Thats correct, no space for 16x @ 16x 2.0 x 2 SliQuote:
Originally Posted by OUTSIDE
Printable View
Thats correct, no space for 16x @ 16x 2.0 x 2 SliQuote:
Originally Posted by OUTSIDE
there is space after them tho, another PCI slot and PCI-EQuote:
Originally Posted by TheHardCase
:up:This is my current rig setup...
http://thumb5.webshots.net/t/72/72/5...5wVYVom_th.jpg
Have any of you guys had issues booting with over 1.1V QPI on F7E?
Setting everything at stock and only changing QPI to just 1.12V, it won't post. It fails at 26 on the debug LED.
Also, is there anyway to change memory timings from within Windows with SB? CPU-Tweaker doesn't work. And finally, is there a way to change BCLK within Windows? Easy Tuner doesn't have a slider for it.
sounds like your chip hates QPI VTT damn
You can change bclk in Windows, but if its not showing up for you in ET6 try this.Quote:
Originally Posted by bassplayer
http://www.xtremesystems.org/forums/...&postcount=444
Has anyone been able to use HDD Erase on this board? My install is screwed but I cannot get HDD Erase to run, worked on my x58A UD7 no issues.
^ Where???? F5m????
Change AHCI to IDE.Quote:
Originally Posted by OC Nub
He copy some my old post.:ROTF:Quote:
Originally Posted by maxxx
Ohh lol. I was confused.
Arrrgghhh.. Thought them changes are out already. Was thinking just in time to test before they start their long holiday.Quote:
He copy some my old post.
Thanks, but I had it set to IDE, still cant get it to run.Quote:
Originally Posted by nijel2@comcast.
If new BIOS didn't come today or 2moro,then we must wait until Chinese New Year end.:rolleyes:Quote:
Originally Posted by bassplayer
Noo!!Quote:
Originally Posted by stasio
F7E is terrible :(
AHCI doesn't work, can't raise QPI...
Yeah,atm nobody happy with BIOS improvement.:shakes:Quote:
Originally Posted by bassplayer
Power ON my rig with stable OC today ,cold boot at least 8 times Off/On.Clear CMOS,set back same setting and boot again.
i am on F7E
have to check AHCI but QPI is no issue for me and clocks wise its thte same as all other ones
Hoping the QPI thing might get fixed, even if it is only on this end.Quote:
Originally Posted by dinos22
Quick question: When holding Clear CMOS on UD9, I shouldn't have to power off the PSU to actually clear it, right? Just another thing to add to the list of reasons I need to RMA...
i usually power off CPU to clear cmos but you can do it without powering off
BIOS department must pull back Hicookie from X58A-OC board and put him on P67.:rofl:Quote:
Originally Posted by dinos22
Edit:
I notice that Asus and MSI (with UEFI) had new beta every 2-3 days with a lot improvements(changelog).From Giga is very hard to get what is change in new beta BIOS.
yea but Gigabyte never really tells you exactly what is changed, just the big points. There aren't that many issues as MSI and ASUS were having.
I can raise qpi/vtt to 1.1+v on F7E.
I agree though waiting for new years to end! gotta wait till feb i guess. February 4th is Chinese New year? correct? I had to wikipedia it.
It's always been common knowledge to power off the PSU when using a clear the cmos jumper or switch. This is to make sure there is no current going to the board. Besides, it takes, what, one second to flip the rocker switch on the PSU, right? Only no name cheap PSU's these days don't come with rocker switches... I've always turned off the PSU and I'll always continue to do it, until a manufacturer tells me not to.Quote:
Originally Posted by bassplayer
I remember back in the day when even decent PSU's didn't have rocker switches, and people complained loudly, then they started putting them back on again...sigh...the things companies do to save a buck....omit a $3 switch....
cookie doesnt do bioses lol
yea i have one or two PSUs without a rocker, and they are OK quality. IDK i have cleared BIOS without shutting off the board's power, but I forgot, nothing went bad, but idk I would turn it off if I remember.
bassplayer i actually had issues where to much qpi/vtt made my oc unstable or cause no post.
