You just replace it with a higher performing CSI link to the southbridge. So no. Also the FSB itself uses very very little power.Quote:
Originally Posted by _Lone_Wolf_
Printable View
You just replace it with a higher performing CSI link to the southbridge. So no. Also the FSB itself uses very very little power.Quote:
Originally Posted by _Lone_Wolf_
I agree, you're just replacing the bus, though it may be faster and more efficient, the actual power consumption wouldn't change that much (in fact it might go up a bit since csi runs at much higher speeds technically)
Absolutely, currently 8.4GB to 10.6GB for Desktop and that's not being stressed. I agree with the Multiple Sockets comments, I think the Single Socket comments are blown out of proportion.Quote:
Originally Posted by StealthyFish
Again, I'm not saying the FSB is Good, I've saying calling it a Big weakness is a stretch at best. Sure IMC will be better than depending on the FSB and it alone would kill the FSB's biggest complaint. I know the Two Cores share one FSB link to the MC to RAM or to I/O via the DMI, I've only been playing computers since 1983. Or that the Controller Hub is just like a Network Hub with each device sucking up its portion of the total Bandwidth. Again, total bandwdith that's not being used up and latency is the only real problem.
NO, I'm not an Engineer or Coder. When I was talking about Branchey or Ramdom vs Streaming type Apps. Or games as compared Video Compression respectively.
All I care about are end results, I don't care about the inner-working of Code or Hardware. As one guy said awhile back, I don't care if the processor was powered by a Gerbil in a wheel. If that architecture argument truly meant something, AMD would be kicking the crap out of Intel with its Native Quad Core and Point to Point Topology. I also know Intel is better at Engineering than anyone posting here. They built Smart cache, Smart Memory Access and etc.. to alleviate the negative effects of the FSB/s.
I'm not sure sure how much better this will work with an IMC???
Please, note to shintai, I asked a question and wasn't arguing with you:)
Some times I think some folks "Can't see the Forrest for the trees".
I'm talking about Intel chipsets.And 65nm ones have yet to be released.Quote:
Originally Posted by Shintai
Btw , a NB does use 25w.
975X uses 13w ; P965 uses 20w , P35~15w , X38 around >20w IIRC.Put that on process parity with the CPU and you cut power by 2-4x.
The discussion was if integrating more functionality in the CPU has an effect on power consumption.Let's not derail from that.Quote:
If you note most of the powersavings is from a faster core. 30% less power consumption for the same performance as a Penryn. And if you have a scenario with 10-100% boost (Lowest single to highest multi). Then you have your 30% there.
I'd be shocked if that figure is absolute and not per clock basis.The Nehalem core is much much bigger than Penryn's , there are tons of extra stuff in the core. ( 22mm2 to 29.6 mm2 , that's a 35% increase while HT takes 5% ).Quote:
Originally Posted by _Lone_Wolf_
IMO , I'd expect it to give Penryn a severe beating in single threaded stuff while trashing it in multithreaded multicpu scenarios.
Intel Vice President Kirk Skaugen said theQuote:
that the CPU core performance jump from the same process Core 2 (Penryn) to Nehalem would be higher than the jump for Netburst to Core 2 itself.
http://www.theinquirer.net/gb/inquir...halem-real-big
Remember that Nehalem is the creation of Intel Oregon , the creators of Netburst.The original Nehalem was Netburst taken to extremes , multipipelined , multithreaded everything.
I'm pretty sure some of the stuff created there got into Nehalem v2 too.
P965 uses LESS than 975X. I think you did like theinq in that calculation...Quote:
Originally Posted by savantu
Also P35 uses a tiny bit more than P965.
My oh my...you forget with who you're arguing.Quote:
Originally Posted by Shintai
By Intel's docs :
975X 13.5w
P35 16w
P965 19w
http://download.intel.com/design/chi...x/31015701.pdf
http://download.intel.com/design/chi...x/31696803.pdf
http://download.intel.com/design/chi...x/31305503.pdf
I was going by memory , but your "polite" reply made my look for references.
