AVS Forum banner

261 - 280 of 375 Posts

·
Registered
Joined
·
21,464 Posts
That's petty funny stuff. I bought a Teranex 3D processor because of people like you. You're entitled to your opinion and I to mine. Here's the good news -- it doesn't really matter that much to me anymore whether Hollywood supports 3D in the future. At this point it seems to me 3D has a fair chance to become a fairly strong niche market. Do I take the above comments personally -- yup! However, I don't wish you any ill will -- I just hope time proves you wrong.

What does this have to do with flavor preferences at an ice cream stand? It ignores so many important factors critical to user enjoyment such as the type of cone or whether a person should order a cup instead of a cone because a person's nose might hit the ice cream stacked in a cone while wearing 3D glasses. Sprinkles or not. Color of the sprinkles. Maybe this thread isn't about ice cream. Maybe its about eye candy.
 

·
Registered
Joined
·
21,464 Posts
Well the Problem is:
there are no official numbers from Epson regardig the other modes... and it would have taken a bit too long to check it out myself :D


I expect it to be 20,000h as this is the number of comparable technologies (Sony FHZ55 / LG Hecto).


Regards,
Ekki
You have my complete trust. I am patient. I will wait for your report on the numbers at 10,000 hours and 20,000 hours.
 

·
Registered
Joined
·
13,035 Posts
Ekki,

Thanks for the great review. Some clarifications / questions please:

1. What was your overall impression of the picture quality in watching some movie clips? Price aside, would you prefer the LS10000 over the JVC or over the Sony 500?

2. It will be great to hear when you determine whether the light path is truly sealed or whether dust blobs are still possible.

3. Thank you for confirming it has both a manual iris and adaptive light control over the laser - not to mention overall lightness setting low/med/high (I almost wrote "bulb" setting- whoops! :) ).

4. I found these two statements to contradict each other. "For this reason, blue must not be greatly corrected by mixing the other colors in color management." But then goes on to say "All deviations are not described as serious, but can of perfections by color management (extended image menu) are corrected". Does this mean that with a Radiance and 3D LUT you would expect to get excellent results across all colors and grayscale/gamma?

5. This was also seems a bit contradictory: "Was implemented this extended color space by a "Cinema Filter", which is automatically pushed upon activation of the cimena presets in the light path and unfortunately also absorded about 70% of the light output..." But then goes on to say "offsets such a wide color gamut at the cimena preset. But this is not generated here by an additional filter, but largely corresponds to the native color space of the laser. Therefore no loss of light is connectd eiwh the Cinema mode..."

So in Cinema mode there is no lumens lost? How about in Dynamic Cinema? If we are looking to use a Radiance with 3D LUT for 1080p material what would be the best choice and are there any lumens lost in that mode?

6. Did you measure lumens at short, long and mid zoom in a few modes? It would be good to see if your results compare to what other reviews have found.

7. "Differences in skin color cannot targetted in color management of the projector" - but with a Radiance it can, yes?

8. Excellent that the manual iris allows you to achieve higher contrast with lower lumens when used in the near throw

9. It says there is an uncalibrated green tint using the dynamic dimming? Or am I not reading that right?

10. I think what Manni is looking for is to see what the saturation sweep charts and ColorChecker SG results from CalMAN look like so we can see how well the colors track within the gamut. For example JVC tends to not track as well as the Sonys. It would be great to see those results if you can still measure them?

11. Any opinions on the overall quality of the lens? Was the focus excellent across the whole screen?

12. How is the color uniformity - any banding or other such issues? Any colorization issues as the static iris is changed?

Thanks again for your excellent review!!
 

·
Premium Member
Joined
·
9,828 Posts
I think this could be difficult. But I do not consider BT1886 useful anyway for this projector. At least not when you are using dynamic contrast control.
You might not consider it useful for this projector, but that means this projector can't be calibrated to the Bluray 4K standard, which is likely to be using BT1886 (a great progress given that there is no gamma standard at all for bluray). Don't you think this would be useful?