I just thought it was odd that the EX58-UD4P, P67A, and RIIIE that I own/have owned cut the power to the board when holding the Clear CMOS button whereas the UD9 requires me to power off the PSU... If I hold down the Clear CMOS on the UD9 it stops the board from booting, but never drains the power. It just becomes an inconvenience to get up every time and switch the PSU off as opposed to just hitting the Clear CMOS which is right next to me when benching.Quote:
Originally Posted by Falkentyne
The board decided it was okay with posting at 1.2V QPI tonight. Didn't change any settings other than that... The inconsistencies are killing me!Quote:
Originally Posted by sin0822
Any idea why the UD7 doesn't have a Clear CMOS button on the back? This one's actually in a case so if I'm benching, the side panel has to come off. It just seems like an odd turn of events. I guess I/O space was limited?
Little question guys, i am in the impossibility to flash the bios with latest one ..
I have downloaded the last bios, extract, put it on my USB flash drive, and using Qflash i have the message " invalid file format " ....
Download latest version of @bios and you should be good to go :up:
After my benching session last night i've notived that i can't keep my CPU stable at my previous stable Speeds/vcore.
I had the CPU @ 4.5Ghz 1.275Vcore (F7e) and it was as stable as a rock. 9 hours prime + Linx stable. Now it won't even get into windows BSOD 0x0000000124 with the same settings. I guess this is a sign that unless i want a new chip i have to take it easy for a while :down:
Another strange thing is i got some red dots (small artifacts, Nothing major just along the waters edge) in Starcraft 2 last night as well as the game and sound freezing/stuttering for a few seconds. I was playing with the driver settings while benching earlier in the ngiht, so i hope a fresh install will clean this up. Anyway i closed the game and ran furmark to test for artifacts, but nothing came up. Which is strange.
I'm a bit put off benching now, i can afford new hardware, but it's a little early to be opening my wallett again :rofl:
Just succes, in reality, i had install the last one, but without unistall the older ( from CD ).. uninstalled it, reinstalled it, and it have work .Quote:
Originally Posted by Johnny87au
thanks Johnny ( i was start to think im completely drunk for don't be able flash this board after many years of using gigabyte boards lol )
( nb: it's crazy how thoses 2600K clock ... 50 multi .... all voltage in auto ( outside DDR@1.66V ) boot... no problem ( i have not test if it's stable ofc but im impressed )
Hmm a chip shouldn't degrade so badly at low volts that you can't get into windows...Quote:
Originally Posted by jackbrennan2008
What is your BIOS setting for the vcore? Was that 1.275v a CPU-Z report?
Is LLC enabled?
If LLC is disabled, you may be trying to get into windows at as low as 1.2v....
In order for a chip to go from being linx/prime stable to a bsod just trying to get into windows, implies at least a 0.05v loss in voltage stability..
It was with LLC level 1.Quote:
Originally Posted by Falkentyne
I've been benchmarking @ 5.5Ghz 1.555v although the systems isn't at those settings for long. so i guess that can contribute to CPU degredation :). It's possible it's the bios though, i'm not really sure.
I tried with 4.5Ghz @ 1.285 last night and it also crashed with the same BSOD 0x0000124. When i have been testing new overclocks i have been testing various BIOS. I've been flashing the new BIOS with Q-Flash, maybe it's better to use @Bios.
0x0000124 mean a problem with QPI voltage too high or too low ..
I just run P67 from few hour, but the first thing i was surprised is the QPI/vtt voltage... it boot with my ram who need normally 1.35V for stability ( cr1, 1600-1700mhz cas6 ) with 1.05V on QPI ... this is really low.
Now i need test this baby, hope i will not face problem with this UD7 ( not much problem, i can afford some, this is part of the fun ) first thing i see, it seems i can't disable competeley the power Intel save settings ( CE1>CE6 ) .
Well i've been running it at 1.080v so maybe it's a little too high. I'll drop it down or even try AUTO tonight and see how i go.Quote:
Originally Posted by Lanek
Thanks for the tip i thought BSOD 124 was Vcore.
124 can be either vcore or an IMC issue. More often than not, though, it's the IMC.
