Nehalem-EP......BLOOMFIELD

[Chri$ch]

But how does it perform when HT is enabled?

all tests @ 1.6Ghz (eist is enabled)













http://valid.x86-secret.com/show_oc.php?id=385481

[Seraphiel]

im glad i didnt buy a new penryn cpu and just wait for nehalem lol. i only wish beckton wasent released in Q4 of 2009. but then again i can always go for dunnington.

image off of xtreview

I thought Beckton was 8 cores / 16 threads

[Shintai]

im glad i didnt buy a new penryn cpu and just wait for nehalem lol. i only wish beckton wasent released in Q4 of 2009. but then again i can always go for dunnington.

image off of xtreview

Dunnington is Penryn based and Beckton is most likely out of your reach finacially. These are 1500-2500$ chips, best used with 4 or more at a time.

I thought Beckton was 8 cores / 16 threads

It is 16 threads.

[Seraphiel]

It is 16 threads.

Thanks

So the infos in the picture regarding that is wrong.

[Drwho?]

Far from your expectation


FYR
...

1) Don't forget your NDA.
2) This is an old silicon.

Thanks for not informing our competitors, you are hurting us when you publish bench results, you all need to understand that.

I know it is fun to publish your proto numbers, just remember that you are as well hurting the competitiveness of my employer.

I am speaking for myself only here, not for my employer here.
It makes my job very hard , when my competitor can know what I am planning, before it is even finish. There are no personal archivement to release confidential information, except taking the risk to be in trouble.
Take it easy, and just think about what you are doing, before you do it.

Just think about it as "you work on something very cool, a good cake for your Daughter birth day, but a dude show the picture of the cake to your daughter before the D day." That kind of suck, isn't it?

Francois/who?

[Seraphiel]

1) Don't forget your NDA.
2) This is an old silicon.

Thanks for not informing our competitors, you are hurting us when you publish bench results, you all need to understand that.

I know it is fun to publish your proto numbers, just remember that you are as well hurting the competitiveness of my employer.

I am speaking for myself only here, not for my employer here.
It makes my job very hard , when my competitor can know what I am planning, before it is even finish. There are no personal archivement to release confidential information, except taking the risk to be in trouble.
Take it easy, and just think about what you are doing, before you do it.

Just think about it as "you work on something very cool, a good cake for your Daughter birth day, but a dude show the picture of the cake to your daughter before the D day." That kind of suck, isn't it?

Francois/who?

You assume too much. JC may not be under a NDA at all, or at least not breaking it. Also, remember the PR campaign before Core 2 Duo was released?

Intel's competition is AMD, with respects to the various benchmarks shown. What does AMD have to offer against Nehalem, with respects to pure performance?

EDIT: The fight for the next gen has already begun, but as usual, AMD is unfortunately not showing its cards yet.

That is how I see it at least.

[Shintai]

Dont wet Dave again...those Boinc results are superb. Remember its per core...and its with SMT. One quad with SMT is near the performance of 2 quads.

[generalpetres]

yeah the pic i had they put the threads in wrong. but i am not going to wait another year for a new comp just for 8 core beckton. although it would be nice, but if i keep waiting and waiting something better then beckton will come out and i would want that lol. why cant they come up with something faster every 2 years instead of every year make my life much easier lol.

[Donnie27]

yeah the pic i had they put the threads in wrong. but i am not going to wait another year for a new comp just for 8 core beckton. although it would be nice, but if i keep waiting and waiting something better then beckton will come out and i would want that lol. why cant they come up with something faster every 2 years instead of every year make my life much easier lol.

Actually if Intel's plans work, they come out with something new once a year that's faster under their Tick Tock scheme. Un talked about are tweaked steppings models of those every six months or so in between the Tick Tock.

[bowman]

im glad i didnt buy a new penryn cpu and just wait for nehalem lol. i only wish beckton wasent released in Q4 of 2009. but then again i can always go for dunnington.

image off of xtreview

Beckton is an MP product based on an entirely different socket, LGA1567. It will not be marketed towards desktops or ordinary consumers in any way and will be extremely expensive.

Dunnington is a Xeon product designed to replace Tigerton Xeon processors in servers - it too is no desktop product, and a quick Google will reveal the bulk prices of these chips, which should put you off the idea quickly. Not to mention that it's just three Penryn dies and massive amounts of L3 on one die, not really worth the cost for regular joes.

According to pcwatch, Westmere will be a six-core product based on the Nehalem architecture. Might want to wait for that instead.

[Movieman]

1) Don't forget your NDA.
2) This is an old silicon.

Thanks for not informing our competitors, you are hurting us when you publish bench results, you all need to understand that.

I know it is fun to publish your proto numbers, just remember that you are as well hurting the competitiveness of my employer.

I am speaking for myself only here, not for my employer here.
It makes my job very hard , when my competitor can know what I am planning, before it is even finish. There are no personal archivement to release confidential information, except taking the risk to be in trouble.
Take it easy, and just think about what you are doing, before you do it.

Just think about it as "you work on something very cool, a good cake for your Daughter birth day, but a dude show the picture of the cake to your daughter before the D day." That kind of suck, isn't it?

Francois/who?

