• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

Date for 955P chipsets?

hardwarephreak said:
Edit: Here is a snipet from Anandtech's article:

The Pentium-D and Pressler are examples of how not to do it: just slap two CPUs on the same die and call it a day. High clocked single cores like the upcoming Athlon 64 FX-57 will eat these massive chips for lunch in almost all benchmarks while consuming less energy. With the exception of some special far-fetched benchmarks, it will be pretty hard to justify the reason behind these dual cores.

Luckily, Intel's Yonah and AMD's Dual Athlon 64 cores show that better multi-core CPUs are on the way. At that point in time, we are entering the multi-core engine for real. And we can only applaud that because it unleashes a massive amount of CPU power upon the developers.


http://www.anandtech.com/cpuchipsets/showdoc.aspx?i=2377&p=5

Go ahead and tell me Anandtech is a division of AMD's marketing dept. :rolleyes:

You left out one Part.

Date: March 14th, 2005
Topic: CPU & Chipset
Manufacturer: Various
Author: Johan De Gelas

Things have changed since then. It was Anand's own review that blew that arguement out the water. The Dual Core 3.2 Pentium can't do single threads faster than an A64-FX57, but would kick the crap out of it multi-tasking and running most multi-threaded apps. That's how it's been for Dualies since the Dual Pentium Pros that I can remember, hehehehe!

Anand's Part 2 of their Dual Core review showed a $240 2.8GHz Intel DC. There's no further need for speculation ;)

Sure I'll say Anandtech is BIASed for AMD but clearly not as much as some others. Wes Fink, one of their writers, is nothing more than an AMD-tool but that's for another thread.

No matter what's said here, the most important performance factor will always be Software support=P No matter if it's Dual Core or Dual Processor, software will determine who runs what faster.

Donnie27
 
robberbaron said:
Yes, but I'm banking on DDR2 being half way decent by then, and I'd rather do DDR3.

Yes, but bear in mind the FX's debuted on the 940, and those are desktop gaming chips.

BTW, it already posted that AMD will use DDR2 long before DDR3 ships. The AMD chipsets for Jan 2006 shows Socket M2 and DDR2 667=P

Donnie27
 
hardwarephreak said:
I would look to find nForce4 SLI boards for Intel out by the end of this month. More specifically the Epox, as they have been all over the place in terms of displays and behind-closed-door demos of the nForce4 architecture for Intel CPUs. I won't speak to the availability of the boards as we all know what is the latest trend with launches (paper?!?)

The 955/945 chipsets are a stop gap solution for Intel. Expect to see new chipsets/sockets (771) with the release of the true dual core chips from them. I like to look at the Pentium D as Intel releasing Pentium4's that are SMP compatible...except no dual socket board...we'll just squeeze them onto one chip...not really dual core.

These boards go on sale next week=P The stop gap solution was 925/915 then their X/XE upgrades. nForce4 IE, 955 and 925 will be around for about a year when they're due to be replaced as well. Socket 771 and AMD's M2 should ship about the same time. Unless that rumor about Intel moving up 65nm to the 3rd Qt proves true. Pressler, Cedar Mills and etc.. will be well known before they ship. No price mysteries like current AMD64 Dualcores. AMD and Intel said they're not looking for a large RAMP UP of Dual cores, just enough to meet demand. Market speak for "we intend to ship a few at higher prices".

For the record, I'm going with AMD on this one.

Donnie27
 
hardwarephreak said:
The 955/945 chipsets are a stop gap solution for Intel. Expect to see new chipsets/sockets (771) with the release of the true dual core chips from them. I like to look at the Pentium D as Intel releasing Pentium4's that are SMP compatible...except no dual socket board...we'll just squeeze them onto one chip...not really dual core.

what do you think dual core is? THIS IS dual core. Dual Core is two processors on a single package, not necessarily on a single die.
There is not much more to it then that.
 
http://www1.us.dell.com/content/topics/segtopic.aspx/xps_gen5?c=us&cs=19&l=en&s=dhs

The NEW DimensionTM XPS Generation 5!


Be the envy of your friends with the latest DimensionTM XPS extreme gaming and enthusiast PC featuring Intel's latest 955X chipset.

