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

Can it be done? (PIII's to 1.2 GHz)

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

Aerosiecki

Registered
Joined
Sep 7, 2001
Location
Atlanta, GA, USA
I know that it's possible, at least it has been done before. My question is whether this is reasonable to consider as a system setup:

I want to take a pair of PIII 850's (which are 100 mhz bus) and take them to at least about 1.13 Ghz (133 bus), possibly higher.

All outside factors being equal (that is, I have a good OCing mobo, memory capable of 150+mhz, and excellent cooling set up) is it reasonable to think I can just by some 850's off the shelf (or from pricewatch) and just zip them up to such a speed?

D
 
Aerosiecki said:
I know that it's possible, at least it has been done before. My question is whether this is reasonable to consider as a system setup:

I want to take a pair of PIII 850's (which are 100 mhz bus) and take them to at least about 1.13 Ghz (133 bus), possibly higher.

All outside factors being equal (that is, I have a good OCing mobo, memory capable of 150+mhz, and excellent cooling set up) is it reasonable to think I can just by some 850's off the shelf (or from pricewatch) and just zip them up to such a speed?

D


Unfortunately, most of P3 800- 850 processors are well known for being bad overclockers. You would have to be very lucky to overclock them where you want i.e. 1100mhz+. None of the P3 850's have cD0 steppings available. But the P3 800's do have some cD0 stepping which may be not easy to find. You might get a little more luck with a P3 800cD0 or maybe cC0. You could overclock these pretty well.. about 1133+/- with good cooling as you mentioned. It's up to you... but I wish you good luck getting a OC'ing chip. any more question, i'd be glad to help =)
 
Last edited:
The cD0 stepping is the best bet to achieve the higher overclocks, but they are only available with a default 133 FSB. These CPUs have a lower multiplier than the 100 FSB CPUs rated at the same speed, so that's something to be aware of when consider how fast they will overclock at 150 FSB. The other thing is that dual procs are MUCH harder to overclock to the same speed than a single CPU. There are no guarantees in overclocking.
 
Last edited:
Thanks for all the great replies! I'm glad I asked ^_^

I suppose I have a new question as a result, then.

I already have my ECS D6VAA motherboard (takes two PIII's, but overclocks really well for a duallie, including per-chip Vcore adjustments!) And I realize I need to get a pair of matched-stepping chips (i've made that mistake before :p )

I guess I am just looking for reccomendations on what chips to throw in the thing. I'd LOVE to run at a memory speed of 150mhz.

From what you've said it sounds like the 800's would be good, but they come in both the 100 and 133 bus flavours, which would be better? The 100 bus gets me a bigger multiplier, so I'd assume it'd go higher. Or I could get an 800 that's already at 133 and bump up to 150. Without actually having overclock it too much.

Or I could go with an entirely faster chip (heard wonderful things about 1000e's before).

Opinions? I just wanna break the 1ghz barrier and have a memory speed that's a good deal above 133 if possible. Well, all that and I don't want to break the bank on them. No $300 chips.

D
 
To achieve your goal of a pair of reasonably priced P-III's that will overclock in the 1.13 gig range with a FSB of 150 MHz, the logical choice would be the 1 gig (1000EB) cD0 stepping CPUs that run at a default 133 FSB. According to price watch, you can find them for about $160 to $170 US shipped. Wasn't sure if when you said you didn't want to spend more than $300 whether you meant each or total. Personally, I think those CPUs would be awesome for a dual proc rig and ideal for your system and if I were doing it, that's the route I'd go. Those cD0 1 gig P-III's are proven to be overclocking friendly.
 
Listen to Batboy.. Those 1gig cD0 have had great overclocking here in the Ocers.com

Lots of people will manage 1.2 Gig or more.
 
batboy said:
the logical choice would be the 1 gig (1000EB) cD0 stepping CPUs that run at a default 133 FSB
I guess I'm down to deciding between the 1000E or the 1000EB . . . I want fast bus speeds, so I'm guessing the EB will be more of a guarantee of that, but the E has a higher multiplier, so has greater potential to get to 1.4+ (if I wanted such an insane thing).

So hmm, guaranteed memory speed, or potential for insane cycles per second . . . decisions, decisions.

