D7 vs. CP990

Introduction:

Many people are looking at the Dimage 7 as an upgrade from the Nikon Coolpix cameras. As a long time Coolpix owner (I've had the 900, 950 and 990), I'll compare the two from the perspecive of a CP990 owner. Warning, I find the Dimage better or at least the equal of the CP990 in most areas. No, I'm not getting rid of my 990 though, I plan to use it as a "take anywhere" back up camera.

Last update, December 10, 2001 Revision #5.

Look and Feel

Well, maybe it isn't fair to compare these camera at all, after all, "look and feel" is purely subjective, one person's art is another's trash and all that, and the "look and feel" does not really matter when you get down to what you can do with the cameras, so keep these things in mind as you read this.

My overall feelings about the two cameras are...

990 - smaller, portable, more rugged, more of a point and shoot type camera. (I know that it can do much more, it just feels that way).  More comfortable to use, carry and grip. I actually like the fact that it does not look like a camera, and that people can be shot unawares with the swivel.

D7 - More of an SLR type camera. Larger and harder to carry. The weight is actually similar to the 990, but the shape isn't as convienent. A "real" camera though, that gives me the "shooting with a real camera" feeling.

The highlights of each camera are...

Nikon

- The swivel. I like it, and it lets me take shots (such as looking straight down at the camera) that even the tilting EVF can't do.  The swivel allows the camera to fold flat to fit into a smaller bag.

- The optical viewfinder. Unless you have an add on lens, using the optical viewfinder is great for moving subjects.

- The 3:2 mode. I just like it. I miss it on the Minolta. Sure, you can crop later, but it is not as nice as seeing the framing while you shoot.

Minolta

- The zoom. After using a "real" zoom ring, it is hard to go back to the limited noisy slow buttons on the 990. No more screwing on add on lenses either.

- The manual focus ring. You just can't compare this to the manual focus on the 990, using the ring with the EVF is faster and more accurate than doing the button and wheel on the LCD on the 990.  The 990 manual focus requires that you use the LCD, which has less resolution than the EVF on the Minolta. People might think that the Minolta EVF is hard to focus, try the 990 LCD, even with the focus assist sharpening feature.

- The flash. The flash on the Minolt pops up, and while I've had some Red eye, it is not as bad as the horrible Red eye that the 990 can produce at the right range. The preflash ADI exposures are always very good, and the range of the flash seems better than the 990 flash.

- The slightly faster lens at zoom range. The Minolta is F3.5 at 110mm zoom, where the Nikon is F4. It stays F3.5 all the way out to 200mm. I now that this does not seem like much, but it makes a difference.

- The larger shoot-ahead buffer. The Minolta can buffer 5 Fine mode shots, in spite of their much larger size, while the Nikon can buffer 3 Fine mode shots. This makes quite a difference when shooting for me, in that I can almost allways just shoot and not worry about waiting for my card. It also allows me to use slower cheaper cards in the Minolta (such as the Mr. Flash compact flash card) because the slow card write speed does not slow me down.

-Illuminated top LCD. Very nice for night work.

- The exposure preview that the EVF shows at all times.

Autofocus accuracy and speed.

Much has been said already about the autofocus and autofocus speed of the D7. It is average by digital camera standards, slow by 35mm autofocus standards. Here we are just comparing to the CP990, this isn't a commentary on how fast you will find the focus if you aren't familiar with the 990.

Compared to the CP990, the autofocus is almost identical in every respect. In a test where I held one camera in each hand, with the Minolta set to beep when it reached focus, and the Nikon set to turn one of the focus bracket Red, I swung the cameras around at various subjects around my house. I gave it a really good workout of hundreds of trials to get a good feeling for what was happening under a variety of lighting conditions. The results were...

* In good light, the two cameras focused in, as closely as I can tell, exactly the same time. The Beep from the Minolta would arrive just as the focus bracket on the 990 LCD turned Red. If one was faster than the other I could not tell.  In no case out of maybe 30-40 trials did the Minolta have trouble focusing or take longer than the CP990.

