Printable View
lol if 60 seems hot you dont want to see what it was before i RMA'd the other CPU.
72-75c under load before hand.
OCCT is about finished. 60c is as high as it got, taht was about 15 mins ago. i think the new paste has spread out a little more and im now only at 56c so its gone down. normally on these stress test it gets hotter and the end of the test.
:shocked: 75c? thats scary stuff.. only time I saw a CPU hotter than that was my old A64 3500+ when the thermal paste underneath the IHS broke down.. sucker got to 101c :p:
thats why i RMA'd taht cpu. this is awsome temps i think. with some more breakin time with this paste i think it will come down some more.
here is occt finished and got the stable sign..lol
and real temp at idle. better now than when i first installed it all
http://i72.photobucket.com/albums/i1...occtstable.png http://i72.photobucket.com/albums/i196/Cincrob/idle.png
Looking good :up: now lets see 4GHz :D
lmao.. i new that was comeing..dont worry in a couple days i will have it 4+ghz.
i never was able to get it stable above 4.1 id sya cause of the heat issue. but im sure no problem this time around.ill see how high i can get it the nback it down to 4.0 i hope for everyday running. in realtemp right now im at 36c i love it..
Nice :cool: I'll put a guesstimation of 4.3GHz on your E8400. Your job is now to prove my guesstimation wrong :D
I learned my lesson with the ds3l after the mosfets on my first one burst into flames. I put some small sinks on those and also replaced the TIM on the nb and have a small fan blowing there. Also I'll avoid running quads in it. ;)Quote:
Originally Posted by LagunaX
seeing i have it here at work and cant really get into it much... i did try it at 4.0 and it wouldnt post.ill try more tomorrow
Have at it man, have at it! Don't let that CPU win! Its sitting there right now.. laughing and giggling to itself like a little school girl.. "heheheeee! I don't POST 4GHz.. I'm not telling him why I didn't tehee!" You can't have that!
If you use RealTemp 2.75 you will love it even more (TJ Max is now right) :up:Quote:
Originally Posted by cincyrob
link me up with a link for 2.75 if ya dont mind....yea im lazy well im at work any how..lol
oh its not gonna get the best of me on the first day out... ill let it run like this for the weekend then sneak up on it like a to of bricks....
Hey guys, how you all doing?
This Gigabyte won't work over 500MHz FSB :(
I tried about a lot of combinations, high and low vPLL, VTT and vNB...
None worked, 510 ~ 520 is the maximum I could even POST.
Any advice for 600MHz FSB ?
Meanwhile, I got 500MHz stable on incredible low volts:
CPU Skew 200ps
vCore 1,23v
VTT 1.22v
vNB 1.4v
vPLL 1.5v
GTLref1 0,79v
GTLref2 0,76v
NBref 0,73v
vMEM 2,16v (1200MHz)
http://img516.imageshack.us/img516/4...oltsqa1.th.gif
well doompc 600 fsb is not eazy to say the least ! but you should be able to eek out 525~545 maybe with high multi
Here it is -» http://www.fileden.com/files/2008/3/...alTempBeta.zipQuote:
Originally Posted by cincyrob
Now you have no excuse to leave it @ 3.6ghz ;)
http://i72.photobucket.com/albums/i196/Cincrob/40-1.png
http://i72.photobucket.com/albums/i1.../40everest.png
http://i72.photobucket.com/albums/i1...40realtemp.png
well there it is. but i got problems. ive been up all night trying eveything i know(which isnt much tbh)
i try and run OCCT Orthos,even the new intelburntest i will get the windows error wanting to shut down the program.all will run for a few second then give me the windows error.
heres what ive done.
i set my FSB overvoltage control to +0.1
i set my MCH overvoltage control to +0.1
CPU voltage to 1.375v
i get it to post, windows opens i can tool around with no problems but when i run the stress test it hits the bricks.
here a screen shot of my bios.
http://i72.photobucket.com/albums/i1...b/100_3368.jpg
there is a new bios update for this board i wonder if updateing it would make a difference?
i have a GA-EP35C-DS3R mobo.just as new as my CPU.
thanks for the link to realtemp 2.75
with this CPU i notice abig difference in how fast stuff opens/loads up compared to the one i RMA'd its alot faster. but i cant seem toget this one to 4.0ghz.
i set the vcore to 1.375v in all the testing programs show actual volts as 1.32v man thats a vdrop isnt it?should i try higher to get the closes to 1.365 as actual volts?
thanks for any help guys
Does VID of 1.1v mean something or not? I've got this E0 E8500, and I just found out it has VID of 1.1v... Kinda nice imho, runs pretty cool at stock speed :) dunno about oc'ing, didn't oc it yet.
doompc, how did you calculate that 0,79v and 0,80v. are 67% of 1,32v ???
