Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Asus P2B rev 1.02 running with a PIII 1ghz?

182 views
Skip to first unread message

oliver

unread,
Dec 5, 2001, 8:35:51 AM12/5/01
to
Hi, this question has probably been asked quite a lot, but my news
server won't show me msgs that are that old.

With the upcomming pricecuts on the somewhat older PIII cpu's, i was
wondering if I could upgrade to a new PIII (currently PIII 500). I know
that i can run anything up to 850Mhz with a slotket (got a soltek 2++
something which supports the Coppermine) now my question is, will it
also run the PIII 1ghz E. I know the FSB of this one is 100mhz, and the
corevoltage is a bit higher making it a little 'easyer' on the CPU. In
other words, wil the BX chipset work with the 10x multiplier the PIII
1ghz wants.

P.S. Does someone also have a copy of the doc of H.oda's rework of an
asus to 'upgrade' it to an asus v1.10? I only have some japanese version
so i only have the pictures.

thank you

Matt Makauskas

unread,
Dec 5, 2001, 4:34:25 PM12/5/01
to
Hey there. Oliver, can you send me he doc of H.oda's rework of an asus to 'upgrade' it to an asus v1.10?

Here's my story:

I just got the SL5QW (P3-1100E) 1.1GHz 100MHz FSB yesterday.

Until then I was running a P3-700E on my ASUS P2B-S rev1.02 @ 112MHz FSB (hardware overclock with jumpers) without any problems. I was using an ABIT SlotKet III adapter set to 1.8v (which is higher than the 1.65v that the 700E is rated at, but is the lowest voltage the p2B-S rev1.02 can run - it wouldn't work set to 1.65v)

The 1100E is rated at 1.75v, so I thought that running at 1.8 would be really safe and stable and running at at 112MHz FSB would just be sweet if it was possible (112MHz FSB is the fastest FSB that the P2B-S supports, I know this is different for the P2B). Then I found this on the web:

http://bp6.com/Forum/topic.asp?topic_id=9731&forum_id=30&Topic_Title=VP6+%2B+SL5QW+%3D+1441MHz+%3D+sweet%21&forum_title=VP6&M=0&S=

And if that wasn't enough to make me want the chip even more, I found it for $258 Canadian!! at one of the computer shops that I always visit.

So I bought it, hoping that it would run, at leat at 11x100MHz.

When I plugged it in, it's wouldn't boot at 1.8v, 112MHz FSB, dropped it to 100MHz FSB and still nothing. I even upped the voltage in incrememnts, but still no go.
it finally booted when I tried 66MHz FSB (back at 1.8v again). yay, it boots, runs fine, but 66MHz FSB was unacceptable. Long ago, I used to use a program called SoftFSB (which works seemlessly, for me) to overclock the 700E; decided to try it and it worked. I can run the chip at 100 and 103 stable, and at 110 and 112, it's not perfectly stable (Quake 3 crashes while loading), but runs. Yay, I'm up over 1.0GHz now, but I'm still unhappy that I have to wait for 98SE to boot and then run SoftFSB to get the chip up to speed.

I still have a few things to try before I close my case up.. And I'm also going to see what the chip can do on my friend's P3B-F and P3V4X.

I don't know what would not let this chip boot at 11x100MHz, maybe a differnt FC-PGA to Slot1 adapter would work, maybe I need a better power supply...

Just as an aside, one of the reasons why I bought the chip was because I didn't want to have to buy a new MB. The P2B-S has and Ultra2Wide SCSI controller on it, which I use, and buying a new MB would mean buying an expensive SCSI controller too. If I was going to buy a new MB, I'g go all the way and get near to the fastest hardware that's available. But for 258 bucks ($CAN), I thought that the 1100E would keep me happy for maybe another year. It probably will, even though it's not as convinient as I though it would be (booting at 66MHz FSB and then using software to overclock back up to 100MHz FSB). It really sucks when things don't work seemlessly.

Pablo

unread,
Dec 5, 2001, 5:19:49 PM12/5/01
to
If you use the registered copy of a number of the software FSB
utilitys....and configure them.....you really dont notice that you had to
bump the FSB from inside windows...except it takes longer to boot....what a
pain!
Im wondering if B21 has some place in this problem. The CPU does run when
started at 66...which makes me wonder if the detection is not picking up the
CPU characteristics......perhaps blocking out the B21 pin like we used to do
on the old 66 mhz CPUs to force them to go to 100 Mhz FSB is in order....I
might be tempted to take alittle fingernail polish to my slotket and see if
it helped with your problem. BXBOARDS has documentation how we did this with
66 Mhz Slot ones....
But it would seem that the slotket is not passing the correct info to the
VID et al
Any one else have a thought on this? I dont have a P2B that I can do this to
at the moment.....nor do I have an 1100 to test with....but it would be a
harmless test....were someone so inclined.
"Matt Makauskas" <mat...@rogers.com> wrote in message
news:3C0E930C...@rogers.com...