The latest Dimension XPS desktop is designed with the power and performance for even the most demanding gaming and multimedia applications. The new Intel 955X chipset provides support for the new dual-core processors with Hyper Threading technology so customers can play intensive 3D games while performing other tasks such as burning music CDs with ease. The new chipset also supports dual channel DDR2 Memory at speeds up to 667 MHz for rich game play and fast performance. Customers can also choose high performance PCI Express graphics cards, 7.1 surround sound, and Microsoft Windows XP Media Center Edition 2005 for a realistic, home-theater experience.


The Dimension XPS Gen 5 will continue to come standard with our dedicated XPS telephone support and stunning arctic silver chassis! Features include the NEW Intel 840 Extreme Edition processor

I bet Kyle is reviewing one right now..... thats what that whole... we're going to be review systems soon thing was about. Probably more money in it!! Get some big Dell advertising $$.
 
A 'real' dual core solution to me at least would involve some actual development into tech that would enhance the ability for the cores to communicate more effectively between each other and the rest of the components on the board, rather than take a dual processor solution and combine it into one package. Obviously Intel feels the same way as their next solution will have some of those features...come to think of it didn't they announce that they were developing a "hypertransport" type of bus interconnect.
 
Donnie27 said:
I bet you're right!

Donnie27

They reviewed the dumb Gen2 1st. I was hoping for the Gen5 XPS review today. :(

They're on sale now:

Dell: http://configure.us.dell.com/dellstore/config.aspx?c=us&cs=19&l=en&oc=DXPS5SAP&s=dhs

Alienware: http://www.alienware.com/product_de...ysCode=PC-AREA51-ALX-R2-D&SubCode=SKU-DEFAULT
- - - - http://www.alienware.com/product_de...SysCode=PC-AREA51-5500-R1&SubCode=SKU-DEFAULT

Velocity: http://www.velocitymicro.com/dualcore.php

I'm starting to get excited thinking about dual core at 4Ghz!!
 
Does anyone know when we will be seeing them on tier 1 motherboard manufactures sites? How about when the Dual Core AMD Athlons are coming out?
 
lelliott731 said:
Does anyone know when we will be seeing them on tier 1 motherboard manufactures sites? How about when the Dual Core AMD Athlons are coming out?

My crystal ball..... says....... 1-2 weeks and you'll see 955's at like $300 from one or 2 manufacturers.

And maybe a month from now, more at reasonable prices.

<pure speculation>
 
They SKUd up yesterday for around $225 (Intel). Nothing from MSI, Asus, or Gigabyte yet. But plenty of 915/925 boards all around :rolleyes:
 
chrisf6969 said:

Would be danged nice!

Donnie27
 
cant wait for my nForce4 SLI Intel Edition DFi LanParty board with a Pentium-D in it!
 
I'm planning my next upgrade right now:

Pentium D (not EE) think the 3.0 or 2.8Ghz (cheaper version)
955 or Nforce 4 SLI mobo
1Gb DDR667
1x PCIe 6800GT (for now, may upgrade to SLI later)
Lian-Li PC-60BPlus
Lite-On 16x DVD burner
use my old WD 74Gb raptor (prevent data/ identity theft)
etc..

And sell my old Lian-Li PC 60, 2.4C @ 3.4Ghz, 6800GT AGP with a brand new harddrive (probably 80Gb) in it.
 
chrisf6969 said:
I'm planning my next upgrade right now:

Pentium D (not EE) think the 3.0 or 2.8Ghz (cheaper version)
955 or Nforce 4 SLI mobo
1Gb DDR667
1x PCIe 6800GT (for now, may upgrade to SLI later)
Lian-Li PC-60BPlus
Lite-On 16x DVD burner
use my old WD 74Gb raptor (prevent data/ identity theft)
etc..

And sell my old Lian-Li PC 60, 2.4C @ 3.4Ghz, 6800GT AGP with a brand new harddrive (probably 80Gb) in it.

You might need to see MaxPC May Issue on page #68, hehehehe!

Donnie27
 
Alright, how about this?

