lol @ that p-2 thing ;)
but... nb IS working.. you can measure it...
lol @ that p-2 thing ;)
but... nb IS working.. you can measure it...
:D
The P0-P3 thing is wierd...but dont do like me and feed the hungry 1066 mems with only 1.95v... it looks like they need 2.08v
Try to fix the another mess first :) , first check memmory with memtest, and after that go to the cpu thing... have you tryied ACC:Disabled/Auto/Fixed?
One more thing can you check one thing for me....i did something stupid this day and changed to singel core instead of auto..so i can have messed up something in windows(nvm looks good now haha)
I always set Vdimm higher right from the start, always around 2.1Vdimm or so.
I cant use ACC since this board is with SB600.
I suspected the memory system however that wouldnt make sense since Ive ran Memtest already. Not very long, but if there would have been serious damage you would find out very soon when starting Memtest.
Ill do a fresh Vista install without anything else, just OCCT and Memtest. What exactly do you mean with 'changed to single core instead of auto'?
Sooner than later Im going to use this board as a frisbee:shakes:
It just wont run stable at 3.7Ghz! And it was sure as hell stable on the 790FXB, un :banana::banana::banana::banana:ing believable. Tried different P-states, Vcore upto 1.4875, lowering Vdimm, using higher NB clocks etc. And it wasnt the memory, ran 125% coverage with 4 instances of Memtest, each 900MB.
This is getting retarded. Ill try and see if this board maybe has some preference for low multipliers or so, as far as a motherboard could have that, I dont know.
installed speedfan for the hell of it and ran it. system froze at scanning ISA bus. cpu did not clock the same afterwards (even a cmos clear did not help). had to do the ol killcmos -> flash -> killcmos -> flash to get it back.
fan on nb w./ velcro is helpin lots. reduced amount of heat gettin to rads internally. man can i hear the intake fan (after cleaning out) and chipset fan slightly. ah well, ill get used to it.
3.82GHZ on fxb-m2rsh. had to increase vdimm slightly to get it to pass after 6.5hrs (i was hoping it wouldn't be CPU-NB volts that needed to go up, and so far don't need to, so thats good..).
http://img128.imageshack.us/img128/7...phenomii3w.jpg
http://img128.imageshack.us/img128/4...lphenomii3.jpg
Oh...
Somehow my CPU has a wall right at 3.7Ghz:confused: 3.7 or over it just doesnt work (thus far).
I ran 15x246 for 3.5h at 1.5V, then I set 15x248 and it restarted in 5 minutes. Now I set 1.55V, if it will still BSOD/restart Im afraid Ive just some retarded batch:(
It's running a few minutes longer thus far, so maybe there's hope:rolleyes: PWM's do get hot though, in the 60's. I dont know what DFI did with PWM cooling but it ain't doing its job very well:rolleyes:
Danger Den Torture Rack
LanParty JR 790GX 1/21/2009 Bios
940 BE
Danger Den Custom Water
CRAP!!!!!!!!! Forgot to change screen resolution ...........any way its 3.8ghz @ 1.44v
M2RS BETA BIOS
well just flashed to the 02/02 lates beta bios and after a VERY preliminary test, it seems MILES better than the A01 version, (may be becuase I usb flashed this time instead of winflash)
but the bios layout seems better and NB multi acrually works.
so far ive had a much better and stable OCing with it.
@ brother esau: nice :D
Oh lol, my this motherboard is rather funny... in a bad way though.
Im slowly boosting up HTT with stock multiplier. Last night I actuall;y had it stable running 15x248, needed 1.55V though. Changed to 15x251, restarted in 10 minutes.
This morning I set 1.575V and it was still going after 40 minutes.
This is on 26/12 beta though, since you guys had issues with 16/01 and 21/01 I didnt try those yet, but could it be that either of those BIOS's has a better OCíng or do I just have a crappy batch:p:
Weird thing is that the PWM temp seems to be different everytime, always too high IMO but well. Ill keep this board for a while though however I do not plan to get a DFI board soon again with all those BIOS issues and bad cooling solutions.
