Then its all about the cache so to say :p:Quote:
Originally Posted by Macadamia
That also explains the "unrealistic" 25% increase from Phenom to Phenom 2 in this test. Not even to talk about the 100% change to the 7000 series Xeons.
Then its all about the cache so to say :p:Quote:
Originally Posted by Macadamia
That also explains the "unrealistic" 25% increase from Phenom to Phenom 2 in this test. Not even to talk about the 100% change to the 7000 series Xeons.
The first 2 AMD Phenom II that will be released on 9th of january will be :
Source here and transalted here !Quote:
1. AMD Phenom II X4 940 Black Edition, Part Number : HDZ940XCGIBOX with 125W TDP, on Socket AM2+. The default speed will be 3000MHz with 8MB Cache.
2. MD Phenom II X4 920, Part Number : HDX920XCGIBOX with 125W TDP, on Socket AM2+. The default speed will be 2800MHz with 8MB Cache.
:)
That's what I thought, and obviously that is what is indicated, but I think you both are talking about the performance numbers in the linked benchmark.Quote:
Originally Posted by Macadamia
:confused:
why so much confusion about the zdnet article... numbers just fit fine...
Queen:
Dunnington:
12core: 57432
1core: 5635
Scaling: 10,2
the only odd thing are the julia benches. :p:
BTW, a Chinese website showed a 3.2Ghz QX9770 getting 24k while a 3.5Ghz Q9300 (6MB Cache) @ 3.5Ghz OC get 20k.Quote:
Originally Posted by Shintai
More food for thought? :p:
Nah, that fits the cache theory :DQuote:
Originally Posted by Macadamia
Yes, the core has gone some optimizations such as fiing coldbug and allowing high clocks(dunno if it is due to 45nm or an actual march fix). IPC-wise there is not that much, except 3x bigger L3 and lower latencies(?).Quote:
Originally Posted by Chad Boga
Do you have some info to link on that? It would be nice to know more about the IPC changes.Quote:
Originally Posted by Calmatory
Does anybody know if AMD still uses SOI in 45nm.? It wasn't mentioned in the press release. The lack of cold bug is pretty fishy :)
Yes it is still SOI.Quote:
Originally Posted by kl0012
Yes they do(use SOI).SOI is even planned for high perf.(read server/desktop) parts on 32nm while bulk and highK is for GPUs and low power CPU parts(mobile/netbooks).Quote:
Originally Posted by kl0012
i would really...but don't know if its worth the trouble if only 2 peoples are interested :)Quote:
Originally Posted by JumpingJack
I'm really interested,but those phenomII results from that blog are kinda fishy so it would be pointless for you to waste your time if we can't compare the results.Thanks anyway :).Quote:
Originally Posted by JumpingJack
You clearly misunderstood me.Bulk is for GPUs and highk is for low power CPU partsQuote:
Originally Posted by LOE
<-respectively.Quote:
while bulk and highK is for GPUs and low power CPU parts(mobile/netbooks).
Err... why can't 32nm SOI HP use HKMG too?
Weren't IBM-AMD working in that direction?
Whether the GPUs have HKMG is up to timeframe, since I think HKMG does not need a big transition in terms of design.
Quote:
Originally Posted by STaRGaZeR
I think you need to calmly wait and see, like for all of us there is an NDA on any numbers being published.
Try to remain focused on the main context of the thread, which was Phenom2 is hitting well over 5GHZ chilled. When the numbers surface detailing performance we will start the debate all over gain.
Nothing much, based on rumours AMD saying "up to 25 %" IPC improvements for Shanghai over Barcelona(NOT for Deneb over Agena, however!). Some leaked Deneb tests whow approx. 5-10 % increases over Agena(C1 AFAIK). It is hard to say how much it is faster "on average" because it is solely up to the tested apps. E.g. I could test only with programs which yield biggest gains and claim that "+15 % increases" whereas usual desktop apps(E.g. WinRAR, divx/video en/decoding, photo editing, audio en/decoding etc) show less than 5 % increases on average (NOTE: I am not claiming that those apps would show less than 5 % IPC improvement, that is just a simplified example! And I have no clue about the IPC improvements on different apps, other than that it should be from 5-10 %, quite average I'd say, sometimes less, sometimes more.).Quote:
Originally Posted by G0ldBr1ck
I'd say that +7 % IPC improvement is slightly pessimistic one. I'd look forward to 8-10 %, but it varies between apps so much that it is impossible to prove/tell/check anyway.
