Measured On/off Contrast Ratios

[The_Preacher1973]

First a bit of background.

It has generally been regarded that looking at a projector's contrast ratio and other specifications is relatively meaningless. This is due to the fact that manufacturers do not all measure their projectors the same way. They "cheat" by measuring the high and low levels using different settings on the projector and not having them calibrated for ideal viewing (D65K).

As such, it was thought that ANSI contrast ratio was the best way of measuring CR as it was produced "in scene" thus making the manufacturer's tricks redundant.

Then along came the JVC HD1/RS1 with a claimed on/off contrast ratio of 15,000:1 without the use of an iris and calibrate to D65K. Everybody who saw it raved about how good the picture was, especially regarding "depth of field" and having a 3 dimensional look. The fly in the ointment was the fact that the HD1/RS1 had quite a poor ANSI ratio when compared to some of the competition that it was supposedly besting in side by side comparisons.

The conclusion that has since been drwn is that on/off contrast ratio is more important than ANSI contrast ratio when viewing low light scenes (an ANSI pattern is said to be 50% APL - average picture level - as it is 50% white and 50% black).

Since most movies predominantly feature scenes with low APL's (less than 10%), on/off contrast ratio is actually more relevant to how a projector will "look" during a movie than ANSI contrast ratio.

That is, so long as you are measuring the native contrast ratio of a projector at D65K.

Having looked at the starfield comparison bewtween the Benq 8720 and W10000 in the 1080/720 comparison thread, I thought it would be interesting to see how the measured contrast ratio of various generations and types of technology stacked up.

The best source I've found for this information is from cine4home. I went through thier English reviews and compiled a list of measured on/off contrast ratios. It provided some surprising results.

All of the figures below are taken with the projector calibrated to D65K and iris's disabled. Where more than one number existed (eg. iris oper or iris closed, low power/high power) I took the higheest measured CR.

Here they are:

1. JVC HD1/RS1 15,200:1

2. Sony VW100 - aka "RUBY" 6,000:1

3. Sony VW50 - "Pearl" 3,400:1

4. Mitsubishi HC3100 (DC3) 3,000:1

5. Mitsubishi HC1100 (DC2) 2,200:1

6. Sony HS60 1,250:1

7. Sanyo Z4 1,100:1

8. Mitsubishi HC5000 720:1

Now obviously CR is only one factor in the overall PQ of a projector. But it is a very important one and the above table does show why the JVC is the "IT" projector of the moment.

The one projector that seems to be out of place on that list is the HC5000. I really expected the D6 1080P panels to outperform the D5's in the Sanyo Z4. Hopefully Cine4home will get either the Panasonic or Epson 1080P offerings in for testing so we can see whether it's an issue with the panels or Mitsubushi's implimentation.

If anybody has some links to other independent tests I'd greatly appreciate it if you could post them here. If they seem rigorous enough, I'll add them to the table.

[benthx]

Thanks Preacher.

Your knowledge and background with CRT lends credance to your information. Very valuable information and thanks. I too get a bit confused at times

Ben

[The_Preacher1973]

Thanks Preacher.

Your knowledge and background with CRT lends credance to your information. Very valuable information and thanks. I too get a bit confused at times

Ben

I'm sure you guys are going to be absolutely stoked with your HD1's. As the above table shows, the JVC has really come from nowhere to produce some really stunning results. Of course, there are no numbers for the new 1080P DLP's in that table yet. When you see how much better the W10000 is than the 8720 in that starfield shot, it's got to be much greater than the 3000:1 achieved by the 720P DC3 chips.

And once again you're really only going to notice the difference bewteen these sort of contrast ratios in an ideal viewing environment. Black tracksuits and faces covered in black shoe polish for all your guests Ben!

