All these tips come down to the same: get the fastest shutter times possible while keeping the picture quality acceptable.

#1: take the highest ISO your camera can afford

If you can use sensitivity ISO 800 instead of ISO 100, you’ll win 3 stops. This is the difference between a picture at 1/40 sec (sharp) and 1/5 sec (blurry), certainly with moving targets (and tango dancers do move). Of course there is a trade-off: you loose colour detail.

This is the main reason why you’ll need a camera with a decent sensor, or in other words, an dSLR instead of a regular point-and-shoot camera. Even my 350D gets colour defects at ISO800. When they’re too bad, I tend to convert the pictures to black-and-white or somewhere halfway (by decreasing the saturation).

The camera I’m drooling over now, the Canon 5D Mk II, can take decent pictures at ISO 3200. That’s another 2 stops faster.

#2: take the lightest lens you can afford

You need a lens that will let in a lot of light, that allows you to use maximum aperture. Your standard kit lens that goes from f/5 to f/3.5 probably won’t do. You will need a better lens and it will probably cost some money.

I started with a Canon fixed 50mm f/1.8. Contrary to what I said before, this is a cheap lens (about €100). It’s a great portrait lens, but it is hard to get a tango couple from head to toe in a picture, you’d have to step back too far (50mm on a dSLR is slighty ‘tele’). It’s great for abrazo (embrace) pictures of the upper bodies.

For the last Brussels Tango festival, I bought a Sigma fixed 30mm f/1.4. It’s pricier, heavier but it fits the purpose better. The f/1.4 lets in more light (it’s a 1/3 stop ‘faster’ than a f/1.8) and the 30mm lens (which, on a digital SLR that does not have a full-size sensor, is equivalent to a 45mm lens) allows me to fit people full-length into the frame.

The drawback: with wide aperture, the depth-of-field becomes very narrow. This means: you focus on something e.g. 1.55m away, and every thing closer than 1.40m or further than 1.70m is already blurry. While this is also a feature (it’s a great effect for portraits up close), this means that if your autofocus is just slightly off, everything will be blurry. The error margin is very narrow.

#3: choose Av or Tv mode

Obviously you’re not going to use the standard ‘portrait’ or ‘landscape’ mode. You should go for either aperture priority “Av” or shutter priority “Tv” mode. The first one allows you to always use maximum aperture and get the best shutter speed that you can. The second one would make sure that you use reasonable shutter times for moving subjects (like 1/20 or 1/40 sec).

I prefer using “Av” and fixing the aperture on f/1.4. I take risks with that, because sometimes I end up with pictures with shutter times of 1/10 or 1/5 sec, but with some luck and good timing, they come out OK. It’s just my default “gimme all you got” setting. I don’t use “M” Manual Mode because I don’t have the time before each picture to decide how much aperture/shutter I want.

#4: use autofocus on 1 (central) point

Put your autofocus on 1 point, I use the middle of the frame. This is important because you want to control on what part of your frame your camera tries to focus. You have to force yourself to the following sequence: focus, reframe, shoot. The point on which you focus will be one with the most contrast. It might be an upperarm, where the skin and clothing touch, or the face, or some pattern with contrasting colours on the back. People who are fully dressed in plain black (and there are a lot of those in tango) are very hard to auto-focus on. The framing you use to focus and the actual frame you want to shoot will seldom be the same. So you need to reframe right after focusing. You focus by pushing the shutter button halfway and then you keep it pushed like that while you reframe. When the composition is right, you click all the way and the actual picture is taken.

This is the main reason why people unfamiliar with my camera have a hard time taking acceptable pictures with it. They point at a dancing couple, they push the button all the way immediately, the camera focuses on something in between the two dancers -typically a far away wall- and the shot is unsharp.

#5: underexpose and fix it later

If the 4 tips above still don’t make your shutter times acceptable, try using exposure compensation (the AV+/- button). I sometimes use a full stop of underexposure and I correct it in ‘post’. I use the contrast maximisation in Picasa to stretch the histogram again.

Also, in tango venues, there are a lot of spot lights (to make it cosy). If I focus on a dark part and then reframe to something that includes some of those lights, they might end up clipping part of the picture. When I underexpose, this becomes less of a problem.

The drawback of this method is very similar to using high ISO values: it messes up your colours. So in a lot of cases, you might have to make the picture black and white. But if the moment, the composition was valuable, and the shutter speed was just good enough, you can still end up with a really nice picture.

