Forum Jump: 
 33Likes
Reply
 
Thread Tools
post #1 of 123 Old 10-06-2018, 06:52 PM - Thread Starter
AVS Forum Special Member
 
Javs's Avatar
 
Join Date: Dec 2012
Location: Sydney
Posts: 7,129
Mentioned: 415 Post(s)
Tagged: 0 Thread(s)
Quoted: 6061 Post(s)
Liked: 5465
Projector ANSI & ADL Contrast Measurements Thread

Ive decided to move the conversation from the new JVC thread to its own thread, because I think this discussion is not done yet, and this is going to be explored much more fully in the coming weeks with new models coming out. I also invite members to post measurements of your own projectors here, and we can discuss things like your room setup, any issues with the ANSI and ADL readings you are getting etc, and what we can do to improve them.

I will quote dump the whole conversation here so we can continue...


---------------


Quote:
Originally Posted by Javs View Post
I had a go at measuring the ADL patterns from Soulnight on the X9500 last night.

I have come to the conclusion that my 9500 must be somehow ANSI degraded. The numbers I got are not matching what this projector should be doing.

I then tried pretty much literally everything to get the highest 50% ANSI reading possible, including velvet behind the meter to eat pretty much all the light, as close as possible etc, turns out my best ANSI readings where about mid way in the room only about 2m from the lens, with a velvet sheet held behind the meter to stop anything at all hitting the screen. Which was about 165:1 ANSI... which is pretty pitiful given its min throw iris wide open.

I dont believe for a second this is the limit of my room. I have essentially a velvet tent! and most all surfaces are black. I have put 1300 hours on this X9500 now, And there are some dust particles inside the light engine stuck on the various panels, I am wondering if this is causing issues with a bloom effect inside the lens polluting the potential. I also tried no lens shift, thought maybe the slight down shift I use is causing reflections right at the point I dont want them to, but no go.

I have tried to measure the units ANSI a number of times over the past 6 months, never really bothering too much though, but the numbers never really added up to what I think this unit should be doing in my room.

So, I am really, really interested to measure the zero hour N7 in my room. Its either going to measure spectacular compared to my 9500, or it may measure the same. But with two meters, one an ID3 with the diffuser, and one a proper LUX meter, both coming to pretty much the same exact numbers, I think there is an issue with the ANSI on my projector rather than my room conditions. I also would have thought 1% ADL would measure far, far higher than I have here...

Here are the numbers anyway out of interest. I think I will send this 9500 back to have its light engine replaced before I sell it (if I decide to sell it). Its funny, because I am NOT seeing anything on screen at all which leads me to believe this is a low ANSI projector, the contrast generally is pretty damned stunning....

-0 Iris Short Throw
All numbers are LUX

---------

Full On/Off

White - 1171.484
Black - 0.028
Contrast: 41,838:1

----------

1% ADL

White: 1218.432 LUX
Black: 0.142
Contrast: 8580:1

----------

2% ADL

White: 1220.591
Black: 0.259
Contrast: 4712:1

----------

3% ADL

White: 1221.576
Black: 0.326
Contrast: 3747:1

----------

4% ADL

White: 1221.780
Black: 0.478
Contrast: 2556:1

----------

5% ADL

White: 1223.042
Black: 0.621
Contrast: 1969:1

----------

10% ADL

White: 1222.361
Black: 1.167
Contrast: 1047:1

----------

20% ADL

White: 1216.309
Black: 2.566
Contrast: 474:1

----------

50% ANSI Center Square Only

White: 1247.491
Black: 8.145
ANSI Contrast: 153:1

----------
Quote:
Originally Posted by Purple X View Post
From the pictures you posted of your setup, your speakers looked like they were picking up a fair amount of reflected light off the screen. Did you try covering the sides close to the screen with velvet?

Also just noticed that you have a white ceiling past the velvet area? do you draw velvet over that? Remember back before LCOS was invented and the big argument was ANSI Contrast of DLP vs LCD.... the conclusion back then was that anything past 100 for the 50% pattern was not so easy to achieve in real world projections and that on off contrast was more important.

I think your 150 reading is pretty good
Quote:
Originally Posted by Javs View Post
Nah, I didn't do this at the screen.

Did it here, speakers made no difference. Even held up a velvet sheet behind the meter and barely a difference in readings. The couch had a black blanket over it, and even my projector shelf was covered in fabric! Something is up with my unit I think.



Fabric on shelf - Old photo (those glass lamps are not there now)

Quote:
Originally Posted by Javs View Post
Mate, the whole entire room closes into a velvet tent! The ceiling and all...

When its closed, I literally cant see a damned thing until I turn a light on... Been there done it all,

165 (the max I was able to measure) is not good, I should be getting well north of 200, simply because I then began to measure the full ANSI from the lens itself from only inches away... velvet lined box, velvet sheet hung behind... I should get max ANSI from that position even in a white room with the lights off I was using so much velvet.

My unit has degraded some how. I think its the particles inside the lens. We will see how the N7 measures, if its well into the 200's at the very least I will know it was an issue with my 9500, if its the same, well, then both my meters are rubbish or my room is significantly less good than I thought (which I kinda doubt!)

Taken with flash on camera..

Spoiler!



Black sheet (not seen here) hung on the shelf kills ALL those reflections from the equipment. I tested it with the 'flashlight' test and with the sheet hung, passes with flying colours.

Spoiler!


Flash

Spoiler!


Same shot No Flash lol

Spoiler!
Quote:
Originally Posted by Purple X View Post
Yeap. I always thought you had the ceiling covered, just couldn't quite make out the rails in the shot before.



Right... I would concur with your observation in that case. But did JVC ever publish ANSI contrast specs? I was pleasantly surprised that they started claiming double ANSI for the new units. but did they say what the actual number is? I think the 1-5% patterns are more important to real world PQ anyways.



lol
Quote:
Originally Posted by Javs View Post
The JVC's should be between 220 and 290:1 depending on iris position. Cine4Home has measured significant quantities of them spanning generations. I have never seen one measure under 200 anywhere personally let alone 165.

The ANSI on the new models is meant to be 50% higher... In relation to the Z1 I think, weather that applies to each and every model in the line rather than just NX9 is another question mark.
Quote:
Originally Posted by Soulnight View Post
Did you remove all the dust from the lens itself. It can make a huge difference!

Do you have an i1 display pro in addition to your minolta t10?

Edit: yes you did try with the iD3.

Did you make 100% sure than you have fully closed/block the gap between the diffuser and the rest of the casing (black cloth with an elastic all around for example)? If not, all the black values you will measure will get wrong with reflected light entering the sensor from this "tiny" gap.
Quote:
Originally Posted by Javs View Post
Yes.

I don't have a Minolta t10.

And yes, I actually taped the gap closed all around on the idea, It's not just that meter anyway, both measured the same thing.
Quote:
Originally Posted by Soulnight View Post
Well if you have perfectly cleaned the lens on the outside, it may well be the inside dust.

Plus you could have a low ansi performer to begin with with maybe only 200:1 Ansi max zoom, Iris open.


I also never measured a x9×××. Maybe the ultimate on-off comes at the cost of Ansi contrast?
Generally when they are unit to unit variation: both Ansi and on-off are a trade... Higher on-off performer gives a lower ansi contrast performer.

Nobody with a x7500 or x7900 in your area?

Another idea:
Are you using a particular calibration eating some contrast away?
Quote:
Originally Posted by Vladimirovich View Post
Did you make measurements when buying?
Can you compare them?
Quote:
Originally Posted by Javs View Post
Never measured ANSI on this projector.

My old Sony in my old apartment with white ceiling and rear walls was measuring 120:1 and the way that room was treated is a joke compared to this one.

On/off Contrast is the same as new.
Quote:
Originally Posted by ARROW-AV View Post
Matt I am thinking that it is most probably a lower range ANSI Contrast unit being exacerbated by the dust particles inside the light engine.

If you do return it to JVC to have the light engine replaced it will be interesting to see what difference it makes to the readings.

Regarding your measurements, is this with or without the Dynamic Contrast functionality enabled?

Quote:
Originally Posted by Javs View Post
Without of course.

Yeah I will only get this unit fixed if I decide to sell it. Which means the N7 would stay in that case.

Otherwise I will probably not bother send it back. In a honestly is still thows an excellent I age and new light engine means new lens.
Quote:
Originally Posted by ARROW-AV View Post
Do you ever watch movies on your projector with the Dynamic Contrast disabled?