Roland Scheidegger

unread,
Dec 5, 2001, 5:26:32 PM12/5/01
to
The 1.1Ghz PIII and 1.1Ghz Celeron will not work in any P2B-xx series
board, this is supposed to be because of a bios bug even in the latest
beta (seems to have something to do with an error calculating the cpu
frequency with multiplier=11 and higher fsb frequencies). Nothing you
can do about it other than trying to send asus an email (or to try to
hack the bios). Those 1.1Ghz chips work fine in the very similar p3b-f
and cubx boards.


Roland

Pablo

unread,
Dec 5, 2001, 6:35:47 PM12/5/01
to
Im aware of the bios bug problem....but if he can get his to actually start
at 66FSB.....then this raises my interest.....always looking for a way
around things like this...esp since I (and you for that matter) seem to be
answering alot of these type questions Roland......guess it couldnt be as
simple as the ol B21 trick or someone would have tried it by now.
"Roland Scheidegger" <rschei...@gmx.ch> wrote in message
news:3C0E9F18...@gmx.ch...

Roland Scheidegger

unread,
Dec 5, 2001, 7:23:24 PM12/5/01
to
I don't think it's possible to solve this with some trick. But there are
people who added k6-III+ cpu detection to t2p4 bios and such things,
so they should be able to correct that one too - without the bios
sources, probably a bit time consuming however.
I think the reason why you can't overclock when using SoftFSB to raise
the FSB when booting with 66MHZ FSB is that it does only change the pci
divider but not the agp frequency divider, and because the 1.02 revision
of the p2b doesn't have the agpfs jumper, this means AGP will run at 1:1
ratio to system bus - not every graphic card will handle 100Mhz or even
more AGP clock. Not sure if you can overclock if you have a revision
with that jumper - don't know if the jumper really does set the agp
ratio to 2/3 if you boot with 66Mhz FSB.

Roland

oliver

unread,
Dec 5, 2001, 7:32:02 PM12/5/01
to
Was it not that there are only a few 1100E's out thre, i actually thought there was 850E and 1000E didn't know about a 1100E. Eitherway, Does this apply to the 1ghz version aswell? the not wanting to boot at 100FSB. I am running a PIII 500 at 112 x 5 fully stable, ram at cas 2 -2-2-2 etc etc.
the thing is, i use llinux a lot to, so doing that softFSB thing isn't possible i think (i did use it to 'test' the 112 speed at first)

BTW, I read about a beta bios on the german asus site for the p2b regarding the 1100E CPU's. That might probably be the thing you guys are talking about. The required 'fix'. I'm still wondering if this applies to a 1ghz CPU aswell, sure i'd like the 1100 aswell.

oh, and pushing your CPU to 11x112 is pretty mad lol
11x112 =1232Mhz and that's bumping it pretty hard. 103FSB is still nice though. what WOULD be a LOT better is a nice 10x112. 1120 Mhz, just a little over spec, and your FSB gets pumped. sucky that they are multiplier locked ... ahh .. the old days ; )

thanks for your reply though
oliver

oliver

unread,
Dec 6, 2001, 6:46:48 AM12/6/01
to
the AGP devidor is set by the BX chipset to either 1:1 or 2:3 as it states in the
manual. So that's not really the problem, besides, if that would be the problem,
NO 100Mhz fsb CPU would work. Aswell as that if it was possible to change the AGP
seperatly from the PCI like that, Asus would have included that in the bios as an
option (or SoftFSB in it's software)
I know the early revisions don't support soft bios yet, but as many know, the mobo
Does support it, only the 'jumper' is solderd to 'jumper mode'

Roland Scheidegger

unread,
Dec 6, 2001, 9:12:25 AM12/6/01
to
But this is exactly the problem, the ratio is only set at boot time. So,
if you boot at 66Mhz and later use SoftFSB to increase the frequency, it
will stay at 1:1. But if you boot at 100Mhz (as you normally do with
100Mhz FSB cpu's) the ratio will be set to 2:3. AFAIK SoftFSB only
programs the clock chip (pci divider is determined by the clock chip)
and doesn't touch agp divider (which is determined by the bx chipset,
not sure if you even can change it when running in protected mode
without crashing).

Roland

Matt Makauskas

unread,
Dec 7, 2001, 3:55:15 AM12/7/01
to
I never thought of the 1:1 AGP problem when booting at 66MHz. It could explain why my system isn't stable at 103MHz, but stable at 100MHz (SoftSFB overclock).
I've read that Geforce based cards, perhaps especially Geforce2, are really stable at higher AGP speed (most often read about 2/3 of 133MHz = 89MHz).
perhaps around 100MHz AGP is the threshold for my Geforce2 MX card and that's why I get crashed doing graphically intense stuff over 100MHz. Funny though, I can do regular windows stuff (browsing my HD, browsing the web, listening to WinAMP) at 112MHz. only hard core CPU and graphics crash the system at over 100MHz.

