Haha. True! If you can isolate equipment from the client, and product the results as desired, that is certainly ideal!
When a customer asks what equipment I use, I'll answer and try not to pick up the job.
For those who like to argue ....
- Thread starter luckyorange
- Start date
When a customer asks what equipment I use, I'll answer and try not to pick up the job.
?? actually I can tell you are not an engineer
ok let me explain more simply. The object of discussion here is FX vs DX, aka sensor size. To do any comparisons, we should choose two cameras as identical as possible with the exception of sensor size. This is because we can safely conclude any observable difference in performance in due to FX vs DX, sensor size.
If you choose two cameras that are very different, then what is the point? Example, D3s vs D90. All you have proven is that the sensor technology in D3s is better than D90. Does that help your arguement in proving that FX is better than DX?
At no point, is TS asking about just the size of the sensor. When considering FX vs DX, the sensor technology is never the same except for the case of D800 vs D7000. Pixel pitch and difference sensor tech is inherent in all FX vs DX comparisons. You can never get away from that.
Even when we compare the D800 vs the D7000, your one stop advantage with the D800 only holds true if at 100%. The sheer resolution of the D800 image, when downsized to a D7000 image size will increase that advantage even more.
You cannot just say that we have to compare FX and DX sensors of the same pixel pitch. One of the major reasons why FX sensors have so much advantage over DX sensors is in the difference in pixel pitch. On the other hand, a camera like D800 with its obscene amount of resolution is designed to overcome the noise issue by providing so much resolution, so that in viewing the entire image, the noise is a lot smaller in relation to the entire image. So your point about comparing D800 vs D7000 also does not hold true, because on the image level, in real world usage, it is a lot more than the 1 stop. Remember, DXOmark completely ignores any advantage in resolution (DXO strongly states this in their explanations).
So looking at the argument from both sides, it is always more than a 1-stop difference in real world usage.
So what is TS really asking about here... if he would benefit moving to FX from DX. If he moves from DX to FX what are his choices? for current lineups, only D600 at 24MP, D800 at 36MP, which both will yield more than 1 stop advantage when downsized to the resolution of D7000 (best of DX currently). Or D4, which you say is a different technology... So in the end, according to your line of reasoning, there is actually no real comparison then.
Last edited:
Ok there is nothing to discuss about this. There is no right and wrong. If you want to do comparison this way, then I can't do anything, except stay out of it. Because it is impossible to have meaningful discussion as you can change 1001 one other variables. You can compared D7000 to D800, D7000 to D3s, D90 to D4, D5200 to D600, etc, etc.
Oh this I can say up front you are flat wrong. The one stop adv of D800 vs D7000 only appears when you downsize the D800 image. In DXOMark, the 100% comparison is called "screen". Click on screen at the top left corner of the charts. You will see that D800 is IDENTICAL to D7000 at 100%. Print means downsizing the pics to allow fairer comparison between cameras of different pixel count.
I am not sure where you read this but as I said DXOMark ONLY ignore adv in resolution if you are at "screen" comparison. Just click "print" and you should see the curves improve for higher pixel count cameras
Last edited:
errr.... just based on the fact that an engineer will not do objective comparisons by changing more than 1 variables at a time? you will basically get nowhere in debugging technical problems and finding root cause if you change everything at one go..
as the posts are deviating further and further away from the original TS' question... i decided to move the posts to a separate thread for you all to continue to argue.
and as a friendly reminder to dniwkh/alantkh - you have been banned before for needlessly picking fights/starting needless arguments with other CSers before. the admins allow you to re-register, assuming that you will behave. if you feel that you need to defend every single position (regardless of whether you are right or wrong), especially when you think the other parties are ignorant buffoons in your own expert opinion, then mebbe other more technical-minded forums might be a better place for your preferences.
and as a friendly reminder to dniwkh/alantkh - you have been banned before for needlessly picking fights/starting needless arguments with other CSers before. the admins allow you to re-register, assuming that you will behave. if you feel that you need to defend every single position (regardless of whether you are right or wrong), especially when you think the other parties are ignorant buffoons in your own expert opinion, then mebbe other more technical-minded forums might be a better place for your preferences.
I see your point but I want to make clear something, I DEFINITELY do not consider others as ignorant buffoons and I am not sure where you get that from.
Is it the engineer thing? There is nothing rude about not thinking like an engineer. Other professions probably needs to deal with more ambiguity and compare complex real life scenarios where many facets are changing. It is probably harder to do that. I don't think Blur shadow was offended...
I just can't "stand it" when I feel something was said is wrong. However, I just "correct" the facts, I don't attack the person. Even when I am "arguing" with Daredevil, I was polite. I never directed any comments against him as a person. I hope he was not insulted or anything.
Last edited:
well... It is pointless to argue any further. dniwkh and me stand on different opinions. And I said opinions, because I know for a fact what he says is not a clear right or wrong. Rather it is an opinion.
To dniwkh,
DXOmark SNR18% is one of the matrix. It is not the ultimate measure of ISO low-light performance. Why? I quote from DXOmark, since you likes DXOmark.
http://www.dxomark.com/index.php/About/Sensor-scores
DXOmark lists the Sports (Low0light ISO scores) for the following cameras as:
D7000 - 1167 ISO
D800 - 2853 ISO
D3s - 3253 ISO
Remember an increase of 25% = 1/3 EV better.
Anyone can do the math.
So which data is accurate? DXOmark will always refer to the ISO score when talking about low-light capabilities. But you use the graph of SNR18%. So what gives?
