Go Back   Steve's Digicams Forums > Digicam Help > What Camera Should I Buy?

Reply
 
Thread Tools Search this Thread
Old Oct 9, 2005, 6:57 AM   #1
Junior Member
 
Join Date: Oct 2005
Posts: 13
Default

Hi,
I hope somebody can help me finding the right camera for a very special setting. For my work, I am taking pics of electrophoresis gels under UV illumination; think of a slab of gelatine-like jelly with little fluorescent dots in it when illuminated from the bottom with a UV source, photographed from 30-50 cm distance with a camera stand.
In the past, this was done with Polaroid large format instant film and very simple stand cameras. Quality was superb, but unfortunately the film is incredibly expensive now ($1.50 a shot and we are doing dozens a day). There are a number of CCD solutions on the market which are very expensive and not very good, so I am trying to build a system from consumer components.

The light conditions are rather low in my setting, and focusing is not trivial as we are talking about a transparent jelly that does not have crisp features (i.e. the AF systems in consumer cameras are surely not primarily designed for this :-)). I am looking for a sensitive, high-res camera that gives me good live viewfinder with control over exposure, shutter, zoom and focus while adjusting and shooting from the camera stand. And I also would like to have a very straightforward way to print the shot immediately.
Many considerations for "normal" photography don't interest me at all, like AF quality, anti-shake, shooting speed, startup time etc.

I guess the first decision I have to make is if this can be a camera/direct printer system, or if it has to be PC driven. I could set up a camera with a monitor on TV out plus a direct printer. Is there a camera could give me detailed and convenient manual control of settings (focus, zoom , exposure, shutter, if possible contrast) and a direct print button?

I'd like to avoid a computer driven system, but have been playing with Canon's remote capture software for my old DigiIxusV3, which is clearly not up to the job. On the "remote" forum I found that the Konica Capture software seems to do most of what I need. I have to decide now whether to buy a Dimage A2 (does the A200 support the software?), a Powershot with 3rd party capture software, any compact or SLR camera with a direct printer, or something else that I have not thought about yet. Price is not a primary consideration, if there is a SLR that can do the job I can spend the money.

I would very much appreciate any input, and apologies for cross-posting.
Best wishes
Andreas
molecularbiologist is offline   Reply With Quote
Sponsored Links
Old Oct 9, 2005, 7:16 AM   #2
Administrator
 
Join Date: Jun 2003
Location: Savannah, GA (USA)
Posts: 22,378
Default

How low of light are you talking about here?

Can you give us an idea of the exposures needed (ISO speed/film speed, aperture, shutter speed)?

If not, can you post a sample image you tried with your Canon? That would be best, so we can figure out why it's not up to the task (focus, distortion, noise, etc.) and can look at the camera settings it was trying to use (these are in the image file itself in what is known as the EXIF).

You don't want to "jump out of the frying pan into the fire". :-)

How small is the subject (and how much of the image frame do you need it to fill)?

What will the images be used for? If prints (I see you want to print immediately now), how large will they be?

Is the live feed absolutely necessary, or can you use an optical viewfinder instead?



JimC is offline   Reply With Quote
Old Oct 9, 2005, 9:22 AM   #3
Senior Member
 
BillDrew's Avatar
 
Join Date: Jun 2002
Location: Hay River Township, WI
Posts: 2,512
Default

The other bit of data that would be usefull is a better idea of the number of shots/month. Taking "dozens per day" to mean something like 50/day - 200/week - 800/month, that means about $1,500/month or $18,000/year.

If that is a reasonable estimate of your costs, you should think about one of the scanning backs, likely with a view camera and bag bellows. Pricy, but still would save you money and produce images way past anything from even the high end dSLRs. At a price past those as well.

A dSRL with a macro lens might do the job for you.

Keep in mind that a scanning back with a 22Mp image has the complete RGB info in each pixel while a CCD/CMOS camera with 22Mp has only the info for Red OR Green OR blue for each pixel. The probably doesn't work out to a factor of three in resolution, but is likely to better than a factor of two.
BillDrew is offline   Reply With Quote
Old Oct 9, 2005, 12:09 PM   #4
Senior Member
 
Join Date: Aug 2005
Posts: 245
Default

I think the only way you'll get a direct print system is if you have a camera with printbridge technology. But, I'm not sure if you can use the camera while it's plugged into a printer (or if it has to be unplugged to use). That might be a hassle.

You'll also want a camera with a rotating LCD. When the camera is on a stand, it will be difficult to see the picture and controls if the LCD is up high (and you have to see on top of it).

Your best bet would be to find a camera and some control software that will allow you to use the camera via a PC. Sure, the PC adds space and cost, but (imo) it will be so much more convenient for you.

The camera's ability to focus on your sample will really depend on the distance and the contrast/brightness of your gel. If you have lots of bands and low background noise, then most camera's should be ok with focusing. You might need a diopter, though if you are too close and/or zoomed in too much.

