Objective look... how many Megapixels are these pics? ** NOW UPDATED **

[dnuggett]

Off of a recent discussion I wanted to post two pics to see if anyone could tell how many MPs the camera is that took these shots.

Yes there is a way to cheat, but please if you know how don't post the answer and don't vote, I want to see if the average person has an eye for these things.

Pic 1

Pic 2


What do you think?

These are taken with the same camera and are the same megapixels. What I'm looking for is your guess on what MP size they are.





UPDATE:
Original is text above.

The purpose of this poll and discussion was to determine if the average person could accurately determine the number of MP a camera is rated when presented on a computer monitor with the exact untouched image that was shot from said digital camera. The image was uncropped, unedited, unressed, un-everything. It was pulled from the camera's memory and uploaded directly for download.

The camera that took these shots is a Sony DSC-P31. The MP rating is 1.9 MP. The picture you see is at full resolution, 1.9MP and untouched be me. The only thing I did was remove the EXIF data. This camera is commonly sold on Ebay for $30-$50.

Total polled: 137

26% were correct. 1-2 MP
73% were incorect. 2-8+ MP

57% were incorrect by more than a 33% margin.

19% were incorrect by more than a 100% margin.

2 people were off by more than 400%. Although they would be considered extreme outliers and thrown out for statistical purposes.

And before some of you say the test was bunk because all of the other factors that come into play are unkown re-read the point of the test. Simply by looking at an unedited picture on a computer monitor can the average person tell the MP rating of the camera that took the shot?

The answer to that question seems to be no. At least not with any confidence. Altough I was very impressed with some who guessed not only the MP size but the camera's manufacturer as well. Good eye.

Some interesting quotes:

At least 3MP if not 4MP. My 2MP can't do that.

I say 3.2... I can do something about that with my Canon A75 3.2MP

3-4

my guess is its a P&S camera and not a DSLR. i also am guessing 2mp because thats seems to be somewhat close to the crappy 2mp Kodak I have.

5.1mp

im guessing 5-6 just for the simple fact if your trying to prove mp isnt everything the best way would be to post crap pictures from low-quality high mp cameras.

Going by simple pixel count it's a 2MP image. But I don't think that the resolution of a 2MP camera would produce these images.

It looks like either digital zoom was used or there was no resizing. It looks like a normal 100% crop from a consumer level P&S. So I'd say 2MP.

I think it's probably between 3 and 4 MP from a point and shoot digital camera, because:

1) The type of noise does not look like something that one would see from a DSLR under these conditions
2) The noise, however, does not look like what I'm used to seeing at 1:1 in a P&S image, hence I think that it was resized down slightly
3) It was not resized down a lot (from a high-MP camera), I suspect, because if it were, even a P&S would appear to have decent noise properties at this resolution under these conditions
4) The areas without detail seem to have a "painterly" appearance to them, which could come from a Kodak 14n or a P&S (I wager the latter, obviously)
5) Finally, I think the images were resized down because I don't see (m)any JPEG artifacts that I would expect if the image were straight from the type of camera that I suspect was used.

I'm going to say that it's a 2 megapixel camera. And the color doesn't look like Canon, so I'm going to guess and say Sony. Want me to get even more specific and make a wilder guess? DSC-F505.

Nice call, very close.

It also looks like the images are a bit blurred, as though they were cropped from a larger image. There's not a lot of detail in them. It could be straight out of a 2MP Camera I suppose, or just that the image is beyond the resolution capabilities of the lens.

I can tell you that your lens isn't as sharp as it could be and that the image sensor is clipping highlights a bit though. My guess is that they were shot on either a pretty good point and shoot or a DSLR with a low-end lens (like a Sigma lens or something slimilar).

My guess is somewhere close to 3 erring more toward 4 than 2.

I'm guessing enlarged by 200%, and a 4-5MP camera.

 

[michaels]

I can't tell a difference.

 

[dnuggett]

Originally posted by: michaels
I can't tell a difference.