For 0,79 or 0.80v GTL you need to set 0,605 and 0,610 in BIOS...
1,32x0,605=0,7986
1,32x0,610=0.8052
instead of 1,32x0,670=0,8844
Finally got around to putting my E8400 back in service. Here's some temperature testing with the IR thermometer pointed at the IHS:
http://www.xtremesystems.org/forums/...postcount=2190
Kartoff, yeah I messed the calculations, I thought the ref values would be locked to 1.2v FSB, but after setting vFSB in the BIOS the values for GTLref are updated.
Luckily, this Gigabyte allows to set the voltages direcly (not in percentages):
http://img175.imageshack.us/img175/3786/gtlrefxw0.jpg
If I increase the vFSB, should I increase the ref values too or stick with the actual voltages (for 1.2v) ?
Thank you :)
Anyone get an e8500 from newegg with e0 yet?
can anyone tell what kind of vcore is needed for 440-450 fsb on a 8400? It seems that i cant get past 430 with the system mentioned in my sig no matter what voltages i put. Also mosfet temps reach 60+ during prime95 stressing, is that normal? thnx
Increase MCH and VTT voltage , I was able to get over 500 fsb with E8500 on IP35 Pro ;)Quote:
Originally Posted by sepultribe
Do Ace's bolt-mod as found in the ip35-pro thread. What do your other voltages look like? I don't think I have an especially lucky e8400 but mine runs 467fsbx9 (4200) stable with 1.344v load on air. What bios are you using?Quote:
Originally Posted by sepultribe
On P5Q Dlx when i change voltage i need to change GTL if i want to stay around 0.8v. So if i increase voltage i need to decrease GTL reference or to leave it untouched to get more GTL volts... I see that on GB this are solved in different much easier way.Quote:
Originally Posted by doompc
Got a good one here and it's not even an E0 stepping :D
http://img380.imageshack.us/img380/9...stclocklm9.jpg
yes thier are some good ones :) makes it hard to want an 8600 :ROTF:
I would like some guidance on how to proceed.
I have an 8500 c0 CPU with a Asus P5W DH Deluxe mb. I have the latest bios, which I think is a beta and there is a problem with early revisions of this mb not being able to set a voltage lower than 1.45!?
Anyhow I've overclocked my 8500 to 3.8Ghz on AUTO CPU voltage and this appears to be stable, however if I try to go higher then it will boot but not load Windows, which blue screens.
Now is this my memory or do I need to apply more volts to the CPU ? At the moment I can't do apply more CPU volts, so I wanted to rule out the memory causing the failure.
Is using memtest a good tool, to see if my memory is the "problem" ?
Any other suggestions ?
Most P5W's are maxing out around 400FSB with these CPU's, some have had more luck, but most users are stuck there... If you want more grab another board... and make sure that NB is well cooled...
OK it don't look like i can get this E8400 to run at 4.0ghz. well not on this board any way. taking this board back to micorcenter and swapping it out for a GA-EP45-DS3R. anyone have any luck with the P45 boards????? i can get it to go up to 3.8ghz but thats it. the vdrop is mad on these gigabyte boards. id probably have to set my vcore to 1.5 to get it running at 1.365v
tried mch up to 1.56, vcore up to 1.5 and tried boosting vtt a bit with no success at all. prime95 fails instantly after several seconds if i go 440+Quote:
Originally Posted by tomati
im using bios 16... im starting to believe that i have a defective chip. i once bumped voltage close to 1.55 when i was on stock cooling trying to get prime stable, but that was for 10minutes max and uguru never showed more than 75-80 C. would that cause degradation or sth on the chip that could justify this kind of behavior?Quote:
Originally Posted by loonym
thnx a lot for the answers
edit : this pic is from 12mins of priming @ 400fsb all stock except vcore which was 1.2850 (it would get errors after the 1st test if i set vcore lower than that). isnt this kind of vcore needed too much? and are these temps showing that im in need of doing the bolt mod on my m/b?
