The Official San Diego Overclocking Thread

[conrad.maranan]

In an effort to maintain the orderliness of the AMD section, I am requesting that all San Diego overclocking results be placed in this official thread. Please refrain from starting a new thread to highlight your overclocking results. If you feel that your overclocking efforts are beyond average and worthy of an individualized thread, please consider doing so in the Xtreme Overclocking section.

We do not need 1,000 new threads on the same topic. Look at what my San Diego can do!...San Diego 4000+ OC Fun...My San Diego hits 3000MHz on air...Mr. Forum Member's Mega San Diego OC Thread...The San Diego/H2O OC action thread... You get the picture.

If you wish to start a thread with regard to the San Diego core which also includes some other issues, dilemmas, questions, and/or concerns that have not been mentioned elsewhere, please feel free to do so.

Your cooperation with regard to this matter allows for a much more organized and systematic way to share information. Yes, this is a group effort.

I will open this thread as soon as a forum member sends me (or any of the AMD section moderators) a PM stating that he/she is ready to share some results.



RULES FOR POSTING IN THIS THREAD
This thread is strictly for overclocking results with a San Diego core processor. Overclocking results must include at least one screenshot along with specific details regarding your accomplishment. This includes, but is not limited to, the following data:

- Stepping and/or CPU Assembly Date
- VCORE
- Method of Cooling
- Temperatures (Software-based or Meter-derived)
- Specifications of System
- Stability Tests (SuperPi, Prime95, 3DMark, etc.)

INQUIRIES:
Data Requests - If you have any requests for additional tests or details to be included in a post, please send a PM to the author of the post to include the information you are seeking. Be sure to include the "post #" or a direct link to the post you wish to see edited.
Data Fulfillment - If you (author of a post) receive a PM from a forum member requesting for additional information regarding your test data, kindly consider editing your post to include the requested information.

[Discipl3]

Picked one of these in Akihabara, Tokyo. It was only 36000 yen so I couldnt refuse.

cpu reads:

ADA3700DAA5BN
CABGE0512XPMW
1230198C50032

The results are as follows:

It does 270x10 on stock volts, and 254x11 on 1.55v on air (zalman CNPS 7000b-cu). My memory wont do 280x10 (its corsair 2x512 xms4400c25p, 275 is the max) so im happy with 275x10 for now. My super pi times seem a little off, compared to the 3800+ at the same speed but I think that this test dosent rightly reflect the 1mb cache.

Concidering the price this has got to be the best value chip out at the moment, It easy did 2.6ghz (fx55) but has the added value of sse3 instructions as well.

[cpulloverclock]

same proc

http://angel.ap.teacup.com/maiko/

[cpulloverclock]

fred is testin' a 3700+ 0512XPMW

263.38 x 11=2,897.17 @1.536Vcore with a XP90

[cpulloverclock]

a few screenies

-300mb/s at sandra and everest in comparison with my 3000+ at same frequency and fsb...

that would give 100MHz+ with my new dod...



max 1M at 3215MHz but bad time cuz maxmem didnt help






tearing, yup, weird, sometimes that just do it, problem between the 4000+ and dfi nF4??, same thing on air

[JNav89GT]

Well obtained a 4000+ San Diego

Seems to run a touch warm on stock cooler, but under an XP-120 and strong fan things are good.

CPU-Z doesn't pick up correct frequency nor does Windows, so I have to use Clock Gen to show freq's

See pics below. Was tasked running other testing for days, so this is first day I've had it in an actual OS

Memory is Mushkin Reline PC4000 btw

[JNav89GT]

yeah it's set to 1.45v in bios
here's one at 1.475 I think

btw clock gen won't set anything else, just locks up with this setup
and sorry, but I can't fit much on screen as I'm working on 5th system and using old ibm 17" crt

[JNav89GT]

lol
just running 1mb as a quick and dirty test
here's another to infuriate you
will test 3dmark in a sec

[JNav89GT]

ok 265x11 3dmark per Bachus's demands
6600gt stock clocks

[JNav89GT]

the money shot

[cpulloverclock]

I can improve with my actual cooling

[JonEapples]

ADA3700DAA5BN
CABGE 0515VPMW
1239326D500xx

Having fun breaking it in, prime stable at these settings:
Rig in sig
271x10 @1.40 Vcore
Ram @ 225.8MHz @2.70Vdimm 2-2-2-6 1T(set 1T using A64 tweaker due to Neo2+Sandy bug)



I will update this post with future results

[charlie]

filthy, stinkin' fast

0515SPMW 4000+
CoolerMaster Hyper6 w/lowspeed 80mm fan
DFI Ultra-D "SLI Mod"
MUSHKIN Redline XP4000
OCZ Powerstream 520W
BBATI X800XL

OH, and don't believe CPU-Z that's only default voltage + 4%!!!