[Foggy]

Another classic Preacher post, great job

I found some more measured On/Off CR's on the Secrets Site:

InFocus 4805 (480p DLP DC2) - 1,550:1

Marantz VP12S4 (720p DLP DC3) - 1,200:1

Painasschronic AE-700 (720p LCD) - 400:1

Painasschronic AE-900 (720p LCD) - 757:1 (Not certain if this is at D65k though)

Painasschronic AX-100 (720p LCD) - 1,320:1 (Not certain if this is at D65k though)

Samsung SP-H710AE (720p DLP DC2) - 1,390:1 (Note: Jason Turk from AVS measured 2564:1)

Sharp XV-Z21000 (1080p DLP DC2+) - 16,000:1

EDIT: Here's a few more from Jason Turk of AVScience's reviews....

InFocus IN76D (720p DLP DC2) - 1,599:1

Sim2 C3X (3 chip DLP 720p) - 4,585:1

Edited February 10, 2007 by Foggy

[The_Preacher1973]

Another classic Preacher post, great job

I found some more measured On/Off CR's on the Secrets Site:

Thanks for the Secrets link Foggy. Some excellent information in there. One thing I noticed though was that you had a very low contrast ration for the Marantz 12S4 (1200:1) This prompted me to read the review and I found the following:

After we calibrated the projector we measured contrast ratios. For On/Off contrast ratio we used a 100IRE full field window and a 0IRE full field window. With the bulb in economy mode and the iris closed down we measured about 2400:1 for contrast.

As such I’ve used this value in my updated table. I’m glad that it was much higher, otherwise my whole perspective on this would have been thrown out of whack!

So, updating the list with this new information gives the following:

Make/Model Contrast Source S=Secrets C=cine4home AVS=AVS

1. Sharp XV-Z21000U 16000:1 S

2. JVC HD1/RS1 15200:1 C

3. Sony VW100 6000:1 C

4. Sim2 C3X Lite 4585:1 AVS

5. Sharp XV-Z12000U 4394:1 S

6. Sony VW50 3400:1 C

7. Mitsibushi HC3100 3000:1 C

8. Marantz 12S4 2400:1 S

9. Mitsubishi HC1100 2200:1 C

10. Optoma H79 2050:1 S

11. Benq 8700 1795:1 S

12. Infocus IN76 1599:1 AVS

13. Infocus 4805 1550:1 S

14. Samsung SP-H710AE 1390:1 S

15. Panasonic AX100 1320:1 S

16. Sony HS60 1250:1 C

17. Optoma HD72 1100:1 S

18. Sanyo PLV-Z4 1100:1 C

19. Panasonic DW 5000U 1036:1 S

20. Panasonic AE900 757:1 S

21. Sanyo PLV-Z1 750:1 S

22. Mitsubishi HC5000 720:1 C

23. Sanyo PLV-Z2 600:1 S

24. Sanyo PLV-Z3 545:1 S

25. Sanyo PLV-70 516:1 S

26. Panasonic AE500 500:1 S

27. Panasonic AE700 400:1 S

28. Hitachi TX100 384:1 S

29. Panasonic AE300 316:1 S

30. Sony 10HT 130:1 S

Now I said earlier that on/off contrast ratio is only one aspect of a projectors picture quality, but the more I look at this list and compare it with the projectors that I’ve actually seen, my preference follows this list exactly.

For example, the first projector that I saw (and the one that made me first realize that I could afford to get something worthwhile for a HT) was the Infocus 4805. And here we are several years later and it’s still sitting up there at number 13 on the list with a lot of more recent big names still trailing behind it. No wonder so many owners raved about this unit at the time.

I also quite liked the Panasonic AE500 when it first came out. I was then convinced that the AE700 was going to be the first projector that I was going to buy as it reviewed as being a big improvement over the AE500 (which I had already seen and liked). All that changed when I finally got to see an AE700 (I actually saw about 5 or 6 of them) and thought it looked terrible in comparison with the AE500. Look at the list. The static on/off of the AE700 is actually worse than that of the AE500.

Later on at work we then got an AE900 which was installed in the conference room next to the one that housed the AE700. I wasn’t expecting anything much from this projector after the disappointment of the AE700 and the fact that reviews had stated that the AE900 was only a slight evolution over the AE700. To my surprise it was much better. I could have lived with an AE900 but an AE700 was just way too “flat” for my tastes. Once again look at the list. The AE900 has a static on/off nearly twice that of the AE700 and believe me it shows.

