Intel 3.40EE & 3.60E - LGA Arrives

MBR writes "MBReview has taken a quick look at Intel's new high-end LGA775 processors, the 3.40GHz Extreme Edition, and the 3.60GHz 'E,' now known as the 560. They've covered some of the questions about pin frailty of the new LGA socket, as well as cooling issues that might arise from these new processors." ("LGA" stands for Land Grid Array, which moves pins from the processor to the socket it sits in.) Update: 06/19 20:50 GMT by T : Reader Chi-Energy points out that besides the new processor packaging, Intel has also just released its i925X and i915 chipsets, PCI Express and DDR2 DRAM for the desktop, and links to this review showcase with benchmarks at HotHardware.

Comments

New pins

[Deltawolf]

Oh wow! Now if your pins snap you have to replace your mobo instead of your processor. Sounds like its begging for trouble.

Re:New pins

[Naffer]

Most home motherboards are cheaper then the processor. Motherboards run around $150 while newer processers run above $200

Re:New pins

[The Living Fractal]

And...

Which costs more?

'nuff said.

Re:New pins

[Anonymous Coward]

So what?

Motherboards stopped being more expensive than processors a long time ago.

Re:New pins

[Anonymous Coward]

I'd rather replace a CPU with a broken pin than tear apart my case and pull out the tray to replace a motherboard.

A CPU can be replaced in just a couple of minutes. A motherboard would take much longer, depending on your case type, how many cards you have, and all the various types of things you're going to have to unplug from it and plug back in.

Re:New pins

[slasher999]

Saw these up close a month or so ago at the ICC. So long as you don't go sticking your fingers in the socket to "check out" the newness, you shouldn't have a problem bending pins. Moving the pins off the CPU and using "pads" in their place should make CPU installation a bit easier, although I neve had a problem with the current layout.

Re:New pins

[Oculus Habent]

Why don't they make a pin grid insert? That way you can just replace the pin set if you bend some, and both the motherboard and th processor are immune from that type of damage.

The only problem you might experience with this is if you break off pins in the socket or the processor, but it seems like a (mostly) unlikely situation.

Re:New pins

[challahc]

But how would that connect? More pins?

Re:New pins

[Junta]

Even in the event of pin breakoff, tweezers would work... A pin insert is a good idea and is used elsewhere in the industry, so it's not exactly new thinking, so I wonder why this LGA came up as intel's 'solution'.

Re:New pins

[The Living Fractal]

Oh stop whining,

Replacing a mobo is a small task, I did it to a few PowerEdge servers yesterday, took about 20 minutes for each one, and there's a lot more on one of those (at least these customized ones I am working with) than on your Compaq Presario.

I could show you a price/time breakdown and explain that by replacing the mobo you are effectively probably making over $100/hr, but then you'd probably reply with some inane comment about how you have so much money that you want to make a mobo out of $100 bills.

Re:New pins

[Anonymous Coward]

I hope you enjoy pulling PCI cards out, cables, and memory and then getting a new mother board for which you will have to repeat the process all over again to assemble it.

How exactly is this new pin system better for us?

Re:New pins

[Anonymous Coward]

I'm not familar with your server, but most servers, especially the rackmount and high end stuff are very modular and alot of times have risers/modules just for CPU's. Thats easy to replace and would never be an issue.

However, for the average PC (or the unaverage build-your-own PC), you have one motherboard with lots of stuff (PCI cards, memory, cable assemblies, and jumpers in some cases still) to contend with.

Especially for those of us in the PC repair business that deal with 10-30 PC's a day, this type of CPU setup being standard would cost me significantly in time if I were replacing or upgrading a CPU.

Money is in the volume, not in the time in the PC repair business. A PC that spends 3 days in the shop does not bring in significantly more money than the PC that took 10 minutes, 24 hours, or 2 days even.

Re:New pins

[toasted_calamari]

I was thinking something along the lines of no pins at all, rather just use flat contacts. there would be a clip that would hold processor in place on motherboard.

(or maybe that wouldn't work, i'm not a processor engineer.)

Re:New pins

[The Living Fractal]

On every PC I've ever worked I can pull a mobo and replace components on it within 5-20 minutes. It's really not that difficult.

Yes, it might end up worse for a PC repair shop, but even then it seems like there would be ways to structure pricing accordingly and, maybe, even make a little bit more money this way.

That is, unless the shop is already charging too much for what is basically a 2 minute job, which isn't something I think belong in business regardless.

Personally, if I ran a shop, and a PC needed a new processor, and the customer brought the PC in and I had the processor on shelf, I would tell them to wait a minute while I changed out the processor, and only charge them for the processor's cost (and my normal percentage markup)

In the case of a mobo change, I would be justified in getting a bit of a labor charge in there...

But this is all assuming a customer was the one who bent the pin, which makes no sense because they're bringing it in to shop afterwards.. so what were they even doing removing their CPU in the first place if they were just going to be bringing it into a shop ?

Hmmm

Re:New pins

[Anonymous Coward]

Sun used to do that for some of their UltraSPARC processors (maybe they still do - haven't touched a new SPARC box in ages)

Worked ok but you had to be real careful installing it to make sure it went on straight -- the processor had several TORX bolts going through it that you had to tighten in a star pattern (just like installing a car tire) and you were supposed to use a torque wrench to make sure they all were the same (although I always just judged it by feel and it went OK)

I don't think a simple "clip" would hold the processor straight enough (especially given all the vibrations caused by the fan and also just the machine being moved around) to make sure the pads stayed lined up year after year -- so you'd have to go to a similar multiple-bolt system which would probably make it too expensive for mass-market CPUs

Re:New pins

[bhtooefr]

http://cpu-museum.de/forum/viewtopic.php?t=927&pos tdays=0&postorder=asc&start=15

The UltraSPARC II used something like what you refer to.