So how much does this all add up to? If you were to take pictures with a compact camera: f/2.8 lens, ISO 100 and shutter time of 1/30 sec, you need a lighting situation of EV 8 (40%). In my situation, a f/1.4 lens, ISO 800 and the same 1/30s shutter allow me to take pictures in a EV 3 (20%) situation a.k.a. “very dark”. I just won 5 stops. With underexposure, I can go lower still, to EV 2. It makes a lot of difference.

Related posts:

1. Point and shoot badly I went to the concert of Bajofondo Tango Club...
2. New lens for my Canon camera My current favourite lens for my Canon 350D is the...
3. Tangotation: writing tango steps I have been looking for a way to write down...

• the logical i’m-used-to-this-digital-stuff version: “could I get those pictures in 1800×1200” / “Can I have at least 2 megapixels“?
• the weird I-used-to-work-in print version: “can you send them in 300dpi“?

DPI (dots-per-inch) only make sense for me if I would know on what size you want to print them. If you’re making an A4 flyer, that’s 8½ × 11 inch, and you need 300 dpi, then that means you need 3300×2550 pixels. If you want to print only an A6 size, that’s 1650×1275 pixels. So don’t tell me what DPI you need, tell me what pixel dimensions you need. Yes, you can save the DPI parameter in a JPEG file, but it changes nothing to the data. My 1200×800 picture with a DPI value of 72 or 300dpi is still, pixel by pixel, the same picture. Your image viewer might decide to show it as a bigger picture on the screen, but the image data is identical.

Beginning to see the pattern? No matter what DPI you set (or, as it happens, what size in inches) so long as you don’t let Photoshop resample the image up or down, it’s STILL 504 by 144 pixels.

Let’s save this one as 12dpi.TIF before we forget, then have a look at the file sizes. You’d think that a 300dpi file would be higher resolution than a 12 dpi file, and because of that a lot bigger, right?

Sorry. All three files are exactly the same size.

(from pptfaq.com)

So, if you’re asking someone for a high-resolution version of a digital picture, ask for minimum pixel dimensions, not for DPI! Otherwise you just show that, while you might have experience in managing print, you have no clue how digital imagery works.

Related posts:

1. To upsize a picture, use the B-spline algorithm If you take a low-res picture, and you want to...
2. Megapixel myth nuances What’s the value of more megapixels? Is 10 megapixels better...
3. What is HD-JPEG? While reading the specs of the PackardBell Store & Play,...

Step 1: the camera

As you might know, I’m a Canonista, I use a EOS 350D digital SLR. I’ve got two CompactFlash cards of 1GB, of which I use the second one only rarely. I don’t take my photos in RAW format, I use the highest JPEG resolution: 3456 x 2304 pixels or 8 megapixels. On a 1GB card I can get between 250 and 300 pictures.

Step 2: import into Picasa

For my photo processing I use Google’s Picasa. I’ve never been fond of software like Photoshop, although I see other people do great stuff with it. I just don’t have the patience. When I connect my 350D to my PC with USB, Picasa starts up automatically and begins downloading the pictures. After 1- 2 minutes the whole card has been read and the photos are shown as a grid of small thumbnails. I typically select all of them (a “roll” of film, in iPhoto speak) and import them into one folder, which I name “YYYYMM-EventName”. The month-based name makes it easier to archive and clean up afterwards. Picasa then deletes all the imported pictures from the memory card.

Step 3: editing in Picasa

Picasa has three different editing menus. The first one is a simple one with 6 buttons and a slider. The second is a more detailed control of colour and brightness. The last one features 12 ‘common effects’ like conversion to B/W, sepia, saturation control …

The sequence I usually follow is:

• if necessary, I crop the picture. This would be to cut out unwanted elements or to reframe for esthetic reasons, like go from portait to landscape. I usually use the 10×15 format (a.k.a. 3:2 aspect ratio).
  It also happens I have to ‘Straighten’ the picture, when vertical lines in the image are clearly not 100% vertical.
• I tweak the colours. This might be just a ‘maximize contrast’ (one click) or a more detailed manual ‘Fill light’. The auto-maximize has a tendency to never use ‘Fill light’ (which makes everything lighter) but always ‘Highlights’ (which makes the light parts lighter, but leaves the dark parts dark). This sometimes leads to exaggerated white spots of people’s foreheads, or leaves the face too dark compared to the rest.
  To compensate for incandescent or other artificial lighting, I sometimes use the ‘Color Temperature’ slider to bring the red down and the other colours up.
• When the colours are not ‘right’ or too distorted from using ISO 800 or higher, I tend to convert to black-and-white, with an optional ‘Sepia’ or ‘Warmify’ afterwards.
• I’ve also started experimenting with decreasing the saturation. I take a lot of tango pictures, and when there are bright red lights around, the photos are already oversaturated.

