Wow sounds promising. Hopefully i get the same thing. Ill get to play with mine weds.Quote:
Originally Posted by gundamit
Printable View
Wow sounds promising. Hopefully i get the same thing. Ill get to play with mine weds.Quote:
Originally Posted by gundamit
DPBW is good, 0524, who knows! and YBBLE is strange. You tell us how it is once you get it :)
I got a DPBW but its week 12 and it does 300*9 @ 1.375V so its a good proc :).
Hope your week 24 does better.
Me too, but I'd be very happy to get 2.7ghz @ 1.375v and water. :) Before I test on Sunday I suppose I can daydream about 3ghz on water. It won't happen but its a nice fantasy. :DQuote:
Originally Posted by 187(V)URD@
Not too shabby for a week 20 :) , my mobo is roooted though, nothing ovrt 290HTT :mad:
http://img340.echo.cx/img340/6359/pi146wz.jpg
I'm at 2.7ghz prime stable why does my friend at 2.7 get 12500mips and I get 11300mips is ther any explanation. Oh yeah memory bandwidth 7.33gb/s
Have you tried booting with a lower multi at something like 9x295 then once in windows change the multi to 10 with clockgen? Works like a charm for me. :)Quote:
Originally Posted by alpha0ne
Yes m8, even tried as low as 6 multi with loose ram and I am not joking.......everything in between with every possible combination of tweaks/settings, I am now certain that it is the mobo but I am still quite happy as I now only need 1.504 ~ 1.520 Vcore (fluctuation seems ONLY when using 1.425 + 10% in bios ??, will reset to similar voltage using diff. combination later) to run @ 2.8Gig torture+ stable :DQuote:
Originally Posted by TheMeatFrog
Quote:
Originally Posted by afireinside
im running phasechange np and had to replace the winchester with this cause of the cold bug so i would know if this vince suffered from it trust me.so i gues the answer to your question is for me its not a problem....
off topic: in my bios (DFI ultra-D) i have two cpu voltate settings one is called CPU stsrt up value wich i set to 1.55v then i have another called CPU VID control and that one has the same settings as the first i can set all the way up to 1.55v!!! i leave it at auto but i have a feeling its why i can't get my voltage to ever change no matter what i do??? anyone help me?
peace perc,
On air its doing about the same as my Winnie. At $115 I can 't complain but this CPU isn't going to bump my Winnie out of my gaming rig.Quote:
Originally Posted by Absolute_0
http://img299.imageshack.us/img299/8...524dpbw4gz.jpg
The LLBLE 0517GPQW I hand picked at PC CLub looks like it might make the Winnie homeless when the new ATI mobo and video cards show up. :)
http://img273.imageshack.us/img273/6...gpqwair9qa.jpg
Nice week 17 there gundamit! Certainly fit to replace the winchester. Your sig says you have a Mach II, i think you should try it out on the GPQW :)
No doubt that will happen, but first I'm going to try a Swiftech kit (leak testing now) so I can see if it scales up a bit with slightly better cooling. I was thinking I'd wait to try the Mach II with the 0517GPQW when the ATI gear becomes availalble. If I get bored or curious it might happen sooner. Really hate to bust my gaming rig down to get at the Mach II. :DQuote:
Originally Posted by Absolute_0
BTW - XP-90/Zalman 92mm combo is my new air cooling champ. Really good value for $30.
Little fun with the 0515 while I slept :p:
http://img342.imageshack.us/img342/2695/img04039ql.jpg
http://img342.imageshack.us/img342/4932/img04072zm.jpg
I have that exact same chip, wondering the volts you use, what volts give you what mhz, my chip does:
1.450 = 2800mhz
1.500 = 2900mhz
1.550 = 2999mhz superpi 1 million
want to try higher but im on air, and it's the middle of summer here in toronto.
If we are talking 32M super PI my chip does 1.52v 2900MHz and 2950 max 1M with 1.55v. These 0515 APAWs run warm as we all know. :) They deserve more than just air cooling.
What was the highest 24/7 stable if you didn't have 24/7 stable then the highest 32mil.
