this is at the max fsb I've tried
http://intenseghetto.com/o/albums/be...300cpuzmem.JPG
my vcore I've been trying 1.50-1.525 w/ +10% so its around 1.70
Printable View
this is at the max fsb I've tried
http://intenseghetto.com/o/albums/be...300cpuzmem.JPG
my vcore I've been trying 1.50-1.525 w/ +10% so its around 1.70
I guess you know about half multipliers. Memory's not working at 300MHz.
300 * 8.5 = 2550
2550 / 9 = 283.3p
Remember, when using half multipliers the memory rounds up .5 .
See ya!
I can run 265x10 stable, 280x10, 279x10 benchable with HTT 4x...
hey guys, try bios 1.51! :D
http://xtremesystems.org/forums/showthread.php?t=47325
fixes nvraid issues and seems to give some people higher ocs :)
hmm... seems interesting... i will check it out... hopefully it doesnt have voltage issues like 1.41 did for me... had to go back to 1.36 just to fix the problems...
And saaya, go get urself a s939 board... or get a cpu for ur dfi board ur poor shmuck!
Quote:
Originally Posted by saaya
tried ! Sad 2 say that no better... ! n Back 2 1..36b. I have many option Vcore with this BIOS ver...
huh, i'm definitely having a hard time wth that board....
I just downloaded and flashed the 1.41 bios linked in the 1rst page and i'm having strange voltage values in corecell or mbm 5. It says 3.xx v for the vcore !!!! I went back in the bios and read 1.45 +10%vid which should be 1.6 or something, definitely strange. Also a strange thing i could not get more than 1.55 vcore voltage....
After fighting a lot to avoid freezes, blue screens and artifacts on the screen, i had to remove nv speed up and agp fastwrites so that the board is stable in 3d graphics. I have a gainward 2600 gs and i can't get more that the defaut frequencies (430/1200) and removing fast write let me loose 200 points in 3dmark 05 falling from 5501 to 5325....
I also have difficulties with o/c since i cannot have more than 220 fsb with htt 4*. The memory runs stable (11hours of memtest) at 2.5-3-2-6 2t but fails at 2-3-2-6. I have the ocz pc3200 el and when i tried to max the fsb to try and be as high possible 1-1 with proco. I was stuck at 260 * 9 * 3htt but with horrible timing (i dont remember but it was horrible like 3-4-4-12 2T). 10 * 260 fails, and i don't know if its a temperature issue or simply a voltage limitation.
Any issue with the 1.41 bios and voltage ? Any new possibility to activate fast write, and why is it failing ? I also search a good bios for my 2600 gs and how i could raise the voltage and have a try to o/c a bit more.
Well, went back to 1.4B1 bios.
Regarding some tips that Pedro Rocha gave us in another forum, I can now get 300HTT and more in my Neo2 with chipset week 10 of 2004.
I raise the AGP voltage to 1.8, I take off the vcore mod and use the core cell for that instead.
With my Mushkin BH5 PC3500 @2:2:2:6 and 2T disabled, configured in AUTO and 166Mhz in bios, and with my Vapo LS I get these values:
Benchable in some way...:
http://clientes.netvisao.pt/ratax/300HTT.JPG
Not benchable, yet :P
http://clientes.netvisao.pt/ratax/310HTT.JPG
I made few tests that were not enough to test rock stability, but will run some more and see what I can get with this chip...
that's what i'm getting so far....
