AMD Shanghai is officially out

[Particle]

Nein. AOD used to have fonts that were like that.

[Macadamia]

I thought was was a photochopped one, since the 4006 and 4009 are quite obviously different sizes.

This is an actual AOD bug, which actually confirms the screenshot's legitimacy, right?

But even if you ignore that, my link is close to that too.

[Jazzman]

The AOD version appears old and the cpu-z version reads the voltage wrong. Mainly, I put the ss up more for entertainment and pushback. There are three or so sources that show similar results but usually the hsf is huge and the test is on an open bench setup or just a standalone board. There is anecdotal evidence to suggest that 4.0Ghz is doable, load stable. My belief from everything I have read is that the chip's tdp is lower than current phenom's, and certainly much lower than Nehalem. I think the Hex test of Nehalem was also on a bench and so temps with a case will probably be higher. Even in the Hex test Nehalem, running at stock clocks with the stock hsf, had temps at 75c under 100% load. It's a great chip just hotter than heill.

[villa1n]

The AOD version appears old and the cpu-z version reads the voltage wrong. Mainly, I put the ss up more for entertainment and pushback. There are three or so sources that show similar results but usually the hsf is huge and the test is on an open bench setup or just a standalone board. There is anecdotal evidence to suggest that 4.0Ghz is doable, load stable. My belief from everything I have read is that the chip's tdp is lower than current phenom's, and certainly much lower than Nehalem. I think the Hex test of Nehalem was also on a bench and so temps with a case will probably be higher. Even in the Hex test Nehalem, running at stock clocks with the stock hsf, had temps at 75c under 100% load. It's a great chip just hotter than heill.

I dont doubt the mhz is possible, just that screen doesn't seem like the best evidence. Glad to know though the AOD checks out in that regard though, havent used it yet so didn't know about that bug :P. Well if shanghai's power consumption is on, it seems like that could be the case. It does seem that nehalem + voltage = a steep temp curve, which is maybe a good reason for the tdp guards etc they have, although it seems stable at temp. Time will tell, it could also just be early steppings... remember the early q6600's vs G0 steppings, and that was a much less complex chip. I think clockwise deneb should be able to hit the same clocks as nehalems are... and maybe a bit cooler, i guess we ll see. lol I m torn, i want both platforms to play with.

[The Coolest]

From some "internal" screenshots that I've seen, retail Shanghai is Rev C2.
So I assume Deneb should also be at least at Rev C2 when they release it.

[Hornet331]

i wonder why noone has posted this...
http://www.eetimes.com/showArticle.j...2002243&pgno=1

They also have some transistor performance numbers in there, to bad its only for Pmos.
Another interesting figure is that the say Ion/Ioff is 10 times better then on the intel proces... to bad there are no numbers.

I hope we see a complete list of the performance indicators for the transistor.

[informal]

i wonder why noone has posted this...
http://www.eetimes.com/showArticle.j...2002243&pgno=1

They also have some transistor performance numbers in there, to bad its only for Pmos.
Another intresstening figute is that the say Ion/Ioff is 10 times better then on the intel proces... to bad there are no numbers.

I hope we see a complete list of the performance indicators for the transistor.

I already posted that link 2 days ago

[Hornet331]

I already posted that link 2 days ago

nothing showed up in search, not for the link nor for eetimes.

[informal]

nothing showed up in search, not for the link nor for eetimes.

Try harder :
http://www.xtremesystems.org/forums/...1&postcount=48

[Hornet331]

Try harder :
http://www.xtremesystems.org/forums/...1&postcount=48

lol no wonder when the url ist different.

[informal]

It is filed under "news" section on eetimes website because it was err news...BTW it's the same article and on top of that you have a whole quote from it in my post...
You could just type eetimes in search box and my post would be in top 3 results...Simple as that.

edit:actually it IS a first result ,before your 1st post above ,that contains eetimes in it...So you can't use search properly...

[accord99]

Hex had a icore7 on a TRUE hitting 83C overclocked.

phenom over clocked only hit's about 60C and lower wit ha decent cooler.

Seeing as how it's often not possible to compare the temperature readings of different stepping CPUs from the same company, I doubt the temperature readings of two completely different CPU families from separate companies really much, especially with reviews that have Nehalem system power consumption comparable to Phenom systems.

[Donnie27]

http://www.eetimes.com/showArticle.j...2002243&pgno=1

Intel's Nehalem is due out soon, but many of the features of what has widely been touted as a "new" architecture are only new to Intel. AMD's previous-generation Barcelona was the first quad-core processor and also included 2 MB of shared L3 cache integrated on the same chip with a Northbridge memory controller. AMD's latest design, Shanghai, triples the L3 cache size to 6 MB, but all the basic building blocks were already incorporated in the 65-nm Barcelona chip.

I'd be ashamed to link to such bad and Mis-information!
Lame since IMC is not "NEW" to Intel LOL!
L3 is OLD NEWS to Intel.
Monolithic Quad-Core processors at 65nm were a bad Idea and Intel was PROVED right for not making the same rushed mistake AMD made. Intel went MCM at the right time (eventhough both cores still sucked) and it seems went *Monolithic at the right time. If it were the other way around many here would be talking about how dumb it was for Intel to try *it @65nm.

Intel maintains a lead in process technology. AMD had a lead in architecture, but Intel's Nehalem will offer many of the innovations AMD introduced, along with a few of Intel's own. Whether you believe AMD or Intel is better at a given point in time, it's clear that competition is driving innovation, resulting in better products and more choice.

First see above?

How are they in the lead architecture wise when they're slower in MOST cases? Point to Point and IMC is AMD's saving grace (thank goodness since NO ONE needs an all powerful Intel without competition). Without the EV6 based Intel licensed tech, AMD would have died years ago. AMD will kick Intel's A$$ when its CORE/S improve IPC wise. When CSI hits the market, that's (core IPC improvement) what they'll need more than anything.

[Katanai]

...

Well you got some points there but I hope this will not start the flame war again. Keep it civilized guys...

[Donnie27]

Well you got some points there but I hope this will not start the flame war again. Keep it civilized guys...

No need for flames, I only gave facts!

Then this one will. RIGHT NOW, Deneb is ahead for my next upgrade! Not because some mis-informed writer. Not because of some misplace Fan loyalty, but from common sense. Example. AMD ships a 2.7GHz X4 Deneb that's the same clock for clock speed as an Intel i7-920. Let's say the benchmarks show AMD has similar performance. Unless Intel can blow them away (I doubt BTW) AMD has a great case for improved sales. IF they ship it with a similar price they'll do well IMHO. I'd buy one if they did. They'll already have cheaper RAM and Motherboards. Less memory bandwidth? Since most Desktop apps don't depend on it now and in the foreseeable future it will still not matter in most cases. Running 8 threads and heavy multitasking isn't that important to me either That's why I'm using a Wolfdale right now.

If things don't improve with the economy though, unfortunately neither will get my business.

[Rammsteiner]

If things don't improve with the economy though, unfortunately neither will get my business.

But z0mfg you n00b! You should spend to boost economy

Well, that's what Im telling myself to justify my hardware addiction