want my turd :rofl:Quote:
Originally Posted by l0ud_sil3nc3
Printable View
want my turd :rofl:Quote:
Originally Posted by l0ud_sil3nc3
I forgot to mention that I'm not having any more BSODs since the latest Bios at 4.2 Ghz CPU speed.
At least not outside of Prime where it happens due to the temps - I need to lap my CPU + Heatsink to get it 100% stable, I'll try that in the future.
I havnt had any problem with running my 12 Gb at 6-8-6-20 T1 timings at 1600 Mhz, but they absolutely still wont go higher than 1800 Mhz @ 8-8-8-24 T2. The 1600 Mhz timings are great though.
hi this is my first post here on these forums but i found an interesting thing with my RIIIE, did anybody tried a memory Benchmark with and without C-States, cause i loose about 1 GB/s in nearly every aspect, (copy,read,write), cannot post any screenshots now atm cause main rig is testin LinX cause i have a annoying problem, i must get rid of the x22 Multi, so my only way to do this right now is to put in the x20 Multiplier but i need more V-Core with these. :(
Could be due to the use of a even multiplier. Has to do with QPI Link speed..read here in depth about why it's better to use odd multis over even:
http://www.xtremesystems.org/forums/...&postcount=877
hmm...my cpu is all over the place atm...well not massively. last week i was testing it at 50% LLC and it was stable all the way up to 4.2 at 1.325 (linx 100 runs, prime, pcmark and 3dmark). Then i moved the gfx card from slot 2 to slot 1 and bam, no longer stable - everything else is fine but linx has started erroring out around run 25-30. At 4.13 it seems to be fine at 1.35V 0%LLC, or 1.325 50% LLC, but even one bclk up (4.16) it errors in linx. no BSODs or anything, just linx errors out. Anybody have any ideas of what to sort out? running QPI at 1.2625 but i only have 6GB of ram.Quote:
Originally Posted by MariaH79
maybe i should post a BIOS template?
The board starts to be really fun :)
http://www.abload.de/thumb2/linx77er.jpg
Here is stats after 1 linx run, sry it aint 20 im still adjusting 0010 seems hotter thou maybe its just shows the right temps..Quote:
Originally Posted by facboy
My cpu voltage doesnt move much at all, just my QPI with LLC OFF.
http://i249.photobucket.com/albums/g...e/1run0010.jpg
My GTX 480s is in slot 1 and 3 (maybe u should try slot 3.. its 16x too), then again i havent passed 4,2 yet cause o temps getting hairy, need to sort my WC before i can check it out.
Not sure if this means anything but my PSU got 2x 12v.. i connected both, used both EZ-plugs inntakes on the MB too but stopped using them ... seems more like those are used for 3way/4way setups.
What really fixed the last bit for me was getting right memory timings, had to up the trfc and get right rtl´s. usualy i set trfc about where i think it would be (mine ended up at 90)
and just leave the rtl´s on auto and adjusted qpi voltage until it hit a sweet spot on the rtl´s like 63-65-66
(not good timings and prbly not the ideal way to adjust them but worked for me getting linx stable)
U could try post a template i guess, thou i think others here can provide better answers since im still new at this ;)
great article, so i guess i have to renounce the C-States.Quote:
Originally Posted by spoolindsm127
It is not issue with R3E, it's like that with all MB's.Quote:
Originally Posted by Scerate
I know that and hate it, as I like seeing high bandwidth in Everest. :D
Got rid of it by buying higher multi CPU's and lowering it. :ROTF:
hum that's strange, my VCore is all over the place if i have LLC off. do you have it set to Auto or to 0%? 'cos Auto is not really off. i have no idea how to set the TRFC, i've just been leaving it at auto (which is 72 according to cpu-z). should i be increasing that? what is it based on?Quote:
Originally Posted by MariaH79
damn so i have to stay with x20 multi and 1.337 V-Core instead of x21 and 1.257 V-Core :(Quote:
Originally Posted by donmarkoni
And what about x21 and 1.337V? As even multi's require more vCore then odd, it just might be possible to run at that voltage, or a little bit higher. It's worth a try.Quote:
Originally Posted by Scerate
If not, test how much voltage you actually need to run it @ 21x and CxE Enabled. It might be worth too. :)
BTW, post mentioned here:
is a little obsolete. :rolleyes: No disrespect to the author meant.Quote:
Originally Posted by spoolindsm127
Did you try to use BIOS FLASHBACK option?
When I try to switch to the second bios I get following screen:
http://a.imageshack.us/img704/4773/recr3ed.jpg
For me the issue looks like a BIOS bug.
