Dismiss Notice
Attention AVForums app / Tapatalk users
Sadly GDPR means that, from 25th, we can no longer offer access to AVForums via the branded app or Tapatalk.
Click here for more information.

Hd2+ v DC3 Rainbow effect

Discussion in 'Projectors, Screens & Video Processors' started by foghorn, Apr 7, 2005.

  1. foghorn

    foghorn
    Active Member

    Joined:
    Mar 13, 2004
    Messages:
    629
    Products Owned:
    1
    Products Wanted:
    0
    Trophy Points:
    21
    Location:
    Central Scotland
    Ratings:
    +22
    Hi guys.

    I have read in a few posts either the reduction or in fact the elimination of the rainbow effect with the DC3 single chip projectors in comparison to the HD2+.

    Can it just be down to the chip itself?

    If this is the case how is it this possible?

    Foghorn
     
  2. Peter Parker

    Peter Parker
    Distinguished Member

    Joined:
    Apr 6, 2001
    Messages:
    13,177
    Products Owned:
    1
    Products Wanted:
    0
    Trophy Points:
    166
    Location:
    UK.
    Ratings:
    +2,714
    Rainbow effect can be reduced by speeding up the colour wheel and by adding more RGB segments, but like you, I don't understand how the chip alone can reduce or eliminate them. HD2+ can be DC2 or DC3 as far as I know, so only the brightness and black level has been improved to increase the contrast capability. In theory, that should make rainbows worse, not better, all else being equal.

    Where did you read about DC3 reducing rainbows?

    Gary.
     
  3. foghorn

    foghorn
    Active Member

    Joined:
    Mar 13, 2004
    Messages:
    629
    Products Owned:
    1
    Products Wanted:
    0
    Trophy Points:
    21
    Location:
    Central Scotland
    Ratings:
    +22
    Nothing official, it has been comments in general on either this forum or over on AVS. I seem to remember someone mentioning that they suffered from rainbow but had found that the Sim2 HT300E was the first DLP they had seen that did not produce them. On AVS they were discussing the Projectiondesign model one mk3 (DC3) and someone had upgraded their Mk 2 (HD2+) to a MK 3(DC3) and commented that the effect had been reduced.

    What started me off thinking about this is that I have recently spent some time with the ProjectionDesign Model one mk2 (HD2+) as a possible replacement to my Davis DL450. When I fired up the Model one I started to see the rainbow effect something I can’t recall noticing on my DL450. Considering the Davis is a 4 Segment 2 - Speed wheel (to the best of my memory) and the Projectiondesign a 7 Segment 5-speed wheel this struck me as a strange development.

    Projectiondesign in Norway thinks it is possible that different configurations of colour segments and wheel speed may produce different degrees of rainbows in different people.

    Foghorn
     
  4. Peter Parker

    Peter Parker
    Distinguished Member

    Joined:
    Apr 6, 2001
    Messages:
    13,177
    Products Owned:
    1
    Products Wanted:
    0
    Trophy Points:
    166
    Location:
    UK.
    Ratings:
    +2,714
    Hi Fog.

    I had a Davis DLS8 and I didn't see rainbows on that either, though I thought it was a single speed 4 sement CW. What frequencies did you run yours at?

    I now have the NEC HT1000 and Optoma H77, and I can't see rainbows on either of those either.

    Is the PD M1 a bright projector and was the screen high gain? I'm wondering of the increased contrast and/or brightness is allowing the rainbows to be more visible, but other than that I wouldn't like to say. Like you say, I would have thought the rainbows would have been more visible on the Davis, all else being equal.

    Gary.
     
  5. foghorn

    foghorn
    Active Member

    Joined:
    Mar 13, 2004
    Messages:
    629
    Products Owned:
    1
    Products Wanted:
    0
    Trophy Points:
    21
    Location:
    Central Scotland
    Ratings:
    +22
    Hi Gary,

    The Davis could well de a single speed. Frequencies?

    The PD M1 mk2 has a brightness range of 500 - 1000 lumens, which you can adjust to your liking, I choose to set at the low setting of 500 lumens.

    The first screen I used was a glass beaded which has a very high gain, however I had another screen which I suspect is a normal gain matt white. I think that the lower gain did slightly help rainbows but it did not eliminate the problem.

    Ken at Ivojo suggested that perhaps a lower gain grey high contrast (0.8) from harkness would be worth trying, however on a 0.5m by 0.5m sample it was a little difficult to tell, however it may have subdued the effect slightly again.

    Joe from the media factory is coming to the house on Sunday with a H77 and the Pd model one along with a different DVD player and screen. Here's hoping!

