Unofficial Opteron 939 Thread

[custom_coms]

Umm..yeah, I have no clue what you are talking about Praz! I followed the rules and the other posts to a tea....

 

Only thing I don't have is Aquamark3d and Everest, which are no longer required per here: http://www.dfi-street.com/forum/showthread.php?t=10192

 

If I missed something let me know, I'm trying to help fellow street members!

[General Septem]

Well for some reason I couldn't change the divider, it wouldn't boot. It would just get to the part where it would normally say "verifying pool data..." but it doesn't say this, it just reboots. It won't boot on any divider other than 1/2. That's not a big problem though as the FSB is so high it already runs at DDR320.

 

So I guess I'm dual priming now. Good night

[Praz]

Did you notice all the other posts look like this?

 

[b]Genie BIOS Settings:[/b]

[color=red]

FSB Bus Frequency.............................  -  290

LDT/FSB Frequency Ratio.......................  -  x3

CPU/FSB Frequency Ratio.......................  -  x9

PCI eXpress Frequency.........................  -  100Mhz



CPU VID StartUp Value.........................  -  1.350v



CPU VID Control...............................  -  1.30v

CPU VID Special Control.......................  -  123%

LDT Voltage Control...........................  -  1.30v 

Chip Set Voltage Control......................  -  1.60v

DRAM Voltage Control..........................  -  2.70v[/color]



[b]DRAM Configuration Settings:[/b]

[color=red]

DRAM Frequency Set............................  -  166=RAM/FSB:05/06

Command Per Clock (CPC).......................  -  Enable 

CAS Latency Control (Tcl).....................  -  2

RAS# to CAS# delay (Trcd).....................  -  3

Min RAS# active time (Tras)...................  -  8

Row precharge time (Trp)......................  -  3

Row Cycle time (Trc)..........................  -  7 Bus Clocks

Row refresh cyc time (Trfc)...................  -  Auto

Row to Row delay (Trrd).......................  -  03 Bus Clocks

Write recovery time (Twr).....................  -  03 Bus Clocks

Write to Read delay (Twtr)....................  -  02 Bus Clocks

Read to Write delay (Trwt)....................  -  03 Bus Clocks

Refresh Period (Tref).........................  -  3120 Cycles

DRAM Bank Interleave..........................  -  Enabled



DQS Skew Control..............................  -  Auto

DQS Skew Value................................  -  0

DRAM Drive Strength...........................  -  Auto

DRAM Data Drive Strength......................  -  Auto

Max Async Latency.............................  -  Auto

DRAM Response Time............................  -  Fast

Read Preamble Time............................  -  Auto

IdleCycle Limit...............................  -  256 Cycles

Dynamic Counter...............................  -  Disable

R/W Queue Bypass..............................  -  16 x

Bypass Max....................................  -  07 x

32 Byte Granularity...........................  -  Disable(4 Bursts)

[/color]

 

Not like this?

 

Genie BIOS Settings:

FSB Bus Frequency - 300

LDT/FSB Frequency Ratio - x3

CPU/FSB Frequency Ratio - x9

PCI eXpress Frequency - 101Mhz

 

CPU VID StartUp Value - StartUp

 

CPU VID Control - Auto

CPU VID Special Control - Auto

LDT Voltage Control - 1.40v

Chip Set Voltage Control - 1.60v

DRAM Voltage Control - 2.90v

 

DRAM Configuration Settings:

DRAM Frequency Set - 150=RAM/FSB:03/04

[burntcircuit]

Kinda weird with the divider. Sure you are not selecting the wrong divider or something. Good Night + Luck:)

[General Septem]

Kinda weird with the divider. Sure you are not selecting the wrong divider or something. Good Night + Luck:)

 

I tried the second to lowest divider and even that wouldn't work. I tried fooling with some other settings but no matter what I did, the BIOS wouldn't boot the system unless the divider was set to 1/2. But with the CPU at 2.88 GHz and the FSB at 320 I'm hardly noticing any RAM defficiency. When I get watercooling I'll be able to raise the FSB even further and get the RAM just a little faster.

[custom_coms]

Fixed. The original OCDB posts look like how I did mine originally b/c I was using an older template. The new one looks better, so i understand!

 

Septem, thats a weird issue. I would look into going to a different bios possibly. I haven't really played with the dividers, since my memory is capable of the speed with a 150 (3/4) divider and a 300 fsb speed. Might become more critical as I push higher. I have the same cpu stepping as you btw, so I'll keep you informed when I go for higher so we can compare!

 

Night all.

[General Septem]

I got my Opty to 2.8 GHz on stock cooling. I'll probably be able to go higher if I had watercooling, but for the time being I can't get it stable at anything above 9*312=2.8 GHz. I'm still very happy though, of course.

 

I'm exceptionally pleased with the stock AMD Opteron cooler. I'm idling low-30s and just about 50 at full load. Now it could be better but considering this is the stock cooler and I'm 0.2V and 1GHz above stock, it's doing great. That 90mm fan thing really helped out a lot.

[jhyunl]

Reporting another opteron 144 overclocking.