Either way it would be useful to see what are the same measurements with the Dynamic Contrast enabled

And personally I would not risk replacing the light engine to gain a slight increase in ANSI but potentially lose your golden lens sample, even if selling it. Tough call though

Quote:
Originally Posted by bobof View Post
Maybe it's a limit of the measurement technique / tools?
I've never measured ANSI before; but I just flipped my Discus colorimeter round to the lens, put the diffuser on and threw up an ANSI and inverted ANSI pattern and took sequential measures just in the bottom left centre square. The meter is positioned so the small diffuser window is in the centre of the square, and it isn't measuring much of the square.
I'm at almost minimum throw, my room isn't perfect (all walls fabric but not velvet, I even have a load of purple on the wall) but I got the following on my new X7900 without any effort to cover the screen or anything:

Measurements in nits with diffuser down (so pretty arbitrary, but fine for CR calc)
Iris 0 110 / 0.37 = 297:1
Iris -8 68.5 / 0.24 = 285:1
Iris -15 17.3 / 0.068 = 253:1

One think which stands out looking at the ANSI squares on the screen; is that there is a lot of bloom at the edges of the dark squares - and I think this is in the lens system as opposed to being contamination at the screen. If your measurement device is very close to the lens then you won't be taking in just the centre area of the square, you'll be taking in quite a lot of this bloom which will kill the black measurement and hence the ANSI ratio. My meter reads dark pretty well and it has a very small diffuser (about 1cm) so it isn't taking in much area at all at that position.

Anyway, interesting measurement. I'm glad I've done this now as I'll be able to track if it has changed over the life of the unit.

Pics of the setup in the room:

Quote:
Originally Posted by markmon1 View Post
How interesting that in the photo, the shadow of the sensor in the white is lighter than the black squares. I'd expect it to be opposite.
Quote:
Originally Posted by ARROW-AV View Post
This is precisely what I was alluding to when we were previously discussing methods of measuring ANSI and why I expressed my preference to measure off the lens at or closer to the screen as opposed to right close to the lens... I remain unconvinced that you can take accurate measurements of the centre of the ANSI rectangles 1 foot away from the lens... but I'm going to evaluate this myself with respect to my own setup and see what's what in this regard.

@Javs can you get your environment absolute pitch black? If so perhaps try taking measurements off the lens closer to the screen so as to ensure your readings are not being contaminated by blooming?

Quote:
Originally Posted by bobof View Post
In real life I think it is - I reckon it might be a trick of the camera, maybe it is polluting the area with the white.

I just moved my sensor to close to the edge of a square but still well within the square and the diffuser is fully covered. In doing that my readings changed for iris 0 to 109/0.52 = 209:1.
Quote:
Originally Posted by Soulnight View Post
Your measurements make sense. :-)
But again this is a x7900.

Maybe the x9900 has higher on-off but lower Ansi. I never measured one myself and did not find a review which did.

Measuring the center black square of the modified ansi needs of course to be well centered and not to measure the edges.

But if you set up your sensor to measure the 1% Adl white square, that's a given since it's significantly smaller than the Ansi field.

Edit: and our adl contrast pattern "pairs " are not susceptible to this "blooming " phenomenon since there is no white field directly close to the center, and also not vertically below or above.
Quote:
Originally Posted by bobof View Post
Perhaps; I'm sure for the reasons discussed before of bright corners etc that any measurements which are averaging a lot of the screen are a bit of an issue (unless that's what you're trying to do, to get a "truer" view of the contrast displayed, rather than the best case 1cm... :-p )

It is all an interesting subject; if bloom were bad but you had a sensor which can read very tightly you could potentially have good ANSI measurements but poor real world performance...

I did do a test close to the edge of a square and it was much worse but not as bad as Javs sees. However I am at min throw / max zoom which should give me an ANSI boost over him, so it might account for it.
Quote:
Originally Posted by bobof View Post
Was I using the wrong image perhaps then? I did just use a standard ANSI chart, not a modified one with a centred measurement field - so my measurement is just one of the offset images.

I saw someone anecdotally posting, claiming the x9XXX had much better ANSI, but when I asked about it of course there were no actually measurements... But I've also never seen any good measurements, which is a shame.

I'd have thought at fully open though it would probably be similar, can't imagine it taking this much of a hit. I thought much of the contrast gain in the x9XXX is an even darker set of fully closed iris positions.
Quote:
Originally Posted by bobof View Post
Of course I guess to some extent it depends on what you're trying to show. If you're looking to see how well black levels in a scene hold up in general if there is some bright areas elsewhere on the screen, what you're doing there makes perfect sense. If you're trying to get a feeling from a number how contrast will work within objects where there are brighter and darker edges closer together it's not so useful.

It's not a criticism by the way, just observation. It's always really interesting to think about exactly what a stimulus can show you and what its limitations might be. In working hard in the design of the images to avoid the effects of blooming you're perhaps giving projectors which exhibit this kind of phenomenon an advantage over units which can control it better but have worse native black levels.
Quote:
Originally Posted by ARROW-AV View Post
Yes and no. Bloom would have to be extraordinarily bad for that to be the case. It is extremely common for there to be an element of blooming with the ANSI checkerboard with respect to projectors, just as it is extremely common for there to be non-uniform luminance resulting in differences in ON/OFF contrast measured at the center as compared with the corners. Wherein the fact of the matter is that if you are not measuring within the centre of the ANSI rectangles but are capturing some of the blooming then your ANSI measurements will be inaccurate. Hence you will need to move the meter closer towards the screen until this is not the case.

This is why ANSI is comparatively much more tricky to measure properly and accurately as compared with ON/OFF. With ON/OFF you can measure much closer to the lens than ANSI before this becomes a problem. And because you can't get as close to the lens when measuring ANSI it is also vital to make sure that the room is not also contaminating the measurements


JVC X9500 (RS620) | 120" 16:9 | Marantz AV7702 MkII | Emotiva XPA-7 | DIY Modular Towers | DIY TPL-150 Surrounds | DIY Atmos | DIY 18" Subs
-
MadVR Settings | UHD Waveform Analysis | Arve Tool Instructions + V3 Javs Curves
Javs is online now  
Sponsored Links
Advertisement
 
post #2 of 123 Old 10-06-2018, 06:55 PM - Thread Starter
AVS Forum Special Member
 
Javs's Avatar
 
Join Date: Dec 2012
Location: Sydney
Posts: 7,129
Mentioned: 415 Post(s)
Tagged: 0 Thread(s)
Quoted: 6061 Post(s)
Liked: 5465
More...

Quote:
Originally Posted by Javs View Post
No, but in regards to measurements, and comparing with pretty much the rest of the world, I turn the DI off because thats how it should be measured. I will do the ADL patterns with the DI on though and see what happens, I am guessing nothing at all will happen past 1%.

And I would not want to give somebody this 9500 with its dust in the lesns and low ANSI, I would feel bad about that honestly. I am torn on what to do there.


Thanks for doing that, howvever without sounding like I am beating a dead horse, cmon guys, I tried to get good readings for a good hour, that includes putting the meter in every conceivable position, I measured well inside the boxes, nowhere near the edges. It so happens I got the best readings in about the middle of the room, with a black blanket over my sofa, and the black sheet over my rear shelf but that was only 15:1 ANSI higher than what I was getting at the screen, so I think that was the simply just the best my projector can do.

I am also at min throw by the way, also the interesting thing about the image above is your sofa looks to be leather like mine, that was the last question mark for me. I can see your couch is reflecting about the same amount of light back off the black leather. So I am even more at a loss here.


Blloming is not the issue mate, yes the room is as black as I can get it, every surface is black, I held up velvet sheets behind the meter even to completely blot out the projector light from even reaching the screen. The meter was well inside the ANSI boxes.

There are a lot of meter positions between 1ft away from the lens and right at the screen I tried them all. Best results were in the middle of the room for me about 2m from the lens, and about 1.6m from the screen.

The sofa was the last thing in the room whcih could be blacker, but I had a black blanket on it

Short of having a MUCH larger room with velvet treatment so light falloff is better due to size, not sure what else I could do here...

Pretty sure Ekki has measured dozens of them. I have found two of his in depth reviews which seem to state its around 290:1 spanning X9000 and X9500 model ranges.... I was expecting a number in that range, and bobof above seems to have hit that. I dont believe the contrast being high is substituted with considerably lower ANSI. Honestly, if that were true, we would have heard this by now.

I am definitely sure I measured your ADL patterns as well as I could, the white box was clearly bigger than the meter, and the sensor was well in the middle.

I am also at min throw, there should be no boosting compared to me


I think I have exhausted the options for making my room more black for this test. Holding up sheets of triple velvet, having black sheets over all my AVR gear elimitnating reflective surfaces... blankets on the sofa... I dont know mate, I really think my JVC is just defective regarding ANSI contrast. My room looks about the same as the Bobof image above to the T when I take these readings. So I dunno.

I have no problem measuring dynamic contrast up to and above 350k:1.

I can measure and blend three measurements each of three different speakers in a three-way to achieve an anechoic response, and design a crossover for them using software which looks like it was written for a NASA engineer, I dont mean to sound arrogant, but ANSI measurements are not hard by comparison whatsoever....

This would be the first time I have read about ANSI taking a nose dive with JVC though, I will say my lens particles have not gone unnoticed on my machine, maybe I should clean the lens again... Funny there is a lot of talk about not needing lens covers and not touching the lens, well I beg to differ!!!

@zombie10k any luck measuring your RS600?

Paper Test... Only the blue orb on the second image is barely visible at the screen, however I didnt even measure that section, I measured many squares, none performed where I expected.



