Is there any v-drop mod yet for this board ?
Printable View
Is there any v-drop mod yet for this board ?
What is your Uncore Mult at? It should be 24 or higher.Quote:
Originally Posted by FuriousSalesman
If you can swing it, Try this.
160 Bclk, 21 CPU Multi, Turbo OFF, 48x QPI, 20X Uncore, 10X Memory
That will give you 3.36 GHz and 1600 Mhz for memory.
you may need to raise Vcore a bit, and Qpi.vtt a tad.
this config should give you a good stable run. You can even go with 21x uncore if you wanted to.
Yeah, not a total BOOM, just a freeze. Like the bios is stuck in a ZOMGWTF loop and can't display the clock :yepp:Quote:
Originally Posted by ReverendMaynard
Update on my progress (i know i should take pics like ya'll but i'm lazy):
Well i was finding max cpu clock. Was stable @ 180 x 23 (permanent turbo on) for 4140mhz and 1.33vcore (1.35v bios) (may be able to lower voltage).
Next step up was 185bclk, for 4255mhz, and damn! it's needing a LOT more volts for the extra 115mhz. My history since stable 180 is:
180 (4140) 9hr 2min stable
185 (4255) BSOD
set 1.362v bios (1.344v actual) = 185 10 mins prime BSOD
Wanted to see if higher bclock with lower multi made a difference here without upping volts:
202 x 21 (4242mhz) = 10 mins blend then BSOD
No cigar!
Back to perma-turbo 23x, and upping volts:
23x 185 1.375v (1.36v real) = instant BSOD. More unstable than before, weird!
set 1.39v bios (1.376v real) = BSOD
set 1.41v (1.392v real) = system freeze after 1hr 30 min blend
set 1.44v (1.424v real) = BSOD after 3hr
Latest setting is 1.46v bios, but for some reason after this bump up i'm still @ 1.424v real. Running another blend at this setting anyway. This is nearly a whole .1v increase for a little over 100mhz clock gain, and while my temps are nice and sub-60 i very much doubt i will settle on that for final clocks. The cpu is definitely reaching its peak here, i'm not QPI, uncore or memory bound.
Vcore Read point.Quote:
Originally Posted by ante_ante
http://img208.imageshack.us/img208/8580/img1167tm5.jpg
Vdroop Mod. 10K or 20K VR set to 1K works great. (Sorry, I'd already put nail polish on it before I took the picture.)
http://img509.imageshack.us/img509/3171/img1171gu0.jpg
Thanks for that pics. Maybe i am a noob, but why nailpolish ? Just for extremebench ?
condensation proofing :)
I've lost my RAID0 twice so far :(Quote:
Originally Posted by ReverendMaynard
Two Windows reinstalls in the past 36 hours. :yepp:
thats the reason why i have switched from raid to singel drives. ;)Quote:
Originally Posted by zlojack
Acronis TI+ NAS box FTW ;)Quote:
Originally Posted by Hornet331
Hi
i flashed with new F4N bios today (UD5) , so far no problems
Quote:
Bios F4N (UX58-UD5) Download
Various optical BIOS improvements ...
New beta BIOS fixed some module if oc 2100MHz can't set memory
timing by manual, also this BIOS add QVL of Corsair 1866 3CH.
regards
great
the bios is here finally :D
anyone find that they need more vcore with f4n than f4k to get stable at same speeds ? or is it just me :confused:
think im gonna go back to f4k
hiQuote:
Originally Posted by freshy
i use the same vcore with F4k and F4N (didnt tested the F4L )
egards
yea but im talking stable , as in 10 to 15 hours stable with prime not 5 minutes prime or playing farcry :D
i usually dont run more than 1 hour prim95 (for my daily use i Know if the system runs about 1hour prime95 i wont get problems) but we all use our systems in different ways/aplications/and so on (i only play 1/2 hours day , Benchmarks , internet , etc) i dont BOINC or Folding !Quote:
Originally Posted by freshy
Here a SS ~ 1H prime95
http://img88.imageshack.us/img88/361...estavelga0.jpg
regards
Now all we need is a new bios for EX58-Extreme as latest bios f4j isn't stable.