Yeah I don't think Im gona go phase anytime soon lol, o well I just don't feel I know 'it really has a place on my daily pc, this computer is already hella fast, 2.9ghz with ram at 7500mb/s for daily use it's think is pretty insane lol.
Maybe if I can learn to build one I would, since I'm very good with that kind of stuff, is phase change loud and how cheap can you build one for?
That's something you would have to ask chilly1. Also remember that PC units add a good bit on your power bill each month. :slap:
My 24/7 stable settings are 2.9GHz at 1.52v. I've ran 32M at 2.91 and 2.92 but that was not good for 24/7 use since I play a good bit of CS:Source.
best i've done for SP1M is 2946 Mhz, ~1.62v. With more volts she'll do more but i'm waiting for the watercooling.
Hm, Im stable at 2.9 like rock stable lol I haven't really test much futher seriouly yet.
Are Pc systems nosiey, does it sound like the centeral a/c unit outside the house lol... thats pretty noisy.
Not much more with water.
http://img336.imageshack.us/img336/7...qwwater3og.jpg
This particular Venice seems volt hungry like the old Newcastles.
I'm going to keep playing with it. I need to test out a bunch of memory on it.
Anyone have a bios suggestion?
I can't get my system over 2650-2700MHz :(
* DFI LP nF4 Ultra-D
* 2x 256MB Infineon PC2700
* Venice 3000+ LBBLE 0517APQW batch 381 and LBBLE 0521BPJW batch 014
* Q-Tec 550W 20p, later replaced by other 500W 24p
CPU 1.61v
mem 3.2v
mem at 133 divider --> 200MHz at HTT of 300MHz, normaly runs 250MHz 2-2-2-5 1T
HTT multiplier 3x
CPU multiplier 9x
at 8x i also get stuck at 2650-2700 (335MHz HTT) so mainboard isnt the problem.
CPU also shouldn't be the problem as both CPUs run the same max speed
tested with sp2004
Boxed cooler and Thermalright SLK-948U + 80mm fan, same result with both coolers.
Not too shabby there gundamit :clap: , and your temps are great considering the huge Vcore you are using, do you use 1.76Vcore 24/7 ??
I am still stuck under 290HTT :mad: , I am now thinking it could even be the CPU that just hits a wall @ anything over 289HTT :slap:
Ok i think i just have bad luck and both CPU's are crap.
At 1.36v it gets unstable at ~2480 :(
This is with the 0521BPJW. Other CPU (0517APQW) i'm going to test after this.
There is NOTHING wrong with that, its about normal for a GOOD chipQuote:
Originally Posted by Erik1
Up the Vcore by to 1.45 and see what you can get too, also after a few days burn in it will be even better :clap:
Strange how the two CPUs I tested in my new testing rig (DFI Ultra-D mobo) both required a lot of voltage but ran cool. Maybe its undervolting. I'm going to break down my game rig so I can test the LBBLE 0517GPQW on my DFI SLI mobo. May not be a good comparison since my game rig has the Mach II.Quote:
Originally Posted by alpha0ne
@Erik1 - Got any other memory you can test with that rig? If not, throw up a screenie with your timings. Also try your best stick solo and see if that makes a difference with the CPU ceiling. Maybe even drop below 200fsb. Anything to eliminate the memory as the suspected weak link in the chain.
