Phenom II 6 GHz+ OVERCLOCK

[charged3800z24]

Sorry .

No website has done a thorough review of Shanghai yet,and that's a sad fact unfortunately... AnandTech tried to review it but failed to deliver a thorough review(they stated BIOS issues so i guess it could be a reason).
Techreport's review is late too,should have been finished a week(or so) ago.

i KNOW.. they (techreport) said a week ago that they had been in testing for 2 weeks and will be wrapped up soon.. what happened.. is AMD hold NDA over parts of Shanghai too...? oh well

[RiverRicer]

Does anyone know of a decent site that did shanghai benchmarks..? more then spec etc..

I've been looking for Shanghai benchies almost daily......not much out there.

I managed to find these 'post-launch' results.....a couple of them even include George 'Intel Fanboi' Ou's i7 babblings.....yeah, couldn't live without George's insightful comments, could we?!
http://news.zdnet.com/2424-9595_22-251980.html
http://www.tecchannel.de/server/proz...i/index20.html
http://formortals.com/Home/tabid/36/...6/Default.aspx
http://www.anandtech.com/weblog/showpost.aspx?i=528
http://www.fluent.com/software/fluen....4.x/index.htm

Bottom line.....Intel holds a smaller (than vs Barcelona) lead in single-thread, SSE performance while AMD extends its lead a bit in FP and energy efficiency.

Waiting for Deneb.........

[Boschwanza]

Anybody saw this ?

http://forums.vr-zone.com/showthread.php?t=359342

http://www.freeimagehosting.net/imag...ea01696aad.jpg
http://www.freeimagehosting.net/imag...fcf31169d9.jpg
http://www.freeimagehosting.net/imag...6e1446313a.jpg
http://www.freeimagehosting.net/imag...a4e1257410.jpg
http://www.freeimagehosting.net/imag...b79b46edf1.jpg
http://www.freeimagehosting.net/imag...3bb1b138aa.jpg

Source seems to be expreview, but not longer available

[RiverRicer]

OOOOHH...........tell me this ain't so!!!!!!!!

http://74.125.93.104/translate_c?hl=...egUoJ0XQ7jbIRw

WTF??????????

I hope their talking about a possible delay to AM3 versions, not AM2+!!!!!

[Zucker2k]

Not bad at all, but it doesn't help that the PII is the highest clocked chip and won 1 test. I'd like to see PII vs. Q9650 for clock to clock comparison test. I doubt AMD will be selling the PII x4 940 for lower than $250. That Q9550 will be a real spoiler for AMD, because it's a drop-in chip for s775.

[Oliverda]

OOOOHH...........tell me this ain't so!!!!!!!!

http://74.125.93.104/translate_c?hl=...egUoJ0XQ7jbIRw

WTF??????????

I think that it is just a guessing. Here is the original article. We've already discussed it.

[RiverRicer]

I think that it is just a guessing. Here is the original article. We've already discussed it.

Yeah, saw the date of 11/26 and thought it was something new........feel better now, facial twitching is subsiding........recovering!

[charged3800z24]

OOOOHH...........tell me this ain't so!!!!!!!!

http://74.125.93.104/translate_c?hl=...egUoJ0XQ7jbIRw

WTF??????????

I hope their talking about a possible delay to AM3 versions, not AM2+!!!!!

Ya, I highly doubt Phenom II 920 and 940 AM2+ chips will be delayed...

[charged3800z24]

Anybody saw this ?

http://forums.vr-zone.com/showthread.php?t=359342

http://www.freeimagehosting.net/imag...ea01696aad.jpg
http://www.freeimagehosting.net/imag...fcf31169d9.jpg
http://www.freeimagehosting.net/imag...6e1446313a.jpg
http://www.freeimagehosting.net/imag...a4e1257410.jpg
http://www.freeimagehosting.net/imag...b79b46edf1.jpg
http://www.freeimagehosting.net/imag...3bb1b138aa.jpg

Source seems to be expreview, but not longer available

well if they were true.. AMD had them pulled...but they look spot on.. if all here say is true.. I am happy with those results.. at least they are more real world gaming and not scaled down 800x600 scores..

Has anyone done any benchmarks of the 9850 9950 with SB750 and higher then 30ghz oc? If I had the software i would run them..but I only have a AMD setup..

[Bgriffs]

If those are in fact true, and AMD gets their pricing right they have have a great shot at pulling of a coup just like the HD4000 series GPU's that came out this summer. I just want to see something that is concrete before I invest in that bad boy though.

[informal]

Anybody saw this ?

http://forums.vr-zone.com/showthread.php?t=359342

http://www.freeimagehosting.net/imag...ea01696aad.jpg
http://www.freeimagehosting.net/imag...fcf31169d9.jpg
http://www.freeimagehosting.net/imag...6e1446313a.jpg
http://www.freeimagehosting.net/imag...a4e1257410.jpg
http://www.freeimagehosting.net/imag...b79b46edf1.jpg
http://www.freeimagehosting.net/imag...3bb1b138aa.jpg

Source seems to be expreview, but not longer available

Seems very plausible so they pulled it out.Thanks for posting these .