^
thats only tdp it doesn't say anthing about the real consumption.
And a 130W TDP QX9650 also uses 130W..right?..or is it more about 55-70W?Quote:
Originally Posted by savantu
And funny that P965 boards are the lowest with power consumption of the 3 aswell.
And just for the beating of it.
http://download.intel.com/design/chi...s/31305302.pdf
Thats what you wish to read. The Ivcc is at 7.2A substained. And the VCC is 1.25V for P965. How tell me what that result to.
Quote:
Originally Posted by Zytek_Fan
Unless Nehalem has a "Conroe" effect on DDR3 prices like it did on DDR2 prices. I dont miss $600+ 2x1gb kits! :shocked:
3 channel DDR3? How does that work?
Same way as dualchannel. Nothing special, just with 3 and not 2.Quote:
Originally Posted by Jimmer411
Think raid with 3 disks and not 2. (Very roughly).
That article is pretty wrong in alot of places. Best ignored.Quote:
Originally Posted by Donnie27
Yorkfield is 214mm2 and 820million. Nehalem is less than 200mm2.Quote:
The quad-core, "extreme" performance CPU codenamed Bloomfield, will be identical to the DP parts above, but will be single socket only and have a single QPI link. This will also be on socket LGA1366. It is suggested to have a 270mm² die size, about the same as current Kentsfield’s 65nm 286mm² die, but it’ll have nearly 150m more transistors at 731m
Great. Bloomfield will be used to suck enthusiasts dry :down:Quote:
Originally Posted by Donnie27
That doesn't take into account device variation.Intel needs to take into account that , you might have a nominal 80w chip with 3 sigma variation where one chip burns 70w running a workload and another 110w.Quote:
Originally Posted by Shintai
Add the relevance of the workload and you might have real problem.Who's to say that Burn or whatever test they use fully loads the CPU ?
Intel uses in house build thermal virus and Linpack to show maximum TDP.
Anyway , that's completly off topic and I fail to see the relevance of a family TDP to a specif product from a different category..
We have your assumptions vs. an Intel thermal guide. ;)Quote:
And funny that P965 boards are the lowest with power consumption of the 3 aswell.
And just for the beating of it.
http://download.intel.com/design/chi...s/31305302.pdf
Thats what you wish to read. The Ivcc is at 7.2A substained. And the VCC is 1.25V for P965. How tell me what that result to.
re dumping x86 for ia64...Quote:
We basicly just need a fresh start in both CPU arch and software to leap ahead instead of doing very small steps towards the dead end.
sorry: idea makes too much sense and has been scrapped/delayed :lol:
what is "ibexpeak"?
nehalem to have less cache due to IMC.Quote:
Yorkfield is 214mm2 and 820million. Nehalem is less than 200mm2.
http://www.anandtech.com/cpuchipsets...spx?i=3102&p=2
re ibexpeak chipset: (on lga1160)
http://www.xbitlabs.com/news/cpu/dis...Necessity.htmlQuote:
Since both Lynnfield and Havendale have memory controller as well as PCI Express interconnection inside, there will be no need for GMCH (or North Bridge) on the mainboard. Instead, the new processors will connect directly to code-named Ibexpeak platform controller hub (PCH) that will carry hard drive controller, wired and wireless network controllers, monitor physical interfaces, PCI controller and other input/output as well as platform-related capabilities.
If today’s mainstream personal computers usually employ three chips that feature the core functionality of the system – CPU, (G)MCH and I/O controller – then in the Nehalem era mainstream systems will be based only on two chips: CPU and PCH. Both Lynnfield and Havendale are projected to emerge in the first half of 2009.
still like the thought of tri channel ram:slobber: lga1366
although i wonder the advantage/s of PCH for graphics (non integrated vs integrated)
i see a new build for me in Q2 "09.
The core might be but what about all if the other stuff in the Die?Quote:
Originally Posted by Shintai
I think they may have gotten a bad translation or something. Follow their link? The layouts don't look that far off.
