AMD going full out and PIII retiring

[Joe Auman]

AMD is already intrducing the Duron and the Thunderbird. The Thunderbird will have 256K of full speed cache, but the new AMD Mustang will have 2Mb of full speed cache. AMD will also be introducing the Spitfire, set to run against the Celeron. In conclusion, AMD will also introduce the Sledghammer, which will be able to handle Tb of memory because of 32 bit compatiblilty.

The PIV (code named "Wimamette") will aim for the mid range of computers set with a 400Mhz bus and speeds of greater than 1.3Ghz. It will also be the first 32 bit cpu since the Pentium Pro. Another additon will be included to Intel. Named the Itanium, it will be the first Intel 64 bit processor. The Intanium will be for servers.

------------------
FATAL ERROR!!

Non-System brain or brain error!!
Replace and strike any key when ready.

[Joe Auman]

I forgot to mention the AMD Corvette. It will (I think) run alond side the Duron. No word yet which chips will take what part of the market.

------------------
FATAL ERROR!!

Non-System brain or brain error!!
Replace and strike any key when ready.

[This message has been edited by Joe Auman (edited 06-15-2000).]

[Danny G]

Sounds good to me - one question though. When will the celerons get that 100 Mhz bus speed or will we have to continue to overclock them ???

------------------
Never trust a man who when left alone in a room with a tea cosy doesn't try it on.

[Joe Auman]

Good question. Maybe the Celeron II has it, but I doubt it. Celeron III hopefully will.

------------------
FATAL ERROR!!

Non-System brain or brain error!!
Replace and strike any key when ready.

[HAL9000]

The release of the 633Mhz and the 667Mhz Celeron is supposed to be tomorrow and they will still be on a 66Mhz bus. The 700Mhz should be out within the next 4 to 8 weeks and it rumored both ways that it will still be at 66Mhz, but other sources are saying 100Mhz. Toaster...what do your Intel buddies say?

------------------
If it ain't broke, you're not pushing hard enough!

[Tiretool]

So as far as overclocking goes... is the 566 the best bang for the buck? Or will they all pretty much go 200-300 over what they're labeled?

------------------

************************************
tiretool@pcmech.com

[HAL9000]

I read a review somewhere on one of the pre-release 633's that was clocked to just over 1000Mhz. So depending on RAM and motherboard combinations, all the Celerons should be pretty good overclockers.

------------------
If it ain't broke, you're not pushing hard enough!

[Joe Auman]

I have actaully heard that a 533 was overclocked to just under 1Ghz. I would expect the same, if not more, for the others.

------------------
FATAL ERROR!!

Non-System brain or brain error!!
Replace and strike any key when ready.

[HAL9000]

Well, I have 800Mhz out of one of my 533's right now. Running a 100Mhz FSB and a standard HS/Fan. Been running almost a week now with no problems and a temp of 43C steady. Have to try the 133Mhz FSB next to see if I can get that sucker to 1066Mhz.

------------------
If it ain't broke, you're not pushing hard enough!

[Joe Auman]

Doubling the intended speed!?!?! That's messed up! I'm surprised Intel hasn't tried to restrict that yet. If you could do that, let me know. I might see what I can do on this 500.

Didn't notice, I'm a grad! Where's my diploma?

------------------
FATAL ERROR!!

Non-System brain or brain error!!
Replace and strike any key when ready.

[This message has been edited by Joe Auman (edited 06-19-2000).]

[Linus]

Hi,

I read with amazement about all your exploits overclocking your celerons. Could you please help me with mine.

mobo : asus p3v4X ( apparently no problems with this bozo, my friend managed to overclock his PIII on this mobo without problems just to test.)
RAM : 2 slabs of 64MB m-tech RAM (brand new)
chipset: celeron 500. 1.5 weeks old.
clock freq : 68+
heatsink + fan standard issue with box
(added thermal compound though)
temp -> hardware monitor pro : approx 42-45
BIOS hardware monitor: approx 49-53

Heat doesnt seem to be the problem. But at any other bus speed, the PC gets unstable and stalls after a while or there will be the infamous windows error message. Pls give me ideas to try and see what exactly is my celeron's problem. Thanks. Am I that unlucky to get a less than perfect celery? oh man.

[Joe Auman]

Did you change the bus speed? This is a problem. Try just changing the multiplier before you mess with the bus speed. Also, I have heard that the Celeron 500 is not the most overclocking friendly of the bunch.

------------------
FATAL ERROR!!

Non-System brain or brain error!!
Replace and strike any key when ready.

[Danny G]

Unfortuantly guys as a fellow owner I can tell you that without some radical cooling most (notice the word most) cel 500s wont go past 7.5*75 - ie 562.5 Mhz.

------------------
Never trust a man who when left alone in a room with a tea cosy doesn't try it on.

[Charliey]

The AMD Duron (aka SpitFire) really kicks the celerons but:AnandTech Duron Review

------------------
-Charlie

[glc]

You might have a PCI device that is very sensitive to overclocking.

