Fastest 66 chip running unmodified on BP6?

Batch codes, RAM specs, BIOS settings, etc..
Post Reply
BigChief
Posts: 1
Joined: Sun Mar 21, 2004 10:56 pm

Fastest 66 chip running unmodified on BP6?

Post by BigChief »

Ok so I inherited a BP6 that right now has two 366's on it, and I was wondering what the fastest 66mhz chips were I could get for it from say, eBay, without having to use adapters to get them to run on it. For instance, I saw some P3 core 533mhz celerons (not sure what exactly they're referred to, as I'm more of an AMD man) and wondered if the BP6 socket supported them, or if maybe it supports anything even faster. Whatever I get I plan on overclocking it, hence the 66mhz bus. I had just been looking for 533 P2 core celerons, that I was gonna bump up to 800s hopefully, but if I could get an SSE chip for not much more, that would be even better. I'm currently browsing around this huge site for the answers myself, but any pointing to topics I might be interested in would be appreciated.
davd_bob
Confused
Posts: 1043
Joined: Fri Feb 13, 2004 2:30 am
Location: Houston, TX

Post by davd_bob »

BigChief,
Welcome to the board.

If you want to use the BP6 with out modification then you can only use the Medocino core celerons. They are the PPGA package. The range is 300 to 533MHz. The Medocino core chip loses stability above 550MHz and that is probably why Intel abandond the line and moved to the Coppermine. Medocinos are the blackish ones with the larger block whereas the Coppermine are greenish with the small block.

The BP6 is primed to OC' best with 366s. That is why there are so many with that combo on eBay. Most 366s will accept 100FSB producing 550MHz. If you want to exceed that you will need some exotic cooling like peilters(sp) or water cooling or something. Experiment with the 366s before you think about changing and you might be pleasantly supprised.

Oh, you can bump up the voltage on the cpus but once you get them stable with your applications usually you can drop the voltage. I traded a pair of 500s for a pair of 366s and all 4 chips worked nicely below 1.95v resulting in lower temps even OC'ed.
There are *almost* no bad BP6s. There are mostly bad caps.

No BP6s remaining
Athlon 2800
Sempron 2000
ViaCPU laptop with Vista.(Works great after bumping ram to 2Gig)
P-III 850@100
Derek
Site Admin
Posts: 2489
Joined: Tue Jul 23, 2002 3:55 pm
Location: Canada
Contact:

Post by Derek »

I second davd's suggestion.

Overclocked 366s easily over-perform 533MHz Celerons. The greatest benefit of overclocked 366s is a 100MHz FSB.
-Derek
purrkur
Linux Guru
Posts: 687
Joined: Fri Dec 12, 2003 5:57 pm
Location: Sweden
Contact:

Post by purrkur »

Derek wrote:I second davd's suggestion.

Overclocked 366s easily over-perform 533MHz Celerons. The greatest benefit of overclocked 366s is a 100MHz FSB.
Well, as you see from my sig, I am running a pair of 533's at 576 MHz and I have yet to see a stability related crash or hang on this machine. I have simply never had it hang on me!

I tried running them at 600MHz and it worked fine until I got the board running some heavy job. After working hard for a while it would lock up. I am pretty sure that I could get it stable at 600 MHz if I was interested in using ultra-fast (and noisy) coolers and maybe some other mods.

As for a 550MHz/100MHz fsb combination over-performing my 576MHz/72 MHz combination I can only say this: Yes, in certain tests it would, but not all. If you are running something that is 100% cpu (like compiling) then my 576MHz will win regardless. A system running with a 100MHz fsb is not overclocking the bus either while I am so I am pretty sure that my combo won't lag too far behind in tests where it comes out under :)

Heh. This would be fun to test actually. Maybe in my next life I will have time!
2x533MHz@544MHz, 2.0V
640MB PC100 memory
Realtek RTL-8139 NIC
Maxtor 6Y080L0 80GB hdd
Debian Linux stable with 2.4.8 kernel
davd_bob
Confused
Posts: 1043
Joined: Fri Feb 13, 2004 2:30 am
Location: Houston, TX

Post by davd_bob »

purrkur wrote:
Heh. This would be fun to test actually. Maybe in my next life I will have time!
I know the feeling.

btw each user needs to understand if they will benifit more from increased FSB or processor MHZ and do their build accordingly. Some folks have the "Midas Touch" and have better luck with equipment the other users. I can get PC equipment to work fine that other people gave up on long ago...in fact thats where most of my equipment came from. I think Quantum_user got a pair of 400s to 660 on a 110FSB(drool) for a beowolf cluster system. He used special cooling.

