Let us now consider real sports surfaces. We will begin with surfaces for sports halls.

It had a positive effect in the practice of building new sports halls in Germany when in 1986 DIN 18032 introduced the distinction between various surface types according to their deformation characteristics. In the introductory part of the standard, the general performance effects are described as related to the use of the surface types:

DIN 18032 Part 2 Paragraph 3.1.2

The theoretical distinction and the information on the possible manifold of surface systems as resulting from theoretical considerations has encouraged the development of mixed-elastic and combined-elastic sports surfaces.

Area-elastic systems are typically produced as shown in fig. 5. They are called 'constructive swing floors' since their resilience is caused by the design of a sleeper system. It is inevitable that they are so-called 'average value' surfaces. This means that they meet the requirements of DIN 18032 by averaging the results of the different test locations. Since the resilience varies relative to the intersections of sleepers and the supporting pads, the uniformity of the mechanic properties across the surface is often insufficient. (From the table we can see the variances of the parameters = still to be drafted).

Figure 6 shows an extremely negative example of an aerea-elastic surface which is produced in Bavaria. This is to abuse of the gaps in the array of the DIN requirements. This example is worthy of mention since this surface is advertised with the slogan 'high tech' and is also accepted in the RAL quality monitoring program. The structure shows that the surface is simply misdesigned. Forces acting on this surface are diverted straight to the rigid pads and will not be cushioned by leading the load through sleepers to the underground as it should be with area-elastic surfaces. Consequently, it is typical for such surfaces that the resilience varies within the surface area between nearly nought and 3+ millimeters. Only by forming the average of all measuring locations/points will this type surface meet the DIN requirements.

This is a twofold violation of DIN 18032: according to the DIN, all points within the sports surface area shall meet the requirements and also the variation of the parameters within the surface area shall be small (this means: surface performance shall be as uniform as possible).

The misdesign is compounded by the sleeper alignment in continuous rows thereby butting the endings of the sleeper elements (about 4 m long) so that double stiff points are produced (see fig. 7).

Combined surface systems are created by installing a point-elastic surface over an area-elastic surface. By this, the advantages of both systems are combined and their disadvantages avoided. The advantages were recognized in Germany, and in Switzerland, these surface systems have been commercially successful.

A considerable improvement of the area-elastic surfaces is achieved through the development of the so-called Sandwich type surfaces. These consist, in principle, of a continuous elastic layer of synthetic foam and wooden or synthetic panels as load distribution slab. With this design, vibrations due to impacts of heavy balls or athletes are highly damped and compliance with DIN requirements is no problem. The effective mass is minimalized. This is achieved by reducing the extent of the deformation trough, the thickness of the load distribution slab and/or the use of light weight wooden panels. Furthermore, the elastic and damping properties of an elastic layer may be adjusted by combining appropriate materials. The uniformity and the damping characteristics can be observed in the recordings of the Artificial Athlete Stuttgart (see Fig. 8 - in prreparation).

The mixed-elastic type was developed in Germany. It is connected with the names of Mr. Höss and Mr. Wilms although others are challenging for patent rights. This surface type is available in different grades of area-elasticity. Depending on the type of use, one or the other variant is more suitable.

The key aspect in the design of mixed-elastic surfaces was: a small dynamically effective mass and a controlled size of the deformation trough. The difficulties to be mastered were mainly due to installation and material problems.

Thus, the development of sports hall surfaces has been so intensively influenced by the existence of DIN 18032 that it cannot be imagined without it. Now, it is necessary to improve the situation by effective application of the standard (i.e. no longer average surfaces). In this process, it should be noted that the existing requirement of the deformation trough size of combined-elastic surfaces of 5 % cannot with good reasons be justified. This was not discussed thoroughly enough when the DIN was prepared. There is no acceptable reason why the W500 value of such surfaces should follow a much tougher requirement than normal area-elastic surfaces (of course: no average compro-mise is applied).

That mixed-elastic surfaces systems function without problems is the result of systematic and uncompromised testing and quality monitoring (not necessarily based on the RAL program) which was carried out voluntarily by the appropriate companies.

At this point, let us refer to a common misunderstanding which has lead to unacceptable sports hall surfaces. In section 3.3.1, DIN 18032 specifies requirements on evenness. It defines tolerances depending on the distance of the 'measuring points'. The terms 'tolerances' and 'measuring points' are confusing and are interpreted arbitrarily in practice. As we understand it - and this was the intention when the standard was prepared -, the measuring point distance is the distance between adjacent peak and valley spots. The tolerance - although not absolutely correct - is the difference in height between adjacent peak and valley spots. When using a straight edge, the tolerance is equal to the gap underneath it. To define measuring point distance as the distance between two adjacent peaks of the surface profile would allow considerable unevenness which would be unacceptable according to the state of the art.

Let us now move to a special sector of sports surfaces: gym mats. Due to the large deformation of such surfaces, we recognize the dynamic behavior in usual video recordings. We clearly see how feet rebound upon landing. The often occurring uncertainties/instabili-ties of gymnasts are caused by too little damping (rebound of feet) and by the fact that the size of the deformation trough is too small in relationship to the magnitude of the deforma-tion. Change in this is unlikely since it is dependant upon the willingness of the manufacturers and representatives of gymnastic associations taking the driver's seat in the appropriate standardisation committee.

