AVS Forum banner

5421 - 5440 of 13667 Posts

·
Registered
Joined
·
28,788 Posts
Thanks Mike. How about my other question here:



Mike - Unless I misunderstood, when we went through the lumens and nit calculation, I think you said you would expect me to get somewhere around 90 - 100 nits on a new bulb. Can you please detail your math and thinking which led to that, as I don't recall. In my quote above you can see my reasoning for why I think it will be in the 70s nits. I could be off base, but would like to know how you arrived at your estimate so I can compare with mine. Thanks!

For quick reference, I have a 0.93 gain 2.37 AR 140" wide screen (XD). Please calculate on that and then I'll reduce your estimated nits by 10% if going with the V6 screen.
When you fill the width of your 140" wide screen, you are at mid throw? Reason I ask, mid throw for a 140" wide image would be (24'-6") at absolute max throw for your 16:9 image (24'-6 3/8") on your 2.37 screen. What is your throw distance? I also used 0.98 gain for the XD screen.
 

·
Registered
Joined
·
18,137 Posts
You are not misunderstanding it at all.... Simple math is your friend. This is why I went with the ISCO IIIL
You're right - math is your friend. If I used a HE lens that stretches the picture 1.33 X, to fill my 118" wide screen, I'd need to zoom the picture down to around 88.75" wide. With my DCR lens, I need to zoom it about as wide open as possible from my throw distance to get 118" wide. Then use the VC lens. The DCR lens in my setup is much brighter on the same size screen than a HE lens would be. 38% brighter as measured by me. A HE lens would not be nearly as bright, since I'd have to start with a much smaller image.
 

·
Registered
Joined
·
4,587 Posts
You're right - math is your friend. If I used a HE lens that stretches the picture 1.33 X, to fill my 118" wide screen, I'd need to zoom the picture down to around 88.75" wide. With my DCR lens, I need to zoom it about as wide open as possible from my throw distance to get 118" wide. Then use the VC lens. The DCR lens in my setup is much brighter on the same size screen than a HE lens would be. 38% brighter as measured by me. A HE lens would not be nearly as bright, since I'd have to start with a much smaller image.
I haven't seen any manufacturer comment, but I've never got the impression that there was a huge delta between the brightness of a VC and HE lens. Both allow you to use the full panel and thus full brightness of the unit. The HE lens is just taking the more concentrated brightness of the narrow image and expanding it. The VC lens is taking the dimmer and bigger image and compressing it. The only reason I have seen the VC lens getting more chatter is that it has one engineered for 17:9. HE lenses generally have been much more common since they fix issues with projector placement where VC does not.

Maybe someone with more knowledge will chime in, but the figures for I've seen over the years for lumen boost have been close for both lens types.
 

·
Registered
Joined
·
2,804 Posts
You're right - math is your friend. If I used a HE lens that stretches the picture 1.33 X, to fill my 118" wide screen, I'd need to zoom the picture down to around 88.75" wide. With my DCR lens, I need to zoom it about as wide open as possible from my throw distance to get 118" wide. Then use the VC lens. The DCR lens in my setup is much brighter on the same size screen than a HE lens would be. 38% brighter as measured by me. A HE lens would not be nearly as bright, since I'd have to start with a much smaller image.
We will try to keep everything equal so there is no question of numbers being skewed. Throw ratios can be a lot lower for both lenses but I wanted a nice conservative number in the middle of zoom range.

Let’s start with a desired scope screen of 150” at 2.35:1 (yes, 1.33x is actually 2.37:1 but most scope screens sold are 2.35:1 so let’s use that as reference, the math should still add up at 2.37:1).

The 150” scope screen will be 138” width by 58.7 height.

Using a HE lens you need to calculate for the height of the 16:9 so that would be 58.7 which is roughly a 120” 16:9 screen.

For a VC lens you need to calculate for the width of the 16:9 so that would be a 138” which is roughly a 158” 16:9 screen.

Let’s use the RS640 for reference on the Web Projector calculator to determine lumens for both.

We will use a 1.3 gain screen at the exact same throw ratio (for this test let’s say 1.7:1 throw ratio).

For the HE lens we need to calculate throw ratio for 120” 16:9 width of 105”. At 1.7 throw ratio that puts us at 178.5”, or 14’10”.

