Nehalem-EP......BLOOMFIELD

Quote:

---

Originally Posted by savantu

Isn't it a bit early ?

---

Umm way too early. I remember people going crazy over Penryn quads expecting crazy overclocks but it didn't happen. Speculation is just that.

Quote:

---

Originally Posted by MomijiTMO

Umm way too early. I remember people going crazy over Penryn quads expecting crazy overclocks but it didn't happen. Speculation is just that.

---

I don't know about that, I don't remember a lot of folks saying something like that? This forum isn't crazy like a few others. Most estimated it would be easier to get Dual Core to 4GHz and Quad to 3.6GHz. Even when others talked about 4GHz for Quad, most folks agreed maybe with water and extra voltage.

Quote:

---

Originally Posted by Pinacolada

I'm passing Nehalem and I see no point in upgrade.

10% on single core and 20% on multicore is not enough to consider me buying new : CPU,RAM,mobo and cooling.
Penryn is good enough until cheap octalcores appear which will offer more than 100% multithreaded improvements than quadcore Penryn.

---

Expect the performance improvements to increase significantly once motherboards have the kinks worked out. Any application that is constrained by memory performance will benefit much more with Nehalem.

Check out how your current memory is being bottle-necked by the FSB and how Nehalem's triple-channel DDR3 IMC really opens the spill-way on RAM bandwidth...

Note that Nehalem's tri-channel DDR3-1333 peak bandwidth is on par with Penryn L2 cache! (the likes of which would never be seen on Penryn with it's highly bottlenecked FSB).

http://members.shaw.ca/virtualrain/n...-fsb-chart.png

Source

DDR3 1333 is pretty cheap. It's the 1800-2000 stuff that gets pricey. Even if the advanced Nehalem offers 1600, that won't be too bad for 6gb triple sticks come Q4.

2.93GHz Rev.B0 :clap:
http://i245.photobucket.com/albums/g...oomfield_1.jpg

...

:eek2:

Around 10% faster than Penryn @the same clock,not bad but not earth shattering.We need something else apart from useless Spi1m(wPrime ie. where multicore shines)

Quote:

---

Originally Posted by informal

Around 10% faster than Penryn @the same clock,not bad but not earth shattering.We need something else apart from useless Spi1m(wPrime ie. where multicore shines)

---

http://www.anandtech.com/cpuchipsets...oc.aspx?i=3326

And thats still singlechannel :rofl:

anyone can upload the pic on another image hoster, photobucket is down for me since yeaterday... :(

Quote:

---

Originally Posted by Hornet331

anyone can upload the pic on another image hoster, photobucket is down for me since yeaterday... :(

---

http://www.abload.de/img/bloomfield_1nel.jpg

Quote:

---

Originally Posted by virtualrain

Check out how your current memory is being bottle-necked by the FSB and how Nehalem's triple-channel DDR3 IMC really opens the spill-way on RAM bandwidth...

Note that Nehalem's tri-channel DDR3-1333 peak bandwidth is on par with Penryn L2 cache! (the likes of which would never be seen on Penryn with it's highly bottlenecked FSB).

---

http://www.anandtech.com/cpuchipsets...oc.aspx?i=2991
"In our recent DDR3 vs. DDR2 review we discovered a 16% to 18% improvement in memory bandwidth with the P35 chipset. This translated into a 2% to 5% increase in real world performance in some computer applications.

This bandwith will be only for overpriced Bloomfield Nehalem,Nehalems in lower price will come in 2H 2009 and will have only dual channel.I expect Bloomfield setup to be 100-150% more expensive than Penryn setup (Intel will position Bloomfield as high-end system and it will cost) with performance around 10-20% single thread ,20-40% multithread.New drivers and better mobos won't help more than 1-3%.

So until you really need new computer or you are into multithreaded apps upgrade to quadcore Nehalem does make sense.Otherwise you will pay 100% more for about 10-20% improvement - not worth IMO for typical user it's much wiser to upgrade GPU or wait for cheap SSDs.
Quadcore penryn is good enough to wait till 2H 2009 for 32nm shrink where price/performance ration against Penryn will be much better.Most probably octalcores will be available then.

People are expecting marvels from Nehalem but it will be more evolution than revolution.

Quote:

---

Originally Posted by Chri$ch

http://www.abload.de/img/bloomfield_1nel.jpg

---

thx mate. :up:

Quote:

---

Originally Posted by Pinacolada

http://www.anandtech.com/cpuchipsets...oc.aspx?i=2991
"In our recent DDR3 vs. DDR2 review we discovered a 16% to 18% improvement in memory bandwidth with the P35 chipset. This translated into a 2% to 5% increase in real world performance in some computer applications.

This bandwith will be only for overpriced Bloomfield Nehalem,Nehalems in lower price will come in 2H 2009 and will have only dual channel.I expect Bloomfield setup to be 100-150% more expensive than Penryn setup (Intel will position Bloomfield as high-end system and it will cost) with performance around 10-20% single thread ,20-40% multithread.New drivers and better mobos won't help more than 1-3%.