Not bad at all, but it doesn't help that the PII is the highest clocked chip and won 1 test. I'd like to see PII vs. Q9650 for clock to clock comparison test. I doubt AMD will be selling the PII x4 940 for lower than $250. That Q9550 will be a real spoiler for AMD, because it's a drop-in chip for s775.

Actually it won FC2 and Crysis warhead ,not 1.It came close second in Dead Space-6% behind and Clear Sky-1fps behind Turbo clocked i940(so that means per clock it was probably faster than i940).World at war and World in Conflict never favored Phenom(I) from the start(compiler?),but Phenom II makes up tremendous ground in these two games.
All in all,if true,in games at least ,Phenom II @ 3GHz is very competitive with both Core2Q and Core i7.We need some OCed benchmarks at say ~4Ghz for all 3 CPUs to see how they scale(in games GPU rules though).

[Zucker2k]

Seems very plausible so they pulled it out.Thanks for posting these .


Actually it won FC2 and Crysis warhead ,not 1.It came close second in Dead Space-6% behind and Clear Sky-1fps behind Turbo clocked i940(so that means per clock it was probably faster than i940).World at war and World in Conflict never favored Phenom(I) from the start(compiler?),but Phenom II makes up tremendous ground in these two games.
All in all,if true,in games at least ,Phenom II @ 3GHz is very competitive with both Core2Q and Core i7.We need some OCed benchmarks at say ~4Ghz for all 3 CPUs to see how they scale(in games GPU rules though).

My bad.

Maybe we're not looking at the same graph so here is what I'm looking at:

STALKER CLEAR SKY, 1680x1050

Phenom II x4 940 3Ghz = 41.7
C2Q Q9550 2.83Ghz = 41.2


CRYSIS WARHEAD, 1680x1050

Phenom II x4 940 3Ghz = 35.7
C2Q Q9550 2.83Ghz = 35.0

DEAD SPACE, 1680x1050

Phenom II x4 940 3Ghz = 219.0
C2Q Q9550 2.83Ghz = 214.4

FAR CRY 2, 1680x1050

Phenom II x4 940 3Ghz = 54.4
C2Q Q9550 2.83Ghz = 46.2

WORLD AT WAR, 1680x1050

Phenom II x4 940 3Ghz = 48.7
C2Q Q9550 2.83Ghz = 57.3

WORLD IN CONFLICT, 1680x1050

Phenom II x4 940 3Ghz = 92.7
C2Q Q9550 2.83Ghz = 95.0

The lower clocked C2Q Q9550 is actually head to head with PII 940 and beats it by a wider margin, Imagine would Q9650 would do at 3Ghz, just like PII 940. Now I'm not a mathematician so you do the math.

Now expect the C2Q line to drop in price around January; I expect the Q9550, the PII 940's main competitor (imho) to drop to around $250, I expect Intel to do nasty things, price-wise with this particular chip against the PII 940. Expect Ci7 920 to be hover just below the Q9550 in price. With the above performance from the Q9550, AMD will always be forced to price the PII 940closely to the Q9550 since the PII 940 offers no performance advantage over this chip; and it is z775 compatible.

I think talks of ATi 4000 series and Nvidia is a little bit premature here; Intel has too many good weapons to fight this battle.

[Caveman787]

weird last I saw at those levels the i7 shouldn't be any bit faster then a yorkfield.

But fairly nice results for the P2 so far...I hope it can match both yorkfield and nehalem clock for clock in games.

[Bgriffs]

My point was that AMD has finally gotten themselves a competetive product. It will also go along way to regaining some cred and building back the market share. It may not be as much of a coup as the ATI/NV battle that happened but if they can launch this right they have a good chance to stay alive and start to come back to the black side of business.

[Zucker2k]

My point was that AMD has finally gotten themselves a competetive product. It will also go along way to regaining some cred and building back the market share. It may not be as much of a coup as the ATI/NV battle that happened but if they can launch this right they have a good chance to stay alive and start to come back to the black side of business.

I'm very impressed with PII in that shootout, the problem though is as you point out, launching it right. For a company that needs to make some decent profit, its going to be a bit tough, out-pricing a competitor's chip that is already below $300 and has no where to go but down.

[DoubleZero]

The lower clocked C2Q Q9550 is actually head to head with PII 940 and beats it by a wider margin, Imagine would Q9650 would do at 3Ghz, just like PII 940. Now I'm not a mathematician so you do the math.

Wasn't PII only suposed to match c2d 65nm?

