5 ms·
But, they did try: http://www.steves-digicams.com/camera-reviews/kodak/ http://www.steves-digicams.com/camera-reviews/kodak/ My understanding is, however, tha
by porsupah 12y ago
But, they did try:
http://www.steves-digicams.com/camera-reviews/kodak/ http://www.steves-digicams.com/camera-reviews/kodak/
My understanding is, however, that they simply weren't all that noteworthy.
Also, bear in mind the degree of pivot required - it's not merely a shift in strategy, but a fundamental transformation, from developing films, into developing entire electronic devices, with all the entirely new skills required in industrial design, UI, and so forth.
It can, of course, be done - Fujifilm has managed to continue in the digital photography sector, although they were rather more broadly based to begin with. Similarly, Nikon and Canon were already acquainted with what was necessary in a DSLR, and were able to transition to the digital world with aplomb.
To provide some context, consider this rec.photo thread from 1990, on the future of photography:
https://groups.google.com/forum/#!topic/rec.photo/ZnqcQVzAln0 https://groups.google.com/forum/#!topic/rec.photo/ZnqcQVzAln...
- ghaff 12y ago> Fujifilm has managed to continue in the digital photography sector, although they were rather more broadly based to begin with Well, they really became broader based in large part when they saw film was going to decline and branched into areas where they could apply their film expertise outside of photography. See e.g. http://www.economist.com/blogs/schumpeter/2012/01/how-fujifilm-survived http://www.economist.com/blogs/schumpeter/2012/01/how-fujifi... They do sell cameras today. (Nice ones; I own one.) But it's a relatively niche business. (And even so they struggled. Film sales absolutely fell off a cliff.)
- Tuna-Fish 12y ago>> To provide some context, consider this rec.photo thread from 1990, on the future of photography: > Right now, 1 Megabyte of memory costs about $45 retail. This will not drop by an order of magnitude in the next decade without a breakthrough, or an economical Gallium-Arsenide process to replace Silicon. Hmm.
- semi-extrinsic 12y agoFortunately the storage business has been ripe with technological breakthroughs since 1990, both for spinning rust and solid state memory. See e.g. the Nobel prize in physics in 2007. I still have my 8MB CF card that cost $40, I think that was around 2000, so for solid state the past 15 years have been insanely good.
- paul 12y agoThat rec.photos thread is great. I love reading old predictions. For some reason many of them got hung up on the idea that there would be crazy compression schemes that eliminate all the creative parts of photos. > O.K. Chemical films can presently store more information than any reasonable digital image-getter can. However, digital may be possible if one has a wristwatch-sized supercomputer doing various shades of image/abberation recognition capable of replacing the actual information with some sort of 'tokenized' equivalent of the original image. Fine. Such data compression/ reduction techniques may alow you to have your little computer-camera. > But wat if you want those abberations, or if you want that blur for that action shot, or if you want that sun/iris flare for a neato scenic shot, or if you want to use some weird lighting effects? Then you are s.o.l. By having the camera identify/correct (tokenize) and remove all of the possible artifacts of using a real-world camera (which is the only way that such a camera is possible), you will also eliminate almost all of your creative power.
- lfam 12y agoSounds like some complaints about digital audio compression schemes.
- rdtsc 12y agoYap. Remember my first digital camera was a Kodak. It wasn't very good compared to others during that time.