Sign in with
Sign up | Sign in
Your question

Woodcrest vs Opteron..Interesting!

Tags:
  • CPUs
  • Opteron
  • Product
Last response: in CPUs
Share
May 23, 2006 7:19:26 AM

Pretty much just a preview, but fun to see the possibilities. I'm sure this

will get the same excuses and FUD as the IDF Conroe tests. :wink:


http://www.techreport.com/etc/2006q2/woodcrest/index.x?...

More about : woodcrest opteron interesting

May 23, 2006 7:27:57 AM

Quote:
Pretty much just a preview, but fun to see the possibilities. I'm sure this

will get the same excuses and FUD as the IDF Conroe tests. :wink:


http://www.techreport.com/etc/2006q2/woodcrest/index.x?...


Odd is it not that AM2 comes out and on the same day Core uArch previews emerge :)  ..... coincidence? I think not.Both Barrels? :wink:
May 23, 2006 7:41:43 AM

Quote:
I am excited because Intel has made a game of it, AMD will need to answer, then intel and so forth and so on.... hobbiest computing will become fun again!!!!

It's also compelling to watch the chess play at work -- go here, move there, release now, talk then, talk now.... process, architecture. Intel has let AMD manhandle them in server and DT for so long it was no longer pushing any limits -- this leap forward is what Intel can do when they refocus their energy and resources to a set of well defined goals.

The struggle in the war means good things are coming....

I am happy NDA's are starting to drop.... we should wake tomorrow morning and see all sorts of info flying around.

Jack

If, by chance, Intel manages to wrestle Server performance away form

AMD too, can you say "standing 8 count"? 8O lol
Related resources
May 23, 2006 7:43:05 AM

OMG, faked, floptimised, incorrect settings, totally wrong.
May 23, 2006 7:49:17 AM

Woodcrest does look impressive. Looks like opterons will have some serious competition.
May 23, 2006 7:52:36 AM

Outrageous power requirements on those FB-DIMMS, though....Ouch

Technological advance has it's price, but having to put a 120mm fan on

the RAM...not cool...(pardon the pun). :?
May 23, 2006 7:58:00 AM

Quote:
Outrageous power requirements on those FB-DIMMS, though....Ouch

Technological advance has it's price, but having to put a 120mm fan on

the RAM...not cool...(pardon the pun). :?


I just hope that they don't screw the pooch with this FB-DIMM, like they did

with RDRAM. :cry: 
May 23, 2006 8:03:37 AM

Its good to see that woodcrest runs faster in 64bit.

I'm pretty impressed overall.
May 23, 2006 8:08:58 AM

Quote:
Its good to see that woodcrest runs faster in 64bit.

I'm pretty impressed overall.
Yes...The fanboys have been putting

alot of weight into Opteron's 64 bit abilities, lately...Like it is there only

remaining arguement. :) 
May 23, 2006 8:10:46 AM

Quote:
will get the same excuses and FUD as the IDF Conroe tests.


Probably not. Intel sent the systems to Tech-Report, and would be stupid for them if the scores went lower. Rather, I expect opposite, it'll be even a bit higher.

They also say: "We marched into Intel's reviewer's workshop with a shedload of workstation benchmarks, ready to test out Woodcrest as advertised, and found out we'd be testing the Bensley server platform."

So they have their own benchmarks.

Quote:
I just hope that they don't screw the pooch with this FB-DIMM, like they did

with RDRAM.


We'll keep going with this DDRx thing, until maybe Intel decides to put integrated memory controller then we go FB-DIMM. The difference from RD-RAM, is that, there is wide industry support, even from AMD.
May 23, 2006 8:12:12 AM

I'd put the single chip setup about 20% faster, but the dual chip, dual core closer to 10%. what's your take?
May 23, 2006 8:13:18 AM

Quote:
Outrageous power requirements on those FB-DIMMS, though....Ouch

Technological advance has it's price, but having to put a 120mm fan on

the RAM...not cool...(pardon the pun). :?


