Sorry in advance for this long post--

Gary did his magic on my G70 last weekend by utilizing a cool device to measure individual R, G, and B output and graph/adjust the same into a much more balance output (can you tell I have no idea what that tri-pod instrument with digital readout monitor and laptop was called?). The long and short of it is that my picture, which I thought was pretty awesome in the first place, tremendously improved in detail, clarity, and depth-perception. The blacks and contrast between colors moved to a new level! All week long I've gone back to movies and HD stuff I've previously watched, only to see a whole new level of detail in the picture I never saw before-- and an even more lifelike image. An example is in the dwarf caverns of Moria in Lord of the Rings-- looks like a new movie in the very dark scenes, with creatures and detail coming out of the woodwork that wasn't clearly visible before.

Because I have a near new G70 (and dumb luck), I have a TON of output so I have been able to drop the contrast to 56 and was keeping the brightness to 35. I am also running a High Power screen which has helped the picture really pop as well. We measured over 15 FL on a 106" screen before we started adjusting. All these levels originally were by guess and set only by the naked eye.

But when we dialed it in with Gary's equipment to get exact measurements, we found a significant hump in the blue all the way up that we eventually smoothed out by defocusing it (see post below to describe). Gary adjusted the individual gains to bring all three as close as possible. By "changing the blue "MAG -- ALL" setting (which controls the full-screen blue focus) from 123 to 153, we defocused the blue significantly, and the blue then fell more in line and in close graph with the R and G at all levels up the IRE curve.
.
Still keeping the contrast at a low 56, we finally checked the black levels and found the brightness needed to be raised substantially, from 35 to 64. But when all those combinations were performed, the picture seems to have turned out AWESOME.

Here's our final gain and bias figures:

.....Before ............After Gary's Tweak
GAIN:
Before: R 223.........After: R 170 (lowered)
Before: G 224.........After: G 224 no change
Before: B 225.........After: B 164 (lowered)

BIAS
Before: R 52.........After: R 52 (no change)
Before: G 60.........After: G 60 (no change)
Before: B 24.........After: B 52 (raised)

Contrast- remained unchanged at 56
Brightness- increased from 35 to 64
Output measured at screen after all adjustments: 14.8 FL

So here's the questions?
Have we hurt the PJ in anyway with these settings? Especially wondering if raising the brightness to 64 is an issue, even though keeping the contrast at 56.

2. Gary's question as to whether there is a better way to defocus the blue than cranking the "Mag-ALL" numbers? Is this amount of blue defocus unusual or out of line on a G70

Hopefully Gary will answer the technical issues on replies. Thanks Gary for an awesome picture-- it certainly was worth more the the brewskey's we drank!!

Blue hump indicates you're overdriving the blue.

Blue CRT phosphor apparently has less light-emitting capacity than R & G. If you run the contrast high enough, you'll drive the blue into a "non-linear" range -- where increasing the input by x% no longer increases the output by x%, but by maybe 60-80% of x%. (That's a simplification, since CRTs don't respond linearly anyway, but hopefully you get the idea.) If you set your color temp at a high IRE, you'll set the blue at the level it takes to get the right temp at that IRE -- but since you're overdriving it to get there, that means the blue is too high at the lower IREs.

(BTW I usually set the temp at 80 IRE instead of 100. 100IRE is not all that common in actual video material. You'll see a lot more 80IRE and I figure it's more important for the colors to be spot-on there, rather than in full-on conditions that are rarely seen outside a test pattern. I think you'll also get more accurate color temps across the IRE range that way, but that's a guess.)

One solution to this blue saturation is to lower the contrast or gain until you get out of the blue-overdrive range. But that might be a lot darker than you want. I just measured a G70 last week and it seemed the blue saturated at a pretty low level.

The other solution is to defocus the blue. Have you ever noticed that, when you focus the blue, you can tell where it's sharpest because it gets *dimmer*? I'm not sure why R & G don't do that, but blue does. So by defocusing the blue you can get higher light output, which means you don't have to overdrive it so much to get the right color temp at 80-100 IRE, which means the blue doesn't run as hot in the lower IREs.

Defocusing the blue won't hurt your PQ. Your eye can't focus very well on blue anyway. You won't notice the defocus except maybe with test patterns.

