"Power requirements and thermal load to be determined by others."
I see that note on most consultant drawing sets I review, and it is fairly common on integrator drawings too. Calculating heat and power is one of the least glamorous parts of AV design, and nobody wants to own it.
On our drawing sets, we own it. Power requirements and heat loads are called out on the drawings, because we procure, assemble, and connect the gear and have the best information about it. Power is the easier half, since most manufacturers publish consumption. Heat is published less often, and when it is missing we calculate a worst-case estimate. Every device in the rack adds heat, including the DSPs, switches, encoders, and power supplies, not only the amplifiers. None of the math is hard. Integrators have to own this number. The consultant's note is understandable on a design-bid-build job, where the mechanical design is finished months before anyone buys the AV gear. The integrator's submittal still has to carry the final number, and the MEP still has to receive it.
Owning the number does not guarantee it reaches the mechanical design. Owners and MEP teams know the MDF has to stay cool, secure, clean, and powered. The AV rack room gets missed, and when it is cooled like an MDF, the design often assumes front-to-back airflow that much convection-cooled AV gear does not have. We are often told the cooling was sized correctly, and we have no way to verify that from our side of the drawings.
On one recent project, no additional cooling was provided for the AV equipment. The general contractor ended up cutting open walls to add ductwork and air handling. On another, the HVAC for our rack room was tied to the occupancy schedule of the room it served. During testing, with the room in use, the rack room was fine. At night it got extremely hot.
I now carry a small temperature logger in my bag. I leave it in a rack room for a week and come back for it, and it shows me what the room does when nobody is there. That occupancy schedule passed every test we ran, because every test ran during occupied hours. I still wonder how many rack rooms we have commissioned behave the same way after everyone goes home.