Maybe a Crosshair II. Sapphire released a new motherboard too, but think it's for AM3. It's a 790GX board with some epic amount of MOSFETs, looks a bit like Jetway's 790GX board. I dont mind squeaking PWM's as long as the cooling is allright:rolleyes:
@pershoot - how did you set your CPU-NB volts - mine is at 1E (is there a way to see volts?) - I tried at 1F but was still not stable with NB @ 2.2 and HT @ 2.0 I matched your other settings but was not stable at 3.7 - passed OCCT at 3.6 (Itried up to +125mV)
As Folding was causing system freeze / crash I have backed down to CPU@3.4 +75mV/NB@2.0/HT@1.8
I think Im getting closer to the problem but there could be various issues;
1) Board/CPU doesnt like high HTT's and/or HTT's around certain speeds (with my Athlon X2 64 it refused to work at 266/267Mhz but over and below it would)
2) Board doesnt like too high multipliers
3) Board has an issue around Vdimm (testing now).
I succesfully ran 15x248 with RAM at DDR533. I set 13x286 and it was quite stable AFAIK. I set RAM at DDR800 and 2.16Vdimm and it all turns black after a while. 18.5x200 never really worked either:confused: And yes, the RAM does work at that speed. It's running for 30 minutes now with high Vdimm, but well, anything can happen in the next hour:rolleyes:
I requested a RMA an hour ago and asking if I could replace it with an other board, awaiting response. That's for my 790FXB-M2RS with the blown MOSFET btw.
Ill also RMA all the other stuff and just sell them.
Rammsteiner,
RE: P states, following your postings and figuring the 790fx-M2R is about the same except for the SB600, I set the P state to P2 in the bios since that was the setting that worked best for you, there seems to be a change in the way the unit runs/reacts (better).....will test further and see what is.....
Interesting stuffs, wonder if it affects the PWM reaction????
Laterzzzzz...............
WTF:confused:
Can anyone with a stable OC try and set a Vdimm of say 2.16V? If your RAM supports that of course, and test for stability again?
Either my PhII doesnt like high Vdimm or my board is just messed up, which might explain why I just got a mail from the store where I send my 9950BE for RMA to saying it ain't broken (memory problems, wouldnt read certain DIMMs etc):confused:
Annoying annoying:shakes:
Well, as a rule I hate going back to an older version if the current one works - anyway we can still hope for a new beta sometime this week :-)
Q1 - have you tried P2 or just P1?
Q2 - do you think ramping up NB and HT is necessary (I ask becuase some do it and some don't)?
1 - i have been on p-1 since i popped the cpu in; when it first landed in my hands. this is the most stable for me (i have tried them all). i have also advised others early on in other threads to play with this setting. the pII brings forth more p-states. i thought it was related to cool and quiet (which i have disabled), so not sure why this works in adding stability. ah well, whetever works, hehe.
2 - NB yea to get down the latencies. If you ramp up NB, HT likes to be at the same speed for most stability, in my experience. I think leaving HT at default (1.8) as well should also be stable with a higher NB.
good luck:)
Rammsteiner,
RE: ram voltage, I have been running 2.3 to 2.4 with my Adata stuffs, no problems, 2.3 is default for 1066 runnings, all four slots, added a .1 just in case.
P2 state still seems to be better....
laterzzzz...................
3.84 looked good when i left my apartment for work (was 3 hours in to occt). 3.86 no go (increasing vdimm to the next level made pc bounce itself during the night while stressing). 3.9 failed in about ~5 mins of stress at this voltage level, so for sure there is clock room left in this gem of a cpu that i have. maybe ill burn the hell out of it when im ready to upgrade next year, hehe.
i had something strange happen...
i was burning a dvd this morning and must have built up a slight static electricity charge from walking over the area rug in the other room and having my coat on. i went to go open the front panel a lil bit more to check on dvd, and felt a slight shock. video went out for about a half a second (then came back), and dvd failed to burn (i reburned and it went fine) and occt was still going strong. hopefully i didnt ruin any components from that one incident. i gotta remember to always ground myself prior to touching the pc!
fired up the Asrock 790GX/SB750 on the test bench this AM...it really can handle the mems....i stuck my pc6400 Adata stuffs in cause i didn't want to rape the main rig of the PC8500 stuffs....the PC6400 stuffs booted and ran the 1066 with pc8500 ADATA settings in the bios, something the DFI M2R just refused to do, period...hmm.............
running on stock cooling (actually a AMD heatpipe unit from and older 939 setup)......will put the huge thermaltake 120 heatpipe on later...
the overclock, meh, just about the same as the DFI except it is on krappy stock cooling, not pelt chilled water.............but I suspect the end result will be the same, just as Brother Esau reported awhile back....
going to tinker for a week or three and see what is....gonna load everest and start tracking the PWM temps....
and, with the bios set to P2 with my setup I can report it definately makes a difference, YMMV, certainly worth a try....
laterzzzz..................