I have been seing this statement around the web quite a bit, not sure what its based on but it sounds way out there..........
Quote:
Beyond that, from a clock-for-clock performance perspective, optimizations have been made to the CPU pipeline, including improvements to branch prediction, pre-fetch mechanisms and TLB (translation look-aside buffer) optimizations. In total, along with larger cache sizes (now 512KB L2 cache per core with a 6MB L3 cache) and higher clock speeds, AMD claims the new Phenom II X4 should offer a 30 – 40% performance increase over all.
Branch prediction - now that was one of the things CPU Queens is supposed to be testing!
(But it ended up loving cache more than BP. :p:)
As much as I want to believe it, I'm thinking that is just a wet dream. Now, it would be nice if I was proven wrong in January but I'm not holding my breath.Quote:
Originally Posted by G0ldBr1ck
Any clue on what has been done for the branch predictor? Core architecture as great BP, for example Core architecture can predict branches perfectly up to 64 loops, while K10 can only predict perfectly up to 9 loops.
I think that "AMD claims the new Phenom II X4 should offer a 30 – 40% performance increase over all." does mean that with higher clocks there can be up to(At least thats what they are saying when/if the claims don't meet with the reality, obviously :p:) 30-40 % performance gains(for same price?), not that IPC improvements would be so great. Don't try this, Shi....tttt. :)
WHY everybody respect the NDA from AMD and nobody respects NDA from Intel??????????? :confused:
BTW, for anyone having trouble with trolls or just someone who isn't exactly trolling but serially upsets you, this forum does sport Annoyance-Be-Gone Technology that is quite helpful for staying calm.
---
Yes, SOI and HKMG are not exclusive. I wonder if AMD would see even further benefits to their already spectacular 45nm process if they utilized HKMG in conjunction with their existing SOI goodness.
it seems AMD has rewrote the book when it comes to SOI.........they may end up having no use for HKMG on desktop CPU's.
How many people saw the results? How many enthusiast people care about AMD? People are after Nvidia and Intel everywhere. :p:Quote:
Originally Posted by vbo
Thats BS, SOI and HK/MG are solutions for totally different problems (channle leakage vs gate leakage).Quote:
Originally Posted by G0ldBr1ck
I say we take it easy. We have had too much crap around.
An example: http://www.theinquirer.net/en/inquir...30000-3dmark06
Is the source different?
Yes.
Is it good enough?
No.
So I say too little known yet.
Yes, indeed! I have had to use it a couple of times over the years. I really hate doing it because my belief is that almost everybody has something to share and are genuinely good people, but there are a few on the Internet who only exist to cause trouble. I now have a grand total of ONE such individual in my Ignore list. I'm not even sure he is still a member here, and I don't wanna know. I have been on forums for over 20 years. Actually longer...try "billboards" for those that know what those were. Rarely have I ever had to do something like that. Like I said, I hate being forced to use that function.Quote:
Originally Posted by Particle
I haven't yet used the ignore list. I've come close once or twice, but then that person does something that redeems himself and I don't do it :lol2:
I agree, i also rearly use the ignore function of a forum.Quote:
Originally Posted by T_Flight
My motto on this: "The bigger your ignore list, the bigger is your own ignorance." :cool:
Even I haven't ignored a certain person, while it would be better for my blood pressure at the times when he starts telling lies again and makes personal attacks which staff around here seem to think is all normal and allowed, but then again I get so much satisfaction being able to read his posts and understanding he has it a lot wurse then I do ;)
So yeah, the iggy bin here is empty as well :up:
Never said it wasnt, if they can manage to make good use with low-k by tweaking there process than no use in the change........and hey, it was only a "you never know" kinda statement.Quote:
Originally Posted by Hornet331
+1 My ignore list has exactly 0 people in it because I like to see what everyone says and use my brain to filter out the nonsense.Quote:
Originally Posted by Hornet331
If Phenom II can hit 4GHz on something like a TRUE (PushPull)/low-mid water (ie Nautilus 500) it'd be frankly a huge step up from K10. I am hoping that the Cryo-Z turns up in the UK sometime soon so I can see how far it can push the Phenom II.