D
 
Yes the 133 FSB P3 will be better..

Grab yourself that and some PC-150 ram.. You'll need it if you want your FSB to reach 150 or so.
 
I guess I'm down to deciding between the 1000E or the 1000EB . . . I want fast bus speeds, so I'm guessing the EB will be more of a guarantee of that, but the E has a higher multiplier, so has greater potential to get to 1.4+ (if I wanted such an insane thing).

Well, first off, there are not many coppermine P-III's that will even go past 1.3 gig, so 1.4 gig is merely a pipe dream. Second, dual procs rarely will overclock as much as each one will individually. Even with a single CPU, 1.2 gig is probably a more realistic goal. With dual procs, 1.13 is a more realistic goal. If you achieve more, then great, consider yourself fortunate. Third, the 1000E is a cC0 stepping which does not overclock quite as high on average as the 1000EB cD0 stepping.

Remember, just because several people reach a certain overclocking speed with a particular CPU does not mean all all those CPUs will do the same. Different sustems using various motherboards, components, cooling, and quality of the CPU core all contribute to the ultimate overclocking success. Often, those really high overclocks are systems with lots and lots of cooling mods, plus, maybe even peltiers and/or water cooling.

Often, some of our more youthful competitive members who want to have top bragging rights will report a speed that they achieved when it's not really stable. I've gotten my P-III 700 up to 1.1 gig before. Was it stable? Heck no. I could not do anything with it, it crashed all the time. But it booted and started Windows. However, I can benchmark my system at 1071 MHz with it being quite stable, but I have to run the voltage higher than I like. So, I normally run at 1050 everyday. Rock solid stability and a cool running system has more appeal to me than running maximum overclock and cussing because it crashes all the time.
 
Last edited:
batboy said:
I guess I'm down to deciding between the 1000E or the 1000EB . . . I want fast bus speeds, so I'm guessing the EB will be more of a guarantee of that, but the E has a higher multiplier, so has greater potential to get to 1.4+ (if I wanted such an insane thing).

Well, first off, there are not many coppermine P-III's that will even go past 1.3 gig, so 1.4 gig is merely a pipe dream. Second, dual procs rarely will overclock as much as each one will individually. Even with a single CPU, 1.2 gig is probably a more realistic goal. With dual procs, 1.13 is a more realistic goal. If you achieve more, then great, consider yourself fortunate. Third, the 1000E is a cC0 stepping which does not overclock quite as high on average as the 1000EB cD0 stepping.

Remember, just because several people reach a certain overclocking speed with a particular CPU does not mean all all those CPUs will do the same. Different sustems using various motherboards, components, cooling, and quality of the CPU core all contribute to the ultimate overclocking success. Often, those really high overclocks are systems with lots and lots of cooling mods, plus, maybe even peltiers and/or water cooling.

Often, some of our more youthful competitive members who want to have top bragging rights will report a speed that they achieved when it's not really stable. I've gotten my P-III 700 up to 1.1 gig before. Was it stable? Heck no. I could not do anything with it, it crashed all the time. But it booted and started Windows. However, I can benchmark my system at 1071 MHz with it being quite stable, but I have to run the voltage higher than I like. So, I normally run at 1050 everyday. Rock solid stability and a cool running system has more appeal to me than running maximum overclock and cussing because it crashes all the time.

Batboy, hi

First of all, you mentioned that only the cC0 steppings are available with the 1000E (10x) processors. I am surprised you said that.. ??.. that is not true. I thought you have read my previous threads about my overclocking success on my P3 1000E@1400+. Maybe you have not read it.. There are many others who are also running their 1000e cD0@ 1333 upto 1370 seriously Stable(including mine). As a matter of fact, my 1000E@ upto 1410mhz, It is so stable running any types of office apps and heavy gaming. At this mhz, I can run all of my games with Zero number of crashes and more stable than when I had my P3 700E@ 1064. Why??.. cauze those 700E's(including yours) are at high FSB of upto 150+ oc'ed and would give unstability in anything! I know it.. you know it.. Whereas running this 1000E@ 140+ FSB is allowing me to have much better stability in gaming. And you know I am running a BX board.