Quote:
Originally Posted by bobof View Post
Sorry, it wasn't meant to seem like criticism. I didn't try and do anything to my room to make it better like covering stuff up - those were literally the first measurements that came out; yes it is a leather sofa.

Apart from anything else I don't want to believe your contrast is that low as obviously it means it could happen to any of us! As I say I'm glad I've measured mine now as I'll be able to tell if it ever happens. How many hours you got on yours?
Quote:
Originally Posted by Javs View Post
I have 1300 hours now....

My room is relatively dusty though in general. Not much I can do about that to be honest. Australia, Dual purpose room with balcony overlooking a lake... I should have measured the ANSI when it was new. I think I have had those dust partucles since about hour 200... So perhaps there was some kind of event that took place which deposited the dust there.

I am happy to see the new models have a proper dust filter on the back which looks like it will actually work. The current models have one on the bottom which is just wire mesh and basically pathetic.

By comparison, my on/off contrast is pretty spectacular really. over 40k iris wide open is pretty great IMO... So I guess this is only affecting my ANSI.
Quote:
Originally Posted by bobof View Post
Yup, dust filter is a joke - it will stop a spider getting in and that's about it. Still makes me laugh the amount of folk (many professional reviewers) claiming these have a sealed design...

I'm a little behind you for on / off - measures a little under 40K, but to be expected I guess for the lower model.

I'm not sure what I'm more upset about... that your unit's had a decay in ANSI, or that now I've measured ANSI off lens and off screen I've discovered I'm going from 295:1 at the lens to 130:1 at the screen... grr! The rabbit hole I'm about to disappear into... My sofa is really close to the screen and the whole width of the room, I'm guessing that is a major contributor to the contrast loss.

At least 130:1 is higher than the recommendations in the DCI Spec V2 (100:1 min):
http://www.dcimovies.com/archives/sp...emSpecv1_2.pdf

Do you measure much worse off the screen?
Quote:
Originally Posted by Javs View Post
In your photos when you had the meter facing the lens, is that when you got 295? Or did you move it a lot closer?

Your 130 screen measurement, was that with your meter facing the screen itself and not the projector?

I didnt measure facing the screen, thats almost always going to yield worse results IMO.

My sofa is exactly the same scenario as yours, its the width of the small room, and my eyeballs are exactly 3m from the 120" screen so it sounds like its about the same distance too.
Quote:
Originally Posted by bobof View Post
No, that photo is when I got 295; much closer to screen than to lens. The photo was taken at the exact position as I was measuring. It is probably around 0.8m from screen and about 2m from lens.
For the 130 screen measurement I turned the sensor and faced the screen with diffuser off. I had hoped it might be better than 130, seems like @Soulnight 's ANSI measures in their supertent are massively higher; but I'm a little confused at how folk are measuring this. I confirmed with 3 meters - i1d3, Discus and JETI spectro (laser aiming is cool!). @Soulnight ; in this image (which is explained as being poor contrast due to poor room) how is the sensor oriented? I can't see it here, whether it is facing the screen or the lens?
Quote:
Originally Posted by Javs View Post
Its facing the lens in that image.
Quote:
Originally Posted by Aztar35 View Post
Javs, I measured 368:1 at the lens on my X990.

By the way, how the heck did you get dust in your lens?
Quote:
Originally Posted by Gary Lightfoot View Post
Two or three years ago there was a video of one of the new JVCs being compared with a Sony and possibly an Epson (can't remember). but they did a 'down and dirty' ANSI CR measurement using just a single white and black square in the centre of the screen. The JVC measured in the mid to low 100s IIRC, which was clearly not representative of what JVCs can do, but as we had no way to communicate with the people doing the testing, we didn't know if it was a bad unit or bad measuring (and no way to get them to double check. It was surprising they just took the reading as it was without double checking tbh). It could have been one or the other or both, but given what you've discovered, it could have been a similar low mechanical ANSI due to an internal issue.

As most people don't measure their projectors there could be more examples like this, albeit a very small percentage. A friend of mine for example had asked me about measuring his JVC to see what kind of lumens he was getting (possibly because he felt it wasn't bright enough - again, can't remember), so I told him what light meter to get and how to measure it - he found it was lower than spec by a fair margin, and he was able to get it replaced. The replacement measured closer to spec. Looks like you've been a bit unlucky on this occasion in that respect, but i think it goes to prove that for movies, ANSI isn't as visually important as on/off can be, but is more of a 'nice to have'. That's not to say it's not important or can't improve the image in some circumstances, but until you measured it I don't think you felt you had any problems there did you?
Quote:
Originally Posted by Aztar35 View Post
Well is there any official literature from the company on what it says is the ANSI spec?
Quote:
Originally Posted by Aztar35 View Post
And he has a good sample 990. It'll hammer any of the new ones when it comes to contrast performance.
Quote:
Originally Posted by Gary Lightfoot View Post
Most companies don't quote ANSI CR and I think JVC is the same. We usually go from reliable sources like Cine4home and Kris Deering for example.
Quote:
Originally Posted by Aztar35 View Post
Did they post anything on the current series' ANSI?
Quote:
Originally Posted by Javs View Post
Mm I dont really believe that number, thats the highest for an eshift JVC ever if so.

You might have some dust too, ... look inside your lens, do a defocuses paper test and see. these light engines are not sealed that well.

As I explained, I live in Australia, near a lake, with a balcony door in my dual purpose room... its not a dedicated temperature and dust controlled room. People around here have fire pits, while I dont run the projector when people next door are having a fire, or with the door open, you never know, could have been a single event where loads of dust made its way into my room and the projector sucked it all in when I powered it up one day.
Quote:
Originally Posted by Javs View Post
Correct, the ANSI in my room when all the curtains are pulled closed feels pretty darned good if Im honest. I certainly notice the low APL contrast performance a lot more often. About 6 months ago though, I did find it odd that I was not able to measure the ANSI spec numbers, ive tried on a few occasions now, but honestly never really thought much of it, because the checkerboard does look pretty nicely black to me.
Quote:
Originally Posted by Gary Lightfoot View Post
The test was two or three years ago so was current for then, but I've no idea if they've done the same thing recently (unless you meant the current eshift range, in which case bobof's numbers are ballpark).

I'm sure someone will be testing the ANSI sometime soon so hopefully we won't have to wait too long and then we'll know how much of an ANSI improvement the new (4K) models have over the eshift.
Quote:
Originally Posted by Aztar35 View Post
Might be the way I measured. I set the glass meter four feet from the lens. Mine was a good sample to say the least...look at that lens in my avatar...come on now; isn't it a beaut...? Then again, I measured 920:1 ANSI contrast on the BenQ HT9050 LED XPR unit...so perhaps my meter is inflating... or not?

Fire pits...? That's scary that these are not hermetically sealed. Are the new ones going to be the same? I mean, I knew the LCDs were risks for dust blobs but these too?
Quote:
Originally Posted by Aztar35 View Post
Got it.
Quote:
Originally Posted by Gary Lightfoot View Post
I never trust colorimiters for measuring contrast (ANSI or on/off) and always use a light meter, but I think that stems from the days when meters weren't as good as they are now and it's just stuck with me (a bit like never emptying the cup when drinking tea or coffee due to tea leaves that get through the strainer - before tea bags were a common thing). Even so, your Benq numbers do seem a bit high as well.
Quote:
Originally Posted by Javs View Post
Yeah sounds like your meter is out of whack with either white or black readings, its either topping out, or bottoming out.

My LUX meter was 30% out after only 1 year! I returned it to the store and compared it to four other meters in store spanning two different meter designs because I was highly suspect of it, sure enough mine was well out and the other 3 all were identical readings essentially. Remember when I said my 9900's were dim as heck? Well that was my meter. Once I swapped it out for a known good one, it read correctly. In saying that, my LUX and my Colorimeter currently seem to agree on contrast measurements and both come up with ANSI numbers within 2 or 3:1 points of each other, so, maybe yours is a bit out on one of the extreme ends of the scale. Even 10% out on either black or white will significantly change the contrast number you get from it.

In any case, you may not / probably dont have dust inside, I dont want to cause alarm here. I just know my room needs a dust down every week or two, so its only going to be a matter of time it seems to me, unless there was an 'event' (balcony door open once when people had fires or the guy across the road was sanding, or even maybe when I was sanding down in the garage!). I cant say for sure if it was that, or a time thing.

The new model, I may take extra steps to beef the whole thing up in that regard, I may well put foam on every intake at the rear (not the front) to try and mitigate anything potentially happening again. Sure as heck my new home and theatre build hopefully beginning next year will have totally controlled environment.

Regardless, I've already looked into it when I first discovered this, and it is indeed a warrantable issue, so JVC would go ahead and replace the light engine for me, I just want to make sure its not a unit I would keep if I did this, because based on the lenses I saw on the new models, I dont believe I would get back a sample as good as what I give them in regards to lens.
Quote:
Originally Posted by bobof View Post
I "don't get it" then if that is what the numbers presented for ANSI all represent then in the projectiondreams stuff. I thought that the ideal room measurements were done facing the lens at the lens, and the living room and velvet tent were done measuring the screen - so I was comparing myself to those.