 

It is settling down to 2.75ghz on air with Zalman 7700 (120mm) cooler under 1.52vcore (now running on prime95 for about 2+ hours stable). I hoped it to be better than this, but 2.7+ will surely do its job. The temperature goes never higher than 48C under full hours load of prime95. When tested under superPi 32M, it does 2.9ghz stable under 1.52vcore, but always fails at prime95. I don't know why, and I just guess prime95 tortures the core much extensively.

 

I tempered around a bit from 2.8-3.1ghz, but was never able to make it stable. The fact that it sporadically boots up and even gets into Windows under 3.0ghz and above, I think this 144 CACJE still has more room to be tweaked, but I am too inexperienced to get more out of it. (And, I have the king of Mr. instability memory, the Crucial Ballistix PC4000 2G kit, which is notorious for its bizaire and erratic performance).

 

It was ordered around 11/12/06 from newegg and shipped on the same day, and price was like 75 bucks. And, it is within the same price range as venice 3200+, and I guess many people actually get stable 2.75Ghz from venice 3200+.

 

Well, so I don't seem to gain nor lose anything by choosing opteron, but anyway am glad it works good now.:shake:

[burntcircuit]

Reporting another opteron 144 overclocking.

 

It is settling down to 2.75ghz on air with Zalman 7700 (120mm) cooler under 1.52vcore (now running on prime95 for about 2+ hours stable). I hoped it to be better than this, but 2.7+ will surely do its job. The temperature goes never higher than 48C under full hours load of prime95. When tested under superPi 32M, it does 2.9ghz stable under 1.52vcore, but always fails at prime95. I don't know why, and I just guess prime95 tortures the core much extensively.

 

I tempered around a bit from 2.8-3.1ghz, but was never able to make it stable. The fact that it sporadically boots up and even gets into Windows under 3.0ghz and above, I think this 144 CACJE still has more room to be tweaked, but I am too inexperienced to get more out of it. (And, I have the king of Mr. instability memory, the Crucial Ballistix PC4000 2G kit, which is notorious for its bizaire and erratic performance).

 

It was ordered around 11/12/06 from newegg and shipped on the same day, and price was like 75 bucks. And, it is within the same price range as venice 3200+, and I guess many people actually get stable 2.75Ghz from venice 3200+.

 

Well, so I don't seem to gain nor lose anything by choosing opteron, but anyway am glad it works good now.:shake:

 

Hey thats still a pretty decent OC on air. Too bad you have the 144 as the 146's were consistently doing 2.9Ghz. Seems like we are both fighting with RAM these days. My rig is stable and 10X257 1:1 all day long but as soon as I put it on a divider and start cranking up the FSB I'll be lucky if I can get 240Mhz out of my sticks. Can't really take a chance on ruining these sticks either as their is a shortage of DDR memory, and the EB Platinums i have are not any better, worse actually.

 

Still you have a nice OC there, enjoy it...

[bchivers]

I just got one of the opty's from Newegg and have a question about the the ram, I have read the stickys but am confused. If I understand it right to get a 1:1 ratio with PC3200 I set the dram frequency to 200 but what about PC4000 how do I get a 1:1 ratio with it.

[jhyunl]

I just got one of the opty's from Newegg and have a question about the the ram, I have read the stickys but am confused. If I understand it right to get a 1:1 ratio with PC3200 I set the dram frequency to 200 but what about PC4000 how do I get a 1:1 ratio with it.

 

Somebody correct me if I'm wrong. As far as I understood, putting 1:1 in bios option for my PC4000 automatically overclocks its paired processor to the corresponding clock speed by 1:1 (200) ratio. Indeed, I see my opteron gets around 2.2ghz speed when I put 200 for the 1:1 ratio. To achieve the stock speed of my opteron, I just downclock the memory to 166 and the processor speed pops up as 1.8G. Then, I simply run my memory as PC3200 by tightening the memory timing like 2-2-2-6 with lower voltage like 2.5v, although no good for a certain type of memory like mine that doesn't like operating speed lower than the spec, DDR500.

 

So, simply the beginning point of overclocking is higher when PC4000 is used than when PC3200 is used. Is there any advantage? The only advantage I can tell is at least I am guaranteed for 250FSB where most of value memories are not able to reach nor stable. How helpful is this for processor overclocking? Practically none as my naive experience tells.

 

Unless I am doing some fancy benchmarking for overall system performance, none of my previous systems appear accomplishing "**practically** better" performance by memory overclocking. If the ram divider is set properly for system stability, the CPU overclocking appears to be the evident way to boost up visible performance (personally, I don't give a damn about 1:1 ratio at all). Losing stability and being limited to lower CPU speed appear much worse than losing a bit of performance gain by sticking to its designated speed of a given memory.

 

Again, correct me if I am wrong !! I am a newbie.

[jhyunl]

The 144 cacje [ke-ci-ji:D or keg-ji :eek2: or sa-si-ji: :nod: or whatever lol] passed 12hrs+ of prime95 blend test. The temperature stayed around below 48c all the time, and when I woke up this morning it stayed around 44-46 (it was cold this morning here). The computer is in my bedroom unfortunately:sweat: , so the the temperature could have been lower if it was placed in the living room.

 

Now, its time for OCCT and 3d gaming test !! (My last build completely failed at 3d gaming even after being stable at every torture test that overclockers exploit.:eek2:)