 |
|
12-24-2002, 03:56 PM
|
#1
|
|
Xtreme Addict
Join Date: Jun 2002
Location: LAS VEGAS
Posts: 1,015
|
Modifying A 133 Fsb T-bred B Into 166 Fsb
According to new information that I have just read you can close the L12 bridge on a 133FSB processor and get a 166 FSB.
Quote:
Flickerdown
We mentioned earlier that there differences between Barton and TBred-B in terms of their L12 bridges. The difference can also be seen with the Athlon XP 2700+ and 2800+, which share a 166MHz FSB. Would it be possible to go straight to a 166MHz FSB by changing the L12 bridges? We decided to test this with a TBred-B Athlon XP 2200+:
To reconnect the two severed L12 bridges, we will use the same method used to unlock the multiplier. Apply correction fluid to the gap in the second bridge from the right (the bridge we will attempt to connect), then tape the adjacent bridges to prevent short-circuiting. Then connect the bridges using a conductive silver pen.
After connecting the bridges, boot up the computer to test. One major worry is that the PC may not boot if it has a high multiplier setting. If at all possible, unlock the Athlon XP’s multipliers as well to prevent booting problems.
In this case the PC booted and the BIOS quite clearly shows a 166MHz FSB. A simple L12 connection to unlock 166MHz FSB speeds seems to be well worth the effort – if you choose to unlock the multiplier, unlock this higher FSB setting as well to give a significant performance boost.
|
|
|
|
12-24-2002, 04:00 PM
|
#2
|
|
Xtreme Addict
Join Date: Jun 2002
Location: LAS VEGAS
Posts: 1,015
|
I haven't actually tested this yet, but according to the article which originated from OC.COM it worked for them. I don't have the time to test this right now, but they used a 2200+ with no problems. Would someone try it on their T-BRED B and see if it does actually work.
|
|
|
12-24-2002, 07:31 PM
|
#3
|
|
Xtreme Member
Join Date: Jun 2002
Location: Ouetr Banks of NC
Posts: 243
|
Quote:
Originally posted by Nohto
I haven't actually tested this yet, but according to the article which originated from OC.COM it worked for them. I don't have the time to test this right now, but they used a 2200+ with no problems. Would someone try it on their T-BRED B and see if it does actually work.
|
The "function" you are relating to is AMD's Front Side Bus Autodetect function, described in datasheet 25175.pdf. Page 7 Logic Symbol Diagram shows 2 outputs, FSB_SENSE[1:0], and page 74 discusses it in more detail, describing the 2 bit code. The pinout diagram/s show FSB1 at pin AH30 and FSB0 at pin AG31, which we believe are the mobo interfacing pins for FSB_SENSE[1:0].
We have written this up with pics and diagrams at
http://www.beachlink.com/candjac/index.htm link to FSB_SENSE article. We also have an FSB_SENSE thread tight here at this forum, asking for someone to verify that the 2 L12 bridges in question, (1st and 3rd from the left), do indeed connect to pins AG31 and AH30. Maybe you will be the one.;-)
You will find that your attached pic of their
(?oc.com.tw?), mod of a 13.5X Tbred 2200 from default 133MHZ to 166MHz agrees with the 2 bit code in the datasheet and the pics and diagrams in our article. So we were very glad to read your post as further confirmation of those 2 bridges as relating to this function. Now if we can just get you to finish the job by tracing the bridges to the pins.;-)
(Of course only the 1 pin, 3rd from left, need be "toggled" to go between 133 and 166, AND, the mobo/bios has to provide "automatic" support for this function, not just a "user-defined" option in the bios).
Thanks, John C.
Last edited by candjac; 12-24-2002 at 07:35 PM.
|
|
|
12-25-2002, 08:08 PM
|
#4
|
|
Xtreme Member
Join Date: Nov 2002
Posts: 740
|
pardon my ignorance but what is the point to making it 166fsb when u can just easily set ur nios to 166fsb in most cases. Of course if ur motherboard doesn't simply allow that setting its a different argument.
