Lousy JPEG engine ?

Started May 7, 2012 | Discussions
Camancha
Contributing MemberPosts: 740
Like?
Lousy JPEG engine ?
May 7, 2012

Hello guys,

I am a K10d owner but looking for a new camera body. I have narrowed it down to Pentax K-01 and Olympus OM-D. I out-ruled Samsung NX200 due to lack of IBIS.

I have more or less figured out the pros and cons of the two remaining bodies, but one thing puzzles me: According to what I have read the K-01 has the best sensor (IQ), but when using the studio comparison tool I got somewhat surpised:

1: The OM-D looks better at ISO 200 and ISO 400, both in RAW and JPEG.
2: At ISO 800, in RAW, the OM-D has a little more noise but also more detail.
3. At ISO 1600 and beyond, In RAW, K-01 is better.
4. At ISO 200-3200 (!), in JPEG, the OM-D is better.

So...the K-01 JPEG engine is mediocre at best and even its sensor (RAW) is worse at base/low ISO (up to ISO 400) ?

Kind regards
.lars

Pentax K-01 Pentax K10D
If you believe there are incorrect tags, please send us this post using our feedback form.
AllBrands
Regular MemberPosts: 481Gear list
Like?
Re: Lousy JPEG engine ?
In reply to Camancha, May 7, 2012

I can hardly wait to see if you'll be told; A) you're only a troll, B) you're wrong or C) JPEG engines don't matter because everyone worth their salt shoots RAW anyway. While I have no problem with either your comparison or your conclusions, I expect you'll be soon pounced upon as though you'd just called someone's child ugly.

It's interesting to me the way so many people accept that the JPEG engine in their favorite camera isn't going to be good enough so one must shoot RAW. Why is that acceptable? Personally, I shoot RAW when the situation warrants it but I'd be VERY unhappy with any camera with a sub-par JPEG engine. Whatever magic Olympus uses in their JPEG engines seems to always be very good. Pentax does a good enough job for me too but I've always liked Olympus JPEG's.

Good luck with your thread.

 AllBrands's gear list:AllBrands's gear list
Canon PowerShot SD550 Canon PowerShot SD1100 IS Olympus E-420 Pentax K-7 Pentax K-5 +10 more
Reply   Reply with quote   Complain
JoeDaBassPlayer
Veteran MemberPosts: 3,583
Like?
Re: Lousy JPEG engine ?
In reply to AllBrands, May 7, 2012

The OMD shots are over processed and oversharpenned. The K 01 results, like other Pentax cameras as before, are better in Natural than in Bright Tone.
--
Variance is Evil!

Reply   Reply with quote   Complain
JoeDaBassPlayer
Veteran MemberPosts: 3,583
Like?
Re: Lousy JPEG engine ?
In reply to JoeDaBassPlayer, May 7, 2012

A lot of what you are seeing on the Oly end is a darkening of light detail to provide a higher contrast.
--
Variance is Evil!

Reply   Reply with quote   Complain
MaxIso
Contributing MemberPosts: 621
Like?
Re: Lousy JPEG engine ?
In reply to AllBrands, May 7, 2012

AllBrands wrote:

I can hardly wait to see if you'll be told; A) you're only a troll, B) you're wrong or C) JPEG engines don't matter because everyone worth their salt shoots RAW anyway.

so u basically list every possible response other than against pentax, and that proves that u r clever? i doubt hes a troll, there is no "wrong" if u have specific needs, and raw is superior to jpeg.

as for the op, if u r looking for pure IQ then shoot raw on the pentax. i dont know enough about the olympus to debate features but raw samples speak for themselves. if u prefer MF the pentax has focus peaking which is super useful. the only reason imo to get the olympus is if u insist on shooting jpeg for whatever reason. in case u havent seen, here is a raw comparison at base iso and 6400. there is another thread a guy posted some k-01 shots at 6400iso and they were not bad, usable for small prints. as u can see in the link, the olympus isnt close.

http://www.dpreview.com/reviews/studiocompare.asp#baseDir=%2Freviews_data&cameraDataSubdir=boxshot&indexFileName=boxshotindex.xml&presetsFileName=boxshotpresets.xml&showDescriptions=false&headerTitle=Studio%20scene&headerSubTitle=Standard%20studio%20scene%20comparison&masterCamera=pentax_k01&masterSample=imgp1365.acr&slotsCount=4&slot0Camera=pentax_k01&slot0Sample=imgp1365.acr&slot0DisableCameraSelection=true&slot0DisableSampleSelection=true&slot0LinkWithMaster=true&slot1Camera=oly_em5&slot1Sample=p1010003.acr&slot2Camera=pentax_k01&slot2Sample=imgp1382.acr&slot3Camera=oly_em5&slot3Sample=p1010017.acr&x=-0.45615433561587765&y=-1.6446010522657952