And then we have my Sim2 SVD-800HD. It probably has an on/off around the 15,000:1 mark (I haven’t tried the gamma tweak yet) and once again it blew away every other projector I have ever seen. Now with the coming of the JVC HD1/RS1 which also has this same on/off CR as the Sim2, we now have just about everybody who’s seen it saying it has the best image they have seen from a projector. Ever.

All of this can’t be a coincidence. As far as I’m concerned, I’ve now found what to me is probably the most important characteristic for a projector with regards to producing a dynamic, life like picture.

[glennb]

no Epson results as yet?

[The_Preacher1973]

no Epson results as yet?

Unfortunately not. Cine4home did review the TW600 but the only contrast ratio measurement they published still had the auto iris operating. They gave the projector 6-8 seconds to react to the difference between a black and white screen before taking measurements (apparently the iris takes up to 20 seconds to fully clos/open). They got 2000:1 doing this so the real on/off will be significantly lower than that. I'm not sure why they tested this way.Perhaps there's no way to disable the auto iris on the TW600?

Anyway, here's hoping they get the new Epson 1080P in soon. It's been shipping for a few months now but I haven't seen any professional reviews on it yet. Just some user reports on AVS.

[glennb]

The auto-iris on the TW600 can be disabled, which is what i've done as it's way too noisey to be used. Would be interested in how they other new Epsons stack up as well.

[Foggy]

I'm still stumped about the HC5000 measurements. I know a couple of guys who are very knowledgeable on projection, and both say that the black levels on a HC5000 are excellent.

I really think that the cine4home measurement is an anomaly brought about by an error in their process, or a faulty projector.

[Gino]

I think so too... the HC3100 performed better yet the manufacturers specs aren't as good as the ir specs for the HC5000

[DonR1503559521]

As a newbie on all this technical stuff, how high can the CR numbers go? It seems to me that there is a big difference between CR's of say 500 to 2000 but when you're talking about 15000:1, how close is that that to the limits a human eye can discern?

[c912039]

As a newbie on all this technical stuff, how high can the CR numbers go? It seems to me that there is a big difference between CR's of say 500 to 2000 but when you're talking about 15000:1, how close is that that to the limits a human eye can discern?

As a cynic, I would suggest that most users will tell you that the human eye can discern a max value CR that equals the CR of their 'pride and joy projector'

[gobberon]

I'm still stumped about the HC5000 measurements. I know a couple of guys who are very knowledgeable on projection, and both say that the black levels on a HC5000 are excellent.

I really think that the cine4home measurement is an anomaly brought about by an error in their process, or a faulty projector.

This is whats bugging me too foggy. But the apparently bad CR quoted looks like it has been done under worst case conditions maybe pushing the limits of the DI too much. Still, if you look at the projectorreviews.com review, the reviewer throws up the menu and naturally we see a drop in CR also, but without seeing it with my own eyes its hard to make a call - for me personally anyway.

I know its making my life hell trying to make a decision because I also like the 8720 with its good native CR.

[c912039]

I'm still stumped about the HC5000 measurements. I know a couple of guys who are very knowledgeable on projection, and both say that the black levels on a HC5000 are excellent.

I really think that the cine4home measurement is an anomaly brought about by an error in their process, or a faulty projector.

Actually, this is the sort of numbers I would expect for a HC5000.

After all, its STILL an LCD projector, and despite improvements in LCD panel design, CR and balck levls for LCD FP are still lower than DLP projectors. I guess its one of the advantages of the DLP technology - its probably easier to rreflect light off the mirrors away from the lens (to create blacks) than it is to completely block the LCD pixel.

What I am amazed about, is how the new-gen D-ILA chip can manage 15000:1 WITHOUT DI.

Im also amazed at the large gap bewteen the Sharp DLP and the CX3 Lite DLP projectors.

Regards

David

[quijibo]

That 720:1 for the HC5000 is with the adaptive iris turned off. I assume it means the iris is fully open... actually it could be closed but thats irrelevant.

