CurtPalme.com Home Theater Forum

[radar]

Often we choose projectors based on a few factors, one of which is sharpness. We are told that the high end units (9") are much sharper than 7" units. The question is (if you've never seen a 9" unit), how sharp is sharp?

Theoretically, a projector that can handle a 1920x1080 source (say a PC or Blu-ray) should see a 1 pixel line (running from top to bottom) of an approximate width of 1mm on a 7' wide 16:9 screen. Alternately, a projector set up for 1024x768 should see a 1 pixel line of approximately 2mm on a 7' wide screen.

So... if I have a 7" unit or a 8" unit or a 9" unit, how wide would that line be? And could that be used to be a benchmark to compare sharpness of a projector?

Also, if I knew that a stock Marquee 8" projector should throw a 1 pixel line and be around 3mm thick (under the same conditions as above on a 1024x768 setup on a 7' screen), then I could compare my setup to the benchmark and tune accordingly. I would also be able to tell if my tubes are getting weak or not performing.

Is there another way to benchmark sharpeness so we can compare our pj's against them?

Any thoughts if we should add a new pj list with regards to sharpness? If we add a new lense to a pj, what value does this change the sharpness by? If we have a benchmark, perhaps we can say "projector A is 33% sharper than the benchmark". So if we knew that the benchmark was 3mm (1 pixel width on a 1024x768 setup on a 7' screen) then projector A should show a 2mm pixel width under the same conditions. If we do go with this idea, what should the benchmark be (pj, screen size and source)? I'm thinking for screen size - 7' and source should be a PC with 1024x768 (in 4:3) and 1920x1080 (in 16:9)... don't know what would make a good pj benchmark.

Cheers,

Kevin.

[stuffandpuff]

Any digital can be your benchmark. A brand new 7" ES tube with good lenses is probably as sharp as an 8" EM tube. At least this seems to be my experience. Problem is, ES tubes don't stay sharp for very long. I would say the NEC PGs I've owned are as sharp as any CRT projector I've owned or saw. I've never seen a 9" so I don't know how it would compare. I know none of this address your questions but it's all I got.

[Person99]

It is very difficult to benchmark sharpness. A CRT does not handle 1 on/1 off patterns well due to the transition. So, it is not really an accurate representation of the PJs sharpness, but can be used for relative comparisons (i.e. most 8" machines will not show a 1-on/1-off 1080p vertical bar pattern whereas many 9" machines will. Also, how well a PJ does this is how "white" the white lines are and how "black" the black lines are. No CRT will do them correctly as white and black lines both due to the nature of them and the very low ANSI CR. So, this is difficult to measure to see which PJ is doing it better.

Additionally, set up skill is a significant factor in CRT sharpness. A "sharper" PJ can look less sharp if not as well set up.

Lastly, the condition of the electronics in the PJ as well as tube condition play a role, as noise in the video chain, etc introduce resolution diminishing artifacts.

In the end, we pretty much use our eyeballs. A lot of the people here have seen most of the major PJs in action. Few side by sides in identical new condition, but many general impressions. We pretty much go on that to set up the "rules of thumb".

[Gary M.]

it is tube dependent, PJ dependent(quality of controls over tubes) and setup ability dependent

impossible to benchmark IMHO

-Gary

[Brian Hampton]

If the scan lines of 1080i remind you too much of "video" so you use 1080p and the image is still sharp... then it's good.

-Brian

[tse]

MTF has been a measurement of resolution for many years. A quick explanation:

http://archive2.avsforum.com/avs-vb/showthread.php?t=633597

Keep in mind the lines are "gaussian" in profile. As they are placed closer and closer together they will overlap and the light at the base will add to the light from the next line. What should be a "black" line will be gray. The less black it is means a lower MTF. It's a way to measure resolution.

Another post showing a way to measure MTF with a camera and oscope.

http://archive2.avsforum.com/avs-vb/showthread.php?t=584081

Scott

[Nashou66]

[tse]

[/quote]

Now that I have a scope I really want to try this, I wonder if it will work with my PAS digital cam in Video mode?

Athanasios[/quote]


You need an analog video signal to connect to the scope. The camera needs to focus close to the screen so not too many lines are displayed.

Scott

[radar]

Person99: Yes, I agree that setup skills is a significant factor. How does a newbie know what to aim for if they don't have a simple benchmark? ie... "I'm aiming for 2mm thick line at 1 pixel wide - easily measureable".... or would the newbie think... "hmmm looks ok" but does not maximise his/her value out of their asset.

TSE: That is a great way to measure resolution! So if I had a scope and found out what the benchmark MTF is for a Marquee 8500 I would be in excellent shape. I used a camcorder on a tripod and a TV to get nice and close to the screen to dial in focus. What did peak my curiosity was the statement "You can increase MTF by overfocussing. This makes the top of the pulse wider but the edges get steeper and the foot of the pulse gets narrower." that one of the guys posted as a response in your referenced thread. I'll have to try this.

Nashou66: I'd be interested in your experiment with your modded 8500 against a stock 8500. That would give some perspective on how your mods improve sharpness.

Rgds,

Kevin.

[Nashou66]