New socket to replace LGA1366 due in 2011

[mbigna]

Meh. I went from Socket A (AMD) to 775 a couple of years ago. It will be another 3 years before I can afford to think about another rig, let alone care about what new sockets there will be.

 

[wingman99]

For Socket 1366, we still have CPU's yet to be released.
Also like I said, 3 years and not having to change my motherboard, I am ok with that. The Fact Intel or AMD cant make a faster product for me for at least 3 years, I am not.


Also 6 Generations have not passed in Socket 775
P68 aka P4 is a 7th Generation CPU
Core 2 is a 8th Generation
Core i7 is a 9th.
I have no idea how you are getting 6 generations...

As i said before I base generation by revolutionary and evolutionary die changes as upgrades. that means a new chip with with a new number or name with a change in architecture of some sort, and so does the rest of the world LINK http://www.google.com/search?num=10...tion+of+cpus&start=100&sa=N&cts=1271539134921
pentium 1,2,3,4 are a generation change.

LGA 1366 has 1 gneration upgrade, i7-980X Gulftown hex core 32nm

Also, How many Chipsets/ Motherboards for Socket 775 are still supported or work with Prescott all the way to a Core 2 Quad?
I think you are over ratting the platform.
A 6 year old motherboard/chipset will not be cutting it today.
Current Chipsets that will not work would be
8XX series
9XX series are hit and miss, but giving the benefit of doubt we will say a
945 or 965 board is still relevant, so a 2006 release, thats 2 years off your timeline right their.

945-955 board was started in 2005 I was 1 year off that makes 5 years on the same board.
and look at these generation of chips 945 board suports, I count about 8 different series.
LINK:http://www.gigabyte-usa.com/Support/Motherboard/CPUSupport_Model.aspx?ProductID=2409&ver=

 

[wingman99]

Prescott and Core/Core2 are the two architectures.
Note the period between "Core2" and "Core2duo".

Core2 was based closely on Core, which was based loosely on Pentium M, which was based on P3, if you want to be exact.

And the core2duo there was second changes in that architectures E7500 to E8400.You have to use all the names and all the numbers like in the old days pentium 1,2,3,4 are a generation change. I know it's complicated these days to keep track form the old days, when the naming and numbering scheme was simple.

 

[Bobnova]

E7500 and E8400 are the same generation.
E7xxx is an E8xxx with some of the cache cut out, and running on 266fsb instead of 333. The core itself is identical.
Same with e5xxx(more cache cut out, 200fsb) and the 45nm e6xxx(266fsb, more cache cut) too.

 

[ihrsetrdr]

Things to replace every 3 yr.s (or sooner):

1. socks
2. underware
3. CPU/motherboard

 

[petteyg359]

Also 6 Generations have not passed in Socket 775
P68 aka P4 is a 7th Generation CPU
Core 2 is a 8th Generation
Core i7 is a 9th.
I have no idea how you are getting 6 generations.

At least five four: 180nm(180nm was 423/478), 130nm, 90nm, 65nm, 45nm.

P4 - Willamette 180nm, Northwood 130nm, Gallatin, Prescott 90nm.
Core 2 - (Conroe. Allendale. Kentsfield.) 65nm (Penryn. Yorkfield. Wolfdale.) 45nm

 

[wingman99]

E7500 and E8400 are the same generation.
E7xxx is an E8xxx with some of the cache cut out, and running on 266fsb instead of 333. The core itself is identical.
Same with e5xxx(more cache cut out, 200fsb) and the 45nm e6xxx(266fsb, more cache cut) too.

Not even close, you have to pay attention to the numbers, Im so glad you went there.

LINK:http://www.xbitlabs.com/articles/cpu/display/intel-wolfdale_2.html
QUOTE: Wolfdale processor core is 107sq.mm big and consists of 410 million transistors. These numbers give us to understand that Wolfdale has been changed significantly since the times of its predecessor - 65nm Conroe with only 291 million transistors. You can even see it from the photographs of Wolfdale and Conroe cores: the positioning of their functional units has slightly changed.


Wolfdale on the left and Conroe on the right (the images aren't scaled).