When I’m happy with the result of my tweaking, I add the picture to my selection with the ‘Hold’ button. I go through a whole 250-items photo set one by one, selecting which ones to use and doing some corrections, in something between 15 minutes and an hour.

Step 4: Export from Picasa

When the selection/edting phase is finished, I push the ‘Export’ button. Picasa will create a copy of each picture, with all the changes I’ve indicated, but in a smaller format. I typically use the ‘1200′ setting: the longest side is never longer than 1200 pixels. This will give me exported pictures that are smaller than 1MB: they upload faster, but are still big enough to print if one wants to. Picasa will put these exported pictures in a separate folder under My Pictures\Picasa Exports.

Step 5: upload to Flickr

When the export is finished, Picasa will pop up the new folder. I select all the images and right-click on ‘Send to Flickr …’. I have used both the Flickr Uploadr v2 and v3, and while the version 3 is more flexible, it sometimes doesn’t work, and the version 2 always works. So, for version 2, I add the tags that will be added to each picture and press ‘Finish’. When the Uploadr is finished, it will give me the option to go to the Flickr website and change titles, descriptions and tags for each picture.

Step 6: archiving and cleaning up

Every month I move all originals older than 1 month to external storage. I’m currently at > 100GB in photos, so you can see why that’s necessary. The Picasa Exports folder I only clean up once a year. For every 1GB of originals, I only generate 25-50MB of exported pictures, so I don’t mind if they linger around a bit longer.

This, in short, is how I process my pictures. Picasa is quite intuitive, so I never really had to read any manual to figure this out. However, good luck trying to do this in iPhoto. My sister gave up on Flickr because she always uploaded the 3MB originals and when I tried to figure out how to select and resize a number of pictures in her iPhoto, I gave up after 1/2 hour. iPhoto is not a workflow tool, it is a simple photo editor.

Related posts:

1. Five tips for taking tango pictures in dark environments I have been taking tango pictures intensively for more than...
2. Point and shoot badly I went to the concert of Bajofondo Tango Club...
3. 100.000 Flickr views Yesterday I reached 100.000 views on my Flickr account....

But a 50mm lens in a small digital SLR is equivalent to a 75mm, and sometimes that’s too much. I have to step 2m backwards to get a portrait of two persons (people ask this, you know), and even further if I want a picture of a tango couple with their feet included as well as their head. So I’d like to find a second lens with at least the same aperture, and with a lower focal distance, at a non-ridiculous price (below 400€).

Candidates seem to be:

• Canon EF 28mm F 1.8 USM 58mm – 411€
• Sigma 28mm F 1.8 EX DG Asf Macro Canon – 295€
• Sigma 24mm F 1.8 EX DG ASF Macro Can AF – 385€
• Sigma 30mm f/1.4 EX DC (HSM) DG Canon – 399€
• Tamron starts at 750€, so yeah …

Does anyone have experience with these lenses? Any suggestions?

Related posts:

1. Five tips for taking tango pictures in dark environments I have been taking tango pictures intensively for more than...
2. Point and shoot badly I went to the concert of Bajofondo Tango Club...
3. The popular Canon 400D The new Digital Rebel It might be me, but...

HD JPEG = HD Photo = JPEG XR (my guess)

Well, it’s not that easy, actually. When you do a search for “HD JPEG” on Google, all you see is promo talk. Vendors like Philips and Panasonic use it as a feature on their devices, but there is no official definition of what it stands for. What I think happened is:

• Microsoft developed a standard “HD Photo” (formerly Windows Media Photo) that corrects some of the disadvantages of JPEG:
  • It can store color information at 32 bit per color instead of 8 or 12. (In current JPEG, “24-bit colour” means: 3 x 8 bits for R, G and B.)
  • All encoding steps of the algorithm are lossless (except for quantisation). So it has a true lossless mode (all coefficients = 1). JPEG does not have this (at least, not implemented).
  • All kinds of tech mumbo-jumbo reasons (read the Wikipedia article)
  • “HD Photo offers image quality comparable to JPEG-2000 with computational and memory performance more closely comparable to JPEG”
• The JPEG committee announced that they would (consider to) adopt HD-Photo as a standard, with as a working name JPEG XR. Provided that, of course, …
  

  One important aspect regarding the standardization of HD Photo is Microsoft’s commitment to make its patents that are required to implement the specification available without charge. Microsoft’s royalty free commitment will help the JPEG committee foster widespread adoption of the specification and help ensure that it can be implemented by the widest possible audience. The JPEG committee hopes and encourages all participants in its meetings to consider this royalty free approach when offering patented technology as a candidate for standardization.

