Wolfdale benchmarked

[Tempered81]

e8400's hit 5.1ghz

http://www.google.com/translat...=ru%7Cen&hl=en&ie=UTF8

 

[Borealis7]

i have an E6550, i expect it to hold for a couple of years (im not that hardcore of a gamer) and i havent O/Ced yet. so these processors, while interesting to read about, do not appeal to me.
my next CPU will be a quadcore, but i dont think my IP-35-E will support Nehalem, so i'll have to settle for a quad Penryn in a year or two.

 

[darkxknight]

sounds good. feel bad for amd though, they're really gonna need some sort of miracle to make it through this year.

 

[21stHermit]

Originally posted by: harpoon84
Indeed it looks like an awesome overclocker although the results are a little random at this stage, ie. the E8200 'only' reached 3.88GHz... perhaps there is some speed binning going on?

To my mind a given. I would expect that all the same size cache E8xxx's are built on a single wafer. Then before the dies are cut from a wafer, each die is probed, tested, laser castrated to its multipler and identified. Doubt it takes more than 10-min a wafer.

 

[lopri]

E8400 is an incredible bargain, IMO. It 'obsoletes' all of the past Core 2 Duo chips, IMO, other than lower end Pentium E series.

 

[Tempered81]

e8400 on air at 4.6ghz 1.6 volts... the 5.1ghz from the post earlier was at 1.75 volts on d-ice, both on foxcon mars motherboards

http://valid.x86-secret.com/show_oc.php?id=278073

 

[21stHermit]

Originally posted by: jaredpace
e8400 on air at 4.6ghz 1.6 volts... the 5.1ghz from the post earlier was at 1.75 volts on d-ice, both on foxcon mars motherboards

1.6, 1.75V NO WAY!!! :Q

Tops for me would be 1.4-1.5V. Actually, I'd like to play with under volting.

 

[aussiestilgar]

Will there be any OEM (no heatsink/cooler) E8400's available at launch? If so, where?

 

[heyheybooboo]

I hate to rain on this parade ...

This time, Core 2 Duo E8500 came to replace Core 2 Duo E6850, Core 2 Duo E8400 steps in for the Core 2 Duo E6750, and Core 2 Duo E6550 gives way to Core 2 Duo E8200.

45nm, 50% more cache, 40% more transistors and when you compare performance clock-to-clocks ...

E6550 (2.33GHz)

E6750 (2.66GHz)
E8200 (2.66GHz)

E6850 (3GHz)
E8400 (3GHz)

E8500 (3.16GHz)

I'm not seeing a real great reason to dump a 65nm CD2.

It's almost like a 'bait & switch' when xbit says, ""Core 2 Duo E8400 steps in for the Core 2 Duo E6750, and Core 2 Duo E6550 gives way to Core 2 Duo E8200."" This is like over-marketing 101.

Clock-for-clock: E6850=E8400. E6750=E8200.

It looks to me like a 10% speed bump (on a sliding scale) where the low-end (E6550) is dropped and a new upper end product (E8500) is added --- And when you compare the *charts and graphs* between similar CPUs I'm not sure I'm seeing a 10% increase in performance.

I'm not saying it's wrong. I'm not dissin' Intel. I'm just saying there are different ways to look at the xbit article from the "marketing" standpoint.

I gots no proiblem increasing performance 10% and cutting price 10% if that's where all this ends up ...

 

[Tempered81]

yeah but 65nm conroes will oc to 3.4ghz while 45nm wolfdales will oc to 4.6ghz

wolfdales = more cache, less heat, higher temp threshold, less power consumed.

now if they fall into the same price brackets as those older chips that are conroe-based, why not get 1, especially if you are in the market to purchase a new cpu

 

[heyheybooboo]

Originally posted by: jaredpace
yeah but 65nm conroes will oc to 3.4ghz while 45nm wolfdales will oc to 4.6ghz

wolfdales = more cache, less heat, higher temp threshold, less power consumed.

now if they fall into the same price brackets as those older chips that are conroe-based, why not get 1, especially if you are in the market to purchase a new cpu

I'm just an old cynical curmudgeon - - pay no attention

There will always be new, faster, cheaper, more efficient, etc. I just pointed out a little ""Marketing 101"" and how it may influence purchasing decisions.

And as noted - if you are looking to spend $180-$190 on a cpu the E8400 is lookin' good

 

[Tempered81]

i would be the younger newbish guy then. im into new & faster, but need the old stability & good job to pay for the upgrade, DAMN!

 

[Toadster]

Originally posted by: jaredpace
so can you buy a 45nm quad today or not?

of course!

http://forums.slickdeals.net/s...hp?sduid=6222&t=553826

 

[v8envy]

Agreed. Everyone in a holding pattern waiting for the new CPUs is going to want one (myself included). The question is not whether to get an E8400 vs. 6750 (vs. anything AMD makes), the question is when will the anticipated price gouging settle down. Intel hasn't had capacity problems in the past, so let's hope they wish to sell as many CPUs as people are willing to buy rather than hand etailers a fat bonus.

But if not, I've resigned myself to waiting until March. My crusty old socket 939 2.4 ghz Venice is still operational, and there are no new games that demand I upgrade right this minute. I may even make it a package deal with the upcoming GPUs. It's just an itch that I'm not going to scratch just yet.

 

[harpoon84]

Originally posted by: heyheybooboo
I hate to rain on this parade ...

....

I gots no proiblem increasing performance 10% and cutting price 10% if that's where all this ends up ...

(Sorry I shortened your quote to save space)

The way I see is is that Intel is effectively giving you about 10 - 15% more performance for the same price, or the same performance at a lower price.

