AVS Forum banner
1 - 20 of 26 Posts

·
Registered
Joined
·
7,349 Posts
Discussion Starter · #1 ·
Since my DIY a-lens slide is finished I was always curious to the PQ of "Native 16:9" vs thru the combo "a-Lens expansion/PJ compression Scaler".

(main reason for having a a-lens slide)


I've not seen this study done here before - did I miss it?

This is purely a subjective visual study at this time, I'm no expert and w/o "real" tools to capture objective data (besides a basic / decent camera).


My Equipment:

PJ: Sony VPL-VW60 (12 months old, less than 500 hours total on it)

a-Lens: Panamorph UH380 (b-stock avs power buy 10 months ago, hopefully PQ does not suffer because it's a b-stock unit)

Canon SD880 IS mounted on tripod, at center screen height/width 10' from screen.

Camera/Tripod of course not moved/zoom stayed same for all shots for relative comparision purposes.

Camera in program mode, flash off, ISO set to 100.

If I had a D-SLR could have totally fixed it's settings and seen a vs b comparision, but this should still capture any major diff's.


Had "Old School" on DVR, HD with some decent shots in it, so watched it tonight after the kids went to bed and put my just finished manual slide to use.

Back and forth, back and forth, cycle testing it....


The first thing I immediately noticed is the "a-Lens/PJ Scaler" image is about 3" skinner than the "native" image.

I put blue painters tape on the screen for the skinner image (a-Lens/PJ Scaler image), you can see that in most of the "native" shots.

Not sure why that's happening, according to what I've read and understood in the Sony manual the "normal" mode compresses 16:9 into 4:3, correct?

This should exactly counteract the a-lens 4/3 expansion ratio to give a net result of 16:9




Besides that look at the picts and make your own subjective visual observations:

Honestly, I was expecting more difference, like the native was obviously sharper/clearer, but I'm pleasantly surprised.

Looks like the "a-lens/PJ scaler" combo does a decent job.

Since this is a $3k lens and a +$4k PJ I'd expect this level of capability.


I'll have to dig into what's going on with the "squeeze" image I'm getting with the a-Lens/PJ Scaler doing their thing.

At this time I have absolutely NO plan to buy a separate scaler box.


Here are 5 comparison series of pictures.


1a native 16:9:



1b a-Lens/PJ Scaler:



2a native 16:9:



2b a-Lens/PJ Scaler:



3a native 16:9:


3b a-Lens/PJ Scaler:



These next two series has "hair detail", which I thought looked fine in both.


4a native 16:9:



4b a-Lens/PJ Scaler:




5a native 16:9:


5b a-Lens/PJ Scaler:




Next, I'll go back to the original pictures and blow them up and look at detail to see if any loss.
 

·
Registered
Joined
·
7,349 Posts
Discussion Starter · #2 ·
Quick zoom, basically same.

What I need is good PQ 16:9 HD material from Blu-Ray source to try this with.

This Cable HD does not have as much detail as I've seen in Blu-ray's.



Full / Native:



Normal / PJ/Scaler
 

·
Registered
Joined
·
866 Posts

Quote:
Originally Posted by jayrader /forum/post/15412977


Nice. I would say that the shots with the scaler / lens have the edge, but definately not enough for me to drop thousands of dollars on the setup for it. Glad I'm still zooming.

I think that the comparison is for those that already have a lens and wondering if they should invest in a slide or leave it in place permanently. The squeezed lens image should look worse than the native image because it's using less pixels (1440x1080 vs 1920x1080) for the same image size. But the screencaps really do seem to show that this difference is really hard (if not impossible) to detect. The 3" skinnier image with the squeezed image is disconcerting, but that may be an issue with the particular projector scaler being used, and not a problem with squeezing in general.


PS - Mike: Nice comparisons, they are very useful.
 

·
Premium Member
Joined
·
1,030 Posts
Ya I leave my lens in place all the time and really like the convenience.


Nate
 

·
Registered
Joined
·
6,298 Posts

Quote:
The squeezed lens image should look worse than the native image because it's using less pixels (1440x1080 vs 1920x1080) for the same image size.

But it's NOT the same image size. The pixel density per horizontal inch of the image is the same when the lens is in place regardless of the AR of the image.


An example:

2.40:1 image dimension 5' x 12' 1920 / 12 = ~160 PPI (Pixels per inch)

1.85:1 image dimension 5' x 9.25' 1500 / 9.25 = ~162 PPI (Pixels per inch)


This is why using a lens slide for 1.85 AR sources with a good quality anamorphic lens does not show any advantage at a reasonable viewing distance when the lens is in use.


