daniel_taylor

"For those who say the filters don't make any difference in their images, how do they know? Do they shoot every image with and without filters?

I've shot enough with/without to know that Hoya S-HMC do not make a noticeable difference.

"Being flat, filters are are a much worse problem for reflections than elements."

I guess we should all avoid Canon teles with flourite up front. The front elements are flat elements that are there only to protect the flourite right behind them.

Moving on...why is it that Canon and Nikon can put excellent coatings on lenses but ship sub-par UV filters? Or could our lens element coatings be better if they licensed technology from Hoya? The test above with the Canon and Hoya S-HMC filters is telling.

Though I like Jean's advice and choices, if you are going to blow it all in one place and need another Canon configuration...

Canon 20D + EF 17-85 f/4-5.6 USM IS $1580: http://www.bhphotovideo.com/bnh/controller/home?O=productlist&A=details&Q=&sku=372827&is=REG&addedTroughType=search

Canon EF-S 10-22 f/3.5-4.5 USM $690: http://www.bhphotovideo.com/bnh/controller/home?O=productlist&A=details&Q=&sku=351542&is=USA&addedTroughType=search

Canon EF 70-200 f/4L USM $545: http://www.bhphotovideo.com/bnh/controller/home?O=productlist&A=details&Q=&sku=183198&is=USA&addedTroughType=search

Leaves you with about $600 left over for tripod and accessories. Pros: good beginner kit covers (35mm equiv) 16mm - 320mm with three lenses, all decent glass. (70-200 f/4L is excellent.) 17-85 is a perfect vacation/walk around lens. Cons: two EF-S lenses that will not work on future full frame. All zooms which are slower than primes and not always optically as good.

"Given the same serious approach, I a photographer learns faster on digital."

That was supposed to read: "Given the same serious approach, I think a photographer learns faster on digital."

Yes digital encourages one to shoot and experiment more. Yes practice makes a photographer better. It's silly to suggest that practice in and of itself is a negative thing. Tell a painter he'll be great by limiting himself to a couple paintings a year.

However, one's approach and level of dedication can override any inherent advantage/disadvantage. A "sloppy" digital shooter might in fact benefit from being forced to slow down and think about shots for a while, but that doesn't mean all digital shooters are sloppy.

Given the same serious approach, I a photographer learns faster on digital.

"Take your pick of favorite classic photographers. Would they be "better" if they'd had a digital camera?"

Many of them shot so much film that they would have kept pace with or exceeded a modern digital shooter. From the Learning section of this site: "Garry Winogrand is famous for having exposed three rolls of Tri-X on the streets of New York City every day for his entire adult life. That's 100 pictures a day, 36,500 a year, a million every 30 years. Winogrand died in 1984 leaving more than 2500 rolls of film exposed but undeveloped, 6500 rolls developed but not proofed, and 3000 rolls proofed but not examined (a total of a third of a million unedited exposures)."

Practice makes perfect, especially in art.

"And all the Canonites wonder why us Nikonians are so obstinate that we won't go back!"

Having helped a friend determine which of her dozen old lenses work on her new D70 (not many), I'm actually glad Canon took the route they did. EF works on EF, simple. For Nikon there are a maze of lens charts one has to wade through only to discover, in the end, that most lenses as old as the FD series won't work on modern bodies, or will be so crippled as to be nearly useless.

Canon bit the bullet and went with a modern mount. It was painful, but it's over. Nikon has the appearance of a good guy, but it's only an appearance. Their mount has evolved enough to make compatibility a nightmare, and Nikon users thinking they can mount their old high school lenses on a modern body have to go through the pain Canon users were over and done with nearly two decades ago.

"Arnab wrote "We don't need any high speed slide (or negative) film anymore. Digital does it better."

That's pure BS: Just ask any professional racing photographer who has to stop a 215 MPH car in color in 60 footcandle Musco lighting. [incidentally, I am... And have]"

Ask anyone who has ever shot a Canon 5D at ISO 1600 and compared it to ISO 1600 film.

It's not BS. Digital has done high ISO better since the D60 days, if not earlier.

