View Full Version : Calibrated RGBYMC vs DVE color filter
nicholc2 02-13-08, 02:19 AM I have just completed calibrating my DLP using HCFR to get my color adjustments as close to CIE diagram levels as possible. I have it pretty close according to HCFR. Well as close as my TV will get it.
Now that the colors are calibrated, I look through my DVE color filter and everything is off looking through that.
Is this normal? Are true calibrated colors going to look off when looking through the DVE color filters? Or should they actually match up?
I have just completed calibrating my DLP using HCFR to get my color adjustments as close to CIE diagram levels as possible. I have it pretty close according to HCFR. Well as close as my TV will get it.
Now that the colors are calibrated, I look through my DVE color filter and everything is off looking through that.
Is this normal? Are true calibrated colors going to look off when looking through the DVE color filters? Or should they actually match up?
In general, color filters aren't as accurate as instrumentation. It is not uncommon for color filters to be off by 5% or so.
What kind of instrumentation are you using?
nicholc2 02-13-08, 10:10 AM In general, color filters aren't as accurate as instrumentation. It is not uncommon for color filters to be off by 5% or so.
What kind of instrumentation are you using?
SpyderTV Platinum
I'd say it's off more than 5%. For the color green especially. Calibrated, it's more of a true dark green. If you go by the color filter, it's a lime green (best comparison I could think of).
SpyderTV Platinum
I'd say it's off more than 5%. For the color green especially. Calibrated, it's more of a true dark green. If you go by the color filter, it's a lime green (best comparison I could think of).
If you are seeing a 20%+ difference, then I would think something is wrong.
The spyder probes don't have a great reputation for accuracy, but I was thinking that the platinum version was "hand picked." You normally trust the meter over the filter, but there shouldn't be that much disagreement between the two unless something is wrong with either the filter or the meter.
rickardl 02-13-08, 12:02 PM You might want to have a look at this thread about filters:
http://www.avsforum.com/avs-vb/showthread.php?t=991485
nicholc2 02-13-08, 01:15 PM It would be easier if I knew someone that had a professionally ISF calibrated set I could look at. Then I'd have a comparison to see what the colors should actually look like. I'm thinking the meter is right, though. Watching HULK HDDVD and he actually looks green now instead of lime green. ;-) But I'm not 100% sure.
It would be easier if I knew someone that had a professionally ISF calibrated set I could look at. Then I'd have a comparison to see what the colors should actually look like. I'm thinking the meter is right, though. Watching HULK HDDVD and he actually looks green now instead of lime green. ;-) But I'm not 100% sure.
If your DLP has a projection lamp (not LED lighting), the Spyders are not terribly accurate. Spyders work best with direct view CRTs and plasmas and have accuracy problems with other technologies. I don't know how the Spyder probe responds to LED lighting.
I've never, EVER, seen the view through a green filter look like it was supposed to look. The view through a red filter is never "right" either, though it tends not to be a "off" as the view through the green filter.
nicholc2 02-15-08, 12:09 PM I do have the Spyder Platinum which is supposed to be a lot better than the regular Spyder. It does a good job with grays. That I can definitely tell. I think the problem is with my TV and the fact that the color decoder is just not accurate. To get the primaries even reasonably close, I have to lower the color saturation so much that it makes the pic look like crap. So I think I'll just have to deal with what I have for color repro. But the grayscale is looking pretty sweet!
I do have the Spyder Platinum which is supposed to be a lot better than the regular Spyder. It does a good job with grays. That I can definitely tell. I think the problem is with my TV and the fact that the color decoder is just not accurate. To get the primaries even reasonably close, I have to lower the color saturation so much that it makes the pic look like crap. So I think I'll just have to deal with what I have for color repro. But the grayscale is looking pretty sweet!
