HOT! (review!) P4 6xx: No good for Intel

[Mephistopheles]

Take a look at <A HREF="http://babelfish.altavista.com/babelfish/trurl_pagecontent?lp=fr_en&trurl=http://www.x86-secret.com/popups/articleswindow.php?id=119" target="_new">x86-secret's</A> review of the 630 processor. They get to the conclusion that it's not at all that good.

It most definitely isn't enough to brag about and is still no match for A64.

Well, at least the 64-bit benchmarks are quite good, so they finally got a working x86-64 mainstream processor out. Just not the best of the ones available.

 

[Vapor]

More interesting is the craptastically slow 3.73EE, IMO. How they never got Gallatin to 3.73GHz is beyond me....

The 6xx and the 3.73 really are f* ups in my opinion, after all, the L2 on the Irwindale is slower than the L3 on the Gallatin! Not only that, but its L2 on Gallatin is even faster than that! How do they expect Irwindale to keep up? Oh yeah, the faster bus has so far proven to be ineffective at doing much of anything.

Now that this cat is out of the bag, it's only time till we see the light of Shitfield over the horizon and finally just get it on the table that this round of Intel's CPUs are just to say they have something new, not for the enthusiast (or even the professional if K8 is as good in dc as it should be [it was designed for dual core from the beginning, after all]).

Maxtor disgraces the six letters that make Matrox.

 

[mozzartusm]

They used the wrong video card to run the test with. The GeForce cards dont allow that MOBO to work around the frequency hurdle nearly as well as the ATI cards. Im not trying to make the argument that this review is sub-par, the CPU's shouldnt be so finicky to start with. It did surprise me to see that they didnt at least throw an ATI into the mix.

My 3.4 775 may not be exactly the same as this line, but the PCIe frequnecy is one of the biggest challenges that Mobo manufactures are faced with when it comes to the 775. ASUS and ABIT are the only 2 that I know of that have been able to run a stable OC much past 10%. I heard about this in an article at Anandtech. I started out with a 6600GT and OC'd the system as far as I could. Now I have an X700 PRO in it and it OC's well beyond what it would when the 6600GT was in it. Let me be clear, the 6600GT is the better card of the 2 IMO, but the X700 is close enough that when you add the much higher OC's that can be obtained with the CPU then it is the only way to go when working with a socket 775 and PCIExpress.

To save us both time, assume I know EVERYTHING :tongue:

 

[P4Man]

I did find the overclocking quite promising. 4.3 GHz aircooled is not too shabby.

= The views stated herein are my personal views, and not necessarily the views of my wife. =

 

[Vapor]

I'm pretty sure they didn't OC anything...

EDIT: they did, but I was talking about non-OCed performance in my first post. Anyway, nvm, they did OC, but it's fairly irrelevant to the point that they still lag behind AMD and don't pull ahead from their own chips by much at all.

Maxtor disgraces the six letters that make Matrox.<P ID="edit"><FONT SIZE=-1><EM>Edited by Vapor on 02/02/05 11:40 AM.</EM></FONT></P>

 

[mozzartusm]

I agree totally. Im not trying to make a case for Intel over AMD, I was throwing that reply in just in case someone that has a 775 would have some info to consider. When I talk to people that are considering buying an Intel, the first thing that I usually tell them is that AMD is the better of the 2 and that building an Intel rig requires alot more planning and consideration. Especially with the heat issue. The info that is out there for someone trying to lean how to OC the 775 is not very easy to find. For a Newbie, it can be frustrating because there is such a limited amount of info on the subject.

To save us both time, assume I know EVERYTHING :tongue:

 

[Schmide]

EMT64 is exactly the same as x86-64 or AMD64 instruction sets, with the exception that the address lines stop at 36 bits instead of 40 bits. Sorry guys no terabyte addressing, only 64GB of memory address for you. The basics of the 64 bit system is: Default pointers are 32bits, Stack is 64bit, several instructions, (at my last count 29) of mostly word and word segment operation, are invalid. Register size is doubled from 8 to 16 for both general and multimedia. (total 32) To use the extended registers you need to add a REX prefix (67h) to operate with or on them. There is a bit more. Some changes in segment register usage were made, but for the most part that’s it.

Dichromatic for your viewing plesure...

 

[Spitfire_x86]

I really liked the "Rejected" seals they used on P4 EE 3.73 GHz at the end of the review

------------
<font color=orange><b><A HREF="http://www.mozilla.org/products/firefox" target="_new">Rediscover the web</A></b></font color=orange>

 

[juin]

i dont even want to look at the benchmark.

i need to change useur name.

 

[Mephistopheles]

Wow, I laughed about that a lot too. I mean, I can almost see someone calmly stamping "Rejected" on P4 6xx and then going over the P4EE and just wildly stamping, as if P4EE were a dangerous bug that needs to be killed ASAP.

x86-secret: "Bad, Bad P4EE!" ***squish***

 

[ytoledano]

Why 4-4-4 latencies? Why not DDR2-711 for the 720? Why be so mean to a processor which has done you no wrong?

This review makes you wonder...

<b>Behold, Mine anger and My fury shall be poured out upon this place upon man and upon beast and upon the trees of the field and upon the fruit of the land and it shall burn and shall not be quenched