I just hope that they don't screw the pooch with this FB-DIMM, like they did

with RDRAM. :cry: 

You or anyone else can argue with me on this, I don't think they screwed the pooch on RDRAM so much as the market simply rejected the cost of the transistion. Intel had no choice but to support DDR as that would have pushed out their customer base.... why they dropped it I do not know, but costs were certainly a factor.

Jack

Yeah, RAMBUS got greedy.. They probably screwed themselves out of

ALOT of money over the years.
May 23, 2006 8:17:34 AM

Quote:
Its good to see that woodcrest runs faster in 64bit.

I'm pretty impressed overall.


I'm very impressed too. Makes me wonder how the desktops will run 64 bit.
May 23, 2006 8:22:56 AM

Quote:

so it was a compromise, get a interconnect technology that helps better than the FSB and the power consumption goes down another 10-20%.

So do you think that FB-DIMM is a must for Intel, to make up for the

inefficiency's of FSB?
May 23, 2006 8:30:36 AM

HMM RDram.
The real problem was that the price of all ram was very high at the time. New ram starts at s premium. It's price only comes down with high production.
Much as I hate RDram the company, It probably could have gone mainstream, if Williamette wasn't such a bad chip.
When I first heard about P4, I was very interested in getting one, and some RDram, if the performance was worth the price.
Unfortunately, with the royalties they get from every stick of memory sold today, I doubt the company cares.
May 23, 2006 8:36:45 AM

I'm going to build one new render farm.
Woodcrest eats my Opty 270 in 3ds Max.

Is Lucasfilm going to do the same?

Rendering, time is money and I could buy new render farm every 6mouths if I get 20% or more time saving.
May 23, 2006 11:38:11 AM

Quote:
With FB-DIMMS you can run the memory a bit faster with less problems and more secure data transfers and at higher capacity (i.e. more memory). There is much more to it than this, physically, I am hoping someone with more knowledge can help out. David C1??? -- you seem to be more knowledgable in this area.


The problem with DDR/DDR2 is that its hard to have enough memory capacity, and that's a must for servers. I read that when there was DDR200, four DIMMs a channel was possible, but with DDR400, it went down to two DIMMs. Likely, DDR2-400 allows four, but DDR2-800 and speeds between will support lower amount of DIMMs. The problem is well explained here:
Part 1: http://www.theinquirer.net/?article=15167
Part 2: http://www.theinquirer.net/?article=15189
Part 3: http://www.theinquirer.net/?article=15214

Basically, it is the shared bus architecture of the DIMM itself, a stub-bus configuration. What is that? Imagine a comb, with the bus being the spine of a comb, a dram on each tine, and the controller on one end. Each point where the tine touches the spine, you get an electrical discontinuity. Each of these little discontinuities is like a speed bump, it causes the signal to degrade.

Just like those nasty pieces of roadway, each of these stubs causes more problems the faster the signal goes. The simple solution is to decrease the number of DRAMs on the bus, so that while cranking up the speed makes each discontinuity worse, it means there are fewer of them. Speed or capacity, pick one.


So basically, by adding a buffer that allows point-to-point connection, the problem with stub bus goes away. That means, you can have many more DIMMs than previously possible.

FB-DIMM also has extensive features to help with data reliability, so the differentiation between ECC and non-ECC now disappears.

It's also good for motherboard manufacturers, since the FB-DIMM DDR2 only requires 69 pin per DIMM. That means vastly reduced design time, and unequal wire traces could be used, unlike now, which needs equal wire traces to ensure signal integrity, and complicates motherboard design.

Basically its an ULTIMATE server memory standard.


For desktop users, it seems we can probably keep going with the normal DDRx and so on. Non FB-DIMM is probably better for us anyway since we don't need huge capacity DIMMs or want to sacrifice latency for capacity.

Now here's a possible advantage that people will care. My understanding is that, it may be possible to not change the memory controller and have support for new memory standards as long as the buffer supports it. In reality, it will probably be more complex, but it looks good.
!