1. A cooled cryostat device comprising:
a cold finger;
an infrared detector means secured to a terminus of said cold finger;
an outer casing surrounding said cold finger;
a stiffener means for stiffening said cold finger, said stiffener means interposed between said outer casing and said cold finger in the vicinity of said terminus,
wherein said stiffener means comprises:
first eyelets distributed around the periphery of a first ring, wherein said first ring is attachable by a first securing means to the inside of said outer casing;
second eyelets distributed around the periphery of a second ring, wherein said second ring is attachable by a second securing means to the outside of said cold finger;
said first eyelets and said second eyelets being mutually angularly offset; and
a cable, passing freely through said first and second eyelets, said cable, once said cold finger has been placed in a balanced situation, being secured under tension, to provide radial support, as a closed loop to said first ring and said second ring.
2. A device according to claim 1, wherein one or both of said first ring and said second ring is made of metal and wherein one or both of said first or second securing means is by weld or solder.
3. A device according to claim 1, wherein one or both of said first or second securing means is bonding by adhesive.
4. A device according to any one of claims 1 to 3, wherein said cable is made of metal.
5. A device according to any one of claims 1 to 3, wherein said cable is made of synthetic material.
6. A device according to claim 1 wherein said cable passes through said first eyelets and said second eyelets in alternation.
The claims below are in addition to those above.
All refrences to claim(s) which appear below refer to the numbering after this setence.
1. A processor comprising:
a cache;
a decoder;
a register file operable to store a first packed data and a second packed data respectively including a first plurality of data elements and a second plurality of data elements, wherein each data element in the first plurality of data elements corresponds to a different data element in the second plurality of data elements, in a respective position; and
at least one functional unit coupled to the cache, the decoder, and the register file, and operable to execute instructions specifying data manipulation operations, including:
a first move instruction that, when executed by the at least one functional unit, causes data to be transferred between a first packed data register and a second packed data register;
a second move instruction that, when executed by the at least one functional unit, causes data to be transferred between the first packed data register and a main memory;
a third instruction of an instruction set of the processor that, when executed by the at least one functional unit, is to cause half of each data element in the first and second pluralities of data elements to be copied into the register file as a third plurality of separate data elements in a third packed data; and
a fourth instruction that is to specify a third packed data register and a fourth packed data register, and that, when executed by the at least one functional unit, is to cause all data elements of the third packed data register to be packed with saturation into a lowest order half of a result, wherein at least one data element of the third packed data register is clamped to a value in the result.
2. The processor of claim 1 wherein the at least one functional unit, is to cause a lowest half of each of the elements of the first plurality of elements and a lowest half of each of the elements of the second plurality of elements to be copied into the register file as the third plurality of separate data elements in the third packed data.
3. The processor of claim 2 wherein the at least one functional unit is to cause a highest half of each of the elements of the first plurality of elements and a highest half of each of the elements of the second plurality of elements to be ignored.
4. The processor of claim 1, wherein the second instruction, when executed by the at least one functional unit, is to clamp the at least one data element to the value which is one of a maximum and a minimum value.
5. The processor of claim 1, wherein the first instruction has a size field to indicate a size of the data elements of the first plurality of data elements.
6. The processor of claim 1, wherein the at least one functional unit is also operable to execute an instruction that, when executed by the at least one functional unit, is to cause a packed data result to be stored that is to include only half of a plurality of data elements of a fourth packed data interleaved with only half of a plurality of data elements of a fifth packed data which are also in corresponding data element positions.
7. A processing device comprising:
a graphics rendering device;
a cache;
a register file operable to store a first packed data that is to include a first plurality of data elements and a second packed data that is to include a second plurality of data elements;
a decoder to decode a first instruction having an opcode, and a second instruction that is to specify a first packed data register in the register file and a second packed data register in the register file; and
at least one functional unit coupled to the decoder, the register file, and the cache, the at least one functional unit, in response to the decoder decoding the first instruction, to cause half of each data element in the first and second pluralities of data elements to be copied into the register file as a third plurality of separate data elements in a third packed data and ignore a second half of each data element in the first and second pluralities of data elements, and
the at least one functional unit, in response to the decoder decoding the second instruction to cause all data elements of the first packed data register to be packed with saturation in a lowest order half of a result, wherein at least one data element from the first packed data register is to be clamped to a value in the result.
8. The processing device of claim 7 wherein the at least one functional unit is to cause a lowest half of each of the elements of the first plurality of elements and a lowest half of each of the elements of the second plurality of elements to be copied into the register file as the third plurality of separate data elements in the third packed data.
9. The processing device of claim 8 wherein the at least one functional unit is to cause a highest half of each of the elements of the first plurality of elements and a highest half of each of the elements of the second plurality of elements to be ignored.
10. The processing device of claim 7, wherein the at least one functional unit, in response to the decoder decoding the second instruction, is to clamp the at least one data element to one of a maximum and a minimum value.
11. The processing device of claim 7, wherein the first instruction has a size field to indicate a size of the data elements of the first packed data.
12. The processing device of claim 7, wherein the decoder is further to decode a third instruction, and wherein the at least one functional unit, in response to the decoder decoding the third instruction, is to cause a packed data result to be stored that is to include only half of a plurality of data elements of a fourth packed data interleaved with half of a plurality of data elements of a fifth packed data in corresponding data element positions.
13. A method comprising:
accessing a first packed data including a first plurality of data elements and a second packed data including a second plurality of data elements in a register file;
decoding a first instruction with a decoder;
in response to the decoding of the first instruction, causing half of each data element in the first and second pluralities of data elements to be copied into the register file as a third plurality of separate data elements in a third packed data and ignoring a second half of each data element in the first and second pluralities of data elements;
decoding a second instruction with the decoder, the second instruction specifying a first packed data register and a second packed data register;
in response to the decoding of the second instruction, causing all data elements of the first packed data register to be packed with saturation in a lowest order half of a result including clamping at least one data element from the first packed data register to a value.
14. The method of claim 13 wherein in response to the decoding of the first instruction causing comprises causing a lowest half of each of the elements of the first plurality of elements and a lowest half of each of the elements of the second plurality of elements to be copied into the register file as the third plurality of separate data elements in the third packed data.
15. The method of claim 14 wherein in response to the decoding of the first instruction ignoring comprises ignoring a highest half of each of the elements of the first plurality of elements and a highest half of each of the elements of the second plurality of elements.
16. A computer readable storage device including code, that when executed, is to cause a system to perform operations comprising:
access a first packed data including a first plurality of data elements and a second packed data including a second plurality of data elements in a register file;
decode a first instruction with a decoder; and
in response to decoding the first instruction, cause half of each data element in the first and second pluralities of data elements to be copied into the register file as a third plurality of separate data elements in a third packed data and ignore a second half of each data element in the first and second pluralities of data elements;
decoding a second instruction with the decoder, wherein the second instruction specifies a first packed data register and a second packed data register;
in response to the decoding of the second instruction, causing all data elements from the first packed data register to be packed with saturation in a lowest order half of a result including clamping at least one data element from the first packed data register to a value in the result.
17. The computer readable storage device of claim 16 wherein to cause the half of each data element in the first and second pluralities of data elements to be copied into the register file as the third plurality of separate data elements in the third packed data comprises to cause a lowest half of each of the elements of the first plurality of elements and a lowest half of each of the elements of the second plurality of elements to be copied into the register file as the third plurality of separate data elements in the third packed data.
18. The computer readable storage device of claim 17 wherein to ignore the second half of each data element in the first and second pluralities of data elements comprises to ignore a highest half of each of the elements of the first plurality of elements and a highest half of each of the elements of the second plurality of elements.