Not so dated, well yesterday...granted it's AMD's own Anandtech again, but it doesn't look like they have changed their tune...I will admit that it isn't that far off from AMD's solution...but AMD actually has had the supporting tech to pull of Dual Cores since the inception of the Athlon64. I am not saying the PentiumD is going to suck by any means, I would want one, it's going to be a bad @$$ chip...but I would look to AMD to have the better first gen dual core solution. Intel however, is definitely more attractively priced compared to what has been released from AMD

AMD's dual core Opteron & Athlon 64 X2 - Server/Desktop Performance Preview
Date: Apr 21, 2005
Type: CPU & Chipset

Even Intel will admit that the architecture of the Pentium D is not the most desirable as is two Pentium 4 cores literally glued together. The two cores can barely be managed independently from a power consumption standpoint (they still share the same voltage and must run in the same power state) and all communication between cores must go over the external FSB.

...Any communication between the two cores has to be done over the external FSB, and obviously, core-to-core communication over an external bus is slow. It particularly doesn't make sense, since the two cores are on the same die.

AMD's architecture is much more sophisticated, thanks to the K8 architecture's on-die North Bridge. While we normally only discuss the benefits of the K8's on-die memory controller, the on-die North Bridge is extremely important for dual core. Instead of having all communication between the cores go over an external FSB, each core will put its request on the System Request Queue (SRQ) and when resources are available, the request will be sent to the appropriate execution core - all without leaving the confines of the CPU's die. There are numerous benefits to AMD's implementation, and in heavily multithreaded/multitasking scenarios, it is possible for AMD to have a performance advantage over Intel just because of this implementation detail alone.
 
I haven't looked at them myself, however in my years of reading sites like Anand, Tom's, yadda yadda, I've learned that benches aren't always as accurate as they can be.
 
hardwarephreak said:
Alright, how about this?

Not so dated, well yesterday...granted it's AMD's own Anandtech again, but it doesn't look like they have changed their tune...I will admit that it isn't that far off from AMD's solution...but AMD actually has had the supporting tech to pull of Dual Cores since the inception of the Athlon64. I am not saying the PentiumD is going to suck by any means, I would want one, it's going to be a bad @$$ chip...but I would look to AMD to have the better first gen dual core solution. Intel however, is definitely more attractively priced compared to what has been released from AMD

WOW! You trust Anand more than the Guys who run HardOCP? In a word, Anand and especially Wes Fink is FOS IMHO! Many forums call them an AMD mouthpiece. The XRays show what Dual Core Opterons are and what Dual Pentium4's are. They even have common parts like, AMD's Crossbar Interface is Similar to Intel's Arbtitration Logic (Interface). The other parts are half of what two Opterons would be. It looses one of its 128bit memory controllers. Again, other crap-o-la about what AMD planned DC since inception is a bunch of Marketing crap to impress Fans and Newbies. Now is the AMD DC good, of course IT IS, it was good as a single processor and as DUAL or more LOL! What? Is it supposed to not be good as a Dual core? That doesn't make any sense. We're supposed to be Geeks who see through AMD's and Intel's marketing gobbly gook!

AMD's biggest problem will be, VOLUME, VOLUME and MORE VOLUME! SO they will NOT be cheap, AMD can't afford to make them cheap. If they had problems building single cores, and they sure as hell did BTW, then building two at a time will be twice as hard.

Go back to Anand and see if you can find the total number of all Hammer of anykind built? See if Anand will tell you that it took AMD a full year to reach 1 million Hammers built? Or that Intel has built more X86-64 in 6 months than AMD has in two years? Search through the Techweb, Cnet, ZDNet and even the Inquirer?

Donnie27
 
AMD's biggest problem will be, VOLUME, VOLUME and MORE VOLUME! SO they will NOT be cheap, AMD can't afford to make them cheap. If they had problems building single cores, and they sure as hell did BTW, then building two at a time will be twice as hard.