I think talks of ATi 4000 series and Nvidia is a little bit premature here; Intel has too many good weapons to fight this battle.

You right, Intel ain't Nvidia, but K10 65nm to K10 45nm, feels a bit like R600 to R700, better then expected.

Overall things are going like expected, AMD is closing the gap in the desktop, Intel will be huge in the server arena.

[informal]

I was looking at both Core i7 and C2Q numbers.With Turbo Ci94 runs at >3Ghz,while C2Q has 6% lower clock than PhII(note that in games,especially in the tested high resolutions,fps doesn't go up linearly with CPU clock ,but it depends more on the uarchitecture itself) :

Clear Sky: PhII 41.7fps, Ci940 42.5fps ,C2Q(45nm) 41.2fps => Phenom II sits between the two chips which is in lign with pure clockspeed since Corei7 is above 3Ghz while C2Q is below(2.83Ghz).Test seems mostly GPU bound at that level with cosmetic difference between the 3 CPUs. Core i7 is just 1.9% faster than PhII while PhII is just 1% faster than C2Q.

Dead Space: PhII 219fps, Ci940 233fps ,C2Q(45nm) 214fps => Similar to clear sky.Ci7 is 6% faster than PhII(while having a slight clock advantage over PhII thanks to Turbo) ,while PhII is 2.3% faster than C2Q @ 2.83Ghz.As you can see,since the res. is 16x10, the fps doesn't rise linearly with CPU clock speed but is more dependent on GPU.The lower res. could have told us another story.

Crysis Warhead is clearly GPU limited at 16x10 ,but still we have some differentiation : PhII on top with 35.7 fps, Ci940 at bottom with 33.9fps ,C2Q(45nm) 2nd with 35fps.This tells us that in that game and game resolution,PhII vs C2Q will be virtually a tie per clock(again i emphasize the results don't scale linearly as you increase CPU clock due to resolution used where GPU is the key factor-not that CPU doesn't play an important role though).

FarCry 2: PhII 54.4fps, Ci940 48.4fps ,C2Q(45nm) 46.2fps .Clear advantage for PhII in this test(no even if we account for clock difference it's quite a lead-11% in front of C2Q per clock).The difference is more pronounced than in any other title tested there(maybe because of the dedicated CPU test just as in Warhead case?) .

World at war and World in Conflict: these two were always "dogs" on previous phenom so i guess there is some special optimization going on there(phenom was running the game just fine,but was lagging in fps when compared to c2q,difference being as high as 50% which is hardly a case in any other game apart from these 2).PhII makes up tremendous ground and ties c2q in WiC(lags 6% in clock for clock terms) and lags 17% clock for clock in World at War. WiC could be cache limited as we see 25% jump(per clock) with PhII,while Ci7 and C2Q still have overall cache advantage over PhII and manage a slight lead.

[qurious63ss]

Wasn't PII only suposed to match c2d 65nm?

No, I think agena was to match 65nm. Deneb is supposed to match c2d 45nm and k10.5 supposed to match nehalem?

[Zucker2k]

Where does it indicate TURBO mode was on for Ci7 940? Are you assuming so? And why is it when PII doesn't fare well, you try to downplay it? I thought there were many core optimizations done over PI?

Despite your effort to explain this away, it is quite clear to any discerning person that a C2Q @ 3Ghz, would win 5 out of those 6 tests, the 6th test could possibly swing either way. THIS IS VERY GOOD FOR PII, NONETHELESS, I don't know why you want to look past the data though.

[informal]

No, I think agena was to match 65nm. Deneb is supposed to match c2d 45nm and k10.5 supposed to match nehalem?

"k10.5" is Deneb.. BTW ,K10.5 is more of a nickname for Deneb/Shanghai.

[qurious63ss]

Based on these benches, Intel can use the Q9550 to compete with PII. The Ci7 looks to be a great workstation part where better multithreaded software is available.

[qurious63ss]

"k10.5" is Deneb.. BTW ,K10.5 is more of a nickname for Deneb/Shanghai.

Hmm.. I thought k10.5 would be the hK metal gates part.

[informal]

Based on these benches, Intel can use the Q9550 to compete with PII. The Ci7 looks to be a great workstation part where better multithreaded software is available.

Games don't scale with CPU clock that well.So i can use that logic and say AMD can use 920 model(2.8Ghz) and match intel's Q9550(2.83Ghz) part.

And no,"K10.5" was a name given by Fudzilla. There is no publicly known "name" for high k 45nm K10 part.

[Bgriffs]

Nope, AMD has the option of adding the HK metal gates for later in the 45nm process but its looking like its going to be 32nm parts that make that switch.

[qurious63ss]

Games don't scale with CPU clock that well.SO i can use that logic and say AMD can use 920 model(2.8Ghz) and match intel's Q9550(2.83Ghz) part.

I agree and that's why I said based on these benches. By match I mean price wise.