Attention'' Extreme ownwers UD5 bios do not work with Extreme boards as they are different
Quote:
Originally Posted by X.T.R.E.M.E_ICE
will release new beta BIOS lafter , both for Extreme an UD5
improve some auto-tweak, memory oc, and warning page.
Thanks champ.Quote:
Originally Posted by hicookie
:up::up::up:
Awesome!!Quote:
Originally Posted by hicookie
F4L and F4N have my multi locked up at 22 on my 940, whether or not I have turbo on.
I absolutely cannot get past 180 mhz bclck, even up to vcore of 1.525 :shocked: ...other voltages I have taken up to (and a bit past) settings I have seen others have here and in the i7 overclock thread. I dont know if I havent already hurt this cpu, or if it is the bios, or why the reason is that I am locked in at 22x.
It feels like I cant get the qpi over ~3300 but without being able to change the multi its hard to tell. I need a x30 qpi setting. Running qpivtt up to 1.535 and qpipll up to 1.3 did not do anything to stabilize it (at 181 bclck).
So I am stuck at 4140 MHz...without HT, b/c even running that I cannot get it stable at up to 1.525 vcore
I think I found my 24/7 clock, lower than I wanted, but hey can't have everything :P Have to do the memory settings next.
180 x 22 Turbo ON (so x23 or 4140 mHz)
QPI x36
Uncore x13 (12 seems unstable)
Mem x6
Vcore 1.4875 :eek: (any lower BSOD during linx)
QPI/VTT 1.34 (see vdimm)
CPUpll 1.88
QPIpll 1.24
IOH Core 1.4
ICH I/O 1.58
ICH Core 1.22
Vdimm 1.8 (older ddr3 ram)
No screenies, but with the high vcore there is nothing new to spam the thread with ;) Temps are at about 80C under load (linx) and 70-75C(prime), with vcore 1.456 load. LLC and turbo are on, HT is OFF. Cooling is a true black 120 (lapped) with a scythe ultra kaze 3000 zip tied to it.
At 1.525 vcore I still cannot turn HT on or go to 181 bclck (even turning turbo off), and Load temps are in the 90C range.
Note on raid: get a separate controller card. Intel/mobo raid is free, and you get what you pay for. I have had no good experiences using built in raid. Just my $.02 ;)
i'd suggest a 2K or 5K for better control though :up:Quote:
Originally Posted by Deanzo
i'm using 2K myself :)
that's normal mateQuote:
Originally Posted by Isepick
you will never be able to select 23x multi in bios on 940
overclocking nehalem in particular is a LOT more complicated than just mimicing other usersQuote:
Originally Posted by Isepick
i spend a good 4 serious testing/overclocking sessions before i could get my head around basic settings and OCing.....it takes time and patience to get it right and voltage tweaking is important but upping voltage does not mean you are going to gain stability....it actually needs to be the right mixture of volts for things to work smoothly
i would suggest higher QPI/VTT >> somewhere 1.5 or 1.55 particularly if you are going to run vdimm at 1.8v or moreQuote:
Originally Posted by Isepick
for high NB speeds you will have to jack up uncore and RAM volts even more
Quote:
Originally Posted by Isepick
sounding like a voltage tweakin issue
start with what i suggested and do some more experiementing on your own
how are you setting RAM timings in bios....try auto and manual set...use latest Memset for nehalem to check values >> there are some issues with reading subtimings properly...software will read them out +2 on most timings
I am having a hard time getting my system past 2.8Ghz maybe I am just doing something dumb, here are my settings + relevant configuration. I would be greatly appreciate any suggestions. I would like to hit a bclk of 175 for a stable 3.5Ghz rig if possible.
Hardware:
12GB OCZ Gold 1600 (8 timiings)
i7 920
GA-EX58-UD5
Settings:
BIOS: F3
Turbo: enabled
EISTE: disabled
C4/5/6: disabled
LLC: disabled
CPU multi: 20
QPI: 36
Uncore: 16
Mem: 8
CPU voltage: 1.26875
QPI/VTT voltage: 1.15
CPU PPL: 1.8
Mem Voltage: 1.64
Bclck: 140
SPD: manual [8-8-8-24]
IOH/ICH voltages: auto / optimized defaults
If I am missing some relevant setting you need to see let me know what and I will add them.