When using auto, it set 1.05V QPI at 5.4GHz for me. To make it through 1M with memory at 2133 8-8-8, I needed 1.2V. It's possible that 1.08V is too little or too much. Try both!
It's the only BSOD I had till now and was related to a zillion of different factors (too high RAM OC, too tight timings, not enough volts, too much volts...)Quote:
Originally Posted by Lanek
any idea if Indigo Xtreme for 1156 will work on 1155? I never messed with 1156 so don't know how much of a size difference there is on the heat shield. or if anyone knows of a 1155 ver. in the works?
Any heatsink with 1156 socket support would/should work on 1155 as well.Quote:
Originally Posted by Kingleo
Is anyone using G Skill ram with this mobo?
I think the bios' have sever incompatability with some RAm sets.
I have used 2 sets 4 X 4GB @ 9-9-9-24-2t (this is the G SKILL Ripjaws X DDR3-1600 set)
and the 4X 4GB @ 7-8-7-24-2t (The G SKill Ripjaws DDR3-1600 set).
Neither will work for me at specified speeds. Only 1333...
Have tried raising voltage (voltage for the latter set is 1.6 anyways)
and have upped the VTT from 1.05 to 1.14 and still nothing. crashes before windows...???
GSkill Tridents work fine, no Ripjaws here hence cannot confirm.Quote:
Originally Posted by btdvox
So just reading the past few pages it seems alot of people can't get their RAM to work, I'm assuming the bios just sucks? is there no point in trying at the moment?Quote:
Originally Posted by thebanik
the Indigo-Xtreme isn't a heatsink.... more like a cpu specific thermal compoundQuote:
Originally Posted by thebanik
2x4GB G SKILL Ripjaws X DDR3-1600 CL9 @1.5V running with 8-9-8-21-1T @1.5V no problems here.
Huum Kingleo ... Are you Kingleo from L2 ( Teon ) living in guadeloupe ? ( Soyana here )Quote:
Originally Posted by Kingleo
Anyway, is it normal to don't be able to cut C1E CE6 etc states, and so run directly with the max clock ? ( bios F7a )
Just figured I'd share this for those who haven't seen it yet. Gigabyte UD7 vs ASUS MIVE.
Tridents fine here as well.Quote:
Originally Posted by thebanik
Interesting review. I guess the Asus board takes a 1.5 to 3% win over the UD7. Not enough for me to consider changing board though :). All i want is a nice stable BIOS from Gigabyte and i'll be even more happy with the board!Quote:
Originally Posted by EnJoY
Either way both boards are extremly nice to look at while being feature packed.
It's a bit dissapointing they didn't test the overclocking capability of the boards as this is quite an important factor when purchasing a high end motherboard!
Initial test @ 4.7ghz and 5ghz ... will test more stability when i m used to this new chips and motherboard
Batch# L042B076 ( too early to say if it is a good one or no )
2x 5870@1010mhz / 2600K@5ghz 1.37V / Mushkin Redline 6-8-6-6-24 1T
http://img255.imageshack.us/img255/5...rkvantagew.png
SuperPi 1M: 5ghz
http://img405.imageshack.us/img405/9785/65669969.png
3Dmark05 @ 5ghz : 44'797 http://3dmark.com/3dm05/5331769
3Dmark06 @ 4.7ghz: 34'074 http://3dmark.com/3dm06/15350297
I dislike motehrboard benchmarks, I feel as they are kind of a waste of everyone's time. There are so many settings in one boards BIOS/different BIOSes for teh boards that can affect internal latency as well as RAM speed.Quote:
Originally Posted by jackbrennan2008
Such as Gigabyte's extreme, turbo, standard in the RAm section, as well how different BIOSes can affet superpi scores. Its very bold to say one motherboard is better than another especially if you don't OC the boards. I know who mia is, but the socket burn thing and their 1.35v limit on sb, its just like they want people to pick at what they write.
I can understand if you give a modest OC to a mainstream board like the UD4 and not test its max OC, but for boards like the UD7 and M4E, if you aren't testing max OC somthing is VERY fishy, especially a site who's reader base is 90% ocers.