* In Poor light, and by that I mean light were the cameras could still focus but were starting to have trouble and taking longer by making more focus passes, the Minolta was faster. Sometimes the 990 would have to make one or more additional passes to achieve focus.

* In Bad light, the two cameras differed. Both cameras would take about 3-4 seconds to focus, but the Minolta was able to focus eventually in almost all cases, even when the 990 could not. The resulting exposures were on the order of 3-4 seconds at ISO 100, so you know that it was dark. In some cases the 990 indicated focus, but was not in focus. Sometimes the 990 could just not focus and gave the flashing Green light out of focus indication. The Minolta could not focus some times too, but it gave the Red can't focus indication in those cases, it never mis-focused. As an incredible fact, I was able to get the Minolta to even focus on the brighter stars in the sky at night!

In all, the Minolta is very comparable to the 990 in autofocus accuracy and speed in reasonable light. In low light the D7 is better. If you can stand to use the 990 as far as speed goes, you will be fine with the Minolta. The 990 is only about average as far as digital camera autofocus speeds go, so if you are coming from some other background you may not be happy with the Minolta.

Shutter lag.

By this I mean the time between pressing the button and taking the picture after the camera is already focused, either by half press or by using manual focus. This time is important for sports action shots. Both cameras are quite fast, I would guess on the order of a tenth of a second or less. I do not have any way of timing these short times accurately, but I will state that I feel that the CP990 has a faster shutter here (and we are talking less than a tenth of a second). This is based on my taking some baseball pictures with both cameras. On the 990, if I press at the batter's swing and it is a strike, which is usually is if it is one of my friends, I often captured the ball in the air between the batter and the catcher. With the D7, the ball was often in the catcher's mitt already. Now that isn't a very long time for the ball to travel that short distance, and I might just be "used" to the CP990 and know when to press and I could get "used" to the D7, but there does seem to be a small practical difference. And yes, I was using the LCD on the 990 and the EVF on the D7, so it could be due to a difference between display speeds.

Use in sunlight.

I'm including this topic because use in bright light has always been a problem for the CP990 when you have an add on lens on. A lot of time has been spent making hoods and viewers for the 990, I made a couple myself. That is because all of the add on lenses block the viewfinder to some degree, and the viewfinder would not be usable for framing anway. The Minolta beats the 990 in two ways in sunlight, first, the rear LCD is more visible, and is recessed a bit which helps at some angles, and second, the EVF is perfectly usable in even the brightest direct sunlight even at reduced brightness settings. No rubber hoods needed for the Minolta.

Add on lenses, telescopes, and filters.

Both cameras have advantages, however the CP990 is the king of add on lenses and use with telescopes. That is because the smaller lens of the CP990 is more suited to the 1/4" exit pupil that is common on optical devices intended for humans such as telescopes and binoculars. The D7 with it's large lens simply can't use them, only a very small highly vignetted circle is visible at the center of the field. This has opened up a whole world of add on lenses and accessories for the CP990 that will not be available to the D7. Yes, the CP990 has some add on disadvantages like the weird 28mm thread size, and the fact that any step up adapter, even 28mm to 37mm blocks the viewfinder and the flash sensor.

The D7 on the other hand, has a standard filter mount of 49mm. Unfortunately, it vignettes badly at wide angle if you attach almost anything to it, so most 49mm filters can't be used. I'm using a 49mm to 62mm step up ring; that works fine.

Lens quality.

The Minolta lens is excellent.

* It has greater zoom range, plus you can zoom fast with the ring, plus it does not drain your batteries to zoom.

* It has less distortion at 28mm than the 990 does at 35mm, and far less than the 990 does with a Nikon wide angle lens on zoomed to approximately 28mm.

* At higher zoom levels it remains F3.5 (the 995 is up to F5.1 at the end of it's range)

* It has a manual zoom ring, and manual focus ring.

* The lens has almost no chromatic aberration. No more purple halo around edges, even at wide angle.

