On the cheap-How to get into the secret club? - Page 9

And that is the story I am sticking to as well

Mark

If you want to use a smilie I don't care. I said VTPete's setup has a larger viewing angle. This is correct.

I'm just wondering if anyone has tested if it's actually using the 810 source lines to do the stretch or the 540 scaled to panel res lines to do the stretch.

You should be able to test it by sending a 1080 resolution test to the projector and a 720 one and comparing them after using the vertical stretch.

Yes, larger than someone who had claimed to have a 180 degree viewing angle. Which is why I'm confused about your "good thing I never said that" statement?

http://www.thx.com/cinema/builtTHX/screen.html

Read the thread.

Firstly, the projector does not actually do a vertical stretch. You can test that by setting the source to 4 x 3 in which case you find that it does not work anymore. The VS image we see is source related and only works wen the projector is designed to allow other viewing modes other then 16:9...


As far as I am aware, it is becuase it simply is clipping to the top and bottom (the black bars) off the image. And setting the source to 1080i on a 720 projector looks better then sending 720 for CIH.

On a 1080 projector, the only thing that would look better is a true anamorphic transfer...

Mark

I'm not really following you here Mark.

The projector is getting a signal of a fixed resolution. If it is HD it is 16x9. When you enable the 'vertical stretch' it keeps the source the same width and stretches the middle 75% of the source to fill the panel.

The 'vertical stretch' mode may be intended for using other viewing modes that 16x9 but whatever its intent is in order from it to take an image that doesn't fill the panel and make it fill it by cropping the top and bottom and just using the middle 75% of the image it has to stretch it.

Image comes in with black bars top and bottom. It's as wide as the panel. Width doesn't change. Middle part of image now fills panel. It had to stretch it.

Some projectors just may not do it if you change the source because it's been programmed to not allow the function when in 4:3 mode because the makers didn't want to include the funtions in that mode to prevent users from doing something they thought they shouldn't.

If it's clipping the top and bottom of the image and filling the panel it has to be stretching the middle. Or if it's a 720 setup you could think of it as compressing the sides. Stretching 1920 pixels to fit on 1280.

But a 1080i source looking better than 720 make me think that your projector is able to use more of the source when scaling so it's scaling the full 1920x1080 image down to the 1280x720 panel. Rather than converting the image to 1280x540 and then upscaling this to 1280x720.

This would explain why the 720p projector users are so adament about how much better the image looks when using the lens setup -vs- zooming. You're actually getting a lot more information out of the source that way. We'll see what things look like when I test out my setup next week.

Do you have any images you'd like me to test from HD-DVDs? Or post any you have that you'd like me to test out.


We can only dream! I wish they would.

Cheers.

But the projector's 16:9 mode also electrically stretches the image. That is why it appears geometry correct. When I choose letterbox mode like 4 x 3 zoom, the image is not geometry correct becuase the projector is not electrically stretching the image. The problem is, some projectors lock to 16:9 for HD. This works however on almost all projectors for SD...


The Letterbox mode or 4 x 3 zoom modes were never intended for 16:9 native or enhances program - that was what the 16:9 mode was for. The reason some projectors don't appear to offer VS is becuase they were designed to prevent consumer confusion by taking out mode choices by the consumer - very plug and play and would always work for HD to give geometry correct images.

So what your seeing when the image appears VS'd is the native format before the display stretched it.


But that is the thing, it is not wide as the panel, and why the 4 x 3 mode works for full screen height 16:9 programing when using a lens. You have to think of it as a 4 x 3 frame, even though it is native 16:9, and what you see then is that the display on letter box mode zooms that frame in to fill the width, then at that new width, a 4 x 3 frame is taller, so it cuts the top and bottom off. That is how and why the scaling works for both SD media and HD media, but the source MUST BE set 16:9 or it will not work...


Correct, and letter box is traditionally a 4 x 3 mode...

Mark

Compressing the sides for HD media is more correct where SD is vertically stretched. To test this, press 4 x 3 mode on the projector. Regardless of the program's AR, you will see that the image is tall and thin. If the program is 16:9 or close, it should fill top to bottom, if Scope, it will have black bars top and bottom. Both will have pillars at the sides. If you use the lens, it should take on the geometry correct look as if your were using 16:9 mode and no lens...


At this stage, I am going to say Player/Projector related. Benq 8720, fed from XE1 HD DVD looked better when the player was set to 1080i that 720P...It shouldn't as 1080i only has 540 lines on the screen at any one time...



Technically at 1080, you've got more lines so some to throw away, but it is still interlaced...


It is an intersting subject, so please post away...

Mark

Hi, all,

I just stumbled across this thread, and, while I haven't had the time this AM to read it all before going off to work, I noted a few posts pointing out that no one knew of any lens users who went back to zooming.

Well, now you know at least one - me.

I've had a 2.35 screen and been doing constant height for at least 6 or 7 years. I have an ISCO II lens purchased new on it's initial release, and used it happily for years. For the past several years now I've been zooming, instead, and my lens is sitting in my living room, looking hi tech. Brightness is not an issue - I use a Da lite hi-power screen. The black bands of the zoomed 2.35 image are masked off and invisible. I started doing this when I went to a Panasonic AE900 and found it had enough zoom range to accomplish this. Have tried an AE2000 as well, and am about to upgrade to it. To my eye (and I'm an ophthalmologist) the image looks great. I just find it a lot less hassle than scaling and using the lens. I think of it as optical rather than electrical scaling. This is not a religious issue, however, and if you like hi tech toys, the lens-scaler route is cool.

