Thanks for continually improving your testing methods.

Low APL contrast is so important for projector viewing in a dark room. The 1% APL method is much more effective than just a simple FOFO measurement due to manufacturers trickery with laser dimming etc.

The way you guys measure contrast, off the screen in a black painted room, is substantially impacting the usefulness of your results.

The easiest way to see how poor your results are is by looking at the ANSI contrast measurements. The highest you have yet to measure is 164:1. This is not a reflection of the projector you have reviewed but the way your room handles light. Your room has a ceiling for ANSI and you will never measure a projector’s true potential in that room. You could put a 1000:1 ANSI Barco or Christie in there and it will most likely be capped at 200-250:1.

If your readers choose to use an ALR screen, or treat their room better than you they will not realize that they are capable of getting more performance from going with a higher ANSI projector. If a 500:1 ANSI projector measures at 160:1 and a 250:1 ANSI projector measures at 135:1 will they conclude that they could see double the performance in a properly treated room? (that’s rhetorical).

I also wonder why you have devised your own ADL patterns and measure them differently than much of the industry. Projection Dream made huge strides in figuring out the usefulness of these patterns and created ADL patterns that operate differently than yours, creating a solid white area in the center of the screen and then dispersing that percentage of white throughout the frame when measuring black. This pattern makes it ideal for measuring off of the lens through a velvet tube - the gold standard way for measuring a projector’s raw performances and removing the unique environment in the room.

I wonder if your methods are actually reflected in real world performance delta’s. The Epson 5050 you measured came in at 1780:1 at 1% which is nearly half of what others, including myself, have measured. Yet a Nexigo and Formovie are around 2500:1. These USTs are in the 3000:1 to 3500:1 native range while the Epson 5050 is usually in the 4000:1 to 5000:1 native range. This means one of two things, you got a really bad Epson 5050 to review or your patterns result in the Epson 5050 not being able to hold peak white in those small circles for some reason. Does that bear out in a 1% ADL real world image? 1780:1 vs 2500:1 should be noticeable when side by side. I’d be curious to hear if you see those differences post a good gamma calibration in a real world image.

Furthermore, your comments about native or FOFO contrast are inaccurate. Projection Dream did an analysis of 50 movies of a variety of content and found that 12% of frames are below 1% ADL at 2.2 gamma. This number will increase at 2.4 gamma. Since you only measure down to 1% ADL, you do not speak to this content. Also, it is very easy to counteract the “undefeatable” dimming on some projectors. All you have to do is put 1 or 4 individual pixels in the corners of the frame and you can measure FOFO. I’d advise measuring off the lens though or with high contrast projectors you may have a bad time measuring off the screen. This FOFO number charted down through 1% ADL will tell you how a projector will handle starfields, basement scenes, night scenes, etc ect. There’s PLENTY of examples of scenes in the 0.5% range https://projectiondream.com/en/movie-brightness-adl-contrast-measurements/

FWIW, I’m not a nobody. I also review projectors and perform ADL measurements while doing so. I help put on a projector comparison every year and have extensive experience with projectors, screens, and the way they interact with each other and the environment.

