Sign in with
Sign up | Sign in
Your question

THE TUALATIN CELERON WHO MURDERED THE P4 - PART II

Last response: in CPUs
Share
August 6, 2002 4:14:58 PM

My Watercooling gear is arriving in two days.

We'll soon be seeing how high my 1400MHz Tualatin Celeron will go.

Any last guesses?

1722 MHz
1736 MHz
1750 MHz
1764 MHz
1778 MHz
1792 MHz
1806 MHz
1820 MHz
1834 MHz
1848 MHz
1862 MHz
> 1862 MHz

Stay tuned ....

<i>/Copenhagen - Clockspeed will make the difference... in the end</i> :cool:
August 6, 2002 4:19:15 PM

1200MHz :tongue: ...

I'll guess >1834 I've seen some PIII 1.4GHz Tuallies get in excess 1600MHz pretty easy. I'm thinking the smaller cache will allow it to go higher and with watercooling.... 1834MHz!!

<i>Past mistakes may make you look stupid, but avoiding future ones will make you look smart!</i>
August 6, 2002 6:39:27 PM

<blockquote><font size=1>Svar på:</font><hr><p>I'll guess >1834 I've seen some PIII 1.4GHz Tuallies get in excess 1600MHz pretty easy. I'm thinking the smaller cache will allow it to go higher and with watercooling.... 1834MHz!!<p><hr></blockquote><p>Well, I'd be grateful for anything above 1800MHz, but of cause I'm hoping for a straight 133MHz FSB, it really needs that memory bandwith to do well. Also the AGP/PCI busses will be running within spec. At FSB's around 120 MHz, my graphics card is suffering causing the screenimage to flash once in a while, really annoying.

<i>/Copenhagen - Clockspeed will make the difference... in the end</i> :cool:
Related resources
Can't find your answer ? Ask !
August 6, 2002 7:59:25 PM

i`ll have to say....1400

"Is Celeron good?"
"No. Celeron is bad."
LOL
August 6, 2002 8:10:29 PM

well acording to <A HREF="http://www.overclockers.com" target="_new">Overclockers.com</A> the only listed 1400MHz Tulyl Celery there runs fastest at 1708MHz at 1.65 vcore........

FPO/Batch #: Q219A222
Speed: 1708MHz
Volts: 1.65v
S Code: SL68G
Plant: Malaysia
Week: 20
Motherboard: ASUS TUSL2-C

so i dunno how ur system differs.....but i suspect the tulyl core maxes otu at around 1700MHz somewhere close to there anyways.......so u MIGHT hit 1722MHz like u want...but i still say u wont....i think ur max speed ull get is 1623MHz.....

<A HREF="http://www.anandtech.com/mysystemrig.html?id=13597" target="_new">-MeTaL RoCkEr</A>
a b à CPUs
August 7, 2002 12:10:03 AM

ER, 1600 maybe if you're lucky.

<font color=blue>By now you're probably wishing you had asked more questions first!</font color=blue>
August 7, 2002 1:03:08 AM

Damnn... can I lower my guess?

<i>Past mistakes may make you look stupid, but avoiding future ones will make you look smart!</i>
a b à CPUs
August 7, 2002 1:10:06 AM

My 1200 wouldn't go over 1500.

<font color=blue>By now you're probably wishing you had asked more questions first!</font color=blue>
August 7, 2002 1:51:24 AM

Any benchmarks?
August 7, 2002 1:51:40 AM

Ummm, 1580 if just water cooled. 1660 if water & peltier cooled.

<b>I have so many cookies I now have a FAT problem!</b>
a b à CPUs
August 7, 2002 2:40:34 AM

Let me look them up if I can. BTW, my PIII 1000@1000 (I sold the 700@933 system to a friend) beat my Celeron Tualatin 1200@1480 by at least 5% in 3D-Mark 2001

<font color=blue>By now you're probably wishing you had asked more questions first!</font color=blue>
a b à CPUs
August 7, 2002 2:45:10 AM

Nope, I guess I must not have submitted them.

<font color=blue>By now you're probably wishing you had asked more questions first!</font color=blue>
a b à CPUs
August 7, 2002 9:18:28 AM

Yep, more cooling couldn't help mine! I had that sucker iced and it still wouldn't clock higher. Raised the core voltage all the way up to 1.85v, it STILL wouldn't clock higher!

