Jump to content

The AMD 3GHz Club


Recommended Posts

Man that suck's I was hoping to see that 3200 up there, But I don't know if i would sell it if i was u. How often do any of us see a chip that will go that high, I think you should keep it around just because of how much work you put into it. Monarch has the 165's OEM for 219.99 right now and I keep trying to talk myself out of buying one.

Share this post


Link to post
Share on other sites

  • Replies 5k
  • Created
  • Last Reply

Top Posters In This Topic

Going to do a little checking. Maybe the block isn't perfectly seated.. I still find it odd that it won't do any better than it did with water. I know it's not bugged at the temps I'm at because I can duplicate my water OC just not get any higher. Might just be the limit of the memory controller, small fft's will run but blend will always drop out on large fft's. It's done it with 3 different sets of memory and every multi/divider combination I can try.

Share this post


Link to post
Share on other sites

See if you see anything. I have been through this thing back and forth for a week. 3 pages of notes on just this jump from 3169 to 3.2. 95% of the settings can be left on auto and the results are same. Where it is now and settings tried in parenthesis

 

FSB Bus Frequency - 291

LDT/FSB Frequency Ration - 3( 2.5 and 4)

CPU/FSB Frequency Ratio - 11X

PCI eXpress Frequency - 101Mhz

 

CPU VID StartUp Value - StartUp

 

CPU VID Control - 1.40v

CPU VID Special Control - Avove VID * 112%

LDT Voltage Control - 1.20v

Chip Set Voltage Control - 1.52v

DRAM Voltage Control - 2.56v

 

DRAM Configuration Settings:

 

DRAM Frequency Set - 180(DRAM/FSB:9/10)

Command Per Clock (CPC) - Enable

CAS Latency Control (Tcl) - 3.0

RAS# to CAS# delay (Trcd) - 04 Bus Clocks

Min RAS# active time (Tras) - 08 Bus Clocks (10)

Row precharge time (Trp) - 04 Bus Clocks

Row Cycle time (Trc) - 08 Bus Clocks ( 10 thru 14 as well)

Row refresh cyc time (Trfc) - 16 Bus Clocks ( 18 thru 24)

Row to Row delay (Trrd) - 03 Bus Clocks (00, 02)

Write recovery time (Twr) - 02 Bus Clocks (03)

Write to Read delay (Twtr) - 02 Bus Clocks

Read to Write delay (Trwt) - 03 Bus Clocks (02, 04)

Refresh Period (Tref) - 3120 Cycles (3684, 4196, 3072, 4078)

 

DRAM Bank Interleave - Enabled

 

 

 

DQS Skew Control - Auto

DQS Skew Value - 0

DRAM Drive Strength - Level 8 and everyone up to it.

DRAM Data Drive Strength - Level 2 (3, 4)

Max Async Latency - 8 (9)

Read Preamble Time - 5.0 (5.5, 6)

IdleCycle Limit - 256 Cycles

Dynamic Counter - Disable

R/W Queue Bypass - 8 x

Bypass Max - 04 x

32 Byte Granularity - Disable(4 Bursts)

Share this post


Link to post
Share on other sites

I think my chip is hitting the limit on the memory controller too, With this uccc i had to set mal to 11ns and read preamble to 9ns to get it stable at 3170. Some of the divider's are a real pain in the a$$ to get working right on my chip, and at least on air 3200 is a no go for me right now. Don't forget I am just keeping that crown warm for ya I want to see u get back up on top. Then I have something to push for besides pi's which are really starting to piss me off lol.

Share this post


Link to post
Share on other sites

Why don't you give it more Vcore? 1.56-1.57 isn't that much really. Try 1.59

 

Also, you -are- starting to hit that range where the high FSB is requiring a bit of chipset teasing. Try 1.3 LTD voltage and 1.6v chipset. Upping those voltages stabilized my chip to get that 3100GHz Pi score.

 

I say you also use those loosened MAL/RP settings.

Share this post


Link to post
Share on other sites

U might try upping the ldt and chipset like Thraxz said it has helped me when using high fsb also. My super pi run at 3270 took the extra voltage on those just to boot and get that 7/10 divider working. (even though i got a worse time than at 3200 with 5/6 divider lol) Worth a shot anyhow.

Share this post


Link to post
Share on other sites

I missed a few of the things I had done. Vcore has been as high as 1.65, LDT 1.4, and up to 1.6 on chipset. I tried quite a few combo's of MA/RP.

 

What's killing me is at 11 X 288 I can pretty much change anything and not affect stability, but bump it to 289 and the problems begin. usually I can do a little tuning and make some progress but this just stops and won't give you a hint either way..

 

Edit.. I do have a CFX3200 arriving tomorrow and another expert on the way so I will give them a shot with the CPU as well.

Share this post


Link to post
Share on other sites

Money goes quick when computer's are involved but when u see a good deal u have to grab it or end up regretting it later. That's how I ended up with this opty, Found a great deal on them and my buddy was still using a 3200 winchester so I was gonna order him one and then thought about how good a deal it was and grabbed me one too. Damn am I ever glad I did 2900 was the max on my old opty, not that it was to bad or anything.:)

Share this post


Link to post
Share on other sites

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now

×
×
  • Create New...