AMD Changes Server Platform Roadmap

[Neuromancer]

...
Infact, here's a simple experiment we did with system power.

45W CPU - X2 4050e inside 780G MB, 330W SS II PSU = 37W idle (AC)
45W CPU - X2 4050e inside 780G MB, 330W SS II PSU = 67W load (AC)
95W CPU - X4 9500 inside 780G MB, 330W SS II PSU = 83W idle (AC)
95W CPU - X4 9500 inside 780G MB, 330W SS II PSU = 172W load (AC)

45W CPU - X2 4050e inside 780G MB, 330W SS II PSU 800MHz 0.75v = 33W idle (AC)
45W CPU - X2 4050e inside 780G MB, 330W SS II PSU 800MHz 0.75v = 48W load (AC)
95W CPU - X4 9500 inside 780G MB, 330W SS II PSU 800MHz 0.8v = 79W idle (AC)
95W CPU - X4 9500 inside 780G MB, 330W SS II PSU 800MHz 0.8v = 89W load (AC)...

The quad core is using 7 Watts less than twice the power of the dual core CPU at 800MHZ .8v loaded

I would say that is something.

Granted the 9X50e cpus are about twice as energy efficient using not much more power than that 4050e you tested

 

[ChanceCoats123]

So you compared a dualcore to a quadcore and you confirmed that the latter draws more power as everyone expected.
How does that prove your point that the Energy efficient models are more than just lower clocked and undervolted prossors ?



In case you are unaware the EE processors are still the same stepping as the others so this proves nothing, again.

You really should read some "botched reviews" as you call them, they even manage to compare the same architectures and steppings to each other.

They basically are undervolted and underclocked versions of the same cpu. I have seen 4x00e's and 5x00e's hit the same if not higher clock speeds as non "e" cpus and they used the exact same voltage on average.

I mean, unless I missed something big, which is definately possible with me. Just point it out please.

 

[solstice]

But you're talking servers here. The server community probably thinks an OC'ers idea of "stability" is laughable. Just like not all (nor even most from what I've seen) DDR2-1066CL5 RAM can run DDR2-800CL4 not all CPU's have consistently low power requirements at slower speeds ...

I must have a bipolar disorder. I think I am the only one I am aware off that overclocks and use ECC memory. I wish I could use chipkill feature on AMD system though. I was limited by the inconsistent implementation of bios from ASUS. Sigh, why must they torture me so.

 

[Kuroimaho]

They basically are undervolted and underclocked versions of the same cpu. I have seen 4x00e's and 5x00e's hit the same if not higher clock speeds as non "e" cpus and they used the exact same voltage on average.

I mean, unless I missed something big, which is definately possible with me. Just point it out please.

That's what I said earlier, E is undervolted, undercloked, but same stepping same rev, but according to KTE the E versions draw less power even with the same clockspeed, NBspeed and V.
Then he starts to compare different arches and rev for some reason but calls those reviews which compare E against non E lame.

I guess I also miss his point here.

 

[Kuroimaho]

I think I am the only one I am aware off that overclocks and use ECC memory. I wish I could use chipkill feature on AMD system though. I was limited by the inconsistent implementation of bios from ASUS. Sigh, why must they torture me so.

You are not alone, anyone who upgraded from K8 opty to the K10 based ones using an L1N64 mobo had to switch to ECC memory as well.

AMD Demonstrates Twelve Core Microprocessors

Now that's something.

 

[Evilsizer]

to me that looks photochopped...

 

[Kuroimaho]

You are right, it is, just checked AMD's opteron page, I thought only the logo was added. ^^;;

 

[KTE]

I'd like to see how much their HT Assist really helps performance of Istanbul. I've heard too many lauds about it already.

 

[Kuroimaho]

Well who doesn't ? What interests me more is to know whether it helps enoguh in 2 way systems that they relaunch Fasn8 ? Maybe next week we will know.

 

[KTE]

I've not heard any such talks currently but I've been staying very aloof from the computing world side of things for 6-7 months now, so who knows. I'm one of those guys who jumped on their old 4x4 and ended up being burnt from many angles so I'm remaining very cautious with such offering now. Very. Specifically the issue of future support, upgrades and compatibility.

The prime reason I'm interested in seeing Istanbul gains, efficiency and scaling, really, is due to the hype I've heard for 3 months now from professional MFG workers. Just want to see how correct they are or if they were exaggerating. Also a good chance to learn something new if it does end up giving a bigger performance boost than expected.

I'd be interested to see how Gulftown scales in comparison (if they do allow Istanbul on the high-end DT) although Gulftown seems to be another year off. By then, AMD is planned to have 8-core and 12-core server CPUs out, which could be a good follow-up and competition if they do decide to launch their FASN8 platform.

Too many IFs

 

[KTE]

Update

Well who doesn't ? What interests me more is to know whether it helps enoguh in 2 way systems that they relaunch Fasn8 ? Maybe next week we will know.

"There is no plan to bring Istanbul to the Desktop although 2P workstations will be available". Confirmed by AMD.

 

[Bobnova]

The line between workstation and desktop is mighty thin these days.
It's not like the modern 2P boards available are designed for desktop use, but people do it anyway.