Sep 27, 2024 10:04 AM By RtingsUser1708119

20

The way you guys measure contrast, off the screen in a black painted room, is substantially impacting the usefulness of your results. The easiest way to see how poor your results are is by looking at the ANSI contrast measurements. The highest you have yet to measure is 164:1. This is not a reflection of the projector you have reviewed but the way your room handles light. Your room has a ceiling for ANSI and you will never measure a projector’s true potential in that room. You could put a 1000:1 ANSI Barco or Christie in there and it will most likely be capped at 200-250:1. If your readers choose to use an ALR screen, or treat their room better than you they will not realize that they are capable of getting more performance from going with a higher ANSI projector. If a 500:1 ANSI projector measures at 160:1 and a 250:1 ANSI projector measures at 135:1 will they conclude that they could see double the performance in a properly treated room? (that’s rhetorical). I also wonder why you have devised your own ADL patterns and measure them differently than much of the industry. Projection Dream made huge strides in figuring out the usefulness of these patterns and created ADL patterns that operate differently than yours, creating a solid white area in the center of the screen and then dispersing that percentage of white throughout the frame when measuring black. This pattern makes it ideal for measuring off of the lens through a velvet tube - the gold standard way for measuring a projector’s raw performances and removing the unique environment in the room. I wonder if your methods are actually reflected in real world performance delta’s. The Epson 5050 you measured came in at 1780:1 at 1% which is nearly half of what others, including myself, have measured. Yet a Nexigo and Formovie are around 2500:1. These USTs are in the 3000:1 to 3500:1 native range while the Epson 5050 is usually in the 4000:1 to 5000:1 native range. This means one of two things, you got a really bad Epson 5050 to review or your patterns result in the Epson 5050 not being able to hold peak white in those small circles for some reason. Does that bear out in a 1% ADL real world image? 1780:1 vs 2500:1 should be noticeable when side by side. I’d be curious to hear if you see those differences post a good gamma calibration in a real world image. Furthermore, your comments about native or FOFO contrast are inaccurate. Projection Dream did an analysis of 50 movies of a variety of content and found that 12% of frames are below 1% ADL at 2.2 gamma. This number will increase at 2.4 gamma. Since you only measure down to 1% ADL, you do not speak to this content. Also, it is very easy to counteract the “undefeatable” dimming on some projectors. All you have to do is put 1 or 4 individual pixels in the corners of the frame and you can measure FOFO. I’d advise measuring off the lens though or with high contrast projectors you may have a bad time measuring off the screen. This FOFO number charted down through 1% ADL will tell you how a projector will handle starfields, basement scenes, night scenes, etc ect. There’s PLENTY of examples of scenes in the 0.5% range https://projectiondream.com/en/movie-brightness-adl-contrast-measurements/ FWIW, I’m not a nobody. I also review projectors and perform ADL measurements while doing so. I help put on a projector comparison every year and have extensive experience with projectors, screens, and the way they interact with each other and the environment.

Hi RtingsUser1708119,

Thanks for taking the time to reach out with your feedback! We’re happy to discuss our contrast testing philosophy & methodology.

Before diving into the matter, it’s important to press that we do not measure ANSI contrast, which must be carried out in a laboratory with specialized equipment to obtain the values stated by manufacturers.

While we could have developed a methodology for this, we chose the route of measuring contrast directly off-screen as this is what is experienced by people at home, as light inevitably makes it way back onto dark areas of the projector screen due to reflections within the viewing environment. While it’s true that our results differ from the ANSI contrast values stated by manufacturers due to real-world environmental influences, we believe this provides a more practical representation of what users can expect in optimized home theater setups.

Our contrast testing patterns were made specifically to simulate a checkerboard at various APLs and to work seamlessly with the 1° measuring angle of our Konica-Minolta LS-100 luminance meter used to carry out contrast measurements. We confirm that the Epson 5050UB held its white luminance as expected throughout all APL test patterns, with an average variation of ~2% between peak all the white measurements at the same location, which falls within the accuracy & repeatability specifications of the LS-100. The differences in measured contrast values between reviewers can be due to differences in testing environments, methodologies, instrumentation and projector settings. Our measurements are done in our recommended SDR settings, post-calibration.

Regarding FOFO contrast, it is defined as the ratio between a completely white image and a completely black image, without any other elements present. Adding individual pixels would no longer fit the definition of FOFO, but instead measure very low APL contrast, well below our current minimum value of 1%, and you’re absolutely correct about many scenes being below 1% ADL. Unfortunately, we needed to limit the scope of tested APL values due to testing resources on our end. The good news is that Contrast Ratios below 1% APL will always be greater than the values we measure at 1% APL.

I’ve added a note to revisit the idea of extending our testing to include ADL values below 1%. We understand that this range provides valuable information about the performance of projectors in dark scenes, and we’ll evaluate how to incorporate this in future updates to our test methodology.

Thanks again for reaching out, and feel free to share any other suggestions or questions you have.

