That is insane. Beautiful write up Payton, very well done. Also the more I read, the more gitty I become because I want one, lol.. jk.

Oh thanks but I didn't write it just passed it along. The thread was lacking specifics, and all the gamers were carrying on like two guys fighting over a woman they'd never seen. LOL

"This is half that of the current Pentium Extreme Edition 955, and less than most AMD dual core CPUs."
It better be. Conroe is a mobile proc, and AMD's current dual cores are for desktop! There will be much less difference when the Turion X2 comes out.

Conroe isn't a mobile processor. Conroe is desktop, Merom is mobile, and Woodcrest is server.

-Pikl

Well, he meant that it's mobile in the sense that it's based off of a mobile architecture. By the way, I'm wondering if anyone can shed some light since we're on the topic of processors and dual-cores. I was checking some things out on a website for a computer store that I shop at every now and then, and I saw that they were selling a Socket 478 Core Duo T2300 1.66GHz CPU. Now, the last time I checked.. I thought 478 was completely phased out.. secondly, I doubt it would work with ay 865, or 875 chipsets because apparently that CPU is only compatible with the 945 chipset. So what's the deal? Is it only for people who want to upgrade their laptops? So laptops supporting the Core Duo come in a 478 pin package? Kinda confusing.

Nevermind, I got my answer. They use the 478 pin package, but it's not compatible with any 478 motherboards.

"There are a lot of Conroe benchmarks. HotHardware, Bit-Tech, PC Perspective and X-Bit Labs all have some Conroe benchmarks, too."

They can come out with 10,000 benchmarks. They mean nothing until the “NEW” CPU's from both parties are in retail & selling.

All these reviewers used Intel's configurations. I don't even agree with the hardware they used for the AMD system. The buggy DFI Radeon Xpress 200 mobo which is still having problems even with the bios update.

The Bios Update may have solved some problems, but the Xpress200 performance issues are still there when using High End Cards like the X1900’s or X1800’s. You can read all about it in DFI Street.

I would like to see both CPU’s benchmarked with new Radeon Xpress3200 chipset or SLI with a much better & stable mobo like the Asus or ABIT.

The 12/23 BIOS offers the following fixes according to DFI LANPARTY UT RDX200.

1. Fix memory Set 2-1-1-1-1 and 4-1-1 mode wrong.
2. Set Cool'n'Quiet default disable.
3. Change the description of DQDRV.
4. Fix Read Preamble Table Error.
5. Shorten the delay time during clock programming loop.
6. Add over clocks step by step.
7. Fix fill 3114 SVID&SSID under Cross fire mode.
8. Fix soft-reset hang on POST code F2h if enable USB mouse support.
9. Change CMOS used to fix some control item can’t save.
10. Add support K8 FX60 CPU.
11. Update SiI3112 Raid ROM.
12. Fix some SATA(DiamondMax 10 (6B160M0) HDD ) HDD detect fail at first time cool boot.

They are still working on a newer BIOS, because there are still performance problems with different system congif's when running 2 x X1900's & X1800's. Though the X1800's seem to perform much better due to it being out longer I guess. Also CCC 6.3's run better on the Xpress200 for both AMD & Intel mobo's.

hey XP do you have any recommended bios settings for the DFI Lanparty UT SLi DR Expert?

I am still trying to figure it out myself LOL,

I did notice that the DFI LAN Party mobo's are extremely sensitive to hardware. So, if you have say a DDR Ram, which for some reason the mobo don't like, forget it, it is a challenge within itself to get that thing working.

With my DFI, even at Bone Stock & default settings it picks & chooses when it wants to boot up. Actually, I have a better chance for it booting up when I configure the bios for OC'ing.

You see, this is what I was trying to explain, the DFI mobo's suck when they run @ default bios settings. Just unstable & lack of performance, well, in my case that is, and from what I've researched back then on the net.

"Conroe isn't a mobile processor. Conroe is desktop, Merom is mobile, and Woodcrest is server."
You're right, I had just always heard that it was a mobile. Still, it's like AMD building their desktop chip on the Turion. Sure, it can be done. But then, it's kinda hard to say that it was a fair match. Like try comparing the Turion to the Prescott. Talk about a blowout.

Payton,
A few years back...about 15 or so, I was working on a project where utilization of cache memory in a dynamic manner in a multi-processor environment was a option. In our configuration it was found that this added overhead, and complexity that caused system instabilities, so we opted for a static assignment. One of the difficulties in multiprocessing designs, historicaly, has been in providing for non-conflicting access to memory, and in direct sharing of the data in memory between two or more processors (the higher the number of procs the more difficult it became by a logrythmic and not a linear progression), without overflowing the pointers/stacks. Given the state of the presently available Desktop OSs, it is my fear that the use of dynamic memory allocation on anything except the most rudementry basis (bank assignment) will probably lead to more vulnerabilities than the efficiency gains can overcome.
It has been my experience that RISC processors scale better than CISC processors, but with modern CISC processors being RISC cores for all intents and purposes with CISC microcode surrounding them, hopefully our friends at Intel and AMD can hash through these issues and provide a truely next generation solution.

