Bikeman, gratz to your new box. I have no idea about Apples, but it sounds like a nice little machine. And so fitting to use it for CPDN after going the "green" route about the old box ;-) I also used to do some work for that project, it's quite nice to never run out of work... but I hope you have a few percent left for Einstein ;-)
It's my first Mac. Wanted something compact, quiet and reasonably fast. Oh, and is does look cooler to sync your iPod with :-).
Don't worry, E@H S5R2 (just to stay on topic here) will benefit from my Mac mini: The Athlon XP 1800+ which used to do CPDN exclusively 24/7 will then do some E@H, and E@H will get half a core on the new Mac. I'm thinking about joining "Malaria" as well, sounds like a worthwhile project to me.
To go a bit off topic ;-) Malaria certainly is a worthwhile project, and it hasn't been too "beta" any more since a few months ago (before, there was trouble with scheduler probs and seemingly unmotivated computation errors) but as I said, you should have enough cache to run it properly. 512 MB is a bit too little, my Venice just sucked there. As I said, the Dothan does fine. No idea how much your Mac has.
So i don't think that the code will be optimised for SSE2. I know the C2D processors would be faster but not significantly. There is no time for this at moment.
It seems that Penryn (I know I keep banging on about this, but I'm seriously comtemplating replacing my Athlon XP 2800 with one of these when they are relased) will bring not only SSE4, but faster SSSE3. Perhaps someone who knows the source code of E@H can guess whether (once the current bugs have been worked out) it will be worth optimising for these instruction sets as well as for SSE1?
(I appreciate that there are far more SSE1 processors than SSE2, SSE3, SSSE3, and SSE4, but when we have this many WUs to crunch, every optimisation helps. And as time passes there will probably be more and more processors supporting these instructions.)
"Intel SSE4 consists of 54 instructions divided into two major categories: Vectorizing Compiler and Media Accelerators, and Efficient Accelerated String and Text Processing.
Vectorizing Compiler and Media Accelerators provide high performance compiler primitives, such as packed (using multiple operands at the same time) integer and floating point operations, that allow for performance optimized code generation. It also includes highly optimized media-related operations such as sum absolute difference, floating point dot products, and memory loads. ...
Penryn supports 47 of the Intel SSE4 instructions including the Vectorizing Compiler and Media Accelerator instructions. ...
Software will be able to programmatically detect which Intel SSE4 instructions are available on the processor.
...
Microarchitecture enhancements in Penryn include:
...
Super Shuffle Engine and Fast Radix-16 Divider: 3X faster shuffles (repositioning of bits, a common operation in image editing) and 1.6X - 2X faster divides. The Super Shuffle Engine will greatly improve the performance of Intel SSE4 and Supplemental Streaming SIMD Extensions 3 (SSSE3) instructions. ...
Enhanced Cache Line Split Load: Greatly improved performance on unaligned loads (those that span across cache boundaries) and optimized store and load operations. ..."
To go a bit off topic ;-) Malaria certainly is a worthwhile project, and it hasn't been too "beta" any more since a few months ago (before, there was trouble with scheduler probs and seemingly unmotivated computation errors) but as I said, you should have enough cache to run it properly. 512 MB is a bit too little, my Venice just sucked there. As I said, the Dothan does fine. No idea how much your Mac has.
You mean KB of L2 cache? The Mac mini's "Yonah" Core Duo has 2MB shared by both cores, and the individual cores are in fact quite similar to a Dothan. So this sounds like it will be OK for Malaria!
I used to have a MIPS R6000 with TTL logic CPU board, no microprocessor, UNIX System V and it had the same speed on Lapack as a VAX 9000. This while working for BULL at Area Science Park in Trieste, but I had to leave it there. Sorry.
Tullio
:-( ... thanks for the info, tho. So maybe Malaria on my Pentium M Banias (1MB L2 cache), E@H on Athlon XP Palomino (Linux), Dual Coppermine (Linux)and half of the Mac mini , CPDN on the other half, and a bit of Rosetta and S@H everywhere as fallback.
:-( ... thanks for the info, tho. So maybe Malaria on my Pentium M Banias (1MB L2 cache), E@H on Athlon XP Palomino (Linux), Dual Coppermine (Linux)and half of the Mac mini , CPDN on the other half, and a bit of Rosetta and S@H everywhere as fallback.
