T 0453/91 31-05-1994

1. The appeal (cf. point III) is admissible.

2. Amendments

2.1. Claim 1 is based on the original Claim 14.

2.2. Dependent Claim 2 is based on the original Claim 15.

2.3. Claim 3, although referring back to (product) Claim 1, is an independent claim insofar as it is of a different category (method).

The method claimed is a method of manufacturing a product (viz. the product claimed in Claim 1). In this respect, Claim 3 is based on, for instance, the reference to "production" in the original description (e.g. on page 6 line 12 and page 10 line 7).

That method comprises a partial method of designing the said product, namely the semiconductor chip claimed in Claim 1. As to this partial method, and the steps it includes, Claim 3 is based, in essence, although with different words, on the original Claims 1 to 5. More particularly, all of the steps of the said partial method in Claim 3 can either directly be refound or are implicit in features (a) to (h) of the original Claim 1 or in the features added by, for instance, Claims 2, 3 and 5.

The concluding step of the said manufacturing method is, again, based on the afore-mentioned reference in the description.

2.4. Similarly, Claim 4 is based,

- as to its category and the last step of the manufacturing method, on the description, and

- as to the partial method of designing the product (chip) and the steps it includes as defined, on the original Claim 6 and its dependent claims, particularly Claims 9 and 11.

2.5. Dependent Claims 5 to 7 are based on the original Claims 10, 12 and 13, respectively.

2.6. The amendments made to the description are, in essence, limited to corrections and modifications rendering the description allowable under Rule 27(1)(b) and (c) EPC.

They are, therefore, also admissible under Article 123(2) EPC.

3. Novelty and Inventive Step - Product Claim

3.1. From the piece of prior art (D) cited as coming nearest to the claimed invention (cf. point VI) a semiconductor chip divided as defined in the introductory passage of Claim 1 is known.

3.2. Feature (a) of Claim 1 is not expressly mentioned in the text of D. However, it is noted that, if the equations (1), for instance y11+y21=y, are taken to be exact, this would point to no gaps being provided between the partitions.

Furthermore, Fig. 1 and 2 of D would seem to give the impression, and thus confirm the view based on the equations (1), that the partitions are intimately attached to each other at their respective adjacent edges leaving no space in between, as claimed by way of feature (a).

3.3. Whereas any partition of a prior art chip would have emerging and ending interconnection lines as a matter of course, no crossing interconnection lines not connected with any of the electrical elements are either mentioned in the text or shown in the figures of D.

It could be argued that it follows from the appearance of feature (a) in Fig. 1 and 2 of D, that

- inevitably no interconnection lines can be placed in channels between the edges of adjacent partitions and

- therefore the partitions in the upper left corner and in the lower right corner (having the dimensions x11.y11 and x22.y22) can, in the absense of a common boundary, or edge, only be connected by interconnection lines drawn either through other partitions (such as that having the dimensions x21.y21 and/or that having the dimensions x12.y12) or around the outer edges of the chip.

However, nothing in D would point to a particular selection of one or the other of these two alternatives.

It is, thus, not possible to clearly and unambiguously derive feature (b) of Claim 1 from D. This feature is therefore to be regarded as new.

3.4. According to D, VLSI chips are designed so that they are "dense". Nothing in D would, however, point to the possibility of partitions being of different density, or porosity.

Thus, evidently, feature (c) cannot be derived from D.

3.5. Thus, the subject-matter of Claim 1, as a whole, is clearly novel.

3.6. As already said above (3.3), nothing in D would point to the selection of one or the other of the two alternatives mentioned.

From general considerations based on the skilled person's general knowledge, he would also not be led to consider choosing specifically the former one of these two alternatives and not the latter one. On the contrary: Given that hitherto such interconnection lines used to be placed in channels between partitions, the skilled person would, if no such intermediate channels exist, be led to look for, and use, channels around the outer edges of the chip.

In D, keeping at a minimum the interconnecting wire lengths between partition centers and bounding I/O's is mentioned as desirable. However, in the absence of any proposal in the prior art to draw an interconnection line through partitions none of whose electrical elements are to be connected to it, the skilled person would not consider attempting to solve the problem of minimizing wire lengths by such a deviation from the prior art.

