Adobe Photoshop Training with Dan Margulis

Does it make sense to outfit an entire production room with Solux lighting? Is there a downside to working this way as opposed to a correct lighting area / lightbox?

For background, we are an outdoor advertising printer - billboards, etc. The lighting in the printing area is vastly different from the prepress area. My goal is to standardize lighting, but I'm not sure if it makes more sense to create a small correct area in each department, or outfit both departments with Solux lighting as the sole light source.

Thoughts?

Josh Hope

___________________________________________________________________________

I did it. One downside is the expense of having to relamp with relatively expensive bulbs that don't last very long. We use the 4700 degree lamps with a 36 degree beam angle and for the most part, I bounce them off the ceiling with a few choice areas lit with direct pools of light. We still use GTI viewers for accurate color viewing however. I don't think that the Solux lamps are good for that purpose at all because they have bad color fringing at the edge of the beam.

___________________________________________________________________________

I moved my photography studio into a new building recently and am in the process of dealing with exactly this issue. The Solux lamps are great but I can't see using them to light any sort of large work space very well. I'm using a mix of Solux lamps and GE Sunshine florescent tubes. The GE tubes are supposedly 5000 degrees Kelvin with a CRI of 90. Much higher than most standard tubes. They cost several times as much as well. But the difference in viewing conditions versus your garden variety cheap tubes is tremendous. So far I've lit my gallery area solely with Solux lamps on track light fixtures. Two production areas (large format Epson printers and print finishing work areas) that connect onto that room are lit with the florescent tubes in standard four tube suspended ceiling fixtures. There is not an obvious shift in image quality as you move from one room to the next. There most definitely is if you move into one of the office areas still lit with "whatever's cheapest" florescent tubes. I'm going to install Solux lamps to light the area around my editing workstations. So far it seems that this mix of Solux and a good color quality florescent is a reasonable compromise between color quality, cost, and efficiency to light a fairly broad area. Solux lamps in the really critical areas with fill from good florescent tubes in the broad areas where I don't absolutely need perfect color but don't want to endure huge color shifts either.

Bob Smith

Accurate Image • Bob Smith Photographer • Waco Texas USA

http://www.accurateimage.org

___________________________________________________________________________

It might be pretty, but also pretty expensive. Each halogen fixture only gives you a small cone of light.

Ciao, Paul D. DeRocco

___________________________________________________________________________

Folks:

I'm sure this will be controversial to some.

I've been through the whole issue in previous lives and tried some of the types of measures proferred below when I ran a photo studio/printing business in the 80's and 90's.

To make a long story short, I think that going to the tremendous expense and trouble of using this type of lighting is unnecessary. I use 6500 kelvin flourescent tubes everywhere now, available very inexpensively and probably almost everywhere. I also match the color temperature of my monitors to this and it seems like a happy balance.

Even though the standard to 5000 kelvins it seems like it's very yellow if you go there. Secondly, as I Josh and Bob have alluded to, it's very difficult to achieve a reasonable brightness level.

Trying to "nail down" the some of the subjectiveness of "color shifts" by attacking the light in which you work, to this extent, is a bit like a dog chasing it's tail.

Yes, some color consistency in these areas is essential, and there are some definite DON'Ts like don't use the way-out "cool-white" ultra cheapo flourescents, don't use wildly varying areas of brightness, don't use tungsten, and especially, don't used mixed sources.

However, trying to bring all the light in the color correction area to this very specific and hard to achieve (costwise, at least) standard is about as sensible as, if I may paraphrase, "tryjng to profile a cloud at sunset".

Just my .02 (Canadian .03)

Ron Kelly

___________________________________________________________________________

I agree with Ron on this; our lab follows the same cost effective approach. All overhead FL Fixtures are fitted with 6500K tubes, including customer service, administrative offices, hallways. We find that consistent lighting is both easier on the eyes - not having to transition and adjust to different temperatures - and allows prints to be viewed anywhere, anytime, without giving thought to (overhead) color temp.

It is important that the tubes have a high CRI index, over 90. As the tubes age, metamerism may rear its ugly head... but we've seen satisfactory life approaching 5 years. In bulk, tubes run from about $7 on up. We spend about $2,000 every few years replacing all 250 tubes at the same time.

For general production area lighting, this has worked well for us. When critical analysis is required, viewing booths are used.

Les De Moss

DigiGraphics

___________________________________________________________________________

What viewing booths have you guys found to work best? I have the GTI TRV, but I find the cover of the lighting to be inconsistent from top to bottom. Are the new GTI's any better?

___________________________________________________________________________

We use GTI transmissive/reflective (combo) boxes.

Les De Moss

DigiGraphics

___________________________________________________________________________

So do we, but I would add that the dimmable fixtures are best for matching to a monitor. The normal box is much too bright.

john c

tech photo & imaging

___________________________________________________________________________

Is there a particular brand of 6500K tubes that you would recommend?