Iris was open and Zoom in mid position
Thanks :)

??
We provide a far more sophisticated DeltaE Color Analysis for the whole Color Palette / Skintones / example pictures in every Review (6 alone in the LS10000 Review)?!? For this Analysis we measure hundreds of Colors / Saturations / brightness Levels...

This is much more precise than the rather primitive Calman Saturation gamut checks, which are still not very detailed at all.

We clearly do show the lead here as long as the Reader knows how to intepret the corresponding Graphs. We explained them in a detailed know how Special here:


http://www.cine4home.de/knowhow/Neue_Farbmessverfahren/Farbraum_Messungen.htm

Come on, Ekki, you know better than that. If your fancy sophisticated analysis is so great, then why do you show a gamut graph at all with only six points?

Let's take this review. After showing a limited 6 point gamut graph, you say the only color that falls out is blue. Why? You are only showing points at 100% sat / 100% stim, where there is very little actual content (except maybe in animation movies). There might be lots of colors out at lower saturation/luminance points, where it matters more for actual content. Your fancy graphs show a lot of green but also a lot of yellow (which means a dE above what and below what? Which dE formula? I have no idea). So provided you tried to correct these issues with the internal controls, where is your fancy analysis showing that there is much less yellow and much more green after correction? I can't see it in your review.


In the link above, for the explanations of your "thermo diagram" in section 4.2, you say that green = good, yellow= acceptable, red=bad. Which dE formula? Which dE threshold for each color-coded zone? I'd like to know. What's good or acceptable to you might not be acceptable to me, depending on the dE formula you use and the thresholds you apply for your color-coded zones. Please could you provide this information?

My point is that either you don't show a gamut graph at all, or you show us a less potentially misleading one. Showing a 6 point gamut graph like the one in your reviews can look perfect but hide a lot of defects. This is what you have done in ALL your reviews of ALL JVCs projectors, showing how the internal CMS allowed to reach a "perfect" gamut at 100% sat using the internal controls, while every competent calibrator and reviewer - which includes you :) - knows that if you target rec-709 at 100% sat to obtain such graphs, your picture is WILDLY undersaturated at 75% sat and below, which causes a visibly undersaturated and inaccurate picture. I'm sorry, but this has to stop, and if you don't start, who will? :)


I know you are competent, and I know you know this. So please show us the saturation and luminance graphs, with associated dEs (telling us which formula you used), especially AFTER you use any internal controls to correct the picture. They might be "primitive", but they don't rely on visual fancy analysis which honestly don't tell me much from a practical point of view (unless everything is green before calibration and I know what green means). By the way, this has nothing to do with Calman. Chromapure, HCFR, Argyll also offers this, and any other calibration software should as well, because that's what you have to use when you use a limited 6 or 7 points CMS to calibrate the display. Even if you gave us more information about the dE formula and thresholds you use to generate them, I have no idea how I could use your fancy analysis charts to reach the best balance and the most accurate results using the internal CMS of say a JVC. And if I only used your 6 point gamut graph to target rec-709 at 100% sat, I'd certainly get a poor calibration.


This was acceptable when there was no consumer-level solution to correct these errors. But now that we can buy an eecolor for a few hundred dollars, and that even the cheapest Radiance model offers a 3D LUT which resolves 90% of these issues, this is not acceptable. As I told Kevin, we are in 2014, not in 2008.


What I want to know when I read a review is if I can get the display close enough using the internal controls so that it doesn't need an external 3D LUT, at least for some of my sources (I use MadVR and its up to 65x65x65 3D LUT for all my movies as my HTPC is my main source). Unless your post calibration fancy analysis shows an all green (which they never do) and green means a dE 2000 under 1.5, I have no idea how far I am, where the problems are and if they can be corrected with the limited internal controls.


So please step up - or clear up - your game :). If you think your fancy visual analysis are good enough, please provide the dE formula used and the threshold for each color-coded zone, and stop showing 6 point gamut charts which are meaningless at best, misleading at worst. And if you feel the need to show something that we can all relate to (and reproduce), then please show us a gamut chart with saturation and luminance tracking along with the data (dE numbers), especially AFTER you've used the internal controls. It might be primitive, but at least it's not misleading and it's meaningful, with hard dE numbers we can all look at and understand.