Does anyone really believe that Epson ink jets limit their dithering in such a ridgid fashion that sending pixels at a multiple of 1440 makes a difference? (Trust me, they don't.) Or that if such was the case Epson's print driver wouldn't scale resolution before dithering? You think you noticed it but none of the engineers at Epson who are paid to notice such things noticed it? After over a decade of research?

Does anyone really believe that feeding an image to your printer vs. wasting time scaling in 10% increments to an exact fraction of the printer's dot resolution would make the difference between "oh, it's OK" and "wow! what a great photo!" to viewers, or that said viewers could even see the difference in a double blind study?

I honestly think that some people's time would be better spent reading The Da Vinci Code and pondering secret Epson societies and special printing secrets that the Catholic church doesn't want you to know about than printing photographs.

Set the image size to your desired print size in your image editor with interpolation off. Is the resolution around 250 or above? Sharpen if you want and print. Is it below? Turn interpolation on, get the resolution up to 250, sharpen if you want, and print. Why interpolate if it's below 250? To smooth out "jaggies", i.e. patterns that the viewer will pick up on because there are too few pixels for the print size.

Those of you who believe interpolating 10% at a time is magic need to do that with a sample, then interpolate the same sample normally, then USM the normal one. Then ponder how you could have spent those wasted minutes of your lives.

Unfortunately, no. The Canon FD mount was abandoned, and there are no digital SLR's to date which can mount FD lenses.

On ebay you can probably get a 2700 ppi film scanner for around $200 though if you need to get your images into the digital domain but can't afford to change to EOS right now.

Set your print size with interpolation off (in whatever image editor you prefer). If your resolution falls below about 250, turn interpolation on and get it at least that high.

Lots of people look for one magic number for color ink jet resolution. There is none for two reasons. One is that different shades require different numbers of dots to produce the shade. Pure black can be layed down at the highest resolution your printer can print because you have a pure black cartridge. An Epson photo printer can easily out perform a laser printer on B&W line art. A particular shade of orange, however, may need a 4x4 or larger grid of dots to simulate the color using the inks available in your printer.

Having printed many Epson prints including a number of resolution tests, I can tell you that the printers would struggle to resolve even 250 ppi with some shades. I have a series of test targets where you can see improvement up to 720 ppi, but only in one area (strands of a model's black hair against a white background), and only because the original source had that much true information to begin with (MF scan). In another area (an actual orange), there is zero improvement from 240 ppi up.

Which brings me to point #2: Epson's print driver is more than capable of scaling your image data before turning it into halftone patterns. You don't need to make sure you're on any multiple of 360 or 720. If you set your print size and end up at 257 ppi, or 301 ppi, or whatever, you're fine. Some really odd geometric test pattern might play with Epson's scaling and reveal a minor difference, but it's nothing you would ever see with a real photograph. The reason you want to interpolate up to 250 ppi if you fall below that is because otherwise the pattern of pixels starts to become apparent. (Actually you can get away with 200 ppi before pixels themselves become really obvious, but below about 250 you can start to see patterns in some areas of prints. Bokeh can get weird, for example.)

Hope this helps!

"I've had a change of heart and I'm going a completly different route. The 17mm of the 17-55 on the 20D still isn't as wide as I would like, and pairing it up with a 70-200 leaves a gap."

I'm not going to tell you your lens choices are right or wrong, but...why do so many photographers feel they have to cover every single mm from 16 to 600 with overlapping zooms?

Your "gap" in 35mm terms is 88mm to 112mm. In other words, it's a step or two on your part. What do you think people did before zooms, when they might have a 85 or 100 and a 135 or 200? Do you really believe that you will lose a shot because it would have been "just perfect" at 93.5mm but sucked at 88 and 112?

I think for weddings the 27-88 (equiv) zoom with f/2.8 and IS would be much more useful, and less swaped, than the slower, no-IS, 16-35 (equiv) zoom.

The bummer is that while this is probably the most valid, useful, and interesting "digital vs. film" thread in a year, it's limited by the realities of web presentation.