So until you really need new computer or you are into multithreaded apps upgrade to quadcore Nehalem does make sense.Otherwise you will pay 100% more for about 10-20% improvement - not worth IMO for typical user it's much wiser to upgrade GPU or wait for cheap SSDs.
Quadcore penryn is good enough to wait till 2H 2009 for 32nm shrink where price/performance ration against Penryn will be much better.Most probably octalcores will be available then.

People are expecting marvels from Nehalem but it will be more evolution than revolution.

---

how can you compare fsb to QPI/IMC ?

Shintai,Pinacolada summed it up well here:

Quote:

---

So until you really need new computer or you are into multithreaded apps upgrade to quadcore Nehalem does make sense.Otherwise you will pay 100% more for about 10-20% improvement - not worth IMO for typical user it's much wiser to upgrade GPU or wait for cheap SSDs.
Quadcore penryn is good enough to wait till 2H 2009 for 32nm shrink where price/performance ration against Penryn will be much better.Most probably octalcores will be available then.

People are expecting marvels from Nehalem but it will be more evolution than revolution.

---

Quote:

---

Originally Posted by informal

Shintai,Pinacolada summed it up well here:

---

So this isn't thread crapping,:rofl:

Quote:

---

Originally Posted by Pinacolada

http://www.anandtech.com/cpuchipsets...oc.aspx?i=2991
"In our recent DDR3 vs. DDR2 review we discovered a 16% to 18% improvement in memory bandwidth with the P35 chipset. This translated into a 2% to 5% increase in real world performance in some computer applications.

This bandwith will be only for overpriced Bloomfield Nehalem,Nehalems in lower price will come in 2H 2009 and will have only dual channel.I expect Bloomfield setup to be 100-150% more expensive than Penryn setup (Intel will position Bloomfield as high-end system and it will cost) with performance around 10-20% single thread ,20-40% multithread.New drivers and better mobos won't help more than 1-3%.

So until you really need new computer or you are into multithreaded apps upgrade to quadcore Nehalem does make sense.Otherwise you will pay 100% more for about 10-20% improvement - not worth IMO for typical user it's much wiser to upgrade GPU or wait for cheap SSDs.
Quadcore penryn is good enough to wait till 2H 2009 for 32nm shrink where price/performance ration against Penryn will be much better.Most probably octalcores will be available then.

People are expecting marvels from Nehalem but it will be more evolution than revolution.

---

Firstly, Bloomfield will launch in the Q9450, Q9550 and QX brackets - if you think a mainstream quad core Penryn is 'overpriced' to begin with then I guess the 2009 Lynnfield one will be too.

For me upgrading the GPU is of no use because it's an 8800GTX, it plays every single game just fine and even folds quite quickly. The GT200 is an overpriced incremental upgrade that brings nothing new to the table, just more of the same.

My CPU, on the other hand, is a 90nm AMD dual-core. Maybe it's not worth it for all your Penryn owners to upgrade, but everyone doesn't have Penryn.

Quote:

---

Originally Posted by Donnie27

So this isn't thread crapping,:rofl:

---

Hmm i quote another user's opinion(which btw is realistic),i don't call out any1 and you call that thread crapping?Get a grip man.I thought the Fugger's post was enough,seems not :confused:

Quote:

---

Originally Posted by bowman

My CPU, on the other hand, is a 90nm AMD dual-core. Maybe it's not worth it for all your Penryn owners to upgrade, but everyone doesn't have Penryn.

---

I agree,in that case upgrade is well worth it.

Quote:

---

Originally Posted by informal

Hmm i quote another user's opinion(which btw is realistic),i don't call out any1 and you call that thread crapping?Get a grip man.I thought the Fugger's post was enough,seems not :confused:

---

Sorry, talking about overpriced without a price list, trying to downplay performance with only estimations, ummm, sounds like crapping to me!

It was a quote from another user.Get it?!This is so amusing.

Quote:

---

Originally Posted by JCornell

2.93GHz Rev.B0 :clap:
...
...

---

Please post some memory scores. :)

Quote:

---

Originally Posted by savantu

Please post some memory scores. :)

---

i second that, maybe some everest scores? :)

Quote:

---

Originally Posted by informal

It was a quote from another user.Get it?!This is so amusing.

---

Yes you are amusing LOL!

Quote:

---

Originally Posted by Hornet331

i second that, maybe some everest scores? :)

---

I'd like to see them as well.

But we'll still run into a problem. Unless the user is stressing the system, most of the time Bandwidth isn't a bottleneck now.

Quote:

---

Originally Posted by Hornet331

i second that, maybe some everest scores? :)

---

Can't run everest, it crash while click the everest.exe :rofl:

I did run CINEBENCH R10 : Rendering (xCPU)

damn would have been to good. :hitself:

same with sisoft sandra?

Quote:

---

Originally Posted by Hornet331

damn would have been to good. :hitself:

same with sisoft sandra?

---

Been a while not using sisoft, Will try it.

Quote:

---

Originally Posted by JCornell

Been a while not using sisoft, Will try it.