With the iris turned on the contrast jumps as high as 5800:1 which would result in quite deep blacks and put it well up the tree in relation to other projectors.

[gobberon]

The native CR of the 8720 is 6200:1, has an iris but not dynamic, I can't find anything that says different. Anyone know worst case CR for it?

[c912039]

That 720:1 for the HC5000 is with the adaptive iris turned off. I assume it means the iris is fully open... actually it could be closed but thats irrelevant.

With the iris turned on the contrast jumps as high as 5800:1 which would result in quite deep blacks and put it well up the tree in relation to other projectors.

My HS60 has a native CR of 1250:1, and up to 10,000:1 using DI.

Many reviews seem to suggest that out of all projectors that use DI, Sony's DI design seems to be the best ( no operational noise, very fast, excellent CR).

Its the huge jump in CR that is prompting me to sell my HS60 and move up to the JVC.

[The_Preacher1973]

A few quick points.

As I mentioned in my first post I too was surprised by the poor ranking of the HC5000, however, the AVS review for this unit gave a static on/off of 516:1 so I don;t think cine4home's example was faulty.

I'm also beginning to think that the AVS numbers are too low, particularly with the high end units. Jason mentions that he doesn't have an ideal testing environment whihch will likely influence the results at the "top" end more than those at the bottom. Once you start trying to measure contrast ratios above 2000:1, a slight problem in measuring can give drastic results. For example AVS only got 1764:1 for the Sony VW50 (Pearl).

When I get a chance I'll split the list. I'll leave the Secrets and Cine4home results together in one table (as they were both tested under better conditions and the numbers seem to correllate quite well) and add a second table with all the AVS results.

In that case it will only be fair to compare the results of projectors in the same list. From this though you should be able to get a good idea as to how they would rank in the other table.

Also, a lot of people here seem to be confusing contrast ratio with black level. Not the same thing at all. For example, most of the reults above were taken in high brightness mode with the iris's open for maximum brightness. That is what gave the highest contrast ratio in most cases. It would not however give the best black level. What it means is that when you use the iris to darken the picture, you are reducing the brightness of bright areas more than you are improving the black level of the dark areas. This will reduce the dynamics of the picture and is commonly referred to as brightness compression and one reason why I'm not a fan of dynamic iris's.

Another example of this is the low ANSI contrast ratio of CRT's. They're down around the 130:1 mark whereas the LCD/SXRD/LCOS are around 250:1 and the DLP's are up around 600:1. This is not because the CRT's can't do black - it's because they can't do white! (and before anyone bashes me over the fact that therefore CRT's must therefore suffer from brightness compression as well - you're right. It's just that CRT's only suffer from it when the APL goes above 25% (they run out of current) which doesn't happen that much in real movies. The ANSI test pattern has an APL of 50% hence when dsiplaying the ANSI checkerboard pattern, they are suffering from brightness compression).

At the end of the day, on its own the list is fairly useless. If, however you have seen a few projectors on the list and know which one you prefer, it can help you decide what you might think about another projector by judging its relative performance against models you are familiar with.

[Foggy]

A few quick points.

hehe... yes, my mistake on the black levels statement... sorry about that chief

[The_Preacher1973]

Right. Here’s the updated lists:

Lab Tested (Secrets/Cine4home)

1. Sharp XV-Z21000U 16000:1

2. JVC HD1/RS1 15200:1

3. Sony VW100 6000:1

4. Sharp XV-Z12000U 4394:1

5. Sony VW50 3400:1

6. Mitsibushi HC3100 3000:1

7. Marantz 12S4 2400:1

8. Mitsubishi HC1100 2200:1

9. Optoma H79 2050:1

10. Benq 8700 1795:1

11. Infocus 4805 1550:1

12. Samsung SP-H710AE 1390:1

13. Panasonic AX100 1320:1

14. Sony HS60 1250:1

15. Optoma HD72 1100:1

16. Sanyo PLV-Z4 1100:1

17. Panasonic DW 5000U 1036:1

18. Panasonic AE900 757:1

19. Sanyo PLV-Z1 750:1

20. Mitsubishi HC5000 720:1

21. Sanyo PLV-Z2 600:1

22. Sanyo PLV-Z3 545:1

23. Sanyo PLV-70 516:1

24. Panasonic AE500 500:1

25. Panasonic AE700 400:1

26. Hitachi TX100 384:1

27. Panasonic AE300 316:1

28. Sony 10HT 130:1

Real World Tested (AVS)

