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."

Re:Celeron 2.6GHz

[Orgazmus]

The cellies lost their value when P3 became outdated.
The Celerons with coppermine cores were kinda fun ;)

Re:Market Statistics

[Anonymous Coward]

Its all the P.O.S. Hp's and compaqs that keep celerons going. I see celerons all the time and its usually the PC's that make me want to rip my hair out. XP is just not made to runs on a celeron with 128 megs of ram.

Re:Market Statistics

[Zorilla]

It seems the usual OEM tactic is to put together a really fast-looking PC and then put a Pic N' Save version of a current generation video card in - one that will be outperformed by a previous generation card that costs less and has more features. A lot of the Dells I've seen recently had GeForce FX 5200s in them, for example.

Submitter is Intel fanboy?

[Vellmont]

This Celeron may be able to compete with AMD's offerings based on more then name brand alone

Ummm.. what? The fastest $117 2.8ghz celeron got the shit kicked out of it by a lowly $55 Athlon 2400XP. Who in their right mind would buy one of these chips? I guess if you really want SSE3 or the only game you play is Quake3 it's a good deal, but otherwise there's no point.

Well, think about who buys them...

[sirGullible]

Not everyone is a geek, and i know plenty of non-geek regular people buy Celerons. From any, say...Best Buy ad, you can see cheap celeron based pc's aimed at families buying their (possibly first) computers. All they need is to browse the internet, listen to some mp3's, instant message, and thats about all. Celerons can accomplish that. They don't really care about overclocking or playing doom 3 or benchmarks or much of that. Also, dell likes to use celeron processors for its lower end systems, so i'm sure dell contributes alot to that.

Nice

[marnargulus]

A speed boost is always nice, but is it really necessary? I think faster RAM would be a better advance, and faster bus speeds for harddrives as well. While the processor might be able to handle more data, we still are having trouble getting data there in the first place. Bring on the 2 gig on-die cache where I run all of my current apps and OS straight on the proc. That is what I'm looking forward to.

Re:Submitter is Intel fanboy?

[Anonymous Coward]

>Who in their right mind would buy one of these chips?

people who don't visit slashdot? people who's never heard of AMD? and believe me, there are many of them out there.

i'd bet that you yourself own many, many things of which there are cheaper and better alternatives than what you have - and you bought what you bought because of lack of research, reliance on brand names, indifference, etc. the same can happen with the general public when it comes to computer chips.

Re:Duron's success

[Neil Watson]

It is also important to note that the average user would not notice any difference between the performance of a budget vs a premium CPU. How much speed does one need to send an email to mom?

If you are buying 300 PCs for an office and can save $20 each buy buying a Celeron or Duron that makes you look good.

Re:Celeron 2.6GHz

[Anonymous Coward]

Agreed. I sell computers for a living and I seem to spend all my time explaining to people why an Athlon 2.0GHz (2400+) out performs the Celeron 2.6GHz.

Intel's insanely high clock frequencies with comparably lower performance are slowly driving me mad from people with questions about the competing Athlon models.

Perhaps I should just raise my prices, use shitty mainboards, less RAM, less HDD space, shared onboard graphics and install 3.2GHz Pentium 4's in all my computers. The scary thing is they'll probably sell better. :-/

Re:Celeron 2.6GHz

[afidel]

Not only that but the Prescott P4 2.4Ghz is only $10 more than the Celeron 2.8, and the P4 part will wipe the floor with the Celeron even giving up 400Mhz. The worst thing you can do to a P4 core is make it stall waiting for reads, and quartering the cache is guarenteed to do that, so why anyone would consider the Celeron for anything other than a web browsing box I can't fathom (and even then you would have to be stupid to use the fastest part).

Re:I don't understand...

[Cat_Byte]

With less than 256M of RAM, XP *will* run that slow. I don't know how much RAM they had, just a thought to keep in mind.

NX command in the Celeron?

[BLKMGK]

AMD is putting the NX processor command into it's low end CPUs, I didn't see any mention of this in the article. Does anyone know if Intel is following suit with it's low end CPUs? Anyone tested the effectiveness of the NX command on an AMD CPU with Linux or the beta SP for XP? IMO if it's as good at stopping overflows as claimed this could provide a competitive edge to the company that has it if the other doesn't....

Comment removed

[account_deleted]

Comment removed based on user account deletion

Re:Submitter is Intel fanboy?

[teg]

Who in their right mind would buy one of these chips?

End users buying the CPU itself (a very minor part of the market)? Not at lot. As part of a system? Quite a few more

One reason is that Dell, the #1 PC manufacturer only ships Intel. And their systems are usually priced pretty competively, at least if you want to use quality components. For companies and non-techies, reliability, support and other parts of the "total" package adds up to be far more important than a few percent performance they wouldn't even notice.

Also, I'd take Intel chipsets over Via or SIS anyday. Nvidia can be painful too, they don't even have an open networking driver (although a reverse engineered one exists for at least the NForce 2).

Re:Celeron 2.6GHz

[Fweeky]

Sounds more like a graphics driver problem than anything CPU related. Pathetic, yes, but not (just) because of the CPU.

