E5 sharper than E3? Not by much.

[rovingtim]

However, the interesting upshot of all this is that in playing with the different RAW developers (and comparing crops to Mr Cat's) I have found different RAW developers work differently on the different cameras.

You must have missed my thread:
http://forums.dpreview.com/forums/read.asp?forum=1022&message=38561969

You must have missed my point. Let me say it as simply as I can:

Raw Therapee good with E5. Raw Therapee not so good with E3. ACR good with E3. ACR not so good with E5.

Because cameras are apples and oranges, using the same developer at the same setting will necessarily favour one of the cameras. Why is this so hard for you to understand?

(rriley clued you in to the MTF50 results using the same lens on the E-3 and E-5. Those tests are incontrovertible proof for the much greater detail that can be obtained from the E-5.)

You really don't get it, do you? First of all, I've always agreed the E5 has more detail. However, mtf is also a measurement of the effectiveness of the chosen RAW developer with the apple and with the orange (also a measurement of false detail). Change a RAW developer and the numbers will changes. So how is mtf 'incontrovertible'?

So, it gets interesting. What is equal?

How about applying equal processing?

This is where you simply don't understand what we are talking about.

Now, here is the problem found with Lightroom (and most other developers): look at the fabric with the leaf pattern closely and you will see artefacts. RawTherapee "amaze" algorithm and Olympus Viewer do not produce such artefacts from the E-5 raw.

No, the artifacts simply turn into colour shifts, except where it is black and white (where RAW therapee is excellent). That is why msusic's E5 crops are going purple whereas the E3's aren't.

This is certainly true if you believe rovingtim's "methodology" is valid.

But it isn't.

You can't judge what you don't understand.

The only valid methodology is to process the files the same way, then compare the results.

What that reveals is how good the 'standard' process is with a given file from a given camera. In order to optimise my files, I have to use different sharpening, levels, noise reduction, etc. depending on the file from the same camera, let alone different cameras.

Remember that this thread is an argument that the E-5 doesn't deliver as much detail over the E-3.

I'm sorry, but I'm getting really tired of saying this: I DIDN'T SAY THIS. Are you purposely being thick?

The method to "prove" this contention is to use different developers for each file then post-process the E-3 result further.

Indeed. All is nonsense to those who don't understand.

 

[SirSeth]

Well then my eyes say the E-5 is noticeably sharper--but I haven't tested it against my E-3 for myself. I've just noticed that the E-5 pictures coming from users seem very nicely sharp across the board compared with E-3 user's pictures and my own E-1, E-330, and E-3 pictures. I do a little sharpening in LR3 now and then, and the E-3 sharpness is just fine, but the E-5 pictures just look more sharp to me without looking over enhanced. Resolution improvement is clear with the E-5, and sharpness is great. That's not a slam to my E-3, just observations. Is it over-hyped? I'm sure it is by some, but for me it will be a worthy upgrade in the future.

Cheers,
Seth

--
What if the hokey pokey really is what it's all about?

--
wallygoots.smugmug.com
wallygoots.blogspot.com

 

[light_bulb]

The best thing you can usually do for brand internal comparison purposes is to deploy the same brand-specific RAW converter that is supplied with the camera. You can expect that the manufactrurer knows his hardware pretty well and will use very similar approaches to get the best out of it.

Ooc pictures may also not be too bad. With similar camera settings you will achieve results that are not necessarily the best but very good for comparison.

While the E-5 has a weaker AA-filter and firmware-based algorithms are used to deal with moire, it is somewhat delicate to use other RAW converters because you can not know whether those are designed in a way that fits the AA filter. However using the same converters for both cameras (E3 and E5) will allow to compare converter specific results.

Doing that same test with all available RAW converters may help to single out the best camera RAW converter combination. Doing this may be rather lengthy already with standard settings but you have to bear in mind that there are so many settings that can be applied during RAW conversion that your time for taking pictures in the first instance may become very limited while you stay busy testing all available combinations.

Thus in some way or another you may find yourself with a second best solution concentrating on increasing your photographic skill which in most cases is the major shortcomming in the entire process.