    Just carried out a search on the web for the Davis and came across a person with a similar problem did not see rainbows on the dl450 however found them on a infocus 4800.

    My theory on the DC3 chip may have fallen through there's a guy on the AVS forum who has just posted a comment that he found more rainbows on a In focus 7210(DC3) compared to 4805 or a 5700.

    Foghorn
     
  6. Peter Parker

    Peter Parker
    Distinguished Member

    Joined:
    Apr 6, 2001
    Messages:
    13,177
    Products Owned:
    1
    Products Wanted:
    0
    Trophy Points:
    166
    Location:
    UK.
    Ratings:
    +2,714
    Hi Fog,

    I asked about frequencies because I thought you might have used an HTPC to drive it rather than a standalone player. I used an HTPC and used 75hz for PAL, and 72hz for NTSC, any frequency change I made resulted in an audible change in the colour wheel speed which is wny I thought it was a single speed wheel. RTFM might know as he used to sell them.

    Quite often lower lumens and/or grey screens are said to help reduce rainbows, but as you saiod before, it's quite an individual thing so who knows how you'll get on!

    Is Joe bringing an H77 or an H78? All new machines should be H78s now I think, and if you don't see rainbows, I think you should be pleasantly surprised. With the H77 I would always add the caveat of occasional panning artefacts, but now I can't say that I'm aware of any major issues with the H78. I've not seen the Action 1, so I'll be interested to hear your opinion of the two machines.

    Now that I think about it, some people did say they saw more rainbows than usual on the HT1000, which is why I made a point of getting a demo rather than buying sight unseen. I guess there's no real formula or guarentee other than getting a demo.

    Gary.
     
  7. MikeK

    MikeK
    Well-known Member

    Joined:
    Oct 9, 2004
    Messages:
    2,231
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    83
    Ratings:
    +316
    My own personal little theory is that the rainbow effect and it's severity may well be related to the brightness (although I still think it's caused by the colour wheel).

    As Gary said, there are anecdotal reports of people suffering less RBE after fitting ND filters (polarizers too - although I suspect it's the brightness reduction from such a filter which is the reason), as well as switching to grey screens.


    Also, many people say they noticed them at first on their new DLP PJ, but then got used to them and either don't see them any more at all, or else see them a lot less. Could this in fact be related, at least in part, to the diminishing light output of the lamp during it's life?
    This seems to be borne out by the experience of a friend of mine who has an X1 - he noticed some mild RBE when he first got the projector, but "got used to it" to the point where he never really noticed it any more - until he had to replace the lamp, and the rainbows were back!

    Food for thought!!!


    PS - Foghorn - isn't the 7205 a DC2 chip model? (I thought it's the new 7210 which is DC3)
     
  8. foghorn

    foghorn
    Active Member

    Joined:
    Mar 13, 2004
    Messages:
    629
    Products Owned:
    1
    Products Wanted:
    0
    Trophy Points:
    21
    Location:
    Central Scotland
    Ratings:
    +22
    MikeK,

    Correction made I meant to say 7210.

    The reduction in brightness on a new bulb over time will lessen the effect, however I was running the PD on its lowest mode (500 lumens), which is probably lower than some other projectors when new.

    I was thinking back to when I got my DL450, and trying to narrow down the reasons for not noticing rainbows with that projector when new, the only other thing was that I was watching a lot off NTSC DVD's at that time. Don't know if that could have made any difference.

    Gary unfortunately didn't get either of the projectors today so as yet nothing to report, well see if anything transpires over the next few days.

    Foghorn
     
  9. foghorn

    foghorn
    Active Member

    Joined:
    Mar 13, 2004
    Messages:
    629
    Products Owned:
    1
    Products Wanted:
    0
    Trophy Points:
    21
    Location:
    Central Scotland
    Ratings:
    +22
    Hi Guys,

    Thanks for the replies.

    Comparing the PD to the H77, I would say that the PD was the better projector.

    PD pros - Natural colours, detailed image/3D, minor tweaking required to get a very acceptable picture.

    PD Cons - No lens shift

    H77 pros - Very quite, good blacks, smooth image (just not as detailed as the PD)

    H77 cons - Processing problems with component and other noticeable well-documented problems. There was also a green push in colour, but managed curb the problem to an acceptable level.

    I am prepared to accept that the above problems will be probably be tackled by the H78, so another demo required.

    On both projectors I saw the rainbow effect, and can’t recall seeing them on the Davis DL450 so I have started to try to understand why this could be.