For the VC lens we need to calculate throw ratio for 158” 16:9 width of 138”. At 1.7 throw ratio that puts us at 234.6”, or 19’6”.

Now, let’s find out the lumens for each of those at their native 16:9 without the lenses in place.

Since we are using a RS640 for reference, let’s figure out the lumens for high lamp calibrated according to the Web Projector calculator.

For the HE at 178.5” (throw distance) or 14’10” with a 1.3 screen at 120" (16:9) we are at 48.0fL

For the VC at 234.6” (throw distance) or 19’6” with a 1.3 screen at 158" (16:9) we are at 27.6fL

So now let’s take into account the numbers with lens in place for each of those locations.

For the HE lens there is a 15-20% light loss due to the HE…but for this test I will use a bit higher loss of 30% to make things more equal. 30% light loss from 48fL is 34fL

For the VC lens there is an increase of 38% due to the VC…. That would make the 27.6fL into 38fL

EDIT - Fixed the adjusted lumens for the VC since I had used LOW lamp instead of HIGH.... thanks to @dlinsley for pointing out my mistake.
 

·
Registered
Joined
·
13,035 Posts
Estimating brightness in HDR

When you fill the width of your 140" wide screen, you are at mid throw? Reason I ask, mid throw for a 140" wide image would be (24'-6") at absolute max throw for your 16:9 image (24'-6 3/8") on your 2.37 screen. What is your throw distance? I also used 0.98 gain for the XD screen.
Nice catch. Based on this:

...Your 140" wide 2.37 screen is 140/2.37 = 59.1 inches visible height.
59.1" visible height translates to a 16:9 width of (59.1/9)*16 = 105.1 inches wide.
This is a 16:9 diagonal of sqrt (59.1^2 + 105.1^2) = 120.6"
2.37" diagonal is sqrt (59.1^2 + 140^2) = 152" diagonal

So you need to look up 16:9 size of 120.6": 120" is in the table at 11.78 to 23.96 ft lens to screen.
2.37 isn't covered, but 2.35 is. 152" is close enough to 150", which is 15.63ft to 31.70 ft lens to screen.
These estimates are based on a 18' throw on a 2.37 140" wide screen (various gains listed below). If you have a different size screen, your nits and ftL will work out differently:

Note that although the spec is 2200, Cine4home reported that 1870 was measured after calibration to D65 (18% light loss from calibration). Reference: http://cine4home.de/neues-jahr-erstes-messergebnis-telegramm-zum-jvc-dla-nx9/ .

16:9 Expected Brightness in HDR:

- 1,870 lumens at shortest throw, high lamp, iris 0
- Lose 5% due to zoom needed to offset the 17:9 panel: 1,870 less 5% = 1,777 lumens
- Lose approximately 15% (guestimate) due to being right at mid throw: 1,777 less 15% = 1,510 lumens
- 1,510 lumens works out to 85 nits / 25 ftL with 0.93 gain XD screen

2.40 Expected Brightness in HDR:

- 1,870 lumens at shortest throw, high lamp, iris 0
- No loss due to 17:9 panel, unlike with 16:9 viewing
- Lose only approximately 5% (guestimate) due to being very close to short throw (just 15% away from shortest throw) 1,870 less 5% = 1,777 lumens
- Lose approximately 37.5% (based on experience with RS500) for being zoomed out for 2.40: less 37.5% = 1,111 lumens
- 1,111 lumens works out to 62 nits / 18 ftL with 0.93 gain XD screen

2.40 Expected Brightness in HDR with DCR A-lens:
- Continue with 1,111 lumens from 2.40 mode above
- Gain 38% from DCR lens: 1,111 plus 38% = 1,533 lumens (I think it is wrong to use 38% because it is already a 17:9 panel, therefore the gain may only be closer to 28%, not sure. Anyone??)
- 1,533 lumens works out to 86 nits / 25 ftL with 0.93 gain XD screen

Note: These units are not out yet, and there are several key assumptions being made, which could very well turn out to be wrong. Therefore please take these estimates with a grain of salt and just to get a rough idea. Actual results will definitely vary. And before making any purchase based on suitability of brightness, please be sure to have your sales person review your estimates to make sure the projector will meet your expected brightness needs. Also you will need to adjust the loss of brightness up or down based on your expected through distance in 16:9 and or 2.40. The math above is based on my specific throw. You'll also need to convert the lumens to the nits / ftL you'll get based on your screen size and gain, which will be vastly different than what I listed above unless you have the same screen size and type.