What is this, the "Ignore List" thread? Back on topic, I'm very interested to see AMDs strategy when the PIIs finally come to market; a lot of folks were making high stock clock predictions (3.6Ghz), but now that we know that:Quote:
Originally Posted by Hornet331
Phenom II x4 940 = 3Ghz
Phenom II x4 920 = 2.8Ghz
It seems that realistically, the Q9550 is going to be enough to take on the entire PII line. Is this too pessimistic of a view?
It was on topic too long, we had to take an OT break :rofl:Quote:
Originally Posted by Zucker2k
Back to your feature presentation :D
Probly all but the unlocked BE. especailly if the price goes down anymore....bang/buckQuote:
Originally Posted by Zucker2k
I dont care how you guys spin this, but 6Ghz+ (hell, even 5Ghz+) is really meaningful. Even if it was just screen stable or quick bench stable.
The other AMD CPU's can barely reach 4.0 with any kind of cooling, due to the arch itself and severe coldbugs.
IF the clocks are true, than they are an clear indication that the arch is not badly OC limited like before anymore, PLUS its faster clock for clock than phenom.
IF the rumors are true it may match current 45nm intel quads in OC, price/performance, etc...Dunno about nehalen, but it may be competitive with it :)
there were never predictions of high stock clock, only on those fx parts.. perhaps your own predictions :DQuote:
Originally Posted by Zucker2k
and that is based on what?Quote:
Originally Posted by Zucker2k
The PII x4 940 at 3Ghz is a BE part @ 125 tdp.Quote:
Originally Posted by G0ldBr1ck
I don't know if this is premature, but Intel's tick-tock cycle probably means they're going to be one step ahead (and a half if you count the launch date of Intel 45nm yorkfields and wolfdale) of AMD, this is going to be AMD's main problem in the future. It seems Intel can keep Nehalem 940 and up, along with Q9650 and up as premium upgrades while AMD struggles to compete with Nehalem 920, and Q9550 down. If you think about it, AMD has to be worried that Intel doesn't get nasty and slash prices for the lower segment processors.
So while AMDs recent showcase has got enthusiasts like us talking, it gives Intel ample time to plan a pricing strategy to counter any strong move by AMD. I think the global economic situation is such that a company would be better off selling a relatively highend cpu for $150 than not selling at all. These companies know they have to make some type of profit to attract investors or make their shares attractive. H1, 2009 is going to be very interesting indeed.
Aren't you being a bit too bold :p: Some numbers have surfaced in this thread and elsewhere which makes me believe based on stock clocks of PII above, to come to that conclusion. One can never tell for sure at this stage, but I'm willing to run tests with my Q9550 all the way from stock (2.83ghz) to 4.2Ghz. :DQuote:
Originally Posted by duploxxx
Actually Q9550 is interesting for upgrade but for a new build i7 920 strongly concurrence it.
After that without knowing PII price it's hard to speculate. We speak alot about PII 940 BE but if PII 920 BE it's price around 200€ and can match (approx. 5%) Yorkfield clock for clock that's can be the real deal.
Yes, for a new build, the i920 is definitely the cpu to get at the moment since it's in the same price bracket as the Q9550, but the Q9550 has the same advantage as PII, it is a drop-in upgrade. You wouldn't need a new board and DDR3.Quote:
Originally Posted by AbelJemka
then what about this shot?Quote:
Originally Posted by Helmore
http://www.coolaler.com.tw/toppc/AMD45NM/4G.JPG
Thats from months ago, not from this last private demo. and that person no doubt was excluded from the last demo
It's an older shot of a C1 stepping from August.C2 should be noticeably better as C0/C1 usually topped at ~3.6-3.8Ghz on air
Are people really expecting screenshots from private show held by AMD for like 20 people which some half of them are AMD employees?Quote:
Originally Posted by G0ldBr1ck
Are people really not believing that AMD can do at least +5 GHz under LN2?