Keep us updated gundamit if it is the board reporting wrong, very interesting :)
I set the voltage in BIOS to 1.475, it says 1,440 in CPU-Z. Max stable freq. is now 2550. Above 2600 i get an error from SP2004 after couple of seconds. This is still with the 0521BPJW, this one isn't burned in well, so i test with my 0517APQW when i can find my thermal compound whitch i lost :p:Quote:
Originally Posted by alpha0ne
I run my memory at the 133 divider, so it always stays under 200MHz when testing my CPU's :) Normally it runs 250MHz. Timings are 2-2-2-5 1TQuote:
Originally Posted by gundamit
Don't have other memory, only slower PC2100 stuff and a Twinmos 2x512 PC3200 kit (not winbond chips) :rolleyes:
Good one m8, looks like you have a good OC'er at last :toast:
BTW I have found, and use as a general rule (though not always completely accurate but good enough) that any Vcore set in bios I always subtract - .0535 to get ~ approx. windows reading :)
My venice 3000+ 0517 DPAW :banana:
http://img312.imageshack.us/img312/4983/16m29ek.jpg
^^ Good one :banana:
Hmm weird, the 0517 does EXACT the same speed as the 0521 at 1.475, both start to run unstable at 2550-2600. Same goes for 1,65v, both max at 2650-2690 :confused:
Is this really the CPU's or is it something else? I tested MB up to 335MHz HTT, memory is at very low divider, bought new PSU that is better than my Q-Tec, temp is always under 60C, so dont know what it could be :confused:
I saw someone else with a 0517APQW go over 2900MHz on air :toast:
Also found this link: http://www.parkoz.com/zboard/view.ph...k_amd&no=19068
That's the same CPU as my 0521, only higher batch nr.
Edit:
Expirimented some more with the 0517APQW in my system, found a stable 24/7 setting: 9x289 = 2601MHz @ 1,55v, mem @ 166 divider --> 237MHz 2-2-2-5 1T (little slow but doesn't matter much)
hey all im @ work right now with access to this stepping of AMD 3200+ Venice
ADA3200DAA4BP
LBBLE 0521 APIW
wondering if this would be a decent upgrade (for free) from my Winchester 3200+ with this stepping
CBBHD 0450 UPIW
Need Help ASAP Thanks
PS. my winchester will do 2.88ghz.
Quote:
Originally Posted by Erik1
I'd really look at other aspects of my system (Have you tried mem @ 100 divider to isolate cpu? Have you made sure that your HT multiplier isn't too high (3x should be fine)?) My weakest cpu was my 3000+ week 16, and it was prime stable @~ 1.675v at 2725mhz. In fact all my venices seem to cap 2700-2850, the only real difference is the voltage they take to get there. Some hit 2600mhz easy then take a lot of voltage to hit 2700+. Some take 2700 easy and take a lot to hit 2775+, one took 2750 easy (~ 1.55v) then hit a wall and wouldn't hit 2800 at any voltage.
My Venice 3000+ with Giga-byte K8NSC-939 (F2)
ADA3000DAA4BP
LBBLE 0521APDW
using stock AMD heatsink.
max Mhz with setting in BIOS (measured)
1.4V (1.376-1.392): 2388 (9x265, 217)
1.5V (1.472-1.488): 2508 (9x278, 228)
1.7V (1.66) : ~2.73Ghz (now only 2.64Ghz) :sad:
All you >2.7ghz overclockers.. you lucky bastards.. and >2.9V on air.. dont get too cheeky ;)
So I got in my SLI-DR board now and I had to give it 1.8v just to get SuperPi 1M stable at 3.2ghz. Maybe both these Venice cores need more juice? Maybe I need better cooling now? :confused:Quote:
Originally Posted by alpha0ne
http://img260.imageshack.us/img260/7...x11sp1m5on.jpg
Your PI sux
Its to slow for 3200mhz
Any idea why that might be the case?Quote:
Originally Posted by OverClocker_gr
thats a pretty blunt statementQuote:
Originally Posted by OverClocker_gr
I dont know...
Maybe its preamable/max asynch...
What timmings you are using?
tCL : 2.5Quote:
Originally Posted by OverClocker_gr
tRCD: 4
tRAS: 6
tRp : 3
tRC : 7
tRFC: 14
tRRD : 2
tWR : 2
tWTR : 1
tRWT : 2
tREF : 3684
tWCL : 1
Dram Bank Interleave: Enable
DQS Skew Control : Auto
DQS Skew Value : 0
DRAM Drive Strength : 7
Dram Data Drive Strength: 2
Max Async Latency : 8
Read Preamble Time : 5
Idle Cycle Limit :256
Dynamic Counter: Disabled
R/W Queue Bypass : 16x
Bypass Max: 7x
32 Byte Granulation: Disable
Any suggestions would be welcome.