• The hardware vendors didn’t like the “JPEG XR” name (it doesn’t really show clearly that it’s an upgrade from JPEG), so they combined it with an acronym that consumers know and associate with “better” and “more expensive”, so was born “HD JPEG”.
• in short: Technically, it does not exist. Practically, it sells flat screens.

Competition

JPEG XR/HD JPEG tries to fill the gap between JPEG (small files, widely accepted standard, but compromises quality) and RAW (huge files, proprietary in a lot of cases but the highest possible quality).

In 2004 Adobe launched DNG (Digital Negative) but this is more an effort to have a ‘unifying camera raw file format’ (uncompressed and compatible between different brands).

OpenEXR is a file format for HDR images (High Dynamic Range), created by Industrial Light & Magic (George Lucas’ company) in 2003. It has lossless compression, and most importantly 32-bit floating point per pixel/color. It is used as a storage format for CGI, that saves as much dynamic detail (dark vs. light) as possible, until the scene is finally rendered onto a more classic format (JPEG2000, JPEG …)

JPEG2000 is the standard developed by the JPEG group in 2000 to allow for higher quality still and moving images. It uses wavelet compression and has excellent image quality. Unfortunately it is also rather slow for both compression and decompression, and most importantly, no on uses it for images. Motion JPEG 2000, on the other hand, is the standard for digital cinema.

PNG (Portable Network Graphics) was developed from 1996 onwards as an alternative for the patent-undermined GIF format. For photos it does not compress as well as JPEG.

TIFF (Tagged Image File Format) is a container format that can store multiple images in 1 file, can do CMYK as well as RGB, but due to old compression algorithms, creates huge files.

Who uses it?

That’s an even better question. Windows Vista can read HD Photo files, some flat-screen TVs can read them from memory cards, some media centers can display them but within popular consumer applications, only Adobe Photoshop can create them. For as far as I know, no camera saves its pictures in the HD Photo/JPEG XR format. So I wouldn’t let this feature decide on what TV you should buy, yet.

Related posts:

1. Adobe authors Digital Negative (DNG) standard Adobe has developed a new format for ‘raw’ camera files,...
2. A JPEG picture doesn’t care about no DPI Every now and then I get a request to use...
3. Megapixel myth nuances What’s the value of more megapixels? Is 10 megapixels better...

She asked me if she could use it, and because An is a friend of mine, because I know Nathalie, because I use a Creative Commons license and maybe also a bit because I like the idea of having an image of mine in the papers, I agreed. So there it is: my first newspaper publication!

(hat tip to Clopin for getting me the PDF version!)

Related posts:

1. A JPEG picture doesn’t care about no DPI Every now and then I get a request to use...
2. To upsize a picture, use the B-spline algorithm If you take a low-res picture, and you want to...
3. Converting a color picture to a stencil Just strolled onto this post: “Converting a colour photo to...

The pictures that came out rather well are in my Flickr Bajofondo Tango Club album.

I did have some fun with a guy in front of me with a ‘point-and-shoot’ camera, it might have been a Canon Ixus or something. He added a twist to the common ‘people-using-a-camera-flash-from-a-distance’ error. He would see a beautiful image in his viewer, push the button and a totally awful grayish picture would show up as a result. For his sake and that of other P&S’ers, here’s two rules for using it in a concert where you’re in the audience.

Rule #1: do not use your camera flash

You have a simple small flash in your camera that might reach as far as 5m maybe, but it is of no use for a podium 20m away. Your camera will think that the subject of your photo will receive a bunch of light from it and choose a faster shutter. The light of your flash will never reach those musicians, but chances are some will fall on members of the audience that are standing nearby, so the only thing that will light up is some bald heads, a fake blonde and a lot of dandruff.

Switch the flash to ‘off’ or use the ‘landscape’ setting! If your camera permits it, set the ISO-setting to as high as 800ISO (if you own a 1000€-plus camera, 1600ISO is safe too, but you shouldn’t be reading this then). This will make the image grainier, but will make the shutter time shorter. If the scene is still too dark and your camera uses shutter times of more than 1/20 sec, try to include more light in the composition. Don’t zoom in too much, it will only make things worse. You can always crop the picture when you’re home.