Vern
 

·
Registered
Joined
·
8,957 Posts

Quote:
The 3" skinnier image with the squeezed image is disconcerting

I would like the OP to confirm that the Panamorph lens is indeed aligned correctly as one of the limitiations of a prisms lens is that rotation affects the optical stretch. Cylindrical lenses do not suffer this...


Mark
 

·
Registered
Joined
·
866 Posts

Quote:
Originally Posted by Vern Dias /forum/post/15419194


But it's NOT the same image size. The pixel density per horizontal inch of the image is the same when the lens is in place regardless of the AR of the image.


An example:

2.40:1 image dimension 5' x 12' 1920 / 12 = ~160 PPI (Pixels per inch)

1.85:1 image dimension 5' x 9.25' 1500 / 9.25 = ~162 PPI (Pixels per inch)


This is why using a lens slide for 1.85 AR sources with a good quality anamorphic lens does not show any advantage at a reasonable viewing distance when the lens is in use.


Vern

Maybe I'm misunderstanding the comparison. I believe Mike was looking at a 1.78 image in two ways: with the lens out of the way using the full 1920x1080 panel, and then with the lens in place while scaling the image to use the central 1440x1080 portion of the panel. If the lens has the correct stretch ratio, shouldn't the projected image in either case be the same size?


Why are you comparing 2.40 vs 1.85?
 

·
Registered
Joined
·
866 Posts

Quote:
Originally Posted by CAVX /forum/post/15420274


I would like the OP to confirm that the Panamorph lens is indeed aligned correctly as one of the limitiations of a prisms lens is that rotation affects the optical stretch. Cylindrical lenses do not suffer this...


Mark

I hope it can be corrected, as the distortion can easily be seen even in the screen shots.
 

·
Registered
Joined
·
6,298 Posts

Quote:
If the lens has the correct stretch ratio, shouldn't the projected image in either case be the same size?

Yes, exactly. However, if you are using an anamorphic lens you are using it to project 2.40 source on a 2.40 screen.


More importantly your seating distance should be optimized for the 2.40:1 scrren width to avoid seeing individual pixels while also maximizing the resolution that is perceived by your final optical system, your eyes.


This means that the resolving power of your eyes will determine the maximum effective resolution that you can perceive and delivering a higher resolution to the screen for a 1.85:1 source by not using the anamorphic lens will result in no visible increase in resolution to your eys/brain....


Unless, of course you change your viewing distance when you watch 1.85:1 sources. (But then why are you wasting time with a CIH set up if you do that???)


Vern
 

·
Registered
Joined
·
7,349 Posts
Discussion Starter · #11 ·
Happy New years!


I have been gone since Tues @ Family, at mother-inlaws now, Thur 1/1/09, just woke up.


Sorry if not clear this original post.

I'll re-state it. (and put in my 1st post)


I posted this in the 2.35 CIH forum because it affect us with scope screens.

Also related to people with anamorphic lens and the decision to buy/make a slide (manual or automatic) to use the a-lens for scope movies and move it out for non scope movies.

I was curious to the PQ of watching 16:9 material with lens in place with my PJ built in scaler vs "virgin" 16:9 material directly onscreen.


This is NOT an offshoot of the zoom for CIH vs a-lens for CIH debate thread.

(at least not intended to be, nor did I read anything in there that led me to do this)


Mark;

I spent lots of time dialing in the a-lens, I believe it is correctly positioned now. I'll take a screen shot of the PJ "green grid".


If my logic is correct and I do my math correctly:

16:9 material, 1920 wide x 1080 tall, with my Sony PJ in normal mode it should re-map those pixels into the middle image chips 1440 pixel for width and still the 1080 for height. I get Black bars on either side.

Then, the a-lens expands those 1440 pixels back to same size as 1920.


So, that picture should look really close to the "virgin" 16:9 material and I was trying to compare both for image PQ, sharpness, brightness, etc.


I'll do some more screen comparisons.

Like I stated, I wish I had a D-SLR to put it into manual mode for screen brightness comparisons.


If possible, it would be nice if 1 or 2 others with lens slide do same and share results.


Why bother doing this?

For me sorta confirmation of why I made my slide in the first place, that the image PQ should be "best" with the "virgin" image.
 

·
Registered
Joined
·
6,298 Posts
Well, the image for 1.85 will be brighter without the lens compared to the 2.40 image with the lens. This will affect your black level calibration, and may require 2 different calibration settings depending on how fussy you are about having it "right".


I see you have a curved screen, which also means that you will not have consistent pincushion compensation unless you always leave the lens in place.


Vern
 

·
Registered
Joined
·
7,349 Posts
Discussion Starter · #13 ·

Quote:
Originally Posted by Vern Dias /forum/post/15429935


Well, the image for 1.85 will be brighter without the lens compared to the 2.40 image with the lens. This will affect your black level calibration, and may require 2 different calibration settings depending on how fussy you are about having it "right".