I get trying to measure the PJ ANSI contrast facing the lens to compare the projector capabilities to another unit. I get trying to see how good your "system ANSI" is by measuring off the screen. I don't get measuring ANSI facing the lens in a suboptimal room.
Quote:
Originally Posted by Javs View Post
His ideal room were done at the lens very close to the lens negating the effects of the room itself (I tried this too), and his optimised and white room measurements were done facing the PJ back at the screen (I did this too), so his measurements actually did include what effects his room has on the measurements.

His suboptimal room measurements are to show the need for a highly treated room, I have even used his data set to argue with an unnamed member here about how a projectors ANSI performance is pretty much irrelevant if your system ANSI cannot pass a certain number. Part of the argument that, for the most part, ANSI is not super important in a reasonable room. In a CRAZY room with insane treatments, sure, having a PJ over 500:1 ANSI at the screen is going to matter, not even my room would reach that level. I really, really think my room is probably a good 350:1 in performance if I had a machine here which could do it.

What we know from your measurements, is your system ANSI actually outperforms your projector, which is GOOD, because it means you are not leaving any ANSI performance on the table as it were due to room compromises. I would expect most/all eshift JVC's to measure 290:1 at wide open iris, and that's exactly what yours did.

Personally I think measuring ANSI off the screen is very difficult. You have many factors, you have the meter blocking light from even hitting the square you are trying to measure, and you have light scatter with close by white boxes potentially skewing your readings due to light scatter on the screen surface and kind of leaking into the light sensor. Unless your meter is almost literally kissing the screen without somehow measuring its own shadow, then the measurements are going to be HIGHLY un-repeatable. I doubt you would come up with identical numbers if you ripped down the whole rig and tried to do it again 10 minutes later. Whereas facing the lens, but meter placed at the screen, will be highly repeatable.

JVC X9500 (RS620) | 120" 16:9 | Marantz AV7702 MkII | Emotiva XPA-7 | DIY Modular Towers | DIY TPL-150 Surrounds | DIY Atmos | DIY 18" Subs
-
MadVR Settings | UHD Waveform Analysis | Arve Tool Instructions + V3 Javs Curves
Javs is online now  
post #3 of 123 Old 10-06-2018, 07:27 PM
AVS Forum Special Member
 
Gary Lightfoot's Avatar
 
Join Date: Jun 2000
Location: UK
Posts: 6,231
Mentioned: 31 Post(s)
Tagged: 0 Thread(s)
Quoted: 1162 Post(s)
Liked: 870
I think my light meter was just fine but the colorimiter was where the inaccuracy was and not the other way round - my contrast readings were always pretty much repeatable for the pj I would be measuring over time and when I compared them against Greg Rogers numbers they were in the same ballpark. The thing with the older colorimiters was getting a good black reading and as you say, it doesn't take much of an error there to skew the readings one way or the other. Taking contrast readings from the screen was pretty much guaranteed to give you a poor black reading (meter could be effectively 14 feet or more from the light source). You really need to have a good solid reading for black that isn't near the meters extremes.

I remember one guy saying that contrast went down as the lamp aged and gave very low measured readings - what was happening was his meter was bottomed out for black, and although the white and black levels were reducing as the lamp aged, the black didn't really change but the white was reducing and that was reading more accurately because it was brighter, so the white level was reducing but the black level was staying the same. Being a fair distance from the screen meant the numbers for white were pretty low too. It never occurred to him that was happening so his conclusion was that contrast reduced as the lamp aged and by quite a lot. I never found that with my projectors, or a great change in the lamps RGB output which may have altered contrast readings after subsequent calibrations over time.

I take it this is to be a JVC thread concentrating on ANSI and on/off CR readings?

Quote:
Originally Posted by elmalloc
Who says Cameron is "right" and why do we care about him so much - lol!

I trust Gary Lightfoot more than James Cameron.
Gary Lightfoot is offline  
Sponsored Links
Advertisement
 
post #4 of 123 Old 10-06-2018, 08:07 PM - Thread Starter
AVS Forum Special Member
 
Javs's Avatar
 
Join Date: Dec 2012
Location: Sydney
Posts: 7,129
Mentioned: 415 Post(s)
Tagged: 0 Thread(s)
Quoted: 6061 Post(s)
Liked: 5465
Quote:
Originally Posted by Gary Lightfoot View Post
I think my light meter was just fine but the colorimiter was where the inaccuracy was and not the other way round - my contrast readings were always pretty much repeatable for the pj I would be measuring over time and when I compared them against Greg Rogers numbers they were in the same ballpark. The thing with the older colorimiters was getting a good black reading and as you say, it doesn't take much of an error there to skew the readings one way or the other. Taking contrast readings from the screen was pretty much guaranteed to give you a poor black reading (meter could be effectively 14 feet or more from the light source). You really need to have a good solid reading for black that isn't near the meters extremes.

I remember one guy saying that contrast went down as the lamp aged and gave very low measured readings - what was happening was his meter was bottomed out for black, and although the white and black levels were reducing as the lamp aged, the black didn't really change but the white was reducing and that was reading more accurately because it was brighter, so the white level was reducing but the black level was staying the same. Being a fair distance from the screen meant the numbers for white were pretty low too. It never occurred to him that was happening so his conclusion was that contrast reduced as the lamp aged and by quite a lot. I never found that with my projectors, or a great change in the lamps RGB output which may have altered contrast readings after subsequent calibrations over time.

I take it this is to be a JVC thread concentrating on ANSI and on/off CR readings?
I thought about putting JVC in the title, but after thinking bout it, actually, I welcome all measurements from all brands honestly. I think the ADL charts which will be used in this thread, and the ones in my first post will be useful for anybody with a projector. And people with similar issues (not getting good ANSI and ADL numbers) will require solutions that would apply to all models.

I would like it if this thread ended up containing a decent database on brands, models, and contrast ADL measures for all of them.

The Projection Dream ADL patterns are here BTW:

http://www.dropbox.com/s/jybiypic3c...terns.zip?dl=0

JVC X9500 (RS620) | 120" 16:9 | Marantz AV7702 MkII | Emotiva XPA-7 | DIY Modular Towers | DIY TPL-150 Surrounds | DIY Atmos | DIY 18" Subs
-
MadVR Settings | UHD Waveform Analysis | Arve Tool Instructions + V3 Javs Curves

Last edited by Javs; 10-07-2018 at 02:42 AM.
Javs is online now  
post #5 of 123 Old 10-07-2018, 01:47 AM
Advanced Member
 
Soulnight's Avatar
 
Join Date: Dec 2011
Location: Germany
Posts: 999
Mentioned: 123 Post(s)
Tagged: 0 Thread(s)
Quoted: 733 Post(s)
Liked: 840
Quote:
Originally Posted by Javs View Post


His ideal room were done at the lens very close to the lens negating the effects of the room itself (I tried this too), and his optimised and white room measurements were done facing the PJ back at the screen (I did this too), so his measurements actually did include what effects his room has on the measurements.

His suboptimal room measurements are to show the need for a highly treated room, I have even used his data set to argue with an unnamed member here about how a projectors ANSI performance is pretty much irrelevant if your system ANSI cannot pass a certain number. Part of the argument that, for the most part, ANSI is not super important in a reasonable room. In a CRAZY room with insane treatments, sure, having a PJ over 500:1 ANSI at the screen is going to matter, not even my room would reach that level. I really, really think my room is probably a good 350:1 in performance if I had a machine here which could do it.

What we know from your measurements, is your system ANSI actually outperforms your projector, which is GOOD, because it means you are not leaving any ANSI performance on the table as it were due to room compromises. I would expect most/all eshift JVC's to measure 290:1 at wide open iris, and that's exactly what yours did.

Personally I think measuring ANSI off the screen is very difficult. You have many factors, you have the meter blocking light from even hitting the square you are trying to measure, and you have light scatter with close by white boxes potentially skewing your readings due to light scatter on the screen surface and kind of leaking into the light sensor. Unless your meter is almost literally kissing the screen without somehow measuring its own shadow, then the measurements are going to be HIGHLY un-repeatable. I doubt you would come up with identical numbers if you ripped down the whole rig and tried to do it again 10 minutes later. Whereas facing the lens, but meter placed at the screen, will be highly repeatable
Exactly like you described. All 3 are measured toward the lens with our adl measurement at projectiondream.

Perfect room is with black tent and at the lens.

Optimized room is 5cm from the screen but facing the lens with black velvet tent closed.


White room is 5cm from the screen but facing the lens with black velvet tent open with white walls and ceiling.

I pretty much got the exact contrast number measuring off the screen for the optimized room and white room with our minolta ls100 with a 30cm pipe flocked on the inside with protostar.

In that case, the minolta is on a tripod at the sitting position 4m away from the screen and facing the screen for very contrast projector. It's set up to measure 1% Adl as well.