Remember: when you’re using a small flash at a concert, or even worse, a mobile phone with the built-in flash from 30m away, you look like an utter amateur.

Rule #2: do not hold your finger in front of the flash

This one cracked me up: the guy in front of me held his left index finger right in front of his flash. So most of his pictures were extremely dark, with the odd one that included a completely white fragment of his finger. The only times that this wasn’t the case, was when he took pictures in portrait, turning his camera 90 degrees to the left, which brought his flash even lower and added quite some very bright shoulders to the composition.

So, if your flash photos are way too dark, and your finger feels very warm whenever you take a picture, check where the flash sits on your compact camera and make sure you put no bodily parts in front of it.

Related posts:

1. Five tips for taking tango pictures in dark environments I have been taking tango pictures intensively for more than...
2. Beyond the megapixel Wired just released an article on digital camera technology:...
3. Megapixel myth nuances What’s the value of more megapixels? Is 10 megapixels better...

Living Portrait of President Woodrow Wilson, for which 21,000 troops assembled at Camp Sherman in Chillicothe, Ohio, in 1918, is the best-known of Mole’s photographs. The image is characteristic of Mole’s work in that it wavers between the compositional effect of the whole (i.e. a portrait of Woodrow Wilson) and the desire to focus upon the obscured individuals who constitute the image, thereby undermining the optical illusion of the totality to a degree. To call this image a portrait would be misleading because the subject of the representation is not so much the countenance of Woodrow Wilson as what he represents and symbolizes.
via Cabinet Magazine

No related posts.

Apart from the fact that the transporter screwed up their tracking of the package, the whole process of showing up at closed doors and going back seems so inefficient. It’s like so much effort has been spent to smoothen out the process of purchasing online, but the physical delivery still works basically the same as twenty years ago, eventhough the drivers now have wireless devices and you have to sign on an electronic sensor.

Let’s describe how I would have preferred to have my goods delivered:

Package Delivery 2.0

• When the order ships:

  From: Vendor Customer Support
  To: Peter
  Subject: Your purchase of 22/3/2007
  Dear Mr. Forret,
  
  we thank you for your business!
  We have just prepared your order and released it for delivery.
  
  The transport will be taken care of by NewDelivery Inc,
  and you can expect an email from them shortly
  to allow you to choose the delivery options.
  
  Your package number is #34556554456 and
  you can track it from this moment onwards(*) on
  http://track.newdelivery.com/(...)
  

  (*) The times I have gotten the message: “this is the tracking URL, but it won’t work yet”. If there is a unique tracking ID, then just display “Currently waiting for package to arrive at our depot” or something.

• From: NewDelivery Customer Support
  To: Peter
  Subject: The delivery of your package #34556554456
  Dear Mr Forret,
  
  We have just received a package for you that
  we want to deliver at your earliest convenience.
  
  Tomorrow between 9h and 12h one of our drivers
  will be visiting the following delivery address:
  (...) 1000 Brussels
  
  If you will be there to accept the package,
  there is nothing you should do now.
  
  If, however, that is not the right time/place to deliver,
  please choose one of the following options
  1) I would like you to come at a later time
  2) I would like you to deliver at a different address
     (e.g. your office)
  3) I would like to pick up the package
      at one of the nearby pick-up points
     (see some suggestions further down)
  
  Click the following link if you want to change your delivery options:
  (...)
  

• Let’s say I change to option “pick up myself”

• From: NewDelivery
  To: Peter
  Subject: The delivery of your package #34556554456
  Dear Mr Forret,
  
  We have changed the delivery options as you requested.
  
  Your package will be available at the following pickup location:
  [Post Office/Fuel Station/Bookshop/...]
  (...) 1000 Brussels
  
  Don't forget: if you pick up the package within the next 2 days
  (before March 26, 20h) you will also receive our PearlyPickup surprise gift.
  

  My gift would be something small like a sticker or a gift certificate, but at least I am rewarded for picking up early and helping keep the storage space limited.

• Or even more sophisticated

  From: NewDelivery
  To: Peter
  Subject: The delivery of your package #34556554456
  Dear Mr Forret,
  
  We have just received a package for you that
  we want to deliver at your earliest convenience.
  
  Since you already have a package stored at
  [Post Office/Fuel Station/Bookshop/...]
  (...)
  we will let this package join the other.
  
  If, however, that is not the right time/place to deliver,
  please choose one of the following options
  1) I would like you to come at a later time
  (...)
  

