Since Excel is based on Double Precision data cells, we can say that that this would be a very poor choice for a traditional business workstation. Maybe ok for Word and Powerpoint.
"the ESC1000 delivers 2.8 teraflops at single precision and 234 gigaflops at double precision"
You *could* get 1.1 teraflops by adding those two together and dividing by two...
I mean an average for that sort of thing would need to be skewed towards whichever it did more often, but a straightforward mean would give you the figure required.
It is extremely difficult to get 100% performance out of these devices. The Teraflop ratings they are given are for an ideal situation where the alu's are fully occupied across all threads and the max mem bandwith is being achieved.
The class of problems which can get you to this level of performance is rather small...
However for their available processing power they use a substantially lower amount of power than a cluster of cpu's, but are admittedly harder to program efficiently.
Gateway DX4831-01e Mini-Tower Desktop (2.93GHz Intel Core i3 530, 6GB DDR3, 1TB HDD, DVD±RW DL, Windows 7 Home Premium)
Dell Inspiron 560 Mini-Tower Desktop (2.93GHz Intel Core 2 Duo E7500, 6GB DDR3, 640GB HDD, DVD±RW, Windows 7 Home Premium)
HP (Hewlett-Packard) Compaq Presario CQ5320F Mini-Tower Desktop (2.8GHz Athlon II X2 240, 3GB DDR3, 500GB HDD, DVD±RW DL, Windows 7 Home Premium)
Apple iMac All-In-One Desktop (3.06GHz Intel Core 2 Duo, 4GB DDR2, 500GB, DVD+-RW DL, Mac OS X v10.5 Snow Leopard, 21.5" LCD)
Gateway DX4300-11 Mini-Tower Desktop (2.5GHz Athlon II X4 805, 8GB DDR2, 1TB HDD, DVD±RW DL, Windows 7 Home Premium)
Goat Jam
Timing FAIL #
Posted Wednesday 28th October 2009 00:04 GMT
Somebody should tell Asus that Vista has been superceded!
Gene Cash
Thank you, El Reg #
Posted Wednesday 28th October 2009 10:11 GMT
For calling "bullshit" on the performance numbers. Everyone else is trumpeting it as something Los Alamos would want for nuclear sims.
david 12
No good as a workstation #
Posted Wednesday 28th October 2009 10:11 GMT
Since Excel is based on Double Precision data cells, we can say that that this would be a very poor choice for a traditional business workstation. Maybe ok for Word and Powerpoint.
John Smith 19
When 234Gigflops inn't a supercomputer #
Posted Wednesday 28th October 2009 10:11 GMT
We've come a very long way from the Cray 1.
Alastair McFarlane
1.1 teraflops #
Posted Wednesday 28th October 2009 10:11 GMT
"the ESC1000 delivers 2.8 teraflops at single precision and 234 gigaflops at double precision"
You *could* get 1.1 teraflops by adding those two together and dividing by two...
I mean an average for that sort of thing would need to be skewed towards whichever it did more often, but a straightforward mean would give you the figure required.
b4k4
errr. #
Posted Wednesday 28th October 2009 11:19 GMT
You *could* get 1.1 teraflops by adding those two together and dividing by two...
by could, do you mean in another universe where 2 x 1.1 doesn't make 2.2?
Anonymous Coward
Performance #
Posted Wednesday 28th October 2009 13:27 GMT
It is extremely difficult to get 100% performance out of these devices. The Teraflop ratings they are given are for an ideal situation where the alu's are fully occupied across all threads and the max mem bandwith is being achieved.
The class of problems which can get you to this level of performance is rather small...
However for their available processing power they use a substantially lower amount of power than a cluster of cpu's, but are admittedly harder to program efficiently.
This forum is now closed for new posts.