Of course, you'll also want a UV filter to block out the UV illumination (and allow your sample to shine through). If you don't, your gel might not have very good contrast with the background.

What about this system from UVP? http://uvp.com/new/index.php?module=...lay&ceid=6

There is 100% control of the camera via software, but it is a PC driven system which you said you'd like to avoid.

Good luck in your search.
swgod98 is offline   Reply With Quote
Old Oct 9, 2005, 3:55 PM   #5
Junior Member
 
Join Date: Oct 2005
Posts: 13
Default

Thanks a lot for your feedback. Here's a bit more info:
The Ixus was not up to it mainly because the software has very limited features like poor live viewfinder and little control of exposure parameters. But this is the RemoteCapture 1.1 I think, pretty outdated....
The UVP system that swgod98 mentions is typical of what is on the market. The reason I don't want to go that way is that they sell you a plastic box, something like a PowerShot G6, and a bit of software that doesn't really help a lot, for 5-6 times the price of the camera.
The commercial systems that I tested have been using a G6, which gives acceptable pix with poor handling, though. Handling is essential for us as we have to do this several times a day and don't want to get an ulcer :-)
I think I could live with a camera quality in the range of the G6 for most specimens in terms of sensitivity, if I can manage to get good handling: Put the gel in the darkbox, adjust area and focus while on white light, switch on UV, adjust exposure, and (ideally direct) print. We could use AF lock to freeze the focus after the first step, where we ususally focus on a piece of printed paper stuck on the top of the gel.
The gel contains an unfriendly carcinogenic chemical for staining, and we have to wear gloves during the whole procedure, so you see where my handling issue comes from. Maybe a computer based system would be best, if we can control the whole setting with a trackball (wrapped in foil :-)).
My guess is that the Polaroid 667 (ISO 3000) that we used is better than the current digitals because of high sensitivity with very little grain. You can basically expose weak signals with high aperture and long shutter without getting grainy background noise. If you overdo it, the print is very light, but because it is not grainy you still can see your signal. Can these scanning back pull this off? Could you point me to a specific make?
As for the UV filter: that's the only bit of kit that I have rescued from the old camera :-)
As for the prints, the Polaroid gave 3x4 prints, but we are flexible on this point if the quality is ok.
Thanks a lot for your help again
best wishes
Andreas
molecularbiologist is offline   Reply With Quote
Old Oct 9, 2005, 4:08 PM   #6
Junior Member
 
Join Date: Oct 2005
Posts: 13
Default

I've just looked at a scanning back website. Wow, that looks like it would be as good as the Polaroid, but also just as expensive :-)
With the handling issue and the additional costs for prints there would be not too much incentive to move away from the instant film, although it would be great for a larger group where the amortisation would be better. Thanks for that suggestion!
Any other suggestion how I can get a "cheapskate" solution with acceptable handling would still be very much appreciated.
Best wishes
Andreas
molecularbiologist is offline   Reply With Quote
Old Oct 9, 2005, 4:25 PM   #7
Administrator
 
Join Date: Jun 2003
Location: Savannah, GA (USA)
Posts: 22,378
Default

The G6 has a lens that's twice as bright as most you'll findin this market niche (starts out at f/2.0, which is twice as bright as f/2.8 where most other models start out).

It's 7MP CCD isn't too bad from a noise perspective compared to many non-DSLR models (especially since you're stuffing 7 Million photosites into a relatively small sensor).

A DSLR would give you much better performance if you increase ISO Speeds or need longer exposures (since they have much larger sensors compared to the models you're looking at, with much larger photosites for each pixel).

But, you'd lose the live feed from the LCD with a DSLR (with rare exceptions), and you may need to stop down the aperture more with one for equivalent depth of field (if needed for the subject type/size and the percentage of the frame you need it to fill). On the other hand, most DSLR models are going to be much easier to use manual focus with, thanks to a true TTL viewfinder.

There are a number of factors that you need to consider for low light photography (noise levels, ability to handle longer exposures without losing too much detail from "dark frame subtraction" noise reduction used by most digicams at exposures over about 1 second), lens brightness, available ISO speed range if you need to go higher due to shutter time limitations, etc.

Can you take the time to try and answer the questions originally posed?

In order to give you better advise, we really need to know those types of things (subject size/percentage of the frame it needs to occupy, max print size you'll need, what exposure is needed, if an optical viewfinder is acceptable).

I see where it looks like your existing prints are small. So, that does help with available options (as smaller print/viewing sizes tend to mask image defects).

Can you post a sample from the G6 (so we can see what you consider to be acceptable quality, and look at the exposure settings the camera was using in the image's EXIF?

If you're using Photoshop to resize, don't use the save for web so we can see the exposure settings. Or, you can post a sample downsized with something like Irfanview (free from http://www.irfanview.com). You'll see a menu choice for doing this (Image, Resize/Resample).

It will need to be around 800x600 or smaller, at a file size of less than 250,000 bytes to post it in the forums (so you may need to change the default image quality when you use "save as" after resizing if the file ends up too large). Make sure to leave the retain EXIF checked (you'll see it when you save an image). It will already be that way by default.