Reply   Reply with quote   Complain
JohnBee
Veteran MemberPosts: 3,932
Like?
Re: Lousy JPEG engine ?
In reply to Camancha, May 7, 2012

I"m going to say yes on the JPG engine. Though I'd remind you that it is possible to adjust(sharpen) output on either camera which would likely leave both in very similar territory. Likewise... I noticed the K-01 suffers(in detail) at the edge of the scene in contrast to the OM5 which may or may not be attributed to optical performance. And to support that, I'd point to you're attention to the coin on top of Mikey's head, where we find more detail from the K-01 than with the OM5.

Whatever the case, I would definitely point to the JPG engine on this one. Not to mention the grain quality on the Oly at this higher base sensitivity.

PS. just for gags I looked at both system fr. RAW and found the K-01 to have much more detail. However, in many ways, the Olympus seems to have better color resolution and contrast(SEE: red channel performance).

Hope this helps.

Reply   Reply with quote   Complain
chiane
Senior MemberPosts: 4,046
Like?
Re: Lousy JPEG engine ?
In reply to Camancha, May 7, 2012

I hate these pixel peeping comparisons. I think you'll never notice a difference in real life use.

Reply   Reply with quote   Complain
rogerstpierre
Veteran MemberPosts: 3,339Gear list
Like?
Irrelevant...
In reply to Camancha, May 7, 2012

The differences observed are so minutes that the variances are irrelevant in all practical applications, and highly subject to PP choices. The Oly probably has a better JPEG engine in camera than Pentax, but I am not certain how the 4/3 fairs in DR compared to the K-01. I suspect not quite as well. Oly always fairs on the more contract/sharpness side and Pentax the other.

-- hide signature --

Roger

 rogerstpierre's gear list:rogerstpierre's gear list
Pentax smc DA 21mm F3.2 AL Limited Pentax Optio 33WR Samsung TL500 Pentax K10D Pentax K-01 +16 more
Reply   Reply with quote   Complain
JoeDaBassPlayer
Veteran MemberPosts: 3,583
Like?
Re: Irrelevant...
In reply to rogerstpierre, May 7, 2012

The JPEG engine is obviously optimized for the low contrast detail like hair. It appears when an area has low contrast, suddenly it gets hyped. The side of one bottle has light text on white. With Oly, the text is quite dark. The shiny label on one liquor bottle has some wrinkles. They are subtle. Some cameras smooth them over. However, the foremost section has exaggerated contrast compared to the rest of the label.

There is some serious cooking going on. It helps bring out detail in hair but it does leave some artifacts. If you like it, fine. Just do not flaunt it as better.

rogerstpierre wrote:

The differences observed are so minutes that the variances are irrelevant in all practical applications, and highly subject to PP choices. The Oly probably has a better JPEG engine in camera than Pentax, but I am not certain how the 4/3 fairs in DR compared to the K-01. I suspect not quite as well. Oly always fairs on the more contract/sharpness side and Pentax the other.

-- hide signature --

Roger

-- hide signature --

Variance is Evil!

Reply   Reply with quote   Complain
ET2
ET2
Senior MemberPosts: 3,748Gear list
Like?
Re: Lousy JPEG engine ?
In reply to Camancha, May 7, 2012

Camancha wrote:

I have more or less figured out the pros and cons of the two remaining bodies, but one thing puzzles me: According to what I have read the K-01 has the best sensor (IQ), but when using the studio comparison tool I got somewhat surpised:

The sensor performance is similar to the other 16 MP sensors; all of these cameras score within 5-points on dxomark (remember, 5 points on dxomark is equal to 1/3 stops).

Oly jpegs are impressive.

IQ is only not the only thing. In real usage you won't even notice these differences. There are many other factors that are just as important (for example auto-focus, lenses, ergonomics).

The weather sealed Oly (with buil-in EVF) is hands down a superior camera, though it's $400 more expensive too.

Reply   Reply with quote   Complain
MaxIso
Contributing MemberPosts: 621
Like?
Re: Lousy JPEG engine ?
In reply to chiane, May 7, 2012

chiane wrote:

I hate these pixel peeping comparisons. I think you'll never notice a difference in real life use.

i dare u to take a lousy iso performing camera and shoot a night game in little league, or an indoor volleyball game. see if u can avoid blurr without getting a picture made of sand. real life use is the only time noise matters. in studio u can do whatever u want, its a controlled environment.

if u shoot landscape or portraits perhaps its not as important, but anybody ive ever known with a camera has used it for something that moves at one point or another. is the olympus bad? no, its a modern high tech piece of engineering, its just not as high tech as some other cameras thats all. this is a review on DXOMark for what its worth. ive already shown u the sample pics, heres what some of the tech nerds say.