---

Please do if you're not bound by any obligations.

CPU tests ( arithmetic , multimedia, multicore efficiency ) and memory BW+latency would be really nice.

Quote:

---

Originally Posted by JCornell

Can't run everest, it crash while click the everest.exe :rofl:

I did run CINEBENCH R10 : Rendering (xCPU)

---

Watching and waiting:shocked:

JC can you tell us if this is still Single Channel memory? what is the issue with triple channel?

Thanks for posting SS of Nehalem, more benchs would be awsome.

From a Yorktown this probably isn't much of an improvement, but from the older first gen Conroe's 2 generations of improvements leading to an on die mem controller is pretty nice.

On the memory front, it's reasonable to expect DDR3 to fall a bit more by the time mainstream Nehalem's are available (early '09). And the prices slightly below now will probably satisfy the first adopters who are going to be buying the premium priced cpu's anyway.

Quote:

---

Originally Posted by Pinacolada

Quadcore penryn is good enough to wait till 2H 2009 for 32nm shrink where price/performance ration against Penryn will be much better.Most probably octalcores will be available then.

People are expecting marvels from Nehalem but it will be more evolution than revolution.

---

I agree with what you said for the most part, but a 20-30% improvement is a very good improvement and I want one.

Now, you said that Nehalem is evolution vs. revolution.

What then would be a revolution?

I see it as both ways. it's a revolution in the fact that it completely revamps Intel memory subsystem. XeonMP? Yeah, not much has changed when it comes to bandwith per core and anandtech said that each core would get maybe 600 MB/sec right now.

You move to a 4 socket Nehalem and each cpu has it's own memory controller and is connected to each other by more than a northbridge. I see that as a combination of evolution and revolution.

it's a revolution in that it's the stepping stone for what's to come.

With the rise of the GPU, I feel as if the way to go is not only adding more cores, but adding dumb cores that crank out FP. Intel said that Nehalem was a modular design, and unused QPI paths could be used for other applications such as FP boosts.

I think was Intel moves toward graphics that we might see some of those cores in CPUs. The role of the GPU is changing in the world. Back in the day, you gamed on your GPU and that's about it. Now that there's actual scientific research to the GPU, I wouldn't be surprised if we say future cpus with a Larrabee partially integrated.

So i feel l like it's both.

However, one thing that i'm wondering. The die size of Nehalem is relatively large for an Intel portfolio. Now they plan on increasing the cache size by 50% and adding two cores for Westmere.

How much cache is too much? Could we not just switch to a faster memory subsystem and rely less on cache. I think that's an important thing to consider. So yeah that's my ramble.

Quote:

---

Originally Posted by Pinacolada

http://www.anandtech.com/cpuchipsets...oc.aspx?i=2991
"In our recent DDR3 vs. DDR2 review we discovered a 16% to 18% improvement in memory bandwidth with the P35 chipset. This translated into a 2% to 5% increase in real world performance in some computer applications.

This bandwith will be only for overpriced Bloomfield Nehalem,Nehalems in lower price will come in 2H 2009 and will have only dual channel.I expect Bloomfield setup to be 100-150% more expensive than Penryn setup (Intel will position Bloomfield as high-end system and it will cost) with performance around 10-20% single thread ,20-40% multithread.New drivers and better mobos won't help more than 1-3%.

So until you really need new computer or you are into multithreaded apps upgrade to quadcore Nehalem does make sense.Otherwise you will pay 100% more for about 10-20% improvement - not worth IMO for typical user it's much wiser to upgrade GPU or wait for cheap SSDs.
Quadcore penryn is good enough to wait till 2H 2009 for 32nm shrink where price/performance ration against Penryn will be much better.Most probably octalcores will be available then.

People are expecting marvels from Nehalem but it will be more evolution than revolution.

---

This extraction is not quite fair, nor correct. The entire memory sub-system for Nehalem has been reworked, you cannot look at the off-die DDR2-DDR3 comparision and make the assumption you are making.

In terms of marvels, who knows ... we have very little info to go by ... however, the hype machine of Geslinger compared the leap of Nehalem over Conroe to the magnitude of Conroe over Netburst, this indeed would be marvelous. My take on it is that it will be less overwhelming in single thread, but multithreaded situations will be impressive on this magnitude... Anand's preview data seems to indicate that this is indeed the potential... 30-50% gains depending on the app (multithreaded of course).

This will be even more impressive in server, where the mem BW issue is much much more pronounced... Kanter (RWT) seems to think Nehalem will be nothing short of a miracle. Time will tell.... however, initial indications are very positive that your downplay of Nehalem is probably not warranted.

Jack

Quote:

---

Originally Posted by JumpingJack

This extraction is not quite fair, nor correct. The entire memory sub-system for Nehalem has been reworked, you cannot look at the off-die DDR2-DDR3 comparision and make the assumption you are making.

In terms of marvels, who knows ... we have very little info to go by ... however, the hype machine of Geslinger compared the leap of Nehalem over Conroe to the magnitude of Conroe over Netburst, this indeed would be marvelous. My take on it is that it will be less overwhelming in single thread, but multithreaded situations will be impressive on this magnitude... Anand's preview data seems to indicate that this is indeed the potential... 30-50% gains depending on the app (multithreaded of course).

