To produce synthetic mixtures for diving techniques several methods are available:

Each method has its pros and cons.

Unfortunately, the most accurate methods are the most expensive or the least practical, or both. 

The system by weight for example seems great, the perfect solution. It does not depend on the temperature, nor by the compressibility, nor by the interaction between the molecular gases, but simply by the management of the molar weight of the gases: two computations, a precision scale and there you go…

But how about when the weight of the whip and the torque  goes under pressure?
And in case of residual gas, what is the weight of the gas and that of the cylinder? Small errors in terms of weight mean big mistakes in percentage terms.

Mixing by volume is a technique where known volumes of each gas, calculated on their desired percentages in the final mixture, are delivered to a common gas chamber at atmospheric pressure and ambient temperature; the final mixture is then compressed into high-pressure cylinders.

At ambient pressure and temperature, the behavior of the gases is pretty close to the Ideal Gas model, thus the procedure is not affected (from the point of view of the composition) by the gas compressibility or by their molecular interactions. 

Although this method is conceptually simple, the equipment needed  is extremely bulky and very difficult to manage.

In short, there are good reasons for thinking that the method of partial pressures have many chances to still be the most widely used for many years.

The partial pressure method is certainly the most diffused technique in tech diving communities: it is simple and practical, and for those who have experience, it generates good results.