I should have just posted the PC.Watch link:)
http://pc.watch.impress.co.jp/docs/2.../kaigai403.htm
It's linked for reference at Bit Tech.
Another mistake on the Diags, is that there aren't two 16X PCI-E lanes. It was supposed to be One 16X, 8 X 8, and 4 X 4 from what I've read. The link clearly shows 16 X 16.
I'm just waiting for E8500 or something similar LOL!Quote:
Originally Posted by Zytek_Fan
Nehalems cores are not much larger. IMC+CSI doesnt take that much space. And cache is 67% of yorkfield. Trust me. Its under 200mm2.Quote:
Originally Posted by Donnie27
Also Intel would never manufactor a 270mm2+ single die chip today unless it was an Itanium grade.
PCwatch also said it was L3 cache or something earlier and other things. They aint much better.
The biggest Nehalem CPU will be an octo core at 32nm.
Huh ? Nehalem cores are 35% bigger than Penryn's.That huge!Quote:
Originally Posted by Shintai
Nehalem is ~270-280mm^2 , you simply cannot cram 4 cores and a shared 8MB L2 cache in under 250mm^2.
As a matter of fact , Nehalem EX Becton 8 core is a 45nm product scheduled for mid-late 09.Intel said they will demo it this year ; there's no way it could be 32nm ( that process goes online in late 09 ).Quote:
The biggest Nehalem CPU will be an octo core at 32nm.
Nehalem cores aint that much bigger. Also 4 cores and 8MB shared L2 cant be in under 250mm2? Funny since yorkfield can cram 4 cores and 12MB shared L2 into 214mm2.Quote:
Originally Posted by savantu
And call me blind. But this Nehalem core dont seem to be Barcelona size.
http://pc.watch.impress.co.jp/docs/2...9/idf01_01.jpg
Compared with:
http://www.legitreviews.com/images/r.../barcelona.jpg
Nehalem has more logic as % of die area vs. Yorkfield.Also , IMC and CSI take a fair bit of space too.Quote:
Originally Posted by Shintai
You're not blind , but you forget that with different package sizes visual comparisons are meaningless.Quote:
And call me blind. But this Nehalem core dont seem to be Barcelona size.
http://pc.watch.impress.co.jp/docs/2...9/idf01_01.jpg
Compared with:
http://www.legitreviews.com/images/r.../barcelona.jpg
OK, I see. One of them obviously also got very fat fingers and the other slim.Quote:
Originally Posted by savantu
Well you will get wiser with time...hopefully. Its a waste of time to continue.
And just for the record, its not that hard to find the package sizes....
And that's single thread IPC improvement.Quote:
Originally Posted by Shintai
If Nehalem scales to higher frequencies, you'd get a further boost.
Wouldn't that be a DMI link to the south bridge? Intel has used DMI since at least i875/65. IIRC DMI is already a Serialized CSI like BUS/Link or whatever you like to call it. I only saw Higher end Bloomfield using CSI or Quick Path to a DT-IOH (Tylersberg), then DMI to the ICH-10 (South Bridge). That is unless PC Watch blew it and the first link here missed that too, I'll check.Quote:
Originally Posted by Shintai
http://pc.watch.impress.co.jp/docs/2.../kaigai397.htm
For Bloomfield yes, since you still have the I/O hub. For the desktop and mobile versions no. ICH10 is the last ICH.Quote:
Originally Posted by Donnie27
I also think HKEPC havent realized it yet. Also DMI is just another word really for a PCIe x4 link.
OK, Cool, thanks!Quote:
Originally Posted by Shintai
Oh....I didn't know that....just looked it up to be sure:Quote:
Originally Posted by Shintai
LinkyQuote:
Originally Posted by Intel (website)
That means that DMI isn't even close to CSI from a technical standpoint....isn't it.....
Well we learn something new every day....