This will be even more impressive in server, where the mem BW issue is much much more pronounced... Kanter (RWT) seems to think Nehalem will be nothing short of a miracle. Time will tell.... however, initial indications are very positive that your downplay of Nehalem is probably not warranted.

Jack

---

Very well said!:up:

Quote:

---

Originally Posted by Pinacolada

http://www.anandtech.com/cpuchipsets...oc.aspx?i=2991
"In our recent DDR3 vs. DDR2 review we discovered a 16% to 18% improvement in memory bandwidth with the P35 chipset. This translated into a 2% to 5% increase in real world performance in some computer applications.

This bandwith will be only for overpriced Bloomfield Nehalem,Nehalems in lower price will come in 2H 2009 and will have only dual channel.I expect Bloomfield setup to be 100-150% more expensive than Penryn setup (Intel will position Bloomfield as high-end system and it will cost) with performance around 10-20% single thread ,20-40% multithread.New drivers and better mobos won't help more than 1-3%.

So until you really need new computer or you are into multithreaded apps upgrade to quadcore Nehalem does make sense.Otherwise you will pay 100% more for about 10-20% improvement - not worth IMO for typical user it's much wiser to upgrade GPU or wait for cheap SSDs.
Quadcore penryn is good enough to wait till 2H 2009 for 32nm shrink where price/performance ration against Penryn will be much better.Most probably octalcores will be available then.

People are expecting marvels from Nehalem but it will be more evolution than revolution.

---

To expect a Core 2 Duo style beatdown for every processor launch would be asking too much. that said, Nehalem looks like a monster on top of a beast (Conroe).

Quote:

---

Originally Posted by JumpingJack

This extraction is not quite fair, nor correct. The entire memory sub-system for Nehalem has been reworked, you cannot look at the off-die DDR2-DDR3 comparision and make the assumption you are making.

In terms of marvels, who knows ... we have very little info to go by ... however, the hype machine of Geslinger compared the leap of Nehalem over Conroe to the magnitude of Conroe over Netburst, this indeed would be marvelous. My take on it is that it will be less overwhelming in single thread, but multithreaded situations will be impressive on this magnitude... Anand's preview data seems to indicate that this is indeed the potential... 30-50% gains depending on the app (multithreaded of course).

This will be even more impressive in server, where the mem BW issue is much much more pronounced... Kanter (RWT) seems to think Nehalem will be nothing short of a miracle. Time will tell.... however, initial indications are very positive that your downplay of Nehalem is probably not warranted.

Jack

---

Absolutely agreed.

Core2 already provides a lot more computational power than the average user needs. Many games cannot fully utilize the power of a Penryn quad and most Microsoft/Adobe apps don't even come close. So unless you are working extensively with video and rendering or folding, Nehalem probably won't benefit you much.

It should turn out some pretty impressive memory benchmarks and I agree with Kanter that it will be a miracle on server loads, particularly multi-socket systems where the new QPI and IMC architecture can really shine.

Can't execute SisoftSandra too :ROTF:
http://i245.photobucket.com/albums/g...oomfield_3.jpg

...

PI32M :clap:
http://i245.photobucket.com/albums/g...oomfield_2.jpg

...

JC, do you have problem running other things. Or basicly just the HW detecting programs.

hmm, ok i know one last thing, can you try crystalmark?

Nehalem at 2.93GHz runs wPrime 32M with the speed of a QX9650 at ~ 4.4GHz.

That's 50% clock/clock advantage in this multithreaded test.

Quote:

---

Originally Posted by informal

Shintai,Pinacolada summed it up well here:

---

informal, Jack summed it up well here:

Quote:

---

This will be even more impressive in server, where the mem BW issue is much much more pronounced... Kanter (RWT) seems to think Nehalem will be nothing short of a miracle. Time will tell.... however, initial indications are very positive that your downplay of Nehalem is probably not warranted.

Jack

---

Quote:

---

Originally Posted by Donnie27

informal, Jack summed it up well here:

---

Good stuff.

I had wished Nehalem would clock a tad higher on air, but then again, if it clocked the same as the quad cores with 8 cores, AMD would be a chip recycling plant by now :p:

Perkam

Quote:

---

Originally Posted by perkam

Good stuff.

I had wished Nehalem would clock a tad higher on air, but then again, if it clocked the same as the quad cores with 8 cores, AMD would be a chip recycling plant by now :p:

Perkam

---

I don't think this Optimized Paralleled performance test will not make that much difference for most Desktop apps. This is what's expected with Hyperthreading 2. I'd much rather see what those poorly optimized and fewer threaded apps will do. I'm not worried about it clock speed or overclocking.

Those Nehalem are really fast:eek:

Quote:

---

Originally Posted by JCornell

I did run CINEBENCH R10 : Rendering (xCPU)

---

Post your CPU scores if you could please :yepp:

Quote:

---

Originally Posted by Shintai

JC, do you have problem running other things. Or basicly just the HW detecting programs.

