Intel Launches New Chipset

mikemuch writes "The new P35 and G33 chipsets, codenamed 'Bear Lake' are now available. They have a new memory controller that supports DDR3 RAM at up to 1333MHz, a new southbridge, and will support the upcoming 45nm Penryn CPUs. They don't yet have an actually new and different GPU — their GMA 3100 is pretty much the same as the GMA 3000 of the G965 chipset." For a little more technical info you can also check out the Hot Hardware writeup.

Comments

What's Different

[Nom du Keyboard]

will support the upcoming 45nm Penryn CPUs.

What does Penryn need that's new and different in the way of support? Is it just a bump in FSB speed?

Re:What's Different

[morgan_greywolf]

Well, for one, Intel's biggest instruction set change in 5 years: SSE4 extensions, an updated to Intel's SIMD instruction set.

I know. I'm not all the excited, either. :)

Re:What's Different

[dgoldman]

Voltage is lower. Existing (pre-P35) boards won't support the Penryn.

Re:What's Different

[dfghjk]

In what way do the new instructions require new chipset support? That is what he asked after all.

Re:What's Different

[IPFreely]

Well, for one, Intel's biggest instruction set change in 5 years: SSE4 extensions, an updated to Intel's SIMD instruction set.

Really? I would have thought going 64 bit would be considered a slightly larger instruction set change than SSE4.

Maybe it does not count since it was an AMD invention rather than an Intel invention?

Re:What's Different

[TheRaven64]

Rumour has it (I haven't kept up, so maybe rumour had it) that SSE4 would include scatter-gather instructions. These allow you to specify multiple memory addresses to be loaded into the same vector. This makes auto-vectorisation much easier for compilers, since your memory layout no longer has to be designed with vectorisation in mind.

If this is true, then it might need co-operation from the memory controller to work effectively. Since Intel's memory controllers are on the north bridge chip, it would need a new chipset.

Of course, I'm just guessing, so don't take this as fact.

Re:What's Different

[cthulhu11]

Marvy. More stuff that compilers won't actually generate code that uses. Yee. Hahh.

Re:What's Different

[Overly Critical Guy]

Considering Intel invented x86 decades ago and AMD just copied that, I'd consider them even. Intel unsuccessfully tried to move the world off of the ancient x86 instruction set, and lazy companies like Microsoft decided to put their weight behind the 64-bit x86 hack that AMD offered. Now we get to suffer another 20 years of working around x86.

Re:What's Different

[Tenebrarum]

>Voltage is lower. Existing (pre-P35) boards won't support the Penryn. Great, so that means my Asus Striker Extreme (which allows one to set the voltage in 0.01v increments from 0.5v to 3.0v, or something very similar to that) will support Penryn with a simple vdrop. Excellent. ... Yeah, right.

Re:What's Different

[ben there...]

Voltage is lower. Existing (pre-P35) boards won't support the Penryn.

Do you have a link for that? (Preferably from Intel, or a motherboard vendor, or a review site that talked to Intel) Because I can't find anywhere that old motherboard incompatibility is stated definitively.

Re:What's Different

[IPFreely]

X86_64 is certainly a cludge on a cludge. But Itanic is no better, and in many ways worse. Given the choice of the two, MS made the right one.

If you really want to get off of the bad hardware, you could maybe go with POWER or Alpha or go invent something else completely new. MS tried with Alpha for a while, but noone bought it. So it looks like it's really our fault, not theirs.

Re:What's Different

[mallawaani]

Quote: "VRM 11 Required For 45-nm Processors" "One reason for varying processor support is the voltage regulator circuit of 3-series motherboards. It needs to be VRM 11.0 compliant, which is key when it comes to 45-nm processor support. Let me say that the problem isn't decreasing voltage levels, but strong power fluctuation due to millions of transistors clocking up and down, or switching on and off. Remember that future quad-core processors will be able to dynamically adjust clock speeds for each core individually, and switch cores on and off depending on the workload." "This also means that any 965 motherboard that is VRM 11 compliant can technically support 45-nm processors. VRM 11 says that the circuit is programmed using 8-bit voltage IDs (VID), allowing for 0.00625 V voltage increments. The minimum operating voltage isn't 0.8375 V (as in VRM 10), but goes down all the way to 0.5 V. VRM 11 also comes with the option to share the load across more phases, and the circuit runs so-called dual edge modulation, which means that the controllers send multiple impulses to the transistors while using smaller capacitors. The goal isn't just to provide smaller voltage increments and less voltage for the 45-nm processor generation, but also to provide sufficient power at voltage levels that may switch frequently. This can be done by specifying tight slew rates." http://www.tomshardware.com/2007/05/21/intel_intro s_3-series_chipsets_with_fsb1333_and_ddr3/page4.ht mlTom's Hardware "Intel Intros 3-Series Chipsets with FSB1333 and DDR3"

Re:What's Different

[Thundersnatch]

MS tried with Alpha for a while, but noone bought it. So it looks like it's really our fault, not theirs.

Amen. Microsoft supported x86, MIPS, PowerPC, and Alpha with the first release of NT 4.0. Nobody bought PowerPC and MIPS, and very very few people bought Alpha. So by the time Win2k came around, Windows was x86 only. I really hoped Alpha or PowerPC would succeed and get us off the multilayered hack that is x86, but the masses did not agree with me.

The good thing about this debacle for MSFT is that today, almost evertyhing in the Windows kernel is platform-agnostic except for HAL.DLL and drivers. This fact enables the Itanium versions of Windows. It also means that Microsoft can produce a version for, say, POWER6 processors with a modest incremental investment, if they see market for it. One can only hope.

Re:What's Different

[WuphonsReach]

Considering Intel invented x86 decades ago and AMD just copied that, I'd consider them even. Intel unsuccessfully tried to move the world off of the ancient x86 instruction set, and lazy companies like Microsoft decided to put their weight behind the 64-bit x86 hack that AMD offered. Now we get to suffer another 20 years of working around x86.

Nothing lazy about it.

Intel offered up the Itanium. A 64bit platform that ran existing 32bit code in slow-motion mode. Folks making purchase decisions looked at that, realized that 32bit apps wouldn't be rewritten overnight for Itanium and decided that a move to Itanium was too risky (and expensive).

AMD comes out with their 64bit chips... which run existing 32bit apps at normal speed, plus provide a seamless upgrade path to 64bit computing down the road. We all look at it and decided it was a much more sensible approach and bought it in droves. There were even rumors that existing 32bit code ran faster on the Athlon64 chips. AMD provided a nearly risk-free migration path towards 64bit computing (and Intel finally climbed on board).

