|
Notes from the Pocket Protector
The following comments are mine and mine alone and the test images are unofficial, not specifically scientific and are not sanctioned by Phase One (based upon my cheap test shot, Phase probably wouldn’t want it either). These shots AND my comments are based on a PRE-PRODUCTION digital back as well as BETA C1 v4.5Pro software. Hence some features are not “turned-on” in the software but read my comments…hopefully you’ll find this interesting. To effectively follow along with the content, I suggest that you download the files and have them open in PS so you can visualize what I am talking about. A link is provided below the images.
Intro
Well, I had a rare opportunity to get my hands on a P65+ and use the Phase One Camera System with their 80mm lens for a couple of days. ProGear held an open house two weeks ago that brought in commercial, architectural and fashion shooters and right off that tells me this has sparked some interest for all kinds of shooters.
P65+ and Phase One System Overview
Not to give in to the marketing hype but the P65+ does sport some pretty impressive numbers. According to the brochure (and we will compare these to my findings later below) the Phase One P65+ is a 60.5 Megapixel, full frame Digital Back capable of shooting 1 fps and has a useable ISO from 50 – 800. With 16 bit captures the shooter has about 12.5 stops of dynamic range. The effective (or usable for light gathering) chip size is 53.9 x 40.4 giving us 8984 x 6732 working “pixels”. Remember that one does not have pixels until the file is processed but for our purposes we’ll leave it at that. Files sizes are an ~173MB at 8-bit RGB (~346MB 16-bit RGB). Even better, the P65+ does NOT contain any transfats or high fructose corn syrup. Very healthy.
Computer and Lighting Specs
Just for the record I wanted to relate some of the particulars to you with regards to the capturing equipment. I was using a Mac Pro Tower 2.8GHz Quad Core Intel with 6 GB of memory, threw in some fresh AA batteries in the Phase 645 and even took a shower before shooting the tests (of which I found my colleagues to be eternally grateful). A Profoto Acute 2400R pack with 1 medium softbox overhead and 1 head with a 20-degree grid spot from the right side illuminated the scene. The distance from the wine bottle in the scene to the Phase One 645 system with the 80mm lens and P65+ chip plane measured at exactly 6 feet.
The Images
Shots were at ISO 50, ƒ/11 at 1/60th of a sec. I did not choose ISO 50 because of the better quality (I do have some samples shot at ISO 100 and can send them to you) but because I wanted to be able push the image a full stop through the C1 software.
Yes I still use a light meter for the mere purpose to compare what photographers are used to reading vs. what the manufacturers are claiming. My first observation is that I think that the chip is more sensitive than it is given credit. I was actually able to eek out almost 3/4 stop difference as opposed to what the calibrated light meter was reading. Remember, sensors do not have an ISO comparable to film but I would like to think that Phase and other digital imaging manufacturers like to keep it simple so we photogs don’t get too confused by associating signals with ISO. The following is a description of the images that will be associated with my comments.
Image 1 – A lo-res image of what I was seeing full-frame in the viewfinder. I knew that I was going to be working with some gi-huge-ic files (a Bush-ism) so I backed off a bit so knowing that I was going to put images on the website and not be bogged down with a ton of download time. Again, the distance from the wine bottle in the scene to the Phase One 645 system with the 80mm lens and P65+ chip plane measured at 6 feet.
Click here to download Image 1
Image 2 – Another lo-res image indicating areas of importance and items I was looking for in response of the chip. Specifically these items are historically problematic reds, yellows, sharpness and moiré patterns.
Click here to download Image 2
Image 3 – Cropped image from a P45+ at ISO 50 at ƒ/11.
For explanation see section below on Color Fidelity and P45+ vs. P65+
click here to download Image 3
Image 4 – Cropped image from a P65+ at ISO 50 at ƒ/11.
For explanation see section below on Color Fidelity and P45+ vs. P65+
click here to download Image 4
Image 5 – Cropped image from a P65+ at ISO 50 at ƒ/11.