Precisely. And thats also why it has outserved its purpose. Specially with 6+ SATA ports now that is a peak of 1.8GB/sec in theory. Plus USB, turboflash etc etc.Quote:
Originally Posted by Helmore
And I´m sure its also easier and cheaper to just link the ibexpeak with CSI/Quickpath. Faster, less traces and most likely better electric properties.
When thinking about it. I would also make a quite real and possible statement that nVidia is out of the chipset business in 2010/2011. VIA knows this and already abandon Intel chipsets. Because look at the Ibexpeak. Its basicly the same AMD will have with their cores in 2010 or so. And the IGP market is also 100% AMD/Intel with on CPU IGPs. I think this is also why AMD puts less and less effort on highend discrete GFX. And another bold statement would be that around 2013-2015 everything the Ibexpeak and AMD equal will be on the CPUs. That can yield some seriously small corporate/parents desktops with a CPU+memory+SSD basicly :D
Quote:
Originally Posted by Helmore
DMI was one of the first fruits of the PCI Sig and that is where it came from. The DMI link is merely a different variation of a PCI-E:D Even the link you posted hints at that. This tech is the root of CSI and I believe if Intel felt it needed it sooner, they would have went with it.Quote:
IIRC DMI is already a Serialized CSI like BUS/Link or whatever you like to call it.
Quote:
Originally Posted by Shintai
i really hope that doesn't means the death of highend discrete cards, cause i doubt we see anythign compareable to a 8800 on a cpu soon.
It won't, not for the next 10-15 years and the graphics chip inside the CPU can always be used for physics calculations.Quote:
Originally Posted by Hornet331
As always even if Geeks tried to make it something it is not, the highly integrated CPU's have always been for Budget systems. That's what Timna was meant as right from the start. Look at the rest of the Layout of the platform for it and it even screams Budget.Quote:
Originally Posted by Hornet331
I´m quite sure the CPU part will handle that...always...physics is and have always been a flop on GPUs ;)Quote:
Originally Posted by Helmore
The only thing it will do is sit idle or do 2D.
I wouldnt say 10-15 years. Looking at how relatively close we are to true raytraycing and abundance of CPU power along with maximum useable resolution in comparision to the human eye. I would say 6-8 years.Quote:
Originally Posted by Helmore
nVidias biggest fear is this. Since their goldmine chipset business is already doomed.
However AMDs and Intels biggest fear is to end in the gadget segment sooner or later.
If these slides are true, then Nehalem looks nothing short of amazing.
The only thing I pray that Intel really works hard on is the IMC. Having a robust 3 channel DDR3 controller would be amazing. By robust, I mean not finicky like that of beloved Athlons of years gone past, but easily overclockable and not requiring 900 bios options to get it stable.
They know how to do it on their chipsets, so I have no worries there...Quote:
Originally Posted by MaxxxRacer
MOST of the time Intel shoots for stability first, at the cost of speed. But what good is fast if its crashing.Quote:
Originally Posted by MaxxxRacer
They're probably well on their way to that by now since they demonstrated Nehalem booting Windows at IDF.Quote:
Originally Posted by Donnie27
True! But this still relates to that other thread that caused all of the fights. If AMD has nothing and Penryn is selling what gives Intel the Push to NOT delay Nehalem for at least the Desktop?Quote:
Originally Posted by Zytek_Fan
Link
Intel Unveils 16 Next-Generation Processors, Including First Notebook Chips Built on 45nm Technology ›INTERNATIONAL 2008 CONSUMER ELECTRONICS SHOW, Las Vegas, Jan. 7, 2008 – Intel Corporation unveiled 16 products today, including the company's first 45 nanometer (nm) processors for Intel® Centrino® Processor Technology based laptops.
See? That is Clearly the Reason most reasonable posters here wanted to see AMD do well with Phenom. That's really doing well and not merely pretending they're doing well. If they don't we'll more than likely see another delay caused by Obscure errata or some other lame excuse to milk more money from the Computer Public. This ain't bad since Penryn is Kick-A$$ but Nehalem will be even better. I know folks who're already waiting on it.
