Well..heres an interesting development. It seems that a lot of my problems may have been from not so good ram settings.
I had basicly copied ram sets for this memory from a screenshot billb had posted here a long time ago.

I knew he ran this ram and had probably tried the hell out of every possible set. So I just copied them.
The problem was I was trying to run settings he was using for breaking Super PI records.
Well we all know that benchmark settings are rarely good for everday ops, but I hadnt really given it much thought.
Turns out this is the case here as well. He popped in over the weekend and clued me in on some better everyday sets.(thanks bill)

Things are progressing much more nicely now Getting stable at 3000 using the 183 divider.
This wasnt possible with the previous sets....using the 183 divider it would just freak out.
Also ran my fas-*test*-('") SuperPi 1M run ever at these sets without even trying at 28.047s.
With a little effort I can see high to mid 27's will not be out of the question.
Gonna try a few passes of 3dmark 06 next...thats where my instability was the worst.
That was what had me running 1.6v....and here it might have been the ram freaking out all along.
Currently at 1.55v....going to try and start dropping volts next.

Good post Merc and also good news for Wanna !

I should join you in no time

Yea lookin forward to seeing your results in this thread as well Crisao23.
I do hope you have as good of luck and are as happy with yours as we are.

As it turns out I seem to have an extra CPU same stepping that is probably just as good.
I bought a second one that I am running currently and labeled the first a dud.
Now come to find out it was probably my bad ram settings holding it back as well.


Now to decide wether to sell it or keep it

Thanks Cris. Just my own theory and maybe when you get up to the 3.0 we can do a -*test*-('") run of air and water with ambient figured in and see if I am full of it or pretty close.

Yes Wanna is now safely at 3.0. Probably can run at lower voltages as well. I am thinking of setting a 8x multi and ramping up the FSB to see if faster RAM speeds would be better than lower latencies.

Do it Merc......I'm dying to see if these boards/CPU's can handle 376MHz CPU speeds.
With all your watercooling goodness on that mobo, if it can be done you can do it!


Currently running Orthos at 1.47.....havent tried 3D06 yet but its looking good for it.
1.5 passed with flying colors. About time to give it up and hit the bed here soon though.....

I'll give it a shot right now and post back.

Didn't like that at all. Got the old Overclock Failed page. Guess 376MHz is too high of a FSB. Overvolted everything and relaxed the RAM but no go.

Well thats no real surprise....thats asking an awful lot of it.
You did try lowering HTT to 2x right?
Although we know its capable of 1168HTT no problem.
So that prolly woudnt matter anyway.

1.475 passed Orthos and 3DMark 06, and is my current Volts for today.
I'll mess with that more tommorow but I dont expect to be able to go much lower.

Yep tried 2X on the HTT but no go. Remember she ran at 1500MHz on the HT that one time I forgot to reset to 3X. How much is that DDR 600 RAM????

Its kinda pricey.......but damn does it do what it claims

$170 for a 1GB DDR 600 kit.

$275 for a 2GB DDR 500 kit.

They dont sell a 2GB DDR 600 kit yet.......

http://www.computergiants.com/items/mfgr_d...id=539&catid=52

I just looked. Damn, I'd need a 2GB kit. It really wouldn't be a bad purchase as I'd like to move my present RAM to my Wife's soon to be installed A8N SLI Deluxe rig.

Well here's some feedback on the TXDR2048M500HC25DC Micron series 8-3-3-3
http://www.teamgroup.com.tw/forums/search.php?searchid=175

Couldnt find any info on Team Group site on the TXDR2048M500HC25DC 8-4-4-2.5
If it exists it's not on their site anywhere....but would be very similiar to what I have.
Same timings and voltages as you would set for my 1GB kit to do 600MHz but 2Ghz@500Mhz

After a little searching the net apparently they do exist and are reviewed here,
http://www.madshrimps.be/?action=getarticle&articID=397
Although the entire article left me a bit confused. To many similiar long assed model names
JNav89GT appears to have -*test*-('")ed them, but the latency's he listed were different then what TG's has.

On a side note JNav89GT is old school Xtreme Forums and his opinion's have my respect.
As a result of that I would believe his specs on the ram before I believed TG's.

It seems the Microns would be the best bet for hitting 273MHz.

dinnertime.....

I guess that it's not a lucky day for me

Opteron fails orthos all the time. Doesn't even run 300x9.

Trying, trying, trying

That stepping should definitely do 2.9. Check that your HT is at 3x, RAM is set to 133, CPU overvolt is enabled and CPU Voltage set accordingly.

Try 1.275v (overvolt enabled) with a 310 FSB 9x multi RAM voltage accordingly and divider at 133, HT3x

Also, run mem-*test*-('") -*test*-('") #5 just to make sure you don't have a bad stick.

No bad sticks friend

Only good old value "miracle" select.

I'm running Orthos for some minutes and posting at 2700 Mhz/300x9 1.37v. Core 1 10°C higher than Core 2.

21+ minutes now ok.

Divider at 166, ram at 245.5.

I did a complete cleanup on mobo and changed chipset cooling.

Merc, which voltage you used for stable 9x334 ? Also have you ever used divider at 150 or 180 on one of the -*test*-('")s/configs ?

24 minutes and running