So whats the diff between a moded 166fsb versus a 133 set to 166? there should be none correct?
__________________
xp2100+@2550mhz 2.03v 210fsb
Abit nf7-s v1.2
ATI R9700
2x512mb Cprsair pc3500 c2
Dual Channel!!!!!WOOT
SATA raid on 80gig maxtors
highest 3dmark 19300
3dmark2k1
|
|
|
12-31-2002, 06:32 AM
|
#5
|
|
Registered User
Join Date: Dec 2002
Posts: 2
|
Will this work with AMD XP chips moded for to AMD MP chips?
|
|
|
12-31-2002, 06:53 AM
|
#6
|
|
Xtreme Mentor
Join Date: Jun 2002
Location: PA, USA
Posts: 2,772
|
iampedro,
Welcome to Xtreme
__________________
| 1.6GHz Intel E2140 @ 3.2GHz 24/7 | Abit IP35-E @ 400MHz FSB | 2GB Micron PC2-5300 @ 800MHz | 320MB EVGA 8800GTS @ 615/1025 | 100% CPU OC on stock cooler & 24hr Prime95 stable |
|
|
|
12-31-2002, 06:55 AM
|
#7
|
|
Registered User
Join Date: Dec 2002
Posts: 2
|
Thanks...
|
|
|
12-31-2002, 08:15 AM
|
#8
|
|
Registered User
Join Date: Dec 2002
Posts: 4
|
question...
Hehe... this seems to be a good place to hang... some familiar faces too (Oppainter, etc).
Well, after reading about this trick on Warp2Search.net, I decided to stop by and pop a question:
I have a XP 2400+ (FSB 133Mhz, Multiplier 15x), an Abit KX7-333R voltage modded and 512MB of Corsair's XMS PC3200 DDR Ram... do you guys think I'll be able to hit 2.5Ghz with this trick???
While keeping this 15x multi I've been only able to push my FSB to 150 (with the 'user defined' option) while maintaining full system stability... anything higher, my PC won't boot or will reboot while loading Windows XP (I think what's limiting me is the fact that I have 4 PCI cards installed... don't you think?).
Hope to hear from you soon.
Thanks in advance...
And, for those celebrating today the coming of a new year, best wishes!
- D24
Last edited by Devil24; 12-31-2002 at 12:08 PM.
|
|
|
12-31-2002, 08:24 AM
|
#9
|
|
Registered User
Join Date: Dec 2002
Location: United States
Posts: 57
|
I doubt it. 2.5GHz is dependant on the stepping. If you cant hit 166Mhz before the trick, changing the auto-detect value of the proccesor wont help it, unless you change (lower) the multiplier too  .
__________________
- Radea
-Specs-
Outdated Crap.
|
|
|
12-31-2002, 10:30 AM
|
#10
|
|
Registered User
Join Date: Dec 2002
Posts: 5
|
I have tried the multiplier unlock on my 2400+ (on a 8K3A+). Its now running at 166 X12.5 , just like a 2600+. I will try the bus change mod when my new cooler arrives. Works great, I was running the processor at 140x15, but the 166 bus gives me an extra 700 3DMarks.
|
|
|
12-31-2002, 11:55 AM
|
#11
|
|
Registered User
Join Date: Dec 2002
Posts: 2
|
Yet another brought to the community via warp2search.net. I have a few questions!
- When you guys are hitting 200+FSB's, what are you using for cooling? Watercooling? If using a fan, what's the best to buy these days?
- How can you tell if you have a RevB processor?