How to defocus? Some projectors (like the NEC XG) have a "blue defocus" mode that does this automatically. I don't know my way around a G70 -- the one I measured last week was the first one I've played with. I found a "blue defocus" but it didn't seem to defocus nearly enough to flatten out the blue hump. I ended up changing the blue "MAG -- ALL" setting (which controls the full-screen blue focus) from 123 to 153. That got rid of most of the blue hump. I'm not sure if that's the right way to do it on a G70, but it seemed to do the job. (If anybody knows the RIGHT way to do it, I'd love to hear it!)

Maybe Powerstrip can control the output curve of individual colors, so you could flatten out the blue without defocusing it. I dunno, I've never used an HTPC. But defocusing the blue is the "traditional" way of handling the blue hump.

Gary

I'm certain the higher Brightness doesn't hurt a thing. That just sets the LOW end -- determining how bright the dim areas are. Too high and your blacks look gray; too low and your shadow areas get crushed. (I think that was happening before, and raising the Brightness is why you're seeing so many orcs in Moria now.) The only "damage" you can do by setting it a bit high is to make your picture look a bit washed out. Setting CONTRAST too high (see below) can cause rapid tube wear, but Brightness won't hurt it.

I am still wondering about the *right* way to do the blue defocus.

The other thing that puzzles me is the light output on this thing. It was pumping 15 ftL onto a 90" wide screen with contrast at only 56! I tried raising it up for a few seconds and it was blasting over 35 ftL at high contrast, with gains at only 225!!?! What a cannon! Is that *normal*?? But I imagine it would fry tubes in VERY short order if you ran it there. I didn't leave it there long enough to check for blooming.

Gary

The other thing that puzzles me is the light output on this thing. It was pumping 15 ftL onto a 90" wide screen with contrast at only 56! I tried raising it up for a few seconds and it was blasting over 35 ftL at high contrast, with gains at only 225!!?! What a cannon! Is that *normal*?? But I imagine it would fry tubes in VERY short order if you ran it there. I didn't leave it there long enough to check for blooming. Gary

Yup-- a 106" diag or 92" wide image. I'd say it came with a modified arc welder and mirror set-up instead of CRT's - :D

The overall color quality has also improved and a much truer white for sure.

Sorry in advance for this long post--
BIAS
Before: R 52.........After: R 52 (no change)
Before: G 60.........After: G 60 (no change)
Before: B 24.........After: B 52 (raised)

Contrast- remained unchanged at 56
Brightness- increased from 35 to 64
Output measured at screen after all adjustments: 14.8 FL

So here's the questions?
Have we hurt the PJ in anyway with these settings? Especially wondering if raising the brightness to 64 is an issue, even though keeping the contrast at 56.

2. Gary's question as to whether there is a better way to defocus the blue than cranking the "Mag-ALL" numbers? Is this amount of blue defocus unusual or out of line on a G70


On most G70's and with new tubes you will end up with the B Gain higher than the G Gain. I surprised you are the other way around.
When I set them up I set contrast and brightness to factory defaults or 80 and 50 and adjust from there to get a grey scale. Normally what happens in the practical setup (ie with the source) is that the brightness is a bit high at 50 and gets reduced a bit.
Your bias numbers look pretty normal.
You can look in the install or service manual and see the starting numbers that Sony uses for Bias and Gain.
On B defocus I usually start by adding 10 in MG-ALL and then add more as needed. IMO it is absolutely necessary to defocus B on most CRT's to get rid of the B hump in the mid ranges.

Terry

Defocusing the blue is normally done electromagnetically rather than a physical defocus. Blue only contributes about 8% or so to the total picture. Blue guns give out more light at the low end than they should (with the same drive current as the green and the red) but much less than is required at the high end. Before blue gamma circuits became available the only way to spread things out and obtain better gray scale tracking was to defocus the blue. CRT FP with blue gamma circuits now apply a correction, cutting the current to the blue tube only at the low end and boosting it a bit towards the upper end. Boost it too much and the mother will bloom.

Looking at a grid, the blue looks unfocused but your eyes are so insensitive to blue, you can't see a defocus of about 20%, just what the doctor ordered. Back in the old days, piss poor set up people would make the blue sharp bragging to their customers how poorly it came from the factory and how they fixed it. All they did was screw up the gray scale tracking.

