Normally fixed crack models need crack tracking strategies for avoiding global stress locking, which is numerically more stable.

Rotating crack models on the other hand are more flexible, some of which can be tricky and very similar to classical smeared crack model (the crack can rotate 90 degree during loading process). I believe there are also some stable versions standing somewhere between rotating and fixed models, such as Cracking Elements method (unfortunately only 8 node quadrilateral element works by now).

Better try different models and try to develop some novel versions.

Some papers are recommended:

[1] 3D modelling of strong discontinuities in elastoplastic solids: fixed and rotating localization formulations

[2] Embedded crack vs. smeared crack models: a comparison of element wise discontinuous crack path approaches with emphasis on mesh bias

[3] Challenges, tools and applications of tracking algorithms in the numerical modelling of cracks in concrete and masonry structures

[4] Strong discontinuity embedded approach with standard SOS formulation: Element formulation, energy-based crack-tracking strategy, and validations

[5] Cracking elements: a self-propagating Strong Discontinuity embedded Approach for quasi-brittle fracture

[6] Geometrically non-linear and consistently linearized embedded strong discontinuity models for 3D problems with an application to the dissection analysis of soft biological tissues

[7] Smeared crack approach: back to the original track

It depends on your study case. Try to check both and compare between.

I think rotating smeared crack is better.

Smeared crack model is obviously has mor ability to simulate the real case. Please refer to this project. Regards

https://tu-dresden.de/zih/hochleistungsrechnen/hpc-monat?set_language=en

You may be interested in

Article Appraisement of planar, bending and twisting cracks in 3D wi...

Normally fixed crack models need crack tracking strategies for avoiding global stress locking, which is numerically more stable.

Rotating crack models on the other hand are more flexible, some of which can be tricky and very similar to classical smeared crack model (the crack can rotate 90 degree during loading process). I believe there are also some stable versions standing somewhere between rotating and fixed models, such as Cracking Elements method (unfortunately only 8 node quadrilateral element works by now).

Better try different models and try to develop some novel versions.

Some papers are recommended:

[1] 3D modelling of strong discontinuities in elastoplastic solids: fixed and rotating localization formulations

[2] Embedded crack vs. smeared crack models: a comparison of element wise discontinuous crack path approaches with emphasis on mesh bias

[3] Challenges, tools and applications of tracking algorithms in the numerical modelling of cracks in concrete and masonry structures

[4] Strong discontinuity embedded approach with standard SOS formulation: Element formulation, energy-based crack-tracking strategy, and validations

[5] Cracking elements: a self-propagating Strong Discontinuity embedded Approach for quasi-brittle fracture

[6] Geometrically non-linear and consistently linearized embedded strong discontinuity models for 3D problems with an application to the dissection analysis of soft biological tissues

[7] Smeared crack approach: back to the original track

I think the smeared crack model has the ability to simulate the real situation

I recommend Fixed crack model because it is easily to use and correct

Rotating smeared crack model is the more recent and advanced crack model. Recommend reading research articles by Prof. Hsu (Houston Univ) or Prof. Collins (Torronto Univ)