    I wondering if it can be down to the fact that most of the newer DLP projectors are much brighter than the models of 4 to 5 years ago.

    The Davis was about 500/600 lumens and after a few hundred hours probably a bit less so hence the reason for a lack of rainbows to my eyes anyway.

    So I am wondering if I was to fit a ND filter to one of the above projectors or any other modern DLP, could I bring the level of brightness down to a level where the rainbows would all but disappear.

    The PD can be set to 500 lumens, not sure what the H77/H78 would be in eco mode

    Any thoughts on this, or any chance someone could try this for me.

    Additionally if anybody out there has a copy of The Missing DVD could they give it a go on their DLP, the opening scene is a single chip killer.

    I think if a filter could tame this film then I think I would be home and dry on the rainbow front.

    Here’s hoping !

    Foghorn
     
  10. Peter Parker

    Peter Parker
    Distinguished Member

    Joined:
    Apr 6, 2001
    Messages:
    13,177
    Products Owned:
    1
    Products Wanted:
    0
    Trophy Points:
    166
    Location:
    UK.
    Ratings:
    +2,714
    IIRC, via DVI colour temp 1 had a red push, and CT3 was blue, but ct2 should have been about right and not too far from D65. If you maxed out the RGBs green was just 5% higher than red, and blue 20% higher, so I don't know why you were getting a green push. What input was it on?

    I've seen two H78s now, and neither showed the panning issues that were present on the H77. Colour balance is even better now, with a higher contrast as a result and more light output. An H78 demo should show better results.

    The H78 measured 550 lumens in hi lamp, and about 435 (IIRC) in low lamp, so is brighter than the H77 was with similar hours on the lamp. Brightness does seem to help some people to see rainbows, so an ND filter is worth a try.

    I know the scene you mean (thanks to RTFM), and although I agree it's guaranteed to make rainbows visible, I'm not sure if it's a fair test as I think everybody will see them there if they look for them on any single chip DLP. More 'normal' DVDs like Gladiator are a fairer test for rainbows on a single chip DLP IMHO. If it's rainbow prone you will see them, if it isn't, then you probably won't. If people used that scene from 'The Missing' as a means to test for rainbows, DLP would no longer have a foothold in the front projector market. :)

    Gary.
     
  11. ZippyCat

    ZippyCat
    Active Member

    Joined:
    Aug 9, 2004
    Messages:
    573
    Products Owned:
    0
    Products Wanted:
    1
    Trophy Points:
    28
    Ratings:
    +22
    I’ve just had a little thought. On a Pal system, the colour wheel spins in multiples of the refresh rate 50Hz. On NTSC systems, the colour wheel spins in multiples of 60Hz. What happens when a DLP projector is connected to a PC running at 75Hz? Can the DMD device & Colour Wheel cope with multiples of 75Hz, or does it just revert to 50 / 60Hz?
     
  12. MikeK

    MikeK
    Well-known Member

    Joined:
    Oct 9, 2004
    Messages:
    2,231
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    83
    Ratings:
    +316
    AFAIK, it depends on the model.

    Some may be fixed speed (eg 60Hz) and do a frame rate conversion on everything else - some may be variable speed, where the actual wheel speed changes between PAL 50Hz input and NTSC 60Hz input (eg 4x speed with NTSC (120rps), and 5x speed with PAL (150rps), using a six segment RGB wheel).
    What such a variable speed model will do specifically with a 75Hz input though, I'm not sure - I suppose it could either frame rate convert it to 50Hz at input, and then triple it for feeding to the DMD, or use it as is and double it. Either way, I doubt you'd be able to tell for movies if it does it right, although interlaced video sourced 50Hz material could be interesting!
     
  13. ZippyCat

    ZippyCat
    Active Member

    Joined:
    Aug 9, 2004
    Messages:
    573
    Products Owned:
    0
    Products Wanted:
    1
    Trophy Points:
    28
    Ratings:
    +22
    What I was trying to get at was; if the DMD / Colour Wheel was capable of speeding up to accommodate 75Hz, then potentially there would be less rainbow effect than a standard 50 or 60Hz signal. Consider a 4x projector, 50 Hz would refresh at 200Hz, 60Hz would refresh at 240Hz & 75Hz would refresh at 300Hz. If this was the case, this could explain why many HTPC users claim the rainbow effect has been reduced or eliminated. However, I do think you are correct in your assumption that the projector would do a frame rate conversion, infact I would expect the unit to default to multiples of the local AC frequency (50Hz UK, 60Hz USA).
     
