This work deals with strategies of risk management techniques in projects and portfolios in the situation of radical innovation. Existing literature suggests different methods of risk management at the level of 1) projects (S1) (unknown reduction by selecting a priori the less uncertain projects, depending on the identified market and technological risk) 2) portfolio (S2) (consists in using an existing platform core to construct several options. This strategy increases chances to succeed by increasing the size of the sample, maximizing the total economic value of the portfolio of derivatives). These methods consider different level of uncertainties and are independent from each other. We will show that there exists another strategy (S3) of working on "common unknown" of multiple options but its managerial implementation is not obvious. By testing the proposed framework in two cases of Advanced R&D (explorative phase of new technologies development for unknown markets with fixed budget) in semiconductor industry, we compare identified S3 strategy with existing S1' lead by S2'. The paper demonstrates that management of "common unknown" is possible and could be implemented in the context of largely unknown exploration. The proposed strategy of working on common unknown opens a new way to portfolio risk management in the context of radical innovation. Using S3 framework of knowledge gap identification to construct common unknown core, company can build its innovative capabilities through knowledge management and better position to innovate in emerging fields.