I'd like to take a sec here, and talk about this CPU. With all the uproar about the "cold bug", I'd like to make a couple notes: this CPU is really fast... I ran some 3D last night, no Ricky tweak @ 2880, 430/560 on the x800xl and I beat my OLD "high" by 500 points. The old config included MUCH higher clocks on the GPU, and the SD just SMOKED it. The #1 '01 score above mine is with a Winny at 3100mHz++ and I already beat his Lobbies at only 2880. After playing for a day, I believe it's faster than FX55 clock for clock... no proof, just a feeling I get. Next question is how it will do on single phase change, if it'll do "near" cascade speed, maybe 3500mHz, it'll give the 130nM FX's a run for their money!

C

[EQuito]

I got mine from Monarch today (0514SPMW).
I'm sure in the right hands and with better cooling this chip can do better but this is all got so far:

[JNav89GT]

Bacchus is going to kill me
but wanted to tell peepsl that my 4000+ is going strong on modded mach1

liking this dfi, A64 4000+ diego, Redline comob

[maxoverclock]

got my 4000+ SD this morning.......

stable as a rock right now. don't have any memory to run 1:1 so running with a divider. still smokin fast.....my fx-55 could only do 2860 @ 1.7V and topped out at 50c under load. the 4000+ so far @1.65V is topping at 100 faster and the temp is 10c cooler under load @ 40c. i'll pump some more volts into it to see if it can go faster in a couple of days....so far so good. glad i've sold my fx.

max

[JNav89GT]

this shot is for Bacchus
j/k but nice to run SuperPi 32M 1:1 memory super tight timings w/Redline 4000 @3.5V, Diego is about 3-5C lower on evap temps compared to my older 4000+ core running same speeds, but of course Diego taking ALOT less voltage at 1.535 vs 1.83 v

I was able to get 3300mhz stable through super pi 32M, was actually a touch slower than my 32xx run as I was runnning 1:1 memory there, and I had to drop memory divider down to explore limits of cpu for my 3300mhz run. Past 3300mhz I haven't had any 32M luck. Might need to really start pushing vcore which I'm a bit leary about. 3300MHz suspected stable is awesome IMO, on a regular single stage phase change setup.

[EQuito]

I made some progress by lowering the timings...

[computersmsa]

Stable
1.4V -> http://www.bel-hardware.be/SD4000/PC_MS_1.4V.PNG
1.5V -> http://www.bel-hardware.be/SD4000/PC_MS_1.5V.PNG
1.55V -> http://www.bel-hardware.be/SD4000/WCM_MS_1.55V.PNG

Bench
SPI 1M -> http://www.bel-hardware.be/SD4000/SPI1M_3118.PNG
SPI 4M & CPUM99 -> http://www.bel-hardware.be/SD4000/SPI4M_CPUM_3G.PNG

[Nettwerk]

SD 4000+ CABHE 0515SPMW

2950 is the max on this chip.. no 3ghz for me this year I guess :/
Ive been messing with it all day today and I dunno the lowest voltage for it so I set it to 1.65. heres a few shots from messing with it, still got alot of work.

http://home.houston.rr.com/nettwerk/sandeigo/265x11.JPG

[EQuito]

Some BH5 action...

[charlie]

Here's my results.... not bad, but my session simply reminds me how much I need to switch over to phase change
CPU temp -6C to -10C, Base temp as low as -58C, booted just fine.

DFI LP Ultra-D "SLI Mod"
4000+
MUSHKIN Redline XP4000
OCZ Powerstream 520W
BBATI X800XL

[B5I8]

Has anyone ran some real stability tests with their overclock?


Sorry, I didn't read the rules. This is a 4000+ 0515 SPMW from Monarch.

[mcbarnet007]

After I remove the IHS I gain another 100mhz. Here is the result. Cooling is in sig. CABGE 0515VPMW serial end in 472. Voltage is 1.65v. Had to run UTT in dividor, else score would be better.

[Faceless Rebel]

San Diego 3700+
ADA3700DAA5BN
CABGE 0515VPMW
1239326D504xx

2750@1.55W
Prime Stable
3DMark 05/03/2001 Stable


other stuff:
Zalman CNPS-7000Cu (idle: 39C, load: 48C)
MSI K8N Neo4 Platinum
GF 6800GT @400/1100
Antec TrueControl 550 (+12V 24A)