So, Wolfdale core is not just a smaller Conroe core resulting from the transition to finer production technology. Intel engineers made a number of innovations in the new processors (you can read more about the Penryn processor family in our article called Second Iteration of Core Micro-Architecture: Intel Core 2 Extreme QX9650 CPU Review).



Most of these innovations should increase the processor performance. Wolfdale’s biggest advantage is the larger 6MB shared L2 cache memory. Moreover, Wolfdale processors support SSE4.1 instructions set including 47 new commands that can speed up 3D graphics and video processing alongside with scientific calculations in case of appropriate applications optimization.

A few changes have been made to the execution units, too. Wolfdale processors have now acquired Fast Radix-16 Divider unit that speeds up division and square root calculations. The new processor also features Super Shuffle Engine mechanism accelerating those SSE instructions that require bit-by-bit shuffling.

The above listed improvements and a few additional ones introduced in Wolfdale processors ensure that these new processors will work faster than the older Conroe CPUs running at the same clock speeds. However, we are not talking about any significant advantage here. Wolfdale offers only minor cosmetic refresh of the Core micro-architecture, that will change dramatically only in the upcoming Intel Nehalem CPUs scheduled to arrive in the end of 2008.

The most important thing about Wolfdale processors is certainly the new 45nm manufacturing process that allowed Intel not only to significantly increase the number of transistors per die without making the die any bigger in size. New production technology uses a new hafnium based material with a property called high-k, for the transistor gate dielectric, and a new combination of metal materials for the transistor gate electrode. This allows increasing processor clock frequencies further without raising their heat dissipation and power consumption. That is why these new processors will be of specific interest to overclockers, who will definitely be able to use them to set new overclocking records.

 

[wingman99]

At least five four: 180nm(180nm was 423/478), 130nm, 90nm, 65nm, 45nm.

P4 - Willamette 180nm, Northwood 130nm, Gallatin, Prescott 90nm.
Core 2 - (Conroe. Allendale. Kentsfield.) 65nm (Penryn. Yorkfield. Wolfdale.) 45nm

I would say that's 7 different generations if you count all the names of the cores and there numbers. It was much eaiser with pentium 1,2,3,4 generation change.

Just like the new Generation coming up is Sandy Bridge architecture.

 

[JaY_III]

Some of you guys are really confusing what constitutes a generation.
A die shrink does not
A Core variety/variant does not.

[fact]The underlining architecture of the core is what defines a generation[/fact]

It does not matter what ever you are defining, if its anything other than an architectural change, it is not a different generation.
It doesn't matter if you add cache/Cores.
Shrink the die, or even add new instructions.
That CPU will still remain a variant of that said architecture and thus generation family.

CPU Generations based on the the x86 architecture are as follows

1st Gen: 8086/8088
2nd Gen: 80286
3rd Gen: 80386
4th Gen: 80486
5th Gen: P5/P54/P54C/P55C
6th Gen: P2/P Pro/P3
7th Gen: P68 (Pentium 4)
7/8th Gen: Yonah (Core)
8th Gen: Core (Core2)
9th Gen: Nehalem (Core i 3/5/7/9
10th Gen: Sandy Bridge

I may have missed on or two derivatives in their, but those are the only generations of CPU's that exist.

@wingman99
The Gigabyte GA-945PL-DS3 (rev. 2.0) linked is a board from LATE 2006.
You would be lucky to have found it in stores Q4 06. It will have its 4th Birthday in Nov 2010
You may count 6 generations of CPU's but it is in fact 2 generations that this board supports.

@petteyg359
Like i said above a die shrink does not constitute a new generation nor do any of the small variants of the core.

 

[wingman99]

Some of you guys are really confusing what constitutes a generation.
A die shrink does not
A Core variety/variant does not.

[fact]The underlining architecture of the core is what defines a generation[/fact]

Yes I pointed that out above the architecture changees

It does not matter what ever you are defining, if its anything other than an architectural change, it is not a different generation.
It doesn't matter if you add cache/Cores.
Shrink the die, or even add new instructions.
That CPU will still remain a variant of that said architecture and thus generation family.]