So what I get from this is not to defocuss the BLUE on a newer ( IE DWIN HD700 ) because it has Bias controls and setting them will take care of the "de focus" for me.

Right?


Because I have actualy not defocused the BLUE lense at all. I have set the internal RGB LEVEL's as they were off when I got the projector and now that its all dialed in I cant complain at all about the picture one bit. Grey scale looks like its suppised to via interneal pattern and color test pattern looks all even. No one color looking like its too hot or over driven.

Back in the day I had a SONY 1031 and it had no gain controls and I tried the blue defocus. Granted I tried to do it with the lenses and it didnt work. I also tried it with the electronic focus control but I was never really happy with it defocused.

Wrong. The Dwin if I remember correctly does not have a blue gun gamma correction. Setting the G2 and drives has nothing to do with the blue tube putting out too much light at the low end and not enough at the high. To get good gray scale tracking you must defocus the blue 20%. You defocus it electronically on the blue tube only. I think you can do it by physically moving the magnets but i dunno. Ignore the blue grid looking unsharp. YOU WON"T SEE IT IN THE PICTURE. It is not like a convergence error. Blue will be 10% wider each line side than the sharp red and green lines. But you can't really see it, your eyes just aren't sensitive to blue. You will get more accurate colors across the spectrum.

I just had my marquee 8000 calibrated by Craig Rounds and he did the same blue defocus , also to get rid of the blue push he saw when he first came. he also explained that the eye cant really notice the defocused blue in normal viewing. He was 100% correct! i thouhgt i had a nice picture before now i see how messed up the colors were , everything now looks natural and the pic is clear and suprisingly sharp for an 8 incher running 1080p/60hz.

Athansios

I totally agree with all this!! In fact, with the XG's I've had...they need a defocus of the most you can do! That is to say...I'd go a little further with the defocus if I could.

Nothing is better than a linear greyscale!

Ben

Defocus the lens? How that get more output? Surely that just spreads the current output over a larger area, and therefore dimmer per area.

I'm a bit lost Mark...

AS far as I understand it *must* be done electronically. NEC even builds in a switch which gets you in the ballpark.

Is this electronic blue defocus done by eye? Or perhaps for example on a marquee, how many clicks up or down from pic>focus?

No Mark(_A_W), mark(_haflich) said you DO defocus electromagnetically. You're right that defocusing optically would not give you any more light output.

Gino, I defocused my Marquee by going to the RGB Focus menu. The *center* setting changes the electronic focus for the entire screen. I changed mine by 10; "perfect" focus was 48, and I boosted it to 58. That gave me a near-perfect grayscale.

Terry, thanks for the input. So the Blue MAG-ALL *is* the right way to defocus it. I dunno why the blue gain ended up lower than the green, but that's what it took to get the color temp right on Mark's G70.

Read Mark H's Dwin post again Gary :)

Gino, I defocused my Marquee by going to the RGB Focus menu. The *center* setting changes the electronic focus for the entire screen. I changed mine by 10; "perfect" focus was 48, and I boosted it to 58. That gave me a near-perfect grayscale.

Silly me, of course you want to change blue focus in service menu, the pic focus will change all tubes.

When you guys say defocus, are you saying underfocus or overfocus or doesn't matter?

Under focus I believe. So the focus point is past the phosphur, in the glass.

In practice I don't think it's critical.


Edit: Actually I'm not sure if "past the phosphur" is under or over. But past the phosphur is NOT the side which will give you a bright core with halo. It is the blurry dot side.

So we want to defocus so we see more of a blurry dot rather than a halo?

So we want to defocus so we see more of a blurry dot rather than a halo?

The blurry dot will give you more light output....hence the tube drive/focus/light output compromise.

So the Blue MAG-ALL *is* the right way to defocus it. I dunno why the blue gain ended up lower than the green, but that's what it took to get the color temp right on Mark's G70.

Yes Blue MG-ALL

Perhaps its a product of the tubes??

Terry

So we want to defocus so we see more of a blurry dot rather than a halo?

Defocus in the direction that the dot gets larger, not in the direction that it creates a donut.

Terry

The 9500LC VIM3 has a blue gamma correction circuit. Not sure about VIM2, but I think it does. Ampro made a special neck card for the blue. It was an option. Ampro replaced one part with about two others, resistors, diodes or caps, I don't remember. It was really simple, with two parts joined as a v above the board surface. I actually modified a stock Ampro neck board in my youth. Ampro sent me the parts. Obviously, with a blue gamma circuit, defocusing is not called for.

