Thank you, appreciated
Printable View
Thank you, appreciated
Anytime! :toast:
I just put in a 3200+ Venice, here's the stepping
ADA3200 DAA4BP
LBBLE 0517GPPW
Z644446E50150
Everything I've read about the GXXX chips is that they don't overclock too good. At 2.4 GHz (240x10) mine locked after just a couple of loops of car/lobby high on 3D01. Been priming stable (small FFTs) for about half an hour now at 245x10 with 1.4 Vcore in BIOS. That seems a little high for that speed, maybe this batch does suck?
ive been having some fun with my 3000+ LBBLE 0517 APJW recently.
my real reason for running this is to find its max speed i can run aquamark in, since im trying to break 14k.
right now i have it running at 2786mhz at 1.504v 1m and aquamark stable. the highest 1m was 2792, but aquamark locked up at this speed. i never tried anything other than 1m, as i just wanted it stable enough to run aquamark.
i was wondering what would be the highest voltage i could run though it on air.
right now it's at 9x309 1:1 with my OCZ memory @ 2.5-4-3-8.
any help to get to my goal of 14k with it would be welcome.
with a TT Big Typhoon, give it some volts. You can go to 1.6V no problems and many people go higher, 1.65(+). Personally i would keep it under 1.65... keep an eye on temps!
Umm, 14k? Isn't an average aquamark score for a system like yours more like 60-75k?
funny it says 70,500 in his sig, 14K must be a typo. Anyway most APJWs are pretty good (except my 2nd one!) so i've got faith that he can do more if it gives it more voltage.
i have a similar cpu:Quote:
Originally Posted by DB72
ADA3000DAA4BP
LBBLE 0517 GPIW
Z644398E51532
stock 1.35V
now at 1.55v 9x280
but i had to get rid off that heatspreader to reach that.
14k wasn't a typo.
@ the end of the test they have a scoring breakdown, giving a graphics, CPU, and Overall score. i am trying to see how high i can push my CPU, and im trying to get over 14k with my venice by that score
my highest score so far is AquaMark Score: 70483 (CPU: 13585, GFX: 9516)
my stock run i made on it was AquaMark Score: 54658 (CPU: 9261, GFX: 7752)
@ex2cib
Those cpu numbers look to be very close to sandra 2005 cpu scores :D at stock 2ghz I get 9300mips and overclocked I get 13500mips wonder if there using the same test.
local store only has 0516 GPBW and 0519 cpu's.... (3000+)
Any chance of a good clocker?
C
forget week 16 =[
both my week16s overclocked better than my week19s,
0516EPEW 2.7GHZ 1.6VCORE
0516EGPW 2.67GHZ 1.6VCORE
0519CPJW 2.6GHZ 1.6VCORE
0519BPEW 2.7GHZ 1.72VCORE
Man I've had nothing but "BL" with these venice chips :slapass: :bsod: .
Just paid for a LBBLE 0517APCW from ebay. I think the guy has more. We'll see how well it does on air and under TEC, but I'll have to wait for one of the two big crossfire boards.
Just bought an Venice 3200+ 0520BPMW
Will post OC potential when I have it up and running :)
9x312 (2808MHz) primestable 15 hours @ 1.5v :woot:
2900 didn't work, tried up to 1.6 volts, which BSOD'd within 5-10 minutes.
Currently at 9x312 @ 1.456v.
Seems like I couldn't even get my 0518APFW LBBLE stable at 2.6GHz even with 1.7+ volts :( I'm getting rather disappointed honestly....as you can see in my sig I have the new socket939 nf3 board from DFI so to say the least it's a bit flakey, but regardless I'm still seeing people hitting 2.7 and 2.8GHz with various venices and san diegos....mines being stubborn and I'm having to prime it at a measly 2.4GHz right now with 1.5 + .4 in the vcore options which registers about 1.675 according to voltage monitors....that much voltage on the earlier venices I've seen are hitting 2.8GHz+ easily and I don't know if that'll be happening for me.....freakin sucks
week 18s and 19s seem worst to me. There are quite a few 16s that are still pretty decent and a few 20s that are pretty good. Its a nice board anyhow.
