1. Apparatus for processing data under control of a sequence of program instructions stored in a memory, said apparatus comprising:
an instruction prefetch unit, coupled to said memory, for prefetching, from a sequence of prefetch addresses within said memory, a sequence of program instructions to be executed;
a static branch predictor for detecting an opcode of a branch instruction within said sequence of program instructions read from said memory and for performing a static prediction independent of previous processing activity as to whether said branch instruction will result in a jump to a branch target instruction stored at a branch target address within said memory, said instruction prefetch unit being responsive to a prediction by said static branch predictor that said branch instruction will result in said jump to alter said sequence of prefetch addresses to prefetch a sequence of program instructions starting from said branch target address;
a static branch prediction cache, triggered by a static prediction by said static branch predictor of a branch instruction that will result in a jump to a target branch address, for storing said static prediction as a branch instruction address of said branch instruction together with said branch target address, said static prediction stored by said static branch prediction cache being unaltered by whether subsequent execution of said branch instruction does result in said jump; and
an address comparitor for comparing a prefetch address of a program instruction being prefetched by said prefetch unit from said memory with said branch instruction address stored in said static branch prediction cache and upon a match to alter said sequence of prefetch addresses to prefetch a sequence of program instructions starting from said associated branch target address stored in said static branch prediction cache.
2. Apparatus as claimed in claim 1, wherein said static branch predictor predicts that a branch instruction will result in a jump if said branch target address would result in a backward jump in program execution flow.
3. Apparatus as claimed in claim 1, wherein said static branch predictor predicts that a branch instruction will not result in a jump if said branch target address would result in a forward jump in program execution flow.
4. Apparatus as claimed in claim 1, wherein when said address comparitor triggers an alteration in said sequence of prefetch addresses associated with a branch instruction, subsequent alteration of said sequence of prefetch addresses by said static branch predictor based upon detection of said opcode for said branch instruction is suppressed.
5. Apparatus as claimed in claim 1, wherein said static branch prediction cache is operable to store a single static branch prediction.
6. Apparatus as claimed in claim 5, wherein said static branch prediction cache is operable to store a most recent static branch prediction.
7. Apparatus as claimed in claim 1, wherein said static branch prediction cache is operable to store a valid flag indicative of whether said static branch prediction cache is storing a valid said static branch prediction.
8. Apparatus as claimed in claim 7, wherein said valid flag is written to indicate invalidity upon one or more of:
a change in memory address mapping; and
a context switch.
9. A method of processing data under control of a sequence of program instructions stored in a memory, said method comprising the steps of:
prefetching, from a sequence of prefetch addresses within said memory, a sequence of program instructions to be executed;
detecting an opcode of a branch instruction within said sequence of program instructions read from said memory and performing a static prediction, independent of previous processing activity, as to whether said branch instruction will result in a jump to a branch target instruction stored at a branch target address within said memory, said prefetching step being responsive to a prediction that said branch instruction will result in said jump to alter said sequence of prefetch addresses to prefetch a sequence of program instructions starting from said branch target address;
triggered by a static prediction of a branch instruction that will result in a jump to a target branch address, storing in a static branch prediction cache said static prediction as a branch instruction address of said branch instruction together with said branch target address, said static prediction stored by said static branch prediction cache being unaltered by whether subsequent execution of said branch instruction does result in said jump; and
comparing a prefetch address of a program instruction being prefetched from said memory with said branch instruction address stored in said static branch prediction cache and, upon a match, altering said sequence of prefetch addresses to prefetch a sequence of program instructions starting from said associated branch target address stored in said static branch prediction cache.
10. A method as claimed in claim 9, wherein said step of performing a static prediction predicts that a branch instruction will result in a jump if said branch target address would result in a backward jump in program execution flow.
11. A method as claimed in claim 9, wherein said step of performing a static prediction predicts that a branch instruction will not result in a jump if said branch target address would result in a forward jump in program execution flow.
12. A method as claimed in claim 9, wherein said step of comparing triggers an alteration in said sequence of prefetch addresses associated with a branch instruction, subsequent alteration of said sequence of prefetch addresses based upon detection of said opcode for said branch instruction is suppressed.
13. A method as claimed in claim 9, wherein said static branch prediction cache is operable to store a single static branch prediction.
14. A method as claimed in claim 13, wherein said static branch prediction cache is operable to store a most recent static branch prediction.
15. A method as claimed in claim 9, wherein said static branch prediction cache is operable to store a valid flag indicative of whether said static branch prediction cache is storing a valid said static branch prediction.
16. A method as claimed in claim 15, wherein said valid flag is written to indicate invalidity upon one or more of:
a change in memory address mapping; and
a context switch.