Re:New pins

[Synkronos]

Yes, it would basically be a widget with pins on both side, one plugging into the mobo and the other into the chip. Then, if you break a pin off, you just replace the cheap adapter. It also means chips would be less susceptible to static shock, since the contacts would not be as exposed.

Re:New pins

[Naffer]

I would think that an array of pins in on a given area on the underside of a chip would have more electrical contact surface area then flat contacts in that same area.

Re:New pins

[Bishop]

A pin insert would connect to pads on the mainboard and pads on the cpu. I worked with some SGI systems that used this kind of setup. It is a good idea.

I am a little disapointed that Intel did not go with a pin insert. However it would have cost more which would have been hard to justify to the mainboard makers and their razor thin margins. In the long run I think that a pin insert would have been a smart move. Judgeing by the reported fragility of the socket 775 I won't be surprised if Intel moves to a new socket as early as next year. I am sure we will see an outcry from the (wanabee) enthusiast crowd due to the shear number of socket 775 mainboards that they will trash. A pin insert may have eased that problem.

Pin inserts could even have been a lucrative revenue stream. It would not take much to convince the enthusiasts that a new pin insert was needed whenever a cpu was upgraded to insure maximum performance. In some respects computer enthusiasts are as bad as audiophiles.

Re:New pins

[Anonymous Coward]

what a rambling load of drivel.

what the hell are you talking about? the poster didnt say it was 'difficult' he suggested that it was trivial to swap a processor and sometimes its one big pain in the ass to swap a complete motherboard including memory cables cards and all the other requirements.

you obviously disagee?
what was all that gibberish in the last paragraph about?

Re:New pins

[JPriest]

A few points: $150 is reasonable for a motherboard, the above listed processors are likely to run closer to $450 and $900.

If 6 minutes of your time is worth $300 - $750 then you obviously make way more than I do.

AMD is going to start using the same technology. When Intel does it, it is a pain in the ass, when AMD does it, it's innovation.

Besides, you have to be pretty careless with your hardware to break a pin.

Re:New pins

[Anonymous Coward]

Within minutes? That depends..
I've spent hours trying to install an overgrown heatsink. It was naturally equipped with some über-strong metal thingy that's supposed to keep it in place nicely coupled with a motherboard feature of having two very loose condensators literally touching it. Careful use of several screwdrivers made it finally possible though.

Re:New pins

[Anonymous Coward]

Yeah, let's all worry about a problem that would affect 1 in 10000 people.

Re:New pins

[Too Much Noise]

There are some points you're missing about this.

Who's paying for the RMA? if the natural life span of the pins is about 8 insertions (as the mobo producers seem to claim), then there would be a large number of legitimate breakages that get sent back to the mobo manufacturer. Now, they can either replace the CPU socket (not very funny, I think) or throw away the whole mobo, including the rest of the perfectly good components on it[*]. As oppose do just discarding a defective CPU if its pins break.

[*]like the spankin' new and expensive Intel chipsets. I doubt $150 will happen anytime soon as a mobo price, as even the chipset estimated price seems to be above that. I also doubt mobo manufacturers getting too many returns due to bent socket pins will be very happy about all this - remember, their margins are quite slim these days. The least hurt by this is probably going to be Intel itself.

Your AMD jab is a troll. As far as they stated so far, the Opteron socket stays put for the foreseeable future (meaning at least one year). They will have no incentive to move to a pinless package unless it shows some solid advantage. Even Intel might have to back down on this if the hw producers get to unhappy (and they already have enough grief with the BTX form factor).

Finally - pins break. It's called mechanical stress. How many times do you think you can 'carefully' insert and remove a CPU in its socket before some pin gives in? At least, for the old sockets, all you had to do is match pins and holes ; now, with only point contacts, bending can come so much easier.

Re:New pins

[Fweeky]

"It would not take much to convince the enthusiasts that a new pin insert was needed whenever a cpu was upgraded to insure maximum performance"

Well, it would take some fairly repeatable benchmarks, at least. Maybe you could sell "higher quality" ones with gold plated pins and a heatspreader or something. Of course, you could always package them with the CPU...

"In some respects computer enthusiasts are as bad as audiophiles."

I resent that; I haven't spent more than a couple of quid on IDE and VGA cables! ;)

Re:New pins

[Anonymous Coward]

Just wait till some genius mobo manufacturer gets the extremely complicated idea of making a) the pins shorter, a much less breakable length, b) thicker and c) possibly conical so they're even less breakable

From what I gather, LGA's only advantage is that it allows higher signal speed...though this is so far unnecessary, as Intel's QDR has shown it can go up to 1200Mhz without sweating...AMD's hypetransport can hit 1600Mhz per pin and is being planned for 2000Mhz...that'll be 4GBPS on a 16 bit wide pipe in each direction...make a 128bit memory controller out of it and you're looking at 32GBps total

Nevertheless, wake me when we get quad core, quad threaded, low power fanless computers that can fit in the palm of my hand...

Re:New pins

[Pieroxy]

How many times do you think you can 'carefully' insert and remove a CPU in its socket before some pin gives in?