Sorry let me clarify...22 is the only multiplier I can select, no less, so I cannot try say, 21x200. I have not seen anyone else having this problem. The moment I change it in the bios it flips right back to 22. So I have only a 22 or 23(with turbo) multi right now.Quote:
Originally Posted by dinos22
I agree, I meant I was looking to them to see how high some of their voltages were without frying their cpus, so I could freely work within those ranges.Quote:
overclocking nehalem in particular is a LOT more complicated than just mimicing other users
185 bclck x22 (turbo on), vcore @ 1.525, qpi/vtt @ 1.54 = no post (all other settings same as above)Quote:
i would suggest higher QPI/VTT >> somewhere 1.5 or 1.55 particularly if you are going to run vdimm at 1.8v or more
for high NB speeds you will have to jack up uncore and RAM volts even more
Settings are 9-9-9-24 with trc at 34 and the rest on auto. Running a bit loose on timings to get the cpu squared away first. Memset annoys me since it only runs at 96 dpi, which is really tiny at 1600x1200 res ;) Ram is 4x1 Corsair TWIN3X2048-1600C7DHXIN 7-7-7-20 @1600 1.8v ...loading the xmp profile sets the vdimm down to 1.55 I think.Quote:
how are you setting RAM timings in bios....try auto and manual set...use latest Memset for nehalem to check values >> there are some issues with reading subtimings properly...software will read them out +2 on most timings
BTW Dinos, as far as I am concered you are THE authority on OCing GB boards. Thank you for your advice.
:up:
OK, I've had my X58 extreme now for almost a week without issue. Today I put on my cpu waterblock and test my 3.6GHz OC I had on my stock cooler and it was fine and just under 70C after an hour or so.
I reboot to try to get a 4.0GHz OC and it gets stuck in this auto-reboot mode. It won't even post a post code to figure out what's up!
I try cmos clear, quick unplug power cord with no luck. I finally leave the power cord unplugged for a while as I surfed for some answers on my other rig. I plug this back in and viola! I'm up and running. I decide to upgrade to latest BIOS as I was on F3, so I download and copy to flash drive. I reboot to flash and get stuck in auto boot mode.
I power down 10 minutes or so and I get back in, load optimized defaults and reboot to flash without my overclock and the stupid thing auto-reboots in less than a minute of being turned on. I power down and get back to flash latest BIOS and it still auto-reboots every time I warm boot or don't wait long enough on a cold boot, but it always stays running without error on first boot?!?!!
HELP!!
-=Mark=-
Aw man thanks:up: Sooner the better.Quote:
Originally Posted by hicookie
Wee update:
Got 185x23 (4255mhz) stable with 1.475v set in bios.
Anyone else getting really random vdroop? Setting 1.44v in bios gives me 1.424v in cpu-z under load, setting 1.46v in bios gives the same. Setting 1.475v in bios gives 1.456v real @ idle, 1.44v under load. Still that means i've nailed 185x23 =4.25ghz 10hr prime stable with 1.44v, which is not bad i guess. Now going for 4.3ghz with 23x187!
yep i get that Xello , and good oc
Glad to know it's not just me, and thanks unlike with my QX9650 i appear to have avoided getting the highest volt-hungry chip in the world :)
I tried taking loadline calibration off just now, out of curiosity, and @ 1.475v set in bios i proceeded to drop to 1.39v under load :shocked:
Pretty crazy, i guess i'll leave it on :D
Grrr, upped the bios voltage to 1.48v and it's STILL setting @ 1.465v idle and 1.44v load... This is doing me head in.