Note: the minolta bottoms out for on-off measurements of the screen. 1% adl are generally correct.

The i1 display pro colorimeter we use for contrast measurements combined with slow mode from chromapure offer very repeatable measurements even for very dark measurements.

On-off contrast is measured pretty much identical close to the lens (very bright) or at screen facing the lens (pretty dark) with our optimized room and the iD3 with diffuser on.

Last edited by Soulnight; 10-07-2018 at 01:56 AM.
Soulnight is online now  
post #6 of 123 Old 10-07-2018, 02:14 AM
AVS Forum Special Member
 
Join Date: Aug 2015
Location: Norwich, UK
Posts: 1,850
Mentioned: 19 Post(s)
Tagged: 0 Thread(s)
Quoted: 1507 Post(s)
Liked: 662
Quote:
Originally Posted by Gary Lightfoot View Post
I think my light meter was just fine but the colorimiter was where the inaccuracy was and not the other way round - my contrast readings were always pretty much repeatable for the pj I would be measuring over time and when I compared them against Greg Rogers numbers they were in the same ballpark. The thing with the older colorimiters was getting a good black reading and as you say, it doesn't take much of an error there to skew the readings one way or the other. Taking contrast readings from the screen was pretty much guaranteed to give you a poor black reading (meter could be effectively 14 feet or more from the light source). You really need to have a good solid reading for black that isn't near the meters extremes.

I remember one guy saying that contrast went down as the lamp aged and gave very low measured readings - what was happening was his meter was bottomed out for black, and although the white and black levels were reducing as the lamp aged, the black didn't really change but the white was reducing and that was reading more accurately because it was brighter, so the white level was reducing but the black level was staying the same. Being a fair distance from the screen meant the numbers for white were pretty low too. It never occurred to him that was happening so his conclusion was that contrast reduced as the lamp aged and by quite a lot. I never found that with my projectors, or a great change in the lamps RGB output which may have altered contrast readings after subsequent calibrations over time.

I take it this is to be a JVC thread concentrating on ANSI and on/off CR readings?
I guess with all measurements it's about understanding the meter capabilities and your technique. i1d3 is good for at least 1000 nits and down to 0.005 nits if you know what settings to use for integration times etc (the low readings particularly need a longer integration time to be reliable). This should let you measure up to around 200,000:1 comfortably if you move your meter so peak white is towards the top of the range. I've had readings as low as 0.0028 out of an i1d3, but not sure I believe it.

I've got a BasICColor Discus which is good for 2500 nits and measures reliably for luminance down to about 0.002 nits. Should be able to register a CR of over 1,000,000:1 on that. Both the above saturation nits can be achieved easily without being right up at the lens (for me they're about 50cm away). Out of that meter I've read as low as 0.0008 nits, not sure I trust that though.

Of course this is no-where near the 29,000,000:1 capability of the Minolta T1 (assuming you could saturate it close enough to the lens). I've not looked into working out whether that is the case or not

I can't see why this would be JVC only. Admittedly the biggest problems for CR ratio measurements are going to be faced by JVC owners because the black level at screen is well below the colorimeter capabilities for reading off screen for most commonly owned meters. It will be a shame if only JVC owners contributed and got benefit as folk coming to a good common understanding of techniques would be good for everyone talking about this.

Quote:
Originally Posted by Javs View Post
His ideal room were done at the lens very close to the lens negating the effects of the room itself (I tried this too), and his optimised and white room measurements were done facing the PJ back at the screen (I did this too), so his measurements actually did include what effects his room has on the measurements.

His suboptimal room measurements are to show the need for a highly treated room, I have even used his data set to argue with an unnamed member here about how a projectors ANSI performance is pretty much irrelevant if your system ANSI cannot pass a certain number. Part of the argument that, for the most part, ANSI is not super important in a reasonable room. In a CRAZY room with insane treatments, sure, having a PJ over 500:1 ANSI at the screen is going to matter, not even my room would reach that level. I really, really think my room is probably a good 350:1 in performance if I had a machine here which could do it.

What we know from your measurements, is your system ANSI actually outperforms your projector, which is GOOD, because it means you are not leaving any ANSI performance on the table as it were due to room compromises. I would expect most/all eshift JVC's to measure 290:1 at wide open iris, and that's exactly what yours did.

Personally I think measuring ANSI off the screen is very difficult. You have many factors, you have the meter blocking light from even hitting the square you are trying to measure, and you have light scatter with close by white boxes potentially skewing your readings due to light scatter on the screen surface and kind of leaking into the light sensor. Unless your meter is almost literally kissing the screen without somehow measuring its own shadow, then the measurements are going to be HIGHLY un-repeatable. I doubt you would come up with identical numbers if you ripped down the whole rig and tried to do it again 10 minutes later. Whereas facing the lens, but meter placed at the screen, will be highly repeatable.
I think I'd use different terminology - I don't see think system ANSI should be referred to as anything other than the ANSI you can measure directly off the screen with appropriately good equipment and technique; so I think my System ANSI is worse than my projector ANSI. My projector is clearly capable of at least 295:1 in spite of the room; it wouldn't surprise me if I worked at the setup eliminating reflections off the back and front wall that I couldn't get much higher. Yet from seating I'm only getting the benefit of 130:1.

Reading from the screen without issue "just" depends on having an accurate enough meter with a narrow enough field of view so it can be situated outside the projected image area and only measuring the very centre of the ANSI square, on axis. I have a JETI 1201 which optically matches but the sensitivity is barely good enough for 130:1 measurement, though it did match exactly the results from my Discus off screen.

What I don't understand (yet, will do some tests) is how comparable the measurement technique done by @Soulnight for "system influenced projector ANSI" (i1d3 facing lens, from close to screen) would be to a perfect off-screen measure with suitably capable equipment. If you think about it - if the only difference is moving the sensor closer to the lens to go to the perfect room measurement; there is also a continuum in the other direction (from the measured position to being within the screen surface). The meter as shown isn't level with the screen surface - it must be about 10 cm away from the screen? I wonder how much of the room effect this distance between screen and meter removes. In the case where most of the pollution is coming from the back wall reflection it will be negligible. But in the case where pollution is coming from the walls, ceiling and floor I believe it will be significant.
bobof is online now  
post #7 of 123 Old 10-07-2018, 02:27 AM
Advanced Member
 
Soulnight's Avatar
 
Join Date: Dec 2011
Location: Germany
Posts: 999
Mentioned: 123 Post(s)
Tagged: 0 Thread(s)
Quoted: 733 Post(s)
Liked: 840
Quote:
Originally Posted by bobof View Post

Reading from the screen without issue "just" depends on having an accurate enough meter with a narrow enough field of view so it can be situated outside the projected image area and only measuring the very centre of the ANSI square, on axis. I have a JETI 1201 which optically matches but the sensitivity is barely good enough for 130:1 measurement, though it did match exactly the results from my Discus off screen.

What I don't understand (yet, will do some tests) is how comparable the measurement technique done by @Soulnight for "system influenced projector ANSI" (i1d3 facing lens, from close to screen) would be to a perfect off-screen measure with suitably capable equipment. If you think about it - if the only difference is moving the sensor closer to the lens to go to the perfect room measurement; there is also a continuum in the other direction (from the measured position to being within the screen surface). The meter as shown isn't level with the screen surface - it must be about 10 cm away from the screen? I wonder how much of the room effect this distance between screen and meter removes. In the case where most of the pollution is coming from the back wall reflection it will be negligible. But in the case where pollution is coming from the walls, ceiling and floor I believe it will be significant.
I already answered above.

Measurement at 5cm from the screen toward the lens with the i1d3 colorimeter or from the sitting position with my tweaked minolta ls100 with a 30cm protostar pipe give the same contrasts! :-)

With your jeti, you also need a pipe or light will scatter within your sensor lens and your contrast off the screen will top out at 150:1 like it did with the minolta without pipe.
bobof likes this.
Soulnight is online now  
post #8 of 123 Old 10-07-2018, 02:32 AM
AVS Forum Special Member
 
Join Date: Aug 2015
Location: Norwich, UK
Posts: 1,850
Mentioned: 19 Post(s)
Tagged: 0 Thread(s)
Quoted: 1507 Post(s)
Liked: 662
Quote:
Originally Posted by Soulnight View Post
Exactly like you described. All 3 are measured toward the lens with our adl measurement at projectiondream.

Perfect room is with black tent and at the lens.

Optimized room is 5cm from the screen but facing the lens with black velvet tent closed.


White room is 5cm from the screen but facing the lens with black velvet tent open with white walls and ceiling.

I pretty much got the exact contrast number measuring off the screen for the optimized room and white room with our minolta ls100 with a 30cm pipe flocked on the inside with protostar.

In that case, the minolta is on a tripod at the sitting position 4m away from the screen and facing the screen for very contrast projector. It's set up to measure 1% Adl as well.

Note: the minolta bottoms out for on-off measurements of the screen. 1% adl are generally correct.