Hello ..
i want ask about my temp it is ok
i have E8500 @ 4.0GHz and thermalright ultra 120 eXtreme
and my ambient = 24c
with blend test core temp don't go higher 53c and small test 60c
http://img260.imageshack.us/img260/5...tempwa5.th.png
Any suggestion?
regarads :)
Quote:
Originally Posted by total90
those are excellent temps. i wouldnt worry about a thing with those. what do you idle at? around 36-38c?
http://i72.photobucket.com/albums/i196/Cincrob/38-1.png
im getting there still cant get 4.0ghz stable. but as you see with the intelburntest its fine at 3.8ghz. but i do have my vcore set higher right now than i hade it for 4.0..lol
im running OCCT right now and all is well.
i aslo updated my bios to F3 from F2. still not sure if i should trade this board in on the GA-EP45-DS3R. I STILL have 24 day to take this board back for a axchange or replacement. the P45 board is the same price i paid for this one.
I see that you use a 2 x 2gb kit of ddr2 PC6400 , do you know their potential ? may be they are the weaker point of your OC ? Did you test their max OC ?Quote:
Originally Posted by sepultribe
With 2 x 2gb kit you will have to increase the Vmch a little bit more than with a 2 x 1gb. I was on the bios 17 final when I tried the E8500, don't know if it's change something ?
thanks cincyrob for replay ...Quote:
Originally Posted by cincyrob
my idle temp are 39c
thank you
max overclock of my ram is around 1080, wasnt even close to that when i tried 450 fsb 1:1 (ram ~900). plus i had relaxed timings (5-5-5-15) and plenty of volts for the dimms (2.1) and mch (tried up to 1.56).Quote:
Originally Posted by tomati
im just trying to understand if its the pwm/nb temps thats causing all this or a dumb cpu chip.
sepultribe, your pwm temp doesn't really look that high in the screenshot, mine was hitting 90 before I redid it. But your vcore also doesn't look too high either. In the uguru screen you posted I see only 1.26 volts. Try just using small fans to cool the pwm and nb sinks. I've really only used quads in my ip35-pro so I can't help much with your settings but I like that bios 16. Maybe if you look in the big ip35-pro thread you can get some settings from there.
in that screenie i posted i was trying to get fsb 400x9 stable with my rams at 1:1. uguru shows 1.26 altho its actually 1.28 from bios. isnt that quite a lot for only 400 fsb? i've read ppl do that with stock voltages...
i'll try 440+ when the weather gets cooler again
Looks like i have just about lost my bragging rights to the E0's :p:
Is it normal that VID is switching when EIST/Speedstep is enabled? I'm running E0 e8500 and when I disable speedstep, it runs with 1.1 VID reported in CoreTemp but when I enable Speedstep, VID jumps from 1.1 to 1.2 under load in CoreTemp (on idle it stays at 1.1)... It also jumps in CPU-Z. Funny enough, it runs just dandy with 1.1 at stock speed when I disable Speedstep, load or not..
ya thats normal
Setup:
IP35-Pro w/ BIOS 17
2x2GB G.Skill DDR2-1000
Silverstone ST50EF-Plus 500W
E8400 E0
TRUE w/ low speed S-Flex
Antec P182 w/ low speed Yate Loons
So, all near-silent air cooling :)
534 x 9 = 4806Mhz SP 1M (haven't tweaked memory at all so time probably is bad)
vcore - 1.45
vtt - 1.30
vmch - 1.64
http://i37.tinypic.com/24m89kx.jpg
http://valid.x86-secret.com/cache/screenshot/412398.png
Passes wPrime 1024 and SP 32M at 518 x 9 = 4662
vcore - 1.38
vtt - 1.20
vmch - 1.48
Definitely happy with a nearly silent PC that does 4.65Ghz on air 24/7 :) All with a $160 CPU. Praise the E0 revision.
no really thats a pretty good pi time :)
eye am getting 9.838 at 4.7 speed
also edit in your batch # please
Anyone comfirmed where to buy an e8500 e0 yet?