I don't know about you, but for my part I have never removed and inserted a CPU into its socket more than a couple of time. And remember that 99% of their revenue doesn't come from people that do that more than once.

So again, don't generalize on the "Geek point of view". They don't give a sh*t if your CPU/mobo breaks

Re:New pins

[CaptnMArk]

Some might like a water cooled one :)

Re:New pins

[ModMeFlamebait]

or throw away the whole mobo, including the rest of the perfectly good components on it[*].

So I guess it's time for chipsets in sockets, too.

Re:New pins

[Fweeky]

I'd imagine the pin lengths are quite important at high speeds, and routing water around all those pins in a tiny little package, under a hsf...

This line of thinking's got me wondering about cooling both sides of a processor... maybe if you made one with the pins at one corner, like a little slot, pack heatsinks to both sides of it; with the right mounting gear and processor package you could effectively double your cooling area. Maybe one way of dealing with the heat of dual/quad core chips and lower surface areas as they get denser :)

Patent! ;)

Re:New pins

[swankypimp]

As the reviewer pointed out, Socket 478 CPUs tend to stick to the heatsink and get yanked out during upgrades. Having built and repaired hundreds of machines in the past year, I have to tell you that I am still a bit scared to change my own machine's aging 865 chipset motherboard. For a business a few dead CPUs due to bent pins is no biggie, but for an individual who doesn't have another $300 it's more than an annoyance.

The advantage of the new setup is that the CPU is locked in place better, so as long as you are reasonably careful it won't come out and bend anything. Also, in 12 months when the chips are still a couple hundred bucks each, I'm sure ECS and others will have $60 value boards on the market. Sure they won't be the baddest mofos on the planet, but it's a nice safety blanket for geeks on a budget.

Re:New pins

[SomeGuyFromCA]

> Socket 478 CPUs tend to stick to the heatsink and get yanked out during upgrades

Seen it happen twice, once with a p4 and once with a p/166.

Both times, the chip was just fine afterwards.

What should we be concerned about?

[irokitt]

I'm not as worried about frail pins as I am about the amount of heat these things push out, the size of the new heatsink/fan assemblies, and the noise they put out. I thought Socket 478 processors were hot, but LGA Prescott processors run even hotter, which makes me think Intel has a point when it says we should switch away from the ATX case factor and adopt BTX for Intel chips..

That said, are the Extreme Edition processors still selling for $900 USD a pop? Hardly seems worth the extra money for gaming, although a server that wants to survive Slashdottings could probably use one...

Re:What should we be concerned about?

[Anonymous Coward]

A server that wants to survive slashdottings would be better served by quad opertons. Not to mention being more cost effective.

Re:What should we be concerned about?

[bhtooefr]

A server that wants to survive slashdottings would be better served by static pages and a large pipe. A P75 with 48MB RAM (kplug.org) survived a /.ing, after all.

Re:What should we be concerned about?

[Anonymous Coward]

Hardly seems worth the extra money for gaming, although a server that wants to survive Slashdottings could probably use one...

The stuff added to the EE wouldn't be all that useful for a server.

But just wait until you see the 4.60GHz Pentium EEEEEE!

Re:What should we be concerned about?

[Garak]

Looks like there should be better heat transfer between the chip and the heatsink with this new package. It looks like the heatsink with make contact with the metal on top which almost covers the entire chip.

The price ofcourse will go down, $900 is about right for a new processor. Give it a few months.

I just wish they had better benchmarks, They just compare the new processors to other intel processors.

Re:What should we be concerned about?

[bhtooefr]

What, the cache? Umm... the EE is essentially a Xeon MP on Socket 478 (and now T) with standard desktop CPU clock speeds. The cache IS of benefit on the server. Something tells me anything that is RAM/disk intensive would benefit from it, as long as it's working in a small area of that data.

Is not

[Milo of Kroton]

Faster importantless? With 64 bit processing power for all available by athlon 64 made available that works with 64 bit yes immdiately we would switch.

These chip make futiliity. Why make processors of like these new when you can improve on 64 bit? The battle is to will be lost to Athlon without 64 bit competition by.

Re:Is not

[irokitt]

Is what country you are from?

Re:Is not

[The Living Fractal]

ROFL... I think you might have a point, but Babelfish made me laugh my ass off too much to catch it.

OTOH, I suspect the babelfish conversion of this reply will probably leave you in about the same boat as me.

Laugh on, my brother.

Re:Is not

[Anonymous Coward]

Is what country you are from?

What you do mean? They are from country the same you as.

Re:Is not

[NanoGator]

"These chip make futiliity. Why make processors of like these new when you can improve on 64 bit? The battle is to will be lost to Athlon without 64 bit competition by."

Are you running 64-bit apps I'm not aware of?

Re:Is not

[drinkypoo]

Guess what? The Hammer core executes 32 bit code faster than a P4 does, clock for clock, and it has all the same MMX (and similar) extensions including SSE2. Meanwhile, a few games are being built as 64 bit and by the time we can actually get an x86-64 Windows, there will be more of them. From what I hear, just recompiling your code for x86-64 rather than x86 nets a ~15% performance increase, and processor-specific optimization ought to net still more. Hence, especially given a price comparison, Hammer is the clear performance winner.

Also, you do have the option to run something other than Windows. There are linux distributions supporting x86-64 today.

Re:Is not

[Lost Engineer]

In other news... Babelfish reduces language barrier to language annoyance - Slashdotters still don't understand one another

Re:Is not

[Anonymous Coward]

Why bother waiting for x86-64 Windows when you can have 64-bit linux right now?