Note: These figures are WITHOUT the BT2020 color filter in place, as I won't be able to use it. With the filter in place, preliminary it sounds like these numbers would all be 1/3 less.

Edit: Reorganized and also updated to account for approximately a 5% drop in lumens in 16:9 viewing due to the 17:9 panel, with no 5% drop in 2.40 viewing.
Edit: Removed nits and ftL for V6 screen since I am no longer considering it and planning to stick with XD, at least for now.
Edit: Change starting lumens from the 2,200 spec to the 1,870 measured lumens after calibrating to D65, per cine4home.
 

·
ABSOLUTE ULTIMATE AV
Joined
·
4,994 Posts
Discussion Starter #5,426 (Edited)
I dunno, are you sure? The blacks on a VPL-1100ES is not that great and that claims 1 million:1. I know Sony fudges its numbers so if that does even 1/4th of what they claim then that's 250K:1. That doesn't feel good enough to me. Do know what what the dynamic contrast is on my 675ES? I'm so unimpressed with the blacks on the 675ES that I had to set up a dual projector setup. Sony claims 350K:1 on that one right?
Yep I am pretty sure.

Someone is welcome to come along with measurements to prove me wrong.

Where is Nigel?
Sorry! Been travelling on business!

Whilst JVC's reported measurements are typically very accurate, the situation is the precise opposite when it comes to SONY's. The reality with respect to the SONY 1100ES is by no means 1,000,000:1 max ON/OFF but in fact measures circa 300,000 - 330,000:1. And with respect to the SONY 675ES which claims max ON/OFF of 350,000:1 this in reality measures circa 80,000:1 - 90,000:1

:wink:
 

·
Registered
Joined
·
13,035 Posts
3840 pixels wide / 2.4 aspect = 1600 pixels vertically are used in a scope image.
(4096 * 2160) / (3840 * 1600) = 1.44, i.e. 44% of the pixels in the panel are black bars.
Right - thanks for providing the math. That makes a lot of sense. That said, I wonder though if there are some other properties related to the zooming / lens so that even though you lose 44% pixels that may not translate directly 1:1 to lose in lumens?

For instance, as mentioned I get 80 nits in 16:9 mode and 50 nits in 2.40 zoom mode. That's with both at 100% at D65. So in that case, going from 80 to 50 nits is a loss of only 37.5%, not 44%. That may not seem like a big deal but since I'm scraping for ever lumen I can get that means my output is 17% brighter than it would be with the theoretical 44% - which is a big deal.

Therefore I'm wondering if it may be realistic to expect my drop from the 2.40 zoom to "only" be 37.5% instead of the theoretical 44%. Because if so, the numbers look much better. It means I can be at 96 nits / 28 ftL with the XD screen or 87 nits / 25 ftL with the V6 screen.

As a side note, you can see how the V6 material really needs to be super special to be worth giving up the additional 10% in brightness to switch to it from XD. In many cases 10% may not matter, but in my case you can see how it makes a significant difference. Not to mention a few hundred hours into the bulb what my lumens will be in either case...
 

·
ABSOLUTE ULTIMATE AV
Joined
·
4,994 Posts
Discussion Starter #5,428
Javs if we see that the tone mapping is not that great do we have the ability to just upload your curves right away..

Sent from my SM-G930T using Tapatalk
I don't see why not.

If the Z1 can do it I'm sure these can. Worst case you can use the autocal versions.
You most certainly can and in fact you can have it automatically switch to whatever selection of your own custom GAMMA curves you like as well :)

See the following video which I filmed whilst at CEDIA where Chris Deutch explains this:


:wink:
 

·
Registered
Joined
·
18,137 Posts
We will try to keep everything equal so there is no question of numbers being skewed. Throw ratios can be a lot lower for both lenses but I wanted a nice conservative number in the middle of zoom range.

Let’s start with a desired scope screen of 150” at 2.35:1 (yes, 1.33x is actually 2.37:1 but most scope screens sold are 2.35:1 so let’s use that as reference, the math should still add up at 2.37:1).