Fry's electronics store- but you have to ask for the smaller box as they have both.Quote:
Originally Posted by scottc19
They clock very well - see here:
http://www.xtremesystems.org/forums/...postcount=4363
I guess thats the only place, to bad it would be a 4 hour round trip to go there :( I called them and they of course said if you order it it is random which one you will get :(
Somone buy one from newegg and see if its e0 or not lol
otherwise I think I will just fork over all my money and get an 8600
Quote:
Originally Posted by LagunaX
I got mine at Fry's also, in Industry Hills, CA. Great clocker, look at my sig.Quote:
Originally Posted by LagunaX
I read on another forum that at the exact same FSB and multiplier that, an E8500 can outperform an E8600 because the E8600 has more on chip L2 cache latency. Does anyone have these two processors available to do some testing on the same motherboard at the same FSB and multi with the same memory / OS / etc? Not much use looking for an E0 if the C0 performs about the same at a lower MHz.
If you see a difference in gaming or other benchmarks then try running Membench in ScienceMark 2.0 or any program that shows L1 / L2 cache latency to see if Intel has increased the latency which could negatively effect overall performance.
RealTemp 2.76 is available which gives you the option to see the Distance to TjMax in the System Tray.
http://www.fileden.com/files/2008/3/...alTempBeta.zip
better test would be E0 8500 vs C0 8500 OR E0 8400 vs C0 8400, imho...
Although if the 8500 performs better than the 8600, then it's worth doing.Quote:
Originally Posted by anzial
Quote:
Originally Posted by unclewebb
thanks for the science mark2.0 unclewebb !! yet another toy to play with :)
i doubt the cache is slower. If anything its faster but I doubt that as well. The difference lies in the yeilds. if everything else is kept constant than the only difference would be yeilds and I would put my money on the 8600.
hmmm , realtemp beta 2.76 raised my temps 5-6C is this right? or just a flaw in the beta?
Yeap,i agree,there could not be any differenceQuote:
Originally Posted by daos
E8400 Q818A027
http://pix.nofrag.com/a/5/5/048d52af...69ae352dca.png
This is with a vDrop mod (p5k & 0.04v ftl) :)
Are most people buying E8400 CPUs getting an E0 revision by now? I have a C0 E8400 that sucks at oc'ing and Fry's has a special right now.... thinking of getting another, but only if I can be (almost) guaranteed to get E0.
There are no guarantees, but Fry's got pretty nice return policy.
my local microcenter has 8 of the SLA9P E0 stepping and the reg slapl C0 so unkess you can look and pick which one it is a toss up for what ya get.
oh got this bad boy up to 4.0ghz now. im figureing out this mobo.. taken some time though
fry's website has e8400 online for 139. Whoa - cheap.
You want an EO guaranteed? Here you go:Quote:
Originally Posted by Extelleron
http://www.tankguys.com/product_info...oducts_id=1828
are E8400 EO comming in a smaller box too????
are you guys all disregarding your CPU (i'm guessing it's T-Junction) temp on your Wolfdales? i'm sure this has been answered a million times, but a google search turned up a bunch of nubs on other forums guessing at the correct answer.
i ask because my CPU temp is regularly 10° higher than my individual core temps. this screenshot is from Everest Ultimate 4.60.1500 final.
http://i21.photobucket.com/albums/b2...01/cputemp.jpg
everest has never read any board eye have had correctly & the few times it did work the numbers were mixed up
Everest reads my individual core temps the same as CoreTemp. it's just the "CPU" temp that bothers me. i ran Prime the other day while it was very hot (90°F outside and all my windows open) and my CPU temp hit 70°C. i'm assuming that is my T-Junction, which is rated at 105°. the individual core temps are the ones i'm guessing are safe up to 72°. the cores stayed below 60°C.
Just got a E8400 from NewEgg today that I ordered 9/3 G0 SLB9J :)
@HDCHOPPER, yep:
http://img255.imageshack.us/img255/5346/caixa2sd3.jpg
oo man cool ! thats gonna make it a lot eazer at fry's now
yea the HSF in there is as big as a golf ball...lol not that we use them fans..lol
well guys i did it. got my E8400 co stepping stable at 4.05ghz.. took some doing but i finally got it. i was challenged to get it to 4.3ghz i dont think it will do it. but im gonna go up a bit at a time and maybe hit 4.2ghz.here is a shot of intelburntest..man i love that program
http://i72.photobucket.com/albums/i1...burntest-1.png
checkout cpu-z validation in my sig.
http://i72.photobucket.com/albums/i1.../superpi-5.png
http://i72.photobucket.com/albums/i1...sandramips.png
http://i72.photobucket.com/albums/i1...t140stable.png
yup im a happy boy...lol suprisingly my temps have only gone up a couple degrees with the 4.0 OC...at idle that is
i think my CPU would melt if i ran it at 1.39 vCore. i'm able to run at 3.9 GHz on just 1.30 vCore (1.27/1.25 actual). overclocking further requires a lot of extra voltage for mine. even at just 3.9 GHz with a decent air cooler, Linpack pushes my CPU to the very limit of what it can handle, heat-wise.
here is my e8500 533x7.5 1.29v
http://img396.imageshack.us/img396/4...torthosfb2.jpg
E8500 C0 Q816A484 (VID 1.1500)
http://www.abload.de/img/4ghz-prime-02gll.png
Chri$ch where do you get all these awesome chips?
i order them online at different stores, it is only luck :D
I need some of that luck! Your E8600 is a gem and that E8500 looks pretty sweet as well!Quote:
Originally Posted by Chri$ch
i second that question....??????Quote:
Originally Posted by zlojack
is there anything other than bumping up the vcore to help get it stable????? i hear alot of you guys talk about the CPU PLL..... what does that do?should i mess with mine? i have my vcore in bios set for 1.44325v to get 1.36-1.39v
i think i can get more out of this chip or even better lower volts if i totally knew what to adjust...lol
xMrBunglex
i agree wit hyou on the temp with intelburntest. it does push up the heat when testing...i will do 5-20 test on it then if it passes that then i will do the normal testing we all use. if it wont pass intelburntest then i no it wont pass the others.
Not true at all.Quote:
Originally Posted by cincyrob
My E8600 could pass several passes (5-10) of IntelBurnTest but failed Orthos Small FFT.
I had to raise vcore from 1.34 to 1.36 for Prime small fft stability on my E8600.
Funny enough, my E8400 was the opposite. Primed all day and night no problem, but wouldn't pass IBT (this was at 4.2GHz 1.36v)
Someone at anand just got in on the $139 Fry's E8400 deal, and it was an E0 stepping!
hey guys i got a xeon 3110. i havent gotten to play with it yet. i killed another mobo. owell i have been thinking of replacing my new ds4 anyway. im am told that this xeon should be able to do 4.3 @ 1.36vcore. now i just need a mobo and ram. lol. ill post some screen shot once i get up and rolling
Quote:
Originally Posted by zlojack
thats just it for me. if it dont pass IBT i know it wont pass orthos. at least it hasnt so far.
question.? is 1.36-1.39v to much for these E8400 co stepping? will the degragation start on it at those volts? i do need this puter/CPU to last a few years. so as much as id like to keep it at 4.0ghz i dont want to fry it up like a sunday egg...lol
this is Xtreem forum .. really if you are talking YEARS then run stock or whatever you can get out of it with the stock volts the bios boots with stock
most of us dont care if it frys within months so long as we can run 5.0:D
It would be nice to have an answer to the question, though. I know nobody can say exactly how long a CPU will last, but it would be good to get at least a couple of years out of it. It seems nobody can say exactly how much voltage is too much, because I see the question being asked constantly and there are never any logical replies.
yes eye did reply to cincyrob's question ... leave the volts stock and clock up to what you get in cpu speed with that
(stable of course) thats the only really safe way... intel says you limit yourself to 1.364 but thier thay dont care about the life of your cpu... feel daring then use 1.364 as your base line and see what you get then then your KINDA safe
good luck
Intel specs the absolute max vcore and max vtt in their product sheet as 1.45VQuote:
Originally Posted by Dee
Is that right, Gurusan? I read somewhere that they said no more than 1.3625 max voltage, which is in line with what HDCHOPPER said.