Edit: Misunderstood possibly. :( Though question remains open. ;)
so that wasn't under NDA?Quote:
Originally Posted by G0ldBr1ck
im sure it was, thats why im assuming they were not envited to this last demoQuote:
Originally Posted by illidan
ok :)
It clearly was a breach of NDA,you don't need to be a genius to figure that one out.It was the only shot of that clock on that particular CPU,nothing else was shared.Now you can say,wow,why didn't he run some tests while he was at it and we will be going in circles.From no CPUz shots of 6+Ghz under LN2 to a CPUZ shot of 4Ghz under air to why no benches of 4Ghz OC to ...Quote:
Originally Posted by illidan
Some facts about AMD PII event:
You have to accept the fact that a dozen of people were at at Phenom II Show.All of those people(from pcper,amdzone,techreport,anandtech(?) etc.) confirmed the chips ran at the presented clocks and confirmed the cooling used(air,water,dice,LN2) and played around with the chips.All of those people are bound by NDA and AMD would know if any of them(20 or so)breached it.THe 6+Ghz statement was pulled later in the day and reverted to 5+Ghz as AMD didn't want to disclose too much details(competitive reasons,not buidling way too much unneeded hype etc.).All the chips/systems ,apart the 6.3Ghz one,ran Crysis and FC2(fuad reported the FC2 runs) and run them in loops without stability problems.AMD confirmed the chips were not specially cherry picked but also that they weren't duds either.They expect retail parts to clock to similar frequencies(your mileage will vary).No perf. numbers from the systems were disclosed although the people at the event saw them obviously.People at the event could play with the chips and clock them .Chips have no cold bug and AMD claims retail parts won't have it either,due to manufacturing and uarchitectural changes.
Does not compute.Quote:
Originally Posted by informal
Does not compute.[/QUOTE]Quote:
Originally Posted by gallag
They tested 20+ chips and found an average?
While all the reasons you give make logical sense, I don't understand the purpose of the demo :shrug: I mean, NDA applies to chosen beta testers and OEMs I presume, but these are people whose jobs are to report the news, right? Why bring in media people and tell them to shut up. I mean this has to be a first; the media under NDA?? :shocked: :shocked: :shocked:Quote:
Originally Posted by informal
Edit: Media outlets that do the bidding of companies should not be trusted. All they have to worry about is report the news, period.
Ummm yes, with the kinds of numbers they use something like 1000 to 10,000 would be an accurate sampling. Man, I thought they were trying to put a lid on this.
they probly cherry picked from a small lot.
not so wise in the ways of the stock market young zucker2k.
Media is under NDA all the time. [H]ard ocp gets a core i7 processor and is told "you cant talk about it for 3 weeks". In that 3 weeks they test the bejesus out of it and drop little hints in various posts. Eventually NDA is lifted and we get 4 articles (ad-farming much? :\). Media go to NDA events all the time and are told you cant talk about this till we tell you that you can, but they do anyways. The idea behind that is just to give journalists something to write about that seems genuine when infact you control every aspect of it.
The funny thing is when a company like AMD says "yeah thats not under NDA" it gets 3 million diggs, and then AMD says "actually we changed our minds you cant talk about that".
But the SKU's tell the whole story. If deneb really can hit 4.0GHz on air I suspect we'll see a 3.2Ghz SKU.
Marketing :shrug:Quote:
Originally Posted by Zucker2k
But it doesn't make sence, well it does create an hype but they will look real bad if they don't deliver. Maybe they want to scare intel in lowering prices in anticipation, maybe they know they have an average product but feel like playing a game, if intel drops prices, they can't jack em up again if deneb dissapoints ( well ofc then can.. but then they would look like the asses ).
Idk, politics and buisiness don't always make sence to me ;)
On the other hand, their getting allot of free press this way, even if it's only among hardware enthousiasts :up:
I for one am looking forward allot to seeing high res vid's of I7's of well over 6ghz running benches ( to show it's just the thermal diode being disabeld and not a totally crippeld chip ).
You know Francois is being attacked for choosing FUGGER who is a known Intel enthousiast, but he already laid out the rules for it and he won't bs anyone about what this cpu is. That's another thing I'm interested in. How high can it run with him behind the scenes working with FUGGER?
And last but not least, Francois got to share a way to disable the diode for every bencher :rotf:
As in.. hey we can't replicate it, so it never happend :sofa:
Media can talk, they just can't talk about all.Quote:
Originally Posted by Zucker2k
Intel let some sites speak about Nehalhem specs way before it launch but they can't tal k about all.
AMD let site preview HD 48xxx but can't speak about architecture.