I see you have a curved screen, which also means that you will not have consistent pincushion compensation unless you always leave the lens in place.


Vern

Agreed;

With lens out of the light path I do get slight barrel at the bottom of the screen.

I was hoping AussieBob would answer Q I asked in another post http://www.avsforum.com/avs-vb/showt...4#post15388954 , but you are also expert:

Quote:
Q:

My UH380 lens is inline with the top of my screen border (20" below my 8.75' ceiling), when I removed the lens I see barrel at the bottom - I expected that, but at the top it was flat, I did NOT expect that.


For true a-lens placement is it best to be centered height wise as well to "balance" the optics?

I could lower my lens 8" more, but lower than that and I'm worried about peoples head hitting possibly.

The tape in the "virgin" 16:9 image is center height/width of screen.
 

·
Registered
Joined
·
736 Posts

Quote:
Originally Posted by mtbdudex /forum/post/15429760



......


Why bother doing this?

For me sorta confirmation of why I made my slide in the first place, that the image PQ should be "best" with the "virgin" image.

Yes. As I see the virgin image is always better for the 16x9 material. Higher resolution, higher brightness and no scaling issues.


I also feel that vw60 normal-mode scaling is not good as anamorphic stretch mode. A good quality A-lens always shines for 2.35:1 material but not the best way to watch 16x9 material, IMO.
 

·
Registered
Joined
·
7,349 Posts
Discussion Starter · #16 ·
Got my Blu-ray DVE HD Basics and spent some time after the kids went to bed.

I reset my PJ to factory settings, then started from scratch following the instructions.


Plus, I went thru the a-lens set-up again, gosh I was having all kinda fun 9pm - midnight while the wife/kids sleep here in minus 10deg f SE Michigan.


About the only time I can do any of this amateur calibration stuff anyways.


Pictures of a-lens "tuned" and some test patterns:

(ceiling paint job looks crappy in these picts, I moved the PJ mount when going from fixed to slide)



Hard to see but image is mostly centered back and front glass:



1.78:1 w/o a-lens, PJ in full mode:



2.35:1 w/a-lens:

w/a-lens and using the Sony anamorphic mode:



Hard to see but my greyscale is dialed in, where the grey bar is just one stop above blending in with the black background.

First time for me to use color filters over my eye to set the color.





The camera saw these morie patterns, my eyes only saw parallel lines, kinda wierd:



We did watch "Meet the Robinsons" on Blu-ray tonight, a so-so movie, but it's 1.78 not 2.35 material so perfect for some comparisions:

w/o a-lens and "full mode" for native 1.78:1



With a-lens and using "normal mode" for end result of 1.78:1 (trying at least): The shift to the left was my cropping the photo.



w/o a-lens and "full mode" for native 1.78:1



With a-lens and using "normal mode" for end result of 1.78:1 (trying at least):



w/o a-lens and "full mode" for native 1.78:1



With a-lens and using "normal mode" for end result of 1.78:1 (trying at least):




I don't know if it's worth me "complaining" my Sony PJ internal scaler normal mode does not correctly compress pictures by 1.333 (4/3) factor.

I'll probably live with my PJ as is.

I should say that other Sony VPL 50/60/70/10 owners should confirm their internal PJ scaler.


At this time I satisfied with my internal scaler and have no plans to spend $$$'s for external scaler.
 

·
Registered
Joined
·
866 Posts

Quote:
Originally Posted by mtbdudex /forum/post/15577810


IMG]


The camera saw these morie patterns, my eyes only saw parallel lines, kinda wierd:

The moire is from the interaction of the test pattern and camera CCD.


Nice followup. Too bad you can't get the correct squeeze, otherwise it looks to me that you could leave the lens in place all the time. As it is now, I would vote for sliding.
 

·
Registered
Joined
·
201 Posts
Sliding Makes Sense


Some movies place subtitles in the bottom black bar, and sliding lets you view 2.35 movies within 16/9 screen area where subtitles are visible.


Movies often offer a choice of French, English, German subtitles. In player Set-Up, the next step needed is an additional choice: Place subtitles wholely in the picture(Yes/No)?


When you choose No, the player uses black bars for subtitles. This choice works well for LCD panels, for plasma screens, and for 16/9 projection screens.


When you choose Yes, the player scoots subtitles upward a line and treats black bars as total black-out and completely devoid of any info, and you might discard those black bars with impunity.
 

·
Registered
Joined
·
1,029 Posts
The image using the lens looks a bit washed out... which is not surprising. I always opt for sliding the lens out of the way for 1.78/1.85 material.
 
1 - 20 of 26 Posts
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top