DP1 Green Corners

Sigma DP1 Street

As the Sigma DP1 moves into users hands, it seems that it’s engineering compromise may be becoming clear- a tendency to produce green tinted vignetting, especially wide open and close focused. This image seems to show the problem pretty well. All four corners, but in this image, more at the lower two corners, there’s a shift toward green.

In the straight conversion from the camera it’s not very noticable, but this image processed through Sigma Photo Pro 3.1 on the Mac and then adusted in Aperture emphasizes the flaw. The bump in saturation and the contrast in the affect areas make it more obvious. As a few early adopters have shown, you can fix this in Photoshop pretty easily with the gradient tool. Carl Rytterfalk has posted a video on the basic method.

The suggestion on the DPReview message boards is that this is a result of the exit pupil of the lens being so close to the sensor. It might be why we have a lens with an f/4 aperture as well. I’m reminded of the Leica M8 release and the discovery of it’s infrared sensitivies. These issues must have been well known to the design teams, but these cameras have been so critical to the companies that they are released presumably pending further fixes.

On the other hand, I think of Olympus who designed the 4/3rds system for digital to avoid engineering compromises but ended up at a disadvantage to companies that made bigger compromises to incorporate larger sensors.

Sigma DP1 First Image



Sigma DP1 First Image, originally uploaded by jjvornov.

After seeing the first images online from the Sigma DP1, I decided I had to at least try it out as a substitute for the film Leica. It’s a landmark camera, being the first APS-C sensor in a compact form. The Foveon chip makes it interesting, but I’d rather have the sensor from my D300 in this form factor. The f/4 lens makes light gathering capability less than ideal. The need for yet another RAW workflow really had me doubting whether I should even bother. In the end, a small, sealed APS-C sensor package was something I couldn’t pass up. None of the compact cameras I’ve tried up until now seemed to be substitutes for the DSLR or Leica/film tools.

This is one of the first images I captured with the camera. It’s an in camera generated JPEG with the flat default settings from the factory.

So far, I like the controls as speed is adequate for my purposes. The LCD suffers by comparison to the D300’s bright, high rez LCD.

I think that once I get the Sigma software loaded and work in some better light I should be able to better take the measure of the camera as a tool for my work.

Work Capacity

Vern Gambetta:

Functional Path Training: Connections: “You will find if you grasp the idea that you will need to do less %u201Cfitness%u201D oriented training when you realize the cumulative effect of all the components of training.”

This was my first week of what I’m thinking of as my second Base period for the cycling year but is really the start of longer rides on the road. The weather here in Baltimore is finally holding above 40 and rain has been no more than intermittent.

For the last month and a half, I’ve taken a few shorter rides out of doors that were generally no more than 45 mintutes at high intensity. I’ve been doing 20 minute high intensity interval sessions (8 sec hard, 12 sec recovery) on the bike on the trainer. It seemed from testing that my VO2max was possibly higher than in late fall. So I was hopeful that I was starting at a higher level of fitness this year than I ended last year.

So, what’s it like to suddenly transition to a 2 hour ride on rolling terrain followed the next day by a similar 90 minute ride? I felt strong both from the perspective of the power I was able to deliver to keep the pedals turning an hills and I was able to place my heart rate where I wanted it pretty easily.

On the other hand I could see that what I lacked was work capacity. I was tired and the level of effort did create a good bit of leg ache on day 3. Not so much that I wasn’t eager to do another 90 minutes on day 4 after some rest. I’m hoping that the winter preparation will allow me to work somewhat harder during these longer rides through the early spring, bringing the whole effort up a notch from last year. Since I expect to be maintaining the power gains I might have made while extending the efforts, I would expect new effort to add. We’ll see how this plays out over the next few months.

At this point, I’m planning to increase intensity again in May and June, using July and August as a chance to increase volume further as the summer schedule allows.

Fitness

It seems that I helped introduce Dave Rogers to gadget enabled fitness. In the year that I’ve had my Garmin Forerunner 305, I too have found the battery dead on a number of mornings. Somehow it’s unpluged from the USB port and turns on or was left on. It’s got it’s own port and tends to need babysitting in order to be ready to go when I need it.