I hope AMD can put out a bug free K10 and get it up to speed. We need competition :up: :clap:Quote:
Originally Posted by Donnie27
I'm not sure if that will give enough competition for Intel to hurry up on Nehalem, maybe K10 on 45 nm. might do the trick....but that's still a year away and thus to late....:down:Quote:
Originally Posted by Zytek_Fan
Yet, if you don't lie for them, you have folks calling others FanBoys.Quote:
Originally Posted by Zytek_Fan
Intel doesn't have to hurry Nehalem no matter what AMD does. I meant Intel would stick to schedule and ship it later.Quote:
Originally Posted by Helmore
I guess people just keep telling themselves it isn't as bad as it seems.Quote:
Originally Posted by Donnie27
Yep, Nehalem will still launch late Q4 08 and Q1 09 (someone said Q2 09 now though... :shrug: ) no matter what :up:
Probably early server CPUs late this year, followed by limited desktop availability early next year...
The 3 dates are for 3 segments. First server/Extreme in Q4 08. Then Performance/mainstream in Q1 09. And value in Q2 09.Quote:
Originally Posted by Zytek_Fan
So for those not buying extreme or server here. Nehalem comes in Q1 2009.
I'm gonna try to grab Extreme :DQuote:
Originally Posted by Shintai
i'll probbly have a penryn for at least one year b4 going nehammel- ioften mistakenly typo nehamel. must be dislexic.
how many years b4 multicore cpu's are mainstream and super cheap.
q'2140' @ <100 bucks mmmm.
i reckon 2011:hehe: 22nm sandybridge projected then... holy moly
...but nehamel...int graphics on core could be a v nice thing to have if it can be used in tandem with g card....yes: old news but a nice step forward for value adding to a cpu line.
http://www.dvhardware.net/article16469.htmlQuote:
AMD's Chief Technology Officer Phil Hester said in an interview that AMD's Fusion project will be in production only in late 2009. This means it won't be available until 2010, which is later than the company initially expected.
I'm gonna go cheap as possible for high end stuff for my Penryn rig...so no SLI, CrossFire maybe...
mm cheap and more often. hopeful, although doubtful of the q9300...thinking that 3.0GHz q9300 [400x7.5?wishful thinking???] > 3.2 GHz q6600; i dont have uber cooling.
i think i'd be happy with a q9300 at 3.2ishGHz...and running cool aswell.
pre nehalem.
I heard the same thing Shintai posted!Quote:
Originally Posted by Zytek_Fan
Same here! WOW, agreeing with everyone LOL!Quote:
Originally Posted by adamsleath
...also might be tempted by q9450 if only for the huge amount of cache [12mb]...just have to waitnsee..teehee the next month or two gunna be raveings about wolfdales...wunda if i'll survive.
This CPU will really push the poor x86 architecture to the limit. Even x86_64 is just flirting with the law of diminishing returns.
Open-source to the rescue! Hopefully Microsoft will recognize its place and be mercifully beheaded by Google in the online advertising market.
Seeing how AMD has stated they are cutting down on R&D (which honestly I think is the worst possible move they could do right now), I don't think we'll be seeing nehalem for a while anyways. Hopefully after fuzion is launched (assuming amd is around to do so) both companies will start looking at new possible architectures, would be interesting to see how long it would take for the software guys to catch up if we do finally make the necessary switch
Quote:
Originally Posted by oshox
Well, 64bit will tide us over nicely for the next couple decades till quantum computers go mainstream.
But, x86 is growing to be a significant handicap.. I'm sure power consumption on P4 would have been a lot lower if it wasn't spending all that time sorting and translating x86.
Any chance Intel might revisit idea of moving consumers over to IA64 (ie Itanium)?
Oh and does anyone know Nehalem core specifics... ie detailed specint/specfp, actual IPC in single-threaded apps? I'm sure server/3Drendering folks will appreciate the 8 cores/16 threads, but it would be nice to know if there will be any benefit for the 95% of consumer apps/games that aren't multi-threaded.