Other posts show there is a Medocino wall between 525 and 600MHz that can only be crossed with exotic cooling...and Intel DID abandon the line without marketing a 566.

purrkur, PM'ed ya.
and before anyone jumps to conclusions, it was a nice PM not a flame.
There are *almost* no bad BP6s. There are mostly bad caps.

No BP6s remaining
Athlon 2800
Sempron 2000
ViaCPU laptop with Vista.(Works great after bumping ram to 2Gig)
P-III 850@100
Dave Rave
G'Day Mate!
Posts: 894
Joined: Thu Sep 26, 2002 4:28 am
Location: Sydney Australia
Contact:

Post by Dave Rave »

I think the 550/100 (5.5) will beat the 576/72 (8.0) in seti as clock multiplier helps a lot on number crunching.
something to do with flops and how long between fsb tics for a number crunch cycle to complete
24seven
IRC Lurker
Posts: 495
Joined: Wed Jul 24, 2002 5:23 pm
Location: Derbyshire UK
Contact:

Post by 24seven »

Let’s face it, there isn’t going to be that much difference. Well to be exact 28 MHz of fsb difference and 26 MHz difference.
davd_bob
Confused
Posts: 1043
Joined: Fri Feb 13, 2004 2:30 am
Location: Houston, TX

Post by davd_bob »

24seven wrote:Let’s face it, there isn’t going to be that much difference. Well to be exact 28 MHz of fsb difference and 26 MHz difference.
Looking from a different point of view its not the number of MHz but the ratio of change.
In the senario indicated the processing decrease is about 5% but the FSB increase is about 50%. Most applications would weigh the FSB increase more heavily then the CPU decrease especially if one is multi tasking and the CPU is sitting there waiting for data from the HD while the bus is busy doing something else. Of course something like a dedicated file server needs the FSB far more then then a stand alone system crunching SETI which would probably notice the loss of compute cycles more.

btw Derek,
Thanks for letting people throw stuff out there. I just realized a few days ago because of some lame discussion or other that running my 366@500on 92 FSB was rock solid stable BECAUSE I was UNDERCLOCKING everything else in the system.
Dang, I wish I was as smart as I think I am.
rosspp
Posts: 2
Joined: Tue Nov 18, 2003 9:39 am

Post by rosspp »

Are there any 100 MHz processors that work on the BP6 without adapters?
Do they work in SMP mode?
purrkur
Linux Guru
Posts: 687
Joined: Fri Dec 12, 2003 5:57 pm
Location: Sweden
Contact:

Post by purrkur »

davd_bob wrote: btw each user needs to understand if they will benifit more from increased FSB or processor MHZ and do their build accordingly. Some folks have the "Midas Touch" and have better luck with equipment the other users. I can get PC equipment to work fine that other people gave up on long ago...in fact thats where most of my equipment came from. I think Quantum_user got a pair of 400s to 660 on a 110FSB(drool) for a beowolf cluster system. He used special cooling.

Other posts show there is a Medocino wall between 525 and 600MHz that can only be crossed with exotic cooling...and Intel DID abandon the line without marketing a 566.

purrkur, PM'ed ya.
and before anyone jumps to conclusions, it was a nice PM not a flame.
Thanks for the PM! PM'ed ya back :)

I agree with what you wrote above. This discussion reminds me of the one I got with a good friend that has an AMD 2500XP (Barton) while I got an AMD 2400XP vanilla flavor. I got higher clock speed while he has got faster FSB. We have done some tests and it is pretty obvious that my machine will be faster in CPU intensive tests while tests that are more general (test more than a single thing) usually go faster on his machine. Higher FSB is generally a good thing to have!

Same goes for our BP6's, although I think 24Seven has it licked. We are talking extremely small differences at best. However, I am very satisfied with my dual 576MHz setup although I would be interested in running two 366MHz Cellys at 100MHz FSB. Wasn't Derek looking into buying and distributing such CPU's some time ago? Did anything come out of that?

Mendochino core did only reach 533MHz which makes me believe that Intel could get it stable at 566MHz or so because you don't sell a processor that is running at a speed where it is close to being unstable. However, anything above that would reaquire extreme cooling like davd_bob pointed out.