Was that Pcclub by any chance? My local Pcclub has a load of week 16s,.....almost all gpbw. Even though it is a week 16, they are probably among the better week 16s available. I bought one, running at 2.88Ghz / 1.63 Vcore on Neo2.Quote:
Originally Posted by charlie
I'd not be surprised.Quote:
Originally Posted by Nubius
My old 3000+ 0516CPBW needed 1.5v for 2.4GHz, and 2.7GHz was a no goal, even with 1.7v++++ it was far from the stability.
Good luck
It's totally different for Aquamark CPU scoring than Sandra, but some processors are more similar than others. What kind of Aquamark3 CPU score are you getting at your overclocked settings, 4rory?Quote:
Originally Posted by 4rory
I heard 20's were pretty bad...Quote:
week 18s and 19s seem worst to me. There are quite a few 16s that are still pretty decent and a few 20s that are pretty good. Its a nice board anyhow
Anywho...I've tinkered with this DFI board more and I think I'm starting to get a feel for her...
When I had the CPU VID Manually set at 1.55 + .4, it wouldn't give 2.4GHz the time of day...
now...I've left the CPU Vid at pass through mode and have simply been raising the voltage adjustment and it seems to be doing a lot better.
Currently running 260x9.5 = 2.475GHz and it's running smoother than it has been any other time...voltage registering at about 1.55 compared to it's 1.7v with the VID manually set and it still wasn't stable at all....it's been running prime95 for 6 hours now while I watch shows on my computer, ran SuperPI, opened CPU-Z and photoshop all at the same time...no doubt this is part because of the BIOS being quirky.
My RAM seems to top out at 275MHz 2.5-3-3-7-2T 3v (1T will NOT work...I've heard some having luck with the yellow slots in this board but I haven't tried it yet) I tried 2.5-4-3-8 timings and it was BSOD's...this was at 275x9....now at 260x9.5 which is about 5Mhz less than 275x9 it's running strong....
My water coolings heatercore barb gets in the way of my RAM slots unfortunately so I don't know how easily I'd be able to change them over.
So perhaps this week 18 venice isn't too shabby....won't start pushing it more until tomorrow but it's doing pretty good at 2.475GHz right now at 1.55v so I'm really hoping I can get to 2.6+
Sorry for reporting a tad more than just the venice OC's :p
Oh sorry i've never run aquamark :S is it worth running how big is the download and installed sizes? And what will it tell me over other programs?
I like aquamark. it isn't a big download and it's really indicative of game performance. It's influenced by CPU, GPU, and memory. Good benchmark IMO. last time i ran it i got 82K but i was totally unoptimized and my drivers weren't good for it.
Well could get my week 19 to run more than 2479 kinda dissapointed. On the other hand atleast im using the rated speed of my memory. Not sure if i want to take a gamble and get another cpu, maybe a 3000+.
Id try giving it more volts but this board wont go over 1.45:mad:
Another thing is i cant get it to run at this speed in the bios, wont even post. Runs fine in windows though, go figure.
I primed it for 8 hours while i slept, sorry no screenshots of that, wasnt thinking.
http://i4.photobucket.com/albums/y14...alsloan/oc.jpg
Well, unfortunately these weeks 19 and 20's won't overclock very well.
The best overclocks I have seen are with Venices week 17, 12 and 15 (week 15 has a poor mem controller in many times).
Good luck
[]'s
there are a few 19's and 20's that have good overclocks not nearly as many as the earlier steppings.Quote:
Originally Posted by higherøx
Im kinda torn as to wether i should get another venice or san diego?
Just got this a few days ago from Sucka. 3500 CBBLE 0512DPAW Great chip. So far max benchable is 2.9ghz Here's a screenie. Primed overnight at 2.8with 1.61v, but decided to back off to 2.75 with less voltage. Funny thing is that with 11 multi I can't get passed 11x255. Tried 11x256 all the way past 1.7v, get bsod before windows. At 10x290, boots just fine. Using 5/10 bois. Any body have any ideas?
4X4N, some chips just like certain multipliers more than others, i can't explain it.
