Update on P4 extreme...

Pirox · Sep 16, 2003

<A HREF="http://www.theinquirer.net/?article=11581" target="_new">Doesn't exist....</A>

---
If you go to work and your name is on the door, you're rich. If your name is on your desk, you're middle class. If your name is on your shirt, you're poor!

Pirox · Sep 17, 2003

Anandtech.com is showing other things concerning this p4 XE..it could be that it does exist??

---
If you go to work and your name is on the door, you're rich. If your name is on your desk, you're middle class. If your name is on your shirt, you're poor!

ad_rach · Sep 17, 2003

From that first article on The Inquirer I would think that it does indeed exist and that Louis Burns may well just be bluffing in his denial of its existence to prevent anyone getting a scoop before he is ready to officially announce it.Having said all that though, it is The Inquirer after all so a hefty pinch of salt may be necessary.All the same, I have never seen them be so wildly innacurate as to fabricate a product like that.

no matter how hard you try, you can't polish a turd. :]

lxccl · Sep 17, 2003

is this article before the announcement because at the end it says, "Burns is due on stage in a couple of hours. Will he talk about an Intel Extreme Edition Pentium 4?"

ad_rach · Sep 17, 2003

<A HREF="http://www.anandtech.com/IT/showdoc.html?i=1870" target="_new">It definitely exists then</A>.

no matter how hard you try, you can't polish a turd. :]

phsstpok · Sep 17, 2003

Well what do you know? Gamers are finally acknowledged as important customers.

<A HREF="http://www.reuters.com/newsArticle.jhtml?type=technologyNews&storyID=3456448" target="_new">Reuters' story</A>

Can the experts here tell us how much performance a 2 MB cache will bring to the table?

56K, slow and steady does not win the race on internet!

jmecor · Sep 17, 2003

i wonder how much that thing would costs?
It could be using a large wafer.

If your nose RUNS, and feet SMELL.
Then you must be born UPSIDE DOWN.

jmecor · Sep 17, 2003

Well what do you know? Gamers are finally acknowledged as important customers.

We gamers are also concern to the pricing.
740 dollars is quite expensive.

If your nose RUNS, and feet SMELL.
Then you must be born UPSIDE DOWN.

ad_rach · Sep 17, 2003

Yeah.I would have thought that the extra cache and the high price would make it more appealing for use in uniprocessor workstations rather than gaming systems.With as much cache as the new 3.06 Xeon and a higher performance 800MHz fsb, this extreme edition would probably divert a lot of people away from buying dual xeon workstations, since it could offer similar performance at a much lower premium.

no matter how hard you try, you can't polish a turd. :]

slvr_phoenix · Sep 17, 2003

I think that the most definitive proof that it exists is from <A HREF="http://www.intel.com/pressroom/archive/releases/20030916corp_b.htm?iid=Homepage+Highlight_030916b&" target="_new">Intel themselves</A>. After all, it's their product so they should know best.

<pre><A HREF="http://ars.userfriendly.org/cartoons/?id=20030905" target="_new">People don't understand how hard being a dark god can be. - Hastur</A></pre>

HardWareBoss · Sep 17, 2003

The best thing about it? There where no rumors that this was coming down the road.

slvr_phoenix · Sep 17, 2003

I have to wonder though if it even makes sense. I mean the Xeons have pretty much proven that an added L3 cache does almost nothing for the P4 core unless you have problems with your memory bandwidth or have some hefty server usage wanting memory hits from all over the place. So I don't get how this will even benefit the P4 for it's supposed target audience. Gamers will have systems with good bandwidth and games are intentionally written to cache well.

And that could be why it hadn't been leaked before. The product probably was never even intended to get released. I have the feeling htat it's just an old proof-of-concept engineering sample that was discarded but later brought out and dusted off just so that Intel could make people pause to think/worry.

I mean once Scotty comes out, who will even care about this? It's all too nonsensical to feel serious. So I personally see it as just a practical joke from Intel.