For explanation see section below on Color Fidelity and P45+ vs. P65+
click here to download Image 5
Image 6 – Cropped image from a P65+ at ISO 50 at ƒ/11.
For explanation see section below called Moiré
click here to download Image 6
Image 7 – Cropped image from a P65+ at ISO 50 at ƒ/11.
See section called Sharpness
click here to download Image 7
Image 8 – Cropped image from a P65+ at ISO 50 at ƒ/11.
See section called Macro Detail
click here to download Image 8
Final Image – Final Cropped image with USM applied (~50MB file)
click here to download Final Image
Shooting and Processing Times
Although I was rather skeptical at the advertised shooting times, I was shooting at 1 frame-per-second. My personal opinion is that the H2 possibly can shoot a tiny bit quicker (based solely on acquiring auto focus, shooting, releasing the shutter button and then reacquiring the focus) but the Phase 645 held up it’s own. One item to note: if you have Auto Focus selected on lens and camera, keep your finger on the trigger, you will get a faster continuous shooting time than if on manual focus shooting continuously.
The P65+ packs a whopping 1.3GB of buffer and it shows. I was able to shoot continuously while tethered and had a difficult time filling the buffer when shooting to the card. The pack recharged quicker than the download of the image so I was not overshooting the pack and I was timing about 1 frame-per-second. Who’d-a-thunk it?
Processing times from RAW to pixel state was approximately 10 – 11 seconds per image (this includes generating a lo-res 72ppi, 8-bit 300ppi and 16bit ppi file all at the same time). Keep in mind that my computer is set up as a production machine as I have the machine stripped of any un-necessary “complimentary applications” (junk) that may interfere with processes.
Color Fidelity and the P45+ vs. P65+
Not because I am a Phase dealer but I have always been a fan of Phase products. That being said, I like a bunch of items and features in other DB’s and their software as well. It’s hard to find one perfect system that accounts for everyone’s needs and shooting styles. One issue that I have struggled with in the past with Phase and their software is saturation in the reds and yellows. I have found the reds and yellows to be a bit oversaturated based on their default settings. This is the first thing I attacked with the P65+.
To visualize what I am speaking about, go ahead and download Image 3, a file from a P45+. This is an 8-bit RGB file and if you scroll through the RGB channels, or Command>1, 2, 3 respectively (in monochrome, not color channels) notice the potential plugging of colors. When the channels are “plugged”, or flat, then you lose shape and detail in reproduction or output. You’ll see this in the blue channel (Command>3) where the bottom left side of the peach has little to no values.
Now refer to Image 4, a file from the P65+. As before, navigate through the RGB channels, or use Command>1, 2, 3 respectively. Notice any difference? Ya, you betcha Margie**. Nice shape and color detail in the blue channel of the peach and that is exactly what separators and color houses (as well as RGB printers) want to see; more gradation in tonal values. Both images were captured using the P45+ and P65+ using standard default settings, same speed, camera, etc. We are off to a good start.
Aside from the reds and the yellows, what about the yellows by themselves? Typically, as with the reds, yellows used to be a bit saturated for my tastes (to pure, no contaminants) and when converting to CMYK even through a color managed workflow, the yellows tended to be a bit of a “greenish” cast because of red in the initial capture. The absence of red promotes a higher cyan count when converting to CMYK. As we all recall from the ads years ago, “yellow and blue (cyan) make green”. Reference Image 5, I do not see this at all in the yellow threads with the P65+ capture.
Let’s not get too cocky though, we are not out of the woods yet. Due to the response of the chip vs. colored dyes, inhibitors, etc., in the products, there will always be colors that are near impossible to capture let alone reproduce. Remember the Hunter greens and Cobalt blues in film? Fun, fun, fun.
**Fargo reference
Moiré
Let’s face facts: as much as any manufacturer claims moiré free capture, all images are at least susceptible to moiré pattern effect. Basically the high-end digital capture devices that are free from “the beast” are scan backs and once super-popular and continual working 3-shot backs. The only time those moiré is when the images they captured had “screen” moiré caused by screen angles from separation when reproducing in print. Heck, even film has that problem.