Oh, and rosspp, the answer is unfortunately "no". You need to modify the BP6 to get it working with any 100MHz fsb processor, even if you are only running a single one.
2x533MHz@544MHz, 2.0V
640MB PC100 memory
Realtek RTL-8139 NIC
Maxtor 6Y080L0 80GB hdd
Debian Linux stable with 2.4.8 kernel
rosspp
Posts: 2
Joined: Tue Nov 18, 2003 9:39 am

Post by rosspp »

Thanks Purrkur
I do have 3 Cel400 chips, 2 of which run at 100 MHz fsb singly.
(The 3rd won't go above around 93 MHz I think)
I can't get the 2 Cel400 chips to run together at 100 MHz on my bp6, so I thought I might just give up overclocking and buy some that were designed to run at 100 MHz. If these won't work in smp, then I'll go back to trying to get my Cel400s to run at 100MHz in smp mode.
I guess it's some sort of voltage problem?
purrkur
Linux Guru
Posts: 687
Joined: Fri Dec 12, 2003 5:57 pm
Location: Sweden
Contact:

Post by purrkur »

rosspp: You are stuck with those Celerons. Depending on what revision your bp6 board is, you might need to beef up the voltage regulators to the cpu's to get them working in SMP. Otherwise run them at 92MHz...
2x533MHz@544MHz, 2.0V
640MB PC100 memory
Realtek RTL-8139 NIC
Maxtor 6Y080L0 80GB hdd
Debian Linux stable with 2.4.8 kernel
davd_bob
Confused
Posts: 1043
Joined: Fri Feb 13, 2004 2:30 am
Location: Houston, TX

Post by davd_bob »

purrkur wrote:rosspp: You are stuck with those Celerons. Depending on what revision your bp6 board is, you might need to beef up the voltage regulators to the cpu's to get them working in SMP. Otherwise run them at 92MHz...
The two that can handle 100fsb in uni-processing could probably go to 96fsb. Try it at 2.1 volts and see. If ok then every week or so drop the voltage in 0.05 increments. You might get them down to 1.90 which will help the temps. Make sure to put HS grease on the 'GREENE.
There are *almost* no bad BP6s. There are mostly bad caps.

No BP6s remaining
Athlon 2800
Sempron 2000
ViaCPU laptop with Vista.(Works great after bumping ram to 2Gig)
P-III 850@100
24seven
IRC Lurker
Posts: 495
Joined: Wed Jul 24, 2002 5:23 pm
Location: Derbyshire UK
Contact:

Post by 24seven »

They might overclock better after theyve been run for a while. My 500's used to strugle posting at 600, but will now post at silly speeds.
purrkur
Linux Guru
Posts: 687
Joined: Fri Dec 12, 2003 5:57 pm
Location: Sweden
Contact:

Post by purrkur »

24seven wrote:They might overclock better after theyve been run for a while. My 500's used to strugle posting at 600, but will now post at silly speeds.
But can you run them successfully without issues at 600MHz?? My 533's will post at 600 and run for some time as long as I don't work them too hard. When I start working them too hard I can expect a lockup at anytime (at least last time I tried this out).
2x533MHz@544MHz, 2.0V
640MB PC100 memory
Realtek RTL-8139 NIC
Maxtor 6Y080L0 80GB hdd
Debian Linux stable with 2.4.8 kernel
jaybird
Posts: 301
Joined: Tue Jul 30, 2002 9:21 pm

433's @ 110 FSB!

Post by jaybird »

Just a little tickler, I just came across an old article that I had saved.

In it, it shows an unmodified BP6 board running Celeron 433's, matched, Maylay cpu's running @ 110mhz fsb!

NOTE: 433's have the "ideal" multiplier, of 6.5 soooo,,,, 6.5 x 110 = 715, this still keeps your AGP card and PCI bus in line with no over-clocking problems!

However, it required pelts and water cooling.

I have another BP6 that has the EC10 mod and the voltage regulator mod, now where did I lay those 95 watt pelts and that spare "mini-peltier" :wink:

Regards,

jaybird
kuun
Post-O-Matic
Posts: 1566
Joined: Sat Nov 02, 2002 5:26 pm
Location: Tennessee
Contact:

Post by kuun »

i will chime in my 2 cents worth

i was an avid 500mhz OC'd to 604

i ran that setup on my SCSI server for ages (yes she still runs)

til one day, i got my second bp6 (w00tness)

i put my remaining 366's (i had 4 that i traded for 2 of my 500 celerons (yeah, i got alot of celeron chips)) and oc'd them to 550, you wouldn't believe the performance of the server with the 366's over the 500's

even at 604 the SCSIserver couldn't handle the workload nor the data rtansfers nearly as smooth as the dual 366's @550

so for you out there that have a choice,

it's clearly the 366's

Kuun
!!! WARNING !!!
The following forums: www.bp6.com
are infected with the following VIRUS(s): Kuun.infected.all.posts.Win2K.user

The following IRC servers has been exploited: irc.bp6.com
with the Following Exploit: Kuun.lurks.using.mIRC.v5.82.exploit
Post Reply