Armour Group Plc v Leisuretech Electronics Pty Ltd

In action HC 2007 C01407 Armour Group plc seeks the revocation of European Patent (UK) No. 1 004 221 B1 (“the Patent”). In action HC 2007 C01673 LeisureTech Electronics Pty (“LeisureTech”) sues Armour Home Electronics Ltd and QED Audio Products Ltd for infringement of the Patent. There is no need to distinguish between the various Armour companies and I will refer to them collectively as “Armour”. There is no dispute that Armour’s system infringes the Patent if it is valid. Armour challenge the validity of the Patent on two grounds, obviousness and insufficiency. Insufficiency was deployed primarily as a squeeze. Armour’s principal attack is obviousness over common general knowledge. They also rely on two items of prior art known as Linn Knekt and MRS, but as discussed below these add little to the principal attack. A number of other issues fell by the wayside in the run up to, and during, the trial.

The Patent is for a distributed stereo audio system, that is, a system for providing stereo sound to several rooms or zones from a single source of audio signals. Distributed audio systems are often referred to as multi-room audio systems, and I shall use both expressions interchangeably. The Patent has a priority date of 15 August 1997. LeisureTech proposes to amend the Patent in the manner set out below. Armour do not object to the amendments. For reasons which it is unnecessary to go into, LeisureTech did not formally apply to make these amendments prior to the trial. It was agreed that the trial would be conducted on the basis of claim 1 as proposed to be amended and that, if the validity of that claim is upheld, directions will be given for the making of a formal application to amend. If claim 1 as proposed to be amended is invalid, LeisureTech does not contend that any of the subsidiary claims is valid.

Armour called as their expert William Miller. Mr Miller studied Electrical and Electronic Engineering at Glasgow University, and obtained his degree in 1977. After completing his degree, he worked for Hewlett Packard for four years as a design engineer. After leaving HP, he spent the next 25 years working for Linn Products Limited (“Linn”). During that period Linn became a well-known manufacturer of hi-fi components and systems, in particular “high end” products targeted at audiophiles. He was a member of the Management and Product Development Group at Linn and was thus involved in technical aspects of the design of a large number of audio systems and components. From 1981 to 1988 he was a design engineer. From 1988 to 2002 he was the Design Engineering Manager. As such, he was responsible for the development of Linn's multi-room audio system, the Linn Knekt. From 2002 to 2006 he was Chief Engineer.

LeisureTech accepted that Mr Miller’s oral evidence was completely honest and fair. They criticised his written reports as being inaccurate in two respects, but in my view these inaccuracies were merely the result of slightly careless wording. More importantly, LeisureTech submitted that Mr Miller was overqualified in that he was more skilled and experienced than the addressee of the Patent and was inventive. Mr Miller repeatedly emphasised in his evidence, however, that he had tried to view matters from the perspective of the ordinary skilled person, and I am satisfied that he did so as well as anyone in his position could. LeisureTech also made the valid point that, since Mr Miller had spent almost the entirety of his career at Linn, which was both a manufacturer of high end equipment and an innovative company, this was likely to have influenced his viewpoint. I have taken this into account in assessing his evidence. I have taken into account the fact that Mr Miller was slightly diffident in his manner, something that I attribute to his personality and his inexperience as an expert witness rather than to his technical ability or his belief in the opinions he expressed.

LeisureTech called as its expert Martin Colloms. Mr Colloms graduated from The Polytechnic (now the University of Westminster) in 1971 with a BSc in Electrical Engineering specialising in Electronics. Following graduation, he was employed as a design engineer in electronic instrumentation at the UK division of Tektronix Inc, a US company best known for its test and measurement equipment (e.g. oscilloscopes and logic analysers). In 1973, he co-founded Monitor Audio Limited, a company specialising in high fidelity loudspeaker designs. His role was as Technical Director of the company with responsibility for the design and quality control of its products. He left the company in 1975. In that year he founded an independent design and evaluation consultancy, Colloms Electroacoustics, through which he provides a range of services to the audio and electronics industry internationally. He has an impressive list of clients in this field. During his career Mr Colloms has designed or co-designed a large number of audio products, and has designed and installed sound distribution systems. He was admitted as a Chartered Engineer of the Institution of Electrical Engineers in 1981.

Mr Colloms is also a prolific author. He has been Contributing Technical Editor of Stereophile in the USA for 15 years and held an equivalent post at Hi Fi News and Record Review in the UK for 25 years. He is the publisher of a journal called HIFI CRITIC, which he founded in late 2006 to provide more serious coverage of stereo systems for enthusiasts and the audio industry. By 2007 he had published more than 3,900 product reviews and articles in a variety of publications. He has also written or contributed to three books, including High Performance Loudspeakers, which was first published in 1976 and is currently in its sixth edition.

An unusual aspect of Mr Colloms’ career is that he has had much more extensive experience with patents than most expert witnesses in patent actions. He is a named inventor on 200 patents worldwide. In addition, he has acted as a consultant intellectual property manager and patent analyst to NXT plc since 1995, in which capacity he has been responsible for advising on over 1,000 patent applications in 260 patent families. The activities which he lists in his CV as having been performed in this role include many activities of the kind normally undertaken by patent attorneys, including patent drafting, analysing prior art, responding to official actions, attending examiner interviews, helping attorneys prepare oppositions and responses to oppositions, revising patent claims and responding to patent queries by licensees. In addition, Mr Colloms has given expert evidence in several previous cases.

