CurtPalme.com Home Theater Forum

garyfritz · Joined: 08 Apr 2006 Posts: 12088 Location: Fort Collins, CO

I've had a Minolta TV2150 colorimeter for a few years. I use it to tweak up my grayscale &etc.

I just got a Spyder2 to play with. In the process of trying it out, I realized something I'd never thought of before: aiming the sensor is critical! Point it a bit to either side and the color temp will be high (blue) or low (red).

With my Minolta I always aimed the sensor for max luminance (ftL). That probably got me close to the right direction. Unfortunately ColorHCFR (freebie colorimeter software I'm using with the Spyder) doesn't seem to support a continuous luminance reading, so I'm not sure how I should aim it.

Somebody who actually knows what they're doing with colorimeters -- what's the proper way to aim your sensor? Do you use a sight to line it up on the green tube, or...? Maybe stick a mirror on the front feet of the Spyder, tape a laser pointer on the green tube, and adjust the Spyder until the mirror bounces the laser back to the green?? (Then remove the mirror before taking any readings!! Smile

)

Gary

Bighitter · Joined: 15 Dec 2006 Posts: 159

If you really want accurate readings of what you are seeing on screen... you should point it at the screen. Try to get the same viewing angle for the meter as you would have from the seating position.

garyfritz · Joined: 08 Apr 2006 Posts: 12088 Location: Fort Collins, CO

Good point. My Minolta colorimeter couldn't do that -- it could only read down to about 0.1 ftL so it couldn't even read IRE 10-20 pointing at the PJ. But with this HCFR application, it will use longer samples and average many samples to read lower light levels. It's a lot slower but it should be a lot more accurate. So maybe it could read off the screen. That should be a lot more accurate in several ways than pointing it at the projector. I'll have to try that. Thanks!

Bighitter · Joined: 15 Dec 2006 Posts: 159

No problem, sencore has even come out with a new video pod to do exclusively front projection setups.

**kal** · Posted: Fri Apr 27, 2007 5:19 pm Post subject:

I've wondered this myself... when I did my Zenith 1200 with a borrowed Minolta CA-100 I wrote that I did the following:

(1) If you've done your homework, you'll realize that the CA-100 was probably only intended to measure direct-view CRTs (not CRT projectors). Probably right, but I was still able to get an image that looks better then using nothing at all. So caveat emptor! Don't use something like a CA-100 if you expect perfection!

(2) Probe orientation plays a big factor in the x/y readings (most likely due to point (1) above). Move the probe slightly and the values change. To make my readings as consistent as possible, I placed the probe on a tripod in the center of the screen and rotated it up/down/left/right until the luminence value (Y) was as high as possible so that I knew it was pointing directly at the green tube. Better then no consistency at all I suppose (?).

Full thread: https://www.curtpalme.com/forum_archived/viewtopic.php@t=210.html

In retrospect, I wonder if I should have only had the green tube on when doing this to make sure that I was pointing as directly to green as possible.

You'd probably also want to make sure that the tripod/sensor is directly in the center of the screen too.

Kal

garyfritz · Joined: 08 Apr 2006 Posts: 12088 Location: Fort Collins, CO

I think maxing the Y, ideally with just the green on, is probably a good approximation. If I was going to get fussy about it, I *would* do the "laser pointer on the green, mirror on the sensor" trick -- that way you'd know you were aimed dead on. At least then you should get consistent results. But correct results? I'm not sure. The sensor is still going to see 3 separate color sources, and who knows how it will respond to that.

If I can actually read off the screen, that's even better. That way you're reading the color of the image itself, not an attempted blend while looking at the 3 color sources. Plus you're setting the color temp of what you actually SEE, instead of the temp of the light that comes out of the projector. If your screen affects your color temp in any way, reading off the screen could make a difference. You just need a sensor that can read really low light levels. It appears that the Spyder CAN do that, if you give it enough time. HCFR spends over 2 minutes averaging multiple long-exposure reads at the lower IREs!

When I get a chance, I'll try measuring off the screen and report back. Maybe tonight but no promises.

**kal** · Posted: Fri Apr 27, 2007 6:24 pm Post subject:

Yeah, my feeling was that only the really super-sensitive sensors can read off the screen below 30-40 IRE. Let us know what you find out.

Bouncing is certainly the most accurate, though I really wonder sometimes how critical this needs to be as as long as it's reasonable close it looks pretty damned good.

Kal

garyfritz · Joined: 08 Apr 2006 Posts: 12088 Location: Fort Collins, CO

Yes, almost anything looks better than most people's by-eye setups. But I discovered last night just how much the results can vary with just a slight misalignment of the sensor. I thought I had it aimed dead at the projector, but the color temps were 1000K off from what I'd measured with the Minolta -- 7500 instead of 6500! (Worse than that, actually, since the readings were well off the black-body line on the CIE diagram that represents color temps.) That's a pretty severe difference for such a small alignment error. Measuring off the screen should eliminate that variable.

I'll have to try measuring 100IRE off the screen with the Spyder and the Minolta and see how closely they match...

studio · Joined: 24 Mar 2006 Posts: 191

Has anyone in the US built one of these? Is it a good one?

AFryia · Posted: Sat Apr 28, 2007 1:02 pm Post subject:

I'm just starting to play around with a HCFR DIY unit.

To pull readings off the screen It needs to be <12" away. Found the same with the Spyder <8".

IMO I get faster and lower IRE reading with the HCFR than with my Spyder.

studio · Joined: 24 Mar 2006 Posts: 191

AFryia · Posted: Sat Apr 28, 2007 1:17 pm Post subject:

Dave Larsen built mine. I'm not sure if he still does. Check his site here.

http://members.aceweb.com/dlarsen/

I didn't have the time or equipment to tackle programming the PID chip.

I like the idea of the calibration file to tune the meter for what ever display you are working with.

garyfritz · Joined: 08 Apr 2006 Posts: 12088 Location: Fort Collins, CO

Well, I tried the read-off-the-screen last night (but couldn't report on it right away, because the forum was down... breathe, breathe, don't hyperventilate... Smile

).

The results I got were crazy. Reading the primaries (just R, just G, just B), the Spyder agreed pretty closely with the Minolta. So that seems to indicate they are reading the same coordinates. But when reading a white IRE 100 window, they differed dramatically. Which I could understand if I was pointed at the projector, but they were vastly different even when reading off the screen!? I'm not sure if I have to calibrate HCFR for the Spyder, or what. Still investigating.

At least I'm getting very similar results from reading off the screen or pointed at the projector -- assuming you aim at the point of max luminance. So reading off the screen should be safer, as long as you avoid seeing the sensor's shadow.