This is an appeal from the decision of the Patent and Trademark Office Board of Appeals affirming the rejection of claims 1-6 and 9-16, constituting all of the claims in appellant’s application serial No. 243,951, filed April 14, 1972, entitled “Method of Operating a Multiprogrammed Computing System.” We reverse.
The Invention
Chatfield’s invention resides in the field of computer technology and specifically in the area of multiprogrammed computer systems. Computer systems generally include one or more central processing units (CPU’s)1 and one or more peripheral units, e. g., input-output (I/O) devices, card readers, etc. In multiprogram systems, the CPU and the peripheral resource equipment may process more than one program concurrently. A problem arises when more than one program requires the use of the same piece of equipment at the same time.
Three earlier solutions of the problem had been employed. One physically builds priority levels into the CPU, so that one particular program will have priority over another particular program. A second solution assigns each program a priority. A disadvantage encountered in both is that *154once the program begins to run, the relative priority is fixed. The third solution specifically writes the programs to take advantage of the priority-assignment capabilities of the particular system used. The disadvantage there encountered is that specific programs must be written for each particular system. If the system is changed, the programs must be changed.
Chatfield’s novel solution dynamically evaluates and reassigns program priorities as the programs execute. No prior art was cited against the appealed claims and, accordingly, Chatfield’s invention must be considered to have been new and unobvious.
The method embraced by the appealed independent claims involves operating a computing machine system for an interval, and accumulating resource use data2 for that operating interval. Periodically, the processing operation is halted and the use data are analyzed. That analysis determines the priorities, and consequent resource access, for the next period of operation. The processing operation is resumed for another period of time, use data accumulated, operation halted, data analyzed, priorities and access determined and assigned, and so on, repetitively. The net effect is an overall gain in operating efficiency, i. e., an increased “throughput.”
The appealed dependent claims describe several variant mathematical algorithms useable in analyzing the resource use data and regulating resource access priority.
Independent claim 1, and dependent claims 2 and 3, are representative and are reproduced below:
1. A method of operating a computing system upon more than one processing program concurrently for improving total resource utilization, said computing system comprising at least one central processing unit, having a logic and main memory function and an interrupt capability, and a plurality of peripheral resources capable of functioning in parallel with the central processing unit, comprising steps for:
(1) accumulating, system utilization data for at least one processing program for at least one resource, said system utilization data comprising resource activity and/or resource degradation data;
(2) (a) at spaced intervals interrupting the processing programs and analyzing the system utilization of at least one processing program;
(2) (b) based on this analysis regulating resource access by assigning an individual resource access priority and/or preventing resource access altogether in an unlike manner to at least two resources for at least one processing program to increase thruput;
(3) resuming the operation of the computing systems on the processing programs; and,
(4) continually repeating steps (1) to (3).
2. A method according to Claim 1 in which the regulation in step (2)(b) comprises regulating resource access substantially to favor the more overlapped programs, said overlapped programs being those that can use two or more resources in parallel.
3. A method according to Claim 2 in which processing program resource access regulated in step (2) is determined according to the following algorithm:
if CPU activity ^-=> the Maximum Peripheral activity n
*>“ PERIPHERAL ACTIVITY,
CPU PRIORITY = 1 ~ 1_
CPU ACTIVITY
where n is equal to the number of peripherals measured, where PERIPHERAL ACTIVITY, is the activity of the Ith peripheral resource, and where C.P.U. ACTIVITY is the activity of the C.P.U.
or if CPU activity the Maximum Peripheral activity
CPU PRIORITY = n + MAXIMUM PERIPHERAL ACTIVITY TOTAL RESOURCE ACTIVITY
where n is equal to the number of peripherals measured, where MAXIMUM PERIPHERAL ACTIVITY is the total activity on the most active peripheral resource, and where TOTAL RESOURCE ACTIVITY Is the sum of all C.P.U. and all peripheral resource activities, and all PERIPHERAL RESOURCE ACCESS PRIORITIES are assigned inversely to the CPU priorities.