Anyone else noticed the somewhat mismatched prices of the new chip range vs old penryn range? It's still something like over Ģ1000 for a QX9770, yet a Ģ200~ 940 will both out-perform and out-clock it :clap:
Quote:
Originally Posted by Isepick
Hi
yep , i aslo noticed that , with F4L & F4N Bios we can only change CPU multi in Advanced CPU features (and need to be the last setting before SAVE)
regards
I always flash, load optimize defaults, then manually put in stock memory speed/timings/volts then save and boot to avoid reboot hell. Sounds like auto settings in bios dont work to well with something and likely the ram. I would try setting correct memory values first.Quote:
Originally Posted by MarkJohnson
I imagine it takes time/difficulty for bios to be capable of making auto settings work on all ram comers, especially this early in game. My OCZ rated 1600, cas 8, 1.65v was auto set to 1.5V Cas 6 1200. With my corsair, auto was getting closer to stock.
man, this thing is freaking me out, from f3 bios i flashed to f4k, got it stable @ 4ghz @ 1.375v vcore and qpi/vtt 1.315v, also got 4.1ghz stable , 4.137ghz stable, and then 4.2ghz crashed after 8 hours 38 mins of prime, then i flashed to f4n , and i couldnt get it stable , or it needed more vcore then with f4k,
now i went back to f4k and it needs more vcore than before to get stable , wtf is going on :shrug:
so now ive come back to f3 just running this again and then i might go back to f4k again:shakes:
is it posible i stufft up my cpu ?
i think i had the same issue until i disabled turboQuote:
Originally Posted by Isepick
maybe it was something elseQuote:
Originally Posted by freshy
you ambients perhaps or your mount has worsened or something not sure
Thank you for that workaround, I thought I was going crazy :p:Quote:
Originally Posted by mascaras
This has allowed me to test some things out, and disprove my theory about the 180 bclck being my ceiling. I can do 200x19 no problems, 200x20 bsod during linpack (at 1 notch below 1.55 vcore!) So I am stable at 180x23, or 200x19, but no higher regardless of Vcore (which suggests something besides cpu holding me back), yet my QPI/Uncore is running faster at 200 bclck than 180, which in turn suggests that QPI/Uncore is not the cuplrit.
I know they are all interconnected, and I (think) I have the basic grasp of how it is all supposed to work together, but this still makes my head hurt :confused: and I am not a stupid guy, really ;)
Hi Guys,
I'm organizing the MIPS Freezer Set Group Purchase for the UD3 and I received a mail from Michael (MIPS) saying that the Gigabyte UD5 X58 Freezer Set POM will be ready next week (1 piece block + mosfet 1 and 2 blocks).
I was wondering if people that have UD5 wanna join us for the same order to get more discounts.
He sent me these pics:
http://i43.tinypic.com/2ahkap.jpg
http://i44.tinypic.com/24cgnzn.jpg
If interested please reply on that thread.
Regards
my temps are still the same, when i was stress testing with f4k trying to get it stable @ 4.2ghz , i pushed it to
cpu vcore 1.55v
qpi/vtt 1.535v
cpu pll 1.88v
qpipll 1.18v
ioh core 1.24v
ich i/o 1.58v
ich core 1.22v
vdimm 1.66v
qpi link was x36
and uncore was running at 3400mhz
the pc rebooted after 8 hours and 38 mins, could running my voltages that high for that long stuff things up ?
^^Did you load optimize defaults when you switched bioses back. I forgot to do that once going from F4 to F3, so was still running with some F4 values which was screwing things up.
yep, i always load optimize defaults after flashing bios, i think it wants 1.4v vcore now instead of 1.375v for 4ghz to be stable :shakes:
but thats crap because i had it 13 hours prime stable @ 1.375v before proof is in my sig :down:
im back on f4k again now
edit ....thats with HT on , turbo on
update new BIOS , F4M for EX58 EXTREME
most problem is VT function, some guys can't run VMWARE when turnon VT
of i7, F4M is fixed this, also this bios improve some other oc and warning page
or multiplier reset when xmp work.
F4M for GA-EX58-EXTREME
http://rapidshare.com/files/174466209/EX58EX.F4m.html
same issue as extreme ,
this one is for GA-EX58 UD5
F4R for GA-EX58 UD5
http://rapidshare.com/files/174467179/EX58UD5.F4R.html
beta testers let me know how this works ;)
well i wont be testing it , lol not after f4l and f4n, il stick with f4k for now :down:
Thanks i will test it.Quote:
Originally Posted by hicookie
Ok just tested the bios has some nice new features but it is un stable compared to .f4j.