Re:Is not

[bobthemonkey13]

Not only does AMD have the only desktop 64-bit offering right now, but their chips are much faster than Intel's at the same clockspeed, even in 32-bit mode. Whereas Intel's engineers are just running their chips at insane clockspeeds, AMD's are actually designing better processors. For the price of a 3.4GHz "800"MHz FSB P4EE ($989 on pricewatch right now), you could buy two Opteron 246s ($441 each) with cash to spare. If you want to talk raw, meaningless numbers, the Opterons still beat the P4EE (4GHz and 2MB cache total). Of course, SMP isn't simply additive like that, but consider the advantages of 64-bit and multiprocessing, and the fact that AMD chips are /much/ faster than Intel's at the same clockspeed (even on 32-bit code), and there's no contest. All halfway-modern Windows versions and Linux kernels can support SMP, and the latest support amd64, too.

Re:Is not

[Bombcar]

[insert mandatory G5 is 64 bit and I'm going to mention that even though it is obvious that you're talking about x86 chips]

Re:Is not

[Anonymous Coward]

He's not using babelfish - well not as needed. From June 9 Milo went from perfect english to pretending to be German using babelfish - check his post history. He is now a troll account.

Re:Is not

[Anonymous Coward]

So called 64-bit processors* don't accelerate existing applications. IT people know all about 'installed base'.

*All P4s and Athlons can work with 64-bit integers. '64-bit processors' only get you a larger memory space of which milions of existing programs do not need.

Re:Is not

[Anonymous Coward]

A higher clockrate is ALWAYS better from a performance standpoint. ALWAYS. ALWAYS. ALWAYS. If you know anything about synchronous logic design you would know there is no debate about this.
AMD chooses to reduce the chip area/power deticated to the clock tree to add more logic. This is a design tradoff they had to make and it not the mark of a better design. Getting CMOS devices to switch at sub 300ps times is an amazing feat and if AMD could design thier clock system to run at those speeds they would.

The main problem with your analysis is that there exist algorithms that mathmatically CANNOT be solved in parallel, making SMP, hyperthreading, clusters all useless.

Oh, and 64-bit only buys you a larger memory space.

Re:Is not

[vadim_t]

Bah. Clock rate means very little.

I'm fairly sure that whatever CPU you have it'd have to work pretty hard to beat a 1 GHz Nehemiah CPU at doing AES encryption. The reason is that somebody had a great idea to integrate it into the CPU, so it can encrypt one block per clock cycle.

Pure brute force of course helps, but what's better? A 1 GHz CPU that can encrypt a block in a clock cycle, using somewhere about 20W for the motherboard and CPU, or a 4 GHz CPU that needs 200 clock cycles to do it in software, and needs more than 100W for just the CPU?

Re:Is not

[pdbaby]

The point of the P4EE is not to be a real competitor against the Opteron, but rather to show just how easy it is for Intel to play in AMD's back garden. Intel can also validly compare P4EE to the Opteron if they want.

It's actually cheaper and faster to run a dual Intel Xeon computer than it is to run a P4EE.

AMD chips are /much/ faster than Intel's at the same clockspeed

But they aren't running and can't run at the same clockspeeds. I suspect that without Marginally Extreme Cooling, AMD's chips would sizzle at Intel clockspeeds
There are no benefits to existing 64-bit hybrid-as-a-whore x86 processors at the minute except more registers (which could be slapped on without 64bit) and more memory space. Now, if only they would create a clean 64-bit from-scratch implementation that removes the hideousness of existing x86 machine code...

Re:Is not

[Alexis de Torquemada]

A higher clockrate is ALWAYS better from a performance standpoint. ALWAYS. ALWAYS. ALWAYS. If you know anything about synchronous logic design you would know there is no debate about this.

True, provided that you are comparing processors with identical design that only differ in clockrate. But of course this is by far not the case, the P4 and Athlon 64 are implemented in fundamentally different ways. For example, in order to achieve the high clockrates with which they want to market their products to the uninformed (obviously), the Intel guys have increased the pipeline length beyond good and bad, with the consequence that mispredictions for out-of-order execution cost some real time. HyperThreading was introduced as sort of a hack for reducing the negative effect of their long pipelines, at least for multi-threaded applications. Running only a single thread, the P4 just has trouble keeping its functional units busy.

The speed of a processor is not measured in GHz. It's measured by the amount of work it gets done in one second. This depends on the application, but it's no secret that AMD CPUs perform substantially more work per processor cycle than Intel CPUs. E.g. my Ahlon XP 2400+ operates at "only" 2GHz. However, I took a the results from comparative Benchmark tests from the German computer magazine c't, and averaged (over all tests) the clockspeed that a Pentium 4 would need in order to be as fast as the Athlon. The result was 2800MHz, so the Athlon XP is on the average 40% faster than a Pentium 4 operating at the same speed. In other words, clockrate isn't everything.

The main problem with your analysis is that there exist algorithms that mathmatically CANNOT be solved in parallel, making SMP, hyperthreading, clusters all useless.

Actually, that's a good argument against Intel's hyperthreading, though there's a problem with it anyway: In practice, the question is not "Is this problem serial or parallelizable?", but how well it can be parallelized. For example, going from 1 to 8 CPUs may allow you to speed up computation of a certain problem by factor 7, however going from 128 to 512 CPUs may give you a speed increase of only 3%, because the communication and syncrhonization overhead becomes the bottleneck.

