Hey, there are free and academic (much lower cost) licenses:
http://www.intel.com/cd/software/products/asmo-na/eng/219771.htm
http://www.intel.com/cd/software/products/asmo-na/eng/219603.htm#academic

This compiler IS THE fastest one. I really can't understand why most of/all boinc projects don't use ICC - benefits are clear - more processing power, more satisfaction of users (faster credits, more units computed etc.).

EDIT:

Guys, I found a person who will compile Rosetta with SSE, SSE2, SSE3, SSSE3. Now I must "only" obtain source code ;) Non-commercial version of ICC will be used.

---

I agree. If someone could start a fund raising campaign for a new compiler, I bet you could raise the money in a couple of weeks. You'd probably need to get the department admins involved so all donations go to this special university "project".

Piotr, you may want to post in David Baker Journal feedback or contact them directly.

As for fund-raising, money is not normally the problem (and I'm sure Washington university have the compilers or can afford them already, certainly the Gates Foundation funding would be enough as well. That and they're from Who?'s post up there, in talks with Intel...

---

Team mauisun.org

That officials think of this all ????
What motives at them not to realise optimisation favourable by it.

---

That officials think of this all ????
What motives at them not to realise optimisation favourable by it.

We would have to wait for the project Admin to comment (maybe a mod could ask them the question and the status of other platforms...

But time and resources are probably a major factor since comments from both Who? (An Intel Senior Performance Analyst) and Mats Petersson (AMD programmer ?) say converting to x64 or utilising SIMD instructions for this project would inclide a major re-write of the code to see any major benefit.

Though Who? still stalks the halls, Mats seems to have stopped crunching now.

---

Team mauisun.org

Is the deal between Rosetta and Intel going to put AMD machines at a disadvantage, or maybe even phase out AMD machines in the future?

---

Is the deal between Rosetta and Intel going to put AMD machines at a disadvantage, or maybe even phase out AMD machines in the future?

Hmm, using ICC10 should speed up computations on both Intel and AMG CPUs. Althrough ICC is known to have some anti-amd rountines (causes program to run not at full speed on AMD). But 1. Maybe in version 10 this problem does not occur 2. There are patches that cause programs to run on full speed on AMD CPUs.

Anyway ICC is worth trying !

Peter

---

Is the deal between Rosetta and Intel going to put AMD machines at a disadvantage, or maybe even phase out AMD machines in the future?

Hmm, using ICC10 should speed up computations on both Intel and AMG CPUs. Althrough ICC is known to have some anti-amd rountines (causes program to run not at full speed on AMD). But 1. Maybe in version 10 this problem does not occur 2. There are patches that cause programs to run on full speed on AMD CPUs.

Anyway ICC is worth trying !

Peter

Part of those "anti-AMD" routines is caused by mistake in detection-code,where it tries to detect old version of Athlon,which had broken support for SSE,but it did inversly and used slower path for CPUs with correct support.This has been lengthly discussed at Einstein.So far nobody to my best knowledge made tests to show,that Intel is slowing AMD chips.

SSE4.1 is comming !!!!!!!!!
BUT Rosetta still use SSE1
OMG!!!!

---

SSE4.1 is comming !!!!!!!!!
BUT Rosetta still use SSE1
OMG!!!!

IIRC Rosetta does not use even SSE !

---

Somehow it looks like the whole SSE4/SSE5 "competition"/implementation, on Intel/AMD respectively, isn't going to matter much to Rosetta! : )

---

Crunching Rosetta as a member of the Whirlpool BOINC Teams

I know everybody's busy and these things take a long time to code, but if the science that Rosetta does is so important, wouldn't speeding things up from now on by x% be very important?

I'm sure this has been considered and that the Rosetta team has looked at the opportunity cost, but maybe there's a third way; maybe it's possible to make a public call for optimizations experts or whatever and try to get this done without taking time away from the people who work on the science code. I'm sure there's a coder out there who would love to add this challenge to his/her portfolio.

I guess it depends on whether what is needed is just a good programmer or someone who also understand the protein science too...

In any case, I certainly would love to see more of the potential of modern CPUs being used by this awesome project. Making better predictions and finding cures faster seems like it's worth the effort.

---

I know everybody's busy and these things take a long time to code, but if the science that Rosetta does is so important, wouldn't speeding things up from now on by x% be very important?

I'm sure this has been considered and that the Rosetta team has looked at the opportunity cost, but maybe there's a third way; maybe it's possible to make a public call for optimizations experts or whatever and try to get this done without taking time away from the people who work on the science code. I'm sure there's a coder out there who would love to add this challenge to his/her portfolio.

I guess it depends on whether what is needed is just a good programmer or someone who also understand the protein science too...

In any case, I certainly would love to see more of the potential of modern CPUs being used by this awesome project. Making better predictions and finding cures faster seems like it's worth the effort.

The beta I'm on has doubled the output by using the sse instructions. I know that the intels cpus are benifiting the most by this beta so far. I think the higher fsb and cache of the newer intels is a larger facter too.

---

Jmarks

The beta I'm on has doubled the output by using the sse instructions.

what beta is that?

---

The beta I'm on has doubled the output by using the sse instructions.

what beta is that?

Folding at Home SMB Windows beta.

---

Jmarks