I think i might be moving back to F3 bios as that was the only bios that i could prime for hours on end.
@hicookie thanks for trying but it seem that i am loosing faith with this version of EX58-Extreme Motherboard.
sorry,bro which part you mean not stable? the CPU oc maximum 24/7 or?Quote:
Originally Posted by X.T.R.E.M.E_ICE
i would like to figure out.
For me this is normal. With my previous rig, Rampage Extreme/Q9550, during an entire testing week priming had a reboot at 15/16 hours. I tried with Linpack and the systems reboot into the first 5 minutes. I need to climb the Vcore three notch more for true stability. I think that a rig is prime stable only passed 24H, or 10h full mem + 10h small ftt.Quote:
yep, i always load optimize defaults after flashing bios, i think it wants 1.4v vcore now instead of 1.375v for 4ghz to be stable
but thats crap because i had it 13 hours prime stable @ 1.375v before proof is in my sig
One of the most ugly problems in this version is that coming to market many boards with a tiny capacitor blade off in the back of the mosfet circuit. Either it off when you're manipulating the board, which was what happened to me.Quote:
@hicookie thanks for trying but it seem that i am loosing faith with this version of EX58-Extreme Motherboard.
I'm just about to flash it on mine (X58 Extreme), will report back soon
happend to your ex58 ?Quote:
Originally Posted by vega22
Hicookie the problem i had with f4n for the ud5 was i lost stability for 24/7 OC is that fixed with the f4R bios ?Quote:
Originally Posted by hicookie
all my prime runs have been small ftt, you would think 14 hours of prime small fft is stable, and my mem hasnt even been running at 1600mhz and i do have it at timings 9,9,9,24 1.64vQuote:
Originally Posted by vega22
do you mind to send me the BIOS profile? let me check it with F4n.Quote:
Originally Posted by freshy
kechinhua@hotmail.com
Maybe it isn't your bios that is unstable. I tried F3 and it isn't stable any more as well. So now i am back to f4j bios seems to be the most stable for me so far. I was running prime to test f4m bios at my normal 4ghz settings but it BSOD after 2 minutes. I will do more testing tonight. I will show some bios setting screen shots later.Quote:
Originally Posted by hicookie
Of course. Too in this thread to Bobbylite. My reseller in Spain confirmed to me yesterday that he had received a batch of x58 Extreme with this problem.Quote:
Originally Posted by hicookie
(sorry by my english)
My X58 setup with 12GB ram passed 53h of prime now.. guess that's enough :D
I won't update bios from F4J though if you guys are having trouble with later versions - also I don't see any bugs with F4J, Raid 0 with SSDs is working, boot menu is working (it wasn't with F3), and the mobo doesn't forget my OC when disconnected from the wall - nice!
Voltage:Quote:
Originally Posted by hicookie
1.375 vcore set in bios
1.315v QPI/VTT
1.64 Vdimm
All other voltages Auto
19x191
Turbo enabled
HT enabled
x36/x18/x8
PCI-E 101
CPU clock drive -800mV
PCI-E clock drive -900mV
i was 13 hours prime stable with f4k flashed to f4n and wasnt stable anymore, wanted more vcore so i flashed back to f4k and at the same settings it wasnt stable so flashed back to f3 then back to f4k :eek:, so now im 5 hours into prime @ 1.38somthing volts im going to let it go for at least 15 hours then try lower vcore again, when i stop it i will save a bios profile as i had it the f4n bios and send it to you
edit.... and in there somewhere i tried f4l but that wasnt stable for 24/7 oc either
In my case I tried in blend, but anyway priming led to a reboot again and again around 16 hours because Vcore, not the ram..Quote:
all my prime runs have been small ftt, you would think 14 hours of prime small fft is stable, and my mem hasnt even been running at 1600mhz and i do have it at timings 9,9,9,24 1.64v
I think the most adequate current stability standar is: Vcore and Vdimm Linpack passed, NB/MCH/VTT/Qpi OCCT-Prime95 test passed (RAM mode)
Great! I like to read something like this...Quote:
My X58 setup with 12GB ram passed 53h of prime now.. guess that's enough
I am flashing to new extreme bios, will see if it changes anything on mine.