Hmm, is that so, I guess I'll have to try some other multis.
Delirious, the Neo4 platinum doesn't go above 1.45 volts? That sounds wrong, are you sure you're adding the %s to it? Maybe it goes to 1.45 but then you can do +3, 5, 8, or 10% to it. That's how it worked on my Neo2 platinum.
I have a week 15 3200+ Venice and it has no problem running 2.8Ghz but i would have to agree that it seems to have a weak memory controller. It doesn't like to run high memory speeds so i sold my TCCD.Quote:
Originally Posted by higherøx
Yes it's like here. My 0515 does 2.7GHz @ 1.47v but simply hates TCCD, so I had to disable the Bank Interleave, and even with it disabled it's not easy to get'em stable.Quote:
Originally Posted by burningrave101
But the war continues :D
Good luck for me... and for you too :toast:
my 0515 had an unbreakable wall at 2.75 Ghz, nothing seemed to help
hmm, maybe I should ask ocz if they will take my rev.2s back and give me some vxs :D Just maybe.. its worth a shot.
heh, if you do tell me if that stuff plays any nicer with the 0515 mem controller. I know *something* had to be giving mine that crazy wall.
Quote:
Originally Posted by Absolute_0
I saw that option in the bios, wasnt sure why they wouldnt just make it an option that scales normaly instead of having to work with percentages.
So the percentages work off of what the vcore is at. So if i have it at 1.4, 5% of that would be .07 ontop of the 1.45, giving me at total of 1.52?
Oh I'll report back thats for sure. I'm getting some results right now, there kinda weird ill post back in a bit with more conclusive results.
That's how it's supposed to work, but go really slowly with it because it doesn't work perfectly. You'll definitely be able to use more Vcore but it isn't going to be the mathematically logical #s they should be.Quote:
Originally Posted by Delirious
yah i figured it wouldnt be perfect. My motherboard seems to have a droop problem, probably a topic for another post, but it bounces from 1.34 to 1.39 when i set it at 1.4. Kinda sketchy, but its perfectly stable :confused:Quote:
Originally Posted by Absolute_0
1.34 to 1.39 is a HUGE variation. Mine has a little "droop" also, right now it is reading at 1.52 volts and when i go under load it reads as 1.504. And those are consistent readings.
Really thats alot of droop? On my neo2 for a few hours now, my readings are low 1.46 high 1.54 but i notice that with amd64 when i go under load vcore rises. When i set 1.5 i get 1.49 tho.
Sounds like RMA time to me
I had the same problem with a different MSI board, Neo 875P FIS2R, made it impossible to get a fully stable overclock :nono:
my old abaltron KX18D Pro II, used to fluctuate .1-.15 volts contiuiously. It OC'd fine :confused:
You should see the fluctuations on the POS board in my sig (NOT using now :D ) when set at 1.6Vcore in bios and FSB anything over 230 it goes from 1.488 upto 1.578, biggest POS I have ever bought :stick:Quote:
Originally Posted by blackjok3r
well right now i have it at 280x9 and 1.45 vcore, and its fluctuating between 1.408 and 1.45 according to cpu-z and speedfan.
Yah ill never get another msi, i already had to rma this board once. Just waiting for the an8 ultra to become available so i can pick one of those up.
Just resived my 3000+ venice. It's a 0520 EPMW, and right now I'm running with the stock cooler and stock voltage. So far I have been able to reach 2,9 GHz with stock voltage :slobber: but then it bluescreened :rolleyes:
I'll be testing it futher tonight, and hopefully I'll mount my Vapochill SE tomorrow :stick:
So after 4 venice CPUs, I have got a nice stepping... :banana:
eheheh... i put together my system couple days ago... at 1.49vcore bios (1.47 CPU-Z) i can stresscpu 2.97ghz forever... so happy withmy luck (i seem to be lucky when it comes to computer things like this... i buy cheap hardware and it ends up being sweet, ie.. i've had 3 a64s... 1 newcastle that did 2.6ghz, 1 winchester that did 2.75ghz, and now this monster) ... this is on hyper6...