I truely wish that Motorola could have stayed in the mix...

Gerritt

I'd like to have a Conroe under my Chilly1 phase change unit and see what I could max it out to, LOL.

To my understanding (been out of the hardware scene for a while), isn't Yonah the Conroe counterpart for mobile chips?

I'd like to ask how comparing a Turion to a Prescott isn't a fair match. They're processors. Comparing a Pentium versus an Athlon 64 is a fair fight. The Pentium will lose, but so be it. Is not the Athlon a superior processor? It is. Same with the Conroe versus the FX-60. Conroe just happens to deliver better performance while consuming less power. Big deal. I could really care less if Conroe was "based off of a mobile chip" or not. The fact is, it performs, and it performs damn well relative to the current fastest offering on the market.

-Pikl

Folks,
Lets not let this degenerate into yet another distance vs. accuracy contest.
The Conroe appears to have the capability of being a very formitable processor, but it is NOT the greatest thing since sliced bread, or cheese!
This is what I was trying to point out!
AMD has incorporated a memory controller on it's processors for a while, and now Intel is doing the same, but one uping the AMD offering by utilizing a shared L2 cache vs. separate caches.
If it works this is a good thing.
It lowers the transistor count for the same amount of L2 memory accessed, but at the cost of more complex micro-code.
Can it be more efficient, evidently so. But until we start testing WITH the technology vs quoting the pundits, we need to be carefull, because like I say in my tag, we can all be wrong.

Gerritt

>The Conroe appears to have the capability of being a very formitable processor, but it is NOT the greatest thing since sliced bread, or cheese!<

I didn't have enough money left this month to buy any cheese, so I've been living on grilled bread sandwiches. To add insult to injury I gave up my 6600GT to my kid and took back my 5900XT sigh.............

With processors already in the $1000 range, I'm curious what these new gen CPUs are going to cost. At the current prices they would have to come with dancing girls to get me to buy one. I'm afraid I'm stuck with dinosaur tech for awhile. So I'll just have to wait and read what is and what others think after they come out. But whatever the plus and minuses, it's nice to see both AMD and Intel are continuing to push the envelope. Imagine where we will be in 5 years, when just 5 years ago we were saying goodbye to the P III

Good info Gerritt

All new Intel CPU's are based on the Meron Core. And these new Intel CPU's are not quite new seeing that it is there P3 design, with a little of the P4 concept stuck together.

Never the less, if they perform, then that is good enough for Intel which should by them some time until they complete there "NEW" CPU design coming in around 2009+ or so. With this new design, I sure hope they scrap there FSB design or they are in trouble IMO.

AMD does have an answer, but are keeping in behind close doors. This is good healthy competition which should drive CPU $priced$ down to the pavement

Payton,
Back when...you should remember the time....
A Great PC cost $3000+ even if it's a IBM PC/XT/AT/386 (all in their times of course). I remember the first 80386 systems with Wietek math coprocessors coming out at about $10K!

This still holds true.
It is only in stepping back a generation or speed on CPU/GPU that you can have a Very Good PC for about half the cost with 80% of the throughput. Now, you can have a PC that has 50% of the throughput of the highest end systems coming in at 10-20% of the cost.

I'm sorry to hear about your grilled bread sandwiches, but they do have over half the nutritional value of a grilled cheese sandwich, so support the pricing deltas between toast, grilled cheese, and a BLT!

Gerritt

So, I'm looking at the benchmarks at Hot Hardware, and about half-way down the page shows the specs of both computers; that's not even a fair fight. I'm an Intel fanboy, but c'mon! The RAM and motherboards don't even match up! For a fair benchmark, all specs should as close to identical as possible, with of course the exception of the piece of equipment being tested, in this case the Conroe @ 2.66 GHz vs. AMD @ 2.8 GHz. Again, I'm an Intel fan, but it's just not fair if you need to put everything at apocalyptic power compared to your competition in order to win. Anyone, AMD or Intel fan, should see this.

http://www.hothardware.com//viewarticle.aspx

Search "Conroe" its the only article that pops up.

I also wanted to point out: Hot Hardware is testing a Dual Core (Conroe) against a Single Core (FX-60). Duh! Granted, today's games don't exactly take advantage of Dual-Core, but the other benchmarks = UNFAIR. I refuse to believe a benchmark where things just aren't even close to fair.