Further updateness....Brother Esau please read....
Okay, I have the 'bad' 940BE in the AsRock, the one that really struggles beyond 3.8/3.9 on chilled water....same chip with less than stock cooling is now running a very neat 20 x 200 for the 4 gig @1.52v......and, i got it to post to 4.1 gig and get into winders, as soon as I clicked on CPU-z it rebooted, hey, it is a start..... again same kcrap AIR cooler....I really haven't spent much time on this, I have been tiling a bathroom at my house today, poking into the office to play for 5/10 minutes at a shot...
So, changing to a SB750 HAS made a huge difference in my case, but it is with a DIFFERENT brand MB, a $120 wonderboard with quality Japanese caps (no kidding)....
Bottom line, the DFI line of SB600, and very possibly the DFI SB750 as Brother Esau reported, while rock solid up to a point, both clock less than there cheaper brethren.......
Different brand MB, different results......we are talking 200MHZ/250mhz more, now I am wondering what the 'good' 940be will do.....
I am pleasantly surprised at the results, and no longer jealous of the 4gig and beyond posters since I can now 'do the deed' too......
For now, I am sticking with the DFI as I want to keep pounding away on the test bench setup to tweak and play and see if the stable clockage is better too......
So, it is the DFI board holding back ultimate clockage....both SB600 and the SB750 versions....doubtful a righteous bios will magically appear to change all this, especially since they have had numerous shots at making updated bios's...........
I spose this will infuriate some, but it is what it is...all this time chasing the 4 gig, all it took was a different board.....
Laterzzzz................
new beta for DFI LP DK 790FXB-M2RS|H
http://img.lanparty.tw/Upload/BIOS/CM/R7SD224A.zip
Major Reasons of Change:
The BIOS for AMD Phenom II 65nm ACC code built.
Saw that....Still no Love for UT 790FX-M2R Owners though:mad:
bios disables the onboard lan adapter (whether its enabled or disabled in bios).
maxmemclk is gone (i have 1066 spd on my sticks, so auto allowed me the tighter timings @ 1066).
acc still causes a no POST
some new DRAM options such as power down or something to that effect
c1e support
did not spend a minute clocking it, due to the network adapter being disabled. back on 1/16.
Contacted Frankie @DFI inc Taiwan 2day and waiting on response from him for my collaboration efforts with D.F.I:)
Anfi-tec water blocks are here & they look fab :up:
nice :D take pic of the equipped board pls :D
be sure to put enuff thermal paste under the pwm cooler (i used arctic ceramique was better then as5 and you dont have to fear conductivity) and do not overtighten the tiny screws that hold the delrin caps... the copper screw holes will be thankful :D
WARNING to all DK 790FXB-M2RS(H) users regarding BETA BIOS 02/24:
Do not flash this beta unless you have AMD 7xx CPU:
Major Reasons of Change:
The BIOS added 65nm ACC code for AMD Phenom II 710 / 720 CPUs(45nm) only.
@Olli.............could you post a pic of your 790FX-M2R with water cooled loop of PWM , NB ,CPU? Wanna get a visual of those two Antitech Blocks run in the loop.:yepp:
Also on new SB750 boards NB/PWM are inter connected and top fins comes off of NB and leaving the rest of the assembly in place you can put the NB Water Block on top of it:yepp: PWM runs much cooler on new boards where the NB runs very Hot
Hi.
Any clues about AM3 boards from DFI ??
I'm waiting the UT model to build a system for my brother. he clearly wants a ddr3 build albeit knowing the small gains compared with the ddr2 variant :rolleyes:
best regards
@ be:
http://www.abload.de/thumb/dsc02566zxdv.jpghttp://www.abload.de/thumb/dsc02565i07d.jpg
thats what i have atm.. you need more?? ;)
http://www.abload.de/thumb/dsc025671uns.jpg
like this?
btw... those harddrives in the first picture... the comp is lying on its side on that pic.. they are not glued there or something xD
hehe nice experimental setup campbell :D
yes i think that kink is a problem depending on what type of tubing u use... but i think the one u use looks like it could do that.
perhaps there are ways to connect in an other way...
