Over the past few days, I have been experimenting with samples of rolled aluminum rustoleum on some different surfaces such as foamcore board, parkland (smooth side) and parkland (rough side). I tried both the rustoleum regular aluminum and the rustoleum high performance enamel. No difference was observable between the two paints. Samples varied from 5-10 sq. ft. in size. I also tried experimenting with the application technique to see if any particular approach resulted in an improved finish. All applications were done with 5" foam rollers.
In past projects using metallic spray paints, I have found consistency was very much an issue, so for this experiment I chose to roll the paint on in the hope that the results would improve and I wouldn't have to worry about cleaning up overspray.
When applied to a textured surface such as the rough side of a parkland sheet (blackout cloth probably gives similar results unless heavily primed), the resulting image is bright but very textured. It is also more difficult to get an even application as some of the paint tends to want to cling to the low areas. The projected image is bright and somewhat watchable, but there is clearly too much sparkle/texture. Lesson learned: the noticability of substrate texture is amplified by high gain paint; texture=bad. Personally, I prefer a typical white wall.
When applied to foamcore board, the solvent in the paint dissolved the adhesive used in the board as it soaked in, caused the card stock outer layer to delaminate from the styrofoam core. Don't go there. This comment does not apply to gatorboard which I understand is nonporous and should not exhibit this problem.
When applied to the smooth side of a sheet of parkland, I learned that a very uniform application could be achieved by rolling a thin layer of paint on with a foam roller. Paint with the screen lying horizontal (on the floor or on trusses). After the initial paint application, continue running the roller over the paint to smooth out any areas where the paint application is thicker or thinner, taking care to have only the weight of the roller pressing down. If you push down on the roller, this tends to cause unevenness. I found that consistently rolling away from my body gave good results. Do this for a few minutes (~ five or six passes). While the paint is wet, you will see roller marks, but as it dries, they tend to fade. If you have a thin evenly applied coat, they will fade completely. If the coat is too thick it will tend to show inconsistency and roller marks. If the application surface is not cleaned beforehand, and if you have any airborne dust in the room, you will see debris in the paint. All that dust that you normally see floating in front of your projector lens is now embedded in your screen.
Even with a smooth, consistent application, there is still some sparkle to the aluminum paint. The sparkle/texture is nowhere near the severity of the rough substrate case, but is still somewhat annoying to my eyes. For those who have not seen it, I would liken it to the effect you get when you moisten your finger and touch the screen of your monitor, although much less severe (the light does not separate into coherent colors) and scattered evenly over the entire screen. It is most noticeable when viewing whites.
Although the aluminum paint clearly provides characteristics that are desirable to some users, I personally find the remaining sparkle to be a bit much. Also I find that in darker scenes with less contrast visible, the image washes out badly (I am using a 400:1 contrast LCD PJ in a mostly white room). I echo the statement made in past threads that the higher gain image makes artifacts and source imperfections more noticeable. In the end, I will probably go with a 'high power' with a filter or buy the HCCV material and DIY a frame for it. The layered technique discussed by Ddog in one of the other threads would probably help resolve these issues, although I am losing my ambition after breathing all these paint fumes :o
This was definitely a worthwhile experiment and although I do not plan to use aluminum paint for a DIY screen, I found that this exercise very educational and encourage others who have been fence sitting like me to take a couple hours to try something similar. If nothing else, it gives some useful insight into how contrast and light levels are perceived. Don't expect this to be your 'final' screen, but it may help you understand what characteristics you want to have in your 'final' screen.
EDIT: after looking at this again a few days later, I think it is more appropriate to characterize the finish of the aluminum rustoleum on a flat surface as having a high sheen, rather than 'sparkles' or 'texture'. The word 'texture' implies an unevenness which is not present.
One additional observation. The surface finish of the aluminum paint is very fragile. Simple handling tends to mar the finish, leaving noticable blemishes in the projected image. I observed this when taking down my samples. Anyone with small children or a pet that takes an interest in their new screen would likely encounter problems with this finish.