*155 The Board Opinion
The board took the position that the claims are drawn to non-statutory subject matter under the Supreme Court’s decision in Gottschalk v. Benson, 409 U.S. 63, 93 S.Ct. 253, 34 L.Ed.2d 273, 175 USPQ 673 (1972). In sustaining the examiner’s rejection under 35 U.S.C. § 101, the board briefly reviewed our opinions in In re Musgrave, 57 CCPA 1352, 431 F.2d 882, 167 USPQ 280 (1970), and In re Benson, 58 CCPA 1134, 441 F.2d 682, 169 USPQ 548 (1971), on the question of non-statutory subject matter and concluded that the Supreme Court had “cast aside” the reasoning in those opinions. Although the examiner had referred to our opinion in In re Christensen, 478 F.2d 1392, 178 USPQ 35 (CCPA 1973), in which this court affirmed the rejection in view of Benson, the board did not consider Christensen pertinent.
Arguments
Chatfield urges that the Supreme Court in Benson foreclosed only the patentability of the particular program then before the court, not patentability of computer programs in general. Chatfield then argues that, unlike those in Benson, the instant claims (1) are not drawn solely to a mathematical formula or to an algorithm for solving that formula; and (2) are specifically drawn to a particular end use. His claims, Chatfield contends, are drawn to a new use of a known machine, i. e., to a patentable process as defined in 35 U.S.C. § 100(b).
The solicitor argues that the Supreme Court’s reference in Benson to “these programs” must be construed as prohibiting the patentability of a “program” such as that presently claimed. The solicitor further urges that because the present claims include a mathematical formulation, and an algorithm for solving it, they are similar to those held to have been drawn to non-statutory subject matter in Benson. The solicitor contends that the antecedent steps of the claims merely refer to gathering information preparatory to processing the algorithm and that the “end use” limitations depend upon the existence of other specific processing programs requiring priority assignment. In this respect, the solicitor’s basic contention is that Chatfield’s claims are no more limited than were the claims in Benson.
OPINION
Issue
Unlike the apparatus claims in In re Noll, 545 F.2d 141 (Cust. & Pat.App. 1976), decided this date, the appealed claims are drawn to a “method of operating a computing system.” Although Chatfield’s method of operating a computing machine system adjusts priorities of computer processing programs, no processing program, by itself, is claimed.
The claimed method operates to make the computing machine system more efficient in executing a plurality of processing programs. Any algorithm developed to be used in the claimed method would only guide the flow of the processing programs through the computing machine system and would not provide the informational input of the processing programs. Thus the issue is not whether a computer processing program, claimed in method terms, is statutory subject matter under 35 U.S.C. § 101.
Because nothing in the statute or in prior judicial opinions either authorizes or precludes patent protection of “computer programs,” the mere labeling of an invention as “a computer program” does not aid in decision making. Of course we follow the Supreme Court in our decisional process. In so doing, however, we take care not to unduly extend that Court’s pronouncements beyond their expressed limits. The Supreme Court having expressly refused to extend its Benson holding to computer programs generally, 409 U.S. at 71, 93 S.Ct. 253, 175 USPQ at 676, we find no warrant for our doing so. In view of the Supreme Court’s expression in Benson that the involved mathematical formula had “no substantial practical application except in connection with a digital computer,” 409 U.S. *156at 71, 93 S.Ct. at 257, 175 USPQ at 676, the fundamental rationale we glean from Benson is that a patent containing Benson’s claims would have preempted all practical use of both the underlying mathematical formula and the involved algorithm.3
The solicitor suggests that the Supreme Court, in its reference in Benson to the question of whether the patent laws should be extended to cover “these programs” [409 U.S. at 72, 93 S.Ct. 253, 175 USPQ at 676] and to the President’s Commission on the Patent System, intended to include all computer programs within the phrase “these programs.”4 We do not agree. “It is a maxim, not to be disregarded, that general expressions, in every opinion, are to be taken in connection with the case in which these expressions are used.” Cohens v. Virginia, 19 U.S. (6 Wheat.) 264, 398, 5 L.Ed.