I have question to you guys about noctua nh-d14 and r3e. How much place dose it take ? Photo wanted :) Big thanks :)
if im exactly i need the same v-core (1.337) with following settings. This Settings are the one i set into the bios.Quote:
Originally Posted by donmarkoni
21x195 = 4095 Mhz 25 passes LinX stable
20x203 = 4066 Mhz 25 passes LinX stable
but you can see that i actually get bit higher with uneven multi as you said. So what should i do? Try 0010 Bios + play around with clock scew ? Temp i got was max 81 °C but my room temp is pretty hot here atm too 32 °C ;)
Quote:
Originally Posted by l0ud_sil3nc3
Unless you are going to be using a 980x save yourself some money and look for a better option for bloomfield, this is not the best board for ocing them. Get a classi or just save your money.Quote:
Originally Posted by l0ud_sil3nc3
Guys i read all thread and after that i now one thing: r3e is not so good for my i920, better stay on r2e - correct me if im wrong :) Because im thinking about new mobo and i dont know which one will be better than my r2e.
I got it once, or twice, I don't remember. Shut down the board, turned it on and it was solved. Or did I cleared the BIOS while shut down, IDK. Try.Quote:
Originally Posted by jacek
I would push BCLK up to 200MHz, trying to get stable @ 4.2GHz, but you'll need to do something about your temps. Wait for the winter, maybe? :D Just kidding. Post your BIOS settings using template and then we'll see.Quote:
Originally Posted by Scerate
BTW, are you using CPU LLC and/or QPI LLC?
I saw only two dissatisfied overclockers. All others are more or less satisfied with it. IMO, it's a great board. But it's all about expectations. If you expect 4.4GHz@1.3V from your 4.2GHz@1.4V CPU, then you'll be disappointed. ;)Quote:
Originally Posted by Malik
@Malik - If u think that R3E is not good for ur chip get Classy than ;]
And if ive got to b honest i dont understand why ppl so much complaining about R3E =] In my opinion is good enough for WC+Bloom's ;]
If i only couled get better chip ;]
ok guys, i just joined the club!
had my r3e laying around some days now, mem arrived today. so i just dropped in a x5570 (had nothing else at hand right now) with an old 775 cpu cooler, my 6GB gt kit and an old PCI (yes pci) vga card. so no pcix cards and no hd's connected right now.
at first power up cpu temps rose quick so i powercycled the system after a few minutes. at next power up cpu temps stayed just above 40C but i could watch ioh temps rising. took about 10 minutes until the ioh reached 100C and the board switched off!
running bios 0704.
chipset heatsink is pretty hot, so i guess contact is not that bad.
anyway, anybody seen this before?
Yeah this board gets hot, I cant touch mine without burning sensation in my fingers, but it has never shut down on me. Maybe its time for a water block on mine since I am not water cooling the 5870.Quote:
Originally Posted by fgw
Quote:
Originally Posted by fgw
You could also try scraping the crap TIM off your heastink assy and replace with non-garbage TIM such as MX-2 or even AS Ceramique.
If the heatsink is very hot, wouldn't that mean that the heat transfer is working great though?Quote:
Originally Posted by turtletrax
Ah like i said im new to this prbly people around with better answers here. LLC in bios is set to auto but jumper is on disabled.Quote:
Originally Posted by facboy
Thought jumper set it off and on and kinda disable it overall if set to off :confused:
About the TRFC.. i think its in the xmp profile but when using it bios dont set TRFC and a few others, but Everest seem to read it "all"
U can see on Extreme Memory Profile there are more settings than the bios preset for u when selecting the profile and at 800 profile suggest 88.
To give myself slight head room i set 90, then again i got some funky memory sticks,so better memory will provide better timeings from the profile and overall.
http://i249.photobucket.com/albums/g...a-Love/mem.jpg
Quote:
Originally Posted by fgw
Using a 775 cooler on a 1366 is asking for trouble. Why would you be surprised that the temps rise quickly???
I know, I know zsam has the exact same thoughts but I figure I want to try a 32nm quad eventually and try to see if my X5660 works as well in that board, even though I am pretty sure it is dead:mad:Quote:
Originally Posted by OC Nub
32nm quad is the only reason i still have the damn board;)Quote:
Originally Posted by l0ud_sil3nc3
Apparently you can't read. It was stated earlier that the IOH chip is the culprit. If English is not your first language, I can understand your lack of understanding. You got a bum chip (IOH). Get over it, get a life and move on... pleaseQuote:
Originally Posted by zsamz_
you must be retarded and didnt readQuote:
Originally Posted by Carrera997
with my 980x it worked perfect
i had no problems going to 220 blk with 980x
and i cant go over 215 with bloomfeild
so go take your 8 posts and troll somewhere else
i know IOH is a problem i dont need a moron to tell me
meanwhile with 32nm with crappy IOH doesn't limit as much;)
Wow, that was a highly uncalled for and unprofessional post.Quote:
Originally Posted by Carrera997
Quote:
Originally Posted by turtletrax
replacing the tim was my first idea also, but as the heatsink is hot, i guess heat transfer is ok! on the other hand, replacing the tim won't hurt either ...Quote:
Originally Posted by SkOrPn
cpu temps are under control now. cpu temp rose high just at first power up. this is just for a first functional test on the desk to check if everything is working. the system will be watercooled when moved into the case.Quote:
Originally Posted by Carrera997
also the x5570 i use has a TDP of 95W! nothing a 775 stock cooler should not be able to handle even under load ...
but the problem is not cpu temp, its ioh temp!