You have to get some sort of data to compare Sin... if you have to wade through all the variables in the bios, take the plunge in a zillion biosses, you won't get the review "properly " done, not even in months after it's initial release... it's really impossible to get it correctly done...
I always put the ram timings and co the same... rest remains on auto. IF there are big discrepancies I would go deeper... you also have to put yourself into the position of the reader : most just buy the board, put the rams in and it has to work...
Plus boards with the same chipsets mostly perform well within a fraction of eachother... Even if they are OCer boards, my X58 high end roundup only had a few discrepancies...
I expect the same for most Sandy boards... most boards will clock most CPU's exactly the same, being it 12 or 64 phase power... maybe with one board you will have to put 1.4Vcore, the other one might get away with 1.39... one board can do 108Bclock , another one of the exact same type only 105... too many variables to make a bulletproof conclusion...
Clock for clock it prolly will be insignificant...
Gigabyte seems to have a strong board for 01 and 03... the Asus and co are slacking there...
Agreed, I know the main thing I would like to see in a comparo is if one board can get any extra Mhz or mem speed over the other. Were there any differences there? That is the main reason why I would buy 1 board over another if they were in the same class. I did find the 3d comparos interesting however. I really didnt expect MIVE to do any better in 3D.
Yes, + get infos on how the bios and voltage ( vdrop? etc ) are working in OC situation... Anyway high end motherboard are closer of each other now, i remember buying DFI NF4 SLI DR and CFX3200 just cause they was the only board where BH5 have enough Voltage for work. ( good old time )Quote:
Originally Posted by Eldonko
I know it can never be perfect, but that is why I also don't like to judge boards like that. You can have baseline benchmarks(with everything the same, down to the SSD/hdd and OS), but that shouldn't be the significant part of judging the boards performance. personally even in my reviews, i don't like running bechmarks on a board. I feel as if you OC the board and then compare the percent improvments with the overclocks ont he same board is a much better guage of the boards performance. i feel they give the user a number dependant on the CPU, RAM, GPU not the board. Its not like one board uses different material for the traces and transfer data differently, nto yet at least. I know its how boards are reviewed, and that is why i feel as though motherboard reviews should be done differently, with more emphasis on features and overclocking, and less on benchmarks.
I also feel as though BLCK has everything to do with the CPU, and not the board. That OC guide for Sb that originally came out said that BLCk was dependent on the board, ASUS themselves said its dependant on the CPU, and i agree with the CPu dependance, after testing more than one CPU, i found that one CPu does 107.5blck and the other does 106.5 the 107.5 has a 4.8ghz limit and the 106.5 has a 5.2ghz limit. Kinda struck me as odd.
i remember the BH5 days, i still have my sticks :)
The bios has problems with all 4 RAM slots full. Running 4 X 4 GB is much different than 2 X 4 GB. I think the bios does not like anything above 1333 with 16 Gigs of ram.
Gonna test more today though.
16GB of memory is going to be a huge stress on the IMC. What kind of QPI are you running?Quote:
Originally Posted by btdvox
Ive tried increments from 1.05 - 1.16Quote:
Originally Posted by bassplayer
Not going to go higher than that. I've decided to return one kit, and admit that 16GB vs 8GB is prob overkill, seeing as I was fine with 6GB on my X58 lol
Decided to buy a C300 to take advantage of SATA 3!
c300 is a great decision, i have one and it rocks! make sure to install RST 10 drivers, as they are better for 4k speeds than the stock msahci.sys, make sure you use Intel ports if you aren't already!
Yea i agree 16 gigs is a huge stress ont eh IMC, I am going to test 4 sticks at 1600 to see if its a 4 stick problem or 16 gigs problem. Also QPI upto 1.25 should be ok for benching.
Thanks Sin,Quote:
Originally Posted by sin0822
Yeah i think it's a 4 stick problems, I tried 2 different g skill sets and 1 corsair set. Might be my bios, but this is in fact the first time I have ever filled up all my memory slots. It's pretty much a rule in the book that occupying all slots means loss in performance (overclock wise).
Decided to go 8 GB, 7-8-7-24-1t DDR3-1600.