I have some BH-5 I'll try next but I'm not sure it'll beat my TCCD.
Hmmm your timings look OK to me :confused:
400MHz :woot:Quote:
Originally Posted by ***Deimos***
Hm maybe you could lower your Vcore a bit :DQuote:
Originally Posted by ***Deimos***
I'm having trouble overclocking my 3000+ Venice.. it's a LBBLE 0522. Is it that the older weeks are better overclockers? I can barely get 1.9 (originally 1.8). I am pretty new at this and I have read like that "isolate and consolidate" article for overclocking athlons. I have a A8N-E mobo which I know isn't the great for overclocking, but others with the same board get at least 2.4.. I also have some dual channel ddr400 patriot 2-3-2-5-t1 memory.. Any tips for me? Oh yeah, even when I change stuff in the BIOS, sometimes when Windows loads up, everything is back at default frequency, voltage, etc.
One thing I know is that it isn't a cooling problem because I have a zalman 7700cu and temps are around 33 idle.
Thanks in advance for help.
Edit: OK I'm just beginning to think my memory and mobo suck badly.
They are OK, a very good setting. Set the Dynamic Counter ENABLE will help for sure, noticable difference. But will affect the stability for sure, play with that by using the A64 Tweaker, you will notice the improvement.Quote:
Originally Posted by gundamit
It seems that newer the worst for OCing. The best so far are 0517DPMW and 0517 EPMW.Quote:
Originally Posted by danskim
To get higher FSB, you need to lower the HTT value to 4X, 3X is recommended.
Just noticed gundamit, you should set DRAM Drive Strength to AUTO or 3, lower is better for TCCD :)
I'll try that whenever I get my TCCD back in the rig. Its too darn hot to be benching so I decided to work on my game set-up which includes 2gigs of the ultra cheap and slow Samsung UCCC. Not much bandwidth but its smoothe in EQ2 and rock stable. Insane volts not required either. :) And before anyone comments that my SuperPi sucks :rolleyes: ... I know. Its actually slower than the one in my sig which is my Winnie/TCCD combo at high fsb and lower CPU clock. I just included the Pi to show stability.Quote:
Originally Posted by alpha0ne
http://img93.imageshack.us/img93/200...0133div8bz.jpg
continuing OC my 0517DPMW step by step.
I changed my 2 x 512adata pc4800 by Geil Ultra-X pc3200:
I have a problem with my mobo because it's impossible for me te desyncro my ram with any bios (I tested 310p, 510-3, 615 and 702-3) :slapass:
Can I send my mobo to DFI for that?
I must overclock my ram and get it stable before I overclock my cpu and it takes a very long time...
my screen:
http://img223.imageshack.us/img223/5...rime4lf.th.jpg
EDIT: desyncro works with bios 702-1 :toast:
i'm getting mad reboots at higher speeds.
For example
Prime95 @ 2.94 15 minutes, good, move on
Prime95 @ 2.95 15 minutes, good, move on
Prime95 @ 2.96 15 minutes, good, move on
Prime95 @ 2.97 INSTANT REBOOT
Then i tried SP
SP1M at 2.96 good
SP1M at 2.97 good
SP1M at 2.98 REBOOT
Is this the PSU causing the reboots? Enermax Noisetaker 485W.
Using my brand new watercooling, temps do not exceed 44C with 1.66 volts and 2.9x Ghz
Overclocking doesnt extend indefinetly.. I think you hit THE max for your chip.Quote:
Originally Posted by Absolute_0
Need a little help from the knowledgable types out there... I just recently did the cap mod and power supply mod for Giga-byte K8NSC-939.. got 3.3V at 3.61-3.62V... was running my 512MB BH5 x 2 beautifully 2-2-6 240Mhz.. and kept priming, and memtesting.. and then noticed some error creep in.. just 1 in 459% of memtest running (like 2 hours). Kept trying to lower clocks to get rid of this error... but it was still there.. long story short, I am now at 200Mhz and 2.5-3-3-8 and STILL its there.. I flashed to F3 BIOS, but no go.