I flashed back to F3, and for raw overclocks seems to do alright.
i7 940 @ 4976.47
http://valid.canardpc.com/show_oc.php?id=465180
SPI32M - 7m 40s 266ms - i7 940 @ 4799MHz
http://img220.imageshack.us/img220/3...s266bp6.th.jpg
Try Linx (or any other linpack client.).Quote:
Originally Posted by jcool
My rig was 8h prime stable... crashed after 1h linx. :p:
Needed one notch more on vcore and now it also runs 8h+ linx. :yepp:
Thanks for that:up:Quote:
Originally Posted by hicookie
Mine too, the top middle of the backside is missing a resistor. Was in the bottom of the mobo box when I opened it :shakes: no problems yet thoughQuote:
Originally Posted by hicookie
So the consensus seems to be to stick with F4j?
That's the bios i flashed when i got the board and i haven't had a single problem in a few weeks of 24/7 torture testing.
Is it fair to say that those of us without problems shouldn't expect any more out of their chips from a future bios?
My chip is 23x187 4.3ghz stable now, with 1.456v. I have to set 1.493v in the bios though, which puts me @ 1.472v while idle and as i said 1.456v under load.
If there is one thing that would make me flash a new bios it would be a fix for this terrible vdroop :mad: Any chance of that?
This file is neither allocated to a Premium Account, or a Collector's Account, and can therefore only be downloaded 10 times.Quote:
Originally Posted by hicookie
This limit is reached. :(
no worry
mirror here :D
http://rapidshare.com/files/174584700/EX58EX.F4m.html
thanks for that hicookie
f4j has been 100% solid for me so far. I'll be able to shed some light on higher clocks with it with the GTZ coming either tomorrow or Monday.Quote:
Originally Posted by Xello
You'd rather have the bios overshoot your voltage to maintain a constant? You're only getting a .016 droop. That is awesome. The vdrop is .021v which is easy to compensate for. Suck it up or hard mod it ;)
It already ran like 10 hours IntelBurntest using 11,5GB ram, no problem. However, linpack currently doesn't really work for i7 CPUs, as it only runs 4 threads (CPU cores at 50% all the time). From what I gather this is intentional, since Intel officially released an "i7 compatible" linpack version - guess they are afraid of people burning out the CPUs using their crappy stock coolers :rofl:Quote:
Originally Posted by Hornet331
CPU is way below its limit anyway, but I can't go higher on the BCLK without using enormous amounts of QPI/VTT voltage, which I won't 24/7 (this is a customer setup after all..)
Settings/Hardware:
- EX58-Extreme Bios F4J
- i7 920, turbo enabled
- 21x170=3570 Mhz, 1,21V load Vcore
- 6x2GB Compustocx DDR3-1600 CAS9 memory, Mem divider 8x, CAS 8-8-8-20 @ 1,60V Vdimm
- Uncore auto (lol I know)
- QPI x36, QPI/VTT voltage 1,455V
- strap setting "Standard"
- all other voltages on auto
Running Vista Ultimate 64, one GTX 280, 2x 128GB Compustocx SSDs in Raid 0 (does around 260MB/s read on the ICH10R) :up:
Also released BIOS F4l for Extreme (TT Forum)
I'd rather set 1.45v in bios and have an actual 1.45v while idle and also 1.45v under load :D I'd love to mod it but i'd end up breaking something..Quote:
Originally Posted by ReverendMaynard
Right now i set value A, get value B at idle and value C under load.
But if they could do an overshoot instead of an undershoot and i could set 1.41v and get 1.45v idle + load, that would be sweet too. It's not the bios offset that bothers me, it's having a reasonable 1.45v stable setting under 100% load but then 1.47v+ while idle :down:
Anyone try F4R? let us know how it goes.
I have flashed to F4R and i can run the exact same settings that i used on F4L atleast. Slighty less volt i think.
