There are limits to what can be achieved with silver halide chemistry, so that an endless march up the ISO hill is not possible. On the other hand, using a single emulsion and rating the EI higher changes the contrast, and at some point shadow detail cannot be recorded any longer, even with infinite developing time. The other major complicating factor is that the size and nature of the camera changes temendously when moving up through the format sizes. Not that this is relevant to your question directly, but what it means is that camera format choice usually is not driven by the grain vs film speed compromise (unlike the digital FF vs APS compromise). It is not only the quantity of grain but also the appearance that is visually important. With different development, the same grains can be made to look sharp and crisp, or bloated and blotchy. The film's ability to record tonality is not directly related to its grain character, which is why "traditional" grain films such as FP4+ and TriX are so popular, despite not matching T-grain Acros and TMax400 purely from a clinical grain comparison. The materials necessary to do a complete comparison do not really exist, and the motivation to do so not really either. We accept that grain will be less intrusive from larger negatives than 35 mm, and so might be willing to use faster films with fewer reservations, but it has little bearing on the photographic purpose. I doubt a very fast film on 4x5 will have the same tonal character as a very slow film on 35 mm. The look will be different, no doubt. In that sense, the 35 mm with slow film may have more grain, but better tonality than the 4x5 with fast film, when printed to the same output size.
Originally Posted by ckpj99
Obviously a major advantage of large format is the ability to develop frames individually, so that the tonality is handled optimally for a given scene. It is often this difference, rather than grain, which is most visible in the end result. Up to ISO 400, most films are capable of recording excellent tonality, but beyond that it becomes progressively more difficult for the silver halide crystals to cope with light starvation, and something has to give. So I would say that using HP5+ in 4x5 will give superior results to using PanF+ in 35 mm, on the basis that grain will be magnified less, with both films capable of handling the tonality well. OTOH, Delta 3200 in 4x5 will give a visually apparent difference in tonality to Delta 400 in 35 mm, when exposed at box speed and printed to the same effective exposure and contrast. That said, a good printer will be able to make them look very similar indeed, if they were exposed and developed acceptably.