---

No problem at all, That's just simply HW & SW crash(can't detect the HW info). :yepp:

CINEBENCH10 Rendering xCPU
http://i245.photobucket.com/albums/g...oomfield_4.jpg

...

Quote:

---

Originally Posted by JCornell

Can't execute SisoftSandra too :ROTF:
http://i245.photobucket.com/albums/g...oomfield_3.jpg

...

---

try pls Wprime 1024M, thx :)

Quote:

---

Originally Posted by FlanK3r

try pls Wprime 1024M, thx :)

---

Thought you never ask :ROTF:
http://i245.photobucket.com/albums/g...oomfield_5.jpg

...

Quote:

---

Originally Posted by nachthymnen

Those Nehalem are really fast:eek:

---

WPrime scores, I did compared with your QX9650 @5.6~5.7G :clap:
Bloomfield platform with 2.93GHz + 1*2G DDR3-1066 ;)

http://www.pctunerup.com/up/results/...14_233.749.jpg

http://www.pctunerup.com/up/results/...2225_7.218.jpg

...

lol, nehalem is a multithreading beast. :shock:

We desperately need memory benchmarks. :)

Run stream , it's a tiny program which should run no matter what.Used for testing the memory system of computers ranging from desktops to servers with thousands of CPUs.

http://www.cs.virginia.edu/stream/FT...32/wstream.exe

You'll need to open the command console ( remember where you saved it ) and type wstream <core frequency , 2933 this case > < iterations , try with 1 , 4 and 8 >.

Something like this c:\...desktop\wstream 2933 4

*I've noticed you're running single channel mode.

Quote:

---

Originally Posted by Hornet331

lol, nehalem is a multithreading beast. :shock:

---

Damn incredible I'd say.The performance is astounding.

Those Wprime benchs are really impresive, compare to a QX9650 below:


So your 2.93GHz (with dual channel?), beat my result at 4.2GHz with Dual Channel (by a fair margin)

http://img167.imageshack.us/img167/7...24m4200id4.jpg

this Nehalem is going to be a monster in multi-threaded apps even if it only does 3.6GHz on air.

Quote:

---

Originally Posted by JCornell

Thought you never ask :ROTF:
<snip screen shot>

...

---

Thanks for all the benchmarks JC. I was going to buil d a Q67000 or Q9450 system for video editing, but now I will wait for Nehalem. As others have noted, this looks like it will be a great multi-thread chip. Just what I am looking for to run Sony Movie Studio (which can generate upto 4 threads :up: )

Quote:

---

Originally Posted by sdsdv10

Thanks for all the benchmarks JC. I was going to buil d a Q67000 or Q9450 system for video editing, but now I will wait for Nehalem. As others have noted, this looks like it will be a great multi-thread chip. Just what I am looking for to run Sony Movie Studio (which can generate upto 4 threads :up: )

---

Can you wait that long :p. I'm toying with the idea of building a cheapie 24/7 render slave, looks like CPU's are getting cheaper :). Nehalem looks tempting, however I cant really believe we'd be able to find it available any time soon after release :(

Quote:

---

Originally Posted by metro.cl

Those Wprime benchs are really impresive, compare to a QX9650 below:


So your 2.93GHz (with dual channel?), beat my result at 4.2GHz with Dual Channel (by a fair margin)
*pic left out*[/URL]

this Nehalem is going to be a monster in multi-threaded apps even if it only does 3.6GHz on air.

---

I think that would be a dream, we should not have to much high hopes on their first true quad core :D

Quote:

---

Originally Posted by Horus´

I think that would be a dream, we should not have to much high hopes on their first true quad core :D

---

It will release at 3.2GHz so 3.6GHz shouldnt be that hard ;)

Quote:

---

Originally Posted by savantu

Nehalem at 2.93GHz runs wPrime 32M with the speed of a QX9650 at ~ 4.4GHz.

That's 50% clock/clock advantage in this multithreaded test.

---

Yup... just about 50%:

http://www.xcpus.com/gallery/d/6740-2/9650_2_93G.PNG

Quote:

---

Originally Posted by Horus´

I think that would be a dream, we should not have to much high hopes on their first true quad core :D

---

Dewd, true Quad is Quad=P Doesn't matter if it is Monolithic or MCM, it is still Quad. The whole "True Quad Core" was crap and nothing more than Marketing Hype or a Gimmick.

When I did my back of the envelop calculation, I used the 32M numbers... so here is what I get...

For Nehalem @ 2.93G
32M = 9.218 Secs
For Yorkie @ 2.93G
32M = 13.859

This translates to a increase in speed of 50.3% (these are time to complete so you need to take [(1/slower)-(1/faster)]/(1/slower) to calculate the actual rate, or you could convert into digitis per second and then it calculates out like FPS in game).

For Nehalem @ 2.93 G
1024M = 288.937
For Yorkie @ 2.93 G
1024M = 439.39

This translates to about 52% speed improvement.