Re:Celeron 2.6GHz

[sglane81]

When trying to teach people about computers, I think it's best to use analogies from things they have at least a partial understanding of. When it comes to CPUs like the Celerons vs P4s, I use the analogy of a Formula 1 car to a school bus. I make it quite clear both vehicles run at the same speed (top end and acceleration). Considering this, you can move more people in the bus than the F1 car. Everyone I've talked to understands this. When it comes to stuff like ports (as in TCP and UDP), I use the analogy of a house with windows and doors.

Re:Market Statistics

[mattyrobinson69]

whats wrong with the GeForce FX 5200? what would you recommend from nvidia, for the same price range?

Re:What's The Point?

[callipygian-showsyst]

Yes! I use Celerons when I need to put a rack of 1/2U machines up to serve web pages. As long as I can keep the ethernet adapter saturated with a good server like thpptd, there's no need for a faster or more expensive processor.

Re:Market Statistics

[Zorilla]

Although I have no firsthand information about it, the comments I usually saw about it showed that it was outperformed by the Geforce4 Ti 4600. It also seems to be the OEM card of choice. Previous cards I usually saw in OEMs included the Geforce2 MX and RIVA TNT2 M64, which were also budget cards.

It's probably true that the performance of the 5200 wasn't neutered as much as it was the case in earlier generation. I don't really think there's anything "wrong" with the 5200, but I very much disliked the TNT2 M64, especially when I saw it get outperformed by my Voodoo 1 (in games that weren't 3dfx biased, like Unreal).

If I turn out to be right about the performance, perhaps the price for a GeForce4 Ti 4600 is less?

Re:Of course we use Celerons.

[Humorously_Inept]

This is definitely true. While Celerons are processors to avoid, as a rule, you can get a great deal out of a Celeron if you know exactly which one(s) to shop for. I decided to build my passively cooled (well, there's the PSU fan which I dare not remove...) around a Celeron 1.1A simply because it would have been impossible with a Duron. In my case I was willing to sacrifice performance for silence and as it happens, the Tualatin-based Celeron has been a pretty good performer! It's no Duron, but it's also only a 30W chip - no Duron can match that. I love all my AMDs, but for this application it simply had to be a Celeron! The new Celerons look like very compelling budget chips. Assuming they don't suffer from Prescott's thermal power issues, they appear to have a lot to offer in competition against Durons and low-end AXPs. You just need to know exactly what Celeron you're buying and you could end up with a good deal. That said, the budget marketspace will change when AMD releases its Sempron.

Re:Celeron 2.6GHz

[blair1q]

>why an Athlon 2.0GHz (2400+) out performs the Celeron 2.6GHz.

Because the Celeron is a deliberately crippled version of the Pentium designed to run slower than the Athlon to attract the same price point while carrying Intel's goodwill, while the Athlon is the best AMD can market?

What's inside the machine doesn't matter any more. There are so many configurations of pipeline, cache, core, memory i/o, etc. that nobody should give the first thought to the numbers of the chip.

Especially when the rest of the mobo and i/o and MII and video disk system are bottlenecking those theoretical burst-rates.

We should be working towards a benchmark of a whole computer, that gauges how all of those parts add up to "hot" or "value" or everything in between.

Instead we have corporate-empire sycophants on all sides whining at each other about the semantics of the flim rate on the franistan.

Market it as a P4 derivative ?

[billcopc]

Being near-jobless (as in, I work long hours at non-tech jobs because my home town sucks), most of my coworkers are tech illiterate. They know what a Pentium is, at least they know the bigger the number, the better it is. They still don't know the clock speed, but if they say it's a P2, then I can safely assume it's between 233 and 450 mhz, faster than their brains, at any rate.

Now comes Celeron. First of all these people will have a hell of a time remembering that name, because it is gibberish. At least "Pentium" sounds like "Uranium", and they all learned THAT word watching Back To The Future movies. Now consider that Celerons aren't adorned by a model ID (Celeron 2 wasn't an official name), and these people won't tell the diff between a Coppermine 300a and a Prescott 2.8ghz. These are people who paid to get an 8meg ATI Rage Pro installed because they heard "sideways monitor plugs are bad".

So why not just call it a Pentium IV Lite or something cute like that ? Or just make the older P4's cheaper and begone with the whole Celeron debacle.

Last time I checked, AMD Durons had vanished from the market. Now that you can get an Athlon XP for about $80 canadian ($60 USD), they've pretty much trumped the whole point of budget cpus. Now I still can't grok how Intel gets away with charging 2-3x the price for roughly equivalent performance, but it's probably thanks to Compaq, HP and Dell who have their established clientele of rich ignorants, not all of them, but with all the government and fortune 500 contracts they've got their steak well covered.

Re:Submitter is Intel fanboy?

[Mesaeus]

The biggest factor for these kinds of perpetuating myths about AMD instability seems to be errors in mainboard chipsets. The thing is, Intel makes their own chipsets and even a lot of their own mainboards, while AMD has (almost) always left this up to third parties like VIA, Sis etc. Now while these firms do their best, they occasionally blunder and the result will be less than stable mainboards. And ofcourse uninformed people will always blame the cpu when it's in fact the mainboard/chipset that's the culprit.