Armour submitted that Mr Colloms was focussed on the higher end of hi-fi. I do not accept this. If anything, Mr Colloms has more experience of the middle and lower ends of the hi-fi market than Mr Miller. More importantly, Armour submitted that Mr Colloms was a patent advocate par excellence. As I have said, I agree that Mr Colloms’ experience of patents was unusual and I have taken this into account in assessing his evidence. Armour did not suggest, however, that Mr Colloms had been partisan in his approach to this case. I have also taken into account the fact that Mr Colloms was more confident in his manner than Mr Miller, which I attribute to his greater experience in this field.

LeisureTech also adduced unchallenged evidence of fact from Paul Tam, the co-owner and director of The Listening Rooms, a home electronics retail business. I have taken this evidence into account, but in my judgment it is of little assistance on the key issues because, although Mr Tam evidently had considerable knowledge of the distributed audio system market at the relevant time, he was not representative of the addressee of the Patent since he did not possess the qualifications and experience of such a person as discussed below.

The Patent is a fairly short document. Paragraph [0001] defines the technical field. Paragraphs [0002]-[0008] describe the background art. This section includes an acknowledgement of the Linn Knekt system. Paragraphs [0009]-[0024] are headed “Summary of the invention”. The key paragraphs are as follows:

The specification then describes two specific embodiments. The first, illustrated in Figure 1, comprises a source of audio signals consisting of several units and a power supply in one room connected by means of a Cat.5 cable to an amplifier and two speakers in another room. The second, illustrated in Figure 2, comprises a source of audio signals and a power supply in one room connected by means of two Cat.5 cables to an amplifier and two speakers in each of two other rooms.

The specification concludes with some more general description, including some technical details regarding Cat. 5 cable. It is explained that Cat. 5 is an abbreviation of category 5 in the Underwriters Labs Level classification system, which corresponds to a category in the American National Standards Institute’s Electronic Industry Association/Telecommunication Industry Association 568A standard.

As proposed to be amended, omitting reference numerals and broken down into integers, claim 1 is as follows:

As can be seen, the amendments explicitly limit the claim to systems in which the amplifier is connected to the source and power supply by means of a single Cat.5 cable. Thus systems with multiple parallel Cat.5 cables of the kind contemplated by the last sentence of paragraph [12] are excluded.

Although the subsidiary claims are not in issue, it is worth noting two features which are the subject of them. First, claim 5 adds the limitation that “the or each amplifier is based on an integrated circuit amplifier”. Secondly, claim 13 adds the limitation that “infrared signals received by the amplifier are transmitted to the source or source components through a fourth twisted pair in the Cat.5 cable”. These signals are control signals for e.g. source selection and/or volume control.

A patent specification is addressed to those likely to have a practical interest in the subject matter of the invention, and such persons are those with practical knowledge and experience of the kind of work in which the invention was intended to be used. The addressee comes to a reading of the specification with the common general knowledge of persons skilled in the relevant art, and he (or, once and for all, she) reads it knowing that its purpose is to describe and demarcate an invention. He is unimaginative and has no inventive capacity.

In the present case there is little, if any, dispute between the parties as to the identity and attributes of the person skilled in the art to whom the Patent is addressed. He is a designer of home audio systems, in particular distributed or multi-room systems. Such a person would have a degree in electronics or electrical engineering, or an equivalent qualification, and some post-qualification experience in designing audio systems. He would know not merely about multi-room audio systems, but about audio systems and components generally. Mr Colloms described the knowledge required of such a person in this way:

The correct approach to identifying the information which constitutes the common general knowledge of the addressee was explained by Aldous LJ in Beloit Technologies Inc v Valmet Paper Machinery Inc [1997] RPC 489 at 494-495. In the present case there was considerable ground between the parties and their respective experts as to the common general knowledge, but also some differences. I find that the common general knowledge included the following matters. For convenience I shall mainly use the present tense, but I am referring to the position in August 1997.

Home audio systems come in many shapes and sizes, but they all have three basic features. First, a source of music and other sounds. This might be a radio tuner, CD player, cassette player, record deck or a combination of some or all of these. Secondly, an amplifier to amplify the signals generated by the source. Thirdly, at least one loudspeaker, or in a stereo system at least two loudspeakers. For many years, these components have been available in various physical configurations. There are “all in one” products where source, amplifier and speaker(s) are all combined in a single unit. These include portable radios and “boomboxes” or “ghettoblasters”. There are also products, often referred to as “music centres”, where the source and amplifier are in one box with separate loudspeakers connected by cables. Then there are “separates”, where each item is contained in a separate box and connected by cables.

As with most kinds of consumer products, so with home audio equipment there are different kinds of product for different segments of the market. The car market has its sports cars, its family saloons and its small hatchbacks. The market for home audio is no different. There is a spectrum of products both in terms of sound quality and expense. Moreover, as discussed below, these are not the only variables. Although, as Mr Colloms pointed out, it is possible to achieve good sound quality on a moderate budget, there is a well-established distinction between two broad segments of the market.

First, there is an “audiophile” market populated by hi-fi buffs. Audiophiles are primarily interested in achieving good sound reproduction. They will generally purchase carefully selected separates with high specifications, often from different manufacturers, so that the CD player is made by A, the tuner by B, the amplifier by C, the loudspeaker by D and the cables by E. Some people with relatively modest budgets might count themselves as audiophiles, but many audiophiles will be prepared to spend hundreds, or even thousands, of pounds on individual components of a system. While audiophiles may be interested in the visual aesthetics of the products they buy, obtrusiveness is not generally a concern for them. Indeed, they may be proud of having a large pair of loudspeakers in their living room and be prepared to tolerate ugly runs of speaker cable being visible.