The 150” scope screen will be 138” width by 58.7 height.

Using a HE lens you need to calculate for the height of the 16:9 so that would be 58.7 which is roughly a 120” 16:9 screen.

For a VC lens you need to calculate for the width of the 16:9 so that would be a 138” which is roughly a 158” 16:9 screen.

Let’s use the RS640 for reference on the Web Projector calculator to determine lumens for both.

We will use a 1.3 gain screen at the exact same throw ratio (for this test let’s say 1.7:1 throw ratio).

For the HE lens we need to calculate throw ratio for 120” 16:9 width of 105”. At 1.7 throw ratio that puts us at 178.5”, or 14’10”.

For the VC lens we need to calculate throw ratio for 158” 16:9 width of 138”. At 1.7 throw ratio that puts us at 234.6”, or 19’6”.


Now, let’s find out the lumens for each of those at their native 16:9 without the lenses in place.

Since we are using a RS640 for reference, let’s figure out the lumens for high lamp calibrated according to the Web Projector calculator.

For the HE at 178.5” (throw distance) or 14’10” with a 1.3 screen at 120" (16:9) we are at 48.0fL

For the VC at 234.6” (throw distance) or 19’6” with a 1.3 screen at 158" (16:9) we are at 20.1fL

So now let’s take into account the numbers with lens in place for each of those locations.

For the HE lens there is a 15-20% light loss due to the HE…but for this test I will use a bit higher loss of 30% to make things more equal. 30% light loss from 48fL is 34fL

For the VC lens there is an increase of 38% due to the VC…. That would make the 20.1fL into 27.7fL
In my case I can't change the throw - the throw is fixed at under 14'. So my projector lens is close to wide open using either the zoom method or the VC DCR lens. Using a HE lens, it would have to be closed down more in order for the picture to fit on the screen. It would be interesting to measure the actual light output with a HE lens, compared to the VC lens. If I had one, but I do not.
 

·
Registered
Joined
·
8,844 Posts
I haven't seen any manufacturer comment, but I've never got the impression that there was a huge delta between the brightness of a VC and HE lens.
...
Maybe someone with more knowledge will chime in, but the figures for I've seen over the years for lumen boost have been close for both lens types.
Yes, both lens types, by themselves, provide similar lumen boost after the effective screen area is taken into consideration. The large difference being discussed here is not due to the A-lens, but due to the projector itself losing lumens when operating at a longer zoom.
 

·
Registered
Joined
·
28,788 Posts
Nice catch. Based on this:



My throw is 18'. So then for 2.40 viewing, either with or without the DCR lens, I will be fairly close to SHORT throw (about 15% into the throw range). So let's take a fresh look then:

- Spec 2,200 lumens at shortest throw, high lamp, iris 0
- Lose 5% due to 17:9. 2,200 less 5% = 2,090 lumens
- Lose say 5% (?) due to only being 15% away from shortest throw 2,090 less 5% = 1,985 lumens
- Lose say 44% for being zoomed out for 2.40, per discussion above: 1,985 less 44% = 1,110 lumens
- Gain 38% from DCR lens: 1,110 plus 38% = 1,532 lumens
- With a 1.0 gain screen I would then be at 92 nits / 27 ftL.
- With the XD 0.93 gain screen I would be 7% less, so that's 86 nits / 25 ftL.
- With the V6 0.84 gain screen I would be 16% less than the 1.0 gain screen, so that's 77 nits / 23 ftL

Does this look correct? Seems like with XD at 86 nits I am roughly in the ballpark of the 100 nit title, however the loss of gain if going to the V6 really seems to set it back at 77 nits.

Thanks!
Black bars on a 16:9 panel are 33% (using 2.37) You used 44%. I have not checked, but 44% is probably the difference between full 17:9 panel and scope with 16:9 panel, but you already reduced the light for the difference between 17:9 and 16:9, so you should be using 33% reduction rather than 44% reduction. If I am following what you are doing.
 

·
Vendor
Joined
·
26,647 Posts
Do we know at this point when JVC expects to start shipping units?
Middle to the end of October........