The 990 lens has a few minor advantages

* The 990 lens is F2.6 instead of F2.8 at wide angle but this hardly worth noting.

* The 990 lens is smaller and a better match to telescopes.

* The 990 lens is internal, so the camera can be solid mounted to a telescope or microscope, it does not move as you zoom.

* It says Nikkor on it, I like that. :)

Overall, the Minolta lens is the clear winner. I don't have to carry my TC-E2 or WC-E24 with me every where I go, and I don't have to mess around screwing the lenses on and off.

 

Image quality:

Everyone's first question is, "what about the images", but I've saved this topic for near the last because it is the most complex and because I still am thinking about it. The D7 Produces superior images to the 990 in many cases, depending on which aspects of image quality are important to you.  However, there is one issue that is making it difficult to compare, and I need to address this first. You can see the problem in the photos below.

D7

CP990

I've adjusted these two exposures by using EV compensation to have the same visual appearance while I was testing high ISO noise. Both cameras are set to ISO 400. The problem is the exposures used. You would expect that with both cameas set for the same ISO that while they would not make the same choice for exposures, that they would use similar EV values for their exposures. However the 990 used 1/37 at F3.1 = EV8.5 and the D7 used 1/64 at F3.4 = EV9.5. This is a difference of 1 EV or 1 stop, or a factor of 2 in light

In fact, the D7 always seems to act like the true ISO is higher than what is set.

Some more examples, here are how the two cameras metered carefully controlled scenes

D7                               990                             EV difference

1/10 F3.5 = EV6.9      1/5 F4.0 = EV 6.3                  0.6

1/90 F8  = EV 12.4     1/136 F5.8 = EV 12.2             0.2

Not only did the D7 always meter higher EV, but the resulting D7 shots are always brighter than the corresponding CP990 shots. If the D7 and the 990 are placed in manual mode, and shots are made until the highlights and shadows look the same, the D7 always meters about 1EV above the 990 at a given ISO.

This makes comparing shots very complicated. First, is a difference seen such as noise due to a difference in the camera sensitivities, or is it something to do with the transfer curves used by both cameras (for example the D7 preserving shadows better), or is it just that the D7 has less dynamic range so that it's shots have to be underexposed to maintain the highlights, or is it just that the camera designers thought that a brighter picture looked better and made some kind of in camera adjustment? I don't really know, and it complicates comparisons horribly. If the D7 really has increased sensitivity, then it isn't fair to compare an "ISO 100" D7 shot to an ISO 100 Nikon shot as far as noise goes, for example, especially if the ISO difference is really a factor of 2. If the D7 is just making different exposure decisions combined with some kind of different transfer curve, then comparing shots right out of the camera without adjusting the exposure on one or the other for similar appearance isn't fair either.

That being said, I'm just going to have to do the best I can. The D7 feels quite "fast" to me, and my feeling about it is that it is a bit more sensitive than the published number, combined with an exposure preference that produces brighter pictures, sometimes at the expense of highlights. My inclination is to take the emperical approach and to try to compare similar looking shots.

 

Introduction.

Now for some comparison. Yes, I am aware that there are software solutions for some of the things discribed here, filters to reduce chromatic aberration and noise, and that you can overlay several shots to reduce noise. I'm just comparing the out of the camera shots. That might not be fair to the D7 which really requires another step to convert the color from the larger color space.

Resolution.

Starting with the clearest topic, the D7 out resolves the 990 in every case. However, let me say that you have to look closely to see the D7 difference, and that your output device monitor or 8x10 print may not show you the difference unless you magnify the pictures. If you usually look at your shots full size on a 19" monitor, and mostly print 4x6 or 5x7 shots, you may not notice the difference.

The increased resolution is more apparent on shots of man made objects with hard edges, and less apparent on say landscapes. Yeah, the extra 2 Million pixels are there, but your eye precieves linear differences as higher resolution. As such on an object that the 990 can place 10 pixels across, the D7 will place 12. This difference isn't always apparent to casual observation, even on 8x10 prints.