All I know is when I look at the pictures, I see at least a 2 stop difference. I have showed you comparisons from dpreview, as well as respected professionals. Even Thom Hogan says D3s and D4 is at least a 2 stop improvement over D7000's ISO. And you insist it is 1.5 stops only. And back to D800, the same respected professionals are also saying that when images are downsized to same sizes, the D800 comes very close to D3s and D4 in ISO performance. And you say it is strictly a one stop advantage. So what gives?
As an engineer myself, I know a lot of things out there depends on data. And a lot of data can be massaged presented to look in a certain way. Testing methods, calculation methods from both the data collection process as well as the data presentation process all come into question. There is no plain or flat wrong in DXOmark's scoring system or charts. This very reason is also the reason why everyone is always debating like we are now, on what is right or wrong, on what is accurate or not. This is also the same reason why so many industry folks are questioning DXO's way of scoring or tests. You can choose to believe that the SNR18% chart is the whole truth, that is your prerogative. I can choose to reject it like any scientist or engineer. I can even counter you with the Sports scoring and the calculation that DXOmark put out (see above) and will give a result very different from your conclusion.
In the end, I choose to depend to my own eyes and experience. You can try to convince me otherwise till cows come home. I will still not buy into what you are saying. So lets just end here.
To dniwkh,
DXOmark SNR18% is one of the matrix. It is not the ultimate measure of ISO low-light performance. Why? I quote from DXOmark, since you likes DXOmark.
http://www.dxomark.com/index.php/About/Sensor-scores
DXOmark lists the Sports (Low0light ISO scores) for the following cameras as:
D7000 - 1167 ISO
D800 - 2853 ISO
D3s - 3253 ISO
Remember an increase of 25% = 1/3 EV better.
Anyone can do the math.
So which data is accurate? DXOmark will always refer to the ISO score when talking about low-light capabilities. But you use the graph of SNR18%. So what gives?
All I know is when I look at the pictures, I see at least a 2 stop difference. I have showed you comparisons from dpreview, as well as respected professionals. Even Thom Hogan says D3s and D4 is at least a 2 stop improvement over D7000's ISO. And you insist it is 1.5 stops only. And back to D800, the same respected professionals are also saying that when images are downsized to same sizes, the D800 comes very close to D3s and D4 in ISO performance. And you say it is strictly a one stop advantage. So what gives?
As an engineer myself, I know a lot of things out there depends on data. And a lot of data can be massaged presented to look in a certain way. Testing methods, calculation methods from both the data collection process as well as the data presentation process all come into question. There is no plain or flat wrong in DXOmark's scoring system or charts. This very reason is also the reason why everyone is always debating like we are now, on what is right or wrong, on what is accurate or not. This is also the same reason why so many industry folks are questioning DXO's way of scoring or tests. You can choose to believe that the SNR18% chart is the whole truth, that is your prerogative. I can choose to reject it like any scientist or engineer. I can even counter you with the Sports scoring and the calculation that DXOmark put out (see above) and will give a result very different from your conclusion.
In the end, I choose to depend to my own eyes and experience. You can try to convince me otherwise till cows come home. I will still not buy into what you are saying. So lets just end here.
Last edited:
I am the same as you. I also cannot stand it when I feel something is claimed as God's truth when I feel it is not. Remember the premise that started all these? You claim DX (or even m43) can effectively shoot any scene as well as a FX cam. I am just telling you it is not so.
I am also telling you "facts" are not as clear cut as you think they are. Maybe when you move up into management level of the engineering field you may finally understand that.
BTW, I am not insulted. I am not peeved. I am just telling you my point of view that I am not wrong as you claim. But I am not saying you are wrong either. I just choose to believe in the results I see for myself than some ratings and charts as controversial as DXOmark's. In the end I believe that the right tool for the right job is the most important. There will be some things FX will do better than DX. There will be some things a DX will be more than enough. There will even be things that a PnS will be enough too. Right tool for the right job.
Last edited:
lets just base it on basic, simple physic law.
1) if all else being the same, same generation of sensor, same mega pixel (number of photosites), same anti-noise algorithm. FX will have better dynamic range since each photosite is larger as mega pixel is the same. in other word, if the DX sensor is great, imagine the same sensor in FF size.
2) to get "better" bokeh, camera should be as near to subject as possible and background as far as possible. so using the exact same lens, same focus length and same camera settings, you can go nearer to subject on FX. and please dont compare a 50mm on FX and 35mm on DX, the results wont be exactly the same.
if you dont need a FX then dont get a FX, saying DX is enough is just saying "DX is enough for me" and shouldnt be a general sweeping comment.
1) if all else being the same, same generation of sensor, same mega pixel (number of photosites), same anti-noise algorithm. FX will have better dynamic range since each photosite is larger as mega pixel is the same. in other word, if the DX sensor is great, imagine the same sensor in FF size.
2) to get "better" bokeh, camera should be as near to subject as possible and background as far as possible. so using the exact same lens, same focus length and same camera settings, you can go nearer to subject on FX. and please dont compare a 50mm on FX and 35mm on DX, the results wont be exactly the same.
if you dont need a FX then dont get a FX, saying DX is enough is just saying "DX is enough for me" and shouldnt be a general sweeping comment.
i know, thats why i use "better" bokeh. i avoid being too technical for beginners, became a habit.
It would not necessarily be better. Could be worse. And saying "better bokeh" has nothing to do with thinner depth of field.
Best to teach beginners the correct terminology from the start.
*popcorn*
:bsmilie:
seriously... like DD123 says, the right tools for the job.
i just don't geddit. i see people out there who use a D4 to take pictures of the same aesthetic quality a PnS beginner would take.
:bsmilie:
seriously... like DD123 says, the right tools for the job.
i just don't geddit. i see people out there who use a D4 to take pictures of the same aesthetic quality a PnS beginner would take.
Similar threads