I use 1/2" (ID) 3/4" (OD) Primo Flex Tubing
I Ran Hosing lengths long so I could re-use when the Right stuff gets here
BTW....The onboard Graphics of 790GX Plays FEAR 2 on high settings without issue and no LAG ...very Impressed with onboard Graphics Processor from AMD
@ Brother Esau
you use the same pump like me..........D5 power :rocker:
and i use this heat exchanger only for cooling the CPU
http://www.abload.de/thumb/morrajyqp.jpg
and now every of the two HD 3870 wil get one of this pump http://www.webshop-innovatek.de/shop...12V-Pumpe.html
and one of this heat exchanger for cooling
http://www.abload.de/thumb/htfstdh9.jpg
maybe next weekend.......... i need 2 x 4 skew (the same of the big heat exchanger) to assemble the two smal heat exchanger on my case....but watercool dont sell the skews alone :-(
where can you find that crazy stuff at!!?? thats a hell of a radiator for just the cpu
You can't beat German Made Tools and Products:up: I have been buying German Crafted Tools and products for many years now and I have always found the the German craftsmen and Engineers to be probably the best in the word and I have owned the absolute best from around the world and Austrian /German is the best :clap:
the big one is called Mora 2
it is the best heat exchanger wich i know.
i have on it 9 x 120mm fans and so i have a water temperature wich ist maximum between 2°C and 3°C higher then the room temperature
if you are interested, i can ask the shop where i bye it if the could send it autside of austria
the big one on the picture costs 85 Euro
but you can bye the same in chrom who lokks so
http://www.abload.de/thumb/20310rc4c.jpg
this one costs 120 Euro
and if you want to use the heat exchanger with fans you need a fanpanel who costs 35 Euro
http://www.abload.de/thumb/205705nf1.jpg
@Olli...........will 1/2 Barbs with 1/2" (ID) 3/4" (OD) Tubing fit on that NB Block? ...Or will I have to use one of these Fittings?
Heres the link for the fitting...............http://www.sidewindercomputers.com/ferocofig1th.html
i think barbs wont fit unfortunately. there's 1.8cm between the middle of those G1/4", calculating 3/4" OD tubing it would sit side by side tightly if fitting at all. So I think mounting the barbs will be complicated at least.
Those sidewinder fittings look good. They will fit i think..
Lol
If you look for example at watercooling products I really do think those German products look a lot more solid and durable.
But they're not so much into highflow items but more into efficient heat transfer. I really wanted some highflow stuff so I got an EK Supreme block and 1/2" tubing. Then again EK is from Slovenia anyway.
Dangerden and all that stuff is good too, however they just look 'cheap':p: And to be honest, if I had the money I wouldnt mind putting 3 TFC rads into my TJ07 with Aquacomputer Sterling .925 Silver blocks:cool:
ok i got it they feel good EVEN in that plastic wrapping :lol:
lol
I swapped boards two days ago, will be keeping an eye on developements, hoping against all hope....Brother Esau, you have the patience of a Mormon....
I have used DFI for so many years, have many different intinerations (754,939 and 940, some work, some are fried, all stellar performers in their day), I feel like I am getting divorced again...
Anyway, I will be keeping an eye on the bios developement, if it can be fixed and Oscar spends some quality time, maybe improvement will happen.....I switched only because of the PWM problems with my board, which may or may not be a quirk particluar to my board....
The PWM and MB temps are much lower and very stable on this cheap replacement (and it has high quality Japanese caps), but, even though I can get the 4 gig now, the ultimate stable clock did not change much, so it was a matter of not wanting to live with the PWM probs that motivated me, all other aspects of the DFI were spot on for me....
Good luck...
Laterzzzz.................
i am on 05/05 bios now and for so far no problems with 3,75ghz @ 1,44v...
just it reads "unknown cpu" lol
http://www.theinquirer.net/inquirer/...atx-x58-boards
Its a cross-breed of an MR2 & DK 790FXB-M2RSH with AM3 DDR3.
Non socketed BIOS :yawn:
Looks liking the heatpipe mounting holes are the same as the MR2 :) .