"Let’s now quit the micro 4:3 universe in search of other alternatives to the OM-D E-M5, starting with one of the most recently-announced models, the Pentax K-01. Pentax also adopted a retro design — and a more daring one at that, thanks to the input of Australian designer Marc Newson. The Pentax K-01 has nothing new to offer with respect to an integrated viewfinder and other ergonomic details important to amateur and expert photographers alike, but as with the OM-D, it offers integrated (if rather conventional) image stabilization.

Short of a truly bad surprise, the Pentax K-01 will provide superior image quality to that of the OM-D E-M5 because it uses a sensor renowned for its sensitivity and dynamic range, the 16Mpix CMOS (which fits into the APS-C’s intrinsically larger body). And even though the Olympus OM-D E-M5 can take advantage of some very nice micro 4/3-compatible lenses (Olympus M.Zuiko Digital and Panasonic Lumix G lines), the Pentax K-01 uses the much broader array of Pentax K-mount lenses".

include focus peaking and almost 3 times as many focus points and it just seems an easy answer. good luck with whatever u pick, but i hope u choose whats best.

Reply   Reply with quote   Complain
Eric O
Regular MemberPosts: 378
Like?
Re: Lousy JPEG engine ?
In reply to MaxIso, May 7, 2012

MaxIso wrote:

in case u havent seen, here is a raw comparison at base iso and 6400. [...] as u can see in the link, the olympus isnt close.

http://www.dpreview.com/reviews/studiocompare.asp#baseDir=%2Freviews_data&cameraDataSubdir=boxshot&indexFileName=boxshotindex.xml&presetsFileName=boxshotpresets.xml&showDescriptions=false&headerTitle=Studio%20scene&headerSubTitle=Standard%20studio%20scene%20comparison&masterCamera=pentax_k01&masterSample=imgp1365.acr&slotsCount=4&slot0Camera=pentax_k01&slot0Sample=imgp1365.acr&slot0DisableCameraSelection=true&slot0DisableSampleSelection=true&slot0LinkWithMaster=true&slot1Camera=oly_em5&slot1Sample=p1010003.acr&slot2Camera=pentax_k01&slot2Sample=imgp1382.acr&slot3Camera=oly_em5&slot3Sample=p1010017.acr&x=-0.45615433561587765&y=-1.6446010522657952

That's a pretty clear difference in favor of the K-01. I would have to say that even if I were planning to record mostly in JPEG only, it would still disturb me quite a lot to know that the original sensor data that the camera was working with had so much noise (especially color noise) on the Oly.

Reply   Reply with quote   Complain
ET2
ET2
Senior MemberPosts: 3,748Gear list
Like?
Re: Lousy JPEG engine ?
In reply to Eric O, May 7, 2012
Reply   Reply with quote   Complain
ET2
ET2
Senior MemberPosts: 3,748Gear list
Like?
Re: Lousy JPEG engine ?
In reply to MaxIso, May 7, 2012

MaxIso wrote:

include focus peaking and almost 3 times as many focus points and it just seems an easy answer. good luck with whatever u pick, but i hope u choose whats best.

Both cameras have contast detect AF system. That is, you can pick any point on the screen as "focus points". On Nex-5N for example, that has a touch screen, you can touch anywhere on the screen and that becomes "the focus point". Same is true for all Panasonic touch screen models.

PDAF AF has limited number of AF points. More "points" don't mean anything on CDAF cameras.

EM-5 AF is known to be among the best for CDAF cameras.

Reply   Reply with quote   Complain
MaxIso
Contributing MemberPosts: 621
Like?
Re: Lousy JPEG engine ?
In reply to ET2, May 7, 2012

just ignore ET2 if u havent met him, he comes to pentax forums to counter anything good we say. the K-01 and k5 offer better raw IQ straight out of camera than any apsc in existence. they were actually smart about it, they offer close to the same sensor IQ in two very different bodies.

Reply   Reply with quote   Complain
MaxIso
Contributing MemberPosts: 621
Like?
Re: Lousy JPEG engine ?
In reply to ET2, May 7, 2012

ET2 wrote:

MaxIso wrote:

include focus peaking and almost 3 times as many focus points and it just seems an easy answer. good luck with whatever u pick, but i hope u choose whats best.

Both cameras have contast detect AF system. That is, you can pick any point on the screen as "focus points". On Nex-5N for example, that has a touch screen, you can touch anywhere on the screen and that becomes "the focus point". Same is true for all Panasonic touch screen models.