Suposably intel was only able to hit a max OC of 5.02 with LN2 but had to go to 4.8 to bench at a recent demo acording to INQ for whatever thats worth. Off toppic i know but I keep reading posts about 6+ Ghz OC on i7.
The thermal diode is not disabled in Denebs,that would never happen BTW.I don't know what Francois was thinking when he said that..
It is just calibrated so that it doesn't lock the chip when temp. drops below certain level or go above certain level.That's all.The thermal monitoring works and is not disabled like Francois claims.This was achieved with some changes in design + a big change in manufacturing(apart from smaller node AMD ,for the first time,applied DFM strategy for it's first gen. 45nm parts,which cuts costs and improves yields).
http://i35.tinypic.com/35m08k2.jpg
posted by Jazzman in AMD thread
Francois wasn't talking about Denebs, he was talking about Nehalems. Francois gave instructions to FUGGER on how to disable the temperature sensor on Nehalem to make sure Nehalem wouldn't crap out under extreme clocks and extreme cooling. Now he will pretty much have to tell everyone how to do such tricks, otherwise it would be a useless achievement to be able to clock it that high while nobody else can.Quote:
Originally Posted by informal
Just on the sensor data that could lock a chip.
If you ahve a 8bit value for the sensor data. its usually calibrated like -127 to +127. So imagine it gets under -127. -128 would turn to +127 and the chip would shut down.
So does this mean that Deneb's sensor is better than i7? :p:Quote:
Originally Posted by Shintai
jk! :lol2:
one-tenth multiplier increments? :o
That pic is actually not the original.. the original had all the steppinginformation and such, just like this one has some of it not blacked out:Quote:
Originally Posted by G0ldBr1ck
http://4.bp.blogspot.com/_lJdgLM3WR3...Y/s1600/11.PNG
Someone might have saved it to their harddrive, or mayby google has cached it? If so please post it, 'cause it say's alot more than the two pics above.
I'd guess 29.5 multi, with a slight clock variation of less than 1MHz, over a high multiplier like that it can end up making that 20MHz difference.Quote:
Originally Posted by biohead
i don't get it what's with a cpu-z 1.46 :)
Actually he was talking about Denebs first.He implied AMD was messing with the chips just for the demo and disabled the thermal monitoring thus allowing high clocks.He then said he will give his "special chip" to world class OCer and show intel can do 6/7Ghz clocks by messing with on die sensors(8bit manipulation he mentioned).You can search his post and find this by yourself.Quote:
Originally Posted by Helmore
Those that leak things repeatedly get cut off. They can cut off places access to media kits, and if you are a review site, that's not something you want to happen.
My lord man, if I was under NDA I would button it. It wouldn't be "It didn't happen", there would be no "It"...as in "What never happened?". If if asked the answer would be "Huh?". hehe
Also, NDA means that info is confidential prprietary info, and while places leak it, they could get into trouble with the law. I doubt they'd go that far unless it was a serious breach of something though, but they *could*. An NDA is a binding legal agreement. It's not just that, it's just that when you sign one, you give your word.
People crying about an COMPANY (money, money, money) hyping their products, being biased (lol), aggressive marketing, saying half truths, exacerbating their own products etc etc, its extremely laughable :D
I know i had nothing to add with this comment but some dudes need to get real... We should be very skeptical about anything an company says.... After all, everything they do is aimed at taking money out of our pockets.
An kind, honest, sincere, company, with high moral standards is almost always destined to fail. Sadly, they must work on the edge, sometimes with borderline legal practices in order to compete. If they dont do so, they end up overwhelmed by the voracious competition. The system unfortunately works like this :(
Want it or not, we live in a semi-liberal CAPITALIST world (most of it at least).
Back to the topic: Interesting speculations so far, and dubious clocks...lets wait and see from trusty independent reviews :)
This much more trustworthy than the blogspot one. 1.6V to reach 4GHz, just like in the recent AMD demo. :up:Quote:
Originally Posted by illidan
:shakes:
Quote:
Originally Posted by G0ldBr1ck
http://i37.tinypic.com/2rm1c20.jpgQuote:
Originally Posted by Mads321
http://4.bp.blogspot.com/_lJdgLM3WR3...0/PICT0470.JPG
Well for keeping their data coherent, the screen of the "system" for that pic, and the desktop seem to match up. Although it looks like they are french (canadian?) given the "Mes Documents" on the desktop. But i do see orthos, and a cpuz file as well. Anyone confirm that card is ATI? Just the 29.5 multi on that score is quite incredible..no htt bump... hard to buy into.. but to their credit, this pic does support the claim moreso, than discredit.