[fact]So is the next generation, it's variant of the last generation they just do die shrinks and change the instructions and Cache that is architecture change [fact].What do you think instructions, Cache and die shrink are, maybe there just playing around. so you can't base generation on what you said above, remember intels tick tock.

[Fact] A new generation is a new product for sale by Intel that supersedes the last product they sold,tick tock.

And they had allot of new cpu's that superseded the old cpu's for the 775 board from 2006-2010 than the LGA1366 gets just 1 i7-980X tick tock.

 

[petteyg359]

Some of you guys are really confusing what constitutes a generation.
A die shrink does not
A Core variety/variant does not.

[fact]The underlining architecture of the core is what defines a generation[/fact]

It does not matter what ever you are defining, if its anything other than an architectural change, it is not a different generation.
It doesn't matter if you add cache/Cores.
Shrink the die, or even add new instructions.
That CPU will still remain a variant of that said architecture and thus generation family.

CPU Generations based on the the x86 architecture are as follows

1st Gen: 8086/8088
2nd Gen: 80286
3rd Gen: 80386
4th Gen: 80486
5th Gen: P5/P54/P54C/P55C
6th Gen: P2/P Pro/P3
7th Gen: P68 (Pentium 4)
7/8th Gen: Yonah (Core)
8th Gen: Core (Core2)
9th Gen: Nehalem (Core i 3/5/7/9
10th Gen: Sandy Bridge

I may have missed on or two derivatives in their, but those are the only generations of CPU's that exist.

@wingman99
The Gigabyte GA-945PL-DS3 (rev. 2.0) linked is a board from LATE 2006.
You would be lucky to have found it in stores Q4 06. It will have its 4th Birthday in Nov 2010
You may count 6 generations of CPU's but it is in fact 2 generations that this board supports.

@petteyg359
Like i said above a die shrink does not constitute a new generation nor do any of the small variants of the core.

By that logic, you need to remove 8th gen from that list, because it is a "variant" of Pentium 3.

 

[wingman99]

By that logic, you need to remove 8th gen from that list, because it is a "variant" of Pentium 3.

There all variants, they just change a little of this, little of that add this add that, more instructions change the number of pipes, registers, Cache, add a new Cache new instructions,Better Branch Prediction, die shrink. My point is they always build from the past and some times they use the past instructions because it works better with the new die shrink, like the p2- p3 instructions used for Core 2 also HT from p4 on the i3,5,7, series with AMD's on-die memory controller.

 

[JaY_III]

Yes I pointed that out above the architecture changees
[fact]So is the next generation, it's variant of the last generation they just do die shrinks and change the instructions and Cache that is architecture change [fact].What do you think instructions, Cache and die shrink are, maybe there just playing around. so you can't base generation on what you said above, remember intels tick tock.

That is completely wrong. You are not pointing out architect changes at all.
The P5 was a revolutionary change over the 486, as was the P5 over the 486 as was the P6 over the P5, ect, ect.

The only one i might somewhat agree with you was the 386 to 486.
This was basically adding the 80387 onto the 80386 die and wholla you have a 486. That is also completely ignoring that multipliers made their 1st appearance at this point in time.

Anyhow, lets not jump back to far and talk about the P68, once again it is nothing at all like the P6. I would call them the differance between night and day actually. With one favoring a short pipeline with a high IPC, witch subsequently prevents a high clock rate. The other, a long pipeline with a low IPC allowing the clock rate to skyrocket.

However this is were things start to confuse people as things start to blur after P68 as the P68 was far to ambitious. Power and leakage got way out of control and we never did see the 10Ghz promised due to this fact.
they took a step back and moved back towards the P6 architecture as it was VERY power efficient and powerful per watt.

So we ended up with a sort of hybrid fusion of the two techs to get the best of both worlds. Intel was building on that until its finally moved to Nehalem.

[Fact] A new generation is a new product for sale by Intel that supersedes the last product they sold,tick tock.

No its not.
Best example is Pentium Pro, Pentium 2 and Pentium III
They are all the same generation.

I have stated above why you can not include Core into the P6 gen.

Lastly, tick tock is a new concept at Intel that was announced in 2007 to state they were moving to a new generation of processors. AKA the move to Nehalem. As such it doesnt apply to anything other than Nehalem and newer CPU's.