I have to say I'm glad I have a nice £50 cooler for my CPU.
Once its properly seated, my temperatures are the same as what I've seen in corsair H70 reviews. Getting a good CPU cooler for I7 is very important for overclocking, also for this motherboard, you can try and look for a nice matching red Titan Fenrir, which is very affordable cooler, an performs brilliantly for its price. The black megahalem cooler would also look great with this board. Mine clashes a lot with the board colors, but it matches the rest of my cooling so its ok - silver CPU, VGA, and Ramsinks :)
Forget that, I disagree, my CPu cooler is just ugly :(
http://a.imageshack.us/img412/7425/dsc00697kz.jpg
But its very effective, and I've not seen anything with better results yet.
Wow, that has to be a record. 8 posts and 9 inflammatory statements, most of them wrong. I think you are looking for [T]ardOCP or RRR, they will love you there :up:Quote:
Originally Posted by Carrera997
i have 3 chips and 1 can do 4.6 no ht on air the other 4.5 so its all about findin the right chip for airQuote:
Originally Posted by bhavv
the last chip just needs water
we always get the trolls like Carrera997 pop their heads in n go away when they setup their rigs:rofl:Quote:
Originally Posted by turtletrax
ah good info about everest, i'll have to check that out.Quote:
Originally Posted by MariaH79
the LLC confused me as well, there's two LLC, QPI and VCore. the jumper only controls QPI LLC, the BIOS setting controls VCore, so if it is on Auto then that usually means it is on :).
@zsamz_
You must understand his anger (refering to Carrera997) if he is new and read the whole thread. Don't misunderstand me, I'm not justifying his behavior, but you and OC Nub are making a torrents of anti-R3E posts, almost nothing constructive. Whatever is asked, you just say the board is bad and it costs a lot of money and it's your answer. Is that really all you have to say? I really don't care, I just skip reading your posts, but you should care, as I respected your opinion before and now don't even bother to read your posts.
If i can make 4.3Ghz 1.4V (21x205) in r3e - i will be mega happy, because this was my max cpu clock on r2e ;)Quote:
Originally Posted by donmarkoni
i stoped biatchin n just advisin not to buy the board lolQuote:
Originally Posted by donmarkoni
why should others waste $ on it
the new rexIII formula looks like redesign:cool:
i just wished i didnt listen to someone respected in the beginning of the thread say it was good for 230blk 24/7 with bloomfeild:mad: he shoulda tested proper before giving his opinion
i'm boxin the board n i'll just take it out when the 32nm quads come out
my rexIII got my cpu to 4.77 linx stable :cool:Quote:
Originally Posted by Malik
It shouldn't be a problem at all. I ran mine Bloomfield 24/7 @ 4.2-4.3GHz (200-205BCLK).Quote:
Originally Posted by Malik
I understand. Happened to me many times to expect more and receive less.Quote:
Originally Posted by zsamz_
At least my board is higher average class, if I assume correctly. It hits a wall at BCLK 225/240 (QPI x18/x12-slow mode) maybe a couple of MHz higher.
Maybe you just have a less then average MB and should sell it and buy another one.
