about | support
home
 


  Reviews by: sfriedberg  

View profile View recent posts View reviews Add sfriedberg to your Buddy List
Canon i9900 Photo

i9900_586x225_1_
Review Date: Jun 27, 2005 Recommend? yes | Price paid: $324.00 | Rating: 10 

 
Pros: Very fast, guiet, high quality prints
Cons:
Red and green inks go unused, print length limited to 23 inches.

This is a very nice 13x19 (Super-A3/Super-B) printer. In full photo glory it is nearly as fast as an HP 960 business inkjet doing text. It is quiet, even without "quiet mode" engaged. The borderless printing feature works well. And it produces very nice prints on good paper.

I usually print on matte rather than glossy stock. Epson Matte Heavyweight has been a long-standing choice, and I've recently gone to Red River Premium Matte. Both papers work very nicely in this printer. Blacks are dense. K blacks are very consistent with CMY blacks.

There is no point to using this printer unless you are on top of your color management. And there is no point to using any particular paper for this printer unless you can acquire (or produce) a good color profile for the specific combination of paper and printer. Running my usual paper options and "Shirley" images without correction, there are some really LARGE print differences, especially in the blue-green range.
I am talking about "that's a completely different color", not
barely perceptible differences. You will not get good, repeatable results unless you use color management.

Canon supplies 5 color profiles. One is for generic BubbleJet, the other four are for Canon paper, presumably tuned for this printer's ink set and FW. The generic is very different from the Canon paper profiles, all of which seem very similar to one another. I am not interested in buying Canon's overpriced paper, so haven't pursued them further, nor rated the paper.

I have this printer on a Linux print server's "raw" queue, using print drivers on my Windows workstations. This configuration is not supported by Canon, but works well except for maintenance functions and the status monitor, most of which seem to require bidirectional direct access to the printer.

The ink set for this printer is "doubly extended" CMYK. Lower-density "photo cyan" and "photo magenta" are a common extension to form a 6 color set. This printer adds red and green to form an 8 color set. Only one black. Standard Canon BCI-6 13ml dye-based ink cartridges throughout. I've only used one set of inks for this printer, so I can't validate cost/page yet. Genuine Canon cartridges are available in the $9-12 range, which seems reasonable for inkjet ink, but hardly a bargain. There are many 3rd party inks at lower prices. I will be trying WeInk's formula at some point.

Not unusually, photo cyan and photo magenta seem to be the most heavily used inks, with yellow and black in second place. Somewhat surprisingly, I seem to have negligible use of the red and green inks. I was sufficiently surprised to call Canon tech support about this. The support rep told me that none of the following affect the use of red and green: paper type choice in driver, use or non-use of ICM in printer driver, use of raw or EMT print queue (with the understanding that I'm not using a supported configuration). I was told (by a man with a lot of uncertainty wavering in his voice) that the printer driver always sends the same 4-color information [funny, I thought inkjets used RGB color data even though they have a CMYK ink set] to the printer, and it's entirely up to the printer firmware to decide how much photo-cyan and -magenta and red and green to use.

The bottom line is the highly touted red and green "enhancement" inks are contributing nothing to my final prints. Other configurations might differ, but only if Canon tech support didn't know what it was talking about. (That would be both beneficial and unsurprising.) I have seen many superficial magazine reviews for this printer talking up how much of a different the red & green inks make. If they aren't being used, they can't be making a difference. I will eventually have to replace even the R & G cartridges, though: cleaning cycles use ink from all 8 tanks.

While there may be some color shift over a few hours of full drying, properly set up prints are dry to the touch as they come out of the printer. A 32lb matte paper (compare to 50 or 65lb "photo weight" paper) fully loaded with ink does feel limp, but there is no danger of smearing. On the hand, if you put plain paper in the feeder, and tell the printer driver you've got a high resolution photo paper, you're going to have a mess.

No Super-B-sized printer will be small. The Canon i9900 doesn't seem to be larger than necessary. If anything, the feeder paper rest (which unfolds, unfolds again, then extends) should be larger to give better support to 19" long sheets. The outfeed paper rest (which unfolds, then extends three times) is OK for size. At first I though it might not be wide enough to properly support a lightweight (32lb or under), fully wet Super-B sheet, but it seems OK.

100 sheets of letter-sized 24lb paper (5 mil? 5.5mil?) will sit in the feeder just fine. I did a double-sized job of 55 sheets (110 sides) of 24lb plain paper and had no feeding issues on first or second pass. Limp, curled or very light-weight paper may pick up some stray ink on the leading edge or on the sides near the leading edge.

