winchester overclock database

awesome viper.
and lithium, it's not the memory (controller?) being pushed too far?
 
i dunno, ill go ahead and drop my multi to 9 and up the bus speed - report back in a few
 
(cf)Eclipse said:
awesome viper.
and lithium, it's not the memory (controller?) being pushed too far?
Click to expand...

Very awesome. Ill get my name on the chart three times... and people will think im rich. Its to bad I didnt know about it when I had my 1st two winnys. One was a 320+ that did 2.6 and one was a 3000+ that also did 2.6, but I never got the stepping. This 3200+ I will try out monday better be a winner. If its a CBBID im gunne be pissed. I've had 5 winnys, im bound to get atleast one 2.8ghz capable chip.
 
eclipse: right now im running a 1m stable 320x7, so i dont think its the memory controller holding me back :p

edit: prime hasnt bailed yet either :D looks like i got a kickass board!
 
I am trying to OC my LBBID winnie 3500+ on an MSI K8N Neo4 Plat w/ Patriot XBL memory (as Samsung TCCD). I get to 205 but anything beyond that makes the system unstable (3Dmark bombs out, but at different times... sometimes it almost makes it all the way through while at 210).. I tried boosting my vcore up one notch (3.3%) and my mem voltage to 1.65 but still no dice... what about the HT multiplier? should I immediately take it down from 5x when I go over 200 on the fsb? What about the nf4 voltage?

oh yeah - it's an LBBID...
 
yeah, though it shouldn't be crapping out already, you'll want to drop the ht link to 4x.
basically, just keep the final ht link speed under 1ghz.
so when you hit 250mhz, go down to 3x.. then 333mhz, use 2.5x, ect

also, did you try raising the timings (though it shouldn't be the ram yet either..)
 
ok - thanks... keep it at or under 1ghz. I did not loosen the memory timings... but I did up the mem voltage to 2.65, and the vcore to 1.45... but I think I will try loosening the memory to cas2.0 3-3-6 and see how it does... All I am looking for here is a stable 2.4 or so... nothing to crazy... I like stability. :)
 
Thats not a winchester. I *think* LBBIDs are newcastles. Someone please correct me if im wrong.
 
i'm not sure actaully. i think i might have seen an LBBID before..

revenant, can we have the full first and second lines of the opn?
 
I remember something about an LBBID winchester chip on anandtech forums. Hope it overclocks well for you!
 
how about the "agressive timings" setting on the Neo4 bios? seems to make things really unstable... wow... I relaxed my memory timings and ran 3Dmark05 without it crashing at 210 on the fsb - now testing 215 (fingers X'd)

my settings are
agressive timing = DISABLE
DRAM = 2-3-3-6
HT = 4x
vcore= 1.44 (3.3% over norm)
vmemory= 1.65
vnf4= 1.55

testing now.. :)

EDIT: w00t! 19027 on 3Dmark03. :cool: relaxing those mem timings was key... I think I can continue on my current timings and voltages on up to 220... what do you guys think?
 
(cf)Eclipse said:
well i still need the week to put you in the database ;)
Click to expand...

heh... it's all good... here are the vitals from my cpu -

ADA3500DIK4BI
LBBID 0507MPMW
 
(cf)Eclipse said:
i'm not sure actaully. i think i might have seen an LBBID before..

revenant, can we have the full first and second lines of the opn?
Click to expand...

uhm, sure... what do you mean? my full settings for dram and cpu in the bios menu? or the Z-CPU info?

EDIT - N/M .. it's in the post above... LOL. I haven't slept much all week... been on call and getting so many calls at night... I hate sitelite!! arg! heh
 
Here is my configuration:

CPU AMD Athlon64 3000+ Winchester 939 1800Mhz --> OC'd up to 2600 a little bit Unstable and perfekt working 2500Mhz
MB MSI K8N Neo4 Platinum with the FSB set to 278 and the HTT 4X
Memory speed set to 133 before OC, 196 after (2x512MB Apacer CL2.5) the memory are not yet OC'd (coming soon :p )
 
thanks reverent, i'll get you up when i have a minute to update.
Br3Ak3R x64, same thing applies to you ;)
 
well, this is the most bizarre thing. I took out that 3200+ today, eager to see its stepping. Well, what do I find? THe heatspreader is totally blank. It appears as tho emachines had lapped it. I will still try my hand and OCing it, but that this point its useless to your chart, since there is no indication of what the hell it even is.

I bet they wanted to hide that its a CBBID :D
 
3000+ Winchester 1.8ghz - CBBID - 0503 - EPHW - 2478mhz - 678mhz gain - 1.68v - stock air - WarriorProphet
 
update:

running @ 2410 now... smooth and silky so far...

dram= 2-3-3-6 @ 2.7 volts (had to up this from 2.65 for stability @ 220mhz)
cpu= 220x11 (HT = 4x) @ 1.44 volts
nf4= 1.55 volts

cpu temp = 34c idle 38c full load (system tmp = 32c)... :) Cooler Master Hyper 6 is awesome. big, but bad-a$$ for cooling on air.
 