The Sony 4K projectors track so well they barely need a CMS or even a 3D LUT, at least OOTB and if you have a fairly neutral screen. The JVCs track ok before correction with the standard color profile, but their CMS breaks any linearity and produces a worse picture after calibration if used incorrectly, ie if made to look accurate at 100% sat (which is the gamut graph most calibrators and reviewers show to their readers/clients). Don't you think it would be good to know where the Epson sits between these two? This has to be taken into account in the overall cost of the projector when comparing models. A projector a bit more expensive which doesn't need a Radiance to produce an accurate picture might be cheaper than a less expensive one that does. I want to know!


I'm calling all reviewers and calibrators to stop the bullsh*t and to give us the info we deserve regarding native gamut, corrected gamut and the effectiveness of internal CMS controls :)


As I said, your reviews are amongst the best, if not the best (especially regarding the on/off, ANSI contrast and brightness at different lens position / throw position which you are the only one to provide consistently and reliably, and for this we'll be eternally grateful), but it doesn't mean there isn't any room for improvement :).
 

·
Guest
Joined
·
867 Posts
So the Cine4Home review of the Epson LS10000 has arrived and according to many it would be the definitive review. How much more positive can a review be? It doesn't go quite as far as the PC review in the use of superlatives but it isn't far off.
Projector Central was far far more positive.

Projector Central review
Contrast
contrast >1,000,0000:1
With in the body of the review (Our database does not accept alpha characters for contrast specs, so it will be designated as ">1,000,000:1" on this site)

Cine4home review
Contrast
cinema mode 17,000:1
with iris closed down up to 65% more contrast, so I guess up to 28,500:1
dynamic contrast x5, so I guess up to 142,500:1
Within a image 10,00:1
ANSI checker board contrast 270:1
 

·
Premium Member
Joined
·
9,828 Posts
The contrast was infinite in the PC review because they measured it from full black to full white (the standard way to measure on/off). As the laser switches itself off on full black, on/off contrast (in a light controlled room) is infinite, so PC review isn't wrong and Epson isn't lying.

Ekki knows better than that though, so he has displayed a bit of white on the black pattern (to keep the laser on), to make sure he could measure the effective native and dynamic on/off contrast ratio, instead of the on/off contrast that is only at work when the screen is fully black or during a fade to black.


Ekki's numbers are lower (and much more realistic) but you have to agree that they don't look as good on a marketing brochure or in an article as "infinite black" :).
 

·
Registered
Joined
·
1,583 Posts
@Manni
dE2000 (3 / 5 / 10)

@fatjulio
Yes, that is actually what we plan to do: Measure the LS9600 to confirm that the eShift is the cause for lower ANSI.
We already did that with JVCs however and it proved to be like that. So we expect it to be the same with Epson.


Anyway, without DCI, less brighntess, no 4K Input (and probably no HDCP2.2), the LS9600 is not really an alternative to the LS10000.




Regards,
Ekki
 

·
Premium Member
Joined
·
9,828 Posts
@Manni
dE2000 (3 / 5 / 10)

Thanks Ekki :).

So this means that quite a bit of the errors in the LS1000 are visible (if yellow is between 3 and 5 dE 2000 and green is below 3).


As you know, dE 2000 and dE94 have much lower dEs than the same errors measured in dE76.


For me, a good dE2000 is under 1.5, under 3 is acceptable and over 5 is bad.


Is there an internal CMS in the LS10000, and when you use it can you make things better?


I haven't seen a post-calibration (to rec-709) thermal-diagram of the LS10000 in the review, with hopefully less yellow and more green. Did I miss it?


Also another thing I wanted to ask you, did you try measuring the laser light with a semi-pro or consumer spectro/meter, like an ipro/i1pro2 spectro and an i1d3 meter?