If Itanium had been able to run existing 32bit x86 code at close to native speed, we might have shifted that way. The performance hit was simply too large, so the market said "meh, not gonna switch".

Yes, x86 is a PITA, and a kludge on top of a kludge, and other various slurs... but for 90%+ of applications, it works and is fast enough.

OT: External Intel(r) gfx?

[Anonymous Coward]

Really, it would be nice if we can get a external gfx (pcie) for "our" systems.

Re:OT: External Intel(r) gfx?

[jandrese]

You want Intel Graphics as a actual video card? You are aware that you can buy low end nVidia and ATI cards for less than $50 that will outperform them right?

Re:OT: External Intel(r) gfx?

[Anonymous Coward]

But do those nVidia and ATI cards have open source drivers? The Intel chips do!

Intel is therefore the best for Linux.

Re:OT: External Intel(r) gfx?

[Anonymous Coward]

What's performance got to do with it? Intel GPUs have Open Source drivers. nVidia and ATI do not. For some people, that is more important, and the Intel GPUs are perfectly capable of handling all the 3D they need.

Re:OT: External Intel(r) gfx?

[awb131]

> You want Intel Graphics as a actual video card? (sic)

Well, not really, no. But huge numbers of run-of-the-mill business PCs, plus the Apple "consumer" line (mini, imac and macbooks), use the standard Intel graphics hardware. It does OK for most people's purposes, and the install base is huge, and for those reasons, a bump in capabilities for the onboard graphics chip would be noteworthy.

Re:OT: External Intel(r) gfx?

[drinkypoo]

I wouldn't go redundant like this but both of the other replies are from ACs and many people will never read them/know they exist.

So far, Intel is the only company with supported OSS drivers. AMD has "promised" to deliver them for ATI cards, but who knows how long that will take? And nVidia has made no such promise.

In addition, if we could get them without shared memory, the performance would likely improve and it wouldn't drag down system performance. So that would be a great thing.

When we get OSS drivers for ATI, it might become possible to use one under Linux (or any other OS but MacOS for which Apple participates in driver development) in a reliable fashion. But ATI's drivers are poop anyway. Regardless, those who want a 100% OSS system can not buy a current nVidia card, as they are unsupported; an older nVidia card still in production is likely to come from one of the least-reputable vendors, so a card supported by the 'nv' driver that's worth using will be hard to come by. Intel is currently the only credible choice for accelerated video with OSS drivers.

Re:OT: External Intel(r) gfx?

[GooberToo]

But ATI's drivers are poop anyway.

Which is probably why they have claimed they will do it. Their drivers stink. If they can get people to code up quality drivers for little to no expense, suddenly they are much more competative with NVIDIA plus they've bought mindshare in the OSS community.

Re:OT: External Intel(r) gfx?

[mrchaotica]

Yep -- a GMA 950 will outperform even a GeForce 8000 GTX, when the GeForce is using the nv driver!

Re:OT: External Intel(r) gfx?

[drinkypoo]

But ATI's drivers are poop anyway.

Which is probably why they have claimed they will do it. Their drivers stink. If they can get people to code up quality drivers for little to no expense, suddenly they are much more competative with NVIDIA plus they've bought mindshare in the OSS community.

Yeah, it sounds like a win-win situation for ATI. All the proprietary, encumbered code in the world hasn't enabled them to create drivers that are worth one tenth of one shit. I've kept trying ATI off and on over the years, and the situation has frankly never changed. It's been this way since the first Mach64 I ever used.

It's becoming increasingly clear that Linux is taking over the world. Dell selling linux. IBM sells more linux than AIX. Sun trying to work Solaris over to be more Linuxlike (a horribly stupid idea - if your customers want linux, sell them linux. But whatever.) Motorola moving towards Linux on all their non-WinCE phones (and how long can it be before they are taken over too?) Palm going linux. And of course, in a way even Microsoft sells Linux now.

Going Open/Free now is a way of riding that particular beast to the top, instead of grabbing on to its tail and getting thrashed around for a while.

Re:OT: External Intel(r) gfx?

[scottnews]

I wouldn't expect that to change. ATI's open drivers will give limited functionality. There will be no 3D acceleration with those drivers.

Re:OT: External Intel(r) gfx?

[mczak]

In addition, if we could get them without shared memory, the performance would likely improve and it wouldn't drag down system performance. So that would be a great thing.

I don't know how much faster the "same" graphic chip would be if it just would get its own ram, but that igps drag down system performance is basically a myth nowadays. Used to be true when single-channel sdram was best you could get, but it's basically a non-issue with todays dual-channel ddr2 memory systems. (Bandwidth needed for scanout, which is basically what slows things down even if you don't do anything graphics related, hasn't increased that much - a 1920x1200x32 display at 60Hz would need roughly 500MB/s, if you have a chipset which provides 1066MB/s (single-channel 133Mhz sdr sdram) this is a lot but if you have a chipset which provides 12.8GB/s (dual-channel ddr2-800) it's just not that much.)

Re:OT: External Intel(r) gfx?

[abundance]

[quote]In addition, if we could get them without shared memory, the performance would likely improve and it wouldn't drag down system performance. So that would be a great thing.[/quote] I think the shared memory issue is a bit overweighted.
With current memory speeds and dual channel bandwidth, system memory can handle the additional traffic load of the graphic subsystem without suffering that much. And for what concerns 3D graphic performance of those budget cards, that's mainly gpu bound, not memory bound.
I don't think you'd see noticeable improvements in system or graphic performance with discrete graphic memory - while the burden of adding it would hurt the price convenience of the integrate solutions.

Of course, if you have a low specced system, say with 512mb of ram, even those 80mb that the graphic card steal for itself hurts. But a 512mb configuration is quite doomed nowadays by itself.

Re:OT: External Intel(r) gfx?

[drinkypoo]

a 1920x1200x32 display at 60Hz would need roughly 500MB/s, if you have a chipset which provides 1066MB/s (single-channel 133Mhz sdr sdram) this is a lot but if you have a chipset which provides 12.8GB/s (dual-channel ddr2-800) it's just not that much.)

The problem isn't just one of overall bandwidth use, but also one of contention. Further, when used for 3D the memory consumption will be greater because not only graphics memory but also texture memory is in system RAM.

And of course, you don't actually get peak bandwidth across the memory bus. It would be nice though.

Re:OT: External Intel(r) gfx?

[mczak]

The problem isn't just one of overall bandwidth use, but also one of contention. Sure, but that can be dealt with. I'm not exactly sure how current chipsets handle it, but they certainly have some cache for display buffer, together with some logic for prioritization (if the display buffer is full, requests from the display controller to the memory controller have low priority, if it gets more empty priority will increase).