It's pretty hard to tell which, among the many variations, looks best. One could look identical to the film shot on screen and then look nothing like it in print. But John's second post seems the closest so far.

FWIW, the "straight from camera" digital looks pretty bad, but I think most of that could be fixed with a proper WB setting on scene.

"And you can go to Wal-Mart and get a REAL 8x10 PHOTO print for $1.99:"

If they're down to $1.99 I applaud them. However, I take exception that an ink jet print is not a "real" photo print, especially when R800/R1800 prints will last longer than the Fuji prints from Walmart. I have my choice of papers if I print at home, and I don't have to suffer automated cropping or edge bleeding at home either. If I have a bulk order, I trust WHCC (and have more paper choices, again) more than Walmart (though there's not a lot that can get screwed up with straight printing).

"The problem with CMYK inkjet printers is that they have a smaller gamut than RGB photo printers. And, to make matters worse, when inkjet printer manufacturers went to 6 colors, they used CcMmYK with c & m being light cyan & magenta, not bright red and bright green? -- And the use of light cyan & light magenta was an admission that flesh tones stunk (were not of professional quality) with just CMYK on these desktop printers."

The R800 ink set uses bright red and blue, not light cyan and magenta. Argue all you want about the gamut, with a scientific test chart you may even be right. But I can see no difference between my R800 output and output from a laser photo printer (Walmart, WHCC). At least none that cannot be attributed to paper differences.

"First is the grouping of all 3rd party inks in one class and calling them substandard. There are some that are junk and there are some that are at least the equivalent of what Epson produces."

Maybe, maybe not. Have they submitted to permanence testing? *Most* of them do not last. I cannot claim that 100% of them do not, I know there are some serious companies that make ink and paper for Epsons, and I know some of the papers are tested (and test very well) for permanence.

"Second is the comment about all Epson dye ink printers clogging when not used very frequently. This is simply not the case. I've had an Epson dye ink printer (890) running 3rd party dye inks for almost five years."

My comment was limited to Epson dye inks. I don't know what the problem is with their formula, but I've performed dozens of cleanings on Epson dye printers and seen three thrown away. For my life I cannot understand why Epson doesn't use their R800/R1800 ink set across the entire line and stick it to HP and Canon on image quality and permanence, and be rid of these stupid clogging problems forever.

Of course I don't know if you can't find anyone because you're not offering enough, but assuming you are, try California.

Heck, I know a spot in Cupertino where you can find some right now.

Let's see...a completely new, complex product has a few bugs and problems in the initial release. How can a company best address this when IT morale is at an all time low due to dropping wages and slim job availability?

A) Open a lab where testers can pound on Aperture and report the bugs to developers sitting right next to them, who can then immediately track down, classify, and in many cases fix said bugs on scene.

B) Offer bonuses to the team for shipping a stable product by a certain date.

C) Focus exclusively on bug fixes for the next 6 months, freezing feature changes until the code base is stable.

D) Fire the team that knows the code base best and bring in a new team of programmers who don't have the faintest clue how the code base is architected, then threaten them with outsourcing if they don't work 14 hour days to ship a perfect product with 100 new features ASAP.

What I want to know is why Jobs wasn't fired during the 5 years it took to make Mac OS X usable if that is the standard practice for buggy software.

Another example of why high school grads should avoid CS and engineering at all costs, and instead turn to law, medicine, or business administration.

I don't know if copying a Disney video for personal use will get you thrown in jail or not, but I'm a strong proponent of the death penalty for spammers.

BTW, "July Tree" can't get an erection, but do you think this link will help him? http://www.viagra.com

;-)

"Trust the market? With only a few manufactuers who all have similar incentives to maximize profit from ink sales, I'd be worried about collusion. The recent Epson lawsuits against the cartridge refill companies is not reassuring."

If you're going to trust the permanence of your prints to a 3rd party refill company, just buy a Lexmark. Or draw the picture with crayons. At least the crayon version won't fade in 3 months like the 3rd party ink. Or the Lexmark.