I believe Account Creation for Malaria is currently closed. So if you're not already signed-up, you're out of luck.
BIOS date is Apr 2001 and
)
BIOS date is Apr 2001 and there are updates into 2002. I'll see about upgrading the BIOS when I get home from work tonight.
Seti Classic Final Total: 11446 WU.
Hi! RE: Bikeman,
)
Hi!
It's my first Mac. Wanted something compact, quiet and reasonably fast. Oh, and is does look cooler to sync your iPod with :-).
Don't worry, E@H S5R2 (just to stay on topic here) will benefit from my Mac mini: The Athlon XP 1800+ which used to do CPDN exclusively 24/7 will then do some E@H, and E@H will get half a core on the new Mac. I'm thinking about joining "Malaria" as well, sounds like a worthwhile project to me.
Cheers
BRM
To go a bit off topic ;-)
)
To go a bit off topic ;-) Malaria certainly is a worthwhile project, and it hasn't been too "beta" any more since a few months ago (before, there was trouble with scheduler probs and seemingly unmotivated computation errors) but as I said, you should have enough cache to run it properly. 512 MB is a bit too little, my Venice just sucked there. As I said, the Dothan does fine. No idea how much your Mac has.
RE: So i don't think that
)
It seems that Penryn (I know I keep banging on about this, but I'm seriously comtemplating replacing my Athlon XP 2800 with one of these when they are relased) will bring not only SSE4, but faster SSSE3. Perhaps someone who knows the source code of E@H can guess whether (once the current bugs have been worked out) it will be worth optimising for these instruction sets as well as for SSE1?
(I appreciate that there are far more SSE1 processors than SSE2, SSE3, SSSE3, and SSE4, but when we have this many WUs to crunch, every optimisation helps. And as time passes there will probably be more and more processors supporting these instructions.)
From http://softwarecommunity.intel.com/articles/eng/1247.htm
"Intel SSE4 consists of 54 instructions divided into two major categories: Vectorizing Compiler and Media Accelerators, and Efficient Accelerated String and Text Processing.
Vectorizing Compiler and Media Accelerators provide high performance compiler primitives, such as packed (using multiple operands at the same time) integer and floating point operations, that allow for performance optimized code generation. It also includes highly optimized media-related operations such as sum absolute difference, floating point dot products, and memory loads. ...
Penryn supports 47 of the Intel SSE4 instructions including the Vectorizing Compiler and Media Accelerator instructions. ...
Software will be able to programmatically detect which Intel SSE4 instructions are available on the processor.
...
Microarchitecture enhancements in Penryn include:
...
Super Shuffle Engine and Fast Radix-16 Divider: 3X faster shuffles (repositioning of bits, a common operation in image editing) and 1.6X - 2X faster divides. The Super Shuffle Engine will greatly improve the performance of Intel SSE4 and Supplemental Streaming SIMD Extensions 3 (SSSE3) instructions. ...
Enhanced Cache Line Split Load: Greatly improved performance on unaligned loads (those that span across cache boundaries) and optimized store and load operations. ..."
RE: To go a bit off topic
)
You mean KB of L2 cache? The Mac mini's "Yonah" Core Duo has 2MB shared by both cores, and the individual cores are in fact quite similar to a Dothan. So this sounds like it will be OK for Malaria!
Cheers
BRM
Yes of course, I meant KB,
)
Yes of course, I meant KB, sorry. One gets so used to large numbers ;-)
I have a "Yonah" here, too, but never tried to run Malaria on it yet.
I used to have a MIPS R6000
)
I used to have a MIPS R6000 with TTL logic CPU board, no microprocessor, UNIX System V and it had the same speed on Lapack as a VAX 9000. This while working for BULL at Area Science Park in Trieste, but I had to leave it there. Sorry.
Tullio
No Malaria for Mac, so far.
)
No Malaria for Mac, so far.
RE: No Malaria for Mac, so
)
:-( ... thanks for the info, tho. So maybe Malaria on my Pentium M Banias (1MB L2 cache), E@H on Athlon XP Palomino (Linux), Dual Coppermine (Linux)and half of the Mac mini , CPDN on the other half, and a bit of Rosetta and S@H everywhere as fallback.
RE: RE: No Malaria for
)
I believe Account Creation for Malaria is currently closed. So if you're not already signed-up, you're out of luck.
Seti Classic Final Total: 11446 WU.