The i1 display pro colorimeter we use for contrast measurements combined with slow mode from chromapure offer very repeatable measurements even for very dark measurements.

On-off contrast is measured pretty much identical close to the lens (very bright) or at screen facing the lens (pretty dark) with our optimized room and the iD3 with diffuser on.
Thanks for this; I'm really interested to try and replicate your measurements. So just thinking about your setup with the i1d3 at screen; you say 5cm off the screen. I meaure i1d3 body height at something like 65mm with diffuser down (diffuser is inset, so perhaps 62mm from base to diffuser surface). You must have had a tripod base on the meter (unless you custom mounted it somehow with the base close to the screen surface) so the diffuser must be at least 62mm off the screen, if not 62mm + tripod base + any air gap left for safety between tripod base and screen surface. What was the distance between the diffuser and the screen surface?

Useful info that the Minolta flocked tube test confirmed your results facing the screen. It is confirmation I guess that the method holds for at least your screen + room combinations.

Of course for anyone with an AT screen (like me) there is additional complication - my screen surface could be responsible for some ANSI loss if there is light passing through and back, and that would definitely only be measurable facing the screen.
bobof is online now  
post #9 of 123 Old 10-07-2018, 02:36 AM
Advanced Member
 
Soulnight's Avatar
 
Join Date: Dec 2011
Location: Germany
Posts: 999
Mentioned: 123 Post(s)
Tagged: 0 Thread(s)
Quoted: 733 Post(s)
Liked: 840
@bobof

Our studies at projectiondream are to convinced people that optimizing the room is priority number 1. :-)

That's why we have 3 curves:

White room
Optimized room
Projector (perfect room)

Our optimized room got about 90% of the max ansi contrast on the screen.

When measuring screen or room performance, direct measurements from the screen with our tweaked minolta ls100 is what we do. :-)
bobof and BattleAxeVR like this.
Soulnight is online now  
post #10 of 123 Old 10-07-2018, 02:43 AM
AVS Forum Special Member
 
Join Date: Aug 2015
Location: Norwich, UK
Posts: 1,850
Mentioned: 19 Post(s)
Tagged: 0 Thread(s)
Quoted: 1507 Post(s)
Liked: 662
Quote:
Originally Posted by Soulnight View Post
@bobof

Our studies at projectiondream are to convinced people that optimizing the room is priority number 1. :-)

That's why we have 3 curves:

White room
Optimized room
Projector (perfect room)

Our optimized room got about 90% of the max ansi contrast on the screen.

When measuring screen or room performance, direct measurements from the screen with our tweaked minolta ls100 is what we do. :-)
Your excellent data is one of the things that got me started with trying to build a good room (within the realms of acceptable to my wife and useable by my kids). I'm trying to figure where I am in my achievement...

Quote:
Originally Posted by Soulnight View Post
I already answered above.

Measurement at 5cm from the screen toward the lens with the i1d3 colorimeter or from the sitting position with my tweaked minolta ls100 with a 30cm protostar pipe give the same contrasts! :-)

With your jeti, you also need a pipe or light will scatter within your sensor lens and your contrast off the screen will top out at 150:1 like it did with the minolta without pipe.
I will build something like this to see the effect and figure it out, first going to measure to the lens at the screen surface to see how much difference I have.
bobof is online now  
post #11 of 123 Old 10-07-2018, 03:22 AM
AVS Forum Special Member
 
Join Date: Aug 2015
Location: Norwich, UK
Posts: 1,850
Mentioned: 19 Post(s)
Tagged: 0 Thread(s)
Quoted: 1507 Post(s)
Liked: 662
So that was interesting; did the first step - Discus basically at the screen surface.
48.5/0.21 nits = 231:1 vs 295:1 measured ~80cm closer to lens (setup below in images).
I do wonder how much worse it would be if the sensor was completely flush to the screen surface.

AM really intrigued by these differences. Particularly by the difference from 130:1 off screen to 231:1 more or less at screen. I wonder how much is light scatter in the reading off screen vs any AT screen contrast reducing effect vs differences in room effect capture due to not measuring co-planar with screen surface.

Sounds like a job for a weekend when I manage to offload the wife and kids on someone...!

Images in spoiler of setup
Spoiler!
Attached Thumbnails
Click image for larger version

Name:	DSC_5025.JPG
Views:	360
Size:	207.5 KB
ID:	2465154   Click image for larger version

Name:	DSC_5023.JPG
Views:	339
Size:	343.4 KB
ID:	2465156   Click image for larger version

Name:	DSC_5024.JPG
Views:	336
Size:	393.6 KB
ID:	2465158  
bobof is online now  
post #12 of 123 Old 10-07-2018, 03:25 AM - Thread Starter
AVS Forum Special Member
 
Javs's Avatar
 
Join Date: Dec 2012
Location: Sydney
Posts: 7,129
Mentioned: 415 Post(s)
Tagged: 0 Thread(s)
Quoted: 6061 Post(s)
Liked: 5465
So I cleaned my lens quite thoroughly.

The highest I can get at the lens now is about 220:1. This is more on the outer edges of the ANSI chart. The middle right it's the worst which I think is due to the dust particle which forms a rather large orb on the screen.

Anyways still not quite what it should be but not so much the disaster it was before at least i can consistently measure over 200:1 at the lens now.

In saying that I need to clean the lens even better, there seems to be a very fine film which is just not coming off. It's kinda like when you try clean your phone screen and you can't quite get it completely flawless clean unless you use screen cleaner.

Might need to invest in some little microfibre pads. Seems like I will need to use a solution of some kind too.
bobof likes this.

JVC X9500 (RS620) | 120" 16:9 | Marantz AV7702 MkII | Emotiva XPA-7 | DIY Modular Towers | DIY TPL-150 Surrounds | DIY Atmos | DIY 18" Subs
-
MadVR Settings | UHD Waveform Analysis | Arve Tool Instructions + V3 Javs Curves
Javs is online now  
post #13 of 123 Old 10-07-2018, 03:38 AM
AVS Forum Special Member
 
Join Date: Aug 2015
Location: Norwich, UK
Posts: 1,850
Mentioned: 19 Post(s)
Tagged: 0 Thread(s)
Quoted: 1507 Post(s)
Liked: 662
Quote:
Originally Posted by Javs View Post
So I cleaned my lens quite thoroughly.

The highest I can get at the lens now is about 220:1. This is more on the outer edges of the ANSI chart. The middle right it's the worst which I think is due to the dust particle which forms a rather large orb on the screen.

Anyways still not quite what it should be but not so much the disaster it was before at least i can consistently measure over 200:1 at the lens now.

In saying that I need to clean the lens even better, there seems to be a very fine film which is just not coming off. It's kinda like when you try clean your phone screen and you can't quite get it completely flawless clean unless you use screen cleaner.

Might need to invest in some little microfibre pads. Seems like I will need to use a solution of some kind too.
Glad you are getting a better result. I have some little clean room manufactured microfiber squares which are typically used for cleaning DSLR sensors.
Might be worth investing in.

Edit: link didn't work. Google Dust-Cloth MF.
Attached Thumbnails
Click image for larger version

Name:	dustclothmf.jpeg
Views:	22
Size:	16.2 KB
ID:	2465162  
Javs likes this.

Last edited by bobof; 10-07-2018 at 03:48 AM.
bobof is online now  
post #14 of 123 Old 10-07-2018, 03:56 AM - Thread Starter
AVS Forum Special Member
 
Javs's Avatar
 
Join Date: Dec 2012
Location: Sydney
Posts: 7,129
Mentioned: 415 Post(s)
Tagged: 0 Thread(s)
Quoted: 6061 Post(s)
Liked: 5465
Quote:
Originally Posted by bobof View Post
Glad you are getting a better result. I have some little clean room manufactured microfiber squares which are typically used for cleaning DSLR sensors.
Might be worth investing in.

Edit: link didn't work. Google Dust-Cloth MF.
Thanks heaps. Will look into that.

JVC X9500 (RS620) | 120" 16:9 | Marantz AV7702 MkII | Emotiva XPA-7 | DIY Modular Towers | DIY TPL-150 Surrounds | DIY Atmos | DIY 18" Subs
-
MadVR Settings | UHD Waveform Analysis | Arve Tool Instructions + V3 Javs Curves
Javs is online now  
post #15 of 123 Old 10-07-2018, 04:47 AM
Advanced Member
 
Soulnight's Avatar
 
Join Date: Dec 2011
Location: Germany
Posts: 999
Mentioned: 123 Post(s)
Tagged: 0 Thread(s)
Quoted: 733 Post(s)
Liked: 840
Quote:
Originally Posted by Javs View Post
So I cleaned my lens quite thoroughly.

The highest I can get at the lens now is about 220:1.
Happy to hear you followed my advice and that it did help! :-)

We had the same experience of "dirty " lens with Anna & Ekki with a Sony Vw5000es measuring surprising low Ansi contrast (sub 300:1). After a bit of cleaning we were back at 330:1

Since then, that's always one of the first thing that comes into my mind for a "used" projector measuring on the low side for Ansi contrast.

