I deleted my previous post because I felt really stupid that I failed to see the obvious: I was closing the wrong L2 bridge.
The L2 bridge that controls the cache is the second from the left but these chips do no have it cut. Instead, they have the first bridge cut which is the one I was closing by mistake.
So what's going on here? I think these chips don't have full cache thus no need to cut the bridge from factory.
looking @ the size of the core in mt's 1st post, i'd say that it does have the cache...but that still doesn't explain why the chip wouldn't give all 512 of the cache.
Looking at this site, it would appear that closing the first L2 would enable the cache, and also explains why the different bridge is cut.. Dunno what's up :[
EQ, just shut up now, OK? You're making me depressed...
What's the stepping on that chip again?
If you're depressed imagine how I feel. 224MHz is the highest FSB my v1.2 board has EVER gone but decided to do it with someone elses chip.
Also, he paid $65 for it plus an ECS board while my $80 2GHz Thorton doesn't even come close... :banghead:
I think I'm gonna disable the extra cache on my chip and give it another try.
I believe Larry's chips can OC so well because the extra cache is disable.
I hear you. My 1700 used to run stable like that (folding 24/7) but like an idiot, I sold it and bought a 2100+ that doesn't even come close.
btw: something weird about Larry's chip: I wanted to go higher but it won't accept any multiplier above 12.5...
0
Geeky1University of the Pacific (Stockton, CA, USA)
edited December 2003
I was happy with my 1800, too... 2.56GHz on air with no FSB mod and generic DDR400 isn't bad... and then you had to go and get 2.7GHz out of that Thorton. Damn you.
0
Geeky1University of the Pacific (Stockton, CA, USA)
edited December 2003
Hey mtgoat... mackanz has that Prometia Mach II... maybe you should send him the chip and find out what it can do...
0
Mt_GoatHead Cheezy KnobPflugerville (north of Austin)Icrontian
I made it to 2.8GHz (224x12.5) but I couldn't take a shot...
If you want me to, I can raise the voltage over 2v as long as temps remain low.
That looks convincing enough for me. But if you don't have much to do you could play with the other one as it is a different batch. Yes, if enableing the extra cache holds back on the clocing then I certainly won't want it now!
Thrax,
Want me to go see if they have any more? But it'll cost ya.
I did play with the other one and it's not as good as it didn't make it to 2.7GHz with the same voltage.
I marked the good one for you to see and I left the L2's closed as it doesn't make any difference anyway.
They're packed and ready to go unless you want me to do more testing, LMK.
btw: they have'em online but obviously you can't check the code...
Comments
I made it to 2.7GHz (216x12.5) but not stable enough to capture it.
Since they are week 22 chips and later week chips have worked do you have any idea why the L2 mod wouldn't work?
2.6GHz on air at less than 2v is just astounding.
The L2 bridge that controls the cache is the second from the left but these chips do no have it cut. Instead, they have the first bridge cut which is the one I was closing by mistake.
So what's going on here? I think these chips don't have full cache thus no need to cut the bridge from factory.
Looking at this site, it would appear that closing the first L2 would enable the cache, and also explains why the different bridge is cut.. Dunno what's up :[
Do you want me to leave the FSB at 166MHz anyway?
Yes, and like you mentioned I can remove it if I change my mind.
I truely appreciate your help and am happy with having good clockers with or without the extra L2.
I'll try to unlock the cache one more time before I give up. I'll send them back on Monday.
Thorton 1.67GHz @ 2.7GHz (216x12.5 / 1.98v / 43c) stable enough to run Sandra:
What kind of cooling is on that thing?
If you want me to, I can raise the voltage over 2v as long as temps remain low.
What's the stepping on that chip again?
Also, he paid $65 for it plus an ECS board while my $80 2GHz Thorton doesn't even come close... :banghead:
He listed the step on the first post:
AXDC2000DUT3C
1052787L30450
AQXDA0322MPMW
I believe Larry's chips can OC so well because the extra cache is disable.
btw: something weird about Larry's chip: I wanted to go higher but it won't accept any multiplier above 12.5...
Thrax,
Want me to go see if they have any more? But it'll cost ya.
I marked the good one for you to see and I left the L2's closed as it doesn't make any difference anyway.
They're packed and ready to go unless you want me to do more testing, LMK.
btw: they have'em online but obviously you can't check the code...
http://shop1.outpost.com/product/15568