257 (1821).5 If, however, doubt should remain with respect to the solicitor’s suggestion, we think it fully removed by the Supreme Court’s unanimous characterization of its holding in Benson as “limited” in Dann v. Johnston, 425 U.S. 219, 224, 96 S.Ct. 1393, 47 L.Ed.2d 692, 189 USPQ 257, 259 (1976).
The unique technological aspects of digital computers and the changes wrought in their operation by associated software programs, coupled with the limited number of precedential court decisions, illustrate the undesirability of deciding the issue of patentability by mere use of the label “computer program.” We join the Supreme Court’s suggestion in Benson that those who seek to preclude the patenting of all software or “computer program” inventions submit an appropriate proposal to the Congress. In *157the meantime, we must and will decide such cases on the law as we find it. In the present case we find no basis for treating methods of operating computing machine systems differently from methods of operating any other form of machine system.
The issue before us is thus whether these particular claims define statutory subject matter.6
Analysis
The statute, 35 U.S.C. § 101, provides that new and useful processes and improvements thereof are patentable subject matter. A “process” is defined in 35 U.S.C. § 100(b) as a “process, art or method, and includes a new use of a known process, machine, manufacture, composition of matter, or material.” Some inventions, however meritorious, do not constitute patentable subject matter, e. g., printed matter, In re Miller, 418 F.2d 1392, 57 CCPA 809, 164 USPQ 46 (1969); methods of doing business, In re Wait, 73 F.2d 982, 22 CCPA 822, 24 USPQ 88 (1934); purely mental steps, In re Prater, 415 F.2d 1393, 56 CCPA 1381, 162 USPQ 541 (1969), In re Musgrave, 431 F.2d 882, 57 CCPA 1352, 167 USPQ 280 (1970); naturally occurring phenomena or laws of nature, O’Reilly v. Morse, 56 U.S. (15 How.) 62, 14 L.Ed. 601 (1853); a mathematical formula and the algorithm therefor, Gottschalk v. Benson, 409 U.S. 63, 93 S.Ct. 253, 34 L.Ed.2d 273, 175 USPQ 673 (1972).
Because Chatfield’s claim 1 defines the inventions as a “method of operating a computing system upon more than one processing program * * * comprising,” the claim prima facie is directed to a “method” and the wording of the claim would bring it within the statutorily defined “process” category. However, a claim to a “method” or “process” may fall within the literal terms of the statute and yet not define proper subject matter for patent protection. Benson and Christensen, supra. The question is whether the claimed method falls within either of two categories judicially determined to be non-statutory, i. e., claims drawn to mathematical problemsolving algorithms or to purely mental steps. We hold that the present claims fall within neither.7
Chatfield’s inventive contribution, as set forth in the appealed claims and described above, is a unique method for improving the operating efficiency of a system of electro-mechanical machines, i. e., a computing system. No formula or algorithm appears in Chatfield’s independent claims. The dependent claims set forth a number of different algorithms which may be used in carrying out intermediate steps in the overall process of operating the computer system. We faced a similar situation in In re Bernhart, 417 F.2d 1395, 57 CCPA 737, 163 USPQ 611 (1969). Bernhart’s apparatus claims included a mathematical equation. The examiner contended that the point of patentability of the invention lay in the equation itself, the remainder of the subject matter in the claims being admittedly old. Because the point of patentability was non-statutory, the examiner argued, the claims failed to define statutory subject matter. The examiner referred to “mental steps,” and the board referred to “printed matter” as the non-statutory subject matter. We found neither relevant. We did, however, consider the principle extracted by the examiner and the board from earlier “mental step” and “printed matter” cases to be as *158follows: “If, in an invention defined by a claim, the novelty is indicated by an expression which does not itself fit in a statutory class (in this case not a machine or a part thereof), then the whole invention is non-statutory since all else in the claim is old.” (417 F.2d at 1399, 57 CCPA at 743, 163 USPQ at 615). In rejecting that view as incorrect, we said:
We think it is clear that in enacting section 101 Congress meant to exclude principles or laws of nature and mathematics, of which equations are an example, from even temporary monopolization by patent. Accordingly, no rule of law should be announced which would impress a monopoly upon all uses of the equations disclosed by appellants here in their patent application. To allow the claims in issue here would not prohibit all uses of those equations. As we have pointed out above, a member of the public would have to do much more than use the equations to infringe any of these claims. He would have to use them in the physical equipment recited in the claim. Moreover, all machines function according to laws of physics which can be mathematically set forth if known. We cannot deny patents on machines merely because their novelty may be explained in terms of such laws if we are to obey the mandate of Congress that a machine is subject matter for a patent. We should not penalize the inventor who makes his invention by discovering new and unobvious mathematical relationships which he then utilizes in a machine, as against the inventor who makes the same machine by trial and error and does not disclose the laws by which it operates. The mandate of Congress in 35 U.S.C. 103 is that “patentability shall not be negatived by the manner in which the invention was made.” For the foregoing reasons, we conclude that under the statute the apparatus herein claimed constitutes statutory subject matter. [417 F.2d at 1399-1400, 57 CCPA at 743-44, 163 USPQ at 616].
We have thus specifically rejected the broad notion that if a portion of a claim be non-statutory the whole claim is ipso facto non-statutory. The requirement is that the invention set forth in a claim be construed as a whole.
In In re Christensen, 478 F.2d 1392, 178 USPQ 35 (CCPA 1973), we were called upon to decide whether method claims reciting a mathematical equation defined statutory subject matter. The claims in Christensen defined the invention as a “method of determining the porosity of a subsurface formation in situ” by solving a particular equation. The invention comprised gathering data and using it to solve a particular equation. In affirming the rejection of the claims, we followed the tenet of Benson that solving an equation constituted non-statutory subject matter. We determined that the several steps antecedent to the solving of the equation in Christensen’s claims merely established the variables for the equation and did not suffice to render the claimed method patentable as a whole. Our reference in Christensen to the mathematical equation as being “at the point of novelty” does not equate to a holding that a claim may be dissected, the claim components searched in the prior art, and, if the only component found novel is outside the statutory classes of invention, the claim may be rejected under 35 U.S.C. § 101. That procedure is neither correct nor within the intent of Congress, for the reasons we stated in Bernhart.
As above indicated, Chatfield’s independent claims contain neither a mathematical formula nor a mathematical algorithm. Mathematical algorithms appear only in the dependent claims and do not themselves constitute the method per se. The dependent claims are drawn to several different mathematical formulae and algorithms and, conceivably, numerous other algorithms could be developed and used in the novel method. The dependent claims are not, of course, claims to the mathematical algorithms appearing therein but are claims to the method of the independent claims in which the analysis steps are carried out by use of the specified algorithm. A patent issuing on Chatfield’s claimed method *159would thus preempt neither the mathematical formulae nor the algorithms specified in such a patent, unless used in the performance of the entire claimed method. Such a patent would thus not be “in practical effect * * * a patent on the algorithm itself.” See note 4, supra.
The appealed claims, analyzed as a whole, simply define a novel method for operating a particular machine system in a particular mode. The method is claimed in a series of discrete method steps of the type one might expect to find governing the operation of a system of non-computing machines, e. g., a milling machine system. The claimed process does not end with solution of a particular equation, as in Christensen. Nor are the appealed claims so “abstract and sweeping” as those in Benson, being limited to the particular operation of a computing machine system as specified in the claims. Thus, the claimed method differs materially and significantly from the invention in either Christensen or Benson. We therefore are not persuaded to discount Chatfield’s novel contribution to the useful arts as outside the field of endeavor intended to be encouraged by our patent laws.