I don't have the CPU in my new motherboard yet but here is the info so that I don't forget it after installation.

3500+
ADA3500DIK4BI
CBBID 0451APAW
1169135L40171

I am putting this in a MSI K8N Neo2 Platinum with a Zalman 7000B-AlCu LED.
 
Viper87227 said:
well, this is the most bizarre thing. I took out that 3200+ today, eager to see its stepping. Well, what do I find? THe heatspreader is totally blank. It appears as tho emachines had lapped it. I will still try my hand and OCing it, but that this point its useless to your chart, since there is no indication of what the hell it even is.

I bet they wanted to hide that its a CBBID :D
Click to expand...
wow! that's something new. is there any copper around the edge of the heatspreader? or is it roughly lapped just to remove the label?

i'm working on getting it updated. hopefully i won't get distracted within the next 5 min or so :D

edit: reverent! 1.44v?! :D
 
Thought I'd add this:

My overclock was very mediocre on origional nf4 drivers, but the march 17th revision 6.53 brought me a LONG way. Some people may want to revisit their results.
 
dmonkey said:
Did increasing NF4 Chipset voltage help much with stability?
Click to expand...

Angry Games at DFI (and he works for them) says you shouldn't need to touch your chipset voltage
 
dmonkey said:
Did increasing NF4 Chipset voltage help much with stability?
Click to expand...


No, with A64s, the chipset plays much less of a roll in OCing. In AXPs, the memory controller was on the chipset, so increasing voltage as well as better cooling helping overclocking... but now that the memory controller is on the CPU, so the chipset doesnt play as big a roll in OCing.
 
is taht second voltage option int eh bios (i cant remember what its called) raise voltage going to the mem controller? i might try for 1T if thats teh case...
 
Viper87227 said:
No, with A64s, the chipset plays much less of a roll in OCing. In AXPs, the memory controller was on the chipset, so increasing voltage as well as better cooling helping overclocking... but now that the memory controller is on the CPU, so the chipset doesnt play as big a roll in OCing.
Click to expand...

Ahh - this is good to know... I had set mine up .05v but put it back down now to stock.@ 1.50v... :)
 
quick questions;

how does the 3500 winni oc compare to the 3500 clawhammer oc? on water? how do i know if mine is a clawhammer or newcastle? cpuz says newcastle, i thought it was a clawhammer
 
pigz said:
quick questions;

how does the 3500 winni oc compare to the 3500 clawhammer oc? on water? how do i know if mine is a clawhammer or newcastle? cpuz says newcastle, i thought it was a clawhammer
Click to expand...


If cpuz says its a newcastle, its a newcastle. as far as OCing goes, I believe a newcastle will usually OC better
 
lithium726 said:
is taht second voltage option int eh bios (i cant remember what its called) raise voltage going to the mem controller? i might try for 1T if thats teh case...
Click to expand...


No. There is no option to raise your memory controllers voltage. The voltage options you get are cpu, memory, LDT (HT), and chipset.
 
Viper87227 said:
If cpuz says its a newcastle, its a newcastle. as far as OCing goes, I believe a newcastle will usually OC better
Click to expand...
wrong. CPU-Z reports claws with half the cache as newcastles. they are rumored to be FX rejects, and clock quote high

No. There is no option to raise your memory controllers voltage. The voltage options you get are cpu, memory, LDT (HT), and chipset.
Click to expand...
whats the "CPU VID"? it goes to 1.55v
 
for clarification, since the memory controller is on die...

Core voltage = Memory Controller voltage ;)


pigz: what's the first line of your OPN? if it ends with "AS" you have a clawhammer, "AX" is the s939 newcastle.
 
whahah

i thought i had a claw hammer, turns out is a newcastle
last letters are AW (newcastle)

now im stuck with a newcastle and artic silver on my fingers.

anyway how do the newcastle oc compare to the winni's. i can get 2.5ghz that i know, possibly 250x11=2.75ghz??? on water?
 
oops, i got ax and aw mixed up. my bad.
anyhow.. i don't really know. 130mn chips love cold and voltage. just don't give it too much of the latter without the former ;)
 
I don't have the CPU in my new motherboard yet but here is the info so that I don't forget it after installation.

3500+
ADA3500DIK4BI
CBBID 0451APAW
1169135L40171

2530Mhz 230x11
Zalman 7000B-AlCu LED heatsink/fan
Gain of 330Mhz
Voltage 1.55
 
You must log in or register to reply here.
Back
Top