I'd have a really good life if the final of BIOS v1014 fixed this problem.

:-)

Matt Makauskas

unread,
Dec 7, 2001, 4:05:31 AM12/7/01
to
I didn't know about the 1100E until I stumble upon this page:
http://support.intel.com/support/processors/sspec/p3p.htm

there is even a 900E - that must be really rare.

as for my issues applying to the 1000E, I'm not sure. I was telling the story to perhaps make you a little more aware or causious. there are some other posts in this thread that seem to say that my problems are related to the BIOS not recognising the 11x multiplier. maybe if it does recognize the 10x multiplier of the 1000E, then you'll be ok.

anyhow, you've got me all excited about reading something on the asus site regarding the 1100E and the beta BIOS. I am using 1014 BIOS that's availible now (which is beta), but maybe there is mention of a future fix for my problem... I'm off to search for it...

Matt

Matt Makauskas

unread,
Dec 7, 2001, 4:25:50 AM12/7/01
to
Hey Oliver, could you point me in the direction of the info you found reagrding the 1100E and the 1014 beta bios. I cant seem to find it.

thanks,
matt

oliver

unread,
Dec 7, 2001, 11:50:48 AM12/7/01
to
Matt,

I don't remember were I read it, I think i searched all the msgs in this newsgroup regarding the P2B adn I think it came up there. they said a beta 2 bios was on the german asus site, but not the US one.
hope that helps

avid

unread,
Dec 7, 2001, 1:01:37 PM12/7/01
to
Matt and Oliver,

I am very interested in this particular subject myself but you guys
are making reading your posts very painful. You need to set the line
length in your newsreader to something less than it is. Right now most
people are probably having to use the scroll bar to read your
messages. How about 75 characters per line and 80 at the max.

TIA


--
avid
email address not valid

Pablo

unread,
Dec 7, 2001, 8:57:02 PM12/7/01
to
I tried modding the bios, and there really is not an entry in the CPU
microcode that will help......I imported from the CUBX which does not have
the bug, despite being a BX board, and gave the bios to a friend who says it
did not solve the 1.1 problem for him. I can only ask him to swap CPU so
many times, so I probably wont get another chance at this...but bios modding
is pretty easy....someone else out can take another shot...but it was a dead
end for me. We also messed with pins on the slotket edge to force
detection.....not working either. Open to more suggestions.
"avid" <av...@my-deja.com> wrote in message
news:7e021u420fvfli8nd...@4ax.com...

s...@west.com

unread,
Dec 9, 2001, 5:02:43 AM12/9/01
to

A great thing about Agent:

Options
Menus and Toolbars
Toolbars
add command "Message Word Wrap" icon to toolbar.

point and click as necessary... problem goes away ;)


s...@west.com

unread,
Dec 9, 2001, 5:16:16 AM12/9/01
to

p2b rev 1.02 will run at 1Gig.

you need FCPGA and a slocket card. You have to set the card to request 1.8
volts for the cpu. The motherboard voltage regulator supplies the 1.8
volts. If you get a cD0 stepping it's only a 0.05 volt "overvoltage".
Not worth mentioning.

you can run 100mhz if you can find the cpu, or you can run 133mhz with the
PCI bus at 33mhz and the AGP bus at 89Mhz.

Your ram should be capable of running at 133 and your video card has to
handle 89mhz AGP.

You can use a PCI video card, but why waste that 89mhz bus ?


Here's some old background information on FC-PGA coppermines on BX boards.
lots of it geared towards overclocking at 133mhz fsb.

You'll need the latest P2B motherboard bios and the latest SBLive drivers
if you use that sound card.

Note that on a P2B motherboard with revision level less than 1.12, the on
board voltage regulator can only go down to 1.8 volts. A FC-PGA coppermine
will boot and run at this voltage, but it is technically an overvoltage
condition. On the other hand, although a 500e is spec'd at 1.6volts, Intel
runs it's 1.13 gig coppermine (same core) at 1.8 volts.

the motherboard revision level is printed between the ISA slots. Some
P2B's prior to rev 1.12 have the newer voltage regulator that can go down
to 1.6 volts. The easy way to check this is to set your slocket voltage
jumpers to 1.65 volts (or appropriate cpu voltage) and see if your FC-PGA
coppermine boots. If it does you have the newer regulator; if it doesn't,
switch off the power and set your slocket voltage jumpers to 1.8 volts and
it should boot.