<font color=blue>By now you're probably wishing you had asked more questions first!</font color=blue>
August 7, 2002 11:54:37 AM

ahhh the golden limit
enjoy

<b>Before visiting THG i was a clueless noob. Now im still clueless, but look at my nice title!<b>
August 7, 2002 2:17:00 PM

FSB bandwidth...so important since these days, the CPUs do fast, while the FSB is still a pinhole.

"Is Celeron good?"
"No. Celeron is bad."
LOL
August 7, 2002 4:10:05 PM

<blockquote><font size=1>Svar på:</font><hr><p>MeTaLrOcKer wrote: "i doubt ull even hit 1722......."<p><hr></blockquote><p><blockquote><font size=1>Svar på:</font><hr><p>tbirdXPplus wrote: "i`ll have to say....1400"<p><hr></blockquote><p><blockquote><font size=1>Svar på:</font><hr><p>Crashman wrote: "ER, 1600 maybe if you're lucky."<p><hr></blockquote><p><blockquote><font size=1>Svar på:</font><hr><p>Chuck232 wrote: "Damnn... can I lower my guess? (1834MHz)"<p><hr></blockquote><p>

<blockquote><font size=1>Svar på:</font><hr><p>phsstpok wrote: "Ummm, 1580 if just water cooled. 1660 if water & peltier cooled."<p><hr></blockquote><p><blockquote><font size=1>Svar på:</font><hr><p>MeTaLrOcKer wrote: "think he'll hit 1623....pretty lucky if he can get ANY higher....."<p><hr></blockquote><p>
Come on guys, have some faith in my little Tually, will you ?

Guess what, while typing this post my watercooling gear just arrived !

Unfortunately I'll be busy doing something else this evening, but I hope to have a go later in the afternoon, i.e within the next 6-8 hours, but no later than 24 hours.

If anybody wants me to make some special benchmarks then drop a line with your suggestion.

Also, will anybody be able to host my pictures of the benchmarks ?

The bloodbath is about to begin, time to bury some P4's, some PIII EB's and maybe some Athlon XP's...



<i>/Copenhagen - Clockspeed will make the difference... in the end</i> :cool:
August 7, 2002 4:19:40 PM

I'm guessing it'll go as high as you can throw it. :) 
How fast it'll run though ... who knows? Maybe you should put your PC in a tub of goo while you're at it.

<A HREF="http://www.nuklearpower.com/comic/171.htm" target="_new">The corpse you find may be your own.</A> - Black Mage
August 7, 2002 7:41:28 PM

U might take soem PIII EB's and ull take all celerons and most likely Duron's....ull take maybe the XP 1500+ and 1600+ i say no more...and ull prolyl take all Williamette P4's....besides that...natta......my opinion though.......keeping my options open still but thats what it hink will happen.....i wanna see Sandra and 3D Mark 2000 and 2001 SE build 330. =)

And ill host ur pics for u......ICQ me at 13418693........

<A HREF="http://www.anandtech.com/mysystemrig.html?id=13597" target="_new">-MeTaL RoCkEr</A>
August 7, 2002 11:37:41 PM

To anyone who feels like submitting their benchmarks of their systems, you can email them to me at imgod2u@cox.net
Please provide your system cooling information, CPU stock speed and overclocked speed and core voltage (and ram speed and make if possible). I'll post a link here of the results I receive.
Preferable more than just a Sandra benchmark. If you can, download the demo of Q3A and run timedemo 01 and 02 (so that everyone will use the same version).
For those of you who have Jedi Knight 2, you can submit benchmarks for timedemo 01 and 02 as well. DX8 games would be prefered too. DroneZ is available for demo and I believe there's a benchmarking test.
Of course, there's the synthetic benchmarks such as PCmark, 3Dmark and Sandra. Those would be nice as well.

<P ID="edit"><FONT SIZE=-1><EM>Edited by imgod2u on 08/07/02 07:59 PM.</EM></FONT></P>
August 8, 2002 10:59:21 AM

Huston, we have a problem. One Celery out of control ... it could leave orbit any time now. STOP

Major part missing in cooling gear. STOP

It's a no go. STOP

We have to await that crucial part, before we dare attempt a rescue, it's too damn hot! STOP

Stay tuned ...