By that logic, you need to remove 8th gen from that list, because it is a "variant" of Pentium 3.

that is somewhat true but more so for 7/8th gen
As its basically combining the lessons learned from P68 with P6.

 

[wingman99]

That is completely wrong. You are not pointing out architect changes at all.
The P5 was a revolutionary change over the 486, as was the P5 over the 486 as was the P6 over the P5, ect, ect.

The only one i might somewhat agree with you was the 386 to 486.
This was basically adding the 80387 onto the 80386 die and wholla you have a 486. That is also completely ignoring that multipliers made their 1st appearance at this point in time.

Anyhow, lets not jump back to far and talk about the P68, once again it is nothing at all like the P6. I would call them the differance between night and day actually. With one favoring a short pipeline with a high IPC, witch subsequently prevents a high clock rate. The other, a long pipeline with a low IPC allowing the clock rate to skyrocket.

However this is were things start to confuse people as things start to blur after P68 as the P68 was far to ambitious. Power and leakage got way out of control and we never did see the 10Ghz promised due to this fact.
they took a step back and moved back towards the P6 architecture as it was VERY power efficient and powerful per watt.

So we ended up with a sort of hybrid fusion of the two techs to get the best of both worlds. Intel was building on that until its finally moved to Nehalem.

It's not about how much progress intel make when they sell a new cpu, it's just a new generation from the last design there all variants from intel's CPU's build from there last architecture, you have it completely wrong, it's about selling and designing CPU's and each new one that supersede the old is a generation change you just cant pick and chose what cpu is a genration change becase of progress or the lack there of.

All cpu's are variants, they just change a little of this, little of that add this add that, more instructions change the number of pipes, registers, Cache, add a new Cache new instructions,Better Branch Prediction, die shrink. My point is they always build from the past and some times they use the past instructions and design and code also transistors because it works better with the new die shrink, like some of the p2- p3 architecture used for Core 2 also HT from p4 on the i3,5,7, series with AMD's on-die memory controller.

No its not.
Best example is Pentium Pro, Pentium 2 and Pentium III
They are all the same generation..

They are two generations didn't you watch the advertizing back then, there was a architecture changes and it was great because I had all 3 p1 p2 p3.

Lastly, tick tock is a new concept at Intel that was announced in 2007 to state they were moving to a new generation of processors. AKA the move to Nehalem. As such it doesnt apply to anything other than Nehalem and newer CPU's.

Intel has always been in a cycle for new product's yearly, although it has slowed down, tic tock is just slang, it has been around from the begging.

example:Intel Next Generation CPU Technology - Penryn and Nehalem http://www.pcper.com/article.php?aid=382

http://www.cpu-world.com/CPUs/Pentium-III/
New generation of Intel Pentium III processors with Coppermine core featured not only better manufacturing technology (0.18 micron as opposed to 0.25 micron for Katmai core), but also different cache architecture and new package type. Smaller size of the new core allowed Intel to put level 2 cache on the the same die as the core. While new cache was two times smaller than the L2 cache on Katmai processors, it was twice as fast and had a few additional performance improvements. Because external cache chips were no longer required for the new core, it became possible to put the core on smaller and cheaper Flip-Chip Pin Grid Array package.Pentium III microprocessor family was an evolutionary upgrade from Pentium II. The first Pentium III core, Katmai, featured SSE instruction set, which allowed SSE-enabled applications to process up to four single-precision floating point numbers at once. Other Pentium 3 cores added other features, like 256 and 512 KB on-die L2 cache memory and smaller package size. During its lifetime, the core of Pentium III microprocessors was shrunk twice - from 0.25 micron to 0.18 micron, and then to 0.13 micron.

http://www.google.com/search?num=10...cts=1271580166687&aq=f&aqi=&aql=&oq=&gs_rfai=

QUOTE:Intel® Core™2 processor family brings the benefits of second generation ... Formerly known by their codenames Conroe and Merom, the Intel® Core™2 processors

 

[EarthDog]

The tick tock refers to a brand new chip (for example Conroe), then its die shrink derivative/update with improvements sometimes (following the same example above, Penryn). Tick tock was around prior to nehalem.