True, but for the guys looking to take advantage of the preorder pricing they will have to take the leap of faith as the pricing will end before reviews are in.
That may not necessarily be true.......
 

·
Registered
Joined
·
13,035 Posts
Black bars on a 16:9 panel are 33% (using 2.37) You used 44%. I have not checked, but 44% is probably the difference between full 17:9 panel and scope with 16:9 panel, but you already reduced the light for the difference between 17:9 and 16:9, so you should be using 33% reduction rather than 44% reduction. If I am following what you are doing.
I am using 44% because of the 44% number dlinsley mentioned in the below quotes:

It should be 2100 - 44% + 38% = 1974lm.

The lens is a vertical compression type, and so you need to be zoomed to your 2.4:1 position. These being the light from above/below the screen back onto the screen, but don't change the horizontal. The DCR is a is 0.8x vertical compression (1.0 horizontal): (4096 / 2160) * (1.0 / 0.8) = 2.37:1

Usually 44% of the light/pixels is lost to the sides/above/below the screen when zooming 3840x1600 scope presentation. The scaling to 4096x2160 gets this back, but the lens has some losses and ends up at 38%. I think you also lose a small amount from having to be zoomed out a little to fit 4096 wide instead of 3840 wide. This may be accounted for in the losses that bring you down to 38%, but I'm not sure. You can also get a 13% increase without the lens just by doing the 17/16*17/16 scaling to panel width while keeping the aspect correct.

The old UH480 lens was a horizontal expansion lens, which you would have used the 16:9 position. This is one of the advantages of VC over HE in that you are starting with a brighter output to begin with.
It sounds like part of the issue is that I already subtracted out 5% loss from the 17:9 panel. But even adjusting for that, it doesn't explain the difference between your 33% and his 44%. Can you let me know how I would adjust my math in the detailed example in my post to provide a more accurate estimate? Also as I noted in a recent post above, I lose 37.5% when zooming from 16:9 to 2.40 (screen is 2.37 but I use a tad bit of overscan). In that case would we use your 33% or my 37.5%? Thanks!
 

·
Registered
Joined
·
2,804 Posts
Yes, both lens types, by themselves, provide similar lumen boost. The large difference being discussed here is not due to the A-lens, but due to the projector itself losing lumens when operating at a longer zoom when using an HE lens.
But if you use an HE lens you may actually be using a shorter throw since you start off a lower 16:9 than a VC lens. See my example above for the exact same sized scope screen.
 

·
Premium Member
Joined
·
1,381 Posts
For the HE at 178.5” (throw distance) or 14’10” with a 1.3 screen at 120" (16:9) we are at 48.0fL

For the VC at 234.6” (throw distance) or 19’6” with a 1.3 screen at 158" (16:9) we are at 20.1fL
Are your numbers correct? When I put those numbers in I do get 20.1fL for VC but only 35.1fL for the HE - I think you substituted the High lamp for the HE and used low lamp for VC. Then doing the lens adjustment of x1.38 for VC and x3/4 for HE, gives me VC=28fL and HE=26fL.
 

·
Registered
Joined
·
2,804 Posts
Are your numbers correct? When I put those numbers in I do get 20.1fL for VC but only 35.1fL for the HE - I think you substituted the High lamp for the HE and used low lamp for VC. Then doing the lens adjustment of x1.38 for VC and x3/4 for HE, gives me VC=28fL and HE=26fL.


Did you enter the distance correctly?
 

·
Registered
Joined
·
2,804 Posts
Are your numbers correct? When I put those numbers in I do get 20.1fL for VC but only 35.1fL for the HE - I think you substituted the High lamp for the HE and used low lamp for VC. Then doing the lens adjustment of x1.38 for VC and x3/4 for HE, gives me VC=28fL and HE=26fL.
Damn it!!! You are right... I did high/low
 

·
Registered
Joined
·
2,804 Posts
Are your numbers correct? When I put those numbers in I do get 20.1fL for VC but only 35.1fL for the HE - I think you substituted the High lamp for the HE and used low lamp for VC. Then doing the lens adjustment of x1.38 for VC and x3/4 for HE, gives me VC=28fL and HE=26fL.
I corrected the low/high for the VC. Thanks for pointing out my error. Credit was given where it is due.
 
5421 - 5440 of 13667 Posts
Top