The decision of the board is reversed.
REVERSED.
. The central processing unit or units perform the main logic and memory functions for the system.
. The resource use data include resource activity data and data respecting resource degradation, i. e., the inefficiency resulting from more than one program competing for the same resource.
. Though the Supreme Court stated in Benson that the process there involved could be performed without a computer or other apparatus, 409 U.S. at 67, 68, 93 S.Ct. 253, 175 USPQ at 675, it recognized that conversion to pure binary numerals was the only practical application of the formula and algorithm, and that binary numerals have their only practical application in connection with a digital computer, and thus that all practical uses of the formula and algorithm were in effect preempted by the claims. 409 U.S. at 71-72, 93 S.Ct. 253, 175 USPQ at 676.
. The dissenting opinion in Noll concludes that Benson stands for the proposition that computer programs are not statutory subject matter under section 101, basing its conclusion on language in Benson that “[i]f these programs are to be patentable, considerable problems are raised which only committees of Congress can manage * * (Emphasis added and footnote omitted.) However, “these programs” refers to the specific type of claimed program involved in Benson and not to computer programs in general, as clearly evidenced by the immediately preceding paragraphs of the Benson opinion (409 U.S. at 71-72, 93 S.Ct. at 257, 175 USPQ at 676-77):
It is argued that a process patent must either be tied to a particular machine or apparatus or must operate to change articles or materials to a “different state or thing.” We do not hold that no process patent could ever qualify if it did not meet the requirements of our prior precedents. It is said that the decision precludes a patent for any program servicing a computer. We do not so hold. * * * What we come down to in a nutshell is the following.
It is conceded that one may not patent an idea. But in practical effect that would be the result if the formula for converting [binary-coded decimal] numerals to pure binary numerals were patented in this case. The mathematical formula involved here has no substantial practical application except in connection with a digital computer, which means that if the judgment below is affirmed, the patent would wholly pre-empt the mathematical formula and in practical effect would be a patent on the algorithm itself.
It may be that the patent laws should be extended to cover these programs, a policy matter to which we are not competent to speak. The President’s Commission on the Patent System rejected the proposal that these programs be patentable * * *. [Emphasis added and footnotes omitted.]
. Over-concentration on the word “algorithm” alone, for example, may mislead. The Supreme Court carefully supplied a definition of the particular algorithm before it, i. e., “[a] procedure for solving a given type of mathematical problem.” The broader definition of algorithm is “a step-by-step procedure for solving a problem or accomplishing some end.” Webster’s New Collegiate Dictionary (1976). It is axiomatic that inventive minds seek and develop solutions to problems and step-by-step solutions often attain the status of patentable invention. It would be unnecessarily detrimental to our patent system to deny inventors patent protection on the sole ground that their contribution could be broadly termed an “algorithm.”
. We are not unmindful of literature discussing broad public policy considerations involved in the question of software patent protection. See, e. g., Keeffe & Mahn, Protecting Software: Is It Worth All The Trouble?, 62 A.B.A.J. 906 (1976). We are limited, however, to determination of issues presented in specific cases. Though we decide the present case, as we must, in the light of our best understanding of the statute and prior decisions of the Supreme Court, we join our brother Rich’s scholarly call for definitive guidance by higher authority.
. The “formula-algorithm” category is treated in the text. We have extensively reviewed the doctrine of “mental steps” in In re Prater, 415 F.2d 1393, 56 CCPA 1381, 162 USPQ 541 (1969) and In re Musgrave, 431 F.2d 882, 57 CCPA 1352, 167 USPQ 280 (1970), and find no need to repeat that analysis herein. It is enough to say that we find the claims herein to be limited to a method of operating a machine system.