We use MR spectroscopic fingerprinting (MRSF) to quantify T1,T2 and concentration addressing the tradeoff between fingerprint lengths and averaging. Methods. MRSF using 25, 50 and 100 fingerprint lengths were compered to inversion recovery (IR) and multi-TE using Monte-Carlo simulations and in-vivo experiments. Bias and variance were estimated for NAA, Creatine and Choline. Results. Simulations of all MRSF sequences show better accuracy and bias over IR. In-vivo experiments show improved T1 and concentration estimation. Conclusion. The low SNR emphasizes the tradeoff between fingerprint length and averaging. The In-vivo results show clear advantage using shorter fingerprint and increasing the SNR.