<i>/Copenhagen - Clockspeed will make the difference... in the end</i> :cool:
August 8, 2002 1:45:29 PM

Did you forget to put the water in your watercooler again?

;) 

Just kidding. :-p

Seriously though, fill us in. What's missing?

<A HREF="http://www.nuklearpower.com/comic/171.htm" target="_new">The corpse you find may be your own.</A> - Black Mage
August 8, 2002 6:55:42 PM

<blockquote><font size=1>Svar på:</font><hr><p>Seriously though, fill us in. What's missing?<p><hr></blockquote><p>The most important part of all, the water block for the CPU. Should be here tomorrow. My weekend schedule is tight, but I'll devote the necessary time get the system up and running and do some initial benchmarks, like SiSoft Sandra.

The P4's, PIII EB's and the Athlon's just bought themselves some extra time in the limelight, but it'll be short-lived, rest assured.

Stay tuned ...


<i>/Copenhagen - Clockspeed will make the difference... in the end</i> :cool:
August 8, 2002 8:45:36 PM

Quote:
The most important part of all, the water block for the CPU.

Doh! That is a rather important part. I guess that means that my guess of a radiator was wrong then. Heh heh. Not having a water block does rather hamper things...

Good luck on both finding the time and getting it to play nice with others.

<A HREF="http://www.nuklearpower.com/comic/171.htm" target="_new">The corpse you find may be your own.</A> - Black Mage
August 10, 2002 7:07:38 PM

Water block arrived.
System up and running... Watercooled !!!


Default CPU speed: 1400MHz, 100MHz FSB.
Idle: 40 C
Prime95: 42 C


Water not completely free of micro-bubles, but improving all the time. The radiator is cooled by a Pabst 120mm fan (1600 RPM, 55.3 CFM, 26 dBA). A higher RPM fan or maybe 2x120mm Pabst would probably result in lower temps. But it's OK, this virtually silent fan is still giving me about 10 C lower temps than the stock heatsink/fan combo. The only things that are audible at the moment are the power-supply (300W Chieftec) and the WD800JB. Subjectively, the power-supply is not as noisy as usual during load, possibly due to cooler air entering the power-supply now where CPU is no longer heating up the air. Still, I think I need to lower the RPM on the power-supply fan and get myself a more silent HD or maybe try to silence the one I have.

Enough said, time to push the envelope... wish me luck.

<i>/Copenhagen - Clockspeed will make the difference... in the end</i> :cool:
August 10, 2002 10:56:08 PM

I've e-mail'ed some SiSoft Sandra CPU and multimedia screenshots to imgod2u. So we'll just have to wait until he makes them awailable on his site.

Actually, I managed to crush some opponents.

Got higher CPU Dhrystone ALU scores than Athlon XP2200+ and Pentium 4-B 2.66GHz 512L2.

Got higher Floating Point iSSE score than Athlon XP2200+ and nearly reached Pentium 4-B 2.4GHz 512L2's Floating-Point <b>iSSE2</b> score.

3D Mark2001 SE doesn't really make sense, because my graphics card is a MX440.

PCMark2002 favors memory bandwith, which puts the Tualatin Celeron in the shade.

All I can say is, that subjectively this CPU feels fast at 1822MHz.

<i>/Copenhagen - Clockspeed will make the difference... in the end</i> :cool:
August 11, 2002 12:07:48 AM

Wow.....im definently impressed =)

Good job man....must say i honestly didnt think ud make it that high.......

What r ur temps like at that speed ??? Chipset/MB and CPU Temps......

I just did some INTENSIVE modifications....and im goign tot ake some pictures soon and show everyone......but lets just say there enough air movement in my case the i can feel HARD BLOWING wind hittign me from 3 feet away =) and thats with the case closed....... heh

<A HREF="http://www.anandtech.com/mysystemrig.html?id=13597" target="_new">-MeTaL RoCkEr</A>
August 11, 2002 7:20:23 AM

i'm not suprised the tully celly is the [-peep-]. my 1ghz is at 1.4 with pin mod... i think ram speed is my prob. i'm at 1.8xxx volts. my benches like a 1.33+ athlon.
August 11, 2002 8:41:25 AM

<blockquote><font size=1>Svar på:</font><hr><p>Yep, more cooling couldn't help mine! I had that sucker iced and it still wouldn't clock higher. Raised the core voltage all the way up to 1.85v, it STILL wouldn't clock higher!<p><hr></blockquote><p>Well mine seemed a little more responsive to high-end cooling. About 100 MHz more than stock air-cooling. For the record Prime95's torture test didn't pass @ 1822MHz, but SiSoft Sandra benchmarks did. I guess Prime95 is VERY intensive on some parts of the CPU, whereas real life task seems to be more gentle. I had to lower the frequency to as low as 1764MHz @ 1,70V (126MHz FSB) in order to pass Prime95's torture test, so this is to be considered my absolutely stable overclock result. During this test the CPU temperature was in the range 48-49 C.