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 zoom lens comprising a plurality of lens groups which are disposed in order from an object along an optical axis, wherein
among the plurality of lens groups, a first lens group which is disposed closest to the object has a positive refractive power,
the first lens group comprises an optical path refraction element for refracting an optical path, and a plurality of lens components which are disposed closer to the object than the optical path refraction element,
the plurality of lens components include at least one positive lens that satisfies a condition of
\u03bdd>50,
where \u03bdd is an Abbe number with respect to the d-line, and
at least one of the lens groups, which are disposed closer to the image than the first lens group, at least partially includes a shift lens group that can move in a direction substantially perpendicular to the optical axis.
2. The zoom lens according to claim 1, wherein a condition of
0.7<\u03b2bw\xd7(1\u2212\u03b2aw)<1.4
is satisfied, where \u03b2aw is a lateral magnification of the shift lens group in a wide-angle end state, and \u03b2bw is a lateral magnification of the zoom lens that is disposed between the shift lens group and the image in the wide-angle end state.
3. The zoom lens according to claim 1, wherein the plurality of lens components are formed of two lenses.
4. The zoom lens according to claim 1, wherein the plurality of lens components are formed of a negative lens and a positive lens which are disposed in order from the object.
5. The zoom lens according to claim 1, wherein the plurality of lens components comprise a negative meniscus lens having a convex surface facing the object, and a positive meniscus lens having a convex surface facing the object, which are disposed in order from the object.
6. The zoom lens according to claim 1, wherein the shift lens group is fixed to a position on the optical axis during zooming from a wide-angle end state to a telephoto end state.
7. The zoom lens according to claim 1, wherein the shift lens group comprises a plurality of lens components.
8. The zoom lens according to claim 1, wherein the shift lens group comprises a cemented lens.
9. The zoom lens according to claim 1, wherein a condition of
0.3<fw|fs|<0.5
is satisfied, where fw is a focal length of the zoom lens in the wide-angle end state, and fs is a focal length of the shift lens group.
10. The zoom lens according to claim 1, wherein the angle of view in the wide-angle end state is 75 degrees or more.
11. The zoom lens according to claim 1, wherein an aperture stop is disposed near the shift lens group.
12. The zoom lens according to claim 1, wherein an aperture stop is disposed near the object side of the lens group that includes the shift lens group.
13. The zoom lens according to claim 1, wherein a condition of
\u03bdd1<50
is satisfied, where \u03bdd1 is an Abbe number, with respect to the d-line, of a negative lens closest to the object among the plurality of lens components in the first lens group.
14. The zoom lens according to claim 1, wherein a condition of
\u03bdd2>50
is satisfied, where \u03bdd2 is an Abbe number, with respect to the d-line, of a positive lens closest to the image among the plurality of lens components in the first lens group.
15. The zoom lens according to claim 1, wherein a condition of
nd1>1.75
is satisfied, where nd1 is a refractive index, with respect to the d-line, of a negative lens closest to the object among the plurality of lens components in the first lens group.
16. The zoom lens according to claim 1, wherein a condition of
nd2<1.65
is satisfied, where nd2 is a refractive index, with respect to the d-line, of a positive lens closest to the image among the plurality of lens components in the first lens group.
17. The zoom lens according to claim 1, wherein the second lens group, which is disposed to the image side of the first lens group, has a negative refractive index, and a condition of
0.4<(\u2212f2)fw<1.9
is satisfied, where fw is a focal length of the zoom lens in the wide-angle end state, and f2 is a focal length of the second lens group.
18. The zoom lens according to claim 1, wherein a condition of
2.4<f1(\u2212f2)<3.7
is satisfied, where f1 is a focal length of the first lens group, and f2 is a focal length of the second lens group which is disposed to the image side of the first lens group.
19. The zoom lens according to claim 1, wherein
the plurality of lens groups comprise at least the first lens group, a second lens group and a third lens group, which are disposed in order from the object along an optical axis and
the third lens group is a lens group that includes the shift lens group.
20. The zoom lens according to claim 1, wherein
the plurality of lens groups comprise the first lens group, a second lens group, a third lens group, a fourth lens group and a fifth lens group, which are disposed in order from the object along an optical axis, and
the third lens group is a lens group that includes the shift lens group.
21. The zoom lens according to claim 20, wherein
the second lens group has a negative refractive power, the third lens group has a positive refractive power, the fourth lens group has a positive refractive power, and the fifth lens group has a negative refractive power.
22. The zoom lens according to claim 1, wherein the optical path refraction element is a prism, and satisfies a condition of
L1Lp<1.0,
where L1 is a distance on the optical axis from the surface closest to the object in the first lens group to the prism, and Lp is a distance of the prism on the optical axis.