From my post...
Intel however, is definitely more attractively priced compared to what has been released from AMD

I know they won't be cheap. I also know why. Hell they even say why in that marketing piece by Anandtech/AMD. Which kind of tells me you haven't even read the article, if you want to debate something, at least read it. Why do you think AMD is releasing the 800 series Opterons first? Because those are the lowest yield, and they also have the lowest demand, and the highest margin. AMD's problem has ALWAYS been manufacturing. Until their new fab comes online w/ 65nm chips on 300mm wafers, dual core chips aren't profitable unless they charge a lot for them. Again why they are releasing the 800 series. They simply don't have the resources of Intel. Whether that be marketing, R&D, OEM tie-ins...you name it Intel is just bigger in every way. It is no contest as to who shipped more chips, I don't need to read all the stuff you suggested to know that Intel owns the desktop market, the server/workstation market, the mobile market, and the video core market. I am not talking about quantity, I AM talking about who has the better first generation dual core CPUs coming to market. That's it. Who has the better DC implementation at this point in time.

I just don't see how you can argue that the Hammer core was not designed for dual core from the get go. I don't believe for one second that AMD made its desktop DC chips socket compatible to earn favor with the end user...sorry I don't buy that. IF anything, they are doing it because the OEMs that AMD has are demanding it. Also if you are AMD, you don't want to completely be out of the dual core desktop market and just let Intel dominate. "For the end user"...that is definitely PR crap. However, AMD is more interested in the server/workstation market and gaining market share there, especially over gaining market share in the desktop. I do think that AMD had DC planned from the get go, to make their Opteron solutions more attractive. Hey buy a two way system now, get a four way later.

Here are two quotes from this article: (http://www.linuxhardware.org/article.pl?sid=05/04/21/1747217&mode=thread)

Below is a block diagram showing how the new dual-core system works. All of the elements below exist inside the processor itself. Notice that the two processor cores sit above what AMD is calling the Crossbar and SystemRequestQueue. These components have existed in the AMD64 architecture from day-1 anticipating dual-core implementation

The first thing you may notice is how the integrated the Opteron core looks versus Intel's core. While AMD integrated the cores using internal logic, Intel integrated the cores by taking two processors, “gluing” them together, and letting the chipset handle the work load.

You don't think that Intel would rather be making "Irwindale" Dual Core Xeons? How big do you think a dual core Xeon would be @ 90nm with all of that cache (2MB/per core)? They are restrained until their 65nm process comes online (which looks good by the way)

Intel is relying on the FSB. Not the most ideal way of supporting dual core. This is a fact. This is not disputed. This is how Intel is doing things. This is also why they are developing their own Hypertransport-type bus interconnect.

A key point to keep in mind is that wheter it be Intel or AMD, both companies have teams of people planning out $hit that we haven't even heard about yet, and how they will counter act the others moves. That's just smart business. Dual core had been a known course of action for both chip companies. I do believe it was Intel however, that most recently decided to go a completely different course...do you remember the cancellation of the 4GHz Prescott and all of the crap that came out about how Intel was going to focus on other factors rather than speed alone. They new all of those changes would happen, maybe they didn't forsee them happening at that point in time, but any good plan has to be flexible.
 
First of all, I did read the article (and others as well) and thought that line about AMD's prices is BS! Even Anand couldn't soften the blow and he knew it! As I said before any prices were released, AMD can't get good enough yields so they're going for the most MONEY=P I told you before any AMD price was release that these bad boys would be "Bank Busters;)" and I'm sure as hell not only talking about the Opterons. The Sad damned part about all of this is that the damned prices are so high that AMD will now not even cut the prices of the Single Core processor prices, that sucks even more. That’s what the hell I was waiting for LOL!

Last but not least, part of a good Core/Processor is the ability of the manufactures to manufacture it=P If these things are hard to build, then that BS about Dual Core from inception is even more BS than many AMD Fans want to admit. That’s why Dell and even folks who buy Opterons don’t invest more time and money into them. They know Opterons are great performers and etc.. AMD has been there and done this before, it's happened the same Frackin’ way with the K6-2 back in 1998 LOL!

Let’s say all the Crap about AMD building this thing to end up this way right from the start is true. If that is the case, then they’re really screwed up over the last 3 years. Yes, 3 years. After 3 years we have overpriced low yield Opterons to show for it, WOW! Then if you and the others are right, AMD adds what amounts to a half of another Venice (minus the Memory Controller and HT links), add it to another processor and then charge and ARM and a Leg for it! Yup, they nutty alright! What were they thinking LOL!

