CHECK OUT THE PIC OF A "LOCKED" DURON/T-BIRD!!!!!

[HardwareAddicted]

http://www.awaremag.com/tweaking/duronpins_1.html

This only a modified pic, but it will show you what to look for.

Notice the pins missing on the edge of the cpu.

 

[johncar]

Sorry HardwareAddicted, but guy at awaremag.com edited out the WRONG pins.

There are "two" Multiplier functions...FID[3:0} Multiplier "ID" which "identifies" multiplier info to the Northbridge for the system to initialize and set timings...and BP_FID[3:0] FSB Multiplier which actually sets the FSB Multiplier inside the cpu.

Both functions are connected to 4 socket pins each...FID to send multiplier info to the Northbridge...and BP_FID to allow connection to mobo dipsw's/jumpers to over-ride the default multiplier settings for oc'g.

You can't remove the FID pins, the system "needs the info". But AMD is "rumored " to be considering removing the BP_FID pins which would prevent mobo hardware from oc'g/changing default multiplier...just like the "open" L1 bridges tried to do.

awaremag.com was obviously "not aware" of all this, as they used the FID Multiplier 'ID" pins from pics, clearly labeled as FID, at Anand article and Dr Tom article. Check aware's links to those pics to verify.

BP_FID pins, in series with the L1 bridges, are along edge on left side of aware's pic, and you can find pics of labeled BP_FID pins at Anand and Dr Tom Duron OC articles. Even Anand was not initially aware of BP_FID pins because AMD did not document them...they hid them as undocumented pins...Marked NC for "no connection" on pinout diagram in Duron datasheet. We rang out L1 circuits and found them...first.

We also posted an "Alert Wrong Pins" thread here a few days ago.

http://members.xoom.com/candjac/index.htm for full oc'g details...all circuits.
John C.

 

[HardwareAddicted]

Thanks John, your a real asset to this forum.

I will check out your info so I can be sure to know what I'm talkin' about...

I wonder if I can delete this thread...never tried b-4.

 

[johncar]

HardwareAddicted et al,
Contacted aware.com and he has researched issue, agreed, and revised pic to show how chips would look with the BP_FID pins removed. Looks OK now, but, still only a rumor that AMD will remove them. And thanks for kind words, we try to be helpful to oc'g community, and it's fun playing point/counterpoint with the chip mfr's.;-)
John C.

 

[Warrenton]

Are the BP_FID directly connected to the bridges on the top of the CPU? I mean do they connect to them. If so would it not be possible to attach wires to the bridges and then run them to the bottom of the motherboard and solder them to the bottom of the socket?

 

[johncar]

Warrenton wrote.....
Are the BP_FID directly connected to the bridges on the top of the CPU? I mean do they connect to them. If so would it not be possible to attach wires to the bridges and then run them to the bottom of the motherboard and solder them to the bottom of the socket? end quote.

Yes to all ????, and czech site has done so, tested.

For "Alternative BP_FID Settings"..."paint job", see http://members.xoom.com/candjac/index.htm page 2....see all 3 pages
for all circuits, details.
John C.

 

[johncar]

Warrenton,
Forgot to mention....
Soldering wires best/easiest done to 1K resistors rather than the L3, L4, bridges. Circuit diagrams at http://members.xoom.com/candjac/index.htm explain that resistors connected to closed bridges are connected to signal busses.
John C.

 

[DaddyG]

John, the bridges are brass i believe, if so solder won't do the trick.

 

[johncar]

DaddyG,
You're absolutely correct, we misposted when we said yes to all. Should have said solder wires to solder tabs on 1K resistor packs. Be sure to solder on to the 4 resistors connected to the 4 closed L3, L4 bridges.

This has been done by tweak site in Holland, tested...OK

Thanks for picking this up DaddyG.
John C.

 

[DaddyG]

John,

As the recognized expert on cutting and hacking BIRDS/DURONS wouldn't you say that cutting the FID bridges with your modified diamond tip burr is better than soldering to the resistor packs ??

Update: Insanehardware has a proc without the BP-FID pins AND the L1 bridges are blown. This seems redundant.

 

[johncar]

DaddyG wrote.....
...re cutting and hacking BIRDS/DURONS wouldn't you say that cutting the FID bridges with your modified diamond tip burr is better than soldering to the resistor packs ??

DaddyG,
First off, you meant BP_FID bridges, as the FID pins connect to Northbridge and won't be removed.
"Better" depends on the operator's skills and attitude towards the oc'g operation. Yes, cutting bridges with a diamond tip tool is easy...but the operator has to remove the hs and chip, and perhaps close some bridges formerly open, for every change in settings during the oc tweaking/testing.

And you're also right that soldering wires to resistors' solder pads requires much more skill...but once the wires are connected to appropriate solder pads on the socket, (via a 4 pin connector for dis-assembly), the operator can use the dipsw's on the mobo, like the A7V, for example, and never have to remove the hs. But right, would not advise beginning solderers to do this mod.

So there are plusses and minusses either way...Now that the users know, we'd let them choose. We haven't decided which way to go yet...it's that close in our minds. But good we explored the issue together.
John C.