The 4 pic matrix results are in 2048 x 1347 px images so we can examine quality.

They have not been processed in any way.

3 prompts:

into the abyss, in the style of salvadore dali --ar 3:2

All of the reults are okay, and on topic, but rather flat.

***

89 Prompts:

Let's try the opposite extreme - 89 prompts (86 copied from a real request):

"into the abyss, in the style of salvadore dali, sharp focus, concept art, extreme details, full shot, Full Body shot, Cinematic, Color Grading, Photography, Depth of Field, hyper - detailed, beautifully color - coded, insane details, intricate details, beautifully color graded, Unreal Engine, Cinematic, Color Grading, Editorial, Shot on 70mm lens, Depth of Field, DOF, Shutter Speed 1/ 1000, F/ 2, White Balance, 32k, Super - Resolution, Megapixel, Pro Photo GB, VR, Lonely, Good, Massive, Half rear Lighting, Backlight, Natural Lighting, Incandescent, Optical Fiber, Moody Lighting, Cinematic Lighting, Studio Lighting, Soft Lighting, Volumetric, Contre - Jour, Beautiful Lighting, Accent Lighting, Global Illumination, Screen Space Global Illumination, Ray Tracing Global Illumination, Optics, Scattering, Glowing, Shadows, Rough, Shimmering, Ray Tracing Reflections, Lumen Reflections, Screen Space Reflections, Diffraction Grading, Chromatic Aberration, GB Displacement, Scan Lines, R a y Traced, Ray Tracing Ambient Occlusion, Anti - Aliasing, Shaders, OpenGL - Shaders, GLSL - Shaders, Post Processing, Post - Production, Cell Shading, Tone Mapping, painted, concept art, insanely detailed and intricate, hyper maximalist, elegant, super detailed, dynamic pose, volumetric, ultra photoreal, ultra - detailed, super detailed, full color, ambient occlusion, volumetric lighting, high contrast --ar 3:2"

Despite the huge number of prompts, the images are no better, and, in fact, are not as good as the quality or detail of picture #2 in the first set.

It appears that in the confusion of a massive number of prompts, some of which might be contradictory, Midjourney basically ignored all of them.

***

5 Prompts

Let's back up and start adding prompts slowly.

To the first request we will select "intricate details", and "cinematic lighting" as additions to get the following request with 5 prompts -

"into the abyss, in the style of salvadore dali, intricate details, cinematic lighting --ar 3:2"

Cleared of all the clutter, Midjourney focussed on the two additional prompts, "intricate details", and "cinematic lighting" and the results are improved greatly.

***

7 Prompts:

For the next set, let's add 2 more prompts, again, selected from those in the long list of 89 prompts. We will add "ultra photoreal", and "beautifully color graded" to get the following request -

" into the abyss, in the style of salvadore dali, intricate details, cinematic lighting, ultra photoreal, beautifully color graded --ar 3:2"

This seems to have had a good effect as well, and, for the first time, we are seeing a color other than the brown and blue used in all images so far. Detail is very good, and Midjourney is handling the 7 prompts well.

***

9 Prompts:

To our last request we will add "volumetric", and "dynamic pose", resulting in this request -

"into the abyss, in the style of salvadore dali, intricate details, cinematic lighting, ultra photoreal, beautifully color graded, volumetric, dynamic pose --ar 3:2"

Three of the four images certainly have dynamic poses, but the overall quality has suffered greatly compared to the previous set.

It appears that "intricate details", "cinematic lighting" and "beautifully color graded" are now being ignored, and the images are flatter, with much less definition.

***

This short experiment, which I would invite others to repeat and have their results published here, would suggest that it is best to stick with about 5 to 7 well-chosen prompts and modifiers in a request, possibly weighting those that are most imporant to you.

You are welcome to send your results, or any observations, to the webmaster.