Dave is working himself very hard, judging from the heart rates he mentions on the site and given that, like me, he’s 50ish. I’ve been heart rate training for many years now and have developed a good sense of how relative percieved exertion (RPE in the trade), heart rate and my recovery state interact.

I’ve probably mentioned it before, but I think that Clarke’s “5K and 10K Training” is the best introduction to self coaching I’ve seen. He uses heart rate, RPE and classic hard-easy patterns to build an adaptable framework  for training. For me it’s allowed me to increase my volume and intensity without being burnt out and getting sick as I had in the past. I’m stronger and more fit this year, even though I don’t weigh any less.

My favorite accessory for the Forerunner is Ascent, a Mac-only GPS/fitness program from Montebello Software. It’s been steadily improved and the developers are responsive to reports of bugs and issues. Another great piece of handmade software made possible, I think, by Cocoa and the Mac development environment.

Window on Lago Maggiore

Window on Lago Maggiore

This is an image I’ve been trying to make for a long time. For a long time I was caught in a semi-abstract mode of image making that I always felt was not quite true to the medium. What I wanted to create were images that were both formal compositions and had three dimensional reality.

I started by avoiding these flat compositions and working in the 3D landscape. Now, returning to these formal compositions with better tools, I find that I can make things work in a way I couldn’t before.

Thanks to a tip from Janet at Tech Ronin , I upgraded my Mac Book Pro to 4 GB for $98 from OWC. The installation took 15 minutes and I have enough RAM in the computer to move through the workflow without memory thrashing.

I have a demanding workflow now. I import, grade and select in Aperture. If I want to post-process, the RAW file is exported out into a work folder on the disk and opened in Nikon Capture NX.

In Capture NX I tend to set white balance, black and white points first. Then I adjust values across the image since the combination of working on the RAW file and the U-Point technology is my best tool kit. The changes are saved in the working NEF file and a TIFF saved of the final image.

The TIFF is opened in Photoshop where I use a growing set of tools to get light, sharpness and local contrast right. The NIK Color Efex filters, NIK Sharpener and curves are the major Photoshop Tools at this point, guided by Vincent Versace’s approach as outlined in Welcome to Oz.

I save a full PSD file with Layers intact from Photoshop along with a flattened version. The flat version gets dragged back into Aperture, completing the round trip manually. I print out of Aperture and use FlickrExport to upload to Flickr.

With 4 GB on board, I can have Aperture, Capture NX and Photoshop all open at the same time with a small RAM cushion still left. The bottleneck is processor speed, especially since NIK Sharpener requires Photoshop to be run in Rosetta emulation mode as it is not Intel native yet.

The Clothes Phantom at the Wall



The Clothes Phantom at the Wall, originally uploaded by jjvornov.

I was relieved to browse the Aperture library and see that the vast majority of images that I’ve captured fit into the SIgma DP1 parameters: ISO 800 or less and F/4 or smaller apertures. With the D300 and its Auto ISO, I’ve captures quite a few outside of it- ISO 1600 at f/2.8. So while I’d rather have the D300 sensor and 24mm 2.8 in a compact package, I’ll give up 2 stops and continue to consider the DP1 sensor with an f/4 lens.

On the other hand, I suspect that the DP1 doesn’t beat a Leica plus Summicron. My f/2.0 lenses give me 2 stops over the f/4 of a Sigma DP1 and my ISO film is just a stop slower than the ISO 800 of the DP1.

Consultation on Rue Huber

Consultation on Rue Huber, originally uploaded by jjvornov.

This is a reworking of one of my favorite street photos. Here’s the previous version from last Febuary posted to Flickr

_DSC2036

It’s almost embarassing to put these two version up against each other, but I’m very pleased to see the progress I’ve made in creating memorable images. Importantly to me, I can see the same idea working in the previous version, I’ve just done a much better job of correcting the color and performing local brightness and contrast to tell a story and guide the reading of the image.