For full new projector which we usually review, the good thing is you don't have to worry about that.

But always be careful not to scratch your lens when cleaning.

First blow some air on the lens to remove as much dust as you can.
Then with camera grade microfiber cloth, gently clean the lens. I am always worried about using chemicals.
bobof likes this.

Last edited by Soulnight; 10-07-2018 at 04:51 AM.
Soulnight is online now  
post #16 of 123 Old 10-07-2018, 04:54 AM
AVS Forum Special Member
 
Join Date: Aug 2015
Location: Norwich, UK
Posts: 1,850
Mentioned: 19 Post(s)
Tagged: 0 Thread(s)
Quoted: 1507 Post(s)
Liked: 662
Quote:
Originally Posted by Javs View Post
Thanks heaps. Will look into that.
No worries. They're pricey but considering you'll probably use one every six months it comes out in the wash - the bag I have is about 5 years old and still got tons.
For cleaning fluid for the lens surface itself it is a hotly debated subject re: alcohol and ammonia content... Most recently i had used Carl Zeiss alcohol wipes after blowing any dust off but they're a little streaky... leading to having to polish more than you'd want to remove streaks.
Maybe something like ROR - as you have what you think is a film over the lens:
http://www.wexphotovideo.com/ror-op...spray-1033171/
I might pick up a bottle myself as WEX are based in my city.
bobof is online now  
post #17 of 123 Old 10-07-2018, 06:07 AM
AVS Forum Special Member
 
markmon1's Avatar
 
Join Date: Dec 2006
Posts: 3,249
Mentioned: 51 Post(s)
Tagged: 0 Thread(s)
Quoted: 2660 Post(s)
Liked: 1539
Quote:
Originally Posted by Javs View Post
So I cleaned my lens quite thoroughly.

The highest I can get at the lens now is about 220:1. This is more on the outer edges of the ANSI chart. The middle right it's the worst which I think is due to the dust particle which forms a rather large orb on the screen.

Anyways still not quite what it should be but not so much the disaster it was before at least i can consistently measure over 200:1 at the lens now.

In saying that I need to clean the lens even better, there seems to be a very fine film which is just not coming off. It's kinda like when you try clean your phone screen and you can't quite get it completely flawless clean unless you use screen cleaner.

Might need to invest in some little microfibre pads. Seems like I will need to use a solution of some kind too.
Ok so is there any noticeable difference in actual picture quality though?

Video: JVC RS640 135" screen in pure black room no light, htpc nvidia 1080ti.
Audio: Anthem mrx720 running 7.1.4, McIntosh MC-303, MC-152, B&W 802d3 LR, B&W HTM1D3 center, B&W 805d3 surround, B&W 702S2 rear, B&W 706s2 x 4 shelf mounted for atmos, 2 sub arrays both infinite baffle: 4x15 fi audio running on behringer ep4000 + 4x12 fi audio running on 2nd ep4000.
markmon1 is online now  
post #18 of 123 Old 10-07-2018, 06:58 AM
AVS Forum Special Member
 
Join Date: Aug 2015
Location: Norwich, UK
Posts: 1,850
Mentioned: 19 Post(s)
Tagged: 0 Thread(s)
Quoted: 1507 Post(s)
Liked: 662
Quote:
Originally Posted by markmon1 View Post
Ok so is there any noticeable difference in actual picture quality though?
Interesting question given the improvement @Javs achieved is around the same order of magnitude as being claimed on the new JVC series vs outgoing (+50%)

Sent from my E5823 using Tapatalk
bobof is online now  
post #19 of 123 Old 10-07-2018, 07:49 AM
AVS Forum Special Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 5,438
Mentioned: 88 Post(s)
Tagged: 0 Thread(s)
Quoted: 3735 Post(s)
Liked: 1164
Quote:
Originally Posted by bobof View Post
AM really intrigued by these differences. Particularly by the difference from 130:1 off screen to 231:1 more or less at screen. I wonder how much is light scatter in the reading off screen vs any AT screen contrast reducing effect vs differences in room effect capture due to not measuring co-planar with screen surface.
Even if you had a zero-thickness meter, you will not get the same results "off the screen" and "at the screen" (facing the projector lens). The screen picks up reflections in all directions, which will be read by the meter facing the screen, whereas at the screen most meters will read predominantly light coming from the front (relative to the meter), even with the diffuser on. If you use a light meter with a semi-dome diffuser the results will be much closer, but most colorimeter diffusers are flat.

Last edited by Dominic Chan; 10-07-2018 at 07:53 AM.
Dominic Chan is offline  
post #20 of 123 Old 10-07-2018, 08:48 AM
Advanced Member
 
Soulnight's Avatar
 
Join Date: Dec 2011
Location: Germany
Posts: 999
Mentioned: 123 Post(s)
Tagged: 0 Thread(s)
Quoted: 733 Post(s)
Liked: 840
Quote:
Originally Posted by Dominic Chan View Post
Even if you had a zero-thickness meter, you will not get the same results "off the screen" and "at the screen" (facing the projector lens). The screen picks up reflections in all directions, which will be read by the meter facing the screen, whereas at the screen most meters will read predominantly light coming from the front (relative to the meter), even with the diffuser on. If you use a light meter with a semi-dome diffuser the results will be much closer, but most colorimeter diffusers are flat.
Well, we did measure both and we almost measured the same at and from the screen.

With the optimized room and with the white room.
(Of course this only works for a lambertian plain screen and a real diffuser)

A diffuser diffuses all the the light coming from 180 degree. And then the sensor reads the brightness from the diffuser.

It is not taking predominantly the light from the front or your diffuser is not a diffuser and is defective. What you describe is what a ND grey filter would do.

Xrite even describe the use of the diffuser to know the ambient light level in your room.

Last edited by Soulnight; 10-07-2018 at 08:55 AM.
Soulnight is online now  
post #21 of 123 Old 10-07-2018, 09:59 AM
AVS Forum Special Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 5,438
Mentioned: 88 Post(s)
Tagged: 0 Thread(s)
Quoted: 3735 Post(s)
Liked: 1164
Quote:
Originally Posted by Soulnight View Post
It is not taking predominantly the light from the front or your diffuser is not a diffuser and is defective.
A flat diffuser follows a cosine response curve:



Quote:
Originally Posted by Soulnight View Post
A diffuser diffuses all the the light coming from 180 degree.
On top of the cosine response, the i1Display Pro diffuser completely cuts off past +/-80 degrees, presumably due to the raised lip around the diffuser.

bobof likes this.

Last edited by Dominic Chan; 10-07-2018 at 11:11 AM.
Dominic Chan is offline  
post #22 of 123 Old 10-07-2018, 10:35 AM
AVS Forum Special Member
 
Join Date: Aug 2015
Location: Norwich, UK
Posts: 1,850
Mentioned: 19 Post(s)
Tagged: 0 Thread(s)
Quoted: 1507 Post(s)
Liked: 662
I must say I'm a little sceptical at the moment; the main reason being that my room isn't anywhere near as well executed a @Soulnight 's - I have black fabric (Camira Lucia, AT and not velvet or matte) on the walls and ceilings for the first 1.5m, a painted black door within this area which is quite reflective, the rest of the room ceiling is painted black (a bit reflective), the back wall and back 2/3 of the side walls are purple fabric. Despite this I'm still measuring 231:1 at the screen pointing to the lens with diffuser, 295:1 a bit closer to the PJ, but 130:1 pointing at the screen.

The reason for my scepticism is that despite the above room limitations my measurement with the meter facing the lens at the screen with diffuser down is more or less the same as the Projection Dreams velvet tent, (231 vs 235:1, you'd be glad to achieve measurement error of that order). This is comparing my X7900 to their X5000 though, this may be unfair - I've not seen an ANSI comparison of X5 vs X7 series units.

I wonder if this measurement technique works in the Projection Dream tent only because the tent is very resistant to causing ANSI reduction from reflections off the side walls - so it doesn't matter if the technique using the diffuser isn't measuring much of the reflected light off the walls as there isn't any in the first place. In my case I wonder if the meter is effectively rejecting much of the light reflected off the walls giving me a much better measurement facing the lens than the reality.

I've picked up some adhesive backed velvet today to make a stray light extinguising tube to see where I get to with "best practice" off screen measurements.

Has anyone seen any good information on negative ANSI effects of AT screens anywhere?
bobof is online now  
post #23 of 123 Old 10-07-2018, 11:18 AM
Advanced Member
 
Soulnight's Avatar
 
Join Date: Dec 2011
Location: Germany
Posts: 999
Mentioned: 123 Post(s)
Tagged: 0 Thread(s)
Quoted: 733 Post(s)
Liked: 840
Quote:
Originally Posted by Dominic Chan View Post
A flat diffuser follows a cosine response curve:
Very interesting. Source for the id3 diffuser?

Still, we got about the same measurement with the tweaked minolta ls100 with the protostar velvet pipe and with the iD3 facing the screen.