Further, when used for 3D the memory consumption will be greater because not only graphics memory but also texture memory is in system RAM. Yes but as said, that doesn't count as "drags down system performance". It will "only" drag down 3d performance. It will eat some ram, true, but as long as you have "enough" (at least 1GB) that shouldn't really be a problem.

And of course, you don't actually get peak bandwidth across the memory bus. It would be closer to peak (at least for intel cpus) if the frontside bus could actually keep up with the memory bandwidth, as FSB bandwidth is quite a bit lower :-). So you actually do have "free" bandwidth you can use without a penalty in theory...
But anyway, you don't need to trust my theoretical ramblings, the benchmarks speak for themselves. Unfortunately they are hard to find, boards with igps get seldomly reviewed (and even less both with and without using the igp). Here's some quick numbers: http://www.digit-life.com/articles2/mainboard/giga byte-965g-ds3-i965g.htmlNotice the non-3d benchmarks are within one percent if igp is enabled or not - though the resolution isn't stated (I'd guess it was 1600x1200 but I could be wrong...). I never said igps are fast for 3d :-).

Re:OT: External Intel(r) gfx?

[drinkypoo]

Notice the non-3d benchmarks are within one percent if igp is enabled or not - though the resolution isn't stated (I'd guess it was 1600x1200 but I could be wrong...). I never said igps are fast for 3d :-).

This is irrelevant because the maximum load to the system only occurs when the system is doing 3D graphics.

What we need is two versions of essentially the same graphics card, one IGP and one standalone. But good luck finding that.

Or more to the point, we need to run another benchmark while a 3d benchmark is running...

Re:OT: External Intel(r) gfx?

[Grishnakh]

Nvidia cards have an open-source driver, called "nv".

Of course, it's 2D-only, so those fancy Nvidia cards are basically worthless for 3D video if you want all open-source drivers.

Kudos to Intel for releasing open-source video drivers; I just wish they'd make stand-alone PCIe boards with their chips.

Re:OT: External Intel(r) gfx?

[Grishnakh]

Well, it does make sense. It's all about not reinventing the wheel over and over again, and working together with others to create something greater than any one of you could alone.

The only people getting shafted by open-source software are the proprietary software vendors that compete directly against OSS solutions. Too bad, so sad. Time to move on to something else; I don't see Cadence, MentorGraphics, or AutoDesk complaining much about OSS software cutting into their business. Although AutoDesk better get its butt in gear and make Linux versions of its products or else they might face OSS competition sooner than they thought. The former two have been delivering products on Linux for years.

Re:OT: External Intel(r) gfx?

[drinkypoo]

Of course, if you have a low specced system, say with 512mb of ram, even those 80mb that the graphic card steal for itself hurts. But a 512mb configuration is quite doomed nowadays by itself.

512MB is adequate for Windows XP (although you will notice a change to 1GB) and for OSX (ditto) but not enough for Vista. It's more than enough for any Linux you care to use.

Note that the only OS with which you are actually doomed with 512MB is Vista...

Now, with that said; OSX is slow no matter how much RAM you have, XP is slow until 1GB (going from 512MB to 1GB actually halved my boot time including login, no joke, on an XP2500+ system) but Linux even with full GNOME or KDE and with Beryl will be fine. I have a ton of stuff running including a 256MB-RAM virtual machine running another copy of Linux and I'm using 660MB of my 2GB RAM, and only 32MB swap (stuff that will get paged out whether you've used up your memory or not.) This is with Ubuntu 7.04/feisty, Beryl running at 1680x1050x32bpp, Firefox with multiple tabs, Thunderbird 2, VMware server 1.0.2 (1.0.3 not being provided by Ubuntu yet), etc. AND I am running a ton of services as well, like Apache2 with PHP5, MySQL, spamassassin, samba, squid, tor, proftpd, and others, so it's not like I have slimmed down my system. If anything, it has been fattened.

No, I don't agree at all that 512MB is not enough any more. It's not the headroom it once was, certainly. But as long as you're not using Windows or OSX, you'll be fine :D

Re:OT: External Intel(r) gfx?

[Drawsalot]

Vista (Ultimate Edition) pretty much dictated 1 GB RAM for me, initially, and started to behave at the 2 GB mark. Once I moved to the 64-bit version of Vista, I went up to 4 GB and so far it just purrs. XP Pro SP2 needed about 2 GB to really rock.

Re:OT: External Intel(r) gfx?

[abundance]

Note that the only OS with which you are actually doomed with 512MB is Vista...
Mmh ya, you're right, the "quite doomed" thing was... quite stretched. =)
After all my main machine at home is also an XP2500+ with just 768mb and it's okay with Windows XP. My sister still use a Duron 600 with 512mb and she's fine surfing and writing.

My point was about the shared memory stuff - its drawbacks would kick only with a limited amount of system memory, and you'd better off simply adding ram than thinking about discrete memory for the integrated graphic card.
And anyway I think nobody would consider buying a brand new machine with less than 1gb ram in the first place.

Re:OT: External Intel(r) gfx?

[MBGMorden]

Yeah. My sister is still using a thrown together Celeron 533 with 256mb of RAM (running Windows XP no less) and she generally hasn't complained about it to me. It lets her check email, download digital camera photos, and surf around the web. That's all a lot of people care about.

My Windows system was honestly a bit sluggish for my tastes with 1GB though (not so much within an app as it was SWITCHING between apps. If I was playing WoW and tabbed out to check something on the web, the system ground to a crawl for 10-15 seconds). With 2GB it's fine.

I also have a PowerMac G4 (Dual 500mhz w/ 512mb ram) that is a bit laggy, but I'm not sure if it's the ram or the processor speed at fault there.

Re:OT: External Intel(r) gfx?

[default luser]

But do those nVidia and ATI cards have open source drivers? The Intel chips do!

Do Intel's X3000 open-source drivers have support for the T&L units and vertex shaders built-in to the X3000? The Windows drivers certainly don't, even after 9 months on the market!

Almost 3 months ago, beta drivers were promised, but they have yet to surface. The X3000 is still using the processor to perform vertex shading / T&L, just like the GMA 900 / 950, and that's why it still gets beat by the old Nvidia GeForce 6150.

Beer Lake?

[Zwets]

*hic* Best name evar!

..oh, wait.

Re:Beer Lake?

[Des Groines]

Informative? This is Slashdot.

anything new?

[CaptainNerdCave]

my overclocked e6300 shows a fsb of 1333 in cpu-z with ddr2-667 that has been stable for almost 9 months, why do i need ddr3? if ddr2 can do these speeds, why should i upgrade? looks like a bunch of hype; i'll wait for tom's hardware to show some comparisons against overclocked, currently available stuff.