my wall is 214blk:(Quote:
Originally Posted by donmarkoni
but still that slowmode suks even @ 225
zsams my wall for my 950 is 214 as well. No matter what I do it just wont boot lol. But I would need to try another R3E to check or another bloomfield...Quote:
Originally Posted by zsamz_
I have next question :)
In manual i see only one usb connector on mobo:
http://www.abload.de/img/usbw8er.png
And that is :down: info for me because i need two. I search some info and i found something like that: You can connect usb to the OC station connector
( in manual is: for OC station ONLY )
True or False Guys :)
IDK. :shrug: I'll test when I get home, if no one else does it in the meantime.Quote:
Originally Posted by Malik
Ok thanks :)
Sure here is my template:Quote:
Originally Posted by donmarkoni
are you using CPU LLC ? Yes 50%
and QPI LLC? Yes
I wanted to get around 4Ghz stable, but after i saw how C-States affect my Memory Benchmark i was a bit like, what the hell. I even wanted to keep C-States on cause sometimes i just surfing around and with C-States on it keeps my idle power consumption and temps down. But the issue with 21x was that when i played old games like System Shock 1 via Dosbox, the Turbo (22x on one core) kicked in and the rig was running around 4,2Ghz but it was unstable, so i got 0x101 (V-Core) BsoDs or the game was closed abnormally, but if i only got 21x Multi everything was fine, 25x LinX, 6h Small FFT P95 and this has kept me to the decission that i have to do something to optimize that issue. :)Code:
Target CPU Frequency 4060MHz
Target DRAM Frequency 1627MHz
LN2 Mode Disabled
QPI Loadline Calibration Enabled
Sync Mode [Enabled]
AI Overclock Tuner [Manual]
OC From CPU level Up [Auto]
CPU Ratio Setting [Auto]
CPU Turbo Power Limit [Disabled]
>CPU Configuration----------------------------
CPU Ratio Setting [Auto]
C1E Support [Enabled]
Hardware Prefetcher [Enabled]
Adjacent Cache Line Prefetcher [Enabled]
Intel(R) Virtualization Tech [Enabled]
CPU TM Function [Enabled]
Execute Disable Bit [Enabled]
Intel(R) HT Technology [Enabled]
Active Processor Cores [All]
A20M [Disabled]
Intel(R) SpeedStep(TM) Tech [Enabled]
Intel(R) C-STATE Tech [Enabled]
C State package limit setting [Auto]
----------------------------------------------
BCLK Frequency [203]
PCIE Frequency [101]
DRAM Frequency [DDR3-1627Mhz]
UCLK Frequency [3256Mhz]
QPI Link Data Rate [7326MT/s]
Memory Configuration Protect [Disabled]
>DRAM Timing Control--------------------------
CAS# Latency [7 DRAM Clock]
RAS# to CAS# Delay [7 DRAM Clock]
RAS# PRE Time [7 DRAM Clock]
RAS# ACT Time [24 DRAM Clock]
RAS# to RAS# Delay [Auto]
REF Cycle Time [Auto]
WRITE Recovery Time [Auto]
READ to PRE Time [Auto]
FOUR ACT WIN Time [Auto]
Back-To-BackCAS# Delay [Auto]
Timing Mode [1N]
Round Trip Latency on CHA [Auto]
Round Trip Latency on CHB [Auto]
Round Trip Latency on CHC [Auto]
WRITE To READ Delay(DD) [Auto]
WRITE To READ Delay(DR) [Auto]
WRITE To READ Delay(SR) [Auto]
READ To WRITE Delay(DD) [Auto]
READ To WRITE Delay(DR) [Auto]
READ To WRITE Delay(SR) [Auto]
READ To READ Delay(DD) [Auto]
READ To READ Delay(DR) [Auto]
READ To READ Delay(SR) [Auto]
WRITE To WRITE Delay(DD) [Auto]
WRITE To WRITE Delay(DR) [Auto]
WRITE To WRITE Delay(SR) [Auto]
----------------------------------------------
CPU Differential Amplitude [Auto]
CPU Clock Skew [Auto]
IOH Clock Skew [Auto]
------------ Extreme Engine Digi+ ------------
Digi+ PWR Mode [T-Balanced]
PWR Volt. Control [Auto]
Load-Line Calibration [50% Calibration]
CPU Voltege OCP [Enabled]
CPU PWM Frequency [Auto]
Extreme OV [Disabled]
Extreme OC [Mode 1]
CPU Voltage [1.33750]
CPU PLL Voltage [1.40450]
QPI/DRAM Core Voltage [1.35000]
DRAM Bus Voltage [1.68275]
>DRAM REF Voltages----------------------------
DRAM DATA REF Voltage on CHA [Auto]
DRAM CTRL REF Voltage on CHA [Auto]
DRAM DATA REF Voltage on CHB [Auto]
DRAM CTRL REF Voltage on CHB [Auto]
DRAM DATA REF Voltage on CHC [Auto]
DRAM CTRL REF Voltage on CHC [Auto]
----------------------------------------------
IOH Voltage [1.17925]
IOH PCIE Voltage [1.51050]
ICH Voltage [1.11300]
ICH PCIE Voltage [1.51050]
---------- Spread Spectrum Control -----------
CPU Spread Spectrum [Disabled]
PCIE Spread Spectrum [Disabled]
***on-board jumper dependant
Try it i a moment hold onQuote:
Originally Posted by Malik
Damn coulnd't test it only have a Connector C joint
but looked like the OC Station Connector have 10 Pins instead of 8-9 like USB
Hmm :( so i will try this tomorrow when my r3e arrive ;)Quote:
Originally Posted by Scerate
You realy got a bad IOH Zsams.