PDAF AF has limited number of AF points. More "points" don't mean anything on CDAF cameras.

EM-5 AF is known to be among the best for CDAF cameras.

what about focus peaking? u should be talking big about this since your sony cameras have it as well. or are u scared to shoot in manual?

Reply   Reply with quote   Complain
ET2
ET2
Senior MemberPosts: 3,748Gear list
Like?
Re: Lousy JPEG engine ?
In reply to MaxIso, May 7, 2012

MaxIso wrote:

just ignore ET2 if u havent met him, he comes to pentax forums to counter anything good we say. the K-01 and k5 offer better raw IQ straight out of camera than any apsc in existence. they were actually smart about it, they offer close to the same sensor IQ in two very different bodies.

IQ differences (and I am sure K01 is slightly better in RAW) should NOT be the sole reason to pick a camera. DPR reviewers have K01, EM-5, 5N, K-5. They used all these cameras extensively. Took thousands of photos. Read their conclusion:

http://www.dpreview.com/reviews/olympusem5/22

"The E-M5 can't completely overcome the light capture disadvantage brought by its smaller sensor, compared to APS-C, but it reduces it to the point that it's irrelevant for almost all practical purposes. At which point we think its size advantage, in terms of both body and lenses, will outweigh that difference for most uses. If you're absolutely unwilling to compromise on image quality then spending twice the money and moving up to the bulk of a full-frame is the only way of gaining a significant step up from the E-M5."

Do you think DPR just made that up? Having all these cameras. They obviously used them all and that was their conclusion.

Reply   Reply with quote   Complain
ET2
ET2
Senior MemberPosts: 3,748Gear list
Like?
Re: Lousy JPEG engine ?
In reply to MaxIso, May 7, 2012

MaxIso wrote:

EM-5 AF is known to be among the best for CDAF cameras.

what about focus peaking? u should be talking big about this since your sony cameras have it as well. or are u scared to shoot in manual?

Focus peaking is cool if you are into manual focusing. That's not for everyone, obviously.

Even then, if you want perfect focus, you get better results with magnification than focus peeking, especially with large aperture lenses (like anything faster than F2).

For example, DPR studio shots are taken in manual focus mode. I will bet that they didn't rely on focus for something that is critical.

Reply   Reply with quote   Complain
ET2
ET2
Senior MemberPosts: 3,748Gear list
Like?
Re: Lousy JPEG engine ?
In reply to ET2, May 7, 2012

ET2 wrote:

For example, DPR studio shots are taken in manual focus mode. I will bet that they didn't rely on focus for something that is critical.

Ops, I meant "I will bet that they didn't rely on focus peaking for something that is that critical."

Reply   Reply with quote   Complain
awaldram
Forum ProPosts: 10,683Gear list
Like?
What a melodrama over nothing
In reply to Camancha, May 7, 2012

Well you've have seen by now nothing like a bit of pixel peeper excites the pinicky and drives their exertions both for and against

The truth of the matter is from the images you've used your findings are deadly accurate..

But you conclusions are wrong (probably)

We wont know how good the jpeg engine is until we see resolution tests and then only if conducted correctly.

All you can really say about your conclusion and theses samples is out of the box your going to prefer the jpegs from the Olympus.

Now if your chasing best IQ from any camera the term 'default settings; makes no sense.

Given the capabilities of all cameras to tune jpeg output to your own taste what is the point of showing out of the box examples.?

Did you know Nikon program different default in their Jpeg output for different markets ?
Olympus default their cameras to deliver P+S type sharpness and saturation.
Pentax default their cameras to Asian sensibilities a more pastel soft image

the reason why pixel peeping and comparing review images is a pretty pointless task is as follows.

1 competence of the photographer
2 was the lens clean
3 was the lens focused
4 Lens quality
5 were lighting conditions the same
6 jpeg engine settings
7 raw converter setting the same
8 lens used
9 environmental

Without knowing all these and many more questions comparing cameras jpeg and RAW output is error prone.

Much as I hate to say it ETs's beloved DXO site gives a better indication of the merits of the two cameras jpeg or otherwise.

But all these reviews only give an indication not rock solid proof of performance.

 awaldram's gear list:awaldram's gear list
Pentax Q Pentax K-3 Pentax smc DA* 55mm F1.4 SDM Pentax smc DA* 16-50mm F2.8 ED AL (IF) SDM Pentax smc DA* 50-135mm F2.8 ED (IF) SDM +11 more
Reply   Reply with quote   Complain
Keyboard shortcuts:
FForum MMy threads