New Celeron D Core gets a Speed Boost

qtothemax writes "The new Celeron core was released on the 25th. The processor, using Intel's new model number naming convention, looks to be quite a bit faster than the old core. The new core is based on the 90nm Prescott, which offers respectable performance, compared to the very slow Northwood based Celeron. It features a 256kB L2 cache, and a 533mhz FSB. Looks like Prescott's longer pipeline is more then offset by the better branch prediction and most importantly the doubled cache when it comes to the smaller cached Celeron. This Celeron may be able to compete with AMD's offerings based on more then name brand alone. Reviews and benchmarks are at Anandtech. I couldn't find any other good reviews, as budget chips rarely generate much excitement."

Market Statistics

[artlu]

Does anyone have a good website which outlines just how many low-end processors are sold every year? From my POV, I cannot understand how the low-end processors survive. Granted, they use less power for mobile applications, but I would rather spend an extra $30-$50 on a processor then most other components of the system.

Or is it all just marketing?
Aj

GroupShares Inc. [groupshares.com] - An Interactive Stock Market Community

Re:Core

[Orgazmus]

Looks like its close to the P4, like the old cellies were P2's with less cache, and then the coppermine cellies with P3 cores.

Remember my old cellie 633 running rock stable at 950 mhz :D

Re:Core

[SatanicPuppy]

In every case, with every Celeron I've ever worked with, I've found "regular" (i.e. non-crippled) chips running anything near half of the Celeron's posted speed, to be far far more capable.

I'd rather be running an old PIII coppermine, or tualatin than any Celeron p.o.s. I've never seen any use for them except to snare uneducated consumers.

Re:Celeron 2.6GHz

[FuzzyBad-Mofo]

Last fall I was checking out a notebook with a Celery 2.x GHz chip in it. The damn thing couldn't even play a Divx-encoded movie fullscreen without stuttering like crazy. Pretty pathetic for any chip over 1 GHz. (Hell, even my P3 650 does better)

Duron's success

[Anonymous Coward]

Budget chips CAN create excitement. At least they should. I remember when Duron was a new thing. I bought the 750MHz model and got 1/3 to 1/2 more speed with the same amount of money..

I was really suspicious about the Duron but later on I learned that it was just a rather cool hack at the time. They removed some expensive gate (or something alike) from the cpu and replaced the same function with some very clever engineering.

They gained some speed and lost one of the most expensive parts of the cpu with one strike. Someone else might be able to recall the details better.

Anyways the point is: The fact that it is a budget chip means nothing. Some budget chips can wipe the floor with some more expensive "premium chips" if they fit your application. I am always interested in the budget versions since that's where you see what the basic technology tweaked to maximum can do.

Budget chips are also a huge market since lots of embed stuff and alike (terminals etc) will in time utilize that. Many people also want to read their email and do their banking and do not care wether it takes 3.5 or 3.2 seconds for the page to render.

Re:What's The Point?

[GreatDrok]

My father-in-law is still using a dual 400Mhz Celeron BP6 based system. I set it up as his NFS/YP/SMB server recently as it had been gathering dust for a while. Turns out that it is really snappy (running RH8 with yum updates via fedora legacy). A pair of 400s in a server seems to be quite nice compared with a single 800Mhz processor.

Ah, the BP6, those were the days :-)

Re:Market Statistics

[gl4ss]

with a celeron big names like compaq & etc can build a box that's both cheap and has a 2.8ghz sticker. who cares about what the box can do, it's 2.8ghz baby!

another reason is that they're good enough for office work by a wide margin.. and cheap..

I don't understand...

[The Hobo]

From TFA:

Holding in the middle of the pack is definitely not a disgrace for these budget processors.

I don't understand, a chip that costs less, has more cache, and has been a proven good chip (the Athlons) beat this new processor which is considered budget...
I myself bought a Duron 650 3 years ago, it lasted me that long. When my PSU died, I decided to upgrade to a 2500+, and left my old computer alone. Last Christmas I went home and set up some new Dell PCs my family bought with 2.4 Celerons, and just from watching a fresh install of XP running (which is usually fast) I almost swore that the 2.4 Ghz Celerons were slower than my rebuilt Duron 650 Mhz, and this is without benchmarks.. it probably wasn't 'factual' by a stopwatch's perspective, but it shows just how bad these chips inherently are.

Re:Submitter is Intel fanboy?

[Neil Watson]

There are many vendors and buyers out there that honestly believe that one should only buy Intel as AMD is unstable. I once had a vendor tell me that. I asked them to cite proof of their claim. They could not.

Re:Submitter is Intel fanboy?

[qtothemax]

Disclaimer: I'm the submitter. I AM NOT an intel fanboy. If I was going to buy a budget processor today it would DEFIATELY be an Athlon. I think a new processor core is always somthing that should be discussed on slashdot. Also, Intel is moving in the right direction by not screwing people who buy Dell and nothing else, or don't know better and think AMD is crap. My girlfriend wants to buy a laptop, and when I told her to get an AMD she kinda sneered because it isn't the intel she is used to. Funny thing is that her desktop is an AMD K6-II, but she doesn't even know it. Like it or not the vast majority of people who don't know better have NO IDEA that the celeron sucks comapred to the Athlon, and that the intel chip is more expensive. This is at least a break to everyone out there who would buy a celeron over an Athlon just because of the vision that intel is the "trusted name brand." Think of it like toothpaste or somthing similar. Do you read up on toothpaste before you go to the store and buy it? I seriously doubt it, but I guarantee there is a dentist somewhere who is seriously pissed off about Crest's poor quality. Most people just want a computer that works, and they buy Intel, because that's what they had before, especially in the budget PC market, just like probably >90% of you just buy the same brand of toothpaste you always get. Woulden't it be nice if they improved Crest with really not much reason to do so, since you're going to buy it anyway? So consider this new celeron as less of a screwing of budget PC buyers, who generally have no clue what they are getting. People who actually follow processor preformance can probably scrape together the extra $100 to get an Athlon64. I personally would still definately go with AMD, but I woulden't have to get in a fight with my GF anymore if she insists on getting a Dell with a celeron in it. I would actually almost consider myself an AMD fanboy, but I found this interesting, and see it as intel throwing to bone to the ignorant.

Re:Submitter is Intel fanboy?

[biz0r]

I had a friend of mine that has consistently told me this...or rather he 'warned' me and cautioned against using them in my servers at work. Well thats interesting as I've got SEVERAL AMD machines that have been up for almost 2 years now (running linux, of course). And the only reason any of them ever get shutdown is for hardware failure (I should note I never have had a CPU related issue) or a kernel update that I just can't avoid.

Re:Submitter is Intel fanboy?

[qtothemax]

One more point: I thought it was an interesting oddity that the Prescott architecture actually IMPROVED preforamance by a decent margin in a Celeron, while it caused a slight decline in the P4. It shows how the preformance gain from cache really is logarathmic, more then offsetting the preformance loss of the extra pipeline stages. Intel just made an interesting statement about the P4 extreme edition.

Re:Core

[drinkypoo]

More to the point, the oldest celerons were P2s with no L2 cache and only the usual (pitiful) amount of L1. Later celerons that got (128kB of) L2 cache had the cache running at full speed and so for tight loops they could execute faster than a P2 of the same clock rate, which had half speed cache.

Incidentally AMD had some interesting cache speed stuff going on then as well. Systems had either half speed or third-speed cache memory on them, the 700 MHz being the last unit with half speed and the 733 and up having third-speed. Hence a 700 was usually faster than a 733 under load.