I posted this a few years back as well, and some of you offered to buy my lens, but at that time I was afraid to sell. I might be willing, now.

Wow. I give up.

Hi Mark
Just wanted to pull you up on this one thing. As the source is progressive (i.e. 24fps film encoded on the disk), then by the time your projector displays it you will have 1080p lines of real information on each frame, not 540. Providing, of course, that something in the chain is doing inverse telecine deinterlacing.
The fact that XE1 looked better at 1080i than 720p points to the fact that the downscaling was performed better by the projector than the XE1. (It has to end up at 720p in the end to be shown by the projector).

Good point about HD in 1080P, but the Benq 8720 would not except anything other than 720P or 1080i/60, so 1080P/24 was not happening.

On the few films I watched, the back grounds shimmered on 720P, but were sharp on 1080i and I am not sure why...

Mark

Mark, 1080i has 1080 lines of vertical resolution. For 24 fps HD DVDs the player interlaces it and the projector deinterlaces it. The AE2000U can deinterlace it back to a 1080p 24fps signal because all of the information is still there. Basically with 1080i 60 fps you get as much information as 1080p 30 fps because the vertical information is divided between the even and odd frame.

http://en.wikipedia.org/wiki/Interlace

As far as what the vertical stretching is actually doing I think we're discussing different formats! If the source is anamorphic DVDs then with no scaling by the projector the image will look vertically stretched. If the source is HD DVD with no scaling the geometry of the image will be correct. In order to use a lens the source must be scaled.

Hopefully I'll get the time next week. I was just hoping you might have a few recommended images to show the difference. Or any that anyone else would like to see.

I have to rearrange the room to get the projector in a position where the lens can be used and I need to get things back together in time for the wife to watch Jeopardy! so I'll need requests before starting not after everything is put back into place.

I'm looking forward to seeing your screen shots, maybe it will shed some actual test data onto this zooming versus lens issue. I don't think the subject matter means a lot as long as the source DVD is of the best quality. I would recommend some landscape or distant type of shots, as well as shots which are close ups of faces. The 40th anniversary DVD of 'The Sound of Music' is superb quality and has a mix of all types of shots. Looks incredible on my zoomed 2.35 . May I also recommend that you post the pictures on a specific new thread - this ones getting really long.

Hopefully soon I will have some shots of HD media scaled for CIH using the projector. I've asked for both the 4 x 3 mode and 4 x 3 zoom mode to show what I am talking about...

Mark

Yes, that's the trouble with the ISCO II, superb lens but since it magnifies by a small amount, if you remove it you need to rezoom to fill the screen.

The ISCO III and just about every other anamorphic lens has no magnification so you can just slide them in and out without having to rezoom.

Yeah, I'll start up a new thread. I'll mainly be using my HD DVDs but I'll try and get a copy of the Sound of Music DVD as well.

Here's the stretching that my AE2000U will do:


The top portion is SD material where if the source is the 'squeezed signal' anamorphic then yes it will need to be scaled to get the geometry correct.

The bottom is 16:9 HD material where for 'cinemascope' the source doesn't need scaled for zooming. V-Fit would be used for a lens setup. I don't know why they show black bars on the side of the 'cinemascope' image and you can also select '16:9' where it just shows the source not scaled.

I had quick play with that projector, and it seems you use Zoom 1 for Scope...
You can use 4 x 3 for 16:9 when leaving the lens in place all the time. This was tested over HDMI fed from a BD player that was set to 16:9...

Mark

Could you explain this in a little more detail? Are you saying that in addition to stretching in the horizontal direction, it also makes the vertical larger as well? By how much?

I thought that the only functional diffference between the ISCO II and III was that the III is larger, allowing more flexibility in placement relative to projector and with projection angle (i.e. short throw vs long throw). But your post would indicate that the ISCO II has a design flaw that prevents it from functioning like a correct anamorphic lens (stretch in one direction and leave the other direction unchanged).

Not sure of the exact amount of increase vertically, but my understanding as to why this is the case is that the ISCO II uses a combination of Cylindrical and Spherical elements, where the ISCO III only uses Cylindrical elements.

The difference between Cylintrical and Spherical lenses is that Cylindrical lenses are flat in one plain, curved in the other, where are Sphericial is curved in both plains.

When orientated correctly, a Cylindrical element is not curved vertcially so it does not affect the height, and why it makes good sense to use it in an anamorphic lens for CIH applications.

You will notice that even the prism based lenses are also flat vertically, therefore they do not alter the image height either...

Mark

According to the manufacturer's specs ISCO II is 1.04X magnification for long throws and upto 1.12X for short throws.

ISCO III is 1.0X magnification.

Not a design flaw, so much as by product of the lens elements used. Magnification is in all directions, not one.

That is vertical, yes?

Mark

The manufacturer's specs simply say "magnification", which I take to mean the same amount in all directions. If it was just vertical I think they would have called it something else like "vertical stretch"

That spec has to be vertical
IE it does not change vertically because horizontally, it is 1.33x stretch.

If it were 1.0x in all directions, then it is not an anamorphic lens, but a piece of flat glass...

Mark

Design flaw or not, it's disappointing that it's can't be used in a sliding lens system without requiring projector zoom adjustment. Unless the projector has power zoom memory settings (and there aren't very many of those), it defeats the purpose of an automated lens setup.

But it is fine for those wanting to leave the lens in place all of the time...

Mark