With respect.
I can understand how you feel and in your position I would feel the same.
No one here wants to hurt any company in any way.
What you see is the love of the hobby and the desire to share that passion with others that feel the same.
The release of information is a two edged sword for companies.
On the one hand it does as you mentioned give info to the competitor before the company would want that info known.
On the other hand as in this case it builds excitement among the potential buyers and in Nehalems case with it's huge increases will result in people saving their money and purchasing your companies product.
Now I am sitting here with a new Harpertown rig(E5450's) and two 8 core clovertown rigs plus a Q6600 so you know where my buying power is spent.
You also saw my excitement at the benchmarks that have been shown.
The bottom line is that although I understand perfectly how you feel I think that in at least this case your company may gain more than it loses.
it looks like a great product and I can't wait to get one.
Any spares laying around?

[karoshi]

1) Don't forget your NDA.
2) This is an old silicon.

Thanks for not informing our competitors, you are hurting us when you publish bench results, you all need to understand that.

I know it is fun to publish your proto numbers, just remember that you are as well hurting the competitiveness of my employer.

I am speaking for myself only here, not for my employer here.
It makes my job very hard , when my competitor can know what I am planning, before it is even finish. There are no personal archivement to release confidential information, except taking the risk to be in trouble.
Take it easy, and just think about what you are doing, before you do it.

Just think about it as "you work on something very cool, a good cake for your Daughter birth day, but a dude show the picture of the cake to your daughter before the D day." That kind of suck, isn't it?

Francois/who?

1) Your multibillion competitor already has this information. If there are a good number of ES flying around (mobo partners, software partners, ...) some of them will spill the beans to your competitor. Not that your competitor would pay for the information, that would illegal and inmoral.

2) In this point in time your competitor is yourself. There is a point in not wanting this information disclosed to the enthusiast communitiy. They should buy a penryn (or 3) now and a Nehalem (or 3) in 9 months.

my €0.02

[JCornell]

See if there is a setting in either Bllomfield or Gainstown to shut off the HT and then try and run that bench in BOINC again.
I think you will be very surprised, as will we all.

HT Disabled


...

[JCornell]

@JCornell

Is it possible to run the new TrueCrypt 6.0 Benchmark function (100 mb buffer) with multi core support on the Nehalem system?
http://www.truecrypt.org/downloads.php

cu, BitpowerPM

...

[bowman]

For comparison..


What does a C2Q do here?

[JCornell]

For comparison..


What does a C2Q do here?

From the screenshot inside the TrueCrypt site (Q9450 @2.66GHz)


...

[Hornet331]

for comparison: QX9650 @ stock 3ghz with DDR3 666mhz @ 8-6-5-16 1T with loads of crap in background running.

that thing is avg. 24% faster then a similar clocked penryn.

[JCornell]

for comparison: QX9650 @ stock 3ghz with DDR3 666mhz @ 8-6-5-16 1T with loads of crap in background running.

that thing is avg. 24% faster then a similar clocked penryn.

Before jumped to the conclusion, re-run with no crap running in the background, just for fair comparison ...
Other else, just wait for the author comment

[Anemone]

Dr W -
Even you would have to admit the risk of competitor knowledge (of which at this point they probably have enough) is likely lower than the benefits of putting some large amount of members of this and other forums on the shopping list for a Nehalem when it arrives. I mean there is a reason you sell the most expensive chip first right? Does not some amusement come from watching us just sitting here dying to run off to some website and hit "add to cart" the moment this chip hits the shelves?

I mean no disrespect, but rather there has to be some light in this tunnel I hope. And since you've been more than straight with us in the past on numerous topics I'm putting this plainly as well. If you find this painful to see, and I have not lightened your day at all, my apologies.

[villa1n]

...

Those are impressive twofish rates...if only they had the mobile quad coming out soon hehe, as i don't lug my desktop around when i m travelling through the states on business. hehe.

[Movieman]

Put's on Mod hat;
Gentleman;
With respect to Dr. Who's comments here:
We may disagree in our pursuit of knowledge and our excitement over the potential of the Nehalem cpu's but at the same time we have to understand his perspective.
He was the project manager for the Skulltrail system for Intel and that should explain his comments fully.
He is someone to be respected and not, I repeat NOT to be ridiculed under any circumstances.
The knowledge he imparted to the users here in the SkullTrail thread was nothing but a plus for anyone that was working with that board.
I ask again, respect the man even if you differ with his opinions.
He deserves that at the very least from all of us.
Thanks for reading.

[Jupiler]

Nothing more I can add to what Dave just wrote.

Some comments on this page were just plain rude and totally uncalled for (you guys know what happens next...).
So, keep it nice and clean, guys.

[Seraphiel]

HT On:
2786 Whet / CPU
7095 Dhry / CPU

HT Off:
3152 Whet / CPU
7065 Dhry / CPU

HT switched off sure didn't give the expected results in BOINC. If anything they confuses.

[Anemone]

If I created offense, it was not intended. But I apologize nonetheless, because that's not what any of us are here for.

[informal]

HT On:
2786 Whet / CPU
7095 Dhry / CPU

HT Off:
3152 Whet / CPU
7065 Dhry / CPU

HT switched off sure didn't give the expected results in BOINC. If anything they confuses.

Hmm exactly,something is off with this bench..Either HT isn't working correctly or it is working stellarly and is producing 2x the scores ..Which in turn is highly unlikely (how can a HT enabled Nehalem chip score 2x the physical scores with HT enabled??).We need actuall points per day the Nehalem puts out,not the built-in tool to benchmark it.The actual ppd will tell us is the HT really this good or is it messing with the scores.