Now-a-days moiré patterns can be reduced or at least predicted to a certain extent. The Phase One P30+ is touted as having a reduced moiré effect and is great for shooting fashion or fabrics and with that statement I have to agree. There has been a reduction of moiré pattern with P30+ vs. the P25+ or the P45+. This has to do with the construction of the chip itself and the incorporation of microlenses. In my personal opinion, repeat, my personal opinion, the microlenses tend to slightly soften the image just enough to break up any patterns caused screen frequency interference (moiré patterns) and are therefore not as noticeable. The problem with the P30+ and its microlenses is that is not a strong, or even recommended candidate for wide angle photographs as used architectural photography.
The P65+ does a great job with wide angle lenses (such as Mamiya 28mm) and at the same time has a drastic reduction in moiré effect. How can this be? There are no microlenses over the sensor and yet the image is tack sharp (see Image 6.) I purposely put in jeans and a tight weave pattern to see how the sensor/lens combination would react. The reduction in pattern has to do with the micron size of the sensor and the resolving power of the lenses. The smaller the size of the photo site, most often incorrectly identified as a pixel, the better for moiré reduction. The lenses have a hard time picking up the detail that can cause patterns. In addition, if there are color moirés as opposed to screen moirés, Phase’s Capture One v4.5 has the ability to reduce the color artifacts that can be mistaken as moiré patterns. Now the P65+ can be used for commercial, fashion and architectural shooters alike.
Sharpness
For you Capture One Pro v3.7x users, Phase has their proprietary sharpening (USM) numbers. In my workflow I use the Phase sharpening numbers to ONLY check sharpness of the image while capturing and never apply USM to the processed file. I apply USM only when I know the final size of the image that will be used. Normally this is done in Photoshop.
Refer to Image 7 and note the shooting distance. Here I was looking for overall sharpness in high-contrast areas to see how far we can push the USM based on the Dalsa sensor and the Phase 80mm lens. The distance from the wine bottle in the scene to the Phase One 645 system with the 80mm lens and P65+ chip plane measured at 6 feet. Not bad 'dere, ay Margie? **
Good news for the new version of Capture One: v4.x. Phase has incorporated similar numbers for USM as those used in Photoshop. Now the user has a comparative standard to use. I am not thoroughly convinced that they are the exact same numbers for effectiveness but that is really not an issue as all images can require different amounts of USM. Bottom line is now there is more rhyme and reason behind the USM numbers vs. C1Pro v3.7.
**Another Fargo reference, thanks Joel and Ethan
Macro Detail
Speaking of detail and sharpness; I replaced the 80mm with the Mamiya 120mm Macro lens to see detail of some Macro stuff. Refer to Image 8. This image is less than 11 inches from the scene and wanted to pick up threads in the dollar bill, jeans and the edge of the Color checker card. Mission accomplished.
Future Of Things To Come
The first time I used and implemented the first release of Capture One v4.x (several months ago) I was less than impressed because that particular version did not have many of the attributes of v3.7 available. Now, even with the Beta v4.5Pro, my opinion has changed. For example, some items that are now available (the complete list is too long to mention):
Sessions
Styles
Multi-format processing (a HUGE feature in my opinion)
CMYK conversions
Shooting to an auto-refreshed hot-folder with your DSLR, and it works!!!
Remember all of the testing and shooting was done with a pre-production P65+ DB and beta software. Not everything is turned on as of yet. I expect to see longer exposures and better quality at higher ISO’s. If Phase can accomplish with exposure time and ISO quality that the competition has not been able to in the past (Kodak chips), give them a little time. I am really looking forward to variable resolution choices on the fly (end of Q1?). Look for some great updates to existing P65+’s that are on the market.
Hope this helps, please let me know your comments and thoughts,
doug
This e-mail address is being protected from spam bots, you need JavaScript enabled to view it
|