There are two factors at work. One is better tools. Thanks to my internet and book instructors like Thom Hogan and Vincent Versace,  I’m using Capture NX and Nik Color Efex in PS to post process. The improved white balance is due to the NX RAW conversion and neutral point tools. The enhanced lighting is from Nik Efex.

The second factor is a more conscious approach to bringing out the light in the photo. The buildings and sky in the background all share the same tone level now with the figures being emphasized in relation. The maps are now at full white with the faces brought up to a similar level, placing the figures in the darker background.

And I thought I used to be pretty good at this. I hope that in a year I’ll have the chance to rework this image again and once again be embarrassed by past efforts.

Does the Tool Matter?



Glow Plug, originally uploaded by jjvornov.

As I work through some of these Tri-X images from San Francisco, I find that the tools used deeply affect the images captured. These are simple, graphic somewhat grainy and low resolution images. I like these simple, almost mysterious black and white images, but their scale and ambition seems smaller than the clean, high resolution DSLR images which seem, by comparison, to be like medium format captures.

To the extent that I want to range widely over my visual vocabulary, it would seem I need to mix up the tools. Film, digital, rangefinder, DSLR.

Technical note on the image: The right side of the hydrant was blown out, or at least the negative was too dense to pick up texture with the Minolta Scan Dual III that I use. I recovered by adding texture lifted from another area of the image. It’s nice to be able to so simply save an area with “real pixels”.

Two



Two, originally uploaded by jjvornov.

I finally retrieved three rolls of Tri-X from a local lab which were shot back in December on a buisness trip to San Francisco. I had the lab (Techlab) here in Baltimore scan them, but the scans are unusable. Many blown out and most full of dust. My usual lab for film, National Photo does a vastly superior job on develop and scan, but they don’t run Tri-X, so I needed an alternative.

The scanning wouldn’t be so bad, but my VueScan/Minolta Dimage III set up is being cranky, with random crashes of VueScan and one crash of the entire system.

Given my experience of traveling with the D300 last month, I’m not eager to keep on trying to use film and scan as much as I like the look of the images. I’d rather drag the D300 and convert from color.

Now that the Sigma DP1 is getting close to release and some images are being posted to the web, there’s the possibility of a larger sensor digital camera for travel to replace the Leica/film setup.

Beyond 8 Second Sprints

In the last two months, I’ve added two workouts to my menu. Both are well suited to the winter weather and the extensive travel I’ve been doing.

First, I’ve been doing a 20 minute body-weight circuit in which I do 10 reps consisting of 30 seconds each of squats, pushups, lunges and jumping jacks. Each circuit then is 2 minutes. It’s a continuous, high intensity effort that pushes both strength and cardiovascular fitness.

Second, I’ve been doing 20 minutes of 8 second all out sprint with 12 seconds of rest on my indoor bicycle trainer. Using a high level of resistance, it’s similar in effect, but cycling specific.

I’m convinced, based on the work specificity that these efforts will push VO2 max and power. I can’t help but wonder how effective endurance builders they will be based on the specificity principle. While with warm up and cool down, a high intensity workout lasts nearly an hour.

As adaptation to a 20 minute workout is achieved, the question becomes whether to push intensity within the 20 minutes or to keep intensity constant and increase duration.

20 minutes is a convenient and achievable duration for high intensity workouts. By using intervals that allow rest (either as a circuit that rotates muscles or as intermittent sprints) sustaining a high level of cardiovascular effort can be combined with high power output. But I think that continuous power output, what Joe Friel calls “muscular endurance” is different from the VO2 max related abilities to clear lactate during short, high intensity efforts.

Yesterday I tried adding another workout on the bicycle trainer- a steady effort for 20 minutes at highest sustainable heart rate, about 90% of my 12 minute maximum effort on the Cooper Test. It felt about as hard as the 8 second sprint workout. I’m going to try to extend that effort from 20 out to 40 minutes for a full 60 minute workout.