In my opinion, this would streamline the delivery process, keep the costs for unresponded house calls smaller, and give the buyer more confidence in the whole process. There is already a pick-it-up-myself service Kiala that works with bookshops, supermarkets and fuel stations (I use them for Proxis deliveries). But now you have to choose for either that mechanism or the delivery by a courrier, whereas it’s easier to choose the best formula once you know which day the package would be delivered. The Belgian Post could only deliver such a full service when the post offices would stay open until 18h or 19h, so that people that are not retired or unemployed would also have a chance to use them. To me, current practices leave something to be desired.

No related posts.

Most viewed pictures

The top five of my most viewed pictures are:

32450 views. I admit, this was total linkbait, used for Size doesn’t matter.

10557 views. A totally ugly tweaked picture about the threesome scam

5787 views and also a favourite of 29 people. My not-so-bad Web 2.0 overview from the early days of the meme (Sep 2005)

5482 views. I have no idea where those came from.

4196 views. A very unappealling Excel scatter plot about anorexia amongst models, used in The next German top model will be thin

One thing you could derive from this top 5 is that I’m better at choosing good illustrations for blog posts than taking pictures myself. None of the top five is a picture that I took with my camera: 2 were made in Excel, 1 with my Get-Remix-Deliver chart generator and 2 are reworked photos from elsewhere.

Interesting pictures further down the list

Just some highlights:

Popular pictures by me

In my list of most viewed pictures, the first one that I actually took myself is:

#19 in the list, 1313 views. What a surprise, eh?

The most ‘interesting’ (according to Flickr definition) picture that I took myself is this one:

214 views, 2 favorites, but nevertheless #6 in my Most popular list. I’ll never understand how ‘interestingness’ is calculated …

The self-made picture with most favorites (a measle 3) is this Nokia N91 moblog picture from my neighborhood:

155 views. The seats are gone now, and they seem to be renovating the building.

My Flickr experience

I became a Flickr Pro user just one year ago, March 19 2006, and I love everything about it. The ’stream’ paradigm instead of a collection of separate albums, the tags/sets/groups, the notes, the clean layout, the AJAX interface, the uploader, the upload allowance, the clean URLs, the easy embedding and even the way they tell you their servers are down for maintenance. The typical complaints I hear from other people are:

• “it’s so slow”: I don’t agree. I am not the world’s most patient person, but I’m never annoyed by its slowness. I think that in the eye of the beholder. I for one always have the feeling that Mac computers have a slow, luggish interface, but I bet a lot of people wouldn’t agree.
• “I can’t figure out how to add comments/become member”: I agree. When I see Flickr photo pages while I’m not logged in, I find them not too inviting. I can see how people have trouble getting over that first treshold.

But overall, for the photography buff that I am: Flickr totally rocks! If you use it a lot too, make yourself a Pro, it’s the best $25 you’ll spend this year!

To those of you who regularly check my Flickr stream: thanks for helping me towards 100K! Do you have other favorites amongst my Flickr collection? Photos that are exceptionally pleasing or awful? Have some good advice? Let me know!

Related posts:

1. Photography workflow with Picasa & Flickr I’ve been thinking about writing this for a while, but...
2. A JPEG picture doesn’t care about no DPI Every now and then I get a request to use...

Contrary to what I expected, it was not the Lanczos filter (slowest and ‘best’) that gave the best results, but the B-spline algorithm that’s supposedly a bit worse. With Lanczos you get an rastering effect (check her hair), with B-spline it’s more even.

If you look at the simulation I made earlier, you see that B-spline blurs out the picture more evenly.

Lanczos creates these circular artefacts around magnification by 8 or 16.

I will do some tests on magnification of high-resolution photos to super-high resolution: e.g. from a 6 megapixel picture to 150 megapixel (width and height x 5 like above). I have a feeling the artefacts of areas that have less entropy/complexity (a 100×100 pixel area might contain only some hairs of the eyebrows instead of a full face) will be different. Why would you want to create a picture bigger than 10 megapixel? An 8-megapixel should not be printed larger than 21.8″ (553mm) x 16.3″ (415mm) at 150dpi. With a 25 megapixel image, you can go to 38.5″ (978mm) x 28.9″ (734mm).

Related posts:

1. How to upsize an image Printing in larger sizes I have a Canon 350D digital...
2. A JPEG picture doesn’t care about no DPI Every now and then I get a request to use...
3. Megapixel myth nuances What’s the value of more megapixels? Is 10 megapixels better...

No related posts.