Re:anything new?

[Anonymous Coward]

why should i upgrade?

Because your 9 month old computer is now a paperweight silly. If you don't have the latest of everything you are "The SuX0r". You'll need all new stuff to overclock, to demonstrate your 'leet skillz that some web site told you was "Da Bombz"

I like Tom's myself, but this guys an idiot asking to be spoon fed

Re:anything new?

[Anonymous Coward]

Anandtech has a decent comparison article for DDR2 vs. DDR3:
http://www.anandtech.com/memory/showdoc.aspx?i=298 9&p=5

TLDR: high latency DDR3 performs about the same as high-end DDR2.
Once DDR3 latency improves (or the price goes down), expect it to be better than DDR2, but not until then.

Re:Yawn

[AKAImBatman]

From what I can tell, this is basically good news for anyone who wants an affordable business machine. The updated specs mean that it will be competitive with the latest technologies, while still offering the savings and simplicity of an integrated design. (Assuming you're not die-hard about Intel integrated graphics sucking.)

On another topic, I love the screenshots of the upcoming motherboards. Computer components are getting so colorful. I remember back when you got a green motherboard with black and white parts. (<grumpy-old-man>And we liked it that way!</grumpy-old-man>) Maybe with a few blues for caps. Now you can really see the different parts as they leap out at you in blues, purples, oranges, yellows, and greens! Yeah, it looks a bit Fisher-Price. But it's kind of refreshing at the same time. :-)

Intel can quit development now

[Anonymous Coward]

1333MHz DDR3 RAM should be fast enough for anyone.

Re:Intel can quit development now

[aarku]

I think they have exactly 4 megahertz to go before they can stop.

Re:Intel can quit development now

[bruno.fatia]

Just overclock your memory to 1337MHz. Same feeling.

Re:Intel can quit development now

[windsurfer619]

I'd get a better feeling overclocking my pentium 2's ram to the same speed.

P35?

[Anonymous Coward]

I thought they were still on the P4.

Or with actual performance testing

[EconolineCrush]

The Tech Report also has coverage, with full application and peripheral performance testing: http://techreport.com/reviews/2007q2/intel-p35/ind ex.x?pg=1

Re:Or with actual performance testing

[Crazy Eight]

Check out those power consumption stats. Ouch.

When Do We Get Onchip DSPs?

[Doc Ruby]

Intel devastated the entire DSP industry in the late 1990s when they staked out the NSP ("Native Signal Processing") strategy of faster clockrates to run DSP in SW instead of in HW. But now they're up against new Cell chips from IBM which multiprocess with parallel DSPs onchip, and even GSPs ("Graphics Signal Processors") threaten new competition from first nVidia, then TI and other old surviving rivals, as GPGPU techiniques become more sophisticated and applicable.

All because DSP is more parallelizable than true general purpose processing, as parallelization is the best solution to increasing CPU power, just as the data to be processed is inherently more parallel, and more simply streams of "signals", as multimedia convergence redefines computing.

So when will Intel reverse its epoch of NSP, and deliver new uPs with embedded DSP in HW?

Re:When Do We Get Onchip DSPs?

[AKAImBatman]

So when will Intel reverse its epoch of NSP, and deliver new uPs with embedded DSP in HW?

Probably about the same time that web application developers realize that their problems (particularly AJAX) can be solved more efficiently with a DSP architecture and start designing tiers of servers in a pipelined DSP configuration. Considering the amount of computer science exhibited by this industry, I'd peg it at sometime around a quarter to never.

Re:When Do We Get Onchip DSPs?

[Anonymous Coward]

Intel has already done R&D on, and even demoed, an 80 core CPU that focuses on floating point performance, and does so better than current GPUs (and Cell). It will presumably still be a few years before it's released, but that's the direction they are headed in.

Re:When Do We Get Onchip DSPs?

[jshriverWVU]

What is a DSP really? Know it stands for digital signal processor, but not really sure what that means. All that comes to mind is sound synthesizers or something that takes in some kind of signal and changes it's outcome. They seem really popular.

Re:When Do We Get Onchip DSPs?

[Doc Ruby]

Intel putting parallel DSPs on their uPs is not driven by mere "efficiency". Intel demonstrates, even defines, both the CS and economics that are forcing competitors, thereby Intel, too, to put DSP in their cores (literally and figuratively).

And there's not much sense in Web apps being processed by an FIR or full-spectrum mixer.

All I really get from your comment is that you don't know what DSPs do, or what Intel does - or maybe how Web apps work.

Re:When Do We Get Onchip DSPs?

[Doc Ruby]

Do you have some citation of something real, rather than just an Anonymous Coward posting an uncited rumor that amounts to nothing more than FUD?

Re:When Do We Get Onchip DSPs?

[Anonymous Coward]

It's a processor. It runs firmware. Its opcodes and overall execution is geared heavily towards processing signals and heavy number crunching. Think of it as a processor where MMX type instructions are the norm rather than an extension.

Re:When Do We Get Onchip DSPs?

[Anonymous Coward]

Will this do?

Alternately, www.justfuckinggoogleit.com, dimwit.

Re:When Do We Get Onchip DSPs?

[Doc Ruby]

What would you suggest I "google", Anonymous idjit Coward? When I google "dimwit" all I get is "Anonymous Coward".

Congratulations! You've won the "Stupidest Passive-Aggressive Slashdotter of the Hour" award. You win a crusty old 1990s joke to add to your collection.

Re:When Do We Get Onchip DSPs?

[Doc Ruby]

DSP is indeed a "Digital Signal Processor". It's a chip specialized for... processing digital signals. Which nowadays nearly always means running a lot of repeated simple linear transformations of the same basic form on a stream of data. Usually it's a "Multiply/Accumulate" ("MAC" or "MADD"), of the form y=m*x+b, run very fast (billions of times a second), with lots of arithmetic support like zero-delay increments/looping. Also usually at the sacrifice of some performance, or even existence, of some logic operations.

It's a microprocessor with a very fast ALU (math) and corresponding IO, but usually with a much slower CLU (branches/loops/comparisons etc). And since so much computation is in the ALU, not the CLU, it's increasingly popular for more processing than just an optional chip on a bus from the CPU can deliver.

Re:When Do We Get Onchip DSPs?

[Anonymous Coward]

I don't know, "intel 80 core processor", maybe? Seriously, are you this stupid?

Re:When Do We Get Onchip DSPs?

[AKAImBatman]

All I get from your comment is that you weren't following what I was saying.