Please note this.. just because you have reached at X number of mhz with your CPU doesn't mean others have to be overclocked to similar mhz as you and be stable as yours. Others **Can** go much higher with **better** stability than yours, no kidding. I am not trying to put down EB chips here.. but I am trying to let you know that not all E chips are bad overclockers. The new stepping on these 1000E's really do overclock well. Now the 1000E cD0 steppings are easy to find.. they are new and start selling well. and please don't take this bad, im just pointing out misinformation here. =)
 
Last edited:
Sorry \(^0^)/, I didn't mean to offend you. I just checked the Intel website again and looked at the S-Spec chart in case I might of been mistaken and I see that I was wrong and you are correct. They do make a 1000E cD0 stepping P-III. Apparently this was done fairly recently. Now that you mention it, I do remember hearing something about it. My bad.

However, just because you and a few others have gotten exceptional CPUs and know all the tweaking tricks and how to properly cool your system, does not mean that everyone can achieve the same results. Overclocking is always a gamble. There will be plenty of CPUs just like yours that will not go as high as you pushed yours. To tell a newbie that they can hit 1.4 gig with any P-III is not wise or correct. Also, a dual proc system will never overclock as high as a single CPU system.

Yes, the potential is there to get an exceptional CPU, but you might also get a dog. A few cC0 700 chips like mine have made it to 1.2 gig. Lots of similar 700 chips barely make it to 933 MHz. Just like lots of 1 gig cD0 CPUs won't go above 1.2 gig. Let's have a reality check here. Also, I never said that the cC0 stepping chips did not overclock well. On the contrary, they are very good overclockers. It's just that the cD0 CPUs, ON AVERAGE, tend to overclock better.
 
Yodums said:
Grab yourself that and some PC-150 ram.. You'll need it if you want your FSB to reach 150 or so.
Check. I've got a stick of Mushkin Rev 3+ that'll not only do 150Mhz, it'll pull CAS 2-2-2 at 150. Hehe

batboy said:
Well, first off, there are not many coppermine P-III's that will even go past 1.3 gig, so 1.4 gig is merely a pipe dream. Second, dual procs rarely will overclock as much as each one will individually.
Oh, I know. Like I originally said, only about 1.13 is my "goal", anything above that is a nicety. And I've been playing with duallie rigs for a while, so I like to think (modesty, be with me!) that I can pull it off. I've put a lot of research into motherboards and chip combos, and made sure that others have pulled it off first, too.

Thanks for all the help guys! I'm going with the 1000EBs so that I know I can get a memory speed that's nice. I'll be back in month or so when I dig up the cash to tell you all how it goes!

D
 
batboy said:
Sorry \(^0^)/, I didn't mean to offend you. I just checked the Intel website again and looked at the S-Spec chart in case I might of been mistaken and I see that I was wrong and you are correct. They do make a 1000E cD0 stepping P-III. Apparently this was done fairly recently. Now that you mention it, I do remember hearing something about it. My bad.

no problem =)

However, just because you and a few others have gotten exceptional CPUs and know all the tweaking tricks and how to properly cool your system, does not mean that everyone can achieve the same results.


Any overclockers including Newbies suppose to use proper cooling. Even with proper cooling ofcourse people may not get the same result. It's given.

Overclocking is always a gamble.


We know that.

There will be plenty of CPUs just like yours that will not go as high as you pushed yours. To tell a newbie that they can hit 1.4 gig with any P-III is not wise or correct.


Not any P3's.. It has to be cD0 1000E for this particular range of overclock (1333mhz) is very feasible and do not need special care to get running at that speed. Again, as I mentioned above, any overclockers including the Newbies with an overclocking board and cD0 1000E will very highly likely do 1333 or more. I know at least couple of guys in Anandtech running their 1000E@ 1350 with air. Not a few, but quite a bunch of people right here in Intel CPU forum who bought SL5QV cD0 1000E chips, they were/are able to overclock to 1333mhz right out of it. Just about every one of them who bought this CPU quite easily able to attain 1333mhz and Up. They didn't need special cooling, not even special air cooling like the Sk-6's. With any descent air cooled HSF's will shove it. The part of reason why I am able to go this high is because of the SK-6 and from little part of my burn-in process. Pecos has the same HSF as mine and he is able to at least run some apps at 1400mhz. Auky has one also and doing 1360 Rock Solid. The rest of those with cD0 1000E with the lesser cooling HSF's doing 1333 and 1350 just fine. The CPU itself is definitely capable of running 1.33Ghz+. They are bound to this level of OC. I haven't seen any Dud 1000E SL5QV chips yet. Oh guess what.. I just bought another 1000E cD0 and I got 1350 out of it right out of the box, first attemp. Yes, I have two chips. Don't you see that it seems that these cD0's are very OC'able? yes.