I am using F4M for extreme, just ran prime 6.5 hrs, first 2 hrs were combined with Linx max memory (that will usually crash my OC if its going to crash, rather than waiting for prime reboot at 8 hr mark)...stability seems same on mine. F4m has pink "warning font" when you get into higher volts. When manually entering mem settings, 8,8,8, instead of scrolling for CAS entry, you can hit 8 enter, which is faster imo for manually entering instead of scrolling. Couple other minor changes. But works fine for me. And does anyone run their ram with different main timings in different 3 channels? Would like to manually enter in 8,8,8, 24 or whatever once, not 3 times.
Also, wish they would start vcore at like .8-.9v. Not supposed to use lower than that anyways...getting tired of scrolling through a mile of useless volts.
I second the vdroop issue. I wish they would make vcore bios = vcore idle = vcore load. Then just tell us the approximate overshoot volts (percentage/formula whatever) over in comments somewhere.... instead of making it bios = max overshoot volts, feed your idle cpu way more vcore than it needs and cut off the juice when you need it at load. If it wasnt for LLC, I would mod every board. LLC is tolerable enough to avoid modding for me anyways, but could be better.
You know you can type in the number instead of scrolling, right? Well, at least it worked with F4j.Quote:
Originally Posted by rge
"Would like to manually enter in 8,8,8, 24 or whatever once, not 3 times."
QFT
Anyone know the change log for F4R???
I'm guessing it's what was posted at TT.Quote:
Originally Posted by GAR
That was posted under the F4R and F4M update.Quote:
Most problem is VT function, some guys can't run VMWARE when turn on VT
of i7, F4m is fixed this, also this bios improve some other oc and warning page
or multiplier reset when xmp work.
F4R seems ok
how running @201 x 20 plus tubo enable(21 x 200) Ht disable
change water block to ek supreme
got fedup waiting for d tek to produce 1355 mounting kit
temps ide 39 - 40
man these water cooling companies really :banana::banana::banana::banana: the bed with the 1366 introduction.Quote:
Originally Posted by beast200
I just ordered the GTZ and the bracket is $13 bones extra. Pitiful.
What are you're loads like? BobbyLite modified his dtek with the pro-mount and it works extremely well.
4 weeks later, ?30 bios resets clearing cmos/bios trials ....no, I didn't know, thanks for telling me. Just tried it, works for VTT and other volts too...guess I should apologize to all GB bios writers for all the non-abridged dictionary words I thought about them while endlessly scrolling in their bios:p:.Quote:
Originally Posted by Lukee
Hehe.. My old P45 board allowed entry. I was used to it and tried it when I first got this board. Yeah, the scrolling does suck.Quote:
Originally Posted by rge
Quote:
Originally Posted by ReverendMaynard
I updated to F4R. I am running memtest right now for 12gb of ram at 1600mhz using a XMP profile. I seem to be having no problems like I did with the other bios.
Looks like i've finally run into this multiple-multi's problem some of you are having. I was trying to find my memory's max stable overclock, and i started with 206bclk x8 (1648) with 18x QPI (3708) and 16x unc (3296), which was stable.
I decided it would be easier to instead reduce bclk and use the 10x memory multi. This of course means using the 20x uncore multi, which is no problem however with the lower bclk. I went for 165bclk, giving 2970 QPI and 3300 unc. I could not get windows to boot :mad:
Running QPI below the uncore clock can't be a problem, because i have run 222bclk with uncore @ 17x and QPI in 'slow mode' which takes it right down to 100mhz~ range, and it was stable. The uncore clock on that setting was 3774, so with only 3300 on the setting above that can't be the problem either.
CPU is lowered all the way, QPI volts are high enough for higher stable clocks than this (previously tested), any suggestions?
I had that same issue yesterday Xello, only worse, mine wouldn't even post when I did that.
This is a common mistake. Vcore bios=Vcore idle (Vdrop) is a good thing, ok. But Intel specification claiming that the Vcore idle is bigger than the load (Vdroop), for safe run of the chip. I seem to remember that AnandTech strongly discouraged in an article using loadlines in 45nm dual o quads cores, i.e. Itīs an ideal exact amount of Vdroop, but not 0. Vdrop=0 is ok.Quote:
I second the vdroop issue. I wish they would make vcore bios = vcore idle = vcore load.