3.0ghz is proving difficult past a certain temperature... and i need 1.56vcore bios to achieve it... but yeah, i need something better than air cooling (not to mention i need to lap the heatsink, haha)...
when i first suicided at 3.15ghz, i was like whoah wwwttfff... i got a sweet-as chip in my pc...
edit: cpu is an 3000+ 0517 EPMW... i had bought it during my exams a fair few weeks ago, cos I was reading of good results... never expected this though...imagine this on a vapochill... other hardware includes DFI NF4 ultra-d, G.Skill GH BH5 kit (very sweet RAM - but was VERY difficult to get it to play nicely with asynchronous overclocking).
@groovetek did you try prime95?, I'm not familiar with that program, nice chip tho man!
groovetek, nice chip :toast:
EPMW is a good stepping
You have any screenshots for us yet?
Hmmmm :( I have a 3200 0520 BPMW that so far have only been able to get 2400 default Vcore (1.344 in CPU-Z) :mad: , still early days thoughQuote:
Originally Posted by Leverpospade
Quote:
Originally Posted by burningrave101
Yes, week 15 seem to have a very weak mem controller. Didn't allow my BH-5 run with 2T timing @ over 245 2-2-2-5. Is there any BIOS can fix this problem ? ?
There are two things that can help you to solve this problem:
- Disable Bank Interleave
- Sell the processor :stick:
Good luck
My venice week 15 can run my Tccd at 290 and my single stick ch-5 at 260 1.5-2-2-5 with 3.4vQuote:
Originally Posted by vozbidder
My tccd at a claw hammer cpu was able to run at 315 at 2.8 volt....but now with this venice i was able for 290 at 3.1 v and 2.5-3-3-5 (always memtest stable)
Here with my week 15 I can run 300 FSB 1:1.
E aeeeee Higher!!! :)
Hi guys,
I was wondering how much voltage is safe to run 24/7 on a Venice 3200? 1.6v? 1.7v?
Thanks!!
AleX
E ae Alex.. hehe.. blz? :) :***
I think 1.65v would be a safe bet! In the air I wouldn't go for more.
Good luck!
I would stick to no more than 1.6V for every day usage.
Depending on coolingQuote:
Originally Posted by higherøx
If you have air cooling i think that 1.55 its ok
For water 1.65
And on dd 1.75
Quote:
Originally Posted by OverClocker_gr
Have you tried to run tight timing @ 250 2-2-2-5 ? Really disappointed about my 3200 week 15, althought it has impressive OC result :(
Do you guy have any BIOS to fix the mem controller problem ? PLzz show me 1 :toast:
Quote:
Originally Posted by vozbidder
So.. yes i tried 260 at 1/1 at 1.5-2-2-5Quote:
Originally Posted by OverClocker_gr
sry bro , i wanted to know about dual , not single stick :D
I think that there will be no difference if the controler cant handle it...
btw with my tccd at 2-3-3-5 i am able to run at 255
I had my vapo SE installede last nigth, but I had a lot of problems getting the 939-kit to fit. After 2 hours I managed to get it to fit, so it could keep an okay temp.
Right now I'm running around 5-10 C idle, but in load it sometimes jumpes up to around 35 C :confused:
But here is what I have been able to do so far:
http://photobucket.com/albums/a120/L...zsuperPI1m.jpg
Anyone know how to bring down the PWMIC temp, cause it's seams quite high..
My mobo won't clock over 350 FSB, so have anyone got suggestion to how i can get higher??
Hope you understand my bad english
ok here what i have done ... i changed my 3500+ (CH) to 3500+ ( Venice )
and iam very happy with until i reached with it to 2.65Ghz @1.43V.
the problem is .. During Prime95 CPU Torture TEST to see the Stability .everything seems OK. ( no Errors for over 2 hours)..but when i reach 2.7Ghz Prime95 does not give me any errors..no.. instead the comp hangs without any warnings..
so i think my PSU is the Problem ..right? i have 18A in 12+ Rail.
help plz
It could be, 18A on the +12V rail is kind of low, as i've said before, it's recommended to have 26 or more amps on the +12V rail for AMD64 Xtreme OCing. I have dual 18A rails and they kind of suck, instead of prime95 erroring i get voltage droops under load and reboots. A more powerful PSU would be a good idea.