So if you have the blue gamma correction circuit there is no need to defocus?

On most G70's and with new tubes you will end up with the B Gain higher than the G Gain. I surprised you are the other way around.
When I set them up I set contrast and brightness to factory defaults or 80 and 50 and adjust from there to get a grey scale. Normally what happens in the practical setup (ie with the source) is that the brightness is a bit high at 50 and gets reduced a bit. Terry

One thing I think we might have done out of sequence was to set the grey scale initially with our stepped down settings of 56/35. Maybe Gary and I can have one more session at some point and re-run the process starting at the factory levels of 80 and 50 to make sure we are accurate there. That might be what has us now at an increased brightness level rather than it being stepped down a bit from the 50 factory setting. This will definitely cost me some more brewsky's, though. (cheap labor, huh?)

By the way, WAY too many Marks here to keep straight--

Mark S.

Read Mark H's Dwin post again Gary :)Um, OK...

Defocusing the blue is normally done electromagnetically rather than a physical defocus. ...
To get good gray scale tracking you must defocus the blue 20%. You defocus it electronically on the blue tube only. I think you can do it by physically moving the magnets but i dunno.I don't see anywhere that he says you should defocus optically. Now *Vic* mentioned defocusing optically, but Mark set him straight. So I'm confused where/why you're confused...

Am I the only one who sees that the High Power is a High Gain (2.8) screen so we are talking about only about 5 to 6 ftL coming from the projector ?

That does not seem extraordinary for me especially not taking into account the fact that the G70 has almost new tubes.

@lasermark4:

Don't worry, you are not overdriving your projector, it will last VERY long with these settings and the best thing about this is that you still get to enjoy a VERY bright picture :)

Um, OK...
I don't see anywhere that he says you should defocus optically. Now *Vic* mentioned defocusing optically, .. dam it Vic, leave it to a former DLP guy to confuse everything. :) Lets get this straight, you NEVER de-focus the lenses on a CRT for normal viewing. We're talking about electronic de-focus on Blue to curb it's output. AFAIK this only applies to machines with EM focus so this doesn't apply to the Dwin 700.
The marquee 02P and 03P VIM's both have gamma correction, but this wasn't a perfect fix. Blue output still changes as you ramp EM focus up/down. IT sounds like to ge the proper underfcous on a Marquee you turn it down below it's optimal setting, not above.

No, Oliver -- the sensor was pointing at the PROJECTOR, not at the screen. It really WAS blasting 35 ftL onto a 92" wide screen.

Mark runs it with 15 ftL hitting the High Power -- it's like a big plasma!

how can you guys run a 2.8 gain screen and not get hot-spotting?

Hot-spotting is more of a problem with angular-reflective screens. With retro-reflective screens like the High Power (or like stop signs and street signs), the important thing is how close your eyes are to the light source. The closer you are to the light source, the brighter the screen reflects, no matter what the angle to the screen is. There may be some slight hot-spotting, but nothing like you'd get with e.g. Pearlbrite.

You still get a bit of colorshift, assuming the projector is floor-mounted. If you sit on the blue side, you're closer to the blue CRT, so blues are stronger. On the other side the reds are stronger. It's noticeable but not dramatically so, unless you walk from one side to the other.

Mark's High Power looks awesome. I'd get a HP if my projector was floor-mounted. But with a ceiling-mounted pj, even with a low ceiling like mine, your eyes are too far from the light source. You lose almost all of the gain.

how can you guys run a 2.8 gain screen and not get hot-spotting?

I don't think you can.....unless it's some form of toridal screen

@lasermark4:
Don't worry, you are not overdriving your projector, it will last VERY long with these settings and the best thing about this is that you still get to enjoy a VERY bright picture :)

Cool-- just making sure I don't toast this thing, as I am a CRT'er for life at this point. Or at least until I see something that comes close to what I enjoy now-- and I haven't' found anything even close at this point digitally.

how can you guys run a 2.8 gain screen and not get hot-spotting?
I have a floor mounted PJ on a slightly elevated platform, so the angle of the throw is very close to eye level. The center seats have just an exceptional view, with the outside seats not very far behind. None of us has really seen much hot-spotting-- some very small amount on the full-on white screen if you move from one side of the room to the other. But certainly nothing noticeable at all with the normal picture playing on any of the sources.