This is estimated because I am using 1.58 Wprime, JC is using 1.55 Wprime.

Jack

wow, this is dropping some amazing numbers!!!! too bad we have to wait till q1 09 for retail :(

Thanks for the tests JC, we all appreciate it :up:

Quote:

---

Originally Posted by JumpingJack

When I did my back of the envelop calculation, I used the 32M numbers... so here is what I get...

For Nehalem @ 2.93G
32M = 9.218 Secs
For Yorkie @ 2.93G
32M = 13.859

This translates to a increase in speed of 50.3% (these are time to complete so you need to take [(1/slower)-(1/faster)]/(1/slower) to calculate the actual rate, or you could convert into digitis per second and then it calculates out like FPS in game).

For Nehalem @ 2.93 G
1024M = 288.937
For Yorkie @ 2.93 G
1024M = 439.39

This translates to about 52% speed improvement.

This is estimated because I am using 1.58 Wprime, JC is using 1.55 Wprime.

Jack

---

Thanks Jack... I realized the error of my post immediately and deleted it so it's not clear what you are responding to now... for everyone else's benefit, I was wondering how the 50% improvement was being calculated.

Quote:

---

Originally Posted by JumpingJack

When I did my back of the envelop calculation, I used the 32M numbers... so here is what I get...

For Nehalem @ 2.93G
32M = 9.218 Secs
For Yorkie @ 2.93G
32M = 13.859

This translates to a increase in speed of 50.3% (these are time to complete so you need to take [(1/slower)-(1/faster)]/(1/slower) to calculate the actual rate, or you could convert into digitis per second and then it calculates out like FPS in game).

For Nehalem @ 2.93 G
1024M = 288.937
For Yorkie @ 2.93 G
1024M = 439.39

This translates to about 52% speed improvement.

This is estimated because I am using 1.58 Wprime, JC is using 1.55 Wprime.

Jack

---

I thought I used v1.63 WPrime ? ...

Quote:

---

Originally Posted by JCornell

I thought I used v1.63 WPrime ? ...

---

:) ooops, you did.... my point, we are not using the same.... :)

Edit ... my mistake was glancing through this page and saw Metro's screen shot which is 1.55 ... sorry.

Jack

Very impressive Wprime speeds..

Would it be possible to change the thread count in ADV settings to 1 Thread.. Then we can really see how much SMT is helping the score?

Quote:

---

Originally Posted by villa1n

wow, this is dropping some amazing numbers!!!! too bad we have to wait till q1 09 for retail :(

Thanks for the tests JC, we all appreciate it :up:

---

You mean Q4 2008...

Quote:

---

Originally Posted by GoThr3k

i know nehalem exists at a freq of 3.2ghz, but i havent seen any higher clocked nehalems, altough i believe they do exist....
i know shintai has a very good knowledge about stuff, altough i dont consider him as a friend
most of the time i can agree with hem, but sometimes he's going too much to the blue side, but i respect someone's opinion (not like you)

---

:confused:
I'm not quite grasping your point... are you trying to say that while we currently don't know what the max Ghz Intel will release the new Nehalem @, it probably can't be faster than the current 3.2Ghz already known?

Knowing Intel always takes a conservative stance, which coincidentally has been proven by the EXTREMELY ingenious people here and elsewhere by overclocking them. Somehow, since nobody has officially posted Ghz in excess of 3.2gig, it somehow impossible or wrong to discuss them?




-Xoulz

Quote:

---

Originally Posted by Donnie27

Dewd, true Quad is Quad=P Doesn't matter if it is Monolithic or MCM, it is still Quad. The whole "True Quad Core" was crap and nothing more than Marketing Hype or a Gimmick.

---

They call it sarcasme ..

Quote:

---

Originally Posted by Neuuubeh

Can you wait that long :p. I'm toying with the idea of building a cheapie 24/7 render slave, looks like CPU's are getting cheaper :). Nehalem looks tempting, however I cant really believe we'd be able to find it available any time soon after release :(

---

Actually I can. I just picked up a refurb Dell Optiplex 755 with a E8500 for just over $600. I tried to build something similar at Newegg, but even using the cheapest components for everything the lowest total was >$700. Plus it comes with a three year on-site (in this case, meaning at home) warranty. This will easily last me until next year until the price/supply for Nehalem has stabilized. After that, I will pass it on to my wife.

Quote:

---

Originally Posted by Horus´

They call it sarcasme ..

---

my bad:(

From another thread I am wondering: Is it confirmed that there will be <600$ Nehalems in Q4 2008?

Quote:

---

Originally Posted by Seraphiel

From another thread I am wondering: Is it confirmed that there will be <600$ Nehalems in Q4 2008?

---

There will be 3 nehalems in Q4. 3.2, 2.93 and 2.66Ghz. The 2.66ghz should be 316$, worst case would be 540$. But 316$ is the most likely since its in the same group as a Q9550 that got a pricedrop in Q3.

Quote:

---

Originally Posted by Shintai

There will be 3 nehalems in Q4. 3.2, 2.93 and 2.66Ghz. The 2.66ghz should be 316$, worst case would be 540$. But 316$ is the most likely since its in the same group as a Q9550 that got a pricedrop in Q3.