Reboots or prime95 hanging is caused by low voltage, give it more and it should be fine... but again, i think the low voltage is associated with teh voltage drop under load, my voltage goes down by ~.016 under load, which doesn't sound like much but *perhaps* that's the difference between fully stable and not fully stable.
so what Psu do you recommand me to buy??Quote:
Originally Posted by Absolute_0
i need the Psu to be fully stabilty at very high overclocking.
thank you
I'm not sure, antec550 and powerstream 520 are pretty popular. Can't go wrong with PC Power and Cooling or a high end Zippy though ;)
will antec550 do the job perfectly??
Look my 3200 has a new friend :banana: I got it earlier today but about 15 mins in I had to step out. Should have some results later tonight! :toast:
http://img214.imageshack.us/img214/4...wfriend8xl.jpg
What PSU you use to get that high of Oc'ing??
Your +12v rail must have more than 25amps for stable o/c
can you guys tell me which is the Best PSU for high overclocking for around 170$ ??
and Btw iam running Q&C = Default, the 12+=11.84v and @ 2.65Ghz ,12+ = 11.56v Idle and 11.36v @ load then boom restart :mad:
I picked up the Ocz modstream 520watt yesterday it has 28amps on the 12volt and a 12volt peak of 40amps, so far so good, I haven't had chance to test it fully yet but yesterday I tried 1.75vcore in corecenter and it didn't shutdown. So mayne soon When the a/c is on max I'll try some higher vcore, does this board undervolt or what? If i set 1.60 in corecenter i get 1.52 in cpu-z whos right and where can i test with Multimeter?
I just noticed that Now with 1.6 in corecenter I get hihger vcores :D 1.55 now lol sweet so I can probably low it abit and still be stable, I had 18amps on my old psu looks like it really was a problem.
i found Vantec VAN-520A.. it got 28A on 12+ rail , 52A on +5 Rail and 26A on 3.3V ,,,what do you think?
other than that they got Enermax 600 and 550 watts PSU's and thos is really bad on 12+ rail ,, i know that they got 2x12+ but i dont need it cause they carry out 18A on both of them.
Yes but 18+18 = 36Quote:
Originally Posted by SNIPoR
thats what iam afraid of...Quote:
Originally Posted by Absolute_0
dual 12s doesn't work, you need to be able to pinpoint a lot of amps right to one thing, the AMD64. 1 strong 12v rail is better than dual weak ones.
my winnie did 2.6 on air and now its doing 2.6 on my water cooling, but i have the feeling im reaching the limit. is there a good chance i will see a real improvement going to a venice? like im talking about a 80-90% chance i'll hit 2.8-2.9 on water
Bad luck that your winny hit a limit there, doesn't scale well with voltage i suppose.
With venice it's hit or miss these days, if you order online chances are you'll end up with a crappy week 22 or something and be stuck at 2.5 Ghz. If you can manage to find the golden week 17, i'd say 2.9 Ghz on water... most of the time. I can almost do 2.9 on air but i'm completely temperature bound. my chip loves volts.
can't you just combine the 12volts rails, I know one of ocz's psus has a switch to do this, wonder if a simple mod would work. like shorting 2 of the 12volt lines.
i'm no PSU expert but i dont think you can just combine dual 12s into one big rail lol.
It would be pretty sick if i could instantly change my dual 18s into a single 36 though...
I think so, since it's just amperage, this way they could just share the load.
I know that if you take to 12volt batteries with 1 amp each and hook them up in perapell you get the same volt with double the amperage, that or seriers... lol its late and i've had a few beers.
luckly i found today Distrubuters for Ocz stuff here.. :banana:
they got Ocz Power Stream 520W for 135$. is this a good deal? and can i make soo sure about getting more than 2.65Ghz without any hangs?? i mean ,, i don have 26A PSU here..
one more thing.. ive done some Testings about my 12+v Rail.