While PP will never allow to convert a mediocre image into an award winning one, the best recommendation is to start with your skills and deal with PP in a more limited way. This is even more true as in the meantime you can expect further improvements being implemented into the next generation of RAW converters. Thus always shoot RAW at least with jpeg and the future is yours.

When I did my first comparison, a group of fine individuals said my finding were invalid because I was using Lightroom on the E5 and Lightroom had poor algorithms for the E5.

As such, my findings were invalidated.

As a direct response, I didn't use Lightroom on the E5. Now I'm being told that my findings are invalid because I didn't use Lightroom on the E5.

Do please tell me how I can possibly do right by this forum?

--
http://home.fotocommunity.de/andreaspastowski

 

[Brent Lossing]

See my reply to D3xmeister.

Makes it kind of silly to hold any kind of position on relative sharpness, then?

Unless I misunderstand your point - that relative sharpness is a result of the software used in post process. For different camera models (even from the same mfg) the results can vary substantially.

I guess if I believe the results of boggis the cat regarding color artifacts and software then this should be an easy one to see after.

So does that mean that people that are seeing differences between the E-3 and E5 - their results are either non-substantial or are the resullt of using software that compliments the E5's sharpness and not so much the E3? Emotional (not necessarily rationally) I find this hard to believe.

thanks,
brent

 

[rovingtim]

See my reply to D3xmeister.

Makes it kind of silly to hold any kind of position on relative sharpness, then?

Well, the final file might be the final measurement. I am arguing that simply using default settings on a single developer is too simple. That will give you relative sharpness under one specific condition.

Unless I misunderstand your point - that relative sharpness is a result of the software used in post process. For different camera models (even from the same mfg) the results can vary substantially.

I wouldn't say 'a result' rather 'influenced by'. But yes to the gist.

I guess if I believe the results of boggis the cat regarding color artifacts and software then this should be an easy one to see after.

The cat is right where he is right, but he's operating within a very narrow band. I do, however, disagree about the number of artifacts/moire generated by the E5. However, since I argue most aren't seeing them (like the purple cast on msusic's crops) perhaps it really isn't important to most shooters here.

They are there, though.

So does that mean that people that are seeing differences between the E-3 and E5 - their results are either non-substantial

'Substantial' is subjective so I couldn't really comment. The E5 does have more detail, but it is my argument that it doesn't 'lift the veil' from 4/3rds lenses.

or are the resullt of using software that compliments the E5's sharpness and not so much the E3?

That is the case with RAW Therapee, but it is opposite with Lightroom -- does good things to E3 files but obvious moire and artifacts litter E5 files. This suggests to me that RAW Therapee was designed for weaker AA's (processes on the assumption there is moire and artifacts to suppress) while ACR is better suited to cleaner files that have been softened a bit by an AA.

Emotional (not necessarily rationally) I find this hard to believe.

Even though I would argue the E3 is close behind the E5, the E3 files do need more sharpening. This means at higher ISO's the E5 has the edge because you don't have the problem of sharpening the noise as much. For high ISO shooting I would choose the E5 in a heartbeat over the E3.

Also, there is some evidence the E5 might be more accurate in focusing. That alone would be worth an upgrade.

 

[odl]

rovingtim wrote:
Ok wife has the baby now, boy was that a late night!

there is more detail in E5, see bottle labelcrop, some small artifacts, but crisp detail. what we expext.

Careful of false detail on the E5 shot, the most obvious being the horizontal lines in the pink cloth.

I wouldnt call it false detail, there is a repeating pattern in the fabric, the weave represents a number of directional repetitions, you can just as easily see a diagonal line as well, it all depends on your attempts at viewing it. When i compare the E-5 with the detail, compared to the E-3 that is lacking it, in all weaves. YOu can see the fabric weave on every colour.

However, I do want to stress that I am not saying the E5 doesn't have more detail, only that the detail increase has been grossly exaggerated. 'Lifting a veil' from the lenses, etc.

