Anyone have a quick performance/watt between 6core xeon and 6core amd?

[aigomorla]

d.
The day AMD releases a 35W six core processor that works at more than 1.6Ghz, you call me okay?

:O

call me too please!!

i would most definitely want one for my car pc.

 

[Lonbjerg]

That's true. Depending on the price you might have to run them for many years before you compensate for the initial cost.

Not if you look at the time factor?
Do you work for free?
Well, neither do I and time spent waiting = more money spent of the same task = no savings, just adding more cost.

I am so tired of the flawed "Bang for Buck"...save now....spend more every second after that.

 

[Idontcare]

I am so tired of the flawed "Bang for Buck"...save now....spend more every second after that.

That's a great philosophy to have when you have infinite resources.

For people with a budget, either personal or at work, decision are required.

Do you drop $1000 on that i980X and pair it with a 5900rpm seagate drive...or do you spend $500 on a cheaper (and slower) cpu so you can also afford to pickup that $400 160GB G2?

I am so tired of the flawed "bang for the buck is flawed" statements.

 

[JFAMD]

Well, that depends on the application.

Applications with long execute times to get to a final answer fall into this. If you are compiling code or rendering video, higher speed can = higher productivity.

However, most of the standard data center apps don't really drive to that metric. If you run exchange, if server A gets you the message in 1 second and costs $5K and server B can do it in a half second and costs $7.5K, is it worth spending a 50% price premium to get that half second? For 99.9% of the customers, no, it is not worth it.

Most server traffic is spikes. Even database tends to be spiky in nature, so it is hard to justify spending a disproportionate amount of money. Typical server utilization is 10-20%, which generally means that there are periods of 50-80% utilization married up with idle time. If your server is sitting idle, then overpaying for higher performance doesn't make sense.

You want to focus the high performance parts at those workloads that are consistently high utilization (typically 10% of the market) and focus the rest of the effort on driving better price/performance for the ~90% of the servers that have lower utilization, spiky performance needs or aren't that critical.

That is why only ~3% of the SKUs sold are top bin - most workloads demand price/performance over raw performance. Bang for the buck.

 

[Markfw]

That's a great philosophy to have when you have infinite resources.

For people with a budget, either personal or at work, decision are required.

Do you drop $1000 on that i980X and pair it with a 5900rpm seagate drive...or do you spend $500 on a cheaper (and slower) cpu so you can also afford to pickup that $400 160GB G2?

I am so tired of the flawed "bang for the buck is flawed" statements.

+1

And I think that this same poster has argued this with me in another thread.

Its really getting old.

 

[The-Noid]

Budget is mostly irrelevant, we have these stupid sustainability and environmental audits though. Have to prove we are a "green company."

 

[Hacp]

Budget is mostly irrelevant, we have these stupid sustainability and environmental audits though. Have to prove we are a "green company."

Hahahaha.

 

[The-Noid]

I could give a two shits, clients like it though .

Don't get me started on the LED lights in the ceiling, makes my office look like the surface of the sun. Most of the time I turn the lights off.

 

[aigomorla]

I could give a two shits, clients like it though .

Don't get me started on the LED lights in the ceiling, makes my office look like the surface of the sun. Most of the time I turn the lights off.

ROFLROFL..

then BORG a bunch of Laptops and say your innovating.

 

[veri745]

that might night be necessarily true. Most chips with different features in the same generation are still the same.

you reallyd ont think some intel chips have VT-x or VT-d on the die, or hyperthreading or not, or turbo or not right. AMD can do the same thing, like shut off extra HT links so that they can segment their product line. it'll still be the same die, they can just gimp it for a certain market.

sure the pinouts might be different because if you dont hook up that extra HT link or extra ram channels you wont need more pins etc... but in the end its still the same die under there.

I think thuban is basically istanbul in opteron terms. Magny cours is like 2 istanbuls on the same die (thus the quad memory channels, which 2 instanbuls would have since each has dual channels and thus the extra required pins for those 2 memory channels).

From the little i've read socket F = ddr2 registered, socket C32 = dd3 registered. so its like am2 and am3 for servers

So they do need a new socket because the memory is registered and whatever extra pins and changes that requires. i am fairly sure ont he cpu die itself, that is probably something theycan turn on and off depending on what pinouts its hooked up to. it would be pretty inefficient to make 2 seperate dies for that, but if you just change the sockets and disable parts of the die, that woudl be the most efficient way to design a die

but that said, you need a new socket with different pins connected.

I don't know how you can say Lisbon and Thuban are the same processor.

Lisbon and Istanbul are pretty much the same. Istanbul is Rev D0. Lisbon is Rev D1. Magny Cours is Rev D1.

Thuban, on the other hand, is Rev E0. Would they be different silicon Revs if they were the same piece of silicon? No.

That's like saying Shanghai and Istanbul are the same silicon.

Same architecture, yes. Same silicon? No.