And that both with and without triple black velvet tunnel/tent.

We measured only so slightly lower of the screen than at the screen which is expected.
bobof likes this.
Soulnight is online now  
post #24 of 123 Old 10-07-2018, 11:26 AM
Advanced Member
 
Soulnight's Avatar
 
Join Date: Dec 2011
Location: Germany
Posts: 999
Mentioned: 123 Post(s)
Tagged: 0 Thread(s)
Quoted: 733 Post(s)
Liked: 840
Quote:
Originally Posted by bobof View Post
I must say I'm a little sceptical at the moment; the main reason being that my room isn't anywhere near as well executed a @Soulnight 's - I have black fabric (Camira Lucia, AT and not velvet or matte) on the walls and ceilings for the first 1.5m, a painted black door within this area which is quite reflective, the rest of the room ceiling is painted black (a bit reflective), the back wall and back 2/3 of the side walls are purple fabric. Despite this I'm still measuring 231:1 at the screen pointing to the lens with diffuser, 295:1 a bit closer to the PJ, but 130:1 pointing at the screen.

The reason for my scepticism is that despite the above room limitations my measurement with the meter facing the lens at the screen with diffuser down is more or less the same as the Projection Dreams velvet tent, (231 vs 235:1, you'd be glad to achieve measurement error of that order). This is comparing my X7900 to their X5000 though, this may be unfair - I've not seen an ANSI comparison of X5 vs X7 series units.

I wonder if this measurement technique works in the Projection Dream tent only because the tent is very resistant to causing ANSI reduction from reflections off the side walls - so it doesn't matter if the technique using the diffuser isn't measuring much of the reflected light off the walls as there isn't any in the first place. In my case I wonder if the meter is effectively rejecting much of the light reflected off the walls giving me a much better measurement facing the lens than the reality.

I've picked up some adhesive backed velvet today to make a stray light extinguising tube to see where I get to with "best practice" off screen measurements.

Has anyone seen any good information on negative ANSI effects of AT screens anywhere?
Even without the tunnel velvet, the measurement from and at the screen match each other.

~50:1 ansi for the white room.

We also measured at a friend home theater directly off the AT screen with our minolta ls100, with and without triple velvet diffuser on the ceiling, and with and without triple velvet curtains on the side.

Surprisingly, the Ansi was already pretty high on the screen with even the black paint. It also got slightly higher with the velvet.

What the triple black velvet mostly changed is immersion for the viewer.

Btw. the contrast gain was much bigger for the ceiling than for the sides.
bobof likes this.

Last edited by Soulnight; 10-07-2018 at 11:29 AM.
Soulnight is online now  
post #25 of 123 Old 10-07-2018, 12:15 PM
AVS Forum Special Member
 
Join Date: Aug 2015
Location: Norwich, UK
Posts: 1,850
Mentioned: 19 Post(s)
Tagged: 0 Thread(s)
Quoted: 1507 Post(s)
Liked: 662
Quote:
Originally Posted by Soulnight View Post
Even without the tunnel velvet, the measurement from and at the screen match each other.

~50:1 ansi for the white room.

We also measured at a friend home theater directly off the AT screen with our minolta ls100, with and without triple velvet diffuser on the ceiling, and with and without triple velvet curtains on the side.

Surprisingly, the Ansi was already pretty high on the screen with even the black paint. It also got slightly higher with the velvet.

What the triple black velvet mostly changed is immersion for the viewer.

Btw. the contrast gain was much bigger for the ceiling than for the sides.
Ok, I am ready to eat my humble pie (or whatever the German equivalent is! )

Rolled up a piece of dc-fix black velvet inside a couple of toilet roll tubes (high tech solution...) and connected it to the front of the JETI - the only probe I have which I'm confident has tight enough viewing to be outside the image shadow and only read centre of ANSI square).

I upped it to high lamp as I'm running out of range at the bottom on the JETI (JETI isn't specced to read that low but if you're happy to wait 145s it does return readings). Now measured 105 / 0.47 = 224:1, which is I guess within a whisker of the measurement of 231:1 facing the lens at the screen. So the method does seen to correlate for my room also.

Pretty pleased with that then, I've managed to have some colour in the room on the walls, not use velvet and still have a good result for ANSI contrast. As I say it is all fabric but just not velvet fabric. Thanks @Soulnight for all the posts and the colossal efforts with your original tests
bobof is online now  
post #26 of 123 Old 10-07-2018, 04:22 PM - Thread Starter
AVS Forum Special Member
 
Javs's Avatar
 
Join Date: Dec 2012
Location: Sydney
Posts: 7,129
Mentioned: 415 Post(s)
Tagged: 0 Thread(s)
Quoted: 6061 Post(s)
Liked: 5465
Quote:
Originally Posted by markmon1 View Post
Ok so is there any noticeable difference in actual picture quality though?
Quote:
Originally Posted by bobof View Post
Interesting question given the improvement @Javs achieved is around the same order of magnitude as being claimed on the new JVC series vs outgoing (+50%)
Hmm cant say for sure. That pattern looked about the same to me after I cleaned the lens, my wife and I even laughed about it.

I did watch Jurassic World, and thought it looked overall fantastic, but thats pretty normal, it always looks fantastic


JVC X9500 (RS620) | 120" 16:9 | Marantz AV7702 MkII | Emotiva XPA-7 | DIY Modular Towers | DIY TPL-150 Surrounds | DIY Atmos | DIY 18" Subs
-
MadVR Settings | UHD Waveform Analysis | Arve Tool Instructions + V3 Javs Curves
Javs is online now  
post #27 of 123 Old 10-07-2018, 06:47 PM
AVS Forum Addicted Member
 
noah katz's Avatar
 
Join Date: Apr 1999
Location: Mountain View, CA USA
Posts: 22,198
Mentioned: 10 Post(s)
Tagged: 0 Thread(s)
Quoted: 1535 Post(s)
Liked: 536
Seems to me that ANSI checkerboard isn't of much use to judge intrascene contrast by eye.

The large white areas make neighboring dark areas look subjectively black even if they're way above the black floor.

A convincing demonstration of how our eyes are fooled by adjacent-area contrast is the below; squares A and B are actually the same.

Dynamic youtube demo:
Attached Thumbnails
Click image for larger version

Name:	contrast_illusion.JPG
Views:	18
Size:	68.8 KB
ID:	2465394  

Noah
noah katz is offline  
post #28 of 123 Old 10-07-2018, 07:08 PM
AVS Forum Special Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 5,438
Mentioned: 88 Post(s)
Tagged: 0 Thread(s)
Quoted: 3735 Post(s)
Liked: 1164
Quote:
Originally Posted by noah katz View Post
Seems to me that ANSI checkerboard isn't of much use to judge intrascene contrast by eye.
Up to a point. That’s why 200:1 looks good for ANSI contrast,
Dominic Chan is offline  
post #29 of 123 Old 10-07-2018, 10:16 PM
Advanced Member
 
descalabro's Avatar
 
Join Date: Aug 2017
Posts: 743
Mentioned: 12 Post(s)
Tagged: 0 Thread(s)
Quoted: 294 Post(s)
Liked: 103
Quote:
Originally Posted by bobof View Post
No worries. They're pricey but considering you'll probably use one every six months it comes out in the wash - the bag I have is about 5 years old and still got tons.

For cleaning fluid for the lens surface itself it is a hotly debated subject re: alcohol and ammonia content... Most recently i had used Carl Zeiss alcohol wipes after blowing any dust off but they're a little streaky... leading to having to polish more than you'd want to remove streaks.

Maybe something like ROR - as you have what you think is a film over the lens:

http://www.wexphotovideo.com/ror-op...spray-1033171/

I might pick up a bottle myself as WEX are based in my city.


There’s one highly regarded photography expert on YouTube that simply uses distilled water and 70% isopropyl alcohol if there’s any kind of grease to clean. He uses pure cotton swabs.

http://youtu.be/yH-N6-lzOgM


Sent from my iPhone using Tapatalk
descalabro is offline  
post #30 of 123 Old 10-07-2018, 10:39 PM
AVS Forum Special Member
 
markmon1's Avatar
 
Join Date: Dec 2006
Posts: 3,249
Mentioned: 51 Post(s)
Tagged: 0 Thread(s)
Quoted: 2660 Post(s)
Liked: 1539
Can someone say what ADL does for actual picture quality and how it differs from ANSI contrast?

Video: JVC RS640 135" screen in pure black room no light, htpc nvidia 1080ti.
Audio: Anthem mrx720 running 7.1.4, McIntosh MC-303, MC-152, B&W 802d3 LR, B&W HTM1D3 center, B&W 805d3 surround, B&W 702S2 rear, B&W 706s2 x 4 shelf mounted for atmos, 2 sub arrays both infinite baffle: 4x15 fi audio running on behringer ep4000 + 4x12 fi audio running on 2nd ep4000.
markmon1 is online now  
Sponsored Links
Advertisement
 
Reply Digital Hi-End Projectors - $3,000+ USD MSRP



Forum Jump: 

Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off