Feature (b) does not, therefore, appear to be obvious from the prior art.

3.7. As already said (3.4), nothing in D would point to partitions being given different porosity.

In effect this means that feature (c) is not only new but also unobvious.

3.8. Therefore, even in the "worst" case, namely if feature (a) were regarded as being known from D, the subject- matter of Claim 1 is to be considered, in agreement with the Examining Division's apparent view (cf. point II), to involve an inventive step.

4. Novelty and Inventive Step - Method Claims

A process of manufacturing the product claimed in Claim 1 in a particular way so that it has the technical features defined in Claim 1, inevitably results in that particular product and not in one which does not have these features but differing ones. This is particularly so if the manufacturing process comprises a partial method of designing the said product, by respective steps, in such a particular way that it has the said technical features.

This being true in the present case for Claims 3 and 4, it follows already from the novelty and inventiveness of the subject-matter of Claim 1, based on the novelty and non-obviousness of at least some of its features, that the subject-matter of Claims 3 and 4, concerned with particular ways of achieving these features, is also novel and inventive.

5. Non-Exclusion from Patentability - Method Claims

5.1. Therefore, the only issue remaining to be decided in respect of these claims is, following the reasons given for the rejection of the design method claims in the decision under appeal, whether the subject-matter of Claims 3 and 4 is an invention within the meaning of Article 52(1) EPC or not (Art. 52(2) in conjunction with (3) EPC).

5.2. In the particular circumstances of the present case, it appears appropriate first to refer, for contrasting Claims 3 and 4 with the version of the method claims underlying the decision under appeal, to that earlier version, even though it has not been maintained.

As far as the kind of method is concerned for which protection was sought, the methods then claimed could be interpreted as delivering a mere "design" in form of an image of something which does not exist in the real world and which may or may not become a real object; i.e. the result of the claimed method would not necessarily be a "physical entity".

In the earlier decision T 208/84 (OJ EPO 1987, 14), it was held that a "physical entity may be a material object but equally an image stored as an electric signal" (cf. Reason 5). In this finding, the fact that the Board equated an "image" with a "material object" was apparently based on the assumption that the said "image" was that of a "material object". For, "in contrast", the Board in the earlier case held (cf. Reason 7) that "a method for digitally filtering (or, more generally: processing) image data (would) remain an abstract notion not distinguished from a mathematical (or, more generally: non-technical) method so long as it is not specified what physical entity is represented by the data and forms the subject of a technical process" (generalizations added). The "methods for design", as formerly claimed in the present case, would seem to fall under this latter case of an "abstract notion" and not under the former of a "physical entity".

Moreover, referring to the individual steps of designing, these would not seem to make a contribution to the art outside the fields of excluded matters, such as performing mental acts and implementing the resulting steps by programs for computers (Art. 52(2)(c) EPC).

Therefore, in accordance with the case law, the Board would agree with the rejection of the method claims in their earlier versions for the reason that they seemed not to relate to a technical process making a contribution to the art in a field not excluded from patentability.

5.3. However, contrary to those earlier versions, Claims 3 and 4 now on file are clearly restricted to a method (process) of manufacturing a real (physical) object having technical features and thus to a technical process.

With particular reference to features (a) and (b) of Claims 3 and 4, the manufacturing processes now claimed no longer comprise only the "designing" method, marked (a), as did the said earlier versions of the method claims, but also, although in most general terms, the "producing" steps proper, marked (b), since in principle, it should be assumed that a claim is directed solely to the combination of all its features (following decision T 175/84, OJ EPO 3/1989, 71).

No objection under Article 52(2), if applied in conjunction with 52(3), EPC arises therefore in respect of these claims.

6. Conclusions

6.1. The independent product Claim 1 and method Claims 3 and 4. are therefore (points 3, 4 and 5, respectively) allowable.

6.2. No objection arises in respect of the dependent product and method Claims 2 and 5 to 7, respectively.

6.3. The same applies to the description and to the drawings.