I've pegged Premium Glossy 8x10's at roughly $2 per print off my R800. That's better than any local photo lab, and about even to an order through WHCC. But of course I don't have to wait on the post office with my R800 like I do with WHCC. If there's greedy price collusion going on, it's not doing a very good job if an ink jet can still beat high speed laser photofinishing equipment in the hands of people not even earning a "living wage" at Walmart.

A dye ink Epson is probably about the same if you don't include the 1/2 cartridge, 10 cycle cleaning flush required for a clean print if you turn on the printer and then do something time consuming before printing like, say, launching Photoshop.

My friend got one of the notices for this lawsuit, having recently purchased an Epson printer. She said it was due to ink remaining in the cartridge when the printer demands a new one. I told her it was frivalous. You would want to change cartridges before sucking out the last drop to avoid a) a ruined last print, and b) the print head mechanism trying to operate with no ink, which could be damaging.

She went to throw the letter away and I told her she might as well use the rebate in case the cash value would otherwise go to the sleazy lawyers.

If the lawyers had taken the time to actually research the issue, they might have learned the real flaw: clogged print heads on the models that use dye based inks.

I now know of a third person who has thrown their Epson dye inkjet in the trash because no amount of cleaning or cleaning solution would save the damn thing. And this is with "genuine Epson cartridges". I love my R800 and have never even had to clean the print heads. But that's Epson's pigment ink. Their dye ink will dry harder than concrete if you leave the printer on but not printing for 5 minutes. The formula must be a close relative of super glue.

Though I think Epsons across the board produce the best photos with the most paper options, it forces me to recommend HP to people unwilling to drop $400 on a pigment ink printer.

Travel, like sports and weddings, often means you won't be able to change your position quickly enough to get the shot. And one stop of light means very little on digital.

70-200 f/4L.

If you can afford it the 70-200 f/2.8L IS is great because of the IS, but it's more than twice as heavy. Not sure you want that while traveling.

You would think that Canon users would be the ones obsessed with FF vs. DX sensors....

Nikon will introduce a FF body shortly after Sony or some other supplier provides them with FF sensors at a quantity and cost they like. It's that simple people. You haven't seen Sony announce a FF sensor for sale yet? Don't count on a FF body from Nikon then.

"Daniel, first, all of your discussion had nothing to do with the question asked in the first place. I didn't see Ron asking for that information."

Then maybe you need to re-read Ron's post:

"If you do the math, it suggests to me that this scanner is capable of 23.5 megapixels (3,946 x 5,959 pixels) for a full-frame 35mm scan. Is this correct? Doesn't this mean the scanner has a higher potential resolution than even the best current DSLR?"

"Your posted image is your posted image. You seem to have an axe to grind, so I am supposed to believe your posts over my own experience?"

No, you're supposed to back up your claims with a sample or two.

We're photographers. This is photography. You think lens A is sharper than lens B? Post a sample. You think film A is more saturated at sunset than film B? Post a sample. You think 35mm film does a better job with color and tonality than digital? Post a sample.

You want to go back and forth, tit for tat, talking on and on about your experience and your clients and "what you do" (which could be completely made up) and how that means people should believe your words over all evidence to the contrary --- please don't post at all.

If you have the experience you claim, then it should be a simple matter to post examples of what you're talking about. If you can't bring yourself to load Photoshop, crop a couple clips, and post them, then I don't want to hear about it. The debates are boring, and are not helpful when trying to answer the original poster's question.

"I would love to be shooting digital all day long! But it is not there yet!"

35mm was never "there" if "there" is defined as "depth". "Depth" in the context you used it (scan depth) refers to how many subtle shades can be represented, how smooth tonal transitions are i.e. the bit depth. Sorry, but there are too few grains on the miniature piece of film we call 35mm to offer really excellent depth and subtle color details. This, more than resolution, is why people used to turn to MF, and now turn to digital.

I really hope you didn't blow $25,000 of a client's bank account trying to achieve ideal depth/tonality using 35mm film. If I was offered $25,000 to get a series of shots with the best possible tonality, 35mm would be dead last on the list of tools. Film would be a contender in larger formats, digital would certainly be a contender, but 35mm wouldn't even be considered.