23. An optical apparatus comprising the zoom lens according to claim 1.
24. A method for forming an image of an object that uses a zoom lens which has a plurality of lens groups which are disposed in order from the object along an optical axis, wherein
a first lens group, which is disposed closest to the object among the plurality of lens groups, has a positive refractive power,
the first lens group comprises an optical path refraction element for refracting an optical path and a plurality of lens components which are disposed closer to the object than the optical path refraction element,
the plurality of lens components include at least one positive lens that satisfies a condition of
\u03bdd>50,
where \u03bdd is an Abbe number with respect to the d-line, and
at least one of the lens groups, which are disposed closer to the image than the first lens group, at least partially includes a shift lens group that can move in a direction substantially perpendicular to the optical axis.
25. The method for forming an image of an object according to claim 24, wherein a condition of
0.7<\u03b2bw\xd7(1\u2212\u03b2aw)<1.4
is satisfied, where \u03b2aw is a lateral magnification of the shift lens group in a wide-angle end state, and \u03b2bw is a lateral magnification of the zoom lens that is disposed between the shift lens group and the image in the wide-angle end state.
26. The method for forming an image of an object according to claim 24, wherein the plurality of lens components are formed of two lenses.
27. The method for forming an image of an object according to claim 24, wherein the plurality of lens components are formed of a negative lens and a positive lens which are disposed in order from the object.
28. The method for forming an image of an object according to claim 24, wherein the plurality of lens components comprise a negative meniscus lens having a convex surface facing the object, and a positive meniscus lens having a convex surface facing the object, which are disposed in order from the object.
29. The method for forming an image of an object according to claim 24, wherein the shift lens group is fixed to a position on the optical axis during zooming from a wide-angle end state to a telephoto end state.
30. The method for forming an image of an object according to claim 24, wherein the shift lens group comprises a plurality of lens components.
31. The method for forming an image of an object according to claim 24, wherein the shift lens group comprises a cemented lens.
32. The method for forming an image of an object according to claim 24, wherein a condition of
0.3<fw|fs|<0.5
is satisfied, where fw is a focal length of the zoom lens in the wide-angle end state, and fs is a focal length of the shift lens group.
33. The method for forming an image of an object according to claim 24, wherein the angle of view in the wide-angle end state is 75 degrees or more.
34. The method for forming an image of an object according to claim 24, wherein an aperture stop is disposed near the shift lens group.
35. The method for forming an image of an object according to claim 24, wherein an aperture stop is disposed near the object side of the lens group that includes the shift lens group.
36. The method for forming an image of an object according to claim 24, wherein a condition of
\u03bdd1<50
is satisfied, where \u03bdd1 is an Abbe number, with respect to the d-line, of a negative lens closest to the object among the plurality of lens components in the first lens group.
37. The method for forming an image of an object according to claim 24, wherein a condition of
\u03bdd2>50
is satisfied, where \u03bdd2 is an Abbe number, with respect to the d-line, of a positive lens closest to the image among the plurality of lens components in the first lens group.
38. The method for forming an image of an object according to claim 24, wherein a condition of
nd1>1.75
is satisfied, where nd1 is a refractive index, with respect to the d-line, of a negative lens closest to the object among the plurality of lens components in the first lens group.
39. The method for forming an image of an object according to claim 24, wherein a condition of
nd2<1.65
is satisfied, where nd2 is a refractive index, with respect to the d-line, of a positive lens closest to the image among the plurality of lens components in the first lens group.
40. The method for forming an image of an object according to claim 24, wherein the second lens group, which is disposed to the image side of the first lens group, has a negative refractive index, and a condition of
0.4<(\u2212f2)fw<1.9
is satisfied, where fw is a focal length of the zoom lens in the wide-angle end state, and f2 is a focal length of the second lens group.
41. The method for forming an image of an object according to claim 24, wherein a condition of
2.4<f1(\u2212f2)<3.7
is satisfied, where f1 is a focal length of the first lens group, and f2 is a focal length of the second lens group which is disposed to the image side of the first lens group.
42. The method for forming an image of an object according to claim 24, wherein
the plurality of lens groups comprise at least the first lens group, a second lens group and a third lens group, which are disposed in order from the object along an optical axis and
the third lens group is a lens group that includes the shift lens group.
43. The method for forming an image of an object according to claim 24, wherein
the plurality of lens groups comprise the first lens group, a second lens group, a third lens group, a fourth lens group and a fifth lens group, which are disposed in order from the object along an optical axis and
the third lens group is a lens group that includes the shift lens group.
44. The method for forming an image of an object according to claim 43, wherein the second lens group has a negative refractive power, the third lens group has a positive refractive power, the fourth lens group has a positive refractive power, and the fifth lens group has a negative refractive power.