IMHO, >2 cores for consumers is a waste of resources/money... I dont think mainstream is silly enough to catch onto move to 8 core/16thread.
EDIT:
Is anybody actually really excited about faster CPUs? What's the point?Quote:
Originally Posted by adamsleath
Games are all GPU limited, and market is moving from PC to consoles. And even the slowest Wolfdale (2.66Ghz) blazes through everything consumer joe throws at it. At times like this Microsoft would usually come to the rescue with the next killer up, or some newer version of Office bloatware.... where is the next killer app? what is it?
motion capture web avatars with 3D ray-trace rendered hair, individually per-pixel shaded and lighted?
Once 64 bit becomes mainstream, then I could see why Intel and AMD would ditch x86, but it will be a very long time until 64 bit becomes mainstream.
x64 is dependent on x86. x64 is like peeing your pants in a cold winterday. it was the easy and quick solution, and everyone loves compability. But it wont take us anywhere as such. We simply need something new instead of trying to keep on patching the old.Quote:
Originally Posted by Zytek_Fan
so everything is in the cpu
Something new as in Cell right?Quote:
Originally Posted by Shintai
ppc isnt new. :p:Quote:
Originally Posted by Zytek_Fan
nehalem spi 1m run.. don't know is it true.. but it's nice score :D
http://infomars.fr/forum/index.php?showtopic=1524
Now now, that is not what XS stands for!Quote:
Originally Posted by MaxxxRacer
:stick:
I believe that A64 clocks RAM as well as it does just because it has those 900 bios options for the user to tamper with. I'm hoping Nehalem's platform will offer atleast 901 bios options to play with.
;)
Nothing has been released yet.Quote:
Originally Posted by ***Deimos***
BLASPHEMY!!!Quote:
Is anybody actually really excited about faster CPUs? What's the point?
OMG!!! :DQuote:
Originally Posted by illidan
That's 2x the performance per clock of Conroe in SuperPI!!
And check the cache size : 64KB L1 and 8MB L2.That's what I wanted to see!! :yepp:
hehe, the CPU-Z shot looks even less believable there then the other one that was posted, but I have no clue and B-0 is available.Quote:
Originally Posted by illidan
I'm really hoping the cache sizes are true in this second photo ..that's all. :)Quote:
Originally Posted by Blauhung
Me too:DQuote:
Originally Posted by savantu
That does look convincing :D
But at the same time backward compatibility to existing equipment/software is extremely important, and vice versa the future upgrade path.Quote:
Originally Posted by Shintai
The same reason is the cause of dominating/monopolistic technologies such as Ethernet, IP, Windows, x86 and GSM.
3 way channel?
return of hyperthreading?
fun stuff.
Maybe I'm wrong, but if you check that cpu-z screenshot you will see:
FSB 444
Multiplier x6.0 (6-10)
Maybe that SPI score is at 4440MHz, because at idle SpeedStep is working.
But Nehalem does not use FSB so...
The CPU-Z on this shot looks very outdated (somewhere ver. 1.3x) and the verification of the SPI checksum says Incorrect checksum!. ;)
:hehe:Quote:
motion capture web avatars with 3D ray-trace rendered hair, individually per-pixel shaded and lighted?
such lofty goals :hehe:
sry i get less excited the older i get :/Quote:
really excited
it seems the games i like such as total war and simcity type things are cpu limited rather than graphics limited.
alan wake will utilise quad core so....i think i want one besides having more things open and all that jazz could be handy.
if they stop developing games for pc i'll cry :'(
Failed maths? ;)Quote:
Originally Posted by savantu
I see about 16%-17% /clock compared with Yorkfield which is pretty decent.
he said conroe though... not yorkfeild
Woo! A.Wake forgot about that! can't wait :D
It's supposed to come out late this year :DQuote:
Originally Posted by Craftyman.