Oh, and 64-bit only buys you a larger memory space.

First of all, this "only" is misleading since even desktop machines will soon reach the 4GB boundary (actually, the 4GB limit virtual memory, which is often required in substantially larger quantities than physical RAM). You can use PAE for up to 64 gigs, but it's a performance killer.

And second, this is not true. AMD64 allows you to use wider adresses as well as wider integers, and this is a great boon for certain types of application, most notably cryptography. I've seen a benchmark that showed an 2GHz Athlon 64 outperform a P4EE 3.4GHz by factor two in AES encryption. Obviously, 64-bit integer operations benefit AES greatly. On 32-bit machines, they have to be split up into sub-operations - e.g. a 64-bit multiplication (discarding the upper 64 bits of the result) requires 3 32-bit multiplications plus several additions. For comparison, the Athlon 64 requires 3 clock cycles for a 32-bit multiplication, but only 4 for a 64-bit multiplication! Compare this to about 11 or more cycles the CPU would have spent on an equivalent sequence of 32-bit operations, which also would have increased code size (more cache misses) and forced you to use more of the already scarce registers (AMD64 doubles the size and number of the general purpose registers, some of which aren't even that general-purpose...).

Re:Dear Apple...

[Anonymous Coward]

Hmm... somebody's woefully informed...

WOW worst grammer ever award

[gnuLNX]

Holly molly bat man...were did you learn english...I hope you understand it better than you write it!

Re:WOW worst grammer ever award

[The Living Fractal]

Holy shit man, where did you learn how to read?

He's using Babelfish...

Re:WOW worst grammer ever award

[SkankhodBeeblebrox]

Talk about the pot and the f'in kettle... no less than 3 spelling errors and a missing question mark...

BTX (which includes LGA) standard resisted

[Synkronos]

Apparently the BTX form factor (of which LGA is a part) has been heavily resisted by many Taiwanese chassis, mainboard and heatsink manufacturers.

But what's new here? Word has it that this time round, the Taiwanese heastink, mainboard and PSU manufacturers - and quite a lot of them it would seem - are being rather less than enthusiastic or co-operative, about the sweeping changes and support that Intel is asking, nay demanding, of them.

I'd be interested to see if Intel can actually strong-arm them into it

Re:BTX (which includes LGA) standard resisted

[foidulus]

I'd be interested to see if Intel can actually strong-arm them into it
With mainland facilities becoming more and more advanced, but without huge increases in cost, I think Intel can get their way. All they need to do is say, "Shanghai" and I bet the Taiwanese manufacturers will change their minds.

Re:BTX (which includes LGA) standard resisted

[irokitt]

Since AMD chipset and motherboard makers are NOT planning to switch over to the BTX form factor anytime soon, it would be awkward for Intel motherboard and case manufacturers to do so, since it would introduce two completely different standards. It's also difficult because BTX focuses on the "desktop" form factor, where the case is a small box you can place under your monitor. I prefer the "server" form factor, a seperate tower that can be placed anywhere. There is a server form factor for BTX, but none of the prototype motherboards I've seen are made for it, and Intel seems to be pushing the smaller form factors (probably with OEM manufacturers like Dell and HP in mind).

Frankly, I like the idea of sticking with ATX and placing PCI Express and LGA sockets on an ATX motherboard, so I can use the ATX tower I have now. I just wish the ATX form factor could be made to accomodate hotter processors easily.

Re:BTX (which includes LGA) standard resisted

[Anonymous Coward]

With mainland facilities becoming more and more advanced, but without huge increases in cost, I think Intel can get their way.

I wouldn't be surprised if most of those mainland facilities are owned by the Taiwanese.

Re:BTX (which includes LGA) standard resisted

[Anonymous Coward]

I very much doubt this. Cite?

Re:BTX (which includes LGA) standard resisted

[Malc]

What do you mean by "server" form-factor? All of our servers are rack-mounted.

Re:BTX (which includes LGA) standard resisted

[iansmith]

I can understand why they are resisting.

Intel has decided that the motherboard/case manufactuers need to shoulder the cost of the cooling required by the newest Intel chips.

This of course is not going over well. Computers are such a mainstream industry that the profit margins are very low and Intel is basicly trying to shift some of the cost away and increase their profits at the expense of another part of the industry.

I only hope they get taught a lesson like IBM did with Microchannel. But I'm not hopefull. The changeover will be slower.. but it will happen.

Re:BTX (which includes LGA) standard resisted

[irokitt]

If you look at advertisements for ATX towers you often see them advertised as "server towers". The ATX standard was created before rack-mounted servers really caught on. A used quotation marks 'cause I don't like the term, I just call 'em towers, but to each his own.

Re:BTX (which includes LGA) standard resisted

[Malc]

Every computer I've had for the last 15 years has been a tower. I prefer my computer out of the way under the desk where I can't hear it so much. The extra box on the desk just annoys me.

Re:BTX (which includes LGA) standard resisted

[Wesley Felter]

BTX isn't needed today, which is why the manufacturers are complaining. But when the processors get up into the 150W range, they may find that BTX systems are either cheaper at constant dB or quieter at constant cost.

Re:BTX (which includes LGA) standard resisted

[Jeff DeMaagd]

I don't blame the resistance. Most, if not all of the "unique" changes that are part of the BTX chassis spec can be adapted to the ATX bolt patterns and other parts of the spec.