Canon is rather conservative about paper weights. "Don't use anything over 28lb - unless it's Canon brand". That's nonsense. 65lb paper (10mil) runs through the i9900 without any trouble. I suspect you can do 75lb super-heavy, too. I have not tried feeding a stack of more than 3 sheets of 65lb paper, so can't comment on feeder reliability with photo-weight sheets.

I am a bit put out by Canon's entirely artificial restriction on paper length. Print length is limited to 23.39" (595mm), which
obviously puts a crimp in banner or large panorama printing. Canon's rationale, so far as anyone can get a straight answer from them, is: they don't currently offer a panorama photo paper, they don't want people to buy non-Canon papers, so they aren't going to allow people to use a feature unsupported by Canon papers. Canon, get a clue!

Hookup and installation is a cinch. The i9900 does have a completely mysterious feature. It has both a USB 2.0 and a USB 1.1 socket, only one of which can be used at a time. But USB 2.0 is backward compatible with USB 1.1, begging the question of what the USB 1.1 socket is really for. My advice: use the USB 2.0 socket and ignore the USB 1.1 socket.


 
Canon EF 24mm f/1.4L USM

ef24mmf_14l_1_
Review Date: Dec 19, 2004 Recommend? yes | Price paid: Not Indicated | Rating: 8 

 
Pros: combination of speed and focal length, build
Cons:
chromatic aberration in bokeh

I rented one of these for a week to take available light shots of various Thanksgiving gatherings and feasts. On a 1.6 crop camera, 24mm seemed a good focal length for interiors, closeup table spreads and the light. The 50mm f1.4 was the other lens in my kit on this trip.

The lens did not disappoint, but as you will see, I can't make a full critical assessment of the lens and even so found some issues.

Because the available light was meagre, most of my shots were taken at f1.4 or f1.8, and many of them were longish exposures. The depth of field at f1.4, even at 24mm (38mm equivalent), is so shallow as to aggravate any softness or shake problem. Many of my shots were, indeed, soft but I can't tell how much is the lens and how much is shaky hands.

Digressing from the lens a bit, the bright pools of available task lighting with much dimmer ambient light, the wide open lens and possibly shaky photographer, gave several of my shots a dreamlike quality. Much like soft-filtered glamour shots. Wish I were trying for that intentionally...

Build is solid, without excessive heft. With the hood reversed, it's about the same dimensions as my walkabout lens (28-135mm f3.5-5.6), just a bit shorter.

Focus was silent, lightning fast, and seems to have been accurate.

Color purity is hard to judge because the available light was heavily tungsten. The digital images certainly corrected to nice color, but I can't say if the lens added any color tone of its own. Color saturation was very good.

Contrast is also hard to judge, but it's certainly superior to my walkabout lens. I'd like to compare it with the 70-200mm f4L, which I consider to have very good contrast, but don't really have the images for a valid comparison. The color saturation suggests good contrast. Also, lots of detail is retained in the darker ambient-only parts of image, even when an exterior daylight window or light fixture is in the frame, which would have been a muddle for a poor contrast lens. However, the flat (if dim) ambient light makes it hard to find real "snap" in the shots I've got.

The best subject I've got for contrast is a wall-hung kimono with both metallic- and self-colored embroidery (yes, I did say Thanksgiving), lit by a task light from above which falls off strongly down the wall. On the one hand, the subtlety of the self-colored embroidery and the specular reflections of the metallic threads are picked up in both well-lit and poorly-lit areas. On the other hand, these photos were taken at f8 to avoid blowing out the pool of light at the top.

Flare: Initially I spotted a definite red halo around blown highlights such as light fixtures or their reflections when shooting wide open. On closer examination of my images, that was actually a real characteristic of one particular light fixture (!). Looking at exterior windows and other blown highlights shows pretty good flare control.

Chromatic aberration: The (shallow) in-focus area looks very well corrected, but the bokeh is noticeably affected by CA. Enough to be really obnoxious if you have background detail with a frequency matching the blur radius. And given the shallow depth of field, practically everything is affected! Again, this was shooting at f1.8, so I can't say how it improves at a smaller apeture.

While I can't give a fair apples-to-apples comparison (subjects and lighting were very different), I got a larger percentage of keepers with the 50mm f1.4 than with the 24mm f1.4, and my impression is that contrast is better with the 50mm. CA is certainly better with the 50mm. On the other hand, the 24mm took pictures the 50mm couldn't frame, and 50mm is notoriously the "easiest" focal length to design for 35mm cameras.

Bottom line: I'd rent it again in a heartbeat, but I would save my purchase money for a different lens. I have mixed feelings about giving this lens 4 stars. If it behaved as observed all the time, it would only be a 3 star. It would be at least 3 stars, because even with the issues I had, it was no disappointment to use. But I only evaluated it in an extreme situation, so presumably it's better than what I saw almost all the time.