Thanks for the tips.
Ok I've been doing a little research now. I think (running LinX now) i've found the culprit.
My ram:
Corsair Vengeance 1600Mhz CL8 8-8-8-24 1T 1.50v
Overclocked to 1866Mhz 9-9-9-24 1T 1.56v
I mentioned earlier that i kept getting the 0x000000124 BSOD. Anyhow after raising the QPI/VTT it has finally gone away. LinX has been going for 3 minutes now without crashing which it didn't do before raising QPI. Lets hope it holds out the full 20 runs with Maximum ram usage. If it does at least i know what kept crashing my PC. Cheers for the help earlier people.
i see, how are those vengance doing? I just exchanged my dominator 1600mhz 4gb kit for some patriot vipers or whatever, same timings, same grade all i need from them is 1866 at 9,9,9,24
LinX still running, 55 minutes and all is well. qpi/vtt increase has stabilized the system :). i love my vengeance they were also a bargain.Quote:
Originally Posted by sin0822
i'd say 99% stable @ 1866Mhz 9-9-9-24 1T. i cant say 100% stable, but its passing LinX :)
New BIOS released.
updating now! thanks!
Quote:
Originally Posted by stasio
where where
Links in my sig.:DQuote:
Originally Posted by GLENBOY
Sweet thanks, updating as well.
Fantatic i'll give it a go right now. About to goto bed, but i can at least flash it :)Quote:
Originally Posted by stasio
:am:
Also i have to admit to being a complete moron. I have been so annoyed that my CMOS reset button on the motherboard hasn't been working. I've been so frustrated. But then i got a flash light and saw there was another button right beneath it.
I've been pressing the reset button
:am:
got it stasio :up: you da man
Any chance of a changelog? It would be nice to know what has been updated?
Don't update, the bios is garbage won't even boot at 4.5.
Going back to F7e:down:
hmm let me see
You sure the chip is getting enough vcore, maybe GB have adjusted the voltaged to eliminate the drop & level out the overshoot with loadline :shrug:Quote:
Originally Posted by TheHardCase
Let us know how you go :)Quote:
Originally Posted by sin0822
i see they removed CPU PLL Overvoltage... so that is probably his concern.
Doesn't affect my OC but I have D1 stepping so cpu pll overvotlage doesn't help. I am check LLC i doubt they did anything for LLC.
Mine boots fine at 4600, pass hyper pi no issues on F7, same voltages as before.
I was 5.0 100% stable now I'm having a hard time getting back there...
Even going back to F7e
Yeah,I don't mind to add some voltage on CPU,if they memory problem gone.:rolleyes:Quote:
Originally Posted by C-N
You cann't expect that BIOS F7e and F7 is the same.Quote:
Originally Posted by TheHardCase
After 30 reboots back to 5.0, nice & stable again...:up:
F7e Rocks
LLC report
1.40 set
Level 1 IDLE 1.405 DMM/ CPU-Z=1.404
Level 1 LOAD 1.381 DMM/ CPU-Z=1.368
Level 2 testing now.
LLC level 2
1.40 set
Level 2 IDLE 1.431 Measured/ CPU-Z 1.428
Level 2 LOAD 1.468 Digital Multi Meter(measured)/ CPU-Z 1.452
LLC Standard
1.40set
Standard IDLE 1.398 DMM/ CPU-Z 1.404
Standard LOAD 1.302 DMM/CPU-Z 1.296
Not that great, LLC level1 is good though, Standard caused BDOS at 5.0ghz all tests. Voltage read point at the inductor.
yeh cpu pll removed as sin says , for me can't boot with the higher multis , but ok on normal multis, used same vcore as i did on f7e to boot from 54 multi and a no go i never checked if the vcore was actually lower than on f7e, might try it again but back to f7e for nowQuote:
Originally Posted by sin0822
might as well stick with F7E for OCes, LLC Level1 seems to be improved pretty good.
Im too lazy to try and find the voltage read points, I wish Giga would do what EVGA, ASUS, and MSI are doing on the high end boards and put easy read points on the board.
Why would they remove cpu pll overvoltage?
O_O
Is that a bug?