Just for fun I'm currently underclocked. I am at 2.56GHz on 0.968V. This maintains a decent RAM speed over DDR1000 and the machine is still fast! It's way quicker than my old AMD 4800+ X2 at 3.00GHz. I ran Prime and the highest temp for each core reached 30/35. :D Does Pi 1M in 18 seconds. For comparison, my friend's Q6600 takes 21 seconds at stock.
Any less than 1.00V set in BIOS and the machine won't boot.
for reference Intel specs the Q6600 max vcore and vtt at 1.55v......and we all know those are totally fine for that voltage as well as even higher.
http://download.intel.com/design/pro...hts/318732.pdf
http://img352.imageshack.us/img352/3901/42307277vn4.png
Those are ABSOLUTE max and min.
There's another document that says 1.36v is the "safe" limit.
I am going to go higher than that once I redo my loop and my CPU gets its own rad, because right now my temps get a bit high when I pump over 1.4v into it.
As Ivan Drago said: If he dies, he dies.
I tried 1.4V and 4.3GHz on this E8600 and it just won't stabilize. 4.25GHz and 1.376 is Prime stable, so what gives? Another 50MHz and a fairly big jump in vCore, yet it errors out in Prime with hardware failure messages. Is it time to mess with other BIOS settings, such as GTL, PLL Termination, etc?
Thats how most chips overclock, you get to a point of diminishing returns. I can honestly say that with the right cooling and lapping all chips will hit a point where they need alot more voltage to move up slightly with prime stability. Its up to the user to decide if the extra voltage and heat is worth the additional 50 - 100mhz. What you are describing happens with every CPU I've ever used.Quote:
Originally Posted by Dee
With the 4 E8400 C0s I had, getting 4.1 was easy, 4.2 need alot more volts and got alot hotter. One would go all the way to 4.3Ghz. Again needing alot of volts to pass that point though.
The quad in my sig will go up to 3.74 @ ~1.38v - to hit 3.8 I need ~1.46v and it runs alot hotter. A full tenth to move 50 more mhz. After that I cannot maintain prime stability.
On my ds3l 4.4ghz on my e8500 e0 requires +1 on fsb termination voltage and 1.4v on the northbridge with active northbridge cooling.Quote:
Originally Posted by Dee
On my asus p5k 4.2ghz e8400 C0 requires 1.4 on the northbridge too.
Funny though on my Mars board I can set the northbridge to 1.2 for 4.3ghz on my e8600es.:shrug:
Quote:
Originally Posted by Dee
ya my bad it's 1.3625 my meds were kickin in and the memory is going
thats why eyem moving up to Domminators 8888 cas4 :rofl:
Quote:
Originally Posted by gurusan
65nm is soooo much forgiving than 45nm on voltage:D
but this is XTREME FORUM'S ppl rock em till thay burn or untill ya see the magic blue smoke :rofl:
Think of it this way. A lot of folks here are pushing these things hard. If they were dying from it I think there would be more posts to that effect.Quote:
Originally Posted by Dee
I got a 8500 for 170 at Newegg and it was E0, got it yesterday
Nice they are $189 now.Quote:
Originally Posted by Quad-Damage
Anyone have results from an E0 E8500 Q822A yet?
Mine is in the mail from SoCal Parcel post - could take another week to get here. :mad:
q822a here:Quote:
Originally Posted by dnottis
http://www.xtremesystems.org/forums/...&postcount=288
Guyz, I juz bought OEM of E8500 batch no Q815A570. Any idea if this is a good chip?
Here's some results of my E8500 Q814A156
tests and screenshots
1.610V @ 5002MHz suicideshot with +25C water.
1.58V @ 4750MHz SP32M (water +25C)
1.490V @ 4503MHz SP32M (aircool)
1.344V @ 4303MHz SP32M (aircool)
Another reference for E8500:
A quick test on the Biostar Tpower I45 motherboard.
E0 Stepping
Q822A535
http://www.xtremesystems.org/forums/...31#post3267731
boble76, thats nice! hyvä suomi :D