Secondly, there is a “mainstream” market in which sound quality is less of a priority. Portable radios, boomboxes, music centres and systems consisting of matched separates all of which are made by the same manufacturer primarily sell into this market. As a broad generalisation, consumers in this segment of the market are more concerned about both cost and obtrusiveness than audiophiles.

At this point it is necessary to say something about the term “hi-fi”. This is an abbreviation of “high fidelity”. The Oxford English Dictionary defines the term “hi-fi” as meaning:

Although some home audio equipment, such as a portable radio, produces mono sound, most produces stereo. For present purposes mono equipment can be disregarded, and in the remainder of this judgment it should be taken as read that I am considering stereo systems even if I do not say so.

Like home audio systems in general, there are different types of distributed audio systems which range in sophistication and cost. The simplest type, which has existed for many years, comprises systems in which the amplifier (or amplifier part of the unit) has connections for two sets of loudspeakers. One set would usually be placed in the same room, while the other could be placed elsewhere. This would enable the user to listen to music in a second room, but would not enable him to control the system (for example to adjust the volume) without returning to the first room. The same result could also be achieved by a switch box external to the amplifier. Then there are “satellite” systems, which have also been around for a considerable time. These systems comprise a central audio system with multiple connections from the amplifier to loudspeakers in various rooms. The desired source (e.g. CD player or tuner) can be selected and controlled from the rooms (e.g. by means of a wall-mounted control unit), but there is no facility to listen to different sources in different rooms. More recently, multi-room systems have emerged which enable the listener to select different audio sources for different rooms using a switching matrix controllable from each room. Such systems employ either a multi-channel amplifier or several amplifiers.

A common requirement of multi-room audio system customers is that the system should be unobtrusive. This often means concealing the amplifiers, the loudspeakers and the cabling. Compact amplifiers and flush-fitting, in-wall or in-ceiling loudspeakers serve this purpose. Such loudspeakers are necessarily of lower quality than separate free-standing or even shelf-mounted loudspeakers. Because these loudspeakers are embedded in the wall or ceiling they have to be quite small. Their size means that their low frequency, bass, performance is compromised. In addition, much of the sound generated is lost into the ceiling or wall cavity because there is no room to construct a speaker enclosure. They are visually unobtrusive, but as a result they give relatively poor sound quality. Nevertheless, there is a substantial market for such systems amongst customers who value unobtrusive aesthetics over sound quality.

The majority of multi-room audio systems are designed to be installed by professional installers. This market is known as the “custom install” market. Custom installers work with clients, for example householders, architects or developers, and their job is to join together a number of off-the-shelf products into a coherent and functional whole. A custom installation will often not only include entertainment systems (audio and video), but also the lighting system, heating and air conditioning and other aspects of home automation. The products in question are inevitably from different manufacturers, perform a wide range of different functions and use different control protocols. Assembling them into a seamless whole is a complex undertaking because it involves creating a user interface that can control all of the components of the system. This typically involves the use of a programmable control system that has to be programmed by the installer to recognise and control the system components. In addition, the aesthetic demands of customers who choose multi-room audio systems, including the need to hide wires and place loudspeakers discreetly about the home, means that practical skills are required to complement the programming skills. For these reasons, trained and experienced professional installers perform nearly all installation work.

Because most multi-room audio systems are installed by custom installers, any designer of such systems has to be well aware of the needs and views of the installers. Installers prefer systems that are easy to install, reliable, expandable and “open”, that is to say, interconnectable with other components.

Mr Miller’s evidence, which I accept, is that the market for multi-room audio systems started to take off in 1988 and that the custom install business grew from almost nothing in 1990 to substantially larger than the “high end” hi-fi retail business by the late 1990s.

Mr Miller gave evidence that the designer of a distributed audio system in 1997 would have to make a number of choices between various well-known options regarding different aspects of the system. These included the following:

It is clear from the evidence that there are, and the skilled person would be aware that there are, trade-offs between many of these choices in terms of functionality, sound quality, cost, unobtrusiveness, ease of installation and so on.

It is common ground that in August 1997 two basic architectures for distributed audio systems were common general knowledge. The first involves “central” amplification: the signal from the source is amplified by a central amplifier and then distributed to speakers in the various rooms at “speaker level”. The second involves “local” or “remote zone” amplification: the signal from the source is distributed to amplifiers situated in each of the rooms at “line level”. It is also common ground that at that time the large majority of systems were central amplification systems.

Although counsel for LeisureTech suggested to Mr Miller in cross-examination that the choice of central amplification was a “one way street” for a moderately priced, unobtrusive system, it was common ground between the experts that each architecture was known to have advantages and disadvantages. Indeed, in the common general knowledge section of his first report Mr Colloms listed no less than six classes of problems with central amplification, namely (i) cable issues, (ii) transformer-related losses, (iii) volume control issues, (iv) signal selection issues, (v) difficulties in automating and remote control and (vi) overall expense. Among the problems he highlighted were the need to use powerful, expensive amplifiers and bulky, expensive speaker cables. Mr Colloms went on:

In the case of local amplification systems, it is common ground that all the distributed audio systems on the market of which the skilled person would be aware from his common general knowledge involved “local power”, that is to say, each amplifier was powered from a supply located in the same room as the amplifier. It is also common ground that there was one system on the market (namely the MRS system discussed below) involving “central power”, that is to say, power distributed to the amplifiers from the central location along the same cable as the audio signal; but the existence of that system was not common general knowledge. Armour contend that, even though he would not have known of any actual example on the market from his common general knowledge, the skilled person would also have been well aware that local amplification with central power was a possible option. LeisureTech disputes this.