I use a Discus trained to an i1pro2 and I'm wondering how they would work with the laser light. It would be good to know if we need to upgrade our meter(s) if we upgrade the projector!


Cheers.
 

·
Registered
Joined
·
1,583 Posts
Thanks Ekki :).

So this means that quite a bit of the errors in the LS1000 are visible (if yellow is between 3 and 5 dE 2000 and green is below 3)..

Yes, it was not perfect, but for an ootb result quite good.


Is there an internal CMS in the LS10000, and when you use it can you make things better?
I haven't seen a post-calibration (to rec-709) thermal-diagram of the LS10000 in the review, with hopefully less yellow and more green. Did I miss it?

The LS10000 has an RGBCMY CMS like all other Epsons, which works very, very well. And you are right, we have not published a post calibration graph yet as we want to check more final series machines first. With the Review sample LS10000, you would get DeltaE
 

·
Premium Member
Joined
·
9,828 Posts
Thanks a lot for all the info above Ekki, much appreciated :)


Sounds like good news, looking forward to your measurements with a production unit.


If the i1pro2 is fine, the Discus trained to it should be fine too.


Did you have a chance to try a K10A with the Epson?
 

·
Registered
Joined
·
5,545 Posts
Projector Central was far far more positive.

Projector Central review
Contrast
contrast >1,000,0000:1
With in the body of the review (Our database does not accept alpha characters for contrast specs, so it will be designated as ">1,000,000:1" on this site)

Cine4home review
Contrast
cinema mode 17,000:1
with iris closed down up to 65% more contrast, so I guess up to 28,500:1
dynamic contrast x5, so I guess up to 142,500:1
Within a image 10,00:1
ANSI checker board contrast 270:1
Here are the actual numbers from the review:

"A calibration of the dynamic mode is alone with the RGB sliders not possible, which is why we resort to the already more realistic cinema / and Natural modes incl. 6500K / D65 white balance. Color Corrects the LS10000 loses about a third of its luminosity, reaching around 1100 lumens maximum with a native contrast ratio of 17,000: 1 and 27: 000: first But this also are still excellent values ​​that are similar to the reference level such as a JVC X500. This is even more striking when one considers that this is the first generation of "Reflective LCDs" from Epson."

and

"With appropriate measurement procedures, we were able to identify the realistic and practical dynamic contrast values ​​of the Epson LS10000: Its dynamic light regulation works around with the factor "x5", so that the dynamic range (/ static iris depending on the zoom) to 120.000: increased 1: 1-220000 will. In view of the fact that this is actually useful contrast values, these results are no less than outstanding."

The PC review may have been more positive; however, this is still a very positive review:

"Black level and contrast are particularly impressive with the use of dynamic light control. All in all, the Epson LS10000 a device with virtually no weaknesses. But this is not to say that it does not exist: in addition to a native 4K resolution would be even more calibrated luminosity (ie less loss through the calibration) is desirable. The interframe calculation is not quite on the cutting edge of technology and also can not be combined with 4K playback. Dimensions and weight are in combination with the matt black finish little living room compatible and the Lens Memory function operates at the present time nor accurate. If Epson improve these complaints at a high level in the coming years, as the R-LCD / laser hybrid technology can undoubtedly put the reference point of the high-end home cinema and very close to get at the 'perfect projection'."

Cine4Home would like to see improvements with the following: Lens Memory; real 4K instead of enhanced 1080p (you'll have to pay for that); white case as well as black; better inter frame calculations; and brighter after calibration.

The Sony VW600 has native contrast of 6000:1 with a 4K panel and retails at $15,000. You could probably find one new for around $11K. Personally I'd want more contrast. Too bad we can't have the brightness, ANSI contrast and 3D of a Super Lumis, the laser with modulation of the Epson, the on/off contrast of a JVC and Sony's 4K resolution all for under $10K. Maybe in a few years -- we can hope.
 

·
Registered
Joined
·
5,190 Posts
Boy exciting stuff, that contrast is nothing to scoff at. Be really interesting to see where we are in a few more years with this tech.
 