---

Excellent! I thoguht it was only the 3.2 bloomfields that would be avail q4 :up:

Quote:

---

Originally Posted by Shintai

There will be 3 nehalems in Q4. 3.2, 2.93 and 2.66Ghz. The 2.66ghz should be 316$, worst case would be 540$. But 316$ is the most likely since its in the same group as a Q9550 that got a pricedrop in Q3.

---

Thank you :)

Quote:

---

Originally Posted by villa1n

Excellent! I thoguht it was only the 3.2 bloomfields that would be avail q4 :up:

---

Seems the FUD did its job.. FudZilla and other sites have been reporting this 'only extreme edition this year' and 'LGA 1366 will only be extreme edition, and only LGA 1366 will overclock' so many times now that I guess people are starting to believe it. :shrug: Maybe LGA1160 won't overclock, but LGA1366 will definitely be affordable by any enthusiast in 2009, if not already this year.

Let's say it again, the roadmaps have been clear as crystal since April and confirmed several times recently. Three price brackets, the ones that Q9450, Q9550 and some QX processor are currently filling, will launch this year. It's also been said by some site (TweakTown I think it was, and yeah, I know.. TT..) that it might come 'earlier' than we expect in Q4 as opposed to December 31. ;)

Quote:

---

Originally Posted by bowman

Seems the FUD did its job.. FudZilla and other sites have been reporting this 'only extreme edition this year' and 'LGA 1366 will only be extreme edition, and only LGA 1366 will overclock' so many times now that I guess people are starting to believe it. :shrug: Maybe LGA1160 won't overclock, but LGA1366 will definitely be affordable by any enthusiast in 2009, if not already this year.

Let's say it again, the roadmaps have been clear as crystal since April and confirmed several times recently. Three price brackets, the ones that Q9450, Q9550 and some QX processor are currently filling, will launch this year. It's also been said by some site (TweakTown I think it was, and yeah, I know.. TT..) that it might come 'earlier' than we expect in Q4 as opposed to December 31. ;)

---

Black Friday:up:

Quote:

---

Originally Posted by Xoulz

:confused:
I'm not quite grasping your point... are you trying to say that while we currently don't know what the max Ghz Intel will release the new Nehalem @, it probably can't be faster than the current 3.2Ghz already known?

Knowing Intel always takes a conservative stance, which coincidentally has been proven by the EXTREMELY ingenious people here and elsewhere by overclocking them. Somehow, since nobody has officially posted Ghz in excess of 3.2gig, it somehow impossible or wrong to discuss them?




-Xoulz

---

You are responding to a guy who has been banned :shrug:

Quote:

---

Originally Posted by Seraphiel

From another thread I am wondering: Is it confirmed that there will be <600$ Nehalems in Q4 2008?

---

This is what we know about this year's Nehalem's products...

http://members.shaw.ca/virtualrain/n...ield-specs.png

Quote:

---

Originally Posted by virtualrain

This is what we know about this year's Nehalem's products...

http://members.shaw.ca/virtualrain/n...ield-specs.png

---

p1 needs to come down to 499 ^^ But i m losing hope that deneb, or any of the 45nm k10 iterations will provide any market pressure on intel to lower prices in the next 6-8 months :(

thanks for clarifying though, i ve read so much stuff back and forth, good to have something solid. Time to plan out the next machine :D

Quote:

---

Originally Posted by villa1n

p1 needs to come down to 499 ^^ But i m losing hope that deneb, or any of the 45nm k10 iterations will provide any market pressure on intel to lower prices in the next 6-8 months :(

thanks for clarifying though, i ve read so much stuff back and forth, good to have something solid. Time to plan out the next machine :D

---

Why bother, the only difference is the multiplier..

I have a feeling the lowest priced one will have enough headroom as it is, a sky-high multiplier and a low reference clock should give us lots of room to expand.

Quote:

---

Originally Posted by bowman

Why bother, the only difference is the multiplier..

I have a feeling the lowest priced one will have enough headroom as it is, a sky-high multiplier and a low reference clock should give us lots of room to expand.

---

well and binning ;) but it remains to be seen how far past stock we can push these guys yet :P I m sure your right though.

Performance should be either lower or higher. 850 or 540$.

rofl L3 is slowish :D

Quote:

---

Originally Posted by Horus&#180;

...

---

I love your useful and enlightening posts.

JCornell thanks for testing. Could you give us a preview of its single core performance, maybe start your own thread in the news/intel section?

Quote:

---

Originally Posted by Jacky

I love your useful and enlightening posts.

---

Thx for letting me know this usefull and enlightening opinion of yours :up:

Quote:

---

Originally Posted by Shintai

Performance should be either lower or higher. 850 or 540$.

---

So does the L3 run at the same speed as the cores? :p:

Pricing seems to be better than i expected. Benches so far havebeen quite promising, but we haven't seen any meaningful workloads tested yet.