I suspect it is, fine lines are VERY fine with the E-5, images are crisp, check out this upscaling vs the a850 which has a strong AA filter. Even upressing from 12-24mp the E-5 holds its own quite well. I think this is what impresses people about the E-5.

I could sharpen up a bit more using lab colour on the lightness layer and do all sorts of things, but these two are just tweaking the basic values to try to bring up detail.

You couldnt upres the E-3 file and get the same clarity.

E-5 left- sony a850 right.









--
“Imagination is more important than knowledge. For
knowledge is limited to all we now know and
understand, while imagination embraces the entire
world, and all there ever will be to know and
understand.” - Albert Einstein

 

[Rriley]

i dont think many (Brent Lossing, Boggis and myself) has caught on how pointless it is gazing at images, it wont give you a hard number to tell you how different the perceived 'sharpness' is, where actually we are talking about resolution

a res chart will give you that, and so will MTF measurement using the same lens at the same settings. As it happens we have access to both, and its conclusive that e5 has higher scores both jpeg and RAW.

As to how much more resolution, well we have to think in relation to what, but we can also calculate that.

Despite canon 7D being higher Mp by a factor of 21%, E5 is within 15% of that resolution delivery.

--
Riley

any similarity to persons living or dead is coincidental and unintended

 

[rovingtim]

I'm not saying the E5 has more detail, but I am saying there is more detail in the E3 than people seem to think.

I would also argue the horizontal lines in the pink fabric of the E5 is false detail. Boggis the cat posted a Pentax image of the same fabric to show what was really there (I used the D3x; the Pentax is better). There is nothing horizontal there.

http://forums.dpreview.com/forums/read.asp?forum=1022&message=38584987

 

[odl]

There is visible detail loss in your copy of my file, compared to when i posted it. YOu can see it in the red fabric. Somehow you have softened the E-5 image.

What save for web settings are you using?

Now show the bottle label, fabric detail and the like allows our vision to "fill in the blanks" but fine line detail will not.

Ab

--
“Imagination is more important than knowledge. For
knowledge is limited to all we now know and
understand, while imagination embraces the entire
world, and all there ever will be to know and
understand.” - Albert Einstein

 

[odl]

i dont think many (Brent Lossing, Boggis and myself) has caught on how pointless it is gazing at images, it wont give you a hard number to tell you how different the perceived 'sharpness' is, where actually we are talking about resolution

But Rovingtim is ignoring those stats, I am just having fun with the files (as i am up in the middle of the night with my baby boy). This thread shows something, that the majority of people here can see the difference. That means it is there. There are only a handful that agree (with reservations) with Tim.

a res chart will give you that, and so will MTF measurement using the same lens at the same settings. As it happens we have access to both, and its conclusive that e5 has higher scores both jpeg and RAW.

I know. Plus you can see with tims PP, jaggies on the bottle label and colour separation on the text stripes. So he has forces a perfectly good resolution beyond its comfort limit.

As to how much more resolution, well we have to think in relation to what, but we can also calculate that.

Despite canon 7D being higher Mp by a factor of 21%, E5 is within 15% of that resolution delivery.

The E-5 is an excellent performer, tot he point where I am adding up the sony sales and working out what i can get in the E-5 + SHG ballpark

--
Riley

any similarity to persons living or dead is coincidental and unintended

--
“Imagination is more important than knowledge. For
knowledge is limited to all we now know and
understand, while imagination embraces the entire
world, and all there ever will be to know and
understand.” - Albert Einstein

 

[rovingtim]

The change would be dpr galleries. I didn't save for web. I simply copied your file. It is as before.

Try downloading the file when it is expanded to full size and you should see both of them more clearly on your computer.

 

[Richard Pavek]

First, as a strong supporter of the E-5 Image being superior to the E-3 I really resent your statement that I don't know how to process E-3 files. I do and long ago figured out that Lightroom is far superior to Olympus Viewer when it comes to getting detail in E-5 images. Your results were skewed by using different processors.

That said, there is no reason that you could not process the E-5 images exactly the same way you did the E-3 images. Personally I never used on any preset protocols for the E-3, preferring to develop my own settings. I use the same settings now for both my E-3s and E-5s.