Also, a dual proc system will never overclock as high as a single CPU system.


Umm.. I was talking about single processors, not dual. and yes, I know that the Duals are harder to OC the same as the single, but you can't say "will never", that's not true. If the user knows what he/she is doing on OC'ing the dualies like the people (Tim(Moderator), Kingslayer) are running Twin 700E's@ 1050 and Twin P3 800EB@ 1008. And there are more people who have managed to run just as high as it can with a single. So you can't say never. ;)

Yes, the potential is there to get an exceptional CPU, but you might also get a dog. A few cC0 700 chips like mine have made it to 1.2 gig. Lots of similar 700 chips barely make it to 933 MHz. Just like lots of 1 gig cD0 CPUs won't go above 1.2 gig. Let's have a reality check here.


umm.. I did not mentioned about the P3 700's that you might get a dog one because you can get dog one quite easiler with those 850's, 800's, 700's (even the cC0's!!). This is something I knew long time ago. And how can you get the 700's to 1.2Ghz?? That's little too much. You meant breaking 1100mhz, right?


Also, I never said that the cC0 stepping chips did not overclock well. On the contrary, they are very good overclockers. It's just that the cD0 CPUs, ON AVERAGE, tend to overclock better.

Not from what I've been seeing with the 1000E cD0's. By the way.. 1000E cD0's are only available in SL5QV S-Spec code. It is not AVERAGE, it's more of "Majority" in this case. ;) :)
 
Last edited:
Pinky said:
I'm glad I've been posting to other threads :).

That's why I'm trying to avoid replying to your post if there's possible arguments that could stir up. After that last incident with you at the Video card section, I don't want to compete with you coz...... coz........ :)

Anything beside arguments, if you need help, I help.. if I need help, then u do the same for me. Will you?.. =) There's always something one knows something that others don't know. :D
 
Last edited:
\(^0^)/ said:


That's why I'm trying to avoid replying to your post if there's possible arguments that could stir up. After that last incident with you at the Video card section, I don't want to compete with you coz...... coz........ :)

Anything beside arguments, if you need help, I help.. if I need help, then u do the same for me. Will you?.. =) There's always something one knows something that others don't know. :D

I guess I don't remember what you're referring to... glad one of us is nursing a healthy grudge :p.

I was only joking around... and no, the argument hasn't started, thought I might drop a line to avoid one :).
 
Jeez \(^0^)/, just because you can quote off a dozen people that hit 1.4 gig or close to that mark with a 1000E cD0 does NOT mean they all will or that the majority of them will. That's hardly scientific evidence. For every person that can go that high, there is some poor guy that can't. That's how averages works. Statistical normal bell curves apply even to overclocking CPUs.

While I whole heartedly agree those 1000E cD0's are great CPUs and I congratulate you on your awesome success, it's still wrong to tell people that they can expect those results. The potential is there, yes, but it's not a sure thing. Not all newbies or even experienced overclockers use proper cooling or use high quality components. Maybe they should, but they don't.

As far as dual procs not overclocking as well as single procs, Tim (who is an expert in dual procs) has repeatedly said the same thing, just do a search of his past posts. You can make "X" speed with a dual proc system, then if you overclock each one of those CPUs individually, generally one of those chips will go higher than the other.

Additionally, I know of a guy that super cooled a 700 to 1.2+ gig and several that made 1.1+ gig. Does that mean they all do that? Nope. Reality check here. I've gotten lots of private emails from people whining because they heard here on the forum they should be able to overclock their computer to a certain level but can't. They wonder why "everyone" else can make it to that level but they can't.