·
Registered
Joined
·
13,035 Posts
Here are the actual numbers from the review:... All in all, the Epson LS10000 a device with virtually no weaknesses....
I'm excited to see the post-calibration results from Ekki and in particular the saturation tracking at 25%, 50% and 75%. Hopefully he publishes the ColorChecker SG results as well.

Regarding weaknesses, I consider the ANSI CR at what it measured to be rather weak and it is a concern for me.

My VW95 is still going strong and throwing a great picture. At this point I wouldn't mind holding on to it another year. But I've said that the past couple years so I'm a bit itchy to get something new.

I likely would have gone with the JVC 4910 last year, but the HDMI handshake issues and freezes and having to pull the plug etc were a major turn off. That combined with its high lag (lots of gaming here) and 3D and motion handling supposedly not quite as good as the VW95 resulted in me holding off.

At this point I do not care that much about 4K. Just looking to get something with better black levels than my 95 without high gaming lag or buggy operation. The Sony 600 would be a great upgrade, but its hard to justify that type of money when laser light sources are starting to emerge and having no support for HDMI 2.0 A at that price... hummm.
 

·
Registered
Joined
·
13,277 Posts
we still have to find out if there's artifacts in gaming with the fast mode that I believe was measured ~ 55ms.

September 2015 Cedia prediction- native 4K panels, full HDMI 2.x support, more light output in DCI mode, full gaming mode, lower price since JVC and Sony will have the guns out next year. :)
 

·
Registered
Joined
·
1,583 Posts
Ekki, did you get a chance to take a look at "native' sharpness on the pixel level with the e-shift device disabled? I've noticed that the past 4 generations of JVC projectors have had excellent pixel sharpness/delineation and can focus right down on each individual pixel. Other 1080p LCoS machines from Sony and others have fell way short of what the current JVC's can do. Where would you place this Epson model in that regard?

Also, regarding motion sharpness. Do you feel that there is a noticeable difference between JVC's current generation D-ILA panels regarding how motion looks. Is the Epson the closest thing to DLP you've seen so far?

Thanks for all your hard work. You have the best reviews out there.



Without eShift, you could see the individual Pixels sharply, however there was some vertical "streaking", which many JVC had also in the past. So optical sharpness was good, but not overwhelming.


Regarding Motion sharpness, the Sony VW500 does definately a better Job. I could not make a direct comparison to the JVC. With FI on the Epson has a good Motion sharpness, but not reference Level.


Regards,
Ekki
 

·
Banned
Joined
·
10,827 Posts
Without eShift, you could see the individual Pixels sharply, however there was some vertical "streaking", which many JVC had also in the past. So optical sharpness was good, but not overwhelming.


Regarding Motion sharpness, the Sony VW500 does definately a better Job. I could not make a direct comparison to the JVC. With FI on the Epson has a good Motion sharpness, but not reference Level.


Regards,
Ekki
Thank you for your thoughts. I would have thought a 480hz device would at least have comparable motion to Sony's SXRD panels which refresh at the same rate. So it's behind Sony on both native and interpolated motion sharpness when you compared them. Interesting.

I have noticed better motion on the current JVC models compared to older JVC models (X3 and X30 for example). I had the chance to take a look at VW90ES and an X500 side-by-side and they looked pretty similar with motion without either having their FI modes turned on. The Sony had a slight edge over the JVC with certain types of motion moving on screen, but the difference was negligible. Both were still noticeably worse than DLP. Hopefully within a few years we'll have thin enough LCoS panels where we can get motion to look almost indistinguishable to DLP. Thinner panels do mean less native contrast but hopefully we have technological breakthroughs to overcome this issue.
 

·
Registered
Joined
·
1,583 Posts
1. What was your overall impression of the picture quality in watching some movie clips? Price aside, would you prefer the LS10000 over the JVC or over the Sony 500?

The LS10000 is very comparable to a JVC-X700 + the benefits of laser source, HDCP2.2 and better Super Resolution.