Intel's designs are driven by what drives the sales of their processors. For right now, that's gobs of desktop and PC Server machines. The alternative architectures are in no danger of knocking Intel out of that position. They will carve themselves a niche for now, which is why Intel has been more worried about AMD than they've been worried about IBM. Which means that Intel will sit up and take notice of the DSP-oriented chips if and when it actually impacts their bottom line.

Desktop machines are unlikely to need DSP processing anytime soon. (Outside of the similarities in GPU design, that is.) So there needs to be an extensive push inside the server arena to make Intel budge. As it happens, there is an issue that DSP designs can address.

Traditionally, DSPs have been used to process sampled analog data in real-time. They are carefully calibrated to stream the samples at high clock rates, with the expectation that the data will be merely transformed rather than processed by means that would require unexpected (and costly) branching. This expectation allows DSP microprocessor makers to design their chips in certain ways.

Of course, the basic tenets behind DSP microprocessor design do not require that the data I/O be analog in nature. Which is why you're starting to see DSP-derived designs (e.g. the Cell) starting to show up in digital devices with no analog sampling. As you said, the economics of the situation are making DSP microprocessors more effective than traditional GP microprocessors.

Now look at the web. What are we passing around like madmen? Streaming data, of course. These data streams are architecture-independent serializations that potentially lend themselves well to parallel processing of the data. In many cases, it's a simple matter of translating a data stream from a serialized form into an memory form or visa-versa. Sun has already realized this and has begun producing multiprocessor chips designed around streaming data. But their solution is highly focused on existing web loads, and will lose effectiveness as multimedia solutions become more common on the web. Cell is more general purpose, being capable of handling integer and floating point data equally well.

But I'm getting ahead of myself. With the current streaming situation on the web, we're starting to run into resource limitations on individual machines. The interim solution is to develop a pipelined processing stack over the network. As the requests stream into the front-end servers, they should be processed only as far as necessary to branch the stream to the correct server that's next in the pipeline. This server then takes the data and either deals with it directly, or does more processing and branching.

To give an example, take an instant messaging site like Meebo. Let's take a look at how it could be architected in a streaming fashion.

A login to the server produces a ripple effect where each person on the buddy list needs to be notified. This event creates an inverse multiplex for N number of signals, where N is the number of "buddies" in the list. Since these buddies are potentially on different communication protocols, the events can be demuxed to different servers, each one dedicated to maintaining links to that protocol. (e.g. Jabber, AIM, MSN, Yahoo, and Meebo)

IM packets continually hitting their servers as parallel streams of data. Similar to generating a login event, this requires that the messages be demuxed to the correct server based on their intended destination. It also requires that the incoming stream be transformed from a web stream (e.g. XML, JSON, or other textual format) into the format of the destination protocol. This can be handled by a separate machine from the one that maintains the backend IM connections. The stream can be transformed ahead of time, then forwarded to the protocol server for actual delivery to the IM network.

This example is a key area where

Re:When Do We Get Onchip DSPs?

[Doc Ruby]

In our universe, everything can be characterized as a signal. In our society, practically all signals can be usefully digitized. That doesn't mean DSPs are right for everything, because DSPs aren't good at all signal processing, just some - repeated loops of simple, if cumbersome, linear equations.

DSP is fast math at the expense of fast logic. Web apps have at least as much logic as math, intractably intertwined. DSP of Web apps is inappropriate. DSPs on a chip with fast logic would be good for Web apps and everything else. Intel sells lots of CPUs to process Web apps. And IBM/Toshiba/Sony is planning to sell lots of Cells to do so.

I know what you're talking about. And I know that you don't.

Re:When Do We Get Onchip DSPs?

[Doc Ruby]

No, you just made some anonymous claim with minimal info content. I politely asked you for a citation so I could learn more.

Then, like the classic Anonymous Coward asshole, you turned totally obnoxious. So I can tell without any further help from you that you are indeed that stupid.

Re:When Do We Get Onchip DSPs?

[Anonymous Coward]

When we get smart programmers. Face it, HW DSP (multithreaded) programming is hell and software to help you program it is not cheap. Ask Sony. Another thing, hardware is cheap nowadays and people are expensive so a solution that is a little slower that can be developed without crazy deadlock bug hunts you can get out through the door while the other guy is still debugging is going to help the bottom line. Yes parallelization is a good solution but not many things can benefit from parallelization. It's not a solution applicable to every problem, just a small subset of problems with many isolated chunks that can be calculated in parallel without any inter-dependency. Any decent programmer would be able to tell you how few problems can be solved that way.

Re:When Do We Get Onchip DSPs?

[LionMage]

Normally, I wouldn't comment, but it seemed to me you were heaping a bit of undue scorn on the AC poster. The 80-core prototype processor was covered previously on Slashdot.

ArsTechnica has good coverage, and you can find more at C|Net. Incidentally, the AC was right -- simply googling for "Intel 80 core processor" yields plenty of results. (Googling for "Intel 80 core processor Slashdot" will find the Slashdot article to which I provided the link.) Instead of ripping on the guy for being passive-aggressive, maybe you should stop and research before posting a comment insinuating that the AC poster is just spreading FUD. This story was covered well enough recently that a hyperlink citation shouldn't have been strictly necessary.

Re:When Do We Get Onchip DSPs?

[Doc Ruby]

Yet Sony has bet its own uP and videogame businesses on putting parallel DSPs on the Cell, challenging Intel.

As I mentioned, increasingly large segments of all PC processing are suited for DSP. So others, like Sony, are putting DSPs on the CPU.

I started programming/designing for DSP over a decade and a half ago, about half my programming/designing career, so I can tell you - as I did. Proper design, info architecture, eliminates "crazy deadlock bugs". Or you can design it wrong, on a regular CPU without DSP HW, and really go crazy.

Re:When Do We Get Onchip DSPs?

[Anonymous Coward]

I agree that everything I wrote after my first post was rude; but my first post was simply and politely pointing out that Intel was doing something very similar to what you talked about in your first post. Had you replied to this politely (and here I mean a realistic definition of polite, not some bizarro world definition of polite where making a baseless accusation is like saying "please"), I would have linked the article and been done with it. Instead, you chose to act like a child. You insulted me and accused me of spreading FUD without taking even 5 seconds of your own time to try and verify the "rumor" I posted. This pretty much exempts me from any responsibility to reply politely; if you choose to treat me like shit solely because I'm posting AC (or solely because you're just a rude idiot in general), then don't expect me to reply with any semblance of politeness.

Re:When Do We Get Onchip DSPs?

[Doc Ruby]

When an Anonymous Coward says something controversial, I ask them for a citation. If I googled every BS an AC posted, especially regarding vaporware, I'd have googled a lot of BS.

When someone makes a counterintuitive claim, they should cite it. When asked politely to cite it, they are obligated to cite it. When they're obnoxious about it, they deserve scorn.