In my judgment the evidence of the experts establishes that using local amplification with central power was part of the common general knowledge in the sense that the addressee would have been well aware that it was one option available. Thus Mr Colloms stated in the section of his first report dealing with common general knowledge:

Like the other parts of a home audio system, amplifiers range widely in their size, performance and cost. A key measure of amplifier performance is the output power expressed in Watts per channel.

LeisureTech contends that the general perception in the art in 1997 was that 30W per channel or so was the starting point for hi-fi sound quality; that it was regarded as desirable for amplifiers to have a certain amount of “headroom” in terms of power so as to be able to deal with sudden dynamic changes without clipping; that most amplifiers described as hi-fi were rated at 30W per channel or more, and in many cases significantly more; and that all of the multi-room audio systems available in 1997 had amplifiers rated at 30W per channel or more. By contrast, Armour contend that the person skilled in art was well aware that the power ordinarily required for normal domestic listening was much less than 30W per channel, with around 1W per channel being adequate; that systems rated at 30W per channel or more mostly operated at much lower power during ordinary use; that many domestic audio systems were rated at only a few Watts per channel; and that (as described in more detail below) integrated circuit amplifiers were available off the shelf which were rated at a few Watts per channel which were suitable for home audio applications.

In my judgment the evidence demonstrated that both parties’ contentions are well-founded, and there is no contradiction between them. LeisureTech’s focus here is upon the audiophile market. I am satisfied that in that market the general perception was as LeisureTech contends, notwithstanding the fact that it was known that normal domestic listening did not require anything like as much power as 30W. But as I have described above, the audiophile market was not the only market for home audio equipment. In the mainstream market the position was different. The person skilled in the art would have been aware of both. In particular, the skilled person would have been aware that, if one was prepared to compromise to some extent on sound quality, lower-powered amplifiers were not merely available, but also perfectly satisfactory.

So far as multi-room audio systems are concerned, it does appear to be the case that all the systems on the market used amplifiers rated at 30W per channel or more, but I do not accept that the skilled person would have thought that it was essential to employ such powerful amplifiers. As discussed above, many customers for such systems are more interested in visual appearance than in sound quality of the kind expected by audiophiles. Mr Colloms’ evidence was that consumers purchasing distributed audio systems “expected at least basic hi-fi quality [my emphasis]”. The skilled person would have been well aware that lower-powered amplifiers could deliver sound of that quality.

An important class of amplifiers for use in home audio systems is integrated circuit or IC amplifiers. These are amplifiers in which most of the components have been fitted into a single integrated circuit. Such amplifiers are referred to in the Patent at paragraph [0016]. It is common ground that such amplifiers were well known and widely available in 1997. Mr Colloms described them in this way:

A considerable range of IC amplifiers was available in 1997. Specification sheets for a number of examples of IC amplifiers which were available then are in evidence, including several manufactured by Toshiba. These include the TA8211AH dual audio power amplifier, which provided an output power of 6W per channel, had a recommended supply voltage of 20V for an 8Ω load resistance and was described as “suitable for power amplifier of TV and home stereo”; and the TA8216H dual audio power amplifier, which provided an output power of 13W per channel, had a recommended supply voltage of 24V for a 4Ω load resistance, and was described as “suitable for power amplifier of music center”. In the case of the TA8216H the specification sheet includes a graph of total harmonic distortion (“THD”) against output power at three different frequencies, which shows that the device has a THD of less than 0.3% when output power is below 10W. Consistently with the third extract from his reports quoted above, Mr Colloms agreed in cross-examination that this was a hi-fi amplifier if used within its limits.

The evidence does not go so far as to establish that either the TA8211AH or the TA8216H was itself common general knowledge in 1997, but I am in no doubt that the existence and general characteristics of IC amplifiers of the type exemplified by the TA8211AH and the TA8216H were part of the common general knowledge of the addressee. Selecting a suitable IC amplifier for a given application from the manufacturers’ catalogues would have been a routine matter for a skilled person. Furthermore, the skilled person could quickly calculate the power requirements of such an amplifier from the information provided in the data sheet.

IC amplifiers of the kind exemplified by the TA8211AH and the TA8216H have a number of advantageous characteristics. First, they are very compact (less than an inch long), which makes them highly suitable for use in applications where space is at a premium. Secondly, they are inexpensive, which makes them highly suitable for use in applications involving a multiplicity of amplifiers. Thirdly, because they are low power devices they do not dissipate a great deal of heat. Fourthly, they are quite tolerant of variations in the power supply.

In order to connect the various parts of a distributed audio system together one needs cables. Cables are used for a number of purposes. If central amplification is used, long runs of cable are needed to take the speaker level signal from the central amplifier to the speakers in the various rooms. If local amplification is used, long runs of cable are needed to take the line level signal from the source to the remote amplifiers and much shorter runs to take the speaker level signal from each amplifier to its speakers. Cable that carries speaker level signals must carry a heavier load than cable that only carries line level signals. This presents a problem for long runs of cable, because resistance is proportional to length. Cable for both purposes should be good at signal transmission without interference.

Cable is also needed for the transmission of electrical power. If local amplification with local power is used, then the cable is likely to be carrying ordinary mains power (i.e. 240V AC) which is transformed in the amplifier. If local amplification with central power is used, then the cable only needs to carry relatively low voltage DC power (e.g. 24V DC).