FYI we had 6 REX III setups working for the Tones OC demo (all retail ones) : one board just like yours had a 215 Bclock wall. All the others went happily over 225. Yep all Bloomfields CPU's: 930, 950 and 975 and one ES GT 970. Biosses varying from 0704 to 0901.... Me and Massman verified the Bclocks on friday... One board had zero IOH contact and was straight on first boot 90°C lol...
My i950 clocks 229 rockstable on my board (never ever did that on one my previous boards), Gamer's Xeon 3540 even does 233 on air (almost same as on his Classified : 235)
My point is : it's not all the boards, it's just the crappy IOH ones that have issues... This same Bclock issue is also on MSI Xpower ( there's it's also related to PCB and co ) and Gigabyte UD9--> first sample had a 220 Bclock wall...
I really wished some binning was possible on the IOH's before putting them into the top of the line boards... my Rampage Gene II had a 225 wall while my REX II was a 222 one...
EDIT : Just assembling one right now : i930 cpu is Wpriming at 225 ( no slo mode ) HT is off as it's only the boxed cooler
Ive not tried going over 200 BCLK because neither my Ram or CPU can handle it. No point in reducing multis when they are working great maxed out at 200 BCLK.
Maybe after I try lapping my CPU + Heatsink, but Im not doing that for a long time.
The best way around this issue that I found is the NZXT IU01. Installed and apparently working fine on my rig.Quote:
Originally Posted by Malik
vCore offset will not be available.
Source TWEAKTOWN
"Regarding the offset vCore question, as of right now we have no plans of implementing this option. We asked a lot of users about their thoughts on having vCore offset and we did not get a lot of feedback. We did get feedback from the extreme community and most said that they did not utilize dynamic VID, and that they are more concerned with maintaining a more consistent voltage using absolute vCore. However, we do have 2 options for load-line calibration (50% and 100%) to give users more voltage control. So as of right now vCore offset will only be implemented in our mainstream board as well as the upcoming Rampage III Formula."
Ok I finally decided to try the same 950 on my E760 and straight boot past 214 bclk no problemo...
Ugh I guess I have a bad IOH chip in my R3E :eek::shakes:
PLL ICS9LPRS918BKL for Setfsb and rampage 3 dosent work.... what correct pll should i have
Well, after reading manual I decided not to test it, as it is not meant for anything else then OC Station.Quote:
Originally Posted by donmarkoniIDK. :shrug: I'll test when I get home, if no one else does it in the meantime.Quote:
Originally Posted by Malik
http://a.imageshack.us/img217/9786/r3eocstation.jpg
I changed some settings in your template and colored them for you to try.Quote:
Originally Posted by Scerate
One important thing: don't lower CPU PLL voltage below default, not good for bloomfield, only good for gulftown, in my experience.
Also, put IOH voltage @ Auto for start and then lower it later, as you don't want some IOH induced instabilities.
Try lowering VTT (QPI/DRAM) as I think it's too high for UnCore @ 3200, but it depends on IMC.
If unstable, raise vCore a little.
BTW, what BIOS are you using? And what cooling? Can't remember if you mentioned before.
:shrug: Works for me. Just tried it again. Anybody else?Quote:
Originally Posted by PytonOrm
Great solution of this problem: NZXT IU01 :up:Quote:
Originally Posted by donmarkoni
Yah, I think you'll like it. The only issue that I really had with it is that the only thing preventing this piece from shorting out is a 1/8" piece of neoprene adhered to the back of the circuit board. It does do the job, but a plastic housing would be nicer.Quote:
Originally Posted by Malik
This "thing" works good ? No problems with that ?Quote:
Originally Posted by vladimir
Bios: 0902Quote:
Originally Posted by donmarkoni
Cooling: Prolimatech Megashadow + Artic Cooling MX-3 + 2x Xigmatek Fans @ 7v (replacing it maybe with 2 Be Quiet Silent Wings)
but i have a question, i won't be unpolite no i even appreciate it that you helping me that much, but why should i give the processor 4,2Ghz ? 4Ghz is enough for me atm cause if i even give him that speed, would be cause of C-States that if i do something one one core only resulting in 22x200 (4,4Ghz)? My rig is perfectly stable now but the only culprit is this x22 multi which gives me headache cause then everything fails, i even tried LinX with only one core resulting like i said in Bsod 101. :(
and btw lowering CPU PLL even helped me lowering my volts 2 notches, cause first i was running 21x191 ~ 4 Ghz. If i set in to default settings 1.8 i got LinX "stopped upon error" warnings. First i set it to only 1.5v but now i lowered it to 1.4 cause i think it even lowered my temps 1-2 °C :)
Edit: I have just tried your optimized template with 4.2 Ghz but i think atm it's a no go, cause i stopped testing after i put in about 1.36v it even then crashed 0x050 BsoD but temps rose to about 93 °C which is damn hot, even now on evening here (have 29 °C room temp lol) but your Loadline Calibration tip is pretty awesome but rose my temps about 3 °C per core with LinX @ 203x20, but i was able to lower the v-core a bit :)
well heres a thing ....!!
dunno if anyones seen this b4 but if youre worried about performance in gaming when mixing x8 and x16 then worry no longer.