So....OK, Dragon-- so NOW we've got your curiosity peaked, and you HAVE to drive north and join Gary, I, and Shaz on the Windsor Exit to see what we have got running in this Martin Logan theater! This will peak your interest even more- click here: LaserMark4's Theatre (http://67.19.167.226/~tdacquis/forum/showthread.php?t=3701)

Let's set a date soon-- for an FAC?? (bring a couple of Denver folks too!)

Mark S.

how can you guys run a 2.8 gain screen and not get hot-spotting?
Another key I have found with the High Power screen is that it picks up any outside or room light and displays an increased level of "white" or hint of luminescence on the blank screen-- I think a bit more than my previous non-reflective screen.

It LOVES a pitch black room to really perform to an extraordinary level. When it's dark in the room, the screen fades away as part of a black front wall. As I have removed the PJ's reflected light from the surrounding walls and ceilings, it continues to look just that much better during performance. That's not to say that it doesn't perform well with moderate light in the room on say a Super Bowl party level. But in pitch black, this thing is really, *WOW*!

Mark S.

Anyone know how the DVDO VP50 gamma circuit works for CRT? If I bump blue up to like 1.10 or 1.20 will this accomplish the same thing as defocusing the blue on my 1292? I dont know what the VP50 gamma adjustments do to the image, anyone talk to them or tested it?

Thanks, this thread is HUGE. I wondered why I had a very blue mid-range (35-70 was like 7500-8000k ) even though 30 and 80 were 6500k. This makes a lot of sense now.

You can't just bump it up. You have to cut it at the low end also, Atr the higher end the blue is quite limited, it is very weak. The defocusing increases the light output across the board. With the right blue gamma correction, there is no need to defocus.

So....OK, Dragon-- so NOW we've got your curiosity peaked, and you HAVE to drive north and join Gary, I, and Shaz on the Windsor Exit to see what we have got running in this Martin Logan theater! This will peak your interest even more- click here: LaserMark4's Theatre (http://67.19.167.226/~tdacquis/forum/showthread.php?t=3701)
Let's set a date soon-- for an FAC?? (bring a couple of Denver folks too!)
Mark S.
i'm game, i'm sure some of the guys from my HT meet would go too. Hopefully some of the tea-totalers so they can drive my drunk ass home :D I've been looking forward to hearing those Martin Logans.

I've got a Marquee 8000 (ancient but ridiculously bright for some unknown reason...) that I've never been able to get decently flat greyscale tracking. Tonight I defocussed blue by about 10 clicks THEN tried the greyscale again.
Measured with my Philips colorimeter, my greyscale is now almost totally flat. It starts about 6475, pretty much stays there right up to a final of about 6380 or so.
Thanks very much for a really useful thread !

The good explanation of what defocussing does to the gamma response of blue is what made me really understand it finally....

No, Oliver -- the sensor was pointing at the PROJECTOR, not at the screen. It really WAS blasting 35 ftL onto a 92" wide screen.

Mark runs it with 15 ftL hitting the High Power -- it's like a big plasma!

Gary, that obviously means that you measured up to 35 ftL coming from the projector.
Can you please tell me if you mounted the colorimeter flush with the screen and also what kind of colorimeter you used ?

Yes, the sensor was right in front of the screen. The colorimeter is an old Minolta TV2150. I've used it to calibrate my NEC XG, Dwin HD700, and Marquee 8500, so I've used it with several "comparison" projectors. Unless I was completely crazy, this G70 was putting out 3x more ftL than any of those other projectors.

Yes, the sensor was right in front of the screen. The colorimeter is an old Minolta TV2150. I've used it to calibrate my NEC XG, Dwin HD700, and Marquee 8500, so I've used it with several "comparison" projectors. Unless I was completely crazy, this G70 was putting out 3x more ftL than any of those other projectors.

Thanks, that puts things into perspective.
While I do not want to question your sanity I will have to say that a new XG also is VERY bright and should not be outputting less than 75% of the G70 - strange.

Alot feel that you can't use the Hi-power with a ceiling mount BUT I run it with a XG75 with the bottom of the lens 6 foot from the floor and the gain is Fantastic!! Espacailly from the back riser [about 12 in high]