Synthetic Surfaces are used for Light Athletic Areas as well as for Team Sports Areas. For the latter in principle, the same considerations apply as for sports hall surfaces. However, the practical realisation is much more difficult because of the fact that they are exposed to outdoor conditions.

The most acute issue are the light athletics surfaces.

Again and again, the question arises: how can we produce an ultra fast track and what should it look like. This is a one-sided question since it refers exclusively to short distance races of high performance athletes. However, there is evidence that light athletics areas are used by students and leisure athletes less frequently: in any case, the frequency and duration of exercise by these persons on synthetic surfaces is much less than those of high performance athletes. Furthermore, the protective function of a synthetic surface is less important since

In order to answer the question on the 'fast surface', very often the rule is cited that - by mechanical reasons - a surface must be faster the harder it is. This rule is not completely wrong. However, it is questioned by what amount the running speed will be changed if the hardness is increased by a certain amount: will the performance significantly or noticeably improved (problem of significant, quantitative effect; sensational threshold).

Fortunately, the apparent contradiction of the protective function and the sports function is eliminated by biomechanical studies of McMahon and Greene. The authors let model athletes run on specialized track surfaces which could be varied in their give. This work was performed in connection with the construction of a tuned track at Harvard University. They found that the performance was not noticeably affected when the surface is rather soft/flexible compared with the stiffness of the mechanical system of the runner. However, this proposition does not exist with the surfaces actually available and used in athletics facilities. But the small improvement of elasticity (in the sense of 'softer') causes a reasonable reduction of the peak forces on the foot when hitting the surface - an advantage towards protective function. This reveals that a reckless hardening of the surface will only harm, not help in any way.

A more detailed consideration of the question 'what an optimal light athletic surface looks like' should take into account two typical stress situations of an athlete running over a track:

For an engineer, the conclusion drawn from this is that a light athletic surface must be designed with at least two layers.:

In any case, the surface must not be designed upside down: this means the upper layer must not be softer than the base layer.

In the USA, another proposal was made on how to improve the performance: forming the surface with an 'Encapsulated Texture" (Gormley Stadium in New Orleans). This is a surface texture similar to a sprayed plaster on walls. The texture is achieved by casting a PUR coating with an excess of EPDM granules. The surface is rough enough to hurt the athletes feet when only running. They will be seriously hurt when struggling and sliding over the surface with bare skin.

I have a sample of such an Encapsulated Texture here. The argument in favor of this texture is that - other than the granular textured surfaces - full performance will be available right after completion of the surface. With granular textured surface, you will have to wait for about a year until the loose surface granules causing the decrease in traction will have been removed. Regardless of this argument, granular textured surfaces can have their full performance after completion of the track if the surface is brushed with a cleaning machine removing all loosely bound granules.

The discussion with experts of the sports governing bodies revealed that such textures cannot be rejected according to the IAAF Performance Specifications. However, the implications should be noted in the assessment to indicate to the readers that liability problems will arise from this situation.

There are two methods available in sports surface technology which have been introduced internationally and shown to be reliable through experience: the Artificial Athlete Stuttgart and the Artificial Athlete Berlin. Both the devices are used from the USA to China. With them, we can test the suitability of athletic surfaces to a rather great extent (see presentation of Dr. Binder).

Beside these, other test methods are used operating with a rigid mass as a missile (Biomechanical Institut in Poitier and Biomechanical Laboratory of the Federal Technical University Zuerich). The performance of the surfaces is described by means of the characteristics Peak Force and Energy Return. Whereas the Peak Force is measured more or less directly by means of a force platform or acceleration pick-ups, the Energy Return must be calculated from the height of rebound after the impact.

Interpretation of these parameters is uncertain. Does an increase of Energy Return mean an advantage under all circumstances and at which quantitative amount does it affect the performance or sensations of the athletes. As far as I know, there is no persuasive answer to this question. Differences in certain characteristics which can be determined with test devices must not be presented as functionally effective without special evidence. An Energy Loss of 75 % (determined by means of a test device) must not be understood as the Energy Loss of an athlete.

There is no special need for the development of light athletic surfaces. The room for such development is limited by physical and functional reasons. It did not incidentally happen that the design or structure of light athletic surfaces has not greatly changed within the last 25 years. We have to acknowledge or admit that surface characeristics might be accessable to technical measuring and can be differentiated by that. However, this does not reflect the real effect on performance in any case. The decision between two very similar surfaces on superiority - in terms of performace/running speed - cannot be judged with any of the available test methods. The limits of technical testing must be recognized.

Thus, what are the factors influencing top athletic performance? Although I am not predestined to render a comment on this matter, I will formulate as follows:

Top athletic performance is achieved if

At this point, (although not actually belonging to the issue of my presentation) I would like to emphasize a very important problem of testing: precision of test methods. There are tests which are performed by a certain lab with the same equipment (precision is controlled by the term Repeatability) and there are tests performed on the same material by different labs (precision is controlled by the term Comparability). Both the aspects of precision have been neglected in the past. This is a vital issue within the European harmonisation process. At issue is not only the determination of the values for Repeatability and Comparability but also their application rules in practice. A typical case is when a surface which was tested as a master sample is retested after installation in a sports facility: what deviation is acceptable for compliance of that surface with the reference report?