Cat.5 is an industry standard for a certain type of cable which contains four sets of twisted pairs of thin conductors. Telephone systems have used twisted pair cables for many years. Twisted conductor pairs have two significant characteristics: (i) with appropriate circuitry at each end, they have a high immunity to electromagnetic interference; and (ii) they have a predictable and consistent characteristic impedance which, with appropriate circuitry at each end, yields good high frequency signal transmission. The former is particularly relevant for the distribution of audio signals; the latter is essential for the transmission of high speed computer networking signals. Because Cat.5 cable has four twisted pairs of conductors, it is inherently suitable for carrying up to four different signals, although in some circumstances it can carry eight.

Cat.5 has certain other advantages. It is inexpensive and widely available; it can be used with simple RJ45 connectors similar to those used to connect telephone cables to sockets; and it is tolerant to bending and abrasion, and so is relatively easy to thread through a building. Mr Miller gave unchallenged evidence that it would have been assumed to have been certified as flame retardant, although the evidence does not establish that it actually was certified in 1997. Certainly, it appears that by then it was being widely installed in newly-built houses as well used for custom installations in existing houses.

Cat.5 was originally developed for data transmission. Indeed, it is probably best known for its use in Ethernet networks. The first standard describing Cat.5 was published in 1991, but it is clear that it caught on quickly and not just in the computer industry. This is illustrated by what happened with the Linn Knekt system. Linn started developing this system in about 1987. To begin with, Linn used FCC68 eight-conductor flat cable terminated with RJ45 plugs. The product was launched in June 1993. Almost immediately after the launch, Linn was informed by custom installers that they wanted to use Cat.5 cable with the system instead of FCC68. Linn was able to achieve this simply by providing installers with a diagram showing how to connect the eight conductors in a Cat.5 cable to the eight terminals of an RJ45 plug for use with the Linn Knekt system. It was Mr Miller’s evidence that by 1993 it was clear that Cat.5 was the preferred cable for use in multi-room audio installations. Mr Colloms agreed that by 1997 it had become the most commonly used cable for wiring homes for communication and control purposes, but not for audio.

LeisureTech contends that the use of Cat.5 cable for transmission of audio signals, as opposed to digital data, was not common general knowledge by August 1997. In support of this contention it points out that the only system in which Cat.5 is known to have been used for audio signals by that date is the Linn Knekt. Against this, Armour contend that the Linn Knekt system was itself well known to skilled persons interested in multi-room systems and that that in itself was sufficient to make the use of Cat.5 for carrying audio signals common general knowledge. In my judgment the evidence supports this contention. In any event, however, I am satisfied that the skilled person would have been well aware of the existence and characteristics of Cat.5 cable, and in particular its suitability for carrying audio signals, particularly line level audio signals. Thus in his discussion of the common general knowledge in his first report Mr Colloms said:

LeisureTech also contends that, even if the use of Cat.5 for the transmission of audio signals was common general knowledge, the use of Cat.5 for the transmission of electrical power, and in particular power together with audio, was not. In support of this contention it points to the fact that, even in the case of the Linn Knekt, Cat.5 cable was not used to carry power to the amplifiers, but only audio and bi-directional control signals. Indeed, although Linn Knekt was a local amplification system, it used local power for the amplifiers (although it did use Cat.5 cable to carry about 3W of power from the amplifier to the room control unit).

In my view one must be careful here to be precise about what one means by “power”. What matters is for present purposes is low level power of the order of 15W delivered by low voltage DC. It is common ground that Cat.5 cable is not in fact capable of carrying any higher power than about 15W over any significant distance, but it had been used in the telecommunications and computing fields for carrying low level power. I agree that the actual use of Cat.5 cable to carry power has not been shown to be common general knowledge in the audio field, but I am satisfied that the skilled person would have been aware that Cat.5 cable was suitable for carrying low level power. After all, it is an electrical cable. As Mr Colloms accepted during cross-examination, the skilled person would readily be able to calculate the power-carrying capabilities of a length of Cat.5 cable.

As for transmission of audio signals and low level power, again I am satisfied that the skilled person would have been aware that Cat.5 was suitable for carrying both. As can be seen from paragraph 35 above, it was Mr Colloms’ evidence that using the multi-room system cable to carry both audio signals and low voltage power was an option that was part of the common general knowledge. It was also his evidence that Cat.5 cable was well known to provide good immunity to electrical noise and interference.

Because custom installations usually include other systems in addition to audio systems, they are somewhat expensive. The evidence suggests that the cheapest custom installations in 1997 cost several thousand pounds. LeisureTech contends that this meant that cost saving was not a particular concern for a designer of a multi-room audio system, and in particular not cost saving on individual components such as amplifiers. I do not accept this. It was common ground between the experts that there was a market for moderately priced multi-room audio systems. Such systems could be sold both to those who only wanted a distributed audio system and to those who wanted a complete custom installation. In the former case, there were customers who could not afford, or did not want, to spend thousands of pounds. Satellite systems were among the products targeted at such customers. In the latter case, some customers might well prefer to spend more money on, say, the video and lighting systems than on the audio system. Furthermore, the designer of a multi-room audio system would appreciate that the cost of the system depended not merely on the cost of the components, but also on the cost of the cables and on the ease of installation.

A moderately priced distributed audio system dating from before August 1997 which was mentioned by both experts in their reports, and by Mr Tam in his statement, is the QED Systemline system. Mr Colloms described Systemline in this way:

It will be noted that Mr Colloms says that the objective of reduced cost was not fully met because the remote amplifiers delivered 30W per channel and required the provision of local mains supply, and that a defect of the system was the use of a proprietary cable. Similarly, Mr Tam said that the system was “quite bulky and expensive because each amplifier required an individual power supply”.