___________________________________________________________________________

Josh:

Presently I have Sylvania Daylight Deluxe F40DX bulbs, but I'm sure there are others. They all have their specific brand and model names, but the essential part is that they are full spectrum, 6500 Kelvins. You may have to press somebody a bit to find out what the colour temperature is, but I found them at a hardware store.

Cheers,

Ron Kelly

___________________________________________________________________________

A few comments:

I use SoLux in my office, and successfully turn the entire work area into a viewing box. I control the amount by adjusting the distances to the lamps. There are four lamps, with some overlap and that provides a good degree of uniformity. They do put out more heat than a fluorescent tube, but they're low voltage and wattage so it's not a big problem with just four bulbs.

The critical factor in lighting conditions is not the color temperature, or really the CRI. It's the spectral power distribution. The CRI is just an overall score. It won't tell you if the bulb got an D in greens, but a bunch of A's and B's elsewhere. It's like your kid coming home with a report card with just one grade on it. If he gets an F in history, you'd probably want to do something about that, but if you didn't know it, how could you?

A big part of the reason 5000K "looks yellow" is when the eye hasn't adapted to that lighting condition because light of another brightness or color temperature is spilling into the vicinity, preventing adaptation. 5000K can work quite well.

Where a well behaved spectral power distribution comes in particular handy, is in two cases: first, avoiding metamerism. This is often induced by certain ink sets, usually the yellow component of the set, and green light. Fluorescent has a nasty green spike. That spike can cause some ink sets to have rather green gray balance, or if it's offset just right (or wrong depending on your point of view), it can cause a lack of green light for a particular ink set resulting in magenta gray balance. Second, nearly all ICC profiles, at least by default, are built predicated on a particular spectral power distribution known as D50 (sometimes 5000K referring to the SPD of a black body heated to 5000K; technically they are a little different but lets not go there). Therefore your ink set will perform as expected under that kind of light source. The more the light source spectral power distribution deviates, the less ideal it will be. It may result in color casts, which are rather obvious, or it can be as simple as things appearing flat.

As for the term "full spectrum" - a flashlight is full spectrum. A candle is full spectrum. Full spectrum simply means if you plot the amount of each visible wavelength coming from the light source, there will be a value. Something that is not full spectrum would be a red LED for example, or a laser.

However, we also have to be practical. It's not always possible to do the technically correct thing. Either it's not feasible to install, or it costs too much, or it doesn't illuminate a wide enough area, or whatever. Most likely, in the print world, NO ONE is really using D50 as a light source. They are all fluorescent based, so the illuminant is closer to F7 or F12 rather than D50. If you adhere to the standard strictly, you may be technically correct, but you will also be technically different than most everyone else who would view your prints or your proofs. This can lead to color miscommunication, which is exactly what we're trying to avoid.

The best advice I read was from Les De Moss, which was the resplacement of *all* bulbs at the same time. That's a fantastic way to ensure consistency. I'd also note here that fluorescent bulbs do last a long time, seemingly. They won't start to noticeably die until around 10,000 hours. However their color rendering behavior does degrade noticeably beyond 2500 hours. I believe most manufacturers recommend about a 2500 to 3500 hour replacement interval for fluorescent tubes. This happens to be in line with a SoLux bulb, which maintains its color rendering to around 100K over its life until it flat out burns out. I see an inordinate number of fluorescent bulbs that are well outside of their useful life, even though they seem to be working just fine. I suggest moving those elsewhere in a building color critical lighting is not important, and keep the new up to date ones where it is important.

Bottom line. In general I recommend for print related companies to use a D50 simulator, fluorescent based. If you have a printer you work with regularly, or big customer you work with regularly - I'd make an agreement with them, maybe even get a bunch of bulbs all from the same lot all at the same time. That consistency will be vastly more important than spectral power distribution. The gotcha is with inkjet proofs, which have ink spectral reflectance different than that of printing press inks. That means the proof and press sheet can match under true D50, but have noticable discrepancies under fluorescent lighting. If you profile the inkjet printer for the actual lighting condition, you can reduce or eliminate this problem. Only the proofer really needs to be profiled in this manner (yes you could do it for the press also since that output will be viewed under the same fluorescent illuminant but I think it's better to target separations to D50, and target proofs to actual lighting conditions and see if that alone solves the problem satisfactorily).

For photographers, painters, and other fine art types, I often recommend SoLux if they can swing it. The materials used by this category of end user are almost always best viewed under a better D50 simulator. Fluorescent based ones, especially 6500K, will have more UV emission which is probably not so great for longevity of your prints. I doubt they're going to fad in a matter of weeks or months, but possibly years of display with fluorescent lighting isn't a great idea unless there is such a thing as a low UV lamp available.

Ok I think this is quite enough!