Donnie27
 
http://www.overclockers.com/tips00765/

Same Old Stunt

Those of you who can remember back a couple years might remember that when Athlon 64s first came out, they were also high-end loaded. AMD started at 2GHz, at very high prices, and didn't release slower ones until long after introduction.

AMD was able to proclaim that they got the product out, but Athlon 64s sales were minimal until AMD eventually released slower processors at much more reasonable prices.


Donnie27
 
So after all that, where did you mention anything about who has the better dual core implementation? Yes AMD is going to be priced higher. That has been reported. Yes they don't have the production capacity. That has been reported as well. Yes they have a history of not having low speed variants out to market quickly or close to launch (which most companies are nowadays...) Again though, as has been with almost every CPU launch from Intel and AMD over the years, the top end chips are available first...after all if you want it badly enough, you can pay for it...Intel and AMD know where the margin is.

Granted the thread title is not really in anyway related to what we are discussing. :p But my point has been to try and discuss who has the better initial implementation of a dual core processor.
 
hardwarephreak said:
So after all that, where did you mention anything about who has the better dual core implementation? Yes AMD is going to be priced higher. That has been reported. Yes they don't have the production capacity. That has been reported as well. Yes they have a history of not having low speed variants out to market quickly or close to launch (which most companies are nowadays...) Again though, as has been with almost every CPU launch from Intel and AMD over the years, the top end chips are available first...after all if you want it badly enough, you can pay for it...Intel and AMD know where the margin is.

Granted the thread title is not really in anyway related to what we are discussing. :p But my point has been to try and discuss who has the better initial implementation of a dual core processor.

Oh brother with all of that said you still don't have phucking clue do you? You're on the Intel side of the forum trying to push AMD! the title of the thread is "Date for 955P chipsets?" :mad: What in the hell do your spamming AMD have to do with that? This thread was nothing about how you love AMD, or who does what best? I answered what you asked now you change subjects to something else! [Canuck=Donnie27]Oh Geesh Eh![/Canuck]

It's not about how badly someone wants. Good Guy AMD is not supposed to *squeeze customers like the Evil Intel. Yet, they are and here they have !!!!!!s already trying to greese their path and make excuses for them. Get over it, AMD is just as slimy and greedy as Intel. What part of that do you not understand? Both are holding back to maximize profits and that does fit right with this guy asking about when will they launch? You are off base here, I'm not on the AMD side of the forum trying to sell Intel stuff! :rolleyes: AMD Fans, sheesh!

Donnie
 
Dude, I own one AMD box. I own four Intel boxes. I am a fan of cool tech, not of an institution looking to take my bank (unless of course I own their stock)

You are the one coming off sounding like a fan boy :rolleyes:
Hell, they don’t need any AMD !!!!!! Zealots to trash them. They’re doing a great job shooting themselves in the foot.

Here is my FIRST post in this thread...notice...NO AMD IS MENTIONED
I would look to find nForce4 SLI boards for Intel out by the end of this month. More specifically the Epox, as they have been all over the place in terms of displays and behind-closed-door demos of the nForce4 architecture for Intel CPUs. I won't speak to the availability of the boards as we all know what is the latest trend with launches (paper?!?)

The 955/945 chipsets are a stop gap solution for Intel. Expect to see new chipsets/sockets (771) with the release of the true dual core chips from them. I like to look at the Pentium D as Intel releasing Pentium4's that are SMP compatible...except no dual socket board...we'll just squeeze them onto one chip...not really dual core.

Here is your reply...NOTICE..you brought up AMD.
IMHO, we'll see at least Asus, MSI, EPOX, ABit and AOpen nForce4 or i855 or both by Friday next week. The rest by the end of the month. They will start showing up this Weekend BTW!

The board I've been seeing the most news on is the MSI board. There's screen shot of it running an Intel 830/3GHz Dual Core so that's not a problem either.