If you can post a full size original somewhere on a web site, that would be even better. If you don't have any space to use from your ISP or elsewhere, you can open a trial account (no cost unless you decide to keep it) at somewhere like http://www.pbase.com (limited to 10mb with a trial account, but plenty for posting a sample). Then, justpost a link to the image (versus trying to include it in a post).




JimC is offline   Reply With Quote
Old Oct 9, 2005, 4:39 PM   #8
Administrator
 
Join Date: Jun 2003
Location: Savannah, GA (USA)
Posts: 22,378
Default

P.S.

If the amount of fluorescence changessignificantly with different subjects, it would be a good idea to know what exposure is going to be needed for a dimmer one (ISO speed, shutter speed, aperture), to make sure the camera you choose is within it's capabilities. Again, this information would be in the images themselves for any photos taken with a digital camera if you don't know the aperture and shutter speeds you're using with the ISO 3000Polaroidto get proper exposure.

If exposures too long are needed, images can get pretty ugly from some camera models (including ones using a Sony 2/3" 8MP CCD like that A2 you're looking at, especially at anything above the lowest ISO speed settings).

But, without a better understanding of what your needs are, it's hard to give you advise on what might be best.
JimC is offline   Reply With Quote
Old Oct 10, 2005, 12:15 AM   #9
Senior Member
 
Join Date: Jul 2003
Posts: 676
Default

The new Sony DSC-R1 may very well be the camera you need.
lucky2505 is offline   Reply With Quote
Old Oct 10, 2005, 7:23 AM   #10
Junior Member
 
Join Date: Oct 2005
Posts: 13
Default

Thank you very much for that detailed input, this is much more than I expected. I'll run you through the old setting and the shortcomings of the new ones in more detail then:
1. The specimen is a slab gel of 5mm thickness and between 5x10 cms up to 20x30 cms. The signal is a cloud of DNA migrating in that gel, stained with Ethidium Bromide so that it fluoresces under UV illumination (260-320 nm). So, our primary focus is that semi-transparent gel of 5mm depth, and the actual signal is a little band of about 3-6 mm width and 1mm height. The band does not have a sharp boundary, so we are focusing on a slightly fuzzy "cloud" in a gel.
2. The standard system was the Polaroid MP4 with bellows, a 1:4.5 35mm Tominon lens, and a large optical view finder. The film was the 3000 ISO Polaroid 667 b/w. For taking the fluorescence image, the aperture was typically set at 4.5, and shutter between 1/30 for very bright signals and between 1/15 -1/4 most of the cases. Weak signals could be teased out by shutter 1/2 or 1/1, which was not too uncommon. Only rarely was there a need to close the aperture because of too much signal. For the UV, we used a R(25A) filter.
3. The MP4/667 combination delivered a picture quality that was possibly too good for the purpose (overkill :-)), with very good sensitivity, very little noise even at high exposure, and a very large dynamic range, i.e. you could see weak bands even with strong bands on the same gel. This feature, and convenient handling, is what I am most interested in. We don't need to replicate the full quality of the 667.
4. The MP4/667 being too good encouraged people about 10 years ago to start using analog video CCD cameras instead. The resolution was about acceptable, but the dynamic range was really bad, so that you had to take separate pics for your strong and your weak signals, and still had less info than on the 667. I hate these setups, but you can't argue with the running costs. Convenience of handling was also bad, as you had to handle both camera and computer (remember the gloves etc.)
5. The video framegrabber systems as well as the more recent digicam solutions deal acceptably with gels that have strong and straightforward signals (which we would have shot with 1/8 - 1/32 on the MP4. They are no match if it comes to large dynamic ranges, and get noisy on high exposures for weak signals. I think that the G6 based solution also has an autofocus problem finding the blurry target. I am sure that setting the focus manually would improve the outcome, but I only had it on demo for an hour and could not try all settings.

I am attaching a representative pic taken with the G6, done with a focal length of 28.2, F/3, shutter 1s, pattern metering, exp compensation 0. As you might have guessed I am not very knowledgeable about digicams (no experience other than point&click with my Ixus), so just optimizing the settings might help.

We can use an optical viewfinder without a problem, if the viewfinder can be turned 90 degrees so that we can actually peek through it while the camera is on the stand :-), and setting focus, aperture and shutter manually on the camera and objective would not be a problem. As for the prints, I think that the print quality is always going to be better than the picture quality I have seen so far, so the printer is not a limiting factor and everything that gives a good representation of the signal intensity in B&W or colour is fine (we don't need colour anywhere...)
I take the point that the sensors of SLRs might be more suitable for our purpose, in that case the task would be to find the right objective and viewfinder.

Thanks again for your help and apologies for that lengthy post.
best wishes
Andreas
Attached Images
 
molecularbiologist is offline   Reply With Quote
 
Reply


Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off



All times are GMT -5. The time now is 11:42 PM.