2. It will be great to hear when you determine whether the light path is truly sealed or whether dust blobs are still possible.

Yes, we will definately check that asap. I remember that the R-series was not totally sealed 4 years ago.




4. I found these two statements to contradict each other. "For this reason, blue must not be greatly corrected by mixing the other colors in color management." But then goes on to say "All deviations are not described as serious, but can of perfections by color management (extended image menu) are corrected". Does this mean that with a Radiance and 3D LUT you would expect to get excellent results across all colors and grayscale/gamma?

We meant that native, optical blue coordinate is matching Rec709, without digital Color correction. With a radiance you sure will be able to get very good results. With our testsample, it was even possible just with the internal CMS.



5. This was also seems a bit contradictory: "Was implemented this extended color space by a "Cinema Filter", which is automatically pushed upon activation of the cimena presets in the light path and unfortunately also absorded about 70% of the light output..." But then goes on to say "offsets such a wide color gamut at the cimena preset. But this is not generated here by an additional filter, but largely corresponds to the native color space of the laser. Therefore no loss of light is connectd eiwh the Cinema mode..."

Nono, the first part was a view back at old Epson homecinema projectors using UHP lamps (TW9200 for example).
We then point out that the new laser projector reaches a wider space without a Cinema filter. That got probably lost in Translation.
The green filter is only used in DCI / AdobeRGB mode.



So in Cinema mode there is no lumens lost? How about in Dynamic Cinema? If we are looking to use a Radiance with 3D LUT for 1080p material what would be the best choice and are there any lumens lost in that mode?

Cinema mode or natural mode. Calibrated brightness is the same with These two.



6. Did you measure lumens at short, long and mid zoom in a few modes? It would be good to see if your results compare to what other reviews have found.

We did not have enough time to measure a full contrast table. You can calculate the light loss with the contrast gain numbers though.






7. "Differences in skin color cannot targetted in color management of the projector" - but with a Radiance it can, yes?

With a radiance, you can do that on any projector, no? ;)



8. Excellent that the manual iris allows you to achieve higher contrast with lower lumens when used in the near throw

Iris and Long throw had the same effect on contrast. Light loss is comparable also.



9. It says there is an uncalibrated green tint using the dynamic dimming? Or am I not reading that right?

No, not dynamic dimming, Dynamic "mode". This is one of the Picture presets:





The dynamic mode shows a 150% green tint. That is why you lose 450lumens when calibrated to D65. Actually, if you used an external Magenta filter, you could get up to 40,000:1 native contrast (hardcore tweaking) ;)


10. I think what Manni is looking for is to see what the saturation sweep charts and ColorChecker SG results from CalMAN look like so we can see how well the colors track within the gamut. For example JVC tends to not track as well as the Sonys. It would be great to see those results if you can still measure them?

We usually do not use Calman for this sort of measurements. Saturation tracking is very good. The DeltaE >3 Areas in our Analysis come from a slight Color shift in blue (bit too greenish) and red (bit too yellowish ).



11. Any opinions on the overall quality of the lens? Was the focus excellent across the whole screen?

See my post above. (#277)



12. How is the color uniformity - any banding or other such issues? Any colorization issues as the static iris is changed?

Color Uniformity was excellent, better than most JVC or Sony. I could see no banding, no noise. The R-LCD Driver is not fully digital but still analog, which I personally consider as a plus.


Regards,
Ekki
 

·
Registered
Joined
·
1,583 Posts
Thank you for your thoughts. I would have thought a 480hz device would at least have comparable motion to Sony's SXRD panels which refresh at the same rate. So it's behind Sony on both native and interpolated motion sharpness when you compared them. Interesting.


"480Hz" with half the vertical Resolution that is... Never trust Marketing numbers for frequencies (just look at the TV market).


;-)


Fast movements got a bit blurry because of Green beeing a bit slower than Red & Blue. But not by a huge margin. That is where the Sony is better.


Regards,
Ekki
Cine4Home
 
261 - 280 of 375 Posts
Top