  14. foghorn

    foghorn
    Active Member

    Joined:
    Mar 13, 2004
    Messages:
    629
    Products Owned:
    1
    Products Wanted:
    0
    Trophy Points:
    21
    Location:
    Central Scotland
    Ratings:
    +22
    What input was it on?

    Can't recall remember the settings that I settled on. Reducing the green and uping the red and blue seemed to help. I am far from up on calibration so i am not sure I was going about it correctly, that's why the Projectiondesign was easier there was no need to delve into that part of the menu.



    I've seen two H78s now, and neither showed the panning issues that were present on the H77. Colour balance is even better now, with a higher contrast as a result and more light output. An H78 demo should show better results.

    That positive to hear.



    The H78 measured 550 lumens in hi lamp, and about 435 (IIRC) in low lamp, so is brighter than the H77 was with similar hours on the lamp. Brightness does seem to help some people to see rainbows, so an ND filter is worth a try.

    I am I correct in saying that if you added a ND2 filter and used in low lamp mode that the lumens would be 215 approx.



    I know the scene you mean (thanks to RTFM), and although I agree it's guaranteed to make rainbows visible, I'm not sure if it's a fair test as I think everybody will see them there if they look for them on any single chip DLP. More 'normal' DVDs like Gladiator are a fairer test for rainbows on a single chip DLP IMHO. If it's rainbow prone you will see them, if it isn't, then you probably won't. If people used that scene from 'The Missing' as a means to test for rainbows, DLP would no longer have a foothold in the front projector market. :)

    I know what you mean about The Missing, the reason I mentioned that one was if the rainbow effect could be elliminated with that movie using filters then I think we could be all home and dry with all other movies.

    Another DVD is Raiders of the lost ark, the begining scene when they are in the jungle.

    What would be an interesting experiment would be to play a scene that is easy to see rainbows and then start to work through ND2 ND4 and ND8,would you then see a gradual reduction and possible elimination of the effect.

    Foghorn
     
  15. foghorn

    foghorn
    Active Member

    Joined:
    Mar 13, 2004
    Messages:
    629
    Products Owned:
    1
    Products Wanted:
    0
    Trophy Points:
    21
    Location:
    Central Scotland
    Ratings:
    +22
    I know I should not have used the Quotes option. :confused:
     
  16. Peter Parker

    Peter Parker
    Distinguished Member

    Joined:
    Apr 6, 2001
    Messages:
    13,177
    Products Owned:
    1
    Products Wanted:
    0
    Trophy Points:
    166
    Location:
    UK.
    Ratings:
    +2,714
    No problem - it took me a few seconds but I realised what was going on and it was easy to read. :)

    Yes, in theory an ND2 shoudl halve the lumens. I have an ND 0.1 (a 3rd of an ND2), so I might try that and see if the rating is correct for that as well.

    It could be possible to use an ND to reduce the RE, but of course there will be a hit in brightness, so screns larger than 7ft wide would probably not be a good idea with the filter as the ft lamberts would be about 7, and only dim further as it ages. Of course, taking the filter off after a 1000hours (or when you feel you need the brightness back) will stop it from getting dim beyond watchable.

    I've found that with a higher contrast image, you can watch a dimmer image without realising it's dim - I'de used a combination of filters to get 2200:1 CR from an HT1000 and it was only putting out 4ft lamberts - the recommended level is 12 to 16 yet this seemed much brighter. That's probably why CRTs with low lumens and poor white level aren't considered dim.

    Gary.
     
  17. Peter Parker

    Peter Parker
    Distinguished Member

    Joined:
    Apr 6, 2001
    Messages:
    13,177
    Products Owned:
    1
    Products Wanted:
    0
    Trophy Points:
    166
    Location:
    UK.
    Ratings:
    +2,714
    Forgot to mention - I had a Davis DLS8 DLP a few years ago, and I ran that at 72hz for NTSC anmd 75hz for PAL (from HTPC). It did seem to have a possitive effect on smoothing the image out for panning etc. You could hear the machine speed up and down as you changed refresh rates, so that makes me think it was a single speed machine - it did have an RGBW colour wheel so it wasn't complicated by having to use multipe passes per frame.

    Gary.
     
  18. foghorn

    foghorn
    Active Member

    Joined:
    Mar 13, 2004
    Messages:
    629
    Products Owned:
    1
    Products Wanted:
    0
    Trophy Points:
    21
    Location:
    Central Scotland
    Ratings:
    +22
    My intention is to change my 6 foot 4:3 screen and replace with a 7 foot 16:9 when I eventually get the new projector.