What happens if you set cpu pll voltage manually from 1.8 to 1.9 ? Does that do the same thing?
I wrote a OC guide/performance review of this board, i have voltage read points in there, CPU Vcore is pretty easy.Quote:
Originally Posted by OC Nub
No CPU PLL overvoltage basically unlocks multipliers. The problem with it is that when its enabled it disabled S3 sleep. So maybe they should make it a special feature only accessible when you disable S3 sleep or something like that. Or have us hit alt+f4 like on the X58 GB boards unlocks spread spectrum.Quote:
Originally Posted by Falkentyne
Thanks.Quote:
Originally Posted by sin0822
Still cannot boot above 46x multi raising CPU PLL to 1.90v. That makes me wonder how high the voltage was with that setting enabled.
Since sin said that the settings aren't the same, and the overvoltage setting was designed only for unlocking higher multipliers, I guess that was still 1.8v, and it actually did something else. Maybe the s3 sleep disabling was related.
Um...now that I'm thinking, what happens if you change the sleep mode in power options?
Does Overvoltage reappear?
If you guys haven't given up in frustration yet, might as well try a hard power off (power off then unplug PSU for a few minutes) and see if you get the "double post" /overclocking failed, or if that has been resolved or not....
I could only boot above 46x multi before using Internal CPU PLL set to enabled or auto. Now it will not boot above 46x multi because the setting no longer exists, the multi in bios was already unlocked, or unlocked as high as I needed, couldnt boot above 52x no matter what voltage was used.Quote:
Originally Posted by Falkentyne
Normally I get the FE post code and it hangs, that happens 2 times then it will go to 70 and shut down, the next time it goes to 75 and allows me to get back into the bios.
i tried everything to get CPu PLL Overvoltage nothing worked, i even did disable S3 sleep to S1 sleep or whatever.
Hmm yea the thing is Cpu PLL Overvoltage might overvolt the CPu pLL, or allow the overvolting of the CPu pLL, i am unsure, either wa you have to have the feature for it to work. BTW 2.0v Cpu pLL is ok for MAX OC like 5.8ghz like i have heard.
I'm confused,
You're saying that 2.0 cpu pll unlocks the higher multipliers?
I have a feeling Ill be back on the X58 by the end of the weekend. Will give things time to mature and try it again down the road.
No that when OCing for really high multi like 57x people have used that high of a CPu PLL voltage.Quote:
Originally Posted by Falkentyne
I think maturity will only come with later revision of the CPU or Ivy Bridge. This is the mainstream product, not meant to be enthusiast, yet we are pushing it to its breaking point. I still think that 4.5-5ghz on air for 24/7 is pretty damn good enough reason to keep the setup.Quote:
Originally Posted by OC Nub
I agree, it oc's better than 1366, or any other platform, voltages are low, temps are great but its not nearly as much fun.Quote:
Originally Posted by sin0822
I didnt spend much time on the Classified 3 board, Id like to see what it can do.
yea its not as much fun, its just too easy ! :)
Unfortunately, I'm completely broke (less than $30 in the bank after this upgrade) so I don't have the luxury of fun for now :( Can't buy any more stuff....Quote:
Originally Posted by sin0822
ehh don't worry just enjoy what you have, it wont get better for a while anyway! Money is just money. it comes and goes.
If you guys have Intel SSD connected to SATA 6GB/s, update its firmware to get almost instant detection at boot up.
http://downloadcenter.intel.com/Deta...&DwnldID=18363Quote:
This firmware revision fixes enumeration and slow-boot issues on SATA 6Gb/s controllers, adds improvements to S.M.A.R.T. attributes for more accurate reporting of drive health, improves NCQ capability, and fixes possible drive hangs when reading S.M.A.R.T. self-test log.
Guys you shouldnt be running the intel drives on Sata 6gb anyways, their 3gbs drives..
Pulled out old cards to try, they dont scale very well. They actually lose to my 460gtx sli setup, but fun to try.
not bad my friend, very nice!
So,with BIOS F7 anybody can reach 5GHz or over?
Only fix S3 bug,with price of CPU PLL overvoltage option.:confused: