![[H]ard|Forum](/styles/hardforum/xenforo/logo_dark.png){kind=logo}
EVGA SLi Mobo discussion thread. Come one, come all.
- Thread starter qbackin
- Start date
tbogstad said:
My prime stable cpu core voltage/highest clock is 2.77ghz (277*10) w/ RAM running @ 230mhz divider is 1.55v set in bios, reading 1.57 in myGuard. Can't run it @ 1:1 @ 277 under any circumstances, even @ 2t with very loose timings. I get errors even @ 1.57v @ anything above 2780mhz. 2800 only takes a minute or 2, 2780 might take 30mins to error out. 2770 is stable 30+hours in dual prime95 though. I think the chip is good for ~ 2750mhz @ 1.5v. I think the motherboard is finally holding me back on my maximum O/C now that i have this pc4000. Would be nice to run it @ 277mhz 1:1 and retain my 2770mhz O/C. Looks like it needs 295fsb to be stable w/ RAM @ 248 or so. Can't play CS:Source @ 300FSB (250ram) -- crashes on finishing loading a map. I'm running 2xprime and 1xOCCT @ 297fsb/243mhz RAM @ 3-4-4-8-1T. This is better ram-wise but i think 230mhz @ 2.5-3-3-8 w/ the CPU @ 2770(277x10) is going to be the better way to go since i'm all the way down @ 2676 (almost a full 100mhz
) now and still not positive it's stable for gaming.Max for the Gskill 2 x 1gb set is 260 cas3,4,4,8,2t and sandra is crappy at 2T, i was getting better bandwidth at 225 cas2.5,3,3,6,1T
My cpu won't go over 2.6ghz either, i tried rasieing the volt core up to 1.65V and it was no help.
i had htt at 310 and ram at 246FSB i think, but with 8 multiplier core was 2.48ghz and 9 multi was too high for the cpu since 2.6ghz is my max.
looks like i will be trying 230fsb with better ram timings, 1T rather than real high FSB and HTT.
My cpu won't go over 2.6ghz either, i tried rasieing the volt core up to 1.65V and it was no help.
i had htt at 310 and ram at 246FSB i think, but with 8 multiplier core was 2.48ghz and 9 multi was too high for the cpu since 2.6ghz is my max.
looks like i will be trying 230fsb with better ram timings, 1T rather than real high FSB and HTT.
My prime95 doing memory testing (had one with small fft all CPU and one blend w/ lots of RAM) died after 58 minutes last night @ 297fsb. I've got her @ 295fsb now and am testing away, but i have a feeling this just isn't stable. I'll probably end up back @ 275*10. Wish i could run my ram 1:1 @ 270fsb. I've heard reports of people running this ram @ stock timings with a small voltage boost @ 280mhz+. Board won't post AT ALL at 1:1 270+fsb, even @ loose timings & 2T.
Moto450fyz
Limp Gawd
- Joined
- Sep 20, 2005
- Messages
- 426
I lost the memory Controller on My MB, Just sent back for RMA, I hope the next 1 last a bit longer
could someone post the JetWay bios that works for this EVGA board, i am fed up with the evga bios, it sucks big donkey balls.
i can't overclock at all without seting 2T, and bank interleave diasabled, which totally destroys ram bandwidth.
I am ready to flash this thing right now!!!!
i can't overclock at all without seting 2T, and bank interleave diasabled, which totally destroys ram bandwidth.
I am ready to flash this thing right now!!!!
Moto450fyz
Limp Gawd
- Joined
- Sep 20, 2005
- Messages
- 426
No the Tech just seemed to think I got a Bad memory Controller, but I dont know they were good about it so we shall see on the next MB and see if it takes a shit too.
peacetilence
2[H]4U
- Joined
- Sep 23, 2004
- Messages
- 2,343
tbogstad said:
How do you raise the vcore above 1.55? In my bios it only goes that high.
Just to let you guys know, EVGA fixed the issue this motherboard was having with the Creative X-Fi soundcard, and released a new BIOS update to patch it.
I just tested this out myself, and the X-Fi finally works. Head over to the eVGA website to grab it.
I just tested this out myself, and the X-Fi finally works. Head over to the eVGA website to grab it.
I flashed to the JEtway BIOS no problem at all.
I can now run a little higher than i could with the EVGA BIOS, but it still leaves a lot of problems.
TO run 1.6V core, you have to set to 1.55V and the set the 7-thing below it to .05V and so on, you can set ip up to 2.0V core i think. it adds the 2 together.
1.55 + .05 = 1.6V core.
i can now run 1T a little higher but no big help, i can also run the cpu over 2.6ghz now, but i definately have the 270FSB problem now, you can boot over 270 1:1 no matter what, it set 275 back down to 270 no matter what you do, i have not tried A64 Tweaker though.\
I can not run 1t at anything of any reasonable speed, even pc4000 at pc3200 speed at 3,4,4,8,1t fails prime 95 in a hour, it only lasted 5 mins wit the evga bios, so this is a little better...
I can now run a little higher than i could with the EVGA BIOS, but it still leaves a lot of problems.
TO run 1.6V core, you have to set to 1.55V and the set the 7-thing below it to .05V and so on, you can set ip up to 2.0V core i think. it adds the 2 together.
1.55 + .05 = 1.6V core.
i can now run 1T a little higher but no big help, i can also run the cpu over 2.6ghz now, but i definately have the 270FSB problem now, you can boot over 270 1:1 no matter what, it set 275 back down to 270 no matter what you do, i have not tried A64 Tweaker though.\
I can not run 1t at anything of any reasonable speed, even pc4000 at pc3200 speed at 3,4,4,8,1t fails prime 95 in a hour, it only lasted 5 mins wit the evga bios, so this is a little better...
acroig said:
yeah i am beginning to think you are kinda right.
i could not run 1T at all with my Gskill pc4000 , but i just put some Corsair pc3200 value ram cas2.5,3,3,7 in and it is running 220FSB cas3,4,4,8,1T and it is prime stable.
where as the pc4000 Gskill was never prime stable with 1T.
so it is looking like the Gskill is not very good, i guess that is why it was so much cheaper then crucial or OCZ pc4000 1gb sticks.
I am wishing i would have just payed the extra $55.00 for Balistix, since it is proven to overclock well.
I tried the Gskill in a asus 875P top of the line board, and it does run 1T, but it does not overlcock well in that board either.
man i wish Newegg did not charge restocking fees, or i would return the Gskill and pay the extra for the Balistix in a second.
tbogstad said:
There's the problem. I bought the OCZ because my brother in law's DFI can take that all the way to 310-315 FSB 1:1 @ 2.7v.! But in the EVGA it stuck at 244. EVGA is well aware of the issue and the have promised a BIOS update to address this issue. I would wait a bit and see what they come up with before I send any more money. My .02.
Well, kept getting prime errors @ 290fsb/~260mhz RAM and i had to loosen timings @ 260 from 2.5-3-3-7-1t to 3-4-4-8-1T just to get it stable for an hour of prime. At the looser timings i was getting the same bandwidth @ 260mhz as i had been @ 230 w/ the looser timings. So i'm back @ 274*10 w/ RAM @ 230mhz or so, 2.5-3-3-7.
Benchmarks in games are pretty uneffected by the ram changes i've been making anyway, i think on an a64 you're just not limited by RAM speed in most cases.
Benchmarks in games are pretty uneffected by the ram changes i've been making anyway, i think on an a64 you're just not limited by RAM speed in most cases.
http://www.rhcf.com/sis-bin/ultimatebb.cgi?ubb=forum;f=14;hardset=0;start_point=0;DaysPrune=0
the stuff this guy say to do helps but does not solve my problems with 2 x 1gb sticks, i changed to 4 trtw, now it last loger, but still crashes eventually, so it helped but did not solve the problem with 2 x 1gb.
the stuff this guy say to do helps but does not solve my problems with 2 x 1gb sticks, i changed to 4 trtw, now it last loger, but still crashes eventually, so it helped but did not solve the problem with 2 x 1gb.
Evil Scooter
Supreme [H]ardness
- Joined
- Sep 22, 2000
- Messages
- 8,172
Moto450fyz said:I lost the memory Controller on My MB, Just sent back for RMA, I hope the next 1 last a bit longer
Then you really do have issues, the memory "controller" is on the CPU...
Moto450fyz
Limp Gawd
- Joined
- Sep 20, 2005
- Messages
- 426
LOL well thats what the tech said and he says its on the MB so who Knows, all I know is it is RMA for a new one And if it is the CPU its going back too for RMA, its only a week old too so no troubles here
And if it is the CPU its going back too for RMA, its only a week old too so no troubles here Moto450fyz said:LOL well thats what the tech said and he says its on the MB so who Knows, all I know is it is RMA for a new one
The issue is that the memory controller is integrated into the CPU on A64 chips, so when they told you it was the memory controller it just didn't make sense.
I've been thinking a lot about flashing to a Jetway BIOS as well, but I'm also afraid to do it. For the most part if seems that its worked for most people, but I've also read on more then one occasion where it hasn't. Is there a particular Jetway BIOS revision that is working on the eVGA board, and certain versions that don't?
Basically, can everyone who has flashed their eVGA N41 motherboard to the Jetway BIOS post which version they flashed to, and if it worked or not
Basically, can everyone who has flashed their eVGA N41 motherboard to the Jetway BIOS post which version they flashed to, and if it worked or not
dont take offence but you guys are retarded for not reading the previous pages. READ the thread first because all these things have been covered. the latest Jetway bios is the same as the E04 that came out. YES the evga's date is earlier they mention that messup in the evga forum. the latest possible bios for this board is the E042 which includes the x-fi fix and a partially working temp bug fix. both of these bioses are made by keith and co at evga and are considered beta. there is no performance improving jetway bios out there that will solve your high end fsb and memory timing issues. if you guys are at all having instability put your memory at a divider and lower your fsb below 270. there isnt that much performance decrease and i get simillar memory bandwidth running my ram at 180mhz vs 200.. the nforce chipset on this board can fry if you run ur fsb @ 290+ i rma'd mine after running at 295 for about two weeks and noticing perminant artifacting in bios even after flashing. please pm me if you have any questions about this board that are not answered so far in the thread and i will try my best to address them. so far my experience with this board has been quite stable and I am impressed at the performance I get out of it. Headless's config may not be reachable for everyone (he has a tasty venice E3 that got blessed by someone -> mine caps at ~2700 or so)
cheers,
-matt
cheers,
-matt
Moto450fyz
Limp Gawd
- Joined
- Sep 20, 2005
- Messages
- 426
Yea I know but thats what he told me, somthing on the MB and I am sure he said some type of Memory controller was bad on the MB Who am I to tell him I know more about thier MB then he does all I know is it will not run in Daul Channel anymore but will in Single Channel so I dont think its my CPU but I could be wrong as I am new to the AMD 64's. One thing for sure the ram is ok I slapped it in my other system and it runs fine. So as long as I get back a MB from RMA that runs in dual channel I will be happy
Who am I to tell him I know more about thier MB then he does all I know is it will not run in Daul Channel anymore but will in Single Channel so I dont think its my CPU but I could be wrong as I am new to the AMD 64's. One thing for sure the ram is ok I slapped it in my other system and it runs fine. So as long as I get back a MB from RMA that runs in dual channel I will be happy Hey guys quick question. I just updated using winflash, are you only supposed to have the first check box checked? WHen I looked on eVGA site they said just to leave everything like it is, load the bios and flash. Which is what I did. Everything is working fine. So will I ever need to check those other like 4 boxes? BTW my temps are reading 29C now, so I guess I got lucky on the partial temp bug fix.
peacetilence
2[H]4U
- Joined
- Sep 23, 2004
- Messages
- 2,343
I checked all the boxes I think. Atleast flashed each bios block, like DMI etc (don't have the program ATM so I can't look at the specifics) no problems yet but I haven't flashed with that beta bios, only the previous one.
hehe never listen to cust service reps. they are your bitches. be like bitch rma my board this warrenty exists for one reason alone. so you can replace my shit.
Moto450fyz said:Yea I know but thats what he told me, somthing on the MB and I am sure he said some type of Memory controller was bad on the MB
WhyYouLoveMe
Supreme [H]ardness
- Joined
- May 24, 2005
- Messages
- 6,513
grendelrt said:
I checked all the boxes when I flashed mine. All is well and stable!
Moto450fyz
Limp Gawd
- Joined
- Sep 20, 2005
- Messages
- 426
LOL it is RMA'ed and the Tech was most helpful and about 1000 times Better then Newegg. Newegg is my Bitch, they lost my MB for 3 days, I had to call them to find out what the hell was going on and they said, Oh We dont have it yet LOL I said check around it was there 3 days ago and I have the Tracking # to prove it and I have a SIG from your Doc person so Now I have to wait another 3 days for them to figure out how to wipe their A- - - man great service I must say and the best part is the MB is on back order so I don't know whats up now. Damn Me But EVGA was Most helpful so its all good there. I just hope I get a MB before the next week is over.
But EVGA was Most helpful so its all good there. I just hope I get a MB before the next week is over.sinewave said:
Hey man. Glad you posted your info here -- i tried running my board @ around 295fsb w/ 9x multiplier to see if i could squeeze some more performance out of my RAM (stupid.) and now my setup isn't stable @ the old 30 hours dual prime setting of 2750 @ 230mhz RAM. In fact, it fails in less than 5 minutes at that setting, and in 1hour @ 2700 2.5-4-4-8. &@$(*@(*$^!. That's what i get for being adventurous. I turned my FSB to 268(2680) and my RAM to 3-4-4-8 @ 224mhz (also lowered my voltage to 1.525 instead of the previous 1.55) and am dual prime stable for about 5 hours now. This is irritating because i can't really tell if it's the CPU that's unhappy (memory controller) or the motherboard... i run 1 small FFT prime and 1 blend -- the blend always fails. I think i'm going to try some memtest86ing w/ multi @ 9 but RAM set to 230-240mhz and see what happens.
i'm trying to decide between blaming 1.57v on the CPU(gradual damage?) or blaming 295fsb testing on my motherboard. I wish i could tell if the fucking voltage to my CPU is causing excessive heat...but i can't. On the plus side, changing from 1.55v to 1.525 dropped my full load temps from 72c in myguard to 70c. LOL. I'm going to try some tighter timings tonight @ 224mhz just to see if it's stable @ 2.5-3-3-7 for example. The RAM is rated @ 3-4-4-8 @ 250mhz so i'm hoping it can do 224mhz @ 2.5-3-3-7 w/ a voltage boost. Part of me wants to say fuck it and order a DFI just so i can see if it's my CPU or the board causing the instability. 2680 is still a good O/C, though, so i'll live.
headless said:
I wish I had a quick fix for you, but all I can think of is what I've said many times, it's all about RAM. Is there someone you know that could lend you some different RAM to try out? Sometimes I thought the BIOS was about to go when I pushed too far with the OCZ. With the GEIL it seems like there's no limit (but still no 1T). The rule seems to be that the RAM with the tightest SPD timings will perform the worse on our boards. Seem like a DFI is looking like the cheapest and best solution.
acroig said:I wish I had a quick fix for you, but all I can think of is what I've said many times, it's all about RAM. Is there someone you know that could lend you some different RAM to try out? Sometimes I thought the BIOS was about to go when I pushed too far with the OCZ. With the GEIL it seems like there's no limit (but still no 1T). The rule seems to be that the RAM with the tightest SPD timings will perform the worse on our boards. Seem like a DFI is looking like the cheapest and best solution.
I already switched from 2x512 of corsair to 2x1gb of crucial
We'll see what testing @ 2750 w/ memory speed of 180 does tommorrow if my testing tonight fails. I've jacked the voltage on the RAM up to 3.1v(myguard) (3.05 in bios) and am re-testing @ 2753mhz (275*10) with RAM @ 229mhz with 2.5-3-3-7-1T timings. It failed on these settings earlier in just a couple of minutes and so far it's been running for 45 minutes with the new voltage boost. So, maybe the RAM just needed some juice. We'll see if it holds out over time....headless said:I already switched from 2x512 of corsair to 2x1gb of crucial
Good luck!