FWIW, "coverage on Slashdot" neither makes a claim common knowledge, nor really offers any substantiation to a claim, unless it that coverage cites reliable sources.

Re:When Do We Get Onchip DSPs?

[Doc Ruby]

Given the vast majority of uncited rumor posted by ACs is merely vaporware/FUD, replying with

Do you have some citation of something real, rather than just an Anonymous Coward posting an uncited rumor that amounts to nothing more than FUD?

was polite.

Then you turned obnoxious asshole, which was completely consistent with that worthless kind of AC posting.

If you'd just cited your original anonymous claim, or just given the citation in the face of the good reason I asked for it, we could have had an interesting discussion. Now I've learned less than I'd like about a possible Intel direction, and more than I'd like about your bad attitude.

Re:When Do We Get Onchip DSPs?

[Anonymous Coward]

He is obviously a fucking troll. Why respond to it. In fact...why am I responding to this.

Re:When Do We Get Onchip DSPs?

[Anonymous Coward]

Are you calling me (the poster of the information) a troll, or Doc Ruby? Just curious.

Re:When Do We Get Onchip DSPs?

[tom17]

DocRuby fo sho.

AC - 1
Brave poster - 0

Re:When Do We Get Onchip DSPs?

[tsalaroth]

Normally, I don't respond to troll-ish comments - not since my first comment on a /.

You're wrong. That wasn't polite. Now that you have a non-AC that disagrees with you on that point, will you please just drop it?

Re:When Do We Get Onchip DSPs?

[Doc Ruby]

I'm going to continue my policy of requiring both ID'ed and AC posters to back up their assertions, especially critical ones. How is my statement impolite?

I remind you that responding to ACs who make incited claims demands only the minimum of politeness.

Re:When Do We Get Onchip DSPs?

[groslyunderpaid]

Why would Intel ever reverse it's position on NSP, when embedded DSP in HW can't hold enough uP's for GSP's to effectively compete with the GPGPU technique's employed by NSP? Someone please correct me if I'm wrong, but it seems top me that as long as NSP's GPGPU'd uPs work at ~115% of the DSP from Nvidia, then Intel will have the corner market.

Re:When Do We Get Onchip DSPs?

[Anonymous Coward]

Hey loser, go slit your fucking wrists so the world will have one less fucktard in it. I shouldn't even have to back up my claims unless you are just too stupid to even exist let alone use a computer.

Re:When Do We Get Onchip DSPs?

[Doc Ruby]

Sony's put Cells into over 3 million PS3s. The Cell is a 3.5GHz PPC on a cache-coherent onchip bus with 6 usable DSPs for over 200GFLOPS. IBM is putting together workstations to supercomputers of these same (and denser) Cells in parallel. Other chipmakers are following suit. These are not GPUs, but CPUs with embedded DSP that can process graphics: "NGPU", if you will. But actual GPUs use even more embedded DSPs to get something like 10x the specialized performance.

NSP is the way to go only when there's CPU power to burn and DSP is a small share of the CPU duty. Now that CPUs without DSPs are hitting sequential walls, and going parallel, they're parallelizing the NSP HW, with all its overhead, just to handle more DSP tasks. To compete with Sony and others (probably including nVidia), Intel will have to do what they do. Thereby flipping the entire CPU industry to mainly embedded parallel DSPs.

The question is when, and in what form. There's already some demos by Intel of MPP DSP CPUs. But they're not x86 compatible, while the Cell is PPC compatible (reuse PPC code on the PPC core, then port some features to use the DSPs). Let's see what Intel does, and what they show in the next few years that look like Sony/IBM/Toshiba will pioneer.

Re:When Do We Get Onchip DSPs?

[Doc Ruby]

Thank you for finally backing up what I've said about you all along.

Cunt.

Re:Yawn

[Akaihiryuu]

Wow, a Pointer Sisters reference in a comment about chipsets.

DDR3 RAM at up to 1333MHz? ::Yawn::

[More_Cowbell]

Thanks, but I think I will wait for the next chipset ... that can support ram to 1337MHz.

Re:DDR3 RAM at up to 1333MHz? ::Yawn::

[Anonymous Coward]

I actually worked at Intel during the development phase of the 1333MHz FSB and memory busses, and there was a little discussion among some teams as to whether we should bump it to 1337MHz for the gaming crowd. Most people took it as a joke, but I know some of us were serious - why the hell not, you know?

Re-state the question.

[DrYak]

A chip set is just supposed to talk to the CPU, and in case of Intel's architecture, talk to the memory.

A new chipset for DDR3 is logical in this situation : the chipset has to handle a different and electrically incompatible memory.

But why does a new CPU needs a newer Chipset ?!?!?

Meanwhile, in AMD's land, there's a standard between the chipset and the CPU called Hypertransport.
As long as both the CPU and the chipset follow the same protocol or compatible variation of (like AM2 being HT/2.0 and AM2+ and AM3 being HT/3.0) you can pretty much pair any thing you want.
The only restriction for a mother board is to have compatible socket (the CPU has on-board memory controller and directly speaks to the RAM sticks. There are different sockets type for different memory combination : 794 is for single channel DDR, 939 is for dual channel DDR, AM2 is for DDR2, Opteron F is for DDR2 and much higher number of Hypertransport lanes), and even that is getting stabilised (future AM2+ and AM3 CPUs can plug in today's AM2 board).

Why can't Intel guarantee the same kind of stability ?!?!?

Oh, yes, I know : they make chipsets and earn money by selling more motherboard.
Even back at the Pentium II/III era they have gone through the same cycle, releasing several incompatible chipsets and slot/socket formats in order to pump up motherboard sales, even if the same slot-1 PII motherboard could last until the last PIII only using adapted slotckets.

Meanwhile AMD is getting recommended on various website (like Ars Technica) as preferred solution for entry-/middle- level machines, because of cheaper board and more stable (and upgradable) hardware.

Stability of AM2/AM2+/AM3 is one of biggest AMD's advantage over LGA775 and should be put forward.

Re:Re-state the question.

[Jeff DeMaagd]

The Slot 1 business was just a temporary thing, pretty much forced by the need for high-clock cache. The silicon manufacturing technology did not yet allow for affordable large on-die cache. I thought the slot/socket adapter was very good idea. I don't think such an equivalent was offered for adapting a socketed Athlon into a slot, which early Athlons had slots, later it was socketed.

Stability of AM2/AM2+/AM3 is one of biggest AMD's advantage over LGA775 and should be put forward.

What do you mean by "should be put forward"?