<pre><A HREF="http://ars.userfriendly.org/cartoons/?id=20030905" target="_new">People don't understand how hard being a dark god can be. - Hastur</A></pre>

HardWareBoss · Sep 17, 2003

Heck Scotty goes to 1MB L2 Plus with Tejas due next year with 1066 Fsb an 65NM.

ritesh_laud · Sep 18, 2003

I mean the Xeons have pretty much proven that an added L3 cache does almost nothing for the P4 core unless you have problems with your memory bandwidth or have some hefty server usage wanting memory hits from all over the place.

Well I don't think that's an accurate assessment (though I know it's what THG said). Let's take a look at the single-processor benchies at Tom's and see what improvement the 1 MB L3 cache makes on the Xeon 3.06/533. Tom's conclusion was that the L3 cache is practically worthless for workstation apps and games, but let's analyze the percentages. I'm only looking at *single-processor non-synthetic* benchmarks, where the single-CPU P4 3.2 beat the dual Xeons so obviously the second processor was not being used:

SPEC view Perf 7.1: no improvement
Pinnacle Studio: 15.4% faster
SETI@Home: 24% faster
Quake 3 640x480 demo001: 3.5% faster
Quake 3 1024x768 demo001: 2.7% faster
Quake 3 640x480 nv15demo: 5.7% faster
Comanche 4: 4.1% slower (something wrong here, maybe THG flipped them?)
UT 2003: 11.7% faster
3D Mark 2001 SE: 7.5% faster
Splinter Cell: 15.8% faster
Serious Sam: 7.8% faster
3D Mark 2003: 1.0% faster
MP3 Maker: ~1% faster
Xmpeg+DivX: 5.2% faster
WinRAR: 7.6% faster
Sysmark 2002: no improvement

So we see that for many apps there actually is a reasonably healthy difference, between 5-10% generally for the 1 MB L3 cache. Now remember that the P4 EE has a *2 MB* L3 cache, not 1 MB like the Xeon tested above. However, since the P4's FSB is 800 MHz it doesn't benefit as much as the Xeon with the 533 FSB. Overall, though, the P4 EE should be noticeably improved for gaming. I think many games will show around 10% more frames.

slvr_phoenix · Sep 18, 2003

Maybe you're right. We'll have to wait and see. I'm just thinking that the better memory performance of the P4 will clip things down to where it's more or less useless. A lot of it comes down to price though. Just how affordable will Intel make that L3 cache? If a higher GHz with the same or more performance costs less than the L3 cache version then would it really be worth the money? :\ Or is Intel dodging that bullet by only offering it for NWC@3.2GHz?

It might be interesting though if it really ends up making a noticable difference and catches on to new products. If it does help then imagine a Scotty with that much L3 or more...

<pre><A HREF="http://ars.userfriendly.org/cartoons/?id=20030905" target="_new">People don't understand how hard being a dark god can be. - Hastur</A></pre>

HardWareBoss · Sep 19, 2003

SE $740 FX $740 or more. Scotty with L3 2MB or more? Not that's interesting. What about Tejas?

HardWareBoss · Sep 19, 2003

So what's the upgrade from this new chip from Intel? Do they boost Scotty and Tejas with these new L3 for gamers.

lhgpoobaa · Sep 19, 2003

yeah...
i could actually see the 2Mb cache working better if the system was running on a slower bus, or the ram was somehow bottlenecked.

I am not a AMD fanboy.
I am not a Via fanboy.
I am not a ATI fanboy.
I AM a performance fanboy.
And a low price fanboy.

Regards,
Mr no integrity coward.

Update on P4 extreme...

Distinguished

Distinguished

Distinguished

Distinguished

Distinguished

Splendid

Distinguished

Distinguished

Distinguished

Splendid

Splendid

Splendid

Splendid

Distinguished

Splendid

Splendid

Splendid

Illustrious

Share this page