Also, remember that when the E-5 was announced Oly they said that now their sensor matched the quality of their top lenses. Their top two lenses are the 150mm f2.0 and the 300mm f2.8. Using anything less as testing lenses may not show much difference.

As you may know I am a wild bird photographer and am always concerned about fine detail, especially in feathers. When I went out with the E-3, I mounted the ZD300 as I always need a long reach. Now with the E-5, I mount the 300 plus the EC14 converter because I can now get the same detail at the greater distance with the convertor, which I couldn't with the E-3.

I crop a lot. Almost all of my in flight photos are cropped. I am able to crop considerably more with the E-5 images than I could or can with the E-3, because the detail in the E-5 images is now sharp enough to permit that.

I have taken and processed several thousand photos with my two E-5s and there is no question but what the E-5 delivers a sharper, more detailed image than the E-3. This is true both with and without the EC14.

(By the way, comparing jpegs just won't cut it, too much detail is lost in converting from tif to jpeg. I make all my large (up to 48" x 64") prints from tif files or psd, not jpeg.
Richard

 

[boggis the cat]

You must have missed my point. Let me say it as simply as I can:

Raw Therapee good with E5. Raw Therapee not so good with E3. ACR good with E3. ACR not so good with E5.

Not true. The E-5 samples have more detail when developed using either RawTherapee or Lightroom or OlympusViewer. I would be very surprised if any developer could produce anywhere close to as much detail from an E-3 raw as an E-5 -- such a claim is simply not consistent with the physical fact of the lighter AA filter in the E-5.

RawTherapee ("amaze" algorithm):





Lightroom:





Olympus Viewer 2:





The premise of this thread -- the claim in the title "E5 sharper than E3? Not by much." -- is clearly not valid. Put simply: you are quite wrong, and demonstrably so.

 

[esco]

in results if you were trying to make a point, if anything you've managed to support his claim of "not by much" with your images. . .

You must have missed my point. Let me say it as simply as I can:

Raw Therapee good with E5. Raw Therapee not so good with E3. ACR good with E3. ACR not so good with E5.

Not true. The E-5 samples have more detail when developed using either RawTherapee or Lightroom or OlympusViewer. I would be very surprised if any developer could produce anywhere close to as much detail from an E-3 raw as an E-5 -- such a claim is simply not consistent with the physical fact of the lighter AA filter in the E-5.

RawTherapee ("amaze" algorithm):





Lightroom:





Olympus Viewer 2:





The premise of this thread -- the claim in the title "E5 sharper than E3? Not by much." -- is clearly not valid. Put simply: you are quite wrong, and demonstrably so.

--
Oldschool Evolt shooter

 

[boggis the cat]

Also, remember that when the E-5 was announced Oly they said that now their sensor matched the quality of their top lenses. Their top two lenses are the 150mm f2.0 and the 300mm f2.8. Using anything less as testing lenses may not show much difference.

You will see significant differences in lens test results for any lens, I think. See rriley's posts concerning the 12-60 MTF50 results on the E-3 and E-5 up-thread.

I crop a lot. Almost all of my in flight photos are cropped. I am able to crop considerably more with the E-5 images than I could or can with the E-3, because the detail in the E-5 images is now sharp enough to permit that.

I have taken and processed several thousand photos with my two E-5s and there is no question but what the E-5 delivers a sharper, more detailed image than the E-3. This is true both with and without the EC14.

(By the way, comparing jpegs just won't cut it, too much detail is lost in converting from tif to jpeg. I make all my large (up to 48" x 64") prints from tif files or psd, not jpeg.

I save JPEG intermediates at 100% setting (whatever that may mean, of course) to try to retain as much detail as possible for any "comparative" purposes. JPEG loses detail and introduces artefacts, of course.

There was a thread not so long ago based on the premise that the Olympus SuperFine JPEG setting was pointless because Fine was "good enough". A lot of the threads on this forum seem to be farcical, with the same people taking logically contradictory positions.