I refuse to believe QC Nehalem runs in early steppings at 4.44GHz. I'd say that is an ES with unlocked multipliers.Probably , it had a 266MHz FSB clock and was OC to 444 , multiplier dropped accordingly to 6.Quote:
Originally Posted by STaRGaZeR
I mean wtf , 20% better than Core per clock and running 40% faster ?
That's 70% better performance than Core.
See above.I must say , I'm perfectly satisfied with only 20% per clock and 40% higher frequency from the start .I must admit such a scenario would be really bad news for the green team.Quote:
Originally Posted by mAJORD
Because SuperPi validation service is broken atm.Quote:
Originally Posted by fellix_bg
Remember the part where Intel said that Nehalem is the biggest change in CPUs since Pentium Pro? :DQuote:
Originally Posted by savantu
And it is very possible that 4.4ghz could be achieved with early silicon. 45nm production was ironed out with Penryn, and remember when we first started seeing results for Conroe? 4ghz was achieved quite easily :up:
ah RPGWizard, you forget that not everyone is like us :yepp:Quote:
Originally Posted by RPGWiZaRD
sell your existing pc when transferring to nehalem,
pick the right time when the average pc user sees your 4Ghz oc'd pc as a godly machine and the new sockets still haven't hit mainstream
I've free upgraded my pc everytime,
eg. I've went from a 7900 to an 8600gts to an ati 3850 without spending a penny ;)
it's all about timing :up:
I think its wrong to say the biggest change is anything regarding the cores themself. What Intel is saying is the move of PCIe, IMC and for lowend GPU to the die/package is the biggest step since the L2 cache that was moved to the PPro.Quote:
Originally Posted by Zytek_Fan
I believe that 4440Mhz can be achievable with the actual silicon revision. Nehalem is Core based, and Core, in its beginnings (65nm) clocked like hell.Quote:
Originally Posted by savantu
If that SS is not fake it is a huge boost IMO, and I'm more than satisfied. But I'll give my final judgement when Intel launches them. Or at least when they release the final specs.
Bad news for the green team since when, almost 2 years? And we will have bad news for another 2 I think...
Known fake...Quote:
Originally Posted by illidan
http://www.xtremesystems.org/forums/...6&postcount=21
Wow, that's quite old then :rofl:Quote:
Originally Posted by Vapor
Savantu, Suggesting it has 70% or whatever IPC improvment over core is as ridiculous as the stupid expectations people had for Phenom. Only it's you getting excited this time.
Face it, it's either clocked to 4.4 ghz (impressive), or it's a BS screenshot.
It's a known fake,as Vapor said it.It dates back from 2006,i now remember that thread i posted in it with my invisible ES keyboard :D => lookie hereQuote:
Originally Posted by mAJORD
BTW,topic title was,back then :"Intel's newest ES chip". LoL
Here's the original quote(the date was 06-29-2006, 08:29 PM):
The link to the post:Quote:
Originally Posted by PkG.1337
http://www.xtremesystems.org/forums/...6&postcount=21
Savantu,in future don't jump so fast :D.
Shintel on the other hand took this "SPi record" and "Nehalem" screen shot quite soberly :)(at least better than the rest of intel fanclub).
EDIT:just to add that it's needless to say Nehalem taped-out quite recently so from that perspective there is no way(at all) this "screen shot" could be real.That dude was just messing around in a thread that started with a joke("Insane inside",first post in that thread).
PS I still type on that ES keyboard of mine,it's still way better than anything out there :D j/k
sorry this is replying to a post you made over a month ago, but i've just read the thread..