Quote:

---

Originally Posted by BrowncoatGR

So does the L3 run at the same speed as the cores? :p:

Pricing seems to be better than i expected. Benches so far havebeen quite promising, but we haven't seen any meaningful workloads tested yet.

---

Yes

Quote:

---

Originally Posted by Blauhung

It's on a separate clock and power domain. so, it will have to be based off the main reference clock just like the CPU, but both will have their own independent ratios.

---

How bout now?

Quote:

---

Originally Posted by BrowncoatGR

How bout now?

---

http://www.anandtech.com/cpuchipsets...spx?i=3326&p=5

Quote:

---

The L3 cache is quite possibly the most impressive, requiring only 39 cycles to access at 2.66GHz. The L3 cache is a very large 8MB cache, 4x the size of Phenom's L3, yet it can be accessed much faster. In our testing we found that Phenom's L3 cache takes a similar 43 cycles to access but at much lower clock speeds (2.0GHz). If we put these numbers into relative terms it takes 21.5 ns to get a request back from Phenom's L3 vs. 14.6 ns with Nehalem's - that's nearly 50% longer in Phenom.

---

And thats for a 2.66Ghz CPU ofcourse.

Quote:

---

Originally Posted by Shintai

http://www.anandtech.com/cpuchipsets...spx?i=3326&p=5



And thats for a 2.66Ghz CPU ofcourse.

---

Hmm looks like i might be wrong if Anand actually tested it or had reliable info on it , but seems to me Anand has a mistake there. Isn't cache latency always measured in relation to core frequency(even when it's async that should be 43 core cycles and not 43 NB/cache cycles)?

Quote:

---

Originally Posted by BrowncoatGR

Hmm looks like i might be wrong if Anand actually tested it or had reliable info on it , but seems to me Anand has a mistake there. Isn't cache latency always measured in relation to core frequency(even when it's async that should be 43 core cycles and not 43 NB/cache cycles)?

---

Hmm? Its for the cache to react and fetch data and send it back.

Quote:

---

Originally Posted by Shintai

Hmm? Its for the cache to react and fetch data and send it back.

---

How would you measure that directly at the L3? You would have to measure how long it takes for the core to fetch the data.

Quote:

---

Originally Posted by BrowncoatGR

How would you measure that directly at the L3? You would have to measure how long it takes for the core to fetch the data.

---

Ofcourse, but its the cache that does the wast majority of the work and you can calculate the cycles used there when you know its speed. Same way we do with main memory. On a 2.5Ghz Phenom it might actually have used 53-54 CPU cycles.

If you compare CPU-Z cycle measurement they varies with others. Sciencemark would say 12 or 13 cycles depending on stride size for a yorkfield.

Looks like my next purchase will be an Intel cpu :)
When is the cpus meant to be hitting the market? November/december time?

Quote:

---

Originally Posted by azza21

Looks like my next purchase will be an Intel cpu :)
When is the cpus meant to be hitting the market? November/december time?

---

'Q4', TweakTown said to expect it 'sooner than expected' whatever that means..

http://www.tweaktown.com/news/9583/i...pei/index.html

Quote:

---

While Intel did not mention anything about clock speeds to us at all, word from other press around Computex is that it could be running at 3.2GHz and it could be ready to rock and roll sooner than expected but you didn’t read that here.

---

I can confirm the CPUs are effectively done. They could be rolled out in under 1 month if Intel wanted to.

The X58 chipset and getting mobos for them are the biggest issue.

Quote:

---

Originally Posted by mstp2009

I can confirm the CPUs are effectively done. They could be rolled out in under 1 month if Intel wanted to.

The X58 chipset and getting mobos for them are the biggest issue.

---

That's great to hear.

We've only heard about Bloomfield desktop chips so far - will there be UP and DP Xeons around too this year?

Quote:

---

Originally Posted by Shintai

Yes

---

Are you suggesting that the L3 will run at core speed?

According to Kanter (RWT), the L3 would most likely run on it's own clock/power domain... and if the Phenom is any indication, it's clock/power might be tied to the IMC clock/power domain.

Quote:

---

Although Intel has not discussed the physical design of Nehalem at all, it appears that the L3 cache sits on a separate power plane than the cores and operates at an independent frequency. This makes sense from both a power saving and a reliability perspective, since large caches are more susceptible to soft errors at low voltage. As a result, the load to use latency for Nehalem varies depending on the relative frequency and phase alignment of the cores and the L3 itself and the latency of arbitration for access to the L3. In the best case, i.e. phase aligned operation and frequencies that differ by an integer multiple, Nehalem’s L3 load to use latency is somewhere in the range of 30-40 cycles according to Intel architects.

---

http://www.nehalemnews.com/2008/04/w...e-nehalem.html

Cheers,
-Chris.

Quote:

---

Originally Posted by mstp2009

I can confirm the CPUs are effectively done. They could be rolled out in under 1 month if Intel wanted to.

The X58 chipset and getting mobos for them are the biggest issue.

---

This was my impression following Computex... it was the motherboards that were holding things up.

Any indications on branding for Nehalem?

Cheers,
-Chris.