    I would hope that even using a filter that the picture would still be adequate. I always watch in total blackout conditions so ambient light is not an issue.

    I guess I will have to invest in a ND2 in time for the next demo.

    I was also just thinking that how many times do you read a post where a person takes in a demo of lets say an Infocus 4805 and then a week later views lets say an Infocus 7205. As far as they are concerned the RBE is less on the 4805. To most people this does not make sense, the 7205 has more segments and a faster wheel.

    However could the answer be that the 4805 has been well run in with quite a few hours on the clock and the 7205 is new and with only a few hours.

    Just a thought.

    Foghorn
     
  19. Peter Parker

    Peter Parker
    Distinguished Member

    Joined:
    Apr 6, 2001
    Messages:
    13,177
    Products Owned:
    1
    Products Wanted:
    0
    Trophy Points:
    166
    Location:
    UK.
    Ratings:
    +2,714
    That's quite possible - brighter tends to make rainbows more visible for some so maybe that's why.

    If you do get an ND2, try to put it onto the pj before it's fired up, and see how it looks. If you see it without the filter at any time you will immediatley think it's better as we tend to prefer brighter over darker. Also try to demo it when projecting a 7ft wide image, even if this means it's projecting outside of the screen if it's smaller or larger, as the brightness is screen size dependant (pretty obvious I know, but just thought I'd mention it).

    I think I I might have an ND2 gel filter somewhere - I'll have a look and send it to you so you don't have to buy one until you're sure you need to. It'll go in an envelope so it'll only cost the price of a stamp to send. When will you be demo'ing again? If you want to borrow the filter, PM an address for me to send it to.

    Gary.
     
  20. MikeK

    MikeK
    Well-known Member

    Joined:
    Oct 9, 2004
    Messages:
    2,231
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    83
    Ratings:
    +316
    I see your point, but 4x speed is 4x speed, which is 240Hz!
    200Hz is 3.33x and 300Hz is 5x!

    But assuming a projector could change wheel speed between 3.33x for a 50Hz input, and 5x for a 75Hz input, then you may well be right that the 75Hz input might then reduce RBE!
    There are potentially other issues with 75Hz frame rates for PAL material though (specifically with interlaced 50Hz video material).
    Just as with deinterlacing, frame rate conversion can often be more complex than it first seems!

    The difficulty in all this though, is finding out exactly how each model works (other than by simply feeding different Vsync rates in, and listening for wheel speed changes) - the manufacturers typically don't make this information available!

    I did read one manufacturers explanation (sadly forget which one now :) ), which stated that NTSC 60Hz signals were dealt with using a 4x wheel speed (120rps, RGBRGB wheel), while PAL 50Hz was dealt with using a 5X wheel speed (150rps, RGBRGB wheel). So you could argue that on this projector, PAL material might be less susceptible to RBE.

    However, while I do believe (as most do) that wheel speed is closely related to RBE, I think there are other factors at work as well!
     
  21. ZippyCat

    ZippyCat
    Active Member

    Joined:
    Aug 9, 2004
    Messages:
    573
    Products Owned:
    0
    Products Wanted:
    1
    Trophy Points:
    28
    Ratings:
    +22
    “I see your point, but 4x speed is 4x speed, which is 240Hz!
    200Hz is 3.33x and 300Hz is 5x!”


    Not technically correct, the colour wheel will synch at a predetermined speed to the incoming source, hence a 4x projector will refresh at 4x the frequency of the incoming signal. Pal is 50Hz and NTSC is 60Hz, thus Pal will refresh at 200Hz and NTSC will refresh at 240Hz. The colour wheel in a DLP does change speed between standard Pal & NTSC, what is not clear however is whether the colour wheel will speed up further to accommodate a 75Hz signal resulting in reduced rainbow effect.
     
  22. MikeK

    MikeK
    Well-known Member

    Joined:
    Oct 9, 2004
    Messages:
    2,231
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    83
    Ratings:
    +316
    Coventionally, wheel speed is quoted in terms of reference to 60rps (3600rpm), which is 1x with an RGB wheel. (60Hz)
    At 120rps (7200rpm) it's 2x with an RGB wheel (120Hz), and 4x with RGBRGB (240Hz).
    At 150rps (9000rpm) it's 5x with an RGBRGB wheel (300Hz).
    At 180rps (10800rpm) it's 6x with an RGBRGB wheel (360Hz).

    But I suppose there'd be nothing to stop it being quoted as a multiple of the incoming signal Vscan rate either, as you say.
     

Share This Page

Loading...
  1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.
    Dismiss Notice