3200+ , bios 1.37beta , adata bh5
Quote:
Originally Posted by jeffstet
ALWAYS disable ati/nv speedup and fast writes :D
and what card do you have? a 2600? :confused: never heard of that... what vpu is it based on?
use the 1.36b or 1.37 bios is 1.41 doesnt work for you, or try the 1.51... ;)
only 2.36b and 1.37 allow for 1.9v vcore afaik, 1.8v for newcastles that is i guess...
and why do you want to activate fast writes? :confused:
Ratax, did you get a higher oc by removing the vcore mod and adjusting the vcore in bios? or a more stable vcore? what did you set it to in bios before? the lowest value and then rose it with the mod?
j0j0, what memory and what timings? what ht multi?
memory 2*256 adata bh5 week 42 , ht @4 , timings @ auto
280 with 4x multi :eek: nice! 1150HT speed, should give you nice numbers in 2k1 :)
@saaya : hi, i have the 2600 golden sample from gainward. Its a 6800ultra board @ 430/1200 with 1.6 memory. I was just wondering why we had to disable fastwrite since i used it with my old fx5900, and that it increases global perfs ? Well anyway i decided to disable it as it was the only way to after a so-so stable PC.
Yestrday i flashed to 1.36bios and averything went much better. I'm currently at 11*225 and ran 6 hours of prime 95 and 5 hours memtest and played 2 hours doom. Everything seemed to be working well. This early morning i decided to reboot the pc, and wanted to run a few more tests and crashes in 10 mins of prime. It's so strange.
My temperatures are 32 idle and 58-60 in burn. The core voltage is at 1.67v and i can't go down unless instability. My memory is at 2.7v. My zalman cnps7000b cu runs at full and i have put artic silver 5 thermal paste. I know the temps are strange with this board and i hope these are not the real one. I definitely have a problem but where? voltage ? temps ? cpu limit ?
Initially i thought there were not enough thermal paste 'cause from 32° idle to 58°-60 (a pike at 62) it seems really a lot !!! More than 25° difference... So i had some more but the problem stayed the same....
post your complete system specs so we can help you.
and about fast writes, many people tested it and it gave them a really small performence boost. well you decide, fast write enabled+low oc+instability or fast write disabled+high oc+stability :D
and the second one is much faster than the first :D
@saya : For sure i'll take the second....In fact i was just wondering what makes fast write so instable. With my a7n8x e-deluxe and my asus FX5900 board i had good o/c and fast write enable. Is this a 6800u issue ? A mb issue ? A chipset issue ?
I was wondering if it was not an issue with the mobo since i had ( and still have a few) many problems with this mobo (unstable voltage (before bios 1.36b, crazy temps, blue screens, reboot, artefacts in 3d games, sound suddenly stopping in games and comes back by just quitting games, etc...). Most of these problems disappeared when i flashed to 1.36b disabled fw and nv speedup but i still have some issues : Temps, not so stable voltage (with my a7n8x and with the same PSU : no problems) e.G. : 11.87-12.26 at +12v 4.5-5.05 at +5 v 3.03-3.25 at 3.3v, memory with bios v1.2 or v1.3 could just stand 2-3-2-6 at 200 fsb and above 220 i had horrible timings such as 4-4-11-4 (before 1.36b). Now with 1.36b its much better my ram runs fine (6hours mem test + 4hours prime95 @ 2.5-3-6-2 which is good for me).
But when i say its instable i meant this morning for example i just booted the pc and ran a single prime just to test and it fails in less than 5mins (with the same settings 6 hours memtest + 4 prime95 +2 hours of doom+ 1 game of fifa and during that i had to quit 1 time doom and 1 time fifa for sound issues. With the mobo chipset i have no prob and the sound drivers are up to date). I just rebooted the pc and prime is carrying on (i started this morning 8 o'clock) and i just connected to my pc and he is still running. That's really strange.
I won't activate fast write nor nvsppedup and idecided to try and o/c a bit, let say up to 2.6 would be great if my proc can get there...
My system specs are :
PSU ENERMAX 600W Noisetaker EG701AX-VE
MSI NEO2 PLATINUM : bios 1.36b (the only that seemed to be stable), htt: 4x FSB: 225 mhz coef : 11 FW disable nv speedup disable temperature in bios : 21°/41° (at bootup) cpu : 1.67 v (1.525 +10%) Memory : 2.7 v agp : 1.5 v timing ram : 2.5-3-6-2.