<i>/Copenhagen - Clockspeed will make the difference... in the end</i> :cool:
August 11, 2002 12:16:05 PM

Wierd, I estimated you'd hit a 130Mhz FSB originally, same as what you posted for your highest general stability, minus Prime95

I need my Hammer now to bash some newbies
August 11, 2002 3:28:27 PM

Ahh...so now ur computer is just Quiter than it was with Air Cooling...not cooler =)

Well i heard the Tualitan Pentium 3 Overclocks better than the celeron Version.....shoulda checked that out!

<A HREF="http://www.anandtech.com/mysystemrig.html?id=13597" target="_new">-MeTaL RoCkEr</A>
August 11, 2002 6:20:03 PM

<blockquote><font size=1>Svar på:</font><hr><p>Ahh...so now ur computer is just Quiter than it was with Air Cooling...not cooler =)<p><hr></blockquote><p>Actually it is cooler. Taking into account that it's a hot day today and I have no AC:

Watercooled: Prime95 @ 1764MHz @ CPU:50C, MB:32C

At default speed and air-cooled it used to go up to about 52C.

Bare in mind also that my radiator only has one slow and quiet 1600 RPM fan attached. A more powerful and flexible solution is to mount 2 powerful fans on each side of the radiator and have them voltage-controlled from a panel in the front becel. The fans RPM can then be increased or decreased according to demand, getting the best of both worlds.

<blockquote><font size=1>Svar på:</font><hr><p>Well i heard the Tualitan Pentium 3 Overclocks better than the celeron Version.....shoulda checked that out!<p><hr></blockquote><p>Can you provide links ? I've not seen anything higher that what I have right now, water-cooled that is.


<i>/Copenhagen - Clockspeed will make the difference... in the end</i> :cool:
August 11, 2002 9:14:33 PM

This has been bothering me for so long. What is MB temp for???
Why would anyone want to know the temp hovering on the circuits of his MB?
Why not indicate or measure CASE temp instead?
If MB temp or System Temp means CASE temp, then man my case at 37 must be hot...!

--
You are about to witness crazy, mindless, eye-gagging, naive programming, welcome to FOX!
August 11, 2002 10:22:45 PM

MB Temp. from what ive understood is the chipset temperature......there is also AMBIENT/CASE temp. thats usually the 3rd temp...well it is on my comp. anyways =)

Oh, and Copenhagen.....when i see the link again ill post it i cant remember where i put it.....saved the HTML file somewhere on my comp.......

<A HREF="http://www.anandtech.com/mysystemrig.html?id=13597" target="_new">-MeTaL RoCkEr</A>
August 11, 2002 11:37:01 PM

Quote:
Well mine seemed a little more responsive to high-end cooling. About 100 MHz more than stock air-cooling. For the record Prime95's torture test didn't pass @ 1822MHz, but SiSoft Sandra benchmarks did. I guess Prime95 is VERY intensive on some parts of the CPU, whereas real life task seems to be more gentle. I had to lower the frequency to as low as 1764MHz @ 1,70V (126MHz FSB) in order to pass Prime95's torture test, so this is to be considered my absolutely stable overclock result. During this test the CPU temperature was in the range 48-49 C.


Copenhagen, I wouldnt use p95 as the ultimate for stability, many times I have had a system which runs p95 100% for hours but crashes on certain games(although it may have been my sick videocard overclock heh).

But it is a good indicator of stability ESPECIALLY in the memory!

:wink: The Cash Left In My Pocket,The BEST Benchmark :wink:
August 11, 2002 11:38:13 PM

eden if you overclock your fsb you will cause the northbridge to get hot(which is the mobo temp iirc) this allows you to determine if lockups are cpu or northbridge related.

:wink: The Cash Left In My Pocket,The BEST Benchmark :wink:
!