[HAL9000]

I don't think it's fair to say that the Duron is kicking the Celeron's butt. Looking at the review, comparing the Duron 700 to a Celeron 566. I would hope it would beat it. It would be a sad day for AMD if it didn't!. But let's flip the coin here. Lets overclock that 566 to 850 on the good old trusty BX chipset and what happens. The Celeron beats the Duron in all of the benchmarks listed except one. Sure, it's comparing apples to oranges, but isn't comparing a 700 to a 566 a bad comparison as well?

------------------
If it ain't broke, you're not pushing hard enough!

[Mattman]

quote:

---

Originally posted by HAL90001:
I don't think it's fair to say that the Duron is kicking the Celeron's butt. Looking at the review, comparing the Duron 700 to a Celeron 566. I would hope it would beat it. It would be a sad day for AMD if it didn't!. But let's flip the coin here. Lets overclock that 566 to 850 on the good old trusty BX chipset and what happens. The Celeron beats the Duron in all of the benchmarks listed except one. Sure, it's comparing apples to oranges, but isn't comparing a 700 to a 566 a bad comparison as well?

---

HAL, you dirty dog! That's the same post I just saw in an Assembly/Overclocking string here: http://www.pcmech.com/ubb/Forum2/HTML/001429.html .

The Duron 700 is a bad comparison to the Celeron 566. But, thinking about the almighty dollar, comparing a Duron 700 to a Celeron 700 favors the Celeron, because they cost 50 dollars more ($155 to $205). When I look at the benchmarks for Duron 700 vs. Celeron 700, the Duron 700 performs 25% better!! So, the Duron 700 is 25% faster and 25% cheaper than the Celeron 700. I know that the Celeron 700 is not overclocked yet, but the Duron 700 isn't either. No one has seen how overclockable this new guy is. Duron might even overclock better than the Celeron, since they aren't multiplier locked.

Two can play at this game .

------------------
Life without computers would be like the world without war.

[This message has been edited by Mattman (edited 06-22-2000).]

[HAL9000]

Like I said in the Assembly/Overclocking forum, there is NO 700Mhz Celeron yet. The prices you are seeing are pre-released prices. I get those at work all the time, but there is no 700 CPU yet.

------------------
If it ain't broke, you're not pushing hard enough!

[Floppyman]

Just thought I throw this in....I read the new Celerons are actually pIII's with half L2 cache...how innovative is that from Intel?

[HAL9000]

From what I have read, I think Flopster is right. Now if we Celeron junkies could just figure out how to get to that cache.

------------------
If it ain't broke, you're not pushing hard enough!

[figarowa]

just to throw some fuel into the fire, supposedly the celerymines are pee threes that have bad cache chips that couldn't hack it as a P3, so intel just disabled half of it, marks em as a celery,
but i have no proof to back that up..

[HAL9000]

Not likely, if you flip the chip over to view the bottom side, you will notice some differences between the PIII and the Celeron. I do believe the PIII has 8 (I think they are resistors, not sure) soldered to the backside of the chip) while the Celeron has (damn..can't remember...I think 4). These "resistors" were rumored to have something to do with the cache connection. An early review on first sample release Celeron II's actually had the mount points visible on the chip where the remaining pieces would get soldered on, but the prodution version, these mount points are not visible, so I would assume that there is a production variance somewhere.

------------------
If it ain't broke, you're not pushing hard enough!

[padawan]

Intel has done more than disable half the L2 cache in the CeleronII. The CeleronII has a different L2 cache association scheme, that induces more latency into the system, not a huge amount, but enough that the performance is no longer as good as the big brother (as was the case with the PII and the old Celerons). There is a lot of benchmarks out there that show in pure FPU, the CeleronII is good, but as soon as the L2 cache comes into play, it tanks. Go over to Tom's Hardware and look what a Duron did to a Celery running at 1Ghz...

As far as a 100Mhz Celeron, I read a few weeks ago that the Celeron will be dropped next year, in favor of a new all-in-one type of chip to really cater to the low end segment of the market. Anyone remember the Cyrix MII?

As far as the 566, I have one, and it will not run at 850. I'm pretty bummed. I am using a massive GlobalWin FDP32 (modified to fit the flip-chip design), and a copper shim plate, to cool the chip. Temp stays in the mid 30's, but no matter how high I run the voltage, she just won't stay stable in games. All my cards are good, because a 100Mhz bus keeps everything in spec, and I'm using Siemens PC-100, pretty good stuff. 638Mhz is ok, as is 740, so that's where is stays. Shoulda waited and got a Duron. Asus has a mobo that unlocks the multiplier on the TBird and the Duron...how sweet is that, no more Goldfingers!

[Floppyman]

Hey padawan,

Guess this just proves that not every cpu Intel or Amd is the same....some will overclock real good some won't. I never buy a cpu just because I want to overclock it, and the chances are high it will overclock, because there is a chance it won't. I buy for performance first, and then I'd try to overclock. I'm still debating whether to oc my Athlon 550 week28 .25 micron...I saw a good goldfinger for $33 and I wanted a new heatsink anyway so I might give it a shot, but I know that I might be dissapointed as it might not overclock at all. oh well, just my $.02