Anyway, it would be nice to have a broader upgrade range. While AMD's pattern is superior, it's still far from ideal, especially when they too have socket variations. There will always be newer memory standards, and I think that's going to cause problems because that means a different socket for the AMD, still meaning a new board is needed, even if the chipset is otherwise compatible.

Re:Re-state the question.

[mrchaotica]

Meanwhile, in AMD's land, there's a standard between the chipset and the CPU called Hypertransport.

Note that that's not just "AMD land," that's IBM land, VIA land, Transmeta land, HP land, SUN land, and every-other-chip-manufacturer-except-Intel land.

Re:Re-state the question.

[Anonymous Coward]

While back in intel land, I can drop in the Quad-core CPU that I just bought for $800 into a brand new motherboard and now I'm using DDR3 and all it's benefits.

Back in AMD-land, I have to buy a whole nother CPU just because my memory changed? Why can't AMD be more modular like intel? I think AMD put their memory controller in the CPU just to force us to upgrade CPU's whenever we change memory types.

Re:Re-state the question.

[meatpan]

Oh, yes, I know : they make chipsets and earn money by selling more motherboard. As a former Intel employee, I can guarantee you that Intel does NOT make money from chipsets and motherboards. The entire purpose of Intel's server and desktop motherboard operation is to enable their new technology through early discovery and elimination of major processor bugs, and to help the actual motherboard/chipset manufacturers to better support Intel architecture.

Why would Intel invest in chipsets and motherboards when the profit margins are slim (as compared to much higher profit margins for a cpu)? For one, the investment in chipsets and motherboards has saved the company from major disasters on several occasions by early detection of obscure bugs. Knowledge of internal problems can allow the company to delay or cancel a product (such as Timna), which is much less harmful to a stock price than shipping a broken product.

By the way, divisions within a company that constitute a material portion of earnings are required to report their revenue. If you want to know whether or not Intel makes money from chipsets, you can look it up in public records.

Re:Re-state the question.

[darkwhite]

Why can't Intel guarantee the same kind of stability ?!?!? You've got to be fucking shitting me. What are you high on? Because I'd like some of that. I can't see a single statement in your post that isn't absurd and that doesn't turn the truth on its head.

There are plenty of reasons to favor AMD over Intel, but sockets are not one of them.

Have you checked the longevity of LGA775, the only desktop and entry-level server socket that matters? And have you compared that to the longevity of AMD's sockets? Have you read the fucking article? Have you looked at Intel's CPU or chipset roadmaps? Do you know how long Intel plans to support LGA775?

AMD may have had its reasons to switch sockets, but it has managed to royally piss off its customers with the way it abandoned s939 and s940 after stating for a long time that it would not do so. AMD motherboard suppliers also had the same compatibility blues when Athlon X2s were coming out. Compare to Intel, which managed a smooth transition to a radically new CPU architecture without socket changes and with many chipsets not even batting an eye, and is now doing the same thing with the DDR3 migration.

I have no time to go through your absurdities one-by-one, so I guess I'll just assume you're one of those terminally brain-damaged fanboys who inhabit hardware forums and spout things that have no connection to reality.

Re:Re-state the question.

[cheater512]

What the hell are you on? GP is completely correct and your just ranting.

Re:Re-state the question.

[mrsteveman1]

"Stability of AM2/AM2+/AM3 is one of biggest AMD's advantage over LGA775 and should be put forward."

Are you SERIOUSLY trying to say that 3 separate AM- systems are more stable than one socket?

Re:Re-state the question.

[chromozone]

Maybe that's why Asus has such sketchy support compared to Intel. Their site is horrid and they released my present mobo (P5W HD) without without testing it with retail C2D cpu's (they used engineering samples). When motherboards wouldn't boot with the cpu's (as advertised) the solution Asus had was to have people buy a new Celeron cpu's so they could update the BIOS and THEN the C2D would work. My next mobo will be intel. However with all manufacturers I learned not to rush to buy new board because many of them have headaches the first few months/revisions.

Re:Re-state the question.

[Anonymous Coward]

AMD fanboy alert! Get over it, fanboy. Intel won this generation. Deal with it.

Re:Re-state the question.

[Anonymous Coward]

your just ranting

That's just you're opinion.

Re:Re-state the question.

[ben there...]

But why does a new CPU needs a newer Chipset ?!?!?

I looked for a definitive answer from nVidia or eVGA, but it's not clear whether nForce 680i boards will support Penryn/Wolfdale or not. FWIW, a moderator at the eVGA forums thinks they will, but nobody knows for sure.

So unless Intel says otherwise, chipsets from other vendors may work with Penryn, regardless of Intel's chipset refresh for DDR3. I mean, most enthusiast boards do well over 1333 MHz FSB, and also have fine-grained voltage adjustments. Unless the rumor mentioned in that thread about a new vreg revision is true, it should work. Theoretically.

Re:Re-state the question.

[(Score.5, Interestin]

Actually both Asus and Gigabyte are shipping boards built using engineering samples (!!!). This is visible in the VR-Zone and OCWorkbench reviews, with the chips marked "Secret" "ES". This is a very dubious way to build a retail product.

Re:Re-state the question.

[WuphonsReach]

However with all manufacturers I learned not to rush to buy new board because many of them have headaches the first few months/revisions.

The real lesson you should've learned is to always buy CPU+MB+RAM in bundled form. Where the retailer has already put the 3 components together and guarantees that they work. (MWave charges all of $9 for the service.) With a motherboard bundle, you eliminate all of the guesswork and you're sure to get a working setup. Some places call this an "assemble & test" option.

(I started buying motherboard bundles about 2 years ago. I'll never, except in a pinch, go back to buying the components separately. Saves me about 15-30 minutes of assembly time and hours of research time. For the most part, the cost is the same as individual parts, except for a small assembly fee.)

I'm still looking for a good retailer who does Opteron motherboard bundles...

Re:Re-state the question.

[Dzonatas]

One wonders why Intel gets any blame for these designs. Games and graphics cards require one to upgrade those many times more than one has to upgrade the motherboard itself. What is the "preferred solution for entry-/middle- level machines" for what audience?

My old BT2 board still performs as good as current boards on the market.

That preferred solution probably doesn't represent an average between GPU driven and CPU driven markets. Hmm.

SSE 4

[serviscope_minor]

Wikipedia has a more useful description of SSE4

As far as I know, gcc only supports up to SSE3 intrinsics. Look in pmmintrin.h

DDR2 and DDR3 on the same mobo

[anss123]

Intel should be applauded for supporting both DDR 2 and 3 on the same chipset, but this isn't anything like the i810 debacle is it? Where the memory controller ended up barely supporting RD-RAM, just so that you could plug in slower-than-anything SD-RAM.