Strange thing: A64 (0.31) doesnt show correct info anymore.. it used to have CAS=2, precharge=2 etc... but now its "reserved11b" and the newer one (A64 v0.5) shows CAS=4.5*... wtf
If you are hitting a wall like that, its probably the HTT/FSB... just happened to me the other day. When I raise 1Mhz too much, get lots of errors.. 2Mhz more and PC freezes... and sometimes reboots.
I got an 3000+ Venice stepping 0517 APAW doing rock stable (prime95 for 29 hours) at 300X9 with 1,45V (Smartguardian) on air with Thermalright XP-90 and Apacer TCCD PC4000 at 245 3-4-4-10 1T divider 5/6...but I can`t go any futher even if I rise the core voltage to 1,56V...I would like to do sth more but don`t want to rise the voltage to prevent overheating...so I keep it at 300X9 I can`t rise the FSB even to 305mhz at 1,56V withouth having an instant reboot when trying to run cpu burn-in or primer...:rolleyes:
Regards,
Primoz
Primoz78,
Sounds like you need to tweak the ram timings in A64 tweaker. Just changing a few things let me go from 15 minutes SP2004 before failing, to having it run for 9 hours and still going at the exact same speed. I think a few off timings could create this wall for you.
I have now changed the 5V jumper to OFF...and have tried some settings to ram...I have reached now 270 3-4-4-10 at 1T with Apacer 4000 before have reached at 1T just 245...but with my CPU nothing new...have tried a few settings at 305 at the same volts and 300 with 1,4V but withouth success...SP2004 gave me an error in 1 hour 20 minutes...will try to tweak the rame...but have now idea what settings to go for... :rolleyes:
Regards,
Primoz
Im not sure if that was typo but 1.4V for >2.7Ghz.. common.. use common sense.
Since you dont have access to higher multiplier, first make sure that 305HTT is feasable. How? Use 5x CPU multiplier, and then try 305 again (you can leave CPU voltage at default since the CPU will work at <1.8Ghz nominal speed). You should definetly be using 3x HTT, and perhaps up the HTT voltage and chipset voltage, and ofcourse make sure memory divider ensures memory runs <200Mhz. There are also some fine-tuning options that might be available in your BIOS. Apply same testing methodology you seem to be using already.. prime95 and SP2004.
Once you found your max fsb/HTT, use similar method to find max memory speed (if it isn't 270 already). And THEN, get back to the 305x9 problem you were having (use 1.6 or even 1.7V)... be patient.. I hope this helps.
Thanx Deimos for your suggestions I will try this today...and then report Ok. :)
Regards,
Primoz
Just a thought.....
It seems like most of the really high Venice overclocks are accomplished with a NF4 mobo.
Has anyone tried a Venice cpu in something like a Neo2 mobo then put it in a NF4 mobo and got better overclocks due to better power regulation or some other factor?
I am on the NF3 platformQuote:
Originally Posted by DrJay
But i agree, it would be nice to see a direct comparison between the OCing ability of NF3 and NF4 with the exact same chip and setup.
Check my sig, I did pretty well with my Venice and my NEO2 Plat. I am running the 1.9 retail BIOS now. I wish I had access to an NF4 board, but I did also use my old Abit AV8 (VIA chipset) and hit basically the same wall. My NEO does 300 FSB but doesnt like to run my ram with 166 divider for some reason. So, I just run it 1:1 and call it good. Making use of the 11X I paid for, lol.
I envy people who are hitting >2.7Ghz with 1.5V or less. Since I have an AGP 6800NU NF4 wasn't an option. Perhaps the overclocking is slightly better, or at least more consistent. However, that would mostly apply to the FSB/HTT. I can already hit almost 310HT. I'm quite sure I can't get to 2.7Ghz because of either heat (using stock HS) or just the limitations of the chip (more probable).Quote:
Originally Posted by Rickster_64
My giga-byte max cpu voltage (1.7V setting) is 1.626-1.648V, and when running prime95 at 2.7Ghz it seems ok for first few seconds, but then as it gets to around 50C, it errors out. 2.62 runs prime95 overnight.. but 2.626 will fail in minutes. I guess it can't be helped. Perhaps I will try to sell this chip and buy new one or get used one guaranteed with certain clocks.