i'm not sure what you're thinking will happen in 2010/11 but i think it looks pretty good for nvidia in the short term.. just referring to the performance/mainstream, cos i'm not sure how things will happen with bloomfield-Quote:
Originally Posted by Shintai
making a southbridge for an IMC cpu should be a lot easier than making a chipset that has to host a memory controller and route all the information between the controller and the CPU via a front side bus, like LGA775, and nVidia might have an opportunity like they did with socket 939 where they had the best chipsets around. the on-die PCI-e would be connected to a nVidia chipset that would then connect to two or more SLI capable PCI-e slots... and nVidia get to keep their SLI chipset monopoly with very few issues, just like they have it on AMD systems. with the latency improvements from an on-die PCI-e link, say it was PCI-e 2.0 x16, i'm pretty sure that two high end nVidia cards would do very well... more than two might get a bit stressed. depends how much bandwidth Intel gives to the on-die PCI-e on the mainstream part.
onboard GFX would also be pretty easy for nVidia with the mainstream nehalem - an onboard GPU connects directly to the on-die PCI-e controller, or is disabled if a seperate GFX card is connected. hybrid SLI would be a piece of cake. nVidia's integrated GPU business will compete with nehalem's on-die GPU eventually... who knows how that'll pan out, but it'd be amazing if nVidia was pushed out of integrated GPUs. an on-die GPU may have some advantages, but nVidia could make a low powered GPU (maybe with its own memory, like the new ATI AM2+ GPU chipset) that'd run rings around any forseeable AMD or Intel on-die GPU with a flick of their wrist. just pair the nVidia onboard GPU with a "on-die GPU"-free nehalem. presumably there'll be some (heat/overclocking?) advantages (at least initially) to having a "on-die GPU"-free nehalem, and many people may prefer to get a plain nehalem with a nVidia GPU chipset over a gfx-nehalem for certain builds (HTPC for example).
it depends how powerful the on-die GPUs from AMD and Intel will be. i'm pretty sure that to start with they'll be absolute rubbish, basically because of TDP restraints. and nVidia have already shown that they're willing to give away onboard GPUs by promising to put them on all their future chipsets.
i can see the market becoming more interesting/diverse, but i can't see why nVidia would become a non-player in chipsets and onboard GPUs. and for SLI, so long as there are discrete graphics cards nVidia can cling to their SLI monopoly by requiring it to run through a certain chipset, and i don't see anything that would stop them. raytracing seems to be a much larger threat, but that's more a long-term issue. it's hard to see what nVidia would sell if i could render high resolution games at high frame rates with raytracing on my 22nm native 16 core 32 thread CPU...
that nehalem superpi run might be legit because if it's running at 444 qpi (fsb), and is 1333 from the cpu-z screen shot, that would indicate that it's probably the highend 4/8core nehalem because it uses triple channel ram, so 444*3 channel = 1332 fsb/ qpi effectivly. i think :p:
thx to vapor and informal who clear this up
comon ppl
lmao
Boogerlad,read my previous post(2 posts above yours)..
It's not true,it's dating back from 2006 and it's a joke made in another "joke" thread.
Damn i hate to be a party pooper.
EDIT:
Thanks for joining the party and clearing things up PkG.1337,you made quite a stir with your post back then :D,now history repeats itself again hehe :D.
Your little chop even made it to the Nordic Hardware :),congrats :p:
QFTQuote:
Originally Posted by informal
5min photochop in 2006 posted in a joke thread turn in to this
:ROTF:
Savantu's IPC and clock dreams have just shattered :).But hey,it was fun back then in 2006,it's even funnier now :D.
PS Who's the girl in your avatar :)
:ROTF: :ROTF:Quote:
Originally Posted by informal
BoA, korean singer
Quote:
Originally Posted by informal
Actually, Nehalem probably taped out last summer, because Intel showed a working demo at the Fall IDF if I recall. http://www.anandtech.com/cpuchipsets...oc.aspx?i=3102 16 threads, 8 cores in a 2P system.
Jack
Hi Jack.Yes I'm aware of that and my point was that it(the thread) was still much earlier than the actual tape-out happened.One year earlier if we want to be exact.
Quite smug for someone whose own clock and IPC dreams were shattered a mere 3 months ago... ;)Quote:
Originally Posted by informal
Glad to see you're still around, jack. Loved to read your posts back on Tom's :clap:Quote:
Originally Posted by JumpingJack