The boys over at Hard OCP have done some pretty convicing testing using sli and crossfire at x16 and x8 and there is almost no difference !!!
Ok the board was not an R3E but I bet the results would be similar, maybe Ill put my Phys-x card back in ( tbh I never did see any drop in performance with it in there, but when I noticed gpu 2 in lane 3 was at x8 I took it out !)
http://www.hardocp.com/article/2010/...6x16_vs_x8x8/1
just put them where they look best :-)
You may have overlooked my post at the top of this page;Quote:
Originally Posted by Malik
Installed and apparently working fine on my rig.
just to report back:
my issues with temperatures are solved now! replaced the tim under the r3e stock sinks and this brought ioh/ich idle temps to about 60C! remember before this temps were rising quickly and reached 100C within minutes, thus the board simply switched off!
replacing the tim with some ocz freeze dropped temps by more than 40C! (expect any tim will do the same - asus stock tim simply is crap) heatsink is still fairly hot. ir thermometer showed about 58C on top of heatsink.
left the stock pad under the vrm sink. would not recommend replacing this as regular tim might not be able to bridge possible gaps between the sink and the mosfets.
reseated cpu cooler too. at first power on, cpu temp again rose slowly up to about 65C. but after the next powercycle cpu temp stayed below 47C. no idea why it ramped up like this at first power on. did the same at my first mount? anyway, as this the 775 cooler is a temporary solution i will ignore this for now and check how this behaves under water.
conclusion: replace stock tim on r3e!
Donmarkoni: That's it exactly. I appreciate your posts here, and on I4memory.Quote:
Originally Posted by donmarkoni
As for the others.., I have been at this for a while... rig is never set up becasue I am always trying to make it better. I spend more money on my hobbies than most here earn in an entire year. I have worked my ass off over many years, spent time away from my family... blah, blah, blah, to be in this position. I visit this board often but get tired of the same two always posting 'the same' crap. Yeah, their opinion is valued but to beat a dead horse, time and time again is absurd... Yes, I can be harsh and I am sorry if some are offended.... but I don't apologize for my opnion.
For me, the REIII has been great. I burned through 4 EVGA Classifieds. Great board but endurance wise it doesn't hold a candle to the REIII. I also like the USB 3 feature. There are USB 3 products available in Akihabara (Tokyo) that are not readily available in the US that I use.
the only reason you went through a classy and maybe 2 is that you couldn't overclock the boardQuote:
Originally Posted by Carrera997
the rexIII is just n00b freindly
Quote:
Originally Posted by zsamz_
Once again, common sense isn't so common.
Also, learn how to write and spell..... if English isn't your first language, I understand.
Zsamz you asked for that...or maybe you wanted that reaction? Can we try and stay OT please. But honestly I see where Carrera and Donmarkoni are coming from.Quote:
Originally Posted by Carrera997
If you are talking about CPU PLL, then for the 980x it should be around the 1.4v range.Quote:
Originally Posted by PytonOrm
@zsamz: I don't think I have read a helpful post from you yet, nothing but pissing and moaning. If you don't like the board then sell it and move on. For me this board runs better than the Classified or any other board I have had and I have had quite a few x58 boards. You keep griping about Bloomfield support and how the Classified supports it, but you fail to understand that every new bios from EVGA for the Classified has been directed at the 980x. Every new bios thread has someone complaining about this simple fact on EVGA forums. If you want Bloomfield support, and I am not saying the RE3 doesn't support it, then go with an older MB. Getting 4.2ghz on any x58 board is a gimmie now, but you shouldn't expect miracles for your chip, after all, you could have a chip that doesn't overclock well or you just might be a poor overclocker, who knows?
Yes. I read the entire article in your link. And the truth is I had already seen this kind of kind of test before.Quote:
Originally Posted by fmsam
In fact I once saw a post even here on our own XtremeSystems Forum where someone had tested a single
card in 16x and then 8x and shown the difference to be no more than about 10 points in some version of 3DMark.
I can't find the link, but I think it may even have been Ghostleader who made the post but not sure.
Only thing I disagree with the article is it says,
"...we are using the GTX 480 SLI because it is the highest "bandwidth" video card configuration there is and
produces the fastest framerates out of all SLI and CFX two-card combinations. "
Well this is not true. Still to date the Radeon HD5970 gives faster overall performance
(view full Fremi Vs. 5970 comparison)
http://www.maximumpc.com/article/fea...ungry?page=0,1
and has a higher memory badwidith than GTX 480 (177.GB/s for GTX 480, 256GB/s for 5970)
http://www.hardocp.com/images/articl...YXfv_1_1_l.gif
http://hothardware.com/articleimages...5970-specs.png
To Who it May Concern: I hate to repeat my same old story, but as you know this relates to my past questions.