The reason for this :shakes:
Just look at the CPUZ window.. notice the "over the top compression effects"? Like the gray window area looking pinkish a lot.. Its seems to happen a lot on the CPUZ window only while the others not much affected. The GPUZ looks clean, while Orthos has a slight "pink" on its left border. :rolleyes:
Notice that? :shrug:
Cpu-z folder on the desktop says 1.40, program says 1.46, no validation ... I don't buy it.
Blogspot debunked. :down:
Original picture..
http://www.ixtremtek.com/ixtek/Overc...0/PICT0470.JPG
Original source... http://www.ixtremtek.com/forums/showthread.php?t=126
LOL! So much for that blog. :ROTF: :rofl:Quote:
Originally Posted by Ghostbuster
Nice catch Ghostbuster
Yeah, hard to believe CPU-Z 1.46 supports Deneb. :shrug:
I zoomed in on that picture and looked at it, all the gray windows have a slight pink mixed in them. The picture I saved of the edges though is BMP to prevent any further JPEG losses but it makes for a large file, too big to post :rolleyes: But the CPU-Z and Orthos windows are the same with the pinkish tint to them. The GPU-Z window is lacking the pink.Quote:
Originally Posted by Ghostbuster
LOL @ blogspot :rolleyes:
The Sinus Julia results look totally bogus. The AES results look pretty bogus too.Quote:
Originally Posted by Hornet331
http://www.amdzone.com/phpbb3/viewto...147957#p147957
http://www.amdzone.com/phpbb3/viewto...147974#p147974
http://www.amdzone.com/phpbb3/viewto...148012#p148012
Boo.......Blogspot :down:
Another fake attempt.Lol how did they think it would pass?? :confused:
Oh well,next please :)
I knew that pix look familiar, it took me just a while to pinpoint it back.. ;)Quote:
Originally Posted by massman
:rofl::rofl::rofl::rofl:
LoL :D
so bobblemagnifique had the 6ghz deneb in 2007? sweet
One thing is for sure, you at least live up to your name :cool:Quote:
Originally Posted by Ghostbuster
How long will it take before another attempt shows up?
No, it means some b!tch stole his picture and faked some Deneb screen to get some attention.Quote:
Originally Posted by jaredpace
Well, your sarcasm detector isn't working today, is it? :DQuote:
Originally Posted by ownage
Empty battery!Quote:
Originally Posted by azraeel101
Catching ectoplasmic entities is my speciality.. :DQuote:
Originally Posted by Helmore
Just noticed it I guess.. Ever wonder about those ads on that blogspot? Fishy.. ain't it? :shakes:Quote:
Originally Posted by Helmore
The screenshot looks too degraded considering its supposed to be "first" source. And the benchmark figures look way off.. First sign was the processor picture and the olde CPUZ. :down:
Well, whoever's behind this is pulling a fast one... But who? :rofl:
nice catch ghost... i'll wait till the 9th befor beliving anything. :up:
:sofa: :nono:
Definitely fake. 1.46 doesn't support Deneb, thus it couldn't detect the "Revision" field. But in the pic the field has content which is blurred == fake.Quote:
Originally Posted by G0ldBr1ck
lol, The markbench guy has been exposed for his fraud on his own site..........look at comments on bottom of page http://markbench.blogspot.com/2008/1...very-step.html
He's even promoting his blog at http://forums.anandtech.com/messagev...=y&STARTPAGE=1 under the name "wooox". :nono:
Even the guys ar Hardspell picked it up http://en.hardspell.com/doc/showcont.asp?news_id=4500 Someone better warn them before... :rofl:
Gotta love the professional paint shop experts who make those wonderful cpu-z shots
That's a paradox in itself :rofl:Quote:
Originally Posted by GAR
There are no Pros who use Jasc/Corel Paint Shop
Ah lol hardspell could have waited a bit longer and saved themselves from the embarrassment.But i guess it's too late for that :D