Same camera, no difference. Just two different shots.

 

[pinion9]

Wouldn't it be more effective if the pics were of the same place?

Also, there is no way to tell the difference if both pictures have higher pixel count than your monitor. My current resolution is 1280*1024 ~= 1.3 MP. Anything over 1.3 MP will be a waste (picture will just appear large and you will need to scroll.)

Now, print me out an 11x14 on high quality photo paper taken with a 3 MP camera and an 8 MP camera and I'll tell you which is which.

 

[pinion9]

Oh, and since we are looking and giving our opinion, this would be a subjective look.

 

[DBL]

are these from a digital camera? looks like film to me.

 

[DBL]

Originally posted by: pinion9
Wouldn't it be more effective if the pics were of the same place?

Also, there is no way to tell the difference if both pictures have higher pixel count than your monitor. My current resolution is 1280*1024 ~= 1.3 MP. Anything over 1.3 MP will be a waste (picture will just appear large and you will need to scroll.)

Now, print me out an 11x14 on high quality photo paper taken with a 3 MP camera and an 8 MP camera and I'll tell you which is which.

Not true. A 1280 x1024 LCD has a true 1.3 MP while a 1.3MP camera only gets that resolution using Bayer interpolation. In reality, your monitor is capable of displaying the picture of a much higher resolution camera. How high depends on the picture and how well the subject is represented by the sensor technology.

 

[dnuggett]

Originally posted by: pinion9
Wouldn't it be more effective if the pics were of the same place?

Also, there is no way to tell the difference if both pictures have higher pixel count than your monitor. My current resolution is 1280*1024 ~= 1.3 MP. Anything over 1.3 MP will be a waste (picture will just appear large and you will need to scroll.)

Now, print me out an 11x14 on high quality photo paper taken with a 3 MP camera and an 8 MP camera and I'll tell you which is which.

Why would I take two pictures of the same place with the same camera?

 

[dnuggett]

Originally posted by: DBL

Originally posted by: pinion9
Wouldn't it be more effective if the pics were of the same place?

Also, there is no way to tell the difference if both pictures have higher pixel count than your monitor. My current resolution is 1280*1024 ~= 1.3 MP. Anything over 1.3 MP will be a waste (picture will just appear large and you will need to scroll.)

Now, print me out an 11x14 on high quality photo paper taken with a 3 MP camera and an 8 MP camera and I'll tell you which is which.

Not true. A 1280 x1024 LCD has a true 1.3 MP while a 1.3MP camera only gets that resolution using Bayer interpolation. In reality, your monitor is capable of displaying the picture of a much higher resolution camera. How high depends on the picture and how well the subject is represented by the sensor technology.

Thank you. Now.. moving on to taking a guess.

 

[dnuggett]

Originally posted by: pinion9
Oh, and since we are looking and giving our opinion, this would be a subjective look.

No, I want you took look at it objectively.

 

[Glavinsolo]

1200x1600 is 2mp

 

[Baked]

At least 3MP if not 4MP. My 2MP can't do that.

 

[dnuggett]

Originally posted by: Glavinsolo
1200x1600 is 2mp

You sure it wasn't resized?

 

[DBL]

Originally posted by: dnuggett

Originally posted by: DBL

Originally posted by: pinion9
Wouldn't it be more effective if the pics were of the same place?

Also, there is no way to tell the difference if both pictures have higher pixel count than your monitor. My current resolution is 1280*1024 ~= 1.3 MP. Anything over 1.3 MP will be a waste (picture will just appear large and you will need to scroll.)

Now, print me out an 11x14 on high quality photo paper taken with a 3 MP camera and an 8 MP camera and I'll tell you which is which.

Not true. A 1280 x1024 LCD has a true 1.3 MP while a 1.3MP camera only gets that resolution using Bayer interpolation. In reality, your monitor is capable of displaying the picture of a much higher resolution camera. How high depends on the picture and how well the subject is represented by the sensor technology.

Thank you. Now.. moving on to taking a guess.