Well! As mentioned above, i jacked the voltage on these crucial chips to 3.1v @ 230mhz 2.5-3-3-7-1T and left dual prime95s running overnight @ 275*10. I reset all my other settings to stock And they're still running! Apparently these crucial chips really like voltage to run at reasonable timings. They're rated @ 250mhz 3-4-4-8-1T... 2.85v wasn't happy at all at these tighter timings even though they were running under spec clocks... I'm going to let prime continue running all day. If it survives without an error for another 8 hours i'll call it a success Glad i didn't assume it was the board now...dropping memory to 180 would have fixed the problem i bet, but i would have got the wrong idea from that success and blamed the mobo or cpu where it was apparently just the RAM being voltage hungry
How about that fellow Leng, eh?
Glad i didn't assume it was the board now...dropping memory to 180 would have fixed the problem i bet, but i would have got the wrong idea from that success and blamed the mobo or cpu where it was apparently just the RAM being voltage hungryHow about that fellow Leng, eh?
headless said:Well! As mentioned above, i jacked the voltage on these crucial chips to 3.1v @ 230mhz 2.5-3-3-7-1T and left dual prime95s running overnight @ 275*10. I reset all my other settings to stock And they're still running! Apparently these crucial chips really like voltage to run at reasonable timings. They're rated @ 250mhz 3-4-4-8-1T... 2.85v wasn't happy at all at these tighter timings even though they were running under spec clocks... I'm going to let prime continue running all day. If it survives without an error for another 8 hours i'll call it a success
How about that fellow Leng, eh?
Good to hear about the RAM, it's a good thing this board can give plenty of juice to the DIMMs. The Leng guy is an ass. Did you see the smackdown I gave him?
acroig said:Good to hear about the RAM, it's a good thing this board can give plenty of juice to the DIMMs. The Leng guy is an ass. Did you see the smackdown I gave him?
Yeah, the DIMM voltage goes waay up on this thing -- that was one of the big reasons i decided to gamble on the 7800gt combo and get it. I figured, 'hey, if it can go past 3.2v in the bios for vdimm it can't be THAT bad of a board
'LOL yeah, i observed with amusement. Like i've never seen a damn soundcard lock with a solid tone before.
I went and looked some shit up on the squeal of death but none of the sites i found actually said with surety what's causing it. Just that it's a hard lock with a repeating tone. Duh.The first couple links were from 2002 and blamed a PCI timing chip on some via motherboards and dell models. I doubt that's the problem since it's still an issue today in 2005 but hey, GOOGLE ENLIGHTENED ME DAWG. Did you see that link to the signature page he posted? It's completely unspecific. it's like me going up and saying "i'm suing that company for! for ! for! STUFF! I DEMAND THEY FIX STUFF."
They got 10,000+ signatures and you can't even tell what the hell you're signing for.
"Better stability from creative products."
My creative product is perfectly stable...
headless said:Yeah, the DIMM voltage goes waay up on this thing -- that was one of the big reasons i decided to gamble on the 7800gt combo and get it. I figured, 'hey, if it can go past 3.2v in the bios for vdimm it can't be THAT bad of a board
LOL yeah, i observed with amusement. Like i've never seen a damn soundcard lock with a solid tone before.
The first couple links were from 2002 and blamed a PCI timing chip on some via motherboards and dell models. I doubt that's the problem since it's still an issue today in 2005 but hey, GOOGLE ENLIGHTENED ME DAWG. Did you see that link to the signature page he posted? It's completely unspecific. it's like me going up and saying "i'm suing that company for! for ! for! STUFF! I DEMAND THEY FIX STUFF."
They got 10,000+ signatures and you can't even tell what the hell you're signing for.
"Better stability from creative products."
My creative product is perfectly stable...
That guy is a low grade moron if you ask me.
acroig said:That guy is a low grade moron if you ask me.
LOL @ your last post over there.
dual Prime95s still running. It really looks like it was just not enough juice for the RAM. I'm excited now -- If it lasts until i get home tonight it'll be about 20 hours. 20 hours is stable imo
i wonder if it could do 1:1 @ 269 FSB (i've not seen anyone doing 1:1 above 270mhz) w/ 3.1v @ stock 3-4-4-8
Doesn't it figure that every time you solve one problem you create another task for yourself...