Radio frequency (RF) wireless power transfer (WPT) is a promising charging technology for future wireless systems. However, low end-to-end power transfer efficiency (PTE) is a critical challenge for practical implementations. One of the main inefficiency sources is the power consumption and loss of key components such as the high-power amplifier (HPA) and rectenna, which must be considered for PTE optimization. Herein, we investigate the power consumption of an RF-WPT system considering the emerging dynamic metasurface antenna (DMA) as the transmitter. Moreover, we incorporate the HPA and rectenna non-linearities and consider the Doherty HPA to reduce power consumption. We provide a mathematical framework to calculate each user's harvested power from multi-tone signal transmissions and the system power consumption. Then, the waveform and beamforming are designed using swarm-based intelligence to minimize power consumption while satisfying the users' energy harvesting (EH) requirements. Numerical results manifest that increasing the number of transmit tones enhances the performance in terms of coverage probability and power consumption since the HPAs operate below the saturation region in the simulation setup and the EH non-linearity is the dominant factor. Finally, our findings demonstrate that a properly shaped DMA may outperform a fully-digital antenna of the same size.