Power regulation depends on the implementation by manufacturer, and less on the chipset (it just sits there).
And, undervolting or unstrable voltages such as in my case, would only decrease max stable overclock by maybe 50-100Mhz. Usually people hit HTT/mem limitation first.. but disregarding them, I think the intrinsic chip limitation is the bigger factor.
Quote:
Originally Posted by Absolute_0
Send me a NEO2 and NF3 whatever and I will gladly make a comparison between them :D
**00ps** I no more have AGP videocard :slap:
It´s the same but of course it vary from board to board.Quote:
Originally Posted by Absolute_0
Weird problem. I am prime stable at 2.7 with 1.552v. However I have one problem.
I can prime at this speed no problem but I have problems booting into windows. When I get to the screen with all the IRQ settings and device listings, sometimes the computer will not initiate:
"Verifying DMI Pool Data" and "Backup CMOS... OK!"
Then the computer will reboot. What could be the possible cause of this problem?
I dont care if it would take my motherboard's maximum of 1.7V (1.626-1.648 actual) to get to 2.7Ghz.. I just wish I could get there at all. I thought perhaps the problem was poor HS contact since I as moving the board around (lifting it by the cpu hs), so took it off, cleaned wth precision cleaner, and applied as5 (perhaps too much??), but my highest stable prime95 overclock is still exactly the same.
Just tested my 3000+ venice week 17 at the buyers request. Worst Venice I've used. Wont run 2600mhz stable at any voltage... every other chip I've owned ran 2700mhz stable minimum.
Did you still manage to sell it ??Quote:
Originally Posted by Lithan
I'm still testing it and keeping him up to date. Looks like nothing above 2550 will be prime stable regardless of vcore. Hits 2550 stable somewhere around 1.5vcore.
How many Venice have you tested? What was average OC (and what voltage)?
Do you think it would be worthwhile to sell my Venice (max 2.62Ghz) and get used or new Venice? Perhaps wait for cheaper San Diego, or just forget about it till 65nm comes?
IF that is completely regardless of Vcore then yes, it is in the lower end but I've seen worseQuote:
Originally Posted by Lithan
For a chip that only costs ~ $220AU doing 2550MHz that has been tested by someone who knows how to torture a chip to verify complete stability is well worth a premium on the original price IMHO :fact: :)
edit:especially when overclocked by 755MHz
Just for information, my old Venice 0516CPBW needed 1.5v for 2.4GHz :stick:
Quote:
Originally Posted by ***Deimos***
Let's see. I've had two 16's a 20 a 18 a 24 and a 15
18, 16's and 24's maxxed at 2700mhz with 1.65-1.7v, any more voltage got almost no gain. All did 2600mhz between 1.5 and 1.55v
20 and 15 were identical except add an extra 100mhz onto both. (So 2700mhz @ 1.5-1.55 and 2800mhz with 1.65-1.7v)
Personally I'd keep your venice.
I'm on MSI K8N Neo2 Platinum-54G now with 1.36b BIOS and Zalman 7700CU cooler.
http://koti.mbnet.fi/dani69/DFI%20nF...ice%200517.jpg
http://koti.mbnet.fi/dani69/MSI%20K8...re_1h32min.gif
..and the same CPU with DFI nF4/VapoChill LS:
http://koti.mbnet.fi/dani69/DFI%20nF...poChill_LS.jpg
I gonna try tomorrow 2.8GHz @ 1.53V AIRcooled :toast:
I've got my Venice running at 2.7GHz 1.47v. What frequency do you think that would compensate to get it running @ 1.55v?
