to run at that speed so badly that it's worth 25c. I hope that makes sense (it did when I thought it...LOL). It's a judgment call for sure but no worries. You're nowhere near a mistake at far lower than Intels worst shipped E8400 (1.36v). No worries None of the other voltages will help ease Vcore. You either need it or don't, raising dram etc isn't going to change a Vcore problem. Only Vcore will if needed. Same with others. If memory volts is needed, you need to raise it period. Raising MCH may ALSO be needed but there's a voltage required for each of these to NOT crash your system and having one higher isn't going to lower some other voltage on another component. Unless somebody has some proof they can point out I wouldn't buy it.
Anyway back to you. Don't be bummed, that's just avg. And yes some people say stock, but their stock is NOT the same as your stock (and we're talking 3ghz not your 3.6!) or anyone else's for that matter. The worst stock on E8400 could be up to 1.36v @ 3ghz (that guy isn't getting to 4ghz+ without some serious cooling). On a E3110 it's only 1.225v. Which is why I'd rather have E3110. Even one more bump on yours would still be pretty damn good (whats that like 1.2525? or something?). Anything below 1.275 is great in my mind at 3.6ghz. Lets just put it that way
The problem comes when your heat starts shooting up because of ANY bump higher. You are reaching diminishing returns then. Like if you prove its still unstable and your next volt bump causes the cpu to go up 25c yet makes it stable, I'd have to question whether I really wanted to run at that speed so badly that it's worth 25c. I hope that makes sense (it did when I thought it...LOL). It's a judgment call for sure but no worries. You're nowhere near a mistake at far lower than Intels worst shipped E8400 (1.36v). No worries None of the other voltages will help ease Vcore. You either need it or don't, raising dram etc isn't going to change a Vcore problem. Only Vcore will if needed. Same with others. If memory volts is needed, you need to raise it period. Raising MCH may ALSO be needed but there's a voltage required for each of these to NOT crash your system and having one higher isn't going to lower some other voltage on another component. Unless somebody has some proof they can point out I wouldn't buy it.Originally posted by: Burrich
Very good point. People run on Auto which Automatically bumps their voltages when they OC.
It wouldn't be bad to up the DDR voltage because most RAM can run around 2.1V. I'm not a huge fan of raising the FSB or MCH voltages. Just stay below 1.38V on your CPU and you should be fine.
oops...Maybe I didn't understand his post. But that's another way to look at "stock" voltage. Yes set to auto it will go up on most boards as they OC the cpu. Good point. I was just meaning stock out of the box if they manually kept it there. Agreed 2.1v, but I'd always want 2.0 for 24/7 or less if possible. But wouldn't sweat memory anyway it's lifetime..LOL. Agreed with 1.38 or less. Intel says 1.45+ will shorten life. So I feel 1.38 ok. But I'd hope to 24/7 at 1.35 or less for my chosen mhz. Or take whatever I could get at 1.35 (since they ship E8400 @ up to 1.36v it's safe). I admit I error on the side of caution where possible.
- Dec 25, 2007
- 1,243
- 3
- 76
Well, I'm getting tired of staring at Prime95 all day long instead of actually using my PC so I think I'll give it the go with this. 3.6GHz @ 1.25000 vcore is decent I guess, I'm not sure what others are using to get it as most OC threads with the pertinent info are about going absolutely nuts for 4GHz+.
It seems "Prime stable" as it can survive 3 hours of Prime95'ing with 4096-16384K FFTs and small FFTs forever. Getting ready to pull out the PSU and RMA it, but I doubt it will help too much with vcore. It might solve my random rebooting issue though.
Finally got around to installing my Accelero S1 on my 8800GT now that this is over, got that OC'd pretty easily to 700/1750/2000 core/shader/memory. I'm not entirely sure how to "stress test" that though, but it sat in ATITool for 10 minutes without a single artifact.
Thanks again guys, I appreciate all the help. Any more information is always welcome of course as there are still a few unanswered questions floating around here.
It seems "Prime stable" as it can survive 3 hours of Prime95'ing with 4096-16384K FFTs and small FFTs forever. Getting ready to pull out the PSU and RMA it, but I doubt it will help too much with vcore. It might solve my random rebooting issue though.
Finally got around to installing my Accelero S1 on my 8800GT now that this is over, got that OC'd pretty easily to 700/1750/2000 core/shader/memory. I'm not entirely sure how to "stress test" that though, but it sat in ATITool for 10 minutes without a single artifact.
Thanks again guys, I appreciate all the help. Any more information is always welcome of course as there are still a few unanswered questions floating around here.
TRENDING THREADS
-
Discussion Zen 5 Speculation (EPYC Turin and Strix Point/Granite Ridge - Ryzen 9000)- Started by DisEnchantment
- Replies: 25K
-
Discussion Intel Meteor, Arrow, Lunar & Panther Lakes + WCL Discussion Threads
- Started by Tigerick
- Replies: 24K
-
Discussion Intel current and future Lakes & Rapids thread- Started by TheF34RChannel
- Replies: 23K
-
-
Facebook
Twitter