AMD 64 3500+ (0.13ù @ 225*11 @1.67v)
2 * 512mb OCZ EL (2.5-3-6-2 @2.7v)
GAINWARD Powerpack 2600 GOLDEN SAMPLE (6800 ultra) bios 15.06 (@430*1200)
Audigy 2 ZS soundblaster
Internal Kortex ADSL Modem
TEKRAM DC 390U SCSI Controller
ide drive : Ibm DTTA 371010 10GO with w2k
Maxtor Diamond max plus 9 80Go
Western Digital 122 Go
SCSI UNIT : ULTRAPLEX 40x
PLEXWRITER 12/10/32s
Cooling : 1 NOISEBLOCKER SX2 (12cm) in front panel at the bottom to cool my 3 ide drives
2 Enermax fans UC-8FAB (8cm) (extracting a)
1 Zalman CNPS 7000b cu + Artic silver5 on the cpu.
I hope you have everything you need.
P.S. : i forgot to say that cool and quiet is disabled in bios and that i had to uninstall the a64 C&Q driver because with it i took me 1min for each rebbot in windows. I have tried the 66.81 66.93 and 67.03 drivers and my pc is best with these last ones. I also have the 5.10 nforce drivers installed. W2k is up to date (sp4 and all the hotfixes).
I've still heard conflicting opinions about that but w/e I was just posting a ssQuote:
Originally Posted by .: Eru :.
also saaya I don't know if my temps are correct but using speefan I get 40-44 idle and then 48-55 load
I have a big Problem, my system runs with 9*300FSB 1:1 prime and superpi stable with 3-4-4-8 and 2.6 VDimm Voltage. With 10*270FSB 1:1 3-4-4-8 i become "not exakt in round" error in SuperPI. I tested it with 2,55 till 2,85 VDimm but the problem is still there. The HT-Link i´m testing on 3x and 4x.
1.37B Mod Bios, 1.41B Mod Bios and 1.51 Bios all testing :(
well, 939-systems can be very strange...my combo (3200+, Neo2, TCCD) gets lots of errors in memtest @ 8x275 and 9X275. While 10x275 is completely stable!! an a higher multi on A64 actually increases the memory bandwith so it should be less stable at a higher multi. On my old Abit AV8-board the memory wouldn't even boot at 260, while 275 worked flawlessly.Quote:
Originally Posted by Riverna
So just keep playing around with your settings until you find one you like that works. But i don't see why you would want 275 when 300 works.
try using the other mem-slots to, helped me a little...
I have found what is my problem precisely but not the answer. In the bios the voltage of the cpu is 1.65. Each time i boot up the pc it seems the voltage get crazy. I open mbm in windows and find that the voltage is 3.3 !!!!!! I thought it was just a visual bug so i try to lauchn prime95 and he fails in less then 5min ===>reboot. Since the pc was not stopped but rebooted i went in the bios to see the voltage and it says 1.65 right!!! then windows boot and mbm shows the right voltage 1.65. So i tried to stop the pc again left it for 5-10 min and booted back and you know what mbm showed me te 3.3 v for core cpu again. I then rebooted directly this time whiyhout going in the bios and mbm shows me 1.65 rightly.
After cold boot i have 3.3v and i need to reboot to have the correct one.... Anybody has an answer for this?
i've a big problem with this motherboard when i insert a floppy disk and i read this appear a message "this disk is not formatted" but the disk function in other system there are a solution for a floppy problem in k8n neo2??? help me!
i've a big problem too, if i want to use my ocz booster and 2*256 A-Date nh5 it doesen't work but if i put one stick in , the ocz booster works fine , need help!
that means , i can increase the voltage on the two stick but nothing happend , no difference between 2.6v and ~3.9v , but with one stick and ~3.7v i can boot and run 32m without problems, and it makes no difference which stick i use.
bios is 1.36b
the QOS thing did not help,
I know you guys think I should just use a NIC, but i spent $130 on a product and it should work like its supposed to :p
Best bios for crap memory on Micron chips (obviously a ratio) with a 3200+ winch, good water? The board is relatively new - 0438 chipset.
0432 - OK week for a winch, or should I look for a later one?