Of course Intel's dual core is just two Presscotts, AMD's dual Core is just two Athlon64s, so? Are we supposed to believe AMD did something different other than linking two A64's on the same package? Too much AMD Fan hype is muddying the information pool.

The Inquirer posted the rumor that mobo makers say Intel is lying about Dual Core not working on i925 and 915. Intel lying, wow, they were surprised?

Donnie27

You brought it up, I was RESPONDING to YOU. Talk about someone needing to get a phucking clue. After you brought it up I thought I might get some kind of interesting stuff going back and forth about the different tech each company is offering, but you are starting to look like the fan boy you are accusing me of being.
 
Hey people, IMO it's not worth getting heated over...

Ilike AMD for my own reasons...You like Intel for your own reasons....


Agree to disagree?
 
steviep said:
I'm sick of seeing "Q2 2005" in release dates for the new Intel chipsets. Does anyone know when the motherboards with the 955P are coming out?
The Asus P5WD2 Premium Intel 955X showed up on a couple e-tailers sites this morning. Retail price is $265. One site says the boards should be shipping by May 9th.

Tom's Hardware reviewed the D840 and the bottom line is that for applicaitons that are not thread optimized (games) the higher clock rate Prescotts are faster and for thread optimized rendering/encrypting apps the Smithfields are 15%-50% faster. Heat dissipation is about the same as the Prescotts in the 130W range.

I think the 915/925 chipset family must have set some kind of record as for the shortest time from introduction to obsolescence. I'm betting on about the same for the 955, maybe only 9 months. I'm a longtime Intel user but it's hard to be happy about they direction they are taking. More to the point, as a business manager it costs us money when Intel changes the chipset platform this frequently, particularly when there is no associated significant performance payoff. From where I sit it appears that someone from AMD is sitting in the drivers seat at Intel. ;)

oc
 
IMHO they are making a mistake not putting hyperthreading on the regular Pentium D now there is not much of a reason not to by an AMD dual core except the chipsets.

they should just make the expensive edition have a faster bus or more cache or something instead of crippling the D
 
There was a recent article stating that ATI has made sure that their multi-VPU solution works on 955 boards with dual PCI Express Graphics slots. Whether or not these boards with multiple graphics slots are available anytime soon is the question.

Would be nice if the nvidia let their drivers allow for SLI on the 955 (like on Tumwater based boards). But that's not going to happen due to P4 nForce4...but damn that would be nice.

ATI and nVidia multi-VPU compatible...we can dream can't we.
 
hardwarephreak said:
There was a recent article stating that ATI has made sure that their multi-VPU solution works on 955 boards with dual PCI Express Graphics slots. Whether or not these boards with multiple graphics slots are available anytime soon is the question.

Would be nice if the nvidia let their drivers allow for SLI on the 955 (like on Tumwater based boards). But that's not going to happen due to P4 nForce4...but damn that would be nice.

ATI and nVidia multi-VPU compatible...we can dream can't we.


Well if it matters or not I take back everything even close to sounding negative towards you=P

Donnie27
 
Itanium has up to 24Mb of cache, and Xeons have up to 8Mb of cache, why can't the extreme expensive edition get some more cache lovin.

The original EE was nice when it had 4x the cache of the regular P4's.

The latest EE's are fairly disappointing.
3.73 ESPECIALLY. (compared to 600 series)
Or Pentium D EE (compared to regular Pentium D's)
 
Well if it matters or not I take back everything even close to sounding negative towards you=P

Absolutely no problems man :p

The only way I see a complete solution is if someone hacks the nvidia drivers...which is a definite possibility ;)
 
Itanium has up to 24Mb of cache, and Xeons have up to 8Mb of cache, why can't the extreme expensive edition get some more cache lovin.

The original EE was nice when it had 4x the cache of the regular P4's.

The latest EE's are fairly disappointing.
3.73 ESPECIALLY. (compared to 600 series)
Or Pentium D EE (compared to regular Pentium D's
)


With the current 90nm process an increase in cache size isn't feasible...the PentiumD is already huge...they do plan on increasing cache sizes...but you won't see that until the newer 65nm process comes online. Which, according to their last earnings report, seems to be moving along extremely well.
 
Back
Top