Even the overclockers.com database is skewed. I cringe everytime someone reports an "average" from the database. It's not really a real world average. The majority of people putting data into there are reporting good success. There is no way to verify that these entries are stable or even correct. Lots of people that failed in their overclocking attempts never enter their data in. At best, it's just an idea what the potential is for that CPU and system. Your mileage will undoubtedly vary.

Enough of this debate in someone elses topic. If you wish to continue this debate, then let's start another topic in the proper forum. I admitted I was wrong about my initial statement where I mistakenly said there were no cD0 1000E CPUs. However, I stand by everything else I said.
 
batboy said:
Jeez \(^0^)/, just because you can quote off a dozen people that hit 1.4 gig or close to that mark with a 1000E cD0 does NOT mean they all will or that the majority of them will. That's hardly scientific evidence. For every person that can go that high, there is some poor guy that can't. That's how averages works. Statistical normal bell curves apply even to overclocking CPUs.

What?.. I did not say "all will or that the majority of them will". You made that up. called creation. What I said was "majority", not "all". all means 100%. Majority means not. Don't twist with words, no add ups or minus stuff. =(

While I whole heartedly agree those 1000E cD0's are great CPUs and I congratulate you on your awesome success, it's still wrong to tell people that they can expect those results. The potential is there, yes, but it's not a sure thing. Not all newbies or even experienced overclockers use proper cooling or use high quality components. Maybe they should, but they don't.


Geez.. you sure do look at worst case scenarios and hope for worst and give negative, discouragement to newbies. What is wrong with giving them hope while offer them advice about proper cooling if they're not doing it? I been reading your posts and seem freakin negative.

As far as dual procs not overclocking as well as single procs, Tim (who is an expert in dual procs) has repeatedly said the same thing, just do a search of his past posts. You can make "X" speed with a dual proc system, then if you overclock each one of those CPUs individually, generally one of those chips will go higher than the other.


Doh... Let me mention it again. I know that. But, you cannot say "will never". That means impossible. I don't even have to explain this cauze we know it!

Additionally, I know of a guy that super cooled a 700 to 1.2+ gig and several that made 1.1+ gig. Does that mean they all do that? Nope. Reality check here. I've gotten lots of private emails from people whining because they heard here on the forum they should be able to overclock their computer to a certain level but can't. They wonder why "everyone" else can make it to that level but they can't.

Why are you mentioning this again? I said about this before in my last post here. We both are thinking the same fact here, so not a problem.

Even the overclockers.com database is skewed. I cringe everytime someone reports an "average" from the database. It's not really a real world average. The majority of people putting data into there are reporting good success. There is no way to verify that these entries are stable or even correct. Lots of people that failed in their overclocking attempts never enter their data in. At best, it's just an idea what the potential is for that CPU and system. Your mileage will undoubtedly vary.

I agree there.

Enough of this debate in someone elses topic. If you wish to continue this debate, then let's start another topic in the proper forum. I admitted I was wrong about my initial statement where I mistakenly said there were no cD0 1000E CPUs. However, I stand by everything else I said.

You can stand by what you want. I'm not stopping you. I just had to point out something isn't true. This is not to just correct you or anything. It's to let other readers in this forum to hear information. I didn't start this to fight you.
 
\(^0^)/ said:
Geez.. you sure do look at worst case scenarios and hope for worst and give negative, discouragement to newbies. What is wrong with giving them hope while offer them advice about proper cooling if they're not doing it? I been reading your posts and seem freakin negative.

I think batboy is being precautionary, that's all. He's not trying to denounce or ridicule anyone or thing, he's just asking that we keep the expectations to a minimum.

If I expect to hit 1.60mhz with my chip, but don't, I'd be disappointed, and might even take bigger risks with voltages/etc to try and achieve what another member was able to accomplish with the same speed chip and setup. In the process of pushing the limits I fry my chip. I get sore at the forums because they "told me" (and I've seen these posts) that I could go this high and they broke my chip!! Batboy is trying to prevent these incidents from occurring. If we say the realistic overclock for any P3 cD0 is 1.20GHZ, but some have made it as high as 1.4, fine. But to use 1.4 as a possibility for most is like promising all children they will be President or superstars.
 
Back