there are three generations of slot 1 to socket adapter cards:

first generation were for Celerons only.
Iwill slocket
various others

second generation support both FC-PGA and celerons.
Asus S370-133
Iwill Slocket II
Gigabyte GA6R7 Pro
MSI MS6905 Master
Soltek SL-02A++

third generation have on card voltage regulators.
Iwill Slocket III (never manufactured ?)
Soltek SL-02D
(Latest was this card would not be manufactured.)
http://www.soltek.com.tw/English/products/02d.htm

PowerLeap PL-iP3 (rather expensive solution)
http://www.friendtech.com
http://www.powerleap.com/Products/iP3.htm

User Survey Chart of Asus P2B-xx motherboards with coppermine support.
http://www.ctechnet.com/hardware/asus.htm#10100

Note: see Celeron/P2B-xx chart on same page.

Anand's page on running 133mhz on the Asus P2B motherboard.
http://www.anandtech.com/showdoc.html?i=34

Anand's guide to overclocking the FC-PGA Coppermine
http://www.anandtech.com/showdoc.html?i=1103

Anand's guide to overclocking the new Celerons
http://www.anandtech.com/showdoc.html?i=1227

Tom's guide to PIII overclocking performance
http://www7.tomshardware.com/cpu/00q2/000511/index.html

Tom's comparing the BX chipset at 133mhz to the newer chipsets.
http://www7.tomshardware.com/mainboard/00q1/000308/index.html

Anand's Roundup of 440BX 133Mhz Motherboards
http://www.anandtech.com/showdoc.html?i=1250

Tom's info on agp cards running at 89mhz with 133mhz FSB settings.
http://www.tomshardware.com/graphic/00q1/000321/index.html

Anand's info on agp cards running at 133mhz and higher.
http://www.anandtech.com/showdoc.html?i=1270

An interesting performance comparison chart from Tom's 150mhz Project Part2
http://www.tomshardware.com/mainboard/00q2/000503/155mhzbus-05.html

Another cpu/chipset performance comparison chart from Anand's.
http://www.anandtech.com/showdoc.html?i=1290&p=6

info on FC-PGA to Slot One converter cards, three reviewed.
http://www.lostcircuits.com/cpu/slocketii/

Anand's review of the IWILL Slocket II converter card.
http://www.anandtech.com/showdoc.html?i=1140

Review of Asus S370-133 Slocket card
http://www.pcstats.com/articleview.cfm?articleID=175

SharkyExtreme review of the 500e/550e coppermines
http://www.sharkyextreme.com/hardware/reviews/cpu/intel_flipchip/

article on FC-PGA Coppermine 500e
http://www.ixbt-labs.com/cpu/intel-fcpga.html

Anand's review of PC133 ram
http://www.anandtech.com/showdoc.html?i=1213

Roland Scheidegger's FAQ on RAM modules compatible with the i440bx chipset
http://n.ethz.ch/student/sroland/ram_bx_faq.html

special concerns with micron memory at 133mhz ??
http://www.cluboverclocker.com/reviews/memory/micron/mt_pc133/micron_pc133_test2.htm

some nice pix of an overclocked BX chipset cooler and various tips.
http://bxboards.com/pc133/p133-8.shtml

Case Cooling Fun (great stuff and pix)
http://www.virtualhideout.net/cool_case/index.shtml

One blower instead of a half dozen fans.
http://www.overclockers.com/tips461/

Preview of Agilent ArtiCooler heatsink/fan for FC-PGA
http://www.hardocp.com/reviews/cooling/arctic_ca/index.html

industrial strength cooling ;)
www.swiftnets.com

Still want to run a slot 1 Coppermine Cpu ?
Check this out...

http://www.hyperformance-pc.com/overclocking_faq.htm#voltageadjustments

PS it's a lot easier to do with a FC-PGA and a slocket card ;)

Last but not least, the PIII/1266/133 (512k cache) CPU you won't run on
your BX:

http://www.tomshardware.com/cpu/01q3/010919/index.html


oliver

unread,
Dec 11, 2001, 8:50:41 AM12/11/01
to
Thanks alot, allthough I found most of this info allready online, but about the 850Mhz
CPU's and was mostly wondering about 10x multiplier (which is sorta confirmed now) also,
I'm running 112Mhz FSB with the jumpers (not SoftFSB anymore) fully stable, no boot
problems no nothing. I have the VooDoo3 3000 and it was even tested at tomshardware at
133FSB (89Mhz AGP) to be fully stable, so that's yet a nother worry less. meaning
basicly I can run anything I want including the 133Mhz FSB's. (which i'm not gonna but
it's nice to know that I can) also, i got some cas2 PC133 ram so that problem should be
solved with that aswell.

Thanks a lot for all the info

0 new messages