 

[JaY_III]

They are two generations didn't you watch the advertizing back then, there was a architecture changes and it was great because I had all 3 p1 p2 p3.

You are now describing Marketing.
P2 and P3 are in fact the same generation
P1 is an earlier gen than p2/p3


Show me one creditable web link that shows P2 and P3 to be a different generation of CPU. I do no not think you can as they are both p68 cpu's.

I would also link to see some links showing how any other of these CPU's you have listed are a new generation.



@EarthDog
Tick tock was 1st coined by Intel in 2007 as i mentioned before.
1 year prior to Nehalem.
So i guess you could take that as intel stating Penryn / Wolfdale are it for core 2 if you want.

 

[Archer0915]

5th Gen: P5/P54/P54C/P55C
6th Gen: P2/P Pro/P3
7th Gen: P68 (Pentium 4)

HMMM??? can anyone tell me how many sockets we had here?

Socket 5 and 7 for the original PI, 1 for PPro, 2 for the PII-III (slot, socket), 2 for the P4 (unless you count the 775 Pentium) and now we have the 1156 and 1366.

The way I see it there is nothing wrong with it from a business stance. Most customers do not upgrade the CPU. I mean the vast and overwhelming majority! We (those who care about this subject) are such a tint fraction of sales that we are moot.

AMD has a different approach and in a way I think it could be hurting them. Personally I look @ where the money comes from and the BULK IS OEM!!!! Who really cares about the enthusiast? The aftermarket parts sector, GFX card companies, memory, HDD,............ So AMD may be making a mistake by sticking with compatibility.

Sorry but the change is good business.

 

[petteyg359]

AMD has a different approach and in a way I think it could be hurting them. Personally I look @ where the money comes from and the BULK IS OEM!!!! Who really cares about the enthusiast? The aftermarket parts sector, GFX card companies, memory, HDD,............ So AMD may be making a mistake by sticking with compatibility.

Sorry but the change is good business.

I disagree. Hypothetical situation: If you own a computer shop, and you've just ordered (in bulk, because it's cheapest, and therefore most profitable in the end) 1000 socket 775 motherboards. You've also ordered 1000 Q9400 CPUs. Most of your client base is somewhat computer literate, and they always want the latest technology, if not the best. They are aware that Intel has just released i7. Now, you're stuck with 1000 boards and CPUs that your customer base doesn't want. If you had gone with AM2+ boards and Phenom 9650 CPUs, and your customers decide they want the latest Phenom II X4 CPU, you can just order a new batch of CPUs and stick them in the old boards. You're stuck with 1000 old CPUs, but you didn't have to write off the boards.

 

[Archer0915]

I disagree. Hypothetical situation: If you own a computer shop, and you've just ordered (in bulk, because it's cheapest, and therefore most profitable in the end) 1000 socket 775 motherboards. You've also ordered 1000 Q9400 CPUs. Most of your client base is somewhat computer literate, and they always want the latest technology, if not the best. They are aware that Intel has just released i7. Now, you're stuck with 1000 boards and CPUs that your customer base doesn't want. If you had gone with AM2+ boards and Phenom 9650 CPUs, and your customers decide they want the latest Phenom II X4 CPU, you can just order a new batch of CPUs and stick them in the old boards. You're stuck with 1000 old CPUs, but you didn't have to write off the boards.

You know as well as I do that things run concurrent for a while. A good business man can read the market and knows previous sales numbers. If someone puts themselves in that situation in this market then they just get burned. There is plenty of forewarning in this market so your point is moot.

I think just HP/Compaq and Dell makes up for more CPU/MB sales than all of the non OEM sector

 

[Audioaficionado]

They're gonna still be selling dirt cheap LGA775 socket Pentiums and low end C2D systems for quite some time to come to most of the Joe Sixpacks and Joe Suits.

It's mainly the few percentage fraction of very demanding geeks and high performance professionals that make up the market for the later sockets.

I'm still on my LGA775 P965 chipset & 65nm Q6600 system and it runs a lot of high hardware demanding software just fine.

I will upgrade to the latest and greatest eventually, but I don't see the need until after the 32nm stuff has been out for long enough to work out any major bugs.