Sleep States

[lmnfrs]

Penryn does C6. I don't know which, if any, requirements are satisfied in current boards.

The subsystems of the board (buses, controllers, GPU, etc.) need to function by themselves while the processor is off. I'd imagine there are also certain hardware requirements to bring the CPU out of C6 that the new boards provide.
The average enthusiast probably doesn't need outstanding battery life, it's just a nice extra. But for business/professional uses, this is a very welcome development.

What's with these codenames?

[p4rri11iz3r]

Seriously? What is the point of giving new hardware/software codenames? We've all seen "Longhorn", "Revolution", and others and nobody ever said, "gee, I wonder what that is?" Why can't they just say "The next windows", "The next Nintendo", or "The next Intel Chip?" Damn marketing FUD... Sorry, needed to rant.

Re:What's with these codenames?

[Anonymous Coward]

Code names aren't supposed to be secretive. They are the names by which the company refers to a product that has not received an official name.

Re:What's with these codenames?

[justinlindh]

Are you serious? It really hasn't occurred to you that there are almost always future projects in the pipeline, and simply resorting to calling one the "next" would be completely ambiguous?

Code names are mostly for internal use anyway. It's a way of referring to a project before marketing gets a hold of it and names it something officially.

I can't believe I need to explain the purpose of code names to someone on Slashdot.

Re:What's with these codenames?

[edwdig]

Because "The next Intel Chip" isn't very descriptive. They have several designs in progress at once. It gets confusing when your project is "The next next next Intel Chip."

Also, not all designs get released. It confuses all references to "the next next next Intel Chip" when "the next next Intel Chip" gets canceled (see 4.0ghz P4). Likewise, "The next Windows" isn't very descriptive, as MS has separate desktop and server lines. Windows ME would've thrown everything off, as there wasn't originally supposed to be a product in that spot in the lineup.

Re:What's with these codenames?

[Anonymous Coward]

The real reason for codenames: we developers expect project t-shirts with the codename :-)

ECC?

[eviljav]

Did anyone happen to see if these chipsets support ECC ram? The hardware review sites didn't mention if they do or not, but that's usually not something they seem to check. (Or if they do check they just whine about not being able to overclock or something)

Minor nitpick: 1333 is CPU bus, not RAM.

[Anonymous Freak]

This new chipset supports 1067 MHz DDR-3 max. 1333 MHz is the CPU bus speed. This chipset will probably be revved to officially support 1333 MHz RAM, but not yet.

But, as many have already discovered, the previous P965 chipset can be made to support DDR-2 faster than its specced 800 MHz, and processors above its specced 1067 MHz, so 1333 MHz RAM will PROBABLY work just fine with minor BIOS tweaking, but its still unofficial.

I'm waiting for X38, with its dual X16 PCI-E 2.0 slots, among other improvements.

Re:Yawn

[Colin Smith]

WTF? Are you an astroturfer or something? This has to be one of the most boring product announcements ever made. I can only assume it's a paid advertisement, it's certainly not news.

Re:Yawn

[hcgpragt]

I'm so excited!

Much better technical & performance analysis h

[Anonymous Coward]

HotHardware.com has significantly more technical and performance analysis, here. The ExtremeTech article is pretty light with just marketing regurgitation....

So AM2 is the right socket for Linux?

[Anonymous Coward]

I'm about to order an asus M2NPV-VM board w/ athlon64 x2 3800+ cpu for my new Linux box.

No

[Wesley Felter]

Of course they don't support ECC; only servers and high-end workstations deserve ECC according to Intel.

Yeah I too wait on even better.

[DrYak]

Anyway, it would be nice to have a broader upgrade range. While AMD's pattern is superior, it's still far from ideal, especially when they too have socket variations.

I agree on this point. Athough, as I said, there's aa good commitment coming from AMD of stabilising the AM2/AM2+/AM3 family, we could hope even better.

Now that the on-CPU-die memory controller has definitely decoupled the CPU/Memory (the fast evolving part) from the northbridge/motherboard (much more constant - except maybe for the graphical connector), it could be realy fun of having motherboard with HTX connectors and CPU/memory slotcket which communicates with hytertransport. May even give good possibilities for graphic cards (stream engines on the same hypertransport bus as the CPU. Cool). Also same multi-HTX board could be both used for small class multi-CPU servers for small offices / home offices AND high-level gaming rig. Or server grade multi-HTX server motherboard could be used for hardcore gaming machines.

Improvement but not much.

[AbRASiON]

I just read the techreport article this morning.

First off the PCI-e 2.0 support is apparently in the X38 'enthusiast' chipset so that's one scratch.
Also the P35 seems fairly good with DDR3 but it's a hell of a price premium and certainly not an insane speed bump.
On top of this, the P35 supports the 45nm Penryn CPU, guess what? the 965 chipset also supports Penryn if the boards are designed with this in mind, some may not work, some may only need a bios update - but you will see Penryn working on 965 boards soon.
Then there's cost - the DDR3 I mentioned plus the board itself is expensive.

I'm sure in 9 months we'll all be using these and the 965 will be gone, however right now it's not something even enthusiasts should care too much about, wait for the X38 which may (partially) future proof your machine or at least wait for a price drop.

I'm sure in another 6->12 months we'll see yet another Intel chipset with improvements (surprise surprise!) hopefully the next round will be something so amazing that it's simply a no brainer to purchase.

Re:Yawn

[AKAImBatman]

My first reaction was the same as yours. The problem is that you can't be negative about everything, or you'll be negative all the time. Thus I decided to take it at face value and try to be positive about it. Ergo, happy bubbly (mildly sickening) post about colorful motherboards. :-P

AMD fanboy alert

[Anonymous Coward]

Get over it, AMD is dying. Intel rose from the ashes with the Core chips that everyone, including Arstechnica, recommends.

You must be new here

[Overly Critical Guy]

This is Slashdot, where all companies are evil, and conspiracies rule the day. Oh, and AMD is always better for some reason.

Re:But I'm getting ready to build an AMD Linux box

[Anonymous Coward]

Simple fucktard.
First of all I call bulshit on you using Linux as you seem too stupid to even exist let alone use a computer.
Secon Intel works with the open source community while Awful Micro Devices works with Micro$hit. Nvidia has drivers that works for Linux, while Awful Technologies, INC. avoids open source.

ATI is pwned by AMD, and they both hate Open source since they have engineers that are like you, too stupid to even exist let alone use a computer. So why don't you and all at AMD/ATI do us a favor and go find a cliff or a bridge somewhere, then take your entire fucktarded families. Have all of them jump off to their deaths and after that jump to yours. Then there will be a whole lot less fucktards in the gene pool.