 

[SirSeth]

I'm a math guy and I revere numbers and measures, but I also know there are a million ways to introduce error into even the simplest tests. The experience of Richard is very weighty indeed. It's not just my eyes, but those of many experienced users shooting thousands of frames.

For me at least, case closed.

Cheers,
Seth

--
What if the hokey pokey really is what it's all about?

--
wallygoots.smugmug.com
wallygoots.blogspot.com

 

[boggis the cat]

in results if you were trying to make a point, if anything you've managed to support his claim of "not by much" with your images. . .

I don't think so. Possibly there is an issue with your browser or something similar. I'll point out what I think is significant, in case you want to re-examine these crops.

RawTherapee ("amaze" algorithm):





Look at the detail present in the fabric. You can easily see the fabric texture on the E-5 sample, but not on the E-3. Also look at the top of the neck of the bottle -- again you see texture in the E-5 where none is present on the E-3.

Lightroom:





The E-5 sample reds are over-saturated here (due to my LR fiddling), but look at the detail on any of those fabric swatches. Again, you can also see texture on the bottle in the E-5 sample.

Olympus Viewer 2:





These samples show the greatest differences. The E-3 sample is entirely devoid of texture.

The stronger AA filter of the E-3 has blurred away fine detail. No amount of processing can restore lost detail.

 

[rovingtim]

I really resent your statement that I don't know how to process E-3 files.

I also said the AF on the E3 could be creating soft images. I have many wildlife images from the E3 where the light wasn't very bright and the focus missed just by a hair and thus critical sharpness was lost. No amount of sharpening can save it then.

Like the front focused image Olympus America used to promote the E3's AF capabilities.

I do and long ago figured out that Lightroom is far superior to Olympus Viewer when it comes to getting detail in E-5 images. Your results were skewed by using different processors.

Sorry, I should have anticipated this. I originally was using Lightroom for both cameras but, in another thread, was told that Lightroom did not do justice to the E5 files and only out-of-camera jpgs would do. A tribunal then declared my results void as a result. So I processed the E5 in Viewer instead. Now, of course, people criticise me for not using Lightroom, though I understand your position better than the other.

Edit: If you want to definitively prove me wrong Richard, set up your tripod and take two identical pictures using the E3 and E5 and then you process the E5 and I'll process the E3 and then we compare.

However, because the E5 images start out sharp, adding more sharpness doesn't really reveal much more whereas on the E3 it reveals a lot.

Also, remember that when the E-5 was announced Oly they said that now their sensor matched the quality of their top lenses. Their top two lenses are the 150mm f2.0 and the 300mm f2.8. Using anything less as testing lenses may not show much difference.

I do agree that the E5 shows more detail, but if it can only be seen using SHG lenses, that doesn't say much about the sharpness of Oly's lenses further down, does it?

I crop a lot. Almost all of my in flight photos are cropped. I am able to crop considerably more with the E-5 images than I could or can with the E-3, because the detail in the E-5 images is now sharp enough to permit that.

I would suggest you may also be enjoying more accurate focus.

I have taken and processed several thousand photos with my two E-5s and there is no question but what the E-5 delivers a sharper, more detailed image than the E-3. This is true both with and without the EC14.

Again ... focus.

(By the way, comparing jpegs just won't cut it, too much detail is lost in converting from tif to jpeg. I make all my large (up to 48" x 64") prints from tif files or psd, not jpeg.

If this is the case, then we really are looking at the last percentile aren't we? And in that last percentile, I would agree with you.

 

[rovingtim]

The E3 does show texture when processed in Lightroom. So how about you repeat this process using Lightroom instead, especially as that is what you espouse as 'fair'?

And if you do this, don't forget to add your 200% crop of the wheel. Lightroom doesn't render the E5 print as well as RAW Therapee ... at least I couldn't get it to do so. The E3 v E5 files look much closer and the E30 v E5 files look nearly identical (the E30 being ever so slightly softer than the E5 but containing more detail than the E3).

 

[rovingtim]

Richard, if you want to definitively prove me wrong, set up your tripod, take two identical pictures, you process the E5 I process the E3 and then we compare.

Simple.