Chromatic aberration.

The D7 lens has much better color correction than the CP990 lens does. Virtually gone are the purple halos around bright objects near the edge of your pictrue. Even at wide angle, the D7 lens is superior. Take a look at the enlarged crops below, the difference is even more apparent on the originals.

CP990

D7

Noise.

This is a more complex issue. Overall the D7 produces shots that are slightly more noisy in some situations. However this issue is rendered almost impossibly complex by the following considerations.

* The D7 tends to expose images brighter, and so reduces shadow areas where noise occurs.

* The D7 ISO rating seems to be conservative, so you are comparing a camera at ISO 100 against one that acts sometimes like it is ISO 200.

* And the most important issue, the D7 default sharpening has a lot to do with the amount and noise seen. The default sharpening level for the D7 is agressive, and some would say that it is too agressive.

 

To illustrate this last point, here are crops from a larger photo, enlarged to a common scale about 300%.

The first shot to the left is the CP990 shot, taken on the Soft sharpening setting. By the way, the 990 gets this Blue color wrong.

Now a crop from the D7 shot, also enlarged and taken on Soft. There IS more noise than the CP990 produced, but is it that bad?

Normal sharpening on the D7. Note the additional noise and the start of a dark halo around the hood of the car due to sharpening.

Lastly the hard setting. You can see a sharpening halo around almost every contrasting area, and there is a lot more noise.

The shots clearly show several things, and I also encourage you to look for yourself at the originals which are in this D7 sharpening series album if you have a high speed connection, the shots lose detail in the enlargement and re-JPEG process for the web. 

First, you can plainly see that the sharpening on the D7 accentuates existing noise. The D7 also has an unfortunate tendency to sharpen individual noise pixels and make them appear to be black noise holes in your shot, which are very apparent.

Second, you can also see that the D7 soft is less soft than the Nikon soft setting, although some of the difference seen here is due to the resolution difference also, the Nikon shot had to be enlarged more to match the size; again check out the originals.

Third, you can see that even the Normal sharpening on the D7 is pretty harsh, it brings out visible noise and it introduces halo artifacts.

Because of all of this, the noise question is hard to answer. If you shoot on Soft on the D7, the shots are still sharper than shots from the 990, and there is comparable or even less noise. If you shoot on hard with the D7, there is going to be a lot of visible noise. At the normal setting there is still noise, but I would not call it objectionable. Yes, the 990 does have less noise on the soft setting than the D7 does on any setting, but I can't imagine actually using the soft setting without some post processing on the 990, it is just too soft.

Also, the 990 seems to have more of what I call "oil slick" type noise. These are larger areas of color, usually Red Green or Blue, that seem to spread over a larger area like an oil slick. I don't see that type of noise in the D7, the D7 noise is more pixle to pixel noise.

One last thing to consider is that the noise that is visible on the D7 is less visible on your typical viewing device due to the large number of pixels in the image.

Color.

This is a fairly easy comparison. Using the default settings for both cameras, the 990 has better looking  but not necessarily more accurate color. The D7 has the correct color on the little Blue car above, for example. The 990 colors are just more saturated and natrual looking, while the D7 shots appear flat. The D7 color can be improved considerably out of the camera by setting -2 contrast, -0.3 EV and +1 or +2 satruation,. Even at saturation +2 on the D7, the 990 shots still appear more saturated. Since the 990 is generally considered to have good color, the D7 shots do indeed have low color. Saturation examples.

However, if you are willing to convert your D7 shots from their wider color space into a color space for your monitor or printer, then the corrected D7 pictures are superior, with very life like skin tones and absolutely spot on green leaves and grass. The drawback is that you have to run a utility on every image to convert them to the better colors. The images straight out of the D7 aren't really bad, and it might be possible to get used to them with + saturation, but the 990 produces some very nice colors. For the D7, I'm using a custom color profile generated by Profile Prism. If you aren't willing to convert the wider color space, then the 990 is the camera for you. The popular "ColorFix" program is also a nice alternative.