I had my two cards in CrossFireX in slots #2 and #4. And though some people with a single card in slots #2 are
clearly getting only 8x
http://www.xtremesystems.org/forums/...1&d=1282361638Quote:
Originally Posted by havu2000
My own results with two cards in Slots #2 and #4 in CrossFireX were ambiguous. And even though I have now put them
in slots 1 and 3 it makes me wonder if in fact this board was somehow accepting to do 16x in slots 2 and 4.
(NOTE: Keep in mind I did have the PCI-E lane switches set to "off" for slots 1 and 3).
I agree. I run 1.35 PLL for my 4.4 daily overclock on my 980x. The same PLL works for my 4.74 benching oc with this Vantage score (57,726), running tri sli 480s:Quote:
Originally Posted by Utnorris
http://service.futuremark.com/compare?3dmv=2441431
Stock tim isn't that bad. At the moment on my test bench my IOH temp idles at 45c and SB is 39c on my board. IOH heatsink measured with a Fluke IR meter is 39.8 and SB is 39.1. Room ambient is 23.8c.Quote:
Originally Posted by fgw
I do however have two fans blowing across the board. One is a 120mm at ~1500rpm and the other is 80mm at ~1600rpm.
The same applied to my R2E. I just made sure contact was good and screws are tightened uniformly, before I started up the boards for the first time.
If you count the mem bandwidth in the same way for the GTX 480 in SLI it would be 2x 177 = 354 GB/s. I think that's probably what they referred to. The GTX 480 in SLI also makes more noise, is hotter, and uses more power. So, I don't think the HD5970 beats them overall :)Quote:
Originally Posted by Samson
:rofl: Haaa! Ha. Ok, I guess you are right. It is a great a great card, mind you, for a single GPU card. But not the best single card forQuote:
Originally Posted by cx-ray
SLI - CrossFire as they say
Excellent motherboard Asus Rampage III, have passed through my hands some of the best X58 boards.
The Rampage III is the card I've gotten cooler, temperatures are the best. A very complete bios.
Some of the results I've achieved via air.
Then use the Noctua NH-14 with an ambient temperature of 28C° ... 3 am ... :shrug:
single vga.
http://www.pctunerup.com/up/results/...733dmark06.jpg
crossfirex
http://www.pctunerup.com/up/results/...0316_32602.jpg
The only negative that I found to the plate, is that when I use 2 gtx 295, so there must be kept in the PCI Express ports 1 and 2. 2 vga being together, practically glued. Generating much heat.
The heat is so much that I was unable to pass the 3DMark Vantage and 3DMark benchs 06
Try to combine ports 1 -3 and 1-4 and it is not possible to run the 4 gpu. :mad:
Hi.Quote:
Originally Posted by CARLOS*A1
I am having a hard time understanding what you are trying to say.
But from your pictures I can see that you testes an Asus ATI Radeon 5800 series video card and you are getting
8x speed. Then you did CrossFireX with two Asus ATI Radeon 5800 series video cards and still getting 8x on both.
To get 16x with any card in SLI or CrossFireX you need to use PCI-E slots #1 and #3. Not slots #1 and #2.
For a single card to get 16x you need to use PCI-E slot #1. Or maybe even #3 (not sure).
What I was saying is that it still "may" be possible to get 16x in slots #2 and #4. But don't
try this at home !! (Result is still uncertain).
It should certainly be possible to run Dual-GPU cards in slots #1 and #3. This is what I have now.Quote:
Originally Posted by CARLOS*A1
I have two Dual-GPU Radeon 5970 in CrossFireX in Slots #1 and #3 now. And even had them working in Slots #2 and #4.
Please varify if you have done something wrong. Made some bad connections or not connected all the power cables to your system.
Also chech to see if by mistake you disabled the PCI-E "on-off" switches on the top right corner of the board.
Sorry for my bad English, and thanks for trying to help.
I use the PCI Express ports 1 and 3 and the computer restarts after 2 minutes.
The only way to hold the 2 gtx 295 was placed on ports 1 and 2 being very junstas.
I think it can be given the post pictures of your 16x and 16x ports 1-3.
It may be something to the picture that this bloked few buttons and some jumper.
http://www.pctunerup.com/up/results/...Escritorio.jpg
do you have the two 4-pin power plugs for the pcie slots pluged in by slot #1 and last slot#7Quote:
Originally Posted by CARLOS*A1
for two gtx295's it may need them
Yes, good idea. For GTX 295 it is probably a good idea to have them pluged, though I know for a fact thatQuote:
Originally Posted by bill_d
it would work with at least one. There was a girl here on this thread that had them working with only one.
