• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

A K7xp2000 Nightmare In Europe... Plesase Help

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

scandisk

Registered
Joined
Jan 27, 2002
Sorry, but our english level it´s a bit low.

Hello from Spain, in Europe. We love this forum, and now we really need help.

We usually works with overcloked K7 processor without problems... after now. We usually close al L1 bridges using an special conductive Silver Paint, with very good results.

For example, we usually close al L1 bridges of a K7 1.800 XP processor, select a correct multiplicator manually on the motherboard (a Gigabyte GA7DXR+), use a bit more of voltage (1,8V), power up the system... and WE HAVE A 1.900 or a 2.000 K7.XP Processor.

We usually use the Gigabyte GA-7DXR+ with a Promise RAID, that works fine and have a nice of features. We upgrade de Bios to the F8 version (the latest version in the gigabyte website).

Today, we buy a GREEN AMD K7 XP 2.000 (1666 Mhz). We buy a GREEN processor because we hear that GREEN processor it´s better than BROWN processor.

We plug in into motherboard, and works fine at 2.000 (1666Mhz). After that, we unplug it and try to do overclocking... We close all L1 bridges, select a correct multiplicator at motherboard (the GA7DXR+ with F8 bios), turn on de system... and... SURPRISE, WE HAVE A K7 1.700 XP ... THAT´S UNDERCLOKING.

After that, we made many experiments... with all the multiplicator positions at motherboard, with automode activated... and WE HAVE NOTHING. WE ALLWAYS WORK AT 1.700

The processor it´s now pluged into the motherboard... and we don´t know the Proccessor Code ( but we remember that´s similar to AGOIA... )

Someone can help us? Where is the problem?....
- Maybe we fail closing the L1 Bridges at processor.....
- Maybe we must to select manually a fixed multiplicator at L3 and L4 bridges at processor.
- Perhaps motherboard don´t support this processor (but hear that yes).
- Perhaps we have a new T-Bred processor.... but we think that new processor it´s 2.100, 2.200 and more. Really exist T-Bred processor at 2.000? We don´t know.

Please, help us... WE WANT OVERCLOKING, no undercloking. If you can help us, please, writte your solutions in this forum.

Thanks to all.
 
Sounds like a sympton of incorrectly connected bridges.

If one of the bridges is touching another, then this is exactly what will occur. For example, the 2000+ is set to 12.5x133 (1.66ghz). Now it's set to 1700+ which is (1.47ghz 11x133).

So one of your bridges is either not entirely connected, or it is touching another bridge.

My only suggestion is that you redo the bridge and try it again, I'm sure it'll work. I had a similar problem with my XP and this was the cause. If you've never had to redo a bridge before, you can use nail polish remover (acetone) to remove the silver paint.

Good Luck! :beer:
 
I've also heard that the green XP's are harder to correctly connect the L1 bridges on; some kind of shiny coating across the top of the proc. This is not from personal experience though as both XP's I have are both brown. I've read here that you have to scrape the coating off of the copper dots for the L1 bridges before you connect them together.
 
muddocktor said:
I've also heard that the green XP's are harder to correctly connect the L1 bridges on; some kind of shiny coating across the top of the proc. This is not from personal experience though as both XP's I have are both brown. I've read here that you have to scrape the coating off of the copper dots for the L1 bridges before you connect them together.

Yea, that's true although I have no experience either. People have sanded their L1's but many have went too far to see copper and there goes a perfectly fine chip.
 
Sounds a bit risky...

I'd still suggest that the first thing you do is clean off the bridges and try one more time, and then try ONE more time if it fails (3rd times a charm ;) )

If all else fails, perhaps its true what they say and you're going to have to sand the connectors down a bit.

Why don't you use a multimeter to test the bridges prior to putting the cpu back into the board?
 
what has been suggested is propably one of the reasons you get 1700, but the 2000+ have the 12.5 multiplier which doesn't get any higher than that, you cannot set anything higher than 12.5, so you will need to do a major bridge mod to get the multiplier to 13 or above, or just overclock with upping the FSB.
Good luck
 
Back