So while I wait for my RS3000 to arrive, I wanted to start an analysis thread of the iris pumping issue(s) on the RS640 I have. In short summary, iris pumping is a visual artifact when using the dynamic iris where the iris acts causing a visible artifact of either dimming or brightening in a distracting manner. There are times where this can be expected such as end credits roll or extremely black scenes like interstellar 53:30 starfield scenes. These are to be expected and not what is being discussed here.
What I am focusing on here is more of an anomaly where the iris pumping is visible when it should not be. This can be a scene where two folks are talking and the camera goes back and forth between them but one is slightly darker than the other. As the iris is not instantaneous like a laser dimmer, it can be very distracting.
For the purpose of this thread, I'm going to call this iris pumping. The pumping seen on credits etc is not what's being discussed so when I say "no iris pumping" it ignores those scenes.
A short history here. I run my projectors at 0 iris auto1
. This is the worst scenario for potential iris pumping because the projector has to react more aggressively to darker scenes. I owned an RS500 prior to my RS640. I never, ever noticed any iris pumping in this manner on my RS500. On my RS640, I notice it daily.
I originally thought perhaps the RS640 firmware was just more sensitive to APL changes and showed more artifacts in its iris operation. But there are some other factors that I think are worth noting and testing.
My RS500 viewing was mostly TV watching at 1080p playing recorded TV programs via plex. Eshift was off, source was PC at 1080p.
My RS640 viewing is mostly the same TV watching but at 4K with e-shift on, source at 4k, not running in plex, but instead running in MadVR with both upscaling and some post processing.
So, I went back to my old RS500 viewing habits tonight on my RS640. Tonight, I ran with e-shift off, 1080p desktop and playing via plex (no MadVR). There was no iris pumping at all. I tested 3 TV shows, one of which was marvels agents of shield which typically does trigger iris pumping. Another show was "Chicago Med" which often displays iris pumping as some of the camera views on faces can be dark. Usually iris pumping is visible within about 30 minutes of viewing. Tonight went over 2 hours and there was no pumping.
It's a bit too soon to call this a complete success. I need to repeat the test again tomorrow and see if there is still no iris pumping to be sure. I'll do so and report back.
If there's no iris pumping tomorrow, then next will be attempting to narrow down what specifically causes the iris pumping. The variables are:
1) 4K eshift playback. The theory here might be that the onboard firmware has a harder time analyzing 4K video as it's 4x bigger and therefore, does a poor job reacting to the content. To test this, we will continue to use plex player but set desktop to 4K and enable eshift. If the iris pumps, this is the likely culprit and more analysis can be done there. If not then onto #2
2) 1080p playback with MadVR and the same post processing. If this exhibits the behavior, then there's something in MadVR's processing that might be causing something and it will need to be tested with various settings in MadVR. I know that others have complained about iris pumping that don't use MadVR so it'll be interesting if this is the case and if we can track it down to something specific.
Another theory if 4K eshift playback is what's required to exhibit the iris pumping, since it always requires about 30 minutes of playback before this begins, perhaps the projector gets very warm from the CPU analyzing each frame and this somehow impacts the behavior. It is odd that (at least on my end) I can never reproduce any pumping until after about the 30 minute mark.
I plan to update this thread with observations and see if we can come to any conclusions.
Some other observations. While running the RS640 at 4K eshift with MadVR and iris pumping, I have noticed that the more you clamp down the iris the less you notice iris pumping. I assume this is because, for example, if the projector is set at iris -8 and auto iris, it only activates the iris when it needs to clamp lower than -8. Where if you set the iris to 0 and auto, it will activate the iris anytime it needs to clamp lower than 0.