Systemline was promoted both as being “almost invisible” and as producing “real hi-fi sound”, but the evidence of both experts was that it did not deliver good sound quality. Thus Mr Colloms said:

Mr Tam’s evidence was that in about 1995-97 Systemline was developed into a central amplification system in which Cat.5 cable was used to carry control signals but not amplifier power. It is not entirely clear from his description whether or not the Cat.5 cable carried audio signals, but my understanding is that it did not.

Subsequently Systemline was developed into the system which is the subject of LeisureTech’s infringement action.

The task for the court when construing a patent claim is to determine what the person skilled in the art would have understood the patentee to have been using the language of the claim to mean: see Kirin Amgen Inc v Hoechst Marion Roussel Ltd [2004] UKHL 46, [2005] RPC 9 at [30]-[35]. In that case the list of principles to be found in the judgment of Jacob LJ in Technip France SA’s Patent [2004] EWCA Civ 381, [2004] RPC 46 at [41] was approved subject to one point.

Unusually, in the present case there is little dispute as to the construction of claim 1. Nevertheless, there are three points of construction which are of some significance. First, it is common ground that claim 1 is not limited to systems which deliver hi-fi sound quality or indeed any particular sound quality at all. Paragraph [0024] of the specification refers to “a reasonable sound level for most domestic requirements [emphasis added]”, which is not the same thing; but that is as far as the specification goes. This is not surprising given that sound quality has a considerable subjective element, although there are some objective measures.

Secondly, LeisureTech contends that claim 1 is inherently limited to systems which deliver the sort of electrical power, and hence the amplifier output power, that a single pair of conductors in a Cat.5 cable is capable of delivering. Armour pleaded as the basis for their insufficiency attack that claim 1 has no limit with regard to the power of the system. As I will discuss below, however, counsel for Armour did not press this attack, submitting instead that the insufficiency objection had done its job in forcing LeisureTech to limit its claim. Accordingly, I accept LeisureTech’s construction.

Thirdly, claim 1 is not limited to systems of any particular size, in particular in terms of either the number of rooms or the lengths of cable required.

A patent will be invalid for lack of inventive step if the invention claimed in it was obvious to a person skilled in the art having regard to the state of the art at the priority date. The familiar structured approach to the assessment of allegations of obviousness first articulated by the Court of Appeal in Windsurfing International Inc v Tabur Marine (Great Britain) Ltd [1985] RPC 59 has recently been re-stated by Jacob LJ in Pozzoli v BDMO SA [2007] EWCA Civ 588, [2007] FSR 37 at [23] as follows:

In both H. Lundbeck A/S v Generics (UK) Ltd [2008] EWCA Civ 311, [2008] RPC 19 at [24] and Conor Medsystems Inc v Angiotech Pharmaceuticals Inc [2008] UKHL 49, [2008] RPC 28 at [42] Lord Hoffmann approved without qualification the following statement of principle by Kitchin J at first instance in the former case:

I was reminded by counsel for Armour of two points. First, what matters is whether or not the invention was technically obvious, not whether it was commercially obvious: see Hallen Co v Brabantia (UK) Ltd [1991] RPC 195 at 213. As counsel for LeisureTech pointed out, however, this does not necessarily mean that commercial considerations are irrelevant. The mindset of the skilled person may be conditioned by commercial considerations only to consider certain types of technical solutions, as in Dyson Appliances Ltd v Hoover Ltd [2001] EWCA Civ 1440, [2002] RPC 22.

Secondly, if a particular route is an obvious one to take, it is not rendered any less obvious from a technical point of view merely because there are a number, and perhaps a large number, of other obvious routes to take: see Brugger v Medic-Aid Ltd [1996] RPC 635 at 661.

I was reminded by counsel for LeisureTech of three points. First, an invention may perfectly validly arise from investigating the prejudices of the prior art: see Terrell on the Law of Patents (16th ed.) at paragraph 7-74, Dyson and Pozzoli at [24]-[29].

Second, simplicity of an invention is not the same as obviousness. The problem for the court in dealing with a simple invention is that, by definition it is extremely easy to understand once one knows it. In other words, one has to be particularly on guard against hindsight with a simple invention: see Terrell at paragraph 7-66.

Third, as Jacob LJ said in Technip at [122].

I would add that it is also necessary to bear in mind Fletcher Moulton LJ’s well-known warning in British Westinghouse Co Ltd v Braulik (1910) 27 RPC 209 at 230 (described by Jacob LJ in Technip at [112] as “as true today as when it was first said”):

In summary, Armour’s case is as follows. Armour posit a skilled person whose brief is to design a moderately priced system in which unobtrusiveness is more important than sound quality. Armour contend that, given such a brief, among the obvious choices open to the skilled person was a system with local amplification and central power using IC amplifiers and Cat.5 cable. Armour put their case in three ways, depending on the order in which the skilled person confronts the choices he has to make: namely, starting with the amplifier, with the cable and with the architecture.

Starting with the amplifier: Armour contend that an IC amplifier of the kind exemplified by the TA8211AH and TA8216H was an obvious choice for use in a multi-room audio system, particularly if one was more interested in unobtrusiveness and cost than sound quality. An IC amplifier would be naturally suited to being used as a local amplifier with central power. One would then need a multi-conductor cable to distribute both line level audio signals and DC power from the source to the amplifier, and Cat.5 would be an obvious choice for this.

Starting with the cable: Armour contend that Cat.5 cable would be an obvious choice for use in a multi-room audio system given the installers’ preference for it. Cat.5 cable is not suitable for use with central amplification since it will not carry speaker level audio signals, at least for any distance, but is suitable for carrying line level signals for local amplification. The skilled person then has to choose his amplifier and decide how to power it. An IC amplifier would be an obvious choice. It would then be obvious that the Cat.5 cable could carry both the audio signals and DC power for the amplifier.