Looks like the cooling improved a lot:rolleyes:
Anyway, it will be a long time before I buy their stuff again. All their boards look nice and solid but their DK boards are just underrated for any serious stuff IMO and their only UT board thus far has a retarded BIOS + scary PWM's which make me pee my pants since they sound like an Atom bomb might go off any second.
very nice board! only the non-socket bios is a drawback really. And i'm glad that it uses digital pwm section again but with better cooling.. Love this board on first sight :peace:
Olli, here is work in progress about another week and a half and it will be show room finish.
Just waiting on German connection and some other stuff;)
Olli. what fittings are those on NB Block please ? Also link to product page would be nice also Look at Performance Pc's and tell me please what Koolance or Bitspower fittings will work on that NB Block;)
Would these fit>>>>>http://www.performance-pcs.com/catal...ducts_id=23846 on the NB Block then add my Koolance Compression fittings>>>>>These >>>>http://www.performance-pcs.com/catal...ducts_id=24526
Just wondering how come they remove the CPU-DID and NB-DID on 2/02? I thought it was nice option to change multipliers by .25 instead of .50 same goes with NB multiplier.
This is on 790FX-M2RS.
Much more to come I promise you that;) Just beginning stage right now but I assure you in the days ahead to come D.F.I will be holding the Crown for AMD Motherboards Once Again:yepp:
I cannot disclose anymore then what I have just said so please be patient and Know that D.F.I does Care and we are in the Process of getting this Situation to do a complete 360 Degree Turn Around:up:
I heard DFI wtfpwn for clocking, and XS was awesome with a DFI board during the Las Vegas event, but hey I guess its my fault for not doing further research before buying a new board? :shrug:
My BIOS is kinda buggy in general and I really hope DFI releases a new one, official or beta i don't care.
i have found 1/16 to be top notch for the most part with a 940. very stable. not sure what they did to the internals after, but i just retested 1/20 official (i redownloaded it just in case) this weekend, and found it to be completely unstable at my current clocks (reboots during first few mins of stress and some other slight general wierdness).
need some helps here guys. i'm using 790fxb w/ beta bios 02/24 & i'm having problem with cpu-nb vid adjustment. it's default @ 1c, whenever any change is made (even 1 notch up to 1d), i can't boot into windows sucessfully (most of the time i'll just get a reboot after a loading bar or blank screen). :shakes:
in order to increase nb vid i gotta use aod, which not giving me any problem at all. in aod the default nb vid is 1.2v (i think it's kinda low, isn't it:confused:). pls help if anyone of you have any input. thanks. :)
http://i41.tinypic.com/1441c9s.jpg
NB Hex @ 13 is usually good for up too 2.6ghz on NB Clock speed:)
I hope everything good comes out of any endeavor that XS puts forth to help out manufacturers, in this case DFI. I loved DFI back in the good ol' Barton core years, I had my single barton 3200 running 2.8ghz @ 1.84v and it topped off at 52 C on AIR. original clock was 1.8ghz, oh man did I love it. Things were simple then though....
Resurrection.
http://www.scan.co.uk/Products/Sapph...-SATA-RAID-ATX
hehe.. yeah.. they were sold on by sapphire all the time :D
But sap stopped for quite some time selling them.
And the MR2 can still be had at the same place for much less.
http://www.scan.co.uk/Products/DFI-L...-SATA-RAID-ATX
sapphire was really stopping them? in germany the m2r was nowhere available but by sapphire...
lol lovely board for that price there :D
what would be considered too much volts on the cpu-nb vid? Right now im at 1d wich is 1.1875. but my nb isnt stable over 2200mhz. if I up the the nb vid higher will it stable it out?
heres what im stable at now.
http://valid.canardpc.com/cache/banner/527036.png
thats with the 8 multiplier on the nb. when I change it to a 9 multi I get instant errors in occt.
at 8x my nb is at 2176, at 9x its almost 2500.
you can give 1,4v...
and thats for the cpu-nb vid in hex right. My nb I normally give 1.42v but never really messed with the one under cpu settings. I dont want to fry something.
EDIT:
sweet, I upped it to hex 13 which is 1.3125. OCCT no long crashes instanly and my mem bandwidth went up. 1.31 safe for air cooling.
think so... but 1,42 on the (board) nb is much... better leave it at default should work also.. Never had to uppen it so far...
http://i158.photobucket.com/albums/t...istic/3510.jpg
here is what i got stable so far.
Stoner@Large what kind of cooling do you have on the NB. i mean the little heatsink between the PWM and the SB.