So, we are talking about the EZ-PLUG_1 and EZ_PLUG_2. Look in the Instruction book on Page 2-46.
Make sure you have them conected. Also make sure you have a strong Power Supply Unit.
A single GTX 295 need from 31-35 Amps of current on the 12 volt rail. Multiply that by two.
Anyone managed to go over 230 blck on this board under water cooling?
yes, with 980 with any problem with ht on and NO slow mode at 245/250 bclk, and with bloomfield 930 but no stable at 230 bclkQuote:
Originally Posted by playground
Only for Pi 1M - bclk 233 HT off. I can boot up to 232 and with TurboV 1MHz more :(.Quote:
Originally Posted by playground
http://a.imageshack.us/img375/7656/pi8219.th.jpg
for all leechers here in forum
Bios 0011
http://filedeck.net/en-us/files/06K3...-ASUS-0011.rar
Ok. Great, but what does this do better (fix or improve) ?Quote:
Originally Posted by Fire4Wire
fixes some oc staion issues
http://img828.imageshack.us/img828/5...xbclktest1.jpgQuote:
Originally Posted by playground
http://img824.imageshack.us/img824/5205/qpitest2.jpg
As above, my REX3 x36 multi QPI limit & my 980X QPI limit.:D
These result are just Air cooling.
You are right, sorry, but 200x21 is so nice and round number. :DQuote:
Originally Posted by Scerate
I was thinking you want to go higher, as you are testing 406X MHz and 409X MHz IIRC. Anyway, try 200x20 and if it gives you trouble try 191x21, or 210x19. Be sure to set CPU multi manually and not leave it on auto, as I never seen it go over set value, at least on this board. What I'm saying is if you set it to 21x it should stay @ 21x and not raise one core to 22x.
Good for you. :up: It never helped me OC, only made it worse and that is on Xeon W3540.Quote:
Originally Posted by Scerate
CPU LLC helps voltage stay spot on, so usually you can lower voltage, as it doesn't drop under load any more.Quote:
Originally Posted by Scerate
thats what my board can do with Bloom's. SlowMode =/ Ill run LinX later on...
Damn than i have a creepy board cause i set it manually to 21x but the x22 or to be exactly the x21.9 multi kicks in :p: but no problem right now im running 203x20 @ 1.3000v set @ bios, but right now im playing around a bit with cpu pll to get 1.29x or even 1.285x stable :cool:Quote:
Originally Posted by donmarkoni
Edit: @ donmarkoni
i owe you one, got stable @ 1.270v first with 1.84 CPU PLL, think 1.3-1.4 was bit too low, and now i tried CPU PLL 1.7 and yeah atm 10passes stable which looks promising, most of the time i pass the first 5 LinX runs the other 20 runs stable too :). Idle temps are pretty low now 36-38-34-32
HA! i was lost to EVGA... (had a RIIE before) but i found my way back to my beloved asus and ze rampage III ... Yoohoo
From what ive read so far it is a good board !
Good to be back !!
...and this is my board with Bloomfield:
http://img94.imageshack.us/img94/154...lk240spi32.jpg
Can't run LinX, as my Xeon is on the shelf right now. :rolleyes:
Good to have you back! :welcome:Quote:
Originally Posted by AceBaran
So if this helps just one person its worth a post,
yesterday windows had an optional update (win 7 64) available which is a "Promise ACHI raid controller driver" .....hmm no idea what this is but foolishly I installed it, upon re start my ICH10R raid array was missing............oh dear !
luckily restore got me up and running but for some if the bootdrive is on the intel controller it could be messy, fortunately my boot drive at this time is back on the marvell.
donmarkoni - can you post up ur BIOS settings please? Really interested in ur settings. Just cant get over 220 blck stable. Thank u very much in advance.
Even though this memory isn't on the Asus Rampage III Extreme list will it still work?
CORSAIR DOMINATOR-GT 6GB (3 x 2GB) DDR3 2000 (PC3 16000) 8-9-8-24
http://www.newegg.com/Product/Produc...82E16820145318
Yes it will work :up:Quote:
Originally Posted by USFORCES
Nice bclk on air kurumi :yepp:Quote:
Originally Posted by kurumi_Japan
Thank you :)Quote:
Originally Posted by kill_a_wat
Don, how high can you get with QPI Slow Mode disabled? I noticed in your screenie that it was enabled.
Anyone know what the best full board water block is for the R3E? Thanks
I think most of us have the EK Full board.Quote:
Originally Posted by SkOrPn
EK waterblock here too, temps stay nicely around 35-39°C level
Ok thanks guys, EK it is then.... Does the Acetal provide any benefit over the all metal one? The black one looks better to me....Quote:
Originally Posted by Leeghoofd