Starting with the architecture: Armour contend that local amplification with central power was an obvious choice of architecture for a multi-room audio system. In choosing the local amplifier for such a system, an IC amplifier would be an obvious choice. One would then need a multi-conductor cable to carry both the audio signals and DC power for the amplifier, and Cat.5 cable would be an obvious choice.

I have listed the three ways in which Armour put their case in the order in which they were set out in Armour’s written closing submissions, but in his oral submissions counsel for Armour made it clear that the primary way was the third. I shall therefore concentrate on that one. I doubt that it makes a great deal of difference, however, which order one considers the choices in.

I have described these above.

The inventive concept is defined by the claim and not by some vague paraphrase based on the extent of the disclosure in the description: see Conor at [19]. In the present case, therefore, the inventive concept is to be found in claim 1 construed as stated above. Nevertheless, this is one of those cases where it is worth trying to express the concept somewhat more shortly. I would put it as follows: a distributed stereo audio system comprising a mains power supply and a signal source in one location and an amplifier and speakers in a remote location in which a single Cat.5 cable is used to carry the two audio channels and DC power from the source to the amplifier. Going further, the key feature of the inventive concept is the use of Cat.5 cable to carry both audio signals and DC power from the source to the amplifier.

Although claim 1 is not limited to systems comprising IC amplifiers, it undoubtedly embraces them. As can be seen, Armour’s attack is focussed upon such systems.

Armour’s case involves combining three elements, each of which was part of the common general knowledge, but had not been previously been combined, namely (i) local amplification with central power architecture, (ii) IC amplifiers of the kind discussed above and (iii) Cat.5 cable. It also involves use of Cat.5 in a way in which it had not previously been used, namely to transmit low level amplifier power as well as audio signals.

In my judgment the approach to the design of a distributed audio system outlined in paragraph 74 above was a technically obvious one for a person skilled in the art to adopt in August 1997. As can be seen, it involves three steps. The first is the choice of local amplification with central power as the architecture. The existence of this option was part of the skilled person’s common general knowledge. The second step is the selection of an IC amplifier of the kind discussed above. Again, the existence and properties of such amplifiers were part of the common general knowledge. IC amplifiers are particularly suitable for use in a local amplification system which is intended to be unobtrusive and moderately priced because they are very small, inexpensive and do not dissipate much heat. Since they only require low DC power, they can be centrally powered. Furthermore, as Mr Miller pointed out, it makes sense to use them in a centrally-powered system because otherwise the cost of the power supply will be disproportionate to the cost of the amplifier. Although counsel for LeisureTech submitted that neither of these two steps was obvious, the submission received little support from the evidence of either expert. I have no doubt that both steps were obvious.

The third step is the use of a single Cat.5 cable to carry both audio signals and low level power to the amplifiers. With slightly more hesitation than in the case of the first two steps, I have concluded that this step was obvious too. The existence and properties of Cat.5 cable, including its suitability to carry both audio signals and low DC power, were again part of the common general knowledge. Given in particular its cheapness and its popularity with the installers, I consider that Cat.5 was an obvious choice of cable for this purpose. As Mr Miller put it in his second report:

In support of its case that this step was not obvious, LeisureTech relied on an article by Bob Abraham of Armour dated 2 February 2005 which was published on a website called Hidden Wires. In the article Mr Abraham describes how Armour looked at using Cat.5 cable about three years before. Having explained the advantages of Cat.5 cable, Mr Abraham says:

I am not persuaded that this article demonstrates that it required invention to send both audio signals and low level DC power down a single Cat.5 cable, particularly if one was not looking for the best sound quality and was not using long lengths of cable. While it is true that Mr Abraham says that using Cat.5 cable “does present some serious technical challenges, and calls for some thinking outside the box”, in context the main problem he is referring to is how to get high sound quality while transmitting power over cables 50m long. Moreover, in order to achieve high sound quality, Armour took additional steps as well as using Cat.5 cable.

LeisureTech advanced a number of other arguments that the invention was not obvious which I should deal with. First, LeisureTech contended that the skilled person would have considered that amplifiers delivering at least 30W per channel were required for hi-fi sound quality and that this would have led the skilled person away from making the invention since a single Cat.5 cable could not deliver that amount of power. This was the primary reason given by Mr Colloms in his first report for saying that in his opinion the claimed invention was not obvious. I do not agree for the reasons discussed above. In short, while the skilled person might well have considered that amplifiers rated at 30W or more per channel were required for an audiophile system, the skilled person would have been well aware that lower-powered amplifiers could be used to produce acceptable sound quality.

Secondly, LeisureTech contended that the skilled person would have been put off using Cat.5 cable for carrying both audio and power because of concerns about interference. This was another reason given by Mr Colloms for saying that the invention was not obvious. Again I do not agree for the reasons given above. The power which needed to be carried was only up to 15W DC power, and Cat. 5 cable was well known to provide good immunity to interference between the pairs even at very high frequencies.

In this connection LeisureTech also relied on Mr Tam’s evidence that, when the system was demonstrated to him, he was very surprised that there was no interference. I do not find this persuasive for a number of reasons. First, as discussed above, Mr Tam was not representative of the skilled person. Secondly, as I read his evidence, he did not at that point appreciate that the Cat.5 cable was only carrying low DC power rather than 240V AC, although he was told this subsequently. Thirdly, Mr Tam was not told what amplifiers were used in the system and it is far from clear that he was aware of the existence and properties of IC amplifiers of the kind discussed above.