The fan duct? Ha. There are a few aftermarket mods for putting fan ducts into an ATX case. Compaq and Dell use fan ducting too, in some models.

Small form factors? There already exist micro ATX and even NLX form factors. OK, the PCI-E video cards do get better cooling under BTX, but there's nothing there that says that an ATX case can't be adapted such that the video card can't be relocated or the card cage given a duct of its own.

Re:BTX (which includes LGA) standard resisted

[Jeff DeMaagd]

I can't say much else, but the Taiwanese drive company Lite-On uses Taiwanese engineers for design, and their Chinese plants for manufacturing.

I'm not sure the Chinese-ownd Chinese plants have much incentive either, I think it is futile on the part of Intel.

Re:BTX (which includes LGA) standard resisted

[irokitt]

Yup, I've always used towers. Again, that's the reason BTX doesn't interest me-unless they offer towers, I'll just stick to ATX and buy AMD processors. More bang for my buck.

More info on the BTX form factor

[Anonymous Coward]

http://www.formfactors.org/developer/specs/BTX_Spe c_1_0.pdf

Re:Land Grid Array

[Anonymous Coward]

Dumb moderator. How the fuck is this a troll?

Wow, look at the length of the 3.40 EE bar!

[Jeremy Erwin]

Oh, wait. It seems that none of those bar graphs include an origin. Never mind.

Re:Wow, look at the length of the 3.40 EE bar!

[ameoba]

Yeah..

nothing like a graph that makes a 10% difference look like a 90% gap.

Pictures of the socket

[athakur999]

Every review I've seen on these chips has pictures of the CPU and the socket with the CPU in it. I haven't seen one with some good pictures of an empty socket.

Anyone have any links to any? Does this new chip just rest on the pins or is there some more positive mounting method (besides that cover that goes over the CPU)?

Re:Pictures of the socket

[irokitt]

The article has some good ones, just click to enlarge them.

Re:Pictures of the socket

[Aphrika]

here's a couple of pictures.

For a more comprehensive overview of the whole BTX, DDR2, Socket 775 and PCI Express malarkey, I'd recommend having a look here. Interesting stuff.

Pins

[bsd4me]

At high frequencies, the pins on a package aren't really short circuits (ie, zero resistance); they have a capactiance and inductance which mess with the signals. Making removable pins would make this a lot worse.

About time for them to use LGA...

[DraconPern]

It's about time they catched up and started to use the LGA connection! The NEC VR10000 MIPS chip had LGA in 1998, as well as LGA contacts on the motherboard. To connect the LGA on the proc against the LGA on the board, a plastic holder with wads of springy gold wires was used. There was no issues with bent pins, etc. The only problem was lossing those wads of gold...

Watercooling

[xsupergr0verx]

This next incarnation of processors could speed up the innovation of watercooling. Just as alternative fuels will be researched quickly and furiously when conventional fuels become very expensive, when these new processors hit the desktop market (think Dell, HP, Compaq, etc.), they are probably going to be loud or obnoxious.

Maybe we are going to see better watercooling systems due to mainstream demand. That would be pretty cool. (NPI)

Re:Watercooling

[MojoStan]

This next incarnation of processors could speed up the innovation of watercooling.

I think it has already started. Did you know that Sony and Apple have both implemented watercooling in their newest systems? 'Looks like that transition to 90nm is a bitch. Here's the links:

Sony VAIO R Series "Features" Page (including "Advanced Liquid Cooling System")

Apple Power Mac G5 "Design" Page (including "liquid cooling system")

another crap review

[phrasebook]

I'm sick of reading reviews that compare new products with other new products. Example on MBReview: comparing P4s that are all pretty much brand new, all expensive, hardly any difference between them. I want to see how it stacks up against my P3-866, not another P4 that I've never even seen. At least throw an older proc in there for comparison. Same with video card reviews. I don't give a hoot how the Radeon 9600 compares with the 9500... how does it compare with my GF3? FFS these reviews suck. At least throw in an older chip just for a relevant comparison. And stop mentioning how Quake 3 is getting old but is still useful: "this benchmark is slowly progressing towards an archaic stage". STFU. Who keeps regurgitating this crap.

Re:another crap review

[qbwiz]

Shouldn't it be obvious how it stacks up against your P3? What would be the point of such a comparison - "Should I get a P3 866 or a P4 3600?" Some thing with the GF3 and Radeon 9600.

Quantitative comparisons between components with such disparate performances are pretty much meaningless - "I'll wait until a new graphics card has exactly 2.5 times the performance of my old to get it." With such new features as Pixel Shader 3.0 coming (even though games may not support it yet), the quality of the picture is also different. You can't quantify that.

A major use of these reviews is letting people discover how much more performance they would gain by paying the extra money for the top-of-the-line components as opposed to second or third best. Comparison to anything too low would be meaningless and pointless; it would just take up space.

A potential buyer should already have narrowed down what they were considering by that they look at reviews. The reviews for old cards/chips/etc. would help them decide what to get if they want the middle or low range, and if they want the high range they pick the new reviews.

Re:another crap review

[Sgs-Cruz]

He's not considering buying a new P3 or something. He's wondering if the new processors give the 2x (or 4x, or 10x, or 1.1x, or whatever...) benefit that would convince him to upgrade. You can't even use the old specs because the spec programs change every year in order to keep up with the latest featuers (at least with graphics spec programs).

look here

[io-waiter]

2004
http://www.aceshardware.com/read.jsp?id=60000301
2003
http://www.aceshardware.com/read.jsp?id=50000356

Not exactly what you are looking for but it will give you a clue how much performance an upgrade will give. Digging further back in the archive to say 2001 might be even better.