The Platinum is just selected from the regular production run and represents those with the smallest error. It doesn't work any better and isn't any more accurate when you are measuring anything besides plasma or CRT displays. It's a light spectrum issue... the Spyder is optimized for the light spectra emitted from red, green and blue phosphors. Projection lamps and backlights have FAR different emitted light spectra and the Spyder is "fooled" into incorrect readings. Nothing you can do about it.
nicholc2 02-16-08, 10:23 PM Actually, I'm thinking about returning it and picking up an i1Pro. I already have HCFR that I am using with the Spyder. As far as the colors go, though, I ended up just using the spyder to set the Y value to the correct 709 value and that helped a lot. I do think the grays are pretty accurate, though, with the spyder. Either way, I guess I'll find out eventually when I remeasure with the i1Pro.
What you said about it only working on plasma and CRT isn't exactly true, btw. It works on LCD and non-plasma displays as well with the baffle on. It works with CRT with the baffle off. The software takes this into account as well when taking the measurements.
What you said about it only working on plasma and CRT isn't exactly true, btw. It works on LCD and non-plasma displays as well with the baffle on. It works with CRT with the baffle off. The software takes this into account as well when taking the measurements.
Well... those measures help, but they don't undo the inherrent response of the device. All filter-based devices have the same limitation... they simply aren't accurate for more than 1 radiation spectrum. One company that introduced a new filter-based device that was pretty expensive (this was Fall 2007) claimed that finally, there was a filter-based product that would measure every type of video display accurately. As soon as it got into the hands of owners, the posts started flying about how this product was not accurate for several types of display. If they can't get it right for something that sells for far more than 10 times the cost of a Spyder Platinum, believe me, the Platinum Spyder is not going to be nearly as good. I have a Spyder Platinum and a 5-figure Chroma Meter and the errors the Platinum has range from very small on phosphor based displays to significant on other display technologies.
primetimeguy 02-17-08, 05:07 PM I do have the Spyder Platinum which is supposed to be a lot better than the regular Spyder. It does a good job with grays. That I can definitely tell. I think the problem is with my TV and the fact that the color decoder is just not accurate. To get the primaries even reasonably close, I have to lower the color saturation so much that it makes the pic look like crap. So I think I'll just have to deal with what I have for color repro. But the grayscale is looking pretty sweet!
Just to be sure, you're not adjusting your Color setting based on what you see on the CIE chart are you? That is not the intent. Adjust color using the filter, your CIE is what it is unless you have specific adjustments for it and few sets do.
Adjusting your primaries, secondaries, grayscale and most user controls doesn't mean you will get a reference picture. Unfortunately, a LOT of displays do not have adjustments for the color decoder and a lot don't have anything for CMS either. When you use the the color setting to correct for a decoder problem you desaturate all colors and not just the offending color. The trick to getting the most out of calibration is to know where any tradeoffs need to be made.
You might try readjusting you primary/seconday colors, contrast, and user controls to find the best compromise. As stated do NOT use a color control to adjust primary/secondary colors if you do not have some type of CMS system. You can use the color control with a filter to somewhat compensate for decoder issues but too much desaturation will cause muted colors. Only take it down enough to make the error be not offensive but you might not be able to eliminate it.
What TV do you have? Some calibrators on this forum might know some tweaks for your TV that might help.
Bob
nicholc2 02-18-08, 12:31 AM I have a Mitsubishi WD-62725 720p DLP. I do absolutely agree with everything posted here. I finally took it for what it's worth and got my color/tint adjusted to what they need to be. Then I took the Y value to what it should be in relation to white's Y value and now the colors are nicely saturated even though they aren't accurate on the CIE chart. I have perfect color with this tv so I can adjust each color's level individually so that's a big help.
I have my grayscale almost completely at .313/.329 across the board from 30 to 100 and that's the most important thing. The pic really pops and it looks great.
I'm excited to get the spyder returned and get my i1pro to remeasure the grays and make sure they're correct. They are definitely close. I'm also interested to see just how off my CMS is for real. I know that the i1Pro will give me an accurate reading on the CIE chart so we'll see.
|