Chris Murphy

Color Remedies (TM)

www.colorremedies.com/realworldcolor

___________________________________________________________________________

Chris

how do you square using 5000k light system with your recommendation to use 6500k on the monitor? Love you guys book by the way.

thank you, john wickham

___________________________________________________________________________

The only justification I can figure out for 6500K is that phosphors tend to have a natural color temp that is on the blue side, so attaining a color temp of 5000K means turning the brightness down. Personally, I prefer having both my viewing area and the monitor at 5000K.

--

Ciao, Paul D. DeRocco

___________________________________________________________________________

Yes, I'm not sure what Chris does, but we match the monitor white point to the D50 light box as much as we can, and it does result in a dimmer monitor, so we have to dim the light box as well. Once the eye adapts, there's no problem working this way.

___________________________________________________________________________

Exactly. They key is having a setup that allows for the eye to adapt, and that's going to be luminance matching first and foremost. I regularly see 5000K calibrated display, and 5000K light boxes work well together, and likewise 6500K calibrated displays with 5000K light boxes working well together.

Chris Murphy

Color Remedies (TM)

www.colorremedies.com/realworldcolor

___________________________________________________________________________

John,

We do explain this in the book :) Luminance matching is more important than color temperature matching. A 6500K calibration for a display will invariably be brighter than a 5000K calibration, and therefore closer to the likely illumination provided by a lighting system.

The recommendation for 6500K is not universal, however. It depends on the ambient environment. If the ambient environment is brightly lit such as office fluorescent lighting, or outdoor light seeping in, the human visual system will always choose to adapter to that light source, rather than a 5000K display. A 6500K display will look better in those situations. In a dim environment, a 5000K display will be more appropriate. Also, there is a continuum. It's not 5000K vs. 6500K. 5500K, or 6200K is perfectly OK as long as it's getting you the results you want in you environment.

Chris Murphy

Color Remedies (TM)

www.colorremedies.com/realworldcolor

___________________________________________________________________________

Say Chris,

A commercial photographer here. At the risk of being castagated would there be any danger in using 6500 for Apple Studio Displays where the only source of light in the room is household Tungsten bulbs. You know, the 2700K version. No windows whatsoever. No Solux here - yet.

My viewing booth is a 5000K photographic light table turned upside down with foam core frame. Seems I remember somewhere that even though photographic light tables claim 5000K they are really much higher effectively and have a green spike as all fluorescent bulbs do.

I have your book and have always used 6500 and 2.2 as per your general suggestion. I have had fairly good results profiling using NWP (6507 on my machine) and 2.2 All profiled using EyeOne Display and iMatch 3 software with the Luminance set to around 130 for LCD.

Or would I be better served to eliminate the tungsten lighting altogether. Read turn off lights. Perhaps profiling to D50 and 2.2.

On this LCD and my Powerbook that type of profile has always seemed real dingy yellow.

Thanks in advance,

Doug Walker, FP

website: http://www.walkerphoto.com

Phone (360) 943-1293

Olympia, WA

member- ASMP, PPW

___________________________________________________________________________

Doug,

We have D50 and monitors set to 6500. Tungsten task lights on desks were a common thing until someone complained that their file had changed and now had a yellow cast on the upper right edge. I would turn off your tungsten light(s).

Lee

P.S.-- Regarding the rest of this thread another suggestion is to mount timers on the fixtures working when the lights are on. We found that 2500 hrs. is a good max time for stable color.

___________________________________________________________________________

Also note that there is a "warm up" delay with fluorescent bulbs that is around 30 minutes, and depending on what you're viewing and how critical you're viewing it, the difference in timing can be quite noticeable. So turning them on only for the 5 minutes when you need to use it, and then off again is probably asking for less consistency even if you might extend the bulb's life somewhat.

Chris Murphy

Color Remedies (TM)

www.colorremedies.com/realworldcolor

___________________________________________________________________________

Would you be extending the life at all? I would think the opposite, that by constantly turning it on and off that you would severely shorten its lifespan. I was always under the impression that bulbs, like motors and many other devices, suffer the majority of their physical wear-n-tear during startup (which is why you are much more likely to blow a bulb when you turn it on, than for it to just suddenly go out).

I am certainly not an expert on bulbs or their physical properties, please correct me if I'm under the wrong impression here!

BRIAN PYLANT

Manager, Electronic Prepress

:::::::::::::::::::::::::::::::::::::::::::::::::::::::: ::::::::::::::::::::::::

Disc Makers

7905 North Route 130 * Pennsauken NJ * 08110-1402

Toll free: 1-800-468-9353 ext. 5539

http://www.discmakers.com

___________________________________________________________________________

Are there any guidelines or standards for the relative luminance between the room lighting, monitor display, transparency illuminator, and reflective copy illuminator? Or are they all supposed to be the same?

Randy Wright

Color Services

___________________________________________________________________________

on 1/21/05 5:35 PM, Pylant, Brianwrote:

Nope, I agree and along with Chris' note about "warm up delay with fluorescent bulbs" the bulbs are best (including life span) cycled on-off as little as needed. Using them 5 minutes when needed is asking for less consistency.

Lee

___________________________________________________________________________