Still could be a paper release?

[Liquidrage]

That was my 1st question. And the review site mentions this as a possibility.

This is something Intel seems to be a master at. Releasing CPU's to review sites that you can't buy for a long time just to get the hype and "title" of the fastest. Other companies do it, just not as bad as Intel.

My personal opinion is /. shouldn't be participating in the hype-machine by promoting CPU's who we can't even be sure will ever be released as reviewed unless they are truely groundbreaking. And this isn't groundbreaking.

I don't hold it against MBR for submitting his own site. But this is not newsworthy. This is "Intel creates a really fast CPU we might one day be able to buy". Maybe as part as a CPU roundup story, but not as the story itself.

Best for use only in winter

[Sivar]

According to Sandpile.org, the 3.4GHz Pentium IV Prescott can use up to 127W, and has a typical power usage of 103W (when browsing the web or reading email).
In my opinion, it is rediculous for a single processor to single-handedly run up your power bill. That's like having two light bulbs on 24/7 (assuming you keep your computer on), not to mention the power needed to cool your PC, let alone your house's air conditioner.

I would take a VIA chip for low-performance stuff, and an Athlon64 for performance computing. support 64-bit software including 64-bit Linux distributions, are faster than Intel's best even running 32-bit software, and they have a maximum power usage of 89W. Because of Cool'n'Quiet mode, they spend most of the time running at 800MHz consuming about 30-35W and generally not requiring a loud and abnoxious cooling fan.

It is actually impressive what the chips can do at 800MHz. You can play a full screen DVD at 1400x1050, and the CPU usage tops out at about 5% (at 800MHz). If, of course, you run something that requires more power, like a video game or a compiler, the processor instantly switches to full speed. Handy, that.

Re:Best for use only in winter

[Anonymous Coward]

That 103W figure of yours is bullshit - at least how you've understood it. That's the typical maximum power consumption, or what Intel call the "Thermal Design Power". While web browsing or reading email, power consumption is down around 30W. You need difficult-to-design test vectors to push the CPU beyond the TDP, and unless you know what you're doing, you're unlikely to get more than a couple of watts beyond it.

One way to prove this to yourself is to simply remove the heatsink from a running Pentium IV Prescott system. Shock, horror: it will continue to run, only slowly. That's the same slowly the system silently goes into and out of depending on the CPU load (it takes about 10000 CPU cycles, or about 3 millionths of a second, to get into or out of this state.) Now: try web browsing or reading email while it's running like this. Tell the difference? Didn't think so. You might notice a _slight_ slowdown (when rendering a complex page, for example): that's because without the heatsink on, the CPU won't go back into the "normal" S0 power state.

So, just as a Pentium 4 doesn't dissipate 103W without a heatsink installed, so to does it not dissipate 103W if you're not doing anything.

This can all be found in http://developer.intel.com
Handy, that.

Re:Best for use only in winter

[Anonymous Coward]

Sorry, that should read "about 1000 CPU cycles"

Re:Best for use only in winter

[Tmack]

Im sorry, but 103 watts (even if thats correct information) is not much. Like you even said, its about the same as 2 50watt bulbs. Thats .103KW (KiloWatt), running 24hrs a day for a month is about 730Hrs, or 75KWH (KW-Hours). At the US avg electric rate of 7cents/KWH (whats used on most appliances) that ammounts to $5 a month. Your air conditioner on the other hand uses quite a bit more power (for a central unit). I dont think any P4 desktop machines require 2phase 30amp breakers with heavy guage power cable to operate. Your stove also produces many times the ammount of waste heat that your PC does. If you are so concerned about the heat it produces heating your house, get a watercooling system and put its radiator outside.

TM

Re:Best for use only in winter

[Sivar]

It looks like I misinterpreted the information that I read. Thankyou for setting me straight.

PARENT IS TROLL ACCOUNT

[Anonymous Coward]

Mod parent down. He is trolling.

He's a native English speaker (check post history prior to June 9) who changed his posting from then on to pretend to be German using Babelfish.

He is not. His previous posts (in prefect English) even state at one point he only did 4 years of German in high school.

Please mod him down, and ignore the troll.

Re:PARENT IS TROLL ACCOUNT

[Anonymous Coward]

If you checked his entire posting history, you fell for the troll even more than he could have hoped.

YHBT. YHL. HAND.

Re:PARENT IS TROLL ACCOUNT

[Anonymous Coward]

What exactly is a troll? Someone who does not conform to the latest groupthink trends?

I always judge what people say based on what they say and not some convoluted ideas of what is and isn't an acceptable way to communicate or dictate.

It seems to me that you are advocating censorship because someone does not meet your standards. Just who is trolling who here?

Re:PARENT IS TROLL ACCOUNT

[Anonymous Coward]

Well the fact that he is offensively promoting a poor stereotype of Germans could be seen as strong troll material.

Of course, maybe faking a nationality to ridicule them with every post doesn't seem offensive to you - or maybe it would only be offensive to you if it was the right nationality...would you be so happy to see people like this modded up if they were pretending to be Jewish? African American? A woman who "just doesn't get technology"?

Where's your limit? Why is it okay to make fun of Germans (many of whom have better English skills than the average slashdotter) in your book?