The E8200 effectively replaces the E6550 ($163), yet outperforms the E6750

The E8400 effectively replaces the E6750 ($183), yet outperforms the E6850

The E8500 effectively replaces the E6850 ($266), outperforms the E6850 even more

Each Wolfdale C2D offers higher stock speed and IPC than the Conroe C2D it replaces. I don't see what is not to like about this?

 

[Tempered81]

does anyone know what the fan will be like? Will it be the aluminum stock intel cooler they packaged with the conroes - http://www.hardwarelogic.com/a...s/INTEL_E6300/HSF2.jpg ?

i hope its something new, with a little copper, and better heat dissipation

 

[Nemesis 1]

So how do you guys think a 2core wolf will stack up against K10 3 cores cpu? Looks to me like Wolf has the edge.

 

[BlueAcolyte]

wolfdale will probably throttle K10 by the throat. (I mean throttle like strangle, not block speed, although that is a possibility too)

 

[harpoon84]

Originally posted by: Nemesis 1
So how do you guys think a 2core wolf will stack up against K10 3 cores cpu? Looks to me like Wolf has the edge.

Definitely. In single threaded performance Wolfdale would simply destroy a tri-core Phenom. Even in multithreaded performance, tests have shown a 'simulated' Phenom X3 @ 2.3GHz is only comparable to an E6750 or E6850 (or effectively E8200) in multithreaded performance.

http://www.erenumerique.fr/tes...l_ve_-art-1804-11.html

Wolfdale may lack a physical core compared to Phenom Tri-Core, but it'll more than make up for it in brute clockspeed scaling and IPC. Kinda like 2 big guys fighting 3 little guys.

 

[Nemesis 1]

Ya thats how I seen it. I don't really want AMD to go under the competition is a good thing. But I see AMD is in big trouble here on pricing. Dell and the rest of Pc makers are going to Go with Price /performance in selecting Cpu's . Dell won't pay AMD a price that is higher than a 2 core processor if performance is = . So I really see K10 3 core cpus not selling very well unless they are under $150.

 

[harpoon84]

Originally posted by: Nemesis 1
Ya thats how I seen it. I don't really want AMD to go under the competition is a good thing. But I see AMD is in big trouble here on pricing. Dell and the rest of Pc makers are going to Go with Price /performance in selecting Cpu's . Dell won't pay AMD a price that is higher than a 2 core processor if performance is = . So I really see K10 3 core cpus not selling very well unless they are under $150.

Yeah, I would agree.

AMD realistically can't afford to charge over $150 for a tricore that barely matches a $163 Wolfdale duallie in multithreaded performance, whilst getting owned in single threaded performance.

This is why I find it funny that certain tech websites are trying to justify the existence / advantages of tri core for AMD when all those so called 'advantages' are rendered obsolete by simple clockspeed and IPC advantages from Intel dual cores. You see a lot of talk about tri-core being an untapped market. I call BS on that one. At the end of the day, it's performance, not core count, that ultimately matters, and Phenom tri-core doesn't offer any performance advantages over Intel dual cores. At least not in the sub 2.5GHz range, which is what the AMD roadmaps suggest.

 

[golem]

Never underestimate the power of marketing. People might just buy the tri core over the dual core because 3 is more than 2, even if the dual core is faster.

 

[Nemesis 1]

Your right a few could be fooled doing self builds . But the PC vendors aren't stupid and AMD will have to price according to performance and effciency. Or the Venders won't use the cpu's its that simple

 

[Rhonda the Sly]

I don't think the Toliman cores will go over $160 at launch, they're the same chips as the x4s but disfunctional. AMD knows this and knows we know this, they'll price them accordingly. What I'm afraid of though is what this pricing means for their X2 cores. Assuming the Phenom X3 8600 and 8400 go for $160 and $130 AMD will have to shove their brand new X2 6050 and 6250 chips to $110, maybe even under $100, for people to buy them. It looks to me like Phenom X4's aggressive pricing ($190 for a 9500) is strangling the rest of the pack. Trying to sell an entire family of processors (two X2s, two X3s, two X4s) under $240 sounds absolutely madness. They have about $80 to play around with pricing of X2 and X3 unless they plan for X2 to launch at Pentium Dual-Core prices, which I never really considered until I wrote this line but a new AMD low-end sounds interesting. Providing they can pump up the cores on X3 to compete against Wolfsdale, this could work, but I'm being optimistic.

If AMD responds to Wolfsdale/Yorkfield with another price drop this could only further the problem, it doesn't look like they have room for that. If we see Phenom dual-cores running with better speed than the quad-cores per core, they might be able to increase the price reasonably. They'll run into the same territory as X3 but price/performance might still leave them solid. How PC vendors see them is entirely dependent on what AMD can do to them between now and march but I'm not sure their being worked on. More attention is probably being put towards the native triple-core.

And just by the way, as for selling people things they don't know about, selling a triple-core over a Intel Dual-Core might be easy. "Hey, both chips are 8400s, the AMD just has more cores! You'll be surfing the web and writing word documents so fast you'll get dizzy!"

This is why I find it funny that certain tech websites are trying to justify the existence / advantages of tri core for AMD

I agree with you but AMD really needs to start making money and these sound like a decent idea. A decent idea I'd never buy into but a decent idea nonetheless.

 

[Nemesis 1]

IF amd tricore can't compete and beat wolf by a pretty good margin in multi threading . AMD shouldn't sell the . It would be smarter from the to dump them .


ITS a marketing nightmare. Intel Dual cores multi task faster than tricores at 1/3 the cost. ( Efficiency)

AMD tricore is an Intel Marketing Dream come true. If AMD doesn't up the freq. on quads . Intel put one out that cleans their close also . Thats why thr duals are coming out first . SO the reviews sites can compare Duals to PHENOMS Its looking worse than bad.