- It seems like it would be the same, but based off of a member's post ... it made me wonder; which is better 12.5 x 133mhz or 10 x 166mhz. They are both 1666mhz (2000+), but is the higher FSB rating better? I could see where this would be preferable if you were using PC2700 SDRAM and were able to run in sync,
Thanks!
|
|
|
12-31-2002, 11:59 AM
|
#12
|
|
Registered User
Join Date: Dec 2002
Posts: 5
|
like i said the 166 bus gave me an extra 700 points, even though the processor is actually running a little slower.
|
|
|
12-31-2002, 12:03 PM
|
#13
|
|
Registered User
Join Date: Dec 2002
Posts: 2
|
Quote:
Originally posted by chrismm27
like i said the 166 bus gave me an extra 700 points, even though the processor is actually running a little slower.
|
You were the member I was referring to and gaining 700pts in 3dmark means next to nothing. What happens in the games? Do you see any noticeable difference in games?
|
|
|
12-31-2002, 12:10 PM
|
#14
|
|
Registered User
Join Date: Dec 2002
Posts: 5
|
I was a bit surprised really that it made such a difference, but its well worth it. You cant be sure your chip will take the extra bus speed though, whatever multiplier you use. I used to have a 950 duron that i was running on a 133 bus. I changed it for a morgan duron 1.3 but that wouldnt take a 133 bus what ever i did to it. The 2400+ is great though. There is a bit of a difference in games yes but thats not why i did it, i had no problem with frame rates.
Last edited by chrismm27; 12-31-2002 at 12:14 PM.
|
|
|
12-31-2002, 12:22 PM
|
#15
|
|
Registered User
Join Date: Dec 2002
Posts: 4
|
Ok...
This is a picture of my CPU right before I installed it... is that a good stepping???
AMD Athlon XP 2400+
|
|
|
12-31-2002, 12:35 PM
|
#16
|
|
Registered User
Join Date: Dec 2002
Location: North Adams, MA
Posts: 69
|
 oh yea thats a good stepping, and a good dig cam too
__________________
AMD Opteron 165 Dual Core * CCBWE:06/04:TPMW
DFI Lanparty SLI-DR
2x512 OCZ PC3200 Gold VX
XP90c (for now)
Ati X1900XT
Hitachi Deskstar 250GB SATAII
Sintek 500W single rail SLI Modular PS
Gigabyte 3d Aurora Black Aluminum Case
Viewsoinc VX922 19" 2ms LCD Monitor
|
|
|
12-31-2002, 12:40 PM
|
#17
|
|
Registered User
Join Date: Dec 2002
Posts: 2
|
Devil - At the same MHz overall CPU speed, the higher FSB setting for the CPU will usually be faster (certain kt-400 setups are the execption)... More memory bandwidth
If you are having trouble pushing your FSB past 150, you still might be able to get away with the 166Mhz FSB trick... You mobo right now is probably limited by the AGP/PCI bus and not the CPU or Memory.
If you push a 133bus with a 1/4 PCI multiplier to 166, you run the 33MHz PCI bus at 42mhz... too mush for most cards. The same hold true for the AGP bus... you end up running the 66Mhz bus at ~80Mhz. Thats why some people have noticed they cant overclock their CPUs as much when they got a Radeon9700... They were never being held back by their CPU's potential, but their AGP bus rather.
If you have a motherboard with a 1/5 multiplier for the PCI, your in there... thats what the L12 trick does... it fools your motherboard into thinking that you have a 166cpu that needs a 1/5 multiplier.
Heres my question now.... do KT-333 chipsets internally recognize the 166Mhz bus? Or do you need a KT-400/Nforce2??
__________________
Aopen AK7-333 @ 162 fsb
AthlonXP1600+ @ 1732
Radeon9700 @ 400/350
SamsungPC2700@ 162 cas2-2-2
14320 3DMark
|
|
|
12-31-2002, 12:44 PM
|
#18
|
|
Registered User
Join Date: Dec 2002
Posts: 5
|
yes the KT333 does, or it does on my Epox 8K3a+
|
|
|
12-31-2002, 12:56 PM
|
#19
|
|
Registered User
Join Date: Dec 2002
Location: North Adams, MA
Posts: 69
|
why a big hype over 166fsb? with proper ram just about any system can get higher than that, it soulds good to non overclockers , but what advantages does it give to us, its not like we can bump it to 233mhz any easier than a 133 would
__________________
AMD Opteron 165 Dual Core * CCBWE:06/04:TPMW
DFI Lanparty SLI-DR
2x512 OCZ PC3200 Gold VX
XP90c (for now)
Ati X1900XT
Hitachi Deskstar 250GB SATAII
Sintek 500W single rail SLI Modular PS
Gigabyte 3d Aurora Black Aluminum Case
Viewsoinc VX922 19" 2ms LCD Monitor
|
|
|
12-31-2002, 12:59 PM
|
#20
|
|
XIP - thisoneisonus.org
Join Date: Jun 2002
Location: Wakefield, West Yorkshire, UK
Posts: 3,405
|
Quote:
|
I have a XP 2400+ (FSB 133Mhz, Multiplier 15x), an Abit KX7-333R voltage modded and 512MB of Corsair's XMS PC3200 DDR Ram... do you guys think I'll be able to hit 2.5Ghz with this trick???