Test 1 - pluged everything on ..then i got 11.65v idle and 11.55v load
Test 2 - removed everything that is connected to 12+v rail exept for cpu fan vga, hdd. .. i got 12.08v idle and 12.02v load..
but still i can exeed 2.65ghz without hangs :(
also to make sure about the my Twinmos Twister ram ...they are matched up, i used it in my old P4 2.4c ,,it was oc'ed to 3.337Ghz without any problems.
Ugh, my comp isn't stable 24/7 like this. I think it's the heat in the mornings that :banana::banana::banana::banana: it up.
A few times a week, when I wake up, I get greeted by a "No signal detected" screen.
Since the sun is heating this room up to 30-33 degrees before I wake up I'm pretty sure that is my problem, since it's ALWAYS stable daytime.
I wrote a small program that writes to a logfile every 5 minutes, and it seems to freeze between 05:30 - 07:00 every time.
Btw my CPU is LBBLE 0519 8PBW..any ideas bout this?
is it actually 8PBW? That's unheard of to me, i would expect BPBW or something but an 8? Strange. Tell us how she clocks
well ..from 2.2 to 2.5 is on Default then 2.5~2.65 is on 1.43. well i can tell when it will hit the max until i get OCZ Power Stream 520W in the of the next weekend.
cant hardly wait .. :devil:
Anyone tried a week 18 CPAW? Saw one at a local store. I like the *PAW part but a little scared about the week 18.
Also, has anybody found that the lot #'s actually make a difference?
Incase anyone in interested in a what might be good 3000+ its a ADA3000AA4BP
LBBLE 0515FPMW
some dude on these fourms had the same stepping at 2.8ghz with 1.6vcore, anyways heres the link i found it on ebay and asked for the stepping and week info.
http://cgi.ebay.com/ws/eBayISAPI.dll...DME:L:RTQ:US:1
cold bug on venice, whats the verdict? Rare? Common? Iffy?
Some are ok for single stage.
Just got my rig put together and I got a crappy chip. Is it safe to put 1.7 to it 24/7 on water cooling. I'll change the sig soon.
So does anyone feel that they've gotten better cpus by picking ones with low lot numbers?
From my reading, lower batch #s did seem slightly better on average, but my personal experience is contrary to this. I have an 0517APJW that does 2.8 Ghz with 1.52 volts, batch 1127, and an 0517APJW that barely does over 2.5 Ghz with 1.52 volts, batch 534. I think the 2nd CPU was a real outlier though... pretty bad for a week 17.
Ok I'm at 2.7 @ 1.7v 45c @ load. How is this for 24/7
here is my 0517DPMW on stock voltage, it was stable OCCT at 2700MHz at 1.34V, i just tested two sp2004 at the same time:
one with small fft and one with large fft:
http://img301.imageshack.us/img301/6...dpmw8ut.th.jpg
now i'll test at 1.34V primestable :banana:
EDIT: sp2004 do not pass at 1.34V
Thanks Absolute. I remember reading in this thread ( I think) when you picked up those cpus. I was wondering with two otherwise identical cpus if the one with the lower # would do better.Quote:
Originally Posted by Absolute_0
If you still have the one that only made it to 2.5, try removing the IHS. Maybe it just isn't making good contact.
The other APJW is stuck at my friend's house (that's where i build my PCs, i got a shop set up in his garage :toast: ). He went on vacation w/out telling me in advance so now it's locked up. Bad contact is a possibility, i thought it was really strange to have same week and stepping and for one to be such a dud. I'm sending it over to MaxxxRacer as soon as i can ge to it though. I have faith that he will push as much out of it as possible :).
I just got this at Ewiz (check out the Deals section for details) and it looks like they have very new weeks. Anyone care to make a prediction?
http://img321.imageshack.us/img321/8...524dpbw2vm.jpg
A quick Google and I see most of the hits for YBBLE are week 23 (no week 24) and are in the Europeon market. Saw one screenshot at 2.7ghz 32M SuperPi stable on air. XS member Fast had a screenshot (suicide?) at 3.3ghz on extreme cooling. Can't play with mine until Saturday. I'll post in the official forum with air results.