Leica S2 discussed by Erwin Puts

[ErikKaffehr]

Hi!

Erwin Puts just posted part 4 of his report on he Leica S2: http://www.imx.nl/photo/leica/camera/page176/s2part4.html

Best regards
Erik

[kers]

With respect to the author, I think his test is not very well made.

testing has to be done very well or they are not worth that much.

Too may factor are not taken into account - lenses for one. ( and on the lipstick series the d3x photo is not sharp -the image has moved)

...But it is obvious that moire is more clear without an AA filter and that the A900 has the strongest one.

( the 4 pictures made neutral with 100% hue saturation in photshop.)

[ErikKaffehr]

Hi,

I may have a tendency not to overly critical about testing that other folks do. I know it isn't easy.

The lipstick shots demonstrate purple fringing pretty well. According to Mr. Puts this is the results under corrected longitudional chromatic aberration. The lipstick pictures demonstrate this cleanly. It is interesting that the M9 has none and and the S2 some.
I do agree that there is camera vibration in the D3X picture, but it does not affect the purple fringing.

There are very few lenses that don't have longitudinal chromatic aberration at large apertures, this is quite obvious from the tests at  http://www.photozone.de .


Regarding Moiré I'd suggest that there are couple of different issues, Moiré patterns arise when a pattern having pitch near the sensor pitch is imaged. So the presence of moiré is very much dependent on the subject distance. Also, moiré patterns and aliasing effects are not just about color, even if color moiré is most disturbing.  

My interpretation of Mr. Puts's tests is that:

- The Nikon D3X gives a remarkably good image
- Under the premises of test the Leica M9 cannot really show significant advantages over the Alpha 900 or D3X, it's the other way around.
- The S2 has real benefits
- The D3X and the Alpha 900 are quite close

It is very interesting that Mr. Puts finds that the DSLRs give superior quality compared to the M9, especially as many authors on this forum have found that the M9 is significantly sharper than DSLRs. This may depend on the use of Lightroom 3 which has a much improved processing pipeline.

One final observation may be that there is often a difference between lab tests and perceived image quality. There may be different explanations for this. There are for instance lenses that seem to impress in the field but less so in formal testing and also the other way round. This may to do with things like field curvature, lens contrast and other factors like focusing accuracy.

Best regards
Erik

With respect to the author, I think his test is not very well made.

testing has to be done very well or they are not worth that much.

Too may factor are not taken into account - lenses for one. ( and on the lipstick series the d3x photo is not sharp -the image has moved)

...But it is obvious that moire is more clear without an AA filter and that the A900 has the strongest one.

( the 4 pictures made neutral with 100% hue saturation in photshop.)

[fredjeang]

Hi,

Just a quick look at those crop comparaisosn and the S2 has a much better transition between the tones, and that has consequences in
the time saved in post production.
Without talking about the minimalist ergonomics, but of course this S2 is far from mature in certain areas as it has been discussed here intensivelly.

But, I would like to introduce some thoughts about these testings in general.

- The first problem that comes to my mind, and frankly weird and anoying, is that those testings are generally done in conditions that have nothing to do with the real world.
The provided files are useless unless you belong to a police department crime investigation.
Are those testings photography? No. Are they unperfect indication of an overall capability of a product? yes, IMO that is all they are when done "good enough". (but it will never be enough)

- The complains generally made are that this is not relaible, that it misses this or that and they should have done those the other way etc...do not bring any
valuable information either because it will never be perfect and each time one will irrupts to bark against the lightning, the post processing or whatever with always good reasons.
This is an ideal perfection freshly imported from the tech-maniac-land thematic park that will never ever happened, because it simply can not.

- I understand why Michael and the team involved in the testings have more or less give up, because in the end, those are a lot of time and energy spent and poorly rewarded.