Thanks
It depends on your CPU...an example...my venice is running at 2.7ghz at 1.456V prime stable 29 hours...but cannot go further even with 1.68V :confused:Quote:
Originally Posted by higherøx
Regards,
Primoz
If it scales normally, it should be hitting 2.8 Ghz.Quote:
Originally Posted by higherøx
Here we go :)Quote:
Originally Posted by Dani
http://koti.mbnet.fi/dani69/MSI%20K8...re_1h25min.gif
Wow, like you can't get up to 2750?? EXACT same with mine!! Tried with 1.648 v at 2750, failed prime in seconds :confused: :confused:Quote:
Originally Posted by Primoz78
Mine is causing reboot in few minutes...If I try to run OCCT or Prime...after a 3-5 minutes I find my comp restarted...:rolleyes: Now is for sale...I will get an 3700+ SD...will see if can get better with thisone :stick:Quote:
Originally Posted by pduan87
Regards,
Primoz
I wonder what causes that, you would think that the chip could run faster with a 0.1v boost...Quote:
Originally Posted by Primoz78
Thanks for your opinion guys... But which would you use daily: 2.7GHz 1.47v DDR600 or 2.8GHz 1.55v DDR620?
Thanks!
PS.: Zalman CNPS7000A-Cu 43ºC full load @ 2.7GHz 1.47v
I'd hit 2.8 buddy. Your temps are good.
Thanks =]
Think I found some good ones....
One local shop had several of each:
0521 BPEW, 0522 GPBW and 3200+ APAW
Another local shop had several of each of these:
0521 BPAW and 0521 DPMW
Went with one of the DPMWs. Results soon.... Wish me luck after a couple of unsuccessful 3200s.....this is my fifth Venice overall :rolleyes: .
The 0521 DPMW is running at 2646 at stock Vcore. Gonna get to some stability testing now........get some screen shots up tomorrow.
2700mhz @ default Vcore.
SuperPi time super slow with mem on 166 divider, 2.5-4-3-7 and 2T. Don't want the mem controller getting in the way just yet . :D
Vcore reading fluctuated slightly but stayed on 1.376 80% of the time.
Not to shabby for a week 21.
EDIT: mobo is Neo2.
How about Prime95/StressPrime2004 ;) ;) ;)Quote:
Originally Posted by DrJay
Don't care :DQuote:
Originally Posted by Dani
I run Pi, 3dmark and occasionaly surf. Thats it.
However, from what I've seen of this cpu so far...and compared to the four other Venices I've been playing with, I'm pretty sure it would pass.
On second thought, since this thread is primarily to share information about which steppings and/or weeks of manufacture are more likely to clock well, I'll run prime95 and maybe occt later tonight. Right now, its about 32 C outside and nearly as warm inside.....I'll let it cool off a little first. ;)
man, i need help
i have a venice 3500+ runnin on an MSI KN8 neo 2 and crucial ballistix pc4000
and the thing wnt do 215fsb no matter what i do....the following is what ive tried so far:
Spread Spectrum disabled
htt multi lowered to 3x
Multi's tried 8x-11x
vcore tried: 1.38-1.47
vdimm: 2.7/2.65
Bios Updated to 1.9 (latest)
ram tried in all 4 slots
still no boot above 215 fsb :[
btw there is this thing in dram config called
maxmemclock which was 200 by default, i figured tht was holding me back so i upped tht to 250, still no diff :S, any 1 got more details on max memclock?
and any 1 else got any bright idea what to do/try?
Max memclock is the divider. If your memory can't keep up with the HTT, you'll need to set a divider that's < 200, like the 183 or 166, to underclock the ram relative to the FSB.
Okay, finished preliminary testing (SuperPi 1 meg.) for 0521 DPMW while the forums where down.
Sorry for he lack of pics....I'm at work. Being very productive... :D
Max SPi for default Vcore (1.376) is 2,745mhz - 9x305
" for 1.425 bios (1.408 actual) is 2,790mhz - 9x310
" for 1.45 bios (1.440 actual) is 2,826mhz - 9x314
" for 1.45+3.3% (1.488 actual) is 2,871mhz - 9x319
" for 1.45+5% (1.536 actual) is 2,925mhz - 9x325
" for 1.45+8.3% (1.584 actual) is 2,971mhz - 9x330
" for 1.45+10% (1.632 actual) is 2,988mhz - 9x332 (at 2,997mhz errors about 3/4 through.)