Cheers!

Oct 02, 2024 07:52 AM By  Pascal Messiaen

Hi RtingsUser1708119, Thanks for taking the time to reach out with your feedback! We’re happy to discuss our contrast testing philosophy & methodology. Before diving into the matter, it’s important to press that we do not measure ANSI contrast, which must be carried out in a laboratory with specialized equipment to obtain the values stated by manufacturers. While we could have developed a methodology for this, we chose the route of measuring contrast directly off-screen as this is what is experienced by people at home, as light inevitably makes it way back onto dark areas of the projector screen due to reflections within the viewing environment. While it’s true that our results differ from the ANSI contrast values stated by manufacturers due to real-world environmental influences, we believe this provides a more practical representation of what users can expect in optimized home theater setups. Our contrast testing patterns were made specifically to simulate a checkerboard at various APLs and to work seamlessly with the 1° measuring angle of our Konica-Minolta LS-100 luminance meter used to carry out contrast measurements. We confirm that the Epson 5050UB held its white luminance as expected throughout all APL test patterns, with an average variation of ~2% between peak all the white measurements at the same location, which falls within the accuracy & repeatability specifications of the LS-100. The differences in measured contrast values between reviewers can be due to differences in testing environments, methodologies, instrumentation and projector settings. Our measurements are done in our recommended SDR settings, post-calibration. Regarding FOFO contrast, it is defined as the ratio between a completely white image and a completely black image, without any other elements present. Adding individual pixels would no longer fit the definition of FOFO, but instead measure very low APL contrast, well below our current minimum value of 1%, and you’re absolutely correct about many scenes being below 1% ADL. Unfortunately, we needed to limit the scope of tested APL values due to testing resources on our end. The good news is that Contrast Ratios below 1% APL will always be greater than the values we measure at 1% APL. I’ve added a note to revisit the idea of extending our testing to include ADL values below 1%. We understand that this range provides valuable information about the performance of projectors in dark scenes, and we’ll evaluate how to incorporate this in future updates to our test methodology. Thanks again for reaching out, and feel free to share any other suggestions or questions you have. Cheers!

Thanks for the reply!

Regarding ANSI - sorry, terminology mishap there. You are not strictly measuring ANSI as in the spec but you are measuring 50% ADL which is what ANSI measures as well. My comments regarding the room/screen combo having a pretty big impact on the results stands.

I really have to disagree with your decision to measure off the screen and the purpose of that being to mimic “optimized theater environments.” As I mentioned in the previous post there are multiple ways to optimize your setup further without much work. Using an ALR or grey screen would impact your ADL measurements and many readers will be buying those screens. All of your UST measurements will be understating the performance if used with a lenticular or especially fresnel screen.

I’d expect Rtings to set up a lab-type of environment and not be victim to the impacts of the environment. Even just measuring off the lens would remove a lot of the environment. There’s just no way to know if your room is a good match to your readers. It could be better, it could be worse. The way your room reflects light could be different for a UST than a long throw based on how each type disperses light into the room. The reason we as reviewers typically want to measure in the most optimal conditions is that is how you level the playing field and tell the reader what is absolutely possible. If a reader has a near perfect room with black velvet then your 50% ADL measurements make it seem like there is very little difference between many projectors - which is just not true.

Back to the lap-type of environment. It’s not hard to set this up, you can do it in a couple different ways. Either use a velvet lined tube or a velvet tent. The tent may be the easiest for alignment with various types of projectors and could be made out of PVC and triple black velvet from a hobby store. You could also just get velvet for your whole room. Or, getting a velvet cover for the screen that can drop down would substantially cut down on light pollution in the room and get you near ideal measurements.

FOFO contrast - if you measured FOFO on a projector that doesn’t automatically dim for a full black field with and without a pixel in the corner(s) you’d find the measured contrast would be either identical or within a few tenths of a percent of each other. FOFO with a few pixels is FOFO. If you don’t want to call it FOFO then call it “0% ADL - 1px”. Doing this will speak to the performance of under 1% ADL.