Usage comparison.

I don't know if it is even fair to be comparing these cameras, one costs twice as much as the other (not any more! Dec '01). In most real world shooting situations, the D7 beats the CP990 and I guess that it should for the price. With the D7 there is no messing around screwing lenses on and off, no problem seeing an LCD at full telephoto in sunlight. I don't know how many shots I have not taken with the 990 simply because I did not have my TC-E2 with me, or was too lazy to attach it (the real story).

In addition, situations that used to give the CP990 trouble, like stage shots of my kids in low light, or sports shots at twilight, or flash shots from 40 feet (I use ISO 800 on the D7), or indoor shots without a flash are all just a little bit easier for the D7. I don't know if it is the underrated D7 ISO, the slightly bigger lens, or the fact that I consider ISO 400 and even 800 usable on the D7, but I can always pull the shot out with the D7 in situations where I would struggle with the 990.

The D7 feels just as fast at focusing, locks on just as well, is faster to zoom, has more zoom range, and just all around beats the pants off the CP990. If is just as light, and just as easy to hold but does not fit in your pocket. The EVF gives a better preview of your shot than the LCD does on the 990.  The D7 has a faster shutter, higher ISO and bigger lens that helps you get the shot.

The 990 does have an edge in battery life, portability and maybe in action shots at 110mm or less because of the optical viewfinder. The 990 feels more rugged, but I would not want to drop either camera or get either one wet. The projecting lens on the D7 is probably more likely to take a hit if you are wearing the camera. The D7 can't reproduce the flexibility of the 990 swivel, those one hand overhead shots and canded waist level shots are out with the D7, I do miss that a bit. The menu based controls on the 990 are faster to work, but the D7 has dedicated buttons for many choices that are just as fast or faster.

The D7 is a "dream" CP990.

A lot of effort has been expended on creating accessories for the CP990 to extend it's abilities. The D7 has most of these things built in. Forget the extra resolution of the D7, lets just look at what you would have to get for your CP990 to get as close as you can to the D7. I actually bought most of these items for my 990, so don't laugh! I'm using the "discount" prices for these figures.

* A wide angle lens adapter such as the WC-E63. This will run you about $130, but will give you about 22mm of wide angle vs. the 28mm on the D7.

* A teleconverter to match the 200mm focal length of the D7. The TC-E2 comes close, giving you 220mm. Add about $130 for this lens.

* An external flash shoe. You will need a bracket and the SK-E900 cable. Add at least $60 for these items.

* An off camera flash to elminate Red eye. The D7 has a pop up flash. The 990 is going to need an external flash. Add at least another $50.

* A standard lens thread. Add $20 for a 37mm step up ring.

* Some way to see the LCD in Sunlight. Add about $50 for the lens hood of your choice for the 990.

* A cable release and interval shooting. The D7 has an external socket built in which uses simple contacts for you homebrewers, and the D7 already has interval shooting built in. The release for the D7 is about $40. The Nikon remote release is about $130 and gives you interval shooting, but which takes a battery.

 

So, what do you get in the end? You have spent $530 on the 990 accessories, going the very cheapest  route, bringing the price of the 990 up much higher than the December '01 pricing of the D7 which is less than $800 at this time. You have a collection of stuff that weighs more than the D7 does, takes a bigger bag, and is too much junk to carry with you all the time. You have lenses that you have to screw on and off to take a wide angle or telephoto shot, and which block the viewfinder.  You can't even use a standard filter without blocking the viewfinder and the flash sensor. You have a lens hood that you will have to use to use any of this stuff in the sunlight but that can't really be left attached if you want to fit the camera back in the bag. You have a flash shoe and fiddly cable that you can't leave attached.

And you still can't use CF II cards, or adjust the color saturation, or use a manual focus ring or zoom without pressing a button or have a chromatic abberation free lens or...

In all the D7 is a 990 owner's dream. It has every single thing that anyone ever wanted for the 990. Some people would have paid $500 for any one of the D7's features when the 990 came out.