1. Sim2 C3X Lite 4585:1

2. Benq W10000 3900:1

3. Marantz VP11S1 3642:1

4. Sharp XV-Z20000 3767:1

5. Sim2 HT3000 2326:1

6. Sony VW50 1764:1

7. Infocus IN76 1599:1

8. Mitsubishi HC5000 516:1

Now I believe the first list is compiled under more “laboratory” like conditions and the readings may in fact be taken by pointing the probe at the projector rather than the screen.

The second table is quite interesting as it gives you more of a guide as to what to expect in the real world as I believe Jason takes his measurements off the screen and does not operate in a completely blacked out room. He says his room more accurately represents how most people use their projectors.

Now someone mentioned before how much contrast ratio can we actually “see” anyway. Well as I’ve pointed out before, people will probably only see the full difference of the HD1 over the likes of the 8720 if they have a completely light controlled environment. That means dark walls, dark furniture, dark carpet and definitely no white T-shirts for people sitting in the front row! If you don’t have that then as the AVS table shows, you will probably hit the limit of your room before you hit the limit of the projector. I say this because if you look at Jason’s results, you’ll see we’re getting a cluster of results around the 4000:1 mark for the really highend gear. (Having said that, what’s the bet Jason will review the hd1/RS1 and get 10,000:1 just too totally destroy my argument! )

Once again though you can see the HC5000 trailing badly in this area. Still ,when I look at the rest of the results, they all seem to fit quite well with what I have seen personally and read about the various projectors. Look at the performance of the HS60 and the Z4 in comparison with their peers. Now cast your mind back 12 months and see how that fits with the reviews.

Let’s put it this way, if I had been asked to rank the projectors without having first tabled these results, I think my list would have been pretty close to the measured “lab” results.

hehe... yes, my mistake on the black levels statement... sorry about that chief

No problem Foggy. Go back and have a look at the AVS results as Jason gives the contrast ratios at both high power and low power. Invariably the higher contrast ratios are seen in "high power" modes.

Not what you may expect at first glance. Quite "counter intuitve" as we say in engineer speak.

(Counter intuitive in engineer speak can also translate to "We totally screwed the pooch with that design but who would have guessed it was gonna do that." )

[Santa1503559644]

As a newbie on all this technical stuff, how high can the CR numbers go? It seems to me that there is a big difference between CR's of say 500 to 2000 but when you're talking about 15000:1, how close is that that to the limits a human eye can discern?

The eye can discern a great range*, but what's important is the gradation level of that range.

No point in having a 100,000:1 contrast ratio, but applied as completely black versus blinding white.

Similarly remember (as so many dont), in the same way that its not just the darkest:lightest that's important, its also important that the levels are uniform across the light output range. Again, little point in having great, deep blacks, if the brightest levels are indistinguishable for each other, or vice versa.

Sadly, most of these factors are not available in the form of sound-bites like the "CR", so the best tool is your own eyes.

_____________________________________________

*Actually, better described as a "range of ranges".

[Gino]

Santa - I think what you are referring to is gamma per se, not the CR. Sequential or on/off CR... the higher the number the better... simple as that. on/off CR is simply a ratio of measured light output (ft lamberts or lumens) of white to black. So 100,000:1 is very good, you just need to set your brightness, contrast and gamma correctly so that you don't crush blacks or whites.

Very good article on How much on/off CR is enough

I have seen some low numbers discussed for how much On/Off CR people can see, but as I will explain below, I have done some testing with a CRT front projector where it was determined that the On/Off contrast ratio was close to 700,000:1 after tweaking. This is very good, yet still left some room for improvement.