Didn't mention the times as they are bad. TCCD @ 2.5-4-3-7, 2T. This is with air cooling on a Neo2 board. I did run Pi @ 2,979 with mem at 2.5-3-3-7, 1T (166 divider) and broke into the high 28's (have SS when I get home).
I think this cpu will clock higher when I do three things:
1. Run on DFI NF4 board - Better PWM an prolly higher HTT with NF4 flip-chip
2. Run on water
3. Adjust the crappy 12v rail on my PS.
Bottom line; there is still hope for us guys that missed the week 15 and 17s ;) .
Now on to prime / 3dmark / OCCT testing.
Jay
EDIT: I know I might get some cr@p for not having the screen-shots ready...that's okay ;) . At least I didn't post suicide windows posting numbers.
EDIT #2 = screen shot attach
Looks like you have a very nice chip there doctor, keep us posted please :)
Re ss shots IMHO they are a complete waste of bandwidth :fact:
my 0517DPMW: 2853MHz 1.45V with zalman 7700cu:
http://img361.imageshack.us/img361/5...mhz145v9gz.jpg
http://img374.imageshack.us/img374/6...pugraph5tg.gif
Hi everybody,
I can't see my rev in cpu-z 1.30 :s
can somebody explain why this is happening?
and I've got stepping 2 while most people have got stepping 0
here's a screenie
http://img390.imageshack.us/img390/2361/cpuzs9395qa.jpg
help me?
xyus
that's probably an E6 core ;)Quote:
Originally Posted by Xyus89
sorry,
I'm still a noob with the hole E3/E6 happening.
what's the difference between E3 and E6. wich one is the best?
(sorry for these noobs questions)
thx in advance
xyus
well my 3200 LBBLE 0515 APAW E3 us a total turd.
2700 tops without prime failing @ 2.7v
just like others mines failes at anything over 2.75ghz.
bah im so peeved.
it could be this dang NEO2 also.. i dont know..
An E3 is th eoriginal Venice chips.
An E6 is 1/2 of an X2 dual core.. correct ? or is that only the E6 sandiego ?
SanDiego are E4 i think :Quote:
Originally Posted by Lestat
Quote:
Winchester (90 nm SOI)
CPU-Stepping: D0
L1-Cache: 64 + 64 KB (Data + Instructions)
L2-Cache: 512 KB, fullspeed
MMX, Extended 3DNow!, SSE, SSE2, AMD64, Cool'n'Quiet, NX Bit
Socket 939, HyperTransport (1000 MHz, HT1000)
VCore: 1.40 V
Power Consumption (TDP): 67 Watt max
First Release: 2004
Clockrate: 1800 - 2200 MHz
[edit]
Venice (90 nm SOI)
CPU-Stepping: E3
L1-Cache: 64 + 64 KB (Data + Instructions)
L2-Cache: 512 KB, fullspeed
MMX, Extended 3DNow!, SSE, SSE2, SSE3, AMD64, Cool'n'Quiet, NX Bit
Socket 939, HyperTransport (1000 MHz, HT1000)
VCore: 1.35 V or 1.40 V
Power Consumption (TDP): 67 Watt max
First Release: April 4, 2005
Clockrate: 1800 - 2400 MHz
[edit]
San Diego (90 nm SOI)
CPU-Stepping: E4
L1-Cache: 64 + 64 KB (Data + Instructions)
L2-Cache: 1024 KB, fullspeed
MMX, Extended 3DNow!, SSE, SSE2, SSE3, AMD64, Cool'n'Quiet, NX Bit
Socket 939, HyperTransport (1000 MHz, HT1000)
VCore: 1.35 V or 1.40 V
Power Consumption (TDP): 67 Watt max
First Release: April 15, 2005
Clockrate: 2200 - 2800 MHz
hmm ,
I'm going to do some research in the X2.
but I think it's a sucky cpu :(
I wanted 2.8ghz on air, maybe I wanted to much.
xyus