Re:PARENT IS TROLL ACCOUNT

[Anonymous Coward]

Let's turn it around. Why is it wrong to make fun of Germans? Heck American Germans make fun of the Polls, what's wrong with that? Can't we make jokes without being accused of aiding opression?

We really need _that_!

[MikTheUser]

Come on, bust out some even faster processors! I'm sure, with PCI and IDE, we'll be able to get the best out of every single additional Mhz in comparison to those slow and obsolete 2Ghz machines!

TIME TO LEARN FROM HISTORY

[Anonymous Coward]

Each time Intel rolls out a new product, an improved version of that product comes about 6 months later that is much better. Early adoption gets you screwed first, quick, and for a higher price. Nothing is better than paying top dollar to be an uncompensated beta tester for the mobo companies.

It seems to me that when Intel develops a product half way through the design process they realize they screwed up but still release the original tech to make cash, then the fixed version of that tech comes out 6-12 months later.

Pentium 60/66 - 1st of the family, used older mobo technology
Celeron 266 - no L2, much better a few months later with the L2
P2 - oops, here is P3, we fixed everything
P4-1.4 - Couldn't beat a P3-800 yet the ram cost 3x more
P4- Needs to be ~1Ghz faster than a P3/PM/AXP/A64 to deliver the same performance (the clock frequency here is a marketing gimmick - and poor design)
Itanium 1 - I-Tanick

I have a feeling a lot of the P4-3.6s are going to fail in the field with stock Intel cooler.

Re:Land Grid Array

[Anonymous Coward]

it's a shared username, why do you care?

This is interesting, but . . .

[DrMrLordX]

Am I the only one here who's goign to ignore everything Intel releases until they start releasing dual-core Dothan-based P4s? Here's Tom's Hardware's take on the new LGA 775 architecture, along with copious comparitive CPU/platform benchmarks. Anandtech has their own entry here. In both cases, the combination of new architectures, cpus, features, etc don't add up to much of an advancement in performance. What you get are a lot of features of questionable value and/or features that have been touted by platforms such as nForce/nForce2 for some time now. Please wake me when the Dothan P4s start coming out.

Re:This is interesting, but . . .

[SD-VI]

Dothan-based P4s? Dothan is the Pentium-M core, and it's based on P6 (not the Pentium 4-exclusive P7). I don't think they're replacing the P4 line with Dothan, they're doing it with its successor (Merom or something, hell if I can keep track of their roadmap at the rate it's been changing). Also, I'd be surprised if they call 'em P4s, seeing as they have way less in common with any of the P4 cores we know than they do with, say, the good ol' Tualatin P3.

Still, nitpicking aside, I get what you're saying. P6 just has more promise than P7, and a P6-based chip with all the fancypants improvements off the P4 with none of the high-speed low-efficiency madness would... actually, what am I saying? It DOES rule. Just look at Dothan: really fast (no Athlon 64, sure, but still really fast) with-- and here's my favorite bit-- a load heat output in the low twenties. Prescott is in the TRIPLE DIGITS. Finally, cooler desktops.

Re:This is interesting, but . . .

[DrMrLordX]

This article indicates that the dual-core cpus will carry the P4 name. I remember seeing some roadmaps for this somewhere, but I can't find them at the moment. As to whether or not the dual-core cpus will carry Dothan cores or cores based on its successor, I do not precisely know. I only know that they will likely be cores based on whatever is the current Pentium-M at the time of release.

Intel 925 chipset feature

[rumpledstiltskin]

One of the most useful things about the 925 chipset, IMHO, is the interesting possibilities it offers for SATA RAID. Say you want the performance capabilities of RAID 0, but at the same time, you need the redundancy for RAID 1. let's also say that you can only afford two SATA drives.
the intel 925 chipset has native support for a mixed raid, where you can create a raid 0 partiiton across two hard drives, using only part of the hdd capacity on each drive for the raid 0 partition. the rest of the unpartitioned space can be set aside as a raid 1 partition. that way you can install the OS and other non-critical files tha can be lost to the raid 0 partition and get the performance, but if one of the drives fail, you can store your important stuff on the raid 1 partition. I'm trying to find a controller card that will do this functionality, but I can't find anyone that claims to explicitly support it. the only reason I know about the 925 features is I got a chance to play with a pre-production board. definitely a cool feature.

Re:Intel 925 chipset feature

[debrain]

FYI, if you're using Linux, you can just use software RAID. It's widely, albeit controversially, regarded as faster than hardware RAID, and is substantially more flexible.

You might've thought of that already, or not be able to use Linux, but nevertheless, thought I'd mention.

Cheers

Intel is begging for trouble in many ways

[Anonymous Coward]

This isn't just some AC's random opinion either. This was the consensus of most of the Taiwan motherboard manufacturers at Computex this year. P4 boards were everywhere, but nobody really expected them to sell. It's like Intel is on a suicide mission to hit the wall at full speed.
They recently has a press release where they stated that they intend to produce processors that consumer 200watts by the end of the year. And that's not peak power as many people naively want to pretend, that's mostly leakage current that is dumped as heat. It has nothing to do with the halt instruction. You can only reduce the power consumption by turning off the machine. They've clearly lost it.

Just look at the benchmarks

[Alexis de Torquemada]

Considering the fact that Athlon 64 FXs running at 2 to 2.4GHz beat the crap out of various P4 and P4EE with up to 3.4GHz, I can't really take your clockrate argument seriously. Ok, the P4 scored better on some closed-source renderers, but then consider that most applications in the benchmarks have not even been optimized for 64 bit yet!