What do you want?

It looks like a 2mp picture. If one were to use a current 8mp DSLR to snap that picture and downsize it, it would look a hell of a lot better than your picture. There is pretty obvious grain throughout the picture, although I still say it looks like a crappy film scan.

 

[nakedfrog]

Dunno if it's really possible to guess, since part of the image quality is now dependent on whatever software you used to resize/convert the image.

 

[Glavinsolo]

Originally posted by: dnuggett

Originally posted by: Glavinsolo
1200x1600 is 2mp

You sure it wasn't resized?

Yes look at the top of the building in the 2nd picture, you can see grainy areas.

 

[dnuggett]

Originally posted by: nakedfrog
Dunno if it's really possible to guess, since part of the image quality is now dependent on whatever software you used to resize/convert the image.

If anything there would be no improvment to the image quality, just extra pixels. Assuming I did resize it. The guess would still be valid.

 

[rezinn]

I guessed 1-2 based on the quality. Seems I was right.

 

[dnuggett]

Originally posted by: DBL

Originally posted by: dnuggett

Originally posted by: DBL

Originally posted by: pinion9
Wouldn't it be more effective if the pics were of the same place?

Also, there is no way to tell the difference if both pictures have higher pixel count than your monitor. My current resolution is 1280*1024 ~= 1.3 MP. Anything over 1.3 MP will be a waste (picture will just appear large and you will need to scroll.)

Now, print me out an 11x14 on high quality photo paper taken with a 3 MP camera and an 8 MP camera and I'll tell you which is which.

Not true. A 1280 x1024 LCD has a true 1.3 MP while a 1.3MP camera only gets that resolution using Bayer interpolation. In reality, your monitor is capable of displaying the picture of a much higher resolution camera. How high depends on the picture and how well the subject is represented by the sensor technology.

Thank you. Now.. moving on to taking a guess.

What do you want?

It looks like a 2mp picture. If one were to use a current 8mp DSLR to snap that picture and downsize it, it would look a hell of a lot better than your picture. There is pretty obvious grain throughout the picture, although I still say it looks like a crappy film scan.

I'm not bagging on you, I was thanking you for explaining to the poster you quoted. So you are saying 2MP eh?

 

[dnuggett]

Originally posted by: rezinn
I guessed 1-2 based on the quality. Seems I was right.

Why do you think you were right, based on the outcome of the poll? I'm getting closer and closer to proving my theory.

 

[alien42]

this thread is bunk for 2 reasons:

1. the pictures are too different in both subject and lighting
2. the only difference MPs make is when blowing the pictures up. if these pics were printed out poster size then there would be a clear difference

 

[Glavinsolo]

Originally posted by: dnuggett

Originally posted by: rezinn
I guessed 1-2 based on the quality. Seems I was right.

Why do you think you were right, based on the outcome of the poll? I'm getting closer and closer to proving my theory.

Your theory as I conclude is this: Take a 2mp picture and see if people can tell the difference without given a higher quality picture taken with a current 8mp. Now after the study group is done tell them it was a 2mp picture and have them go ohh and ahhh and cry out, "Why oh why are the Digicam people telling me to get an 8mp camera when all I need is a 2mp!"

I sold digicams at circuit city, I was hated by the managers because I told the customer "Why are you buying that 5mp camera? Are you going to be printing posters? And they just looked at me.

I told them to put money where the lens came from

 

[Howard]

I'm not sure it's possible to judge accurately.

 

[nakedfrog]

Originally posted by: dnuggett

Originally posted by: nakedfrog
Dunno if it's really possible to guess, since part of the image quality is now dependent on whatever software you used to resize/convert the image.

If anything there would be no improvment to the image quality, just extra pixels. Assuming I did resize it. The guess would still be valid.

Exactly. I wouldn't expect there to be improvement, only varying levels of degradation from resizing, and the quality of the JPG encoder and it's settings. Seems more of a guessing game than anything else to me.

 

[PurdueRy]

2-3