|
I run at 2.5Ghz 24/7 with my 2400+. Watercooled. Epox 8k5a3+. 512Mb Corsair XMS3500... 200x12.5, Cas 2-2-5-2-1T-8Level-2-2T
|
|
|
12-31-2002, 12:59 PM
|
#21
|
|
Registered User
Join Date: Dec 2002
Posts: 1
|
Hi,
This worked to unlock my 2400+ so I have the full range of low & high multipliers. It's so simple and so easily removed.
For the wire, I used the copper-wire from a failed ATA100 cable, perfect size. It needs to be just thick enough so that when the CPU is inserted, there is just the slightest resistence, too thick and you could damage 462 socket or CPU itself.
Cheerz
|
|
|
12-31-2002, 01:02 PM
|
#22
|
|
Registered User
Join Date: Dec 2002
Posts: 4
|
razor_dsm:
 My Olympus C-3020 Zoom DigiCam rocks
Wonder420:
Yes they do (I think)... although, I haven't tried a CPU that runs at that bus by default.
Whenever I've tried to run this CPU with a FSB of 166Mhz, I've always set the PCI divisor to 5:2:1 as that option is present in the BIOS.... so that shouldn't be a prob... my last CPU (a 1600+) ran without a prob at 166Mhz and beyond (180Mhz was the maximum bus I'd manage to achieve while maintaining full stability), so I KNOW this mobo can run a 133Mhz CPU @ 166Mhz very easily - or should I rephrase that to... 'I knew/supposed'  .
Last edited by Devil24; 12-31-2002 at 01:10 PM.
|
|
|
12-31-2002, 01:03 PM
|
#23
|
|
Registered User
Join Date: Dec 2002
Posts: 5
|
some of us use there PC for more than just gaming or internet. I use mine as a tv/video, as a sound system, movie player, server.. many things. Some of the cards i use will not take the very high FSB's (tv card especially). When i remove them i can get 200+ fsb, but this way i get good speed, reliability and fuctionality.
|
|
|
12-31-2002, 01:07 PM
|
#24
|
|
Registered User
Join Date: Dec 2002
Posts: 4
|
Merci:
Thanks for your reply... and good job!
But I was talking about keeping the 15x and bumping the FSB to 166Mhz.
Im really thinkg about buying a new mobo (nForce2)... they seem to be very o'clocking friendly and with the PCI-bus being locked, I think my chances of getting more out of my 2400+ should be a lot better.
Regards,
D24
Last edited by Devil24; 12-31-2002 at 01:12 PM.
|
|
|
12-31-2002, 01:09 PM
|
#25
|
|
LOVE=MC²
Join Date: Dec 2002
Posts: 3,444
|
What's up guys?
My cousin has a palomino 1900+ and he almost destroyed all the L1 bridges. I tried to unlock it for him but the lowest i got was 11.
Is it possible to get 166fsb on a palomino just by connecting a bridge? If, which bridge?
Thanx.
|
|
|
 |
|
| Thread Tools |
|
|
| Display Modes |
Rate This Thread |
Linear Mode
|
|
Posting Rules
|
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts
HTML code is Off
|
|
|
All times are GMT -8. The time now is 04:18 AM.
|