LeisureTech’s third argument was that the skilled person would also have been put off using a long run of Cat.5 cable to carry power to the amplifier because this would introduce a high source impedance which would adversely affect the amplifier. This was a further reason mentioned by Mr Colloms. Again I do not agree. The experts agreed that the effect of this would be similar to using a poorly regulated power supply. No doubt it would mean that high sound quality would be difficult to achieve, but it would not be a particular concern with a system that used IC amplifiers (which, as noted above, are tolerant of power supply variations) to deliver reasonable sound quality. Still less so if the system was a small one which did not involve long cable runs.

Fourthly, LeisureTech argued that Armour’s case of obviousness was based on hindsight. I have considered this argument with particular care. One must always try to avoid hindsight in considering an allegation of obviousness. As discussed above, this is particularly important with apparently simple inventions and even more so when considering an allegation of obviousness over common general knowledge. In the present case I do not believe my reasoning or conclusion is based on hindsight. As I have pointed out, the key step is the choice of Cat.5 cable to carry both audio signals and low DC power. For the reasons I have explained, I am satisfied that this was obvious.

Fifthly, LeisureTech pointed out that Armour’s case on obviousness requires a number of steps to be taken and argued that, even if each individual step was obvious, it did not follow that the sequence as a whole was obvious. LeisureTech reinforced this argument by putting out that each step involves choosing an untried option in preference to tried and tested ones. This is true at least so far as the second and third steps are concerned (the local amplification central power architecture had been used before in the MRS system, but that was not common general knowledge). Nevertheless, I cannot see that this means that invention was required. This is not a case where the skilled person would have needed to carry out an experiment to determine whether the system would work. At most, he might have to carry out a test to confirm his calculation that Cat.5 cable could carry 15W of DC power or to check that the sound quality of the system was acceptable.

Sixthly, LeisureTech suggested that there had been long-felt want for a multi-room system which was moderately priced, compact, easy to install and gave reasonably good sound quality. In my judgment the evidence simply does not establish this. I would add that LeisureTech pleaded and opened a case of commercial success, but abandoned it during the course of the trial. Moreover, only a subset of systems falling within claim 1 delivers all these advantages.

Seventhly, LeisureTech asked why, if it was obvious to make a system falling within claim 1, it had not been done before. Before answering this question, it is important to note that Mr Miller’s first report suggested that Linn had had the idea of such a system prior to August 1997, but following cross-examination it did not appear that this was in fact the case. Nevertheless, I am not persuaded that the fact that it had not been done before demonstrates that the claimed invention was not obvious. The market for multi-room audio systems only started to take off in 1988. Even by the mid 1990s there were relatively few players in the field. Although there was a market for moderately priced distributed systems, it does not appear to have been a large one: most of the systems available in the mid 1990s seem to have been relatively high end ones. Moreover, most of them were based on central amplification. Cat.5 cable was only specified in 1991 and the evidence does not establish that it had become popular much before June 1993. In these circumstances I do not find it surprising that no one marketed a system falling within claim 1 between then and August 1997.

Lastly, LeisureTech relied on the fact that Armour had not come up with the invention before August 1997, but rather had used Cat.5 cable to develop Systemline into a central amplification system. This does not persuade me that the invention was not obvious. No doubt Armour had both commercial and technical reasons to develop Systemline in the way that they did. That does not prove that the invention was not technically obvious. Indeed, I would go further. In my view, it is arguable that it would have been obvious in August 1997 to modify the local amplification Systemline system described by Mr Colloms to address the shortcomings he identified by substituting lower-powered amplifiers and using Cat.5 cable to carry both the audio signals and the power to the amplifiers. Since no such argument was advanced by Armour, however, I shall not consider this any further.

I have described the essential features of the Linn Knekt system above. Armour did not in the end rely on Linn Knekt as a separate starting point for its obviousness attack, but only relied on Linn Knekt in support of its case of obviousness over common general knowledge. Accordingly I need say no more about it.

MRS was a relatively obscure Australian system. It was a local amplification central power system in which a multi-conductor cable was used to carry audio signals from four sources and power to each room. A key feature of the system was the use of an unusual 40 conductor flat cable. Armour’s case with regard to MRS was not that obvious modification of the MRS would bring one within claim 1, but rather than MRS would teach the skilled person a concept which made claim 1 obvious, namely the concept of carrying both audio signals and amplifier power (and indeed control signals) in the same multi-conductor cable.

If I am right that claim 1 was obvious in the light of common general knowledge, then MRS adds nothing. If I am wrong about that, then I do not consider that claim 1 was obvious over MRS. The 40 conductor cable was an important feature of MRS which had both advantages and disadvantages. It was rather different to Cat.5 cable in a number of respects. In particular, it enabled power to be fed in at any point. As Armour accepts, it would not have been obvious to modify the MRS system to bring it within claim 1. That being so, it seems to me that Armour’s conceptual argument is one based on hindsight.

It is common ground that a single pair in a Cat.5 cable cannot transmit enough electrical power to drive an amplifier rated at 30W per channel over any significant distance, let alone more powerful amplifiers. Mr Colloms demonstrated this by means of a simple calculation which is exhibited to his first report. As noted above, Armour plead that the Patent is insufficient because claim 1 covers systems with amplifiers rated at 30W per channel and higher. LeisureTech’s answer to this is that the claim is inherently limited to systems with a power of the magnitude that a single Cat.5 cable can carry. As I have said, counsel for Armour did not press the argument that the claim extends beyond this or that the Patent was insufficient as a result. Rather, he submitted that on LeisureTech’s construction the claim was obvious. Since I have adopted that construction, the insufficiency plea falls away.

The Patent is invalid and must be revoked.