Already with my new EX58, with all his capacitors in his right place :up:
With the loadline off i get nearly a whole .1v vdroop, it's not usable for any kind of decent OC :shrug:Quote:
Originally Posted by vega22
.1v vdroop is NOT a ideal amount :D
EDIT: ...or yes?
Actually if you read the entire article, including the cartoonish waveforms which are proof of nothing, the whole point of even that article is overshoot volts. If I set bios to 1.4, and vdrop is to 1.35 idle, and vdroop is to 1.3v load, then providing mobo manufacturer followed intel specs, overshoot spikes are to 1.4 (bios setting) or .1v.Quote:
Originally Posted by vega22
Now if I use loadline enabled, and set bios to 1.35 and idle is 1.33 and load is 1.3, the overshoot from same load is still going to be .1v or to 1.4v from 1.3. As long as you can do the simple math and not fool yourself into thinking the overshoots are not there...LLC by itself is not going to kill anything. But if you are using 1.5LLC vcore with overshoots to 1.65V thinking it is same as setting 1.5 bios without LLC, then you have a problem.
According to the anandtech article the above is correct in terms of overshoot volts. There conclusion based on 1 mobo, 1 older vrm circuit that volts were more erratic b/c overclock needed little more vcore, is not going to persuade me of anything. 50 mobos and take an average with 50-100K multi capable of accurately measuring waveforms and detecting slight volt differences, and we will talk. But even then, I can tell you on my mobo, LLC makes it easier not more difficult to OC, so I have different experience than them.
All of which is irrelevant anyways. Until real world experience shows me that vcore I am using with LLC enabled is harming my chips, I will continue to do so.
As an aside feeding a loaded chip less volts than unloaded is idiotic. It is simply cheaper to implement that piss poor strategy via circuitry than better much more expensive one of designing circuitry with less overshoot. But as long as I can add and subtract, I will be using LLC.
I like your style rge.Quote:
Originally Posted by rge
Cheers:toast2:
From Intel Specifications for 775. Only to illustrate the chat:
http://static.zooomr.com/images/6537...d5e90596_o.jpg
for vdrooping is very painful for this motherboard. here is the pencil-modQuote:
Originally Posted by Xello
can fix vdrooping. hope this can help you to oc. i duno if someone already post this.
hicookie
vdrooping pencil-mod for GA-EX58 series.
http://triton.imageshack.us/Himg218/...=578&ysize=480
i'm on f4r guys. flashed from f4f.
quite the same actually. performance wise.
wont post results till i max it. so far my chip is 5ghz 3dm06 but the score is a little low for my liking :D
Mh yeah, the Vcore does fluctuate a bit.. undervolting is quite annoying, Vdroop with LLC is quite ok.
Still too much if you come from a DFI that does bios=idle=load if you want it to.. or even better, works with EIST at any overclock so you actually get considerably lower idle Vcore. Hope they hurry the f up with the mobo.. my 965 is waiting :>
Can't believe i forgot about EIST, i'll be using that so i won't actually have to deal with the 1.47v at idle.
GA-EX58 series all(6) models:
GA-EX58-EXTREME
GA-EX58-UD5
GA-EX58-UD4P
GA-EX58-UD4
GA-EX58-UD3R
GA-EX58-DS4
Well EIST doesn't work when you set Vcore manually. Only DFI and Abit can/could do it.. no Asus no Gigabyte. They will never learn I fear...Quote:
Originally Posted by Xello
nice shotQuote:
Originally Posted by hicookie
did you take it eek
hmmmm
just testing the i7 965 chip in my UD5 on F4f and the RAM seems to clock a bit higher than with 920
just trying to get 1100MHz 8-9-8-20-1T stable atm
[nah still no good will try the new bios] 1080 Max so far
There is very much confusion with the bios some gigabyte ex58 extreme! But the good one is the 4l or the 4m? Thanks! Excuse my English. :shrug:
Huh? It was working fine at stock before i started tweaking i'm sure.. Will test later.Quote:
Originally Posted by jcool