In real life, the ratio of luminance between white paper in sunlight and white paper in moonlight is approximately 1,000,000:1, while the ratio for white paper in sunlight to white paper in starlight is approximately 100,000,000:1. However, white paper in sunlight is too bright for comfortable reading levels, and the white levels that are used with projected images in commercial theaters are much closer to those comfortable reading levels. Even bright plasma screens are much closer to comfortable reading levels than white paper out in sunlight. In a dark theater, going from the white levels we see outside and back and forth with very dark images would bother a lot of people and result in eyestrain. We generally don't go from very bright to very dark conditions and vice versa as quickly and as often in real life as many movies do. For those who want to set up with white levels for movie watching that go much brighter than comfortable reading levels and who would get eye strain from transitions between very bright levels and very dark levels, backlighting can be used to keep things from going too dark in the dark scenes. This can be something as simple as a small table lamp behind the projection screen. However, under these conditions, the On/Off CR of the projector can become mostly irrelevant as the backlighting limits the absolute black level off the screen, and realistic blackouts become impossible regardless of the capabilities of a projector.

For those happy with white levels that are close to commercial theaters or at least in that ballpark, a question then becomes how black is black enough for blackout scenes. For those who want blackouts to hold so that they cannot see the screen (this means the screen also has to be invisible in the room when the projector is off, and the rest of the room is set up for viewing), we can look at some data and then examine some test results in more detail.

The lower cone threshold (for color vision) is about at the white paper in starlight level (3.2 x 10-4 cd/m2), but we can still see about 100 times lower than that with the rods (black & white vision) in our eyes. They are thought to go down to approximately 3.2 x 10-6 cd/m2. For reference, the upper limit where rods tend to get saturated is around 3.2 cd/m2, so by themselves they have a range of about 1,000,000:1.

If we use a fairly common white level for movies of about 40 cd/m2 (a little under 12 ft-lamberts), then matching that absolute lower limit for our vision would imply an On/Off CR of 12,500,000:1. But it takes time for our eyes to adjust (including dark adaptation after entering a dark theater), and when going from bright scenes to a blackout, it doesn't take nearly that much to give the perception of no light on the screen.

[The_Preacher1973]

Santa - I think what you are referring to is gamma per se, not the CR. Sequential or on/off CR... the higher the number the better... simple as that. on/off CR is simply a ratio of measured light output (ft lamberts or lumens) of white to black. So 100,000:1 is very good, you just need to set your brightness, contrast and gamma correctly so that you don't crush blacks or whites.

Very good article on How much on/off CR is enough

Yeah, I've read that whole thing. Basicly it concludes that at times, 700,000:1 isn't quite enough!

[norpus]

Now someone mentioned before how much contrast ratio can we actually “see” anyway. Well as I’ve pointed out before, people will probably only see the full difference of the HD1 over the likes of the 8720 if they have a completely light controlled environment. That means dark walls, dark furniture, dark carpet and definitely no white T-shirts for people sitting in the front row! If you don’t have that then as the AVS table shows, you will probably hit the limit of your room before you hit the limit of the projector. I say this because if you look at Jason’s results, you’ll see we’re getting a cluster of results around the 4000:1 mark for the really highend gear.

I am battling with this very issue myself preach with your exact example. Will my room be able to do the jvc justice. My room is not dedicated and cannot easily be made a bat cave - a rear curtain is the best I will be doing, but the ceiling reflection is not something I can do much about (seeing as others live here too )

[The_Preacher1973]

I am battling with this very issue myself preach with your exact example. Will my room be able to do the jvc justice. My room is not dedicated and cannot easily be made a bat cave - a rear curtain is the best I will be doing, but the ceiling reflection is not something I can do much about (seeing as others live here too )

It's hard to know norpus. All I can say is that I got a significant increase in contrast in my previous setup when I put up dark curtains in all the walls and painted the ceiling a very dark shade of grey.

Darrin P (the guy that wrote the article that Gino linked too and the guy that measured a CR of 700,000:1 for a G70), has a completely blacked out room. Black velvet on the walls, ceiling and floor!

Personally I think you'd probably be better trying to get your CIH setup sorted. Remember that the JVC will need an external scaler whereas the 8720 doesn't.