What I believe would be awesome for you guys to do is to instruct readers on how a projector’s contrast is influenced by the environment and screen. Projection Dream did this in a fantastic study but it is only known to enthusiasts in the community. You guys have a platform to show what a projector’s ideal contrast is, how it performs in a living room, with a grey screen, and ALR screen, a lenticular screen etc etc (for those projectors that work with such screens, of course). Projection Dream documented this with ADL graphs showing the impacts of the environment. If you made a video and an article educating the readers then they could make much more informed decisions. If they have a white living room they could see how a white screen would essentially make ADL contrast differences null. If they had a perfect black velvet room they could see the inverse.

There’s no one better than you guys to take a scientific approach to this and really help members of the community understand the relationship of room vs projector vs screen.

https://projectiondream.com/en/contrast-projector-environment/

Oct 09, 2024 10:05 AM By RtingsUser1708119

20

Thanks for the reply! Regarding ANSI - sorry, terminology mishap there. You are not strictly measuring ANSI as in the spec but you are measuring 50% ADL which is what ANSI measures as well. My comments regarding the room/screen combo having a pretty big impact on the results stands. I really have to disagree with your decision to measure off the screen and the purpose of that being to mimic “optimized theater environments.” As I mentioned in the previous post there are multiple ways to optimize your setup further without much work. Using an ALR or grey screen would impact your ADL measurements and many readers will be buying those screens. All of your UST measurements will be understating the performance if used with a lenticular or especially fresnel screen. I’d expect Rtings to set up a lab-type of environment and not be victim to the impacts of the environment. Even just measuring off the lens would remove a lot of the environment. There’s just no way to know if your room is a good match to your readers. It could be better, it could be worse. The way your room reflects light could be different for a UST than a long throw based on how each type disperses light into the room. The reason we as reviewers typically want to measure in the most optimal conditions is that is how you level the playing field and tell the reader what is absolutely possible. If a reader has a near perfect room with black velvet then your 50% ADL measurements make it seem like there is very little difference between many projectors - which is just not true. Back to the lap-type of environment. It’s not hard to set this up, you can do it in a couple different ways. Either use a velvet lined tube or a velvet tent. The tent may be the easiest for alignment with various types of projectors and could be made out of PVC and triple black velvet from a hobby store. You could also just get velvet for your whole room. Or, getting a velvet cover for the screen that can drop down would substantially cut down on light pollution in the room and get you near ideal measurements. FOFO contrast - if you measured FOFO on a projector that doesn’t automatically dim for a full black field with and without a pixel in the corner(s) you’d find the measured contrast would be either identical or within a few tenths of a percent of each other. FOFO with a few pixels is FOFO. If you don’t want to call it FOFO then call it “0% ADL - 1px”. Doing this will speak to the performance of under 1% ADL. What I believe would be awesome for you guys to do is to instruct readers on how a projector’s contrast is influenced by the environment and screen. Projection Dream did this in a fantastic study but it is only known to enthusiasts in the community. You guys have a platform to show what a projector’s ideal contrast is, how it performs in a living room, with a grey screen, and ALR screen, a lenticular screen etc etc (for those projectors that work with such screens, of course). Projection Dream documented this with ADL graphs showing the impacts of the environment. If you made a video and an article educating the readers then they could make much more informed decisions. If they have a white living room they could see how a white screen would essentially make ADL contrast differences null. If they had a perfect black velvet room they could see the inverse. There’s no one better than you guys to take a scientific approach to this and really help members of the community understand the relationship of room vs projector vs screen. https://projectiondream.com/en/contrast-projector-environment/

Hi RtingsUser1708119,

While our current projector test bench is limited in scope, we are looking to expand it and improve our tests going forward, so we really appreciate you taking the time to share all your thoughts and insights with us about how we could improve our testing environment and contrast tests.

It’s given us a lot to consider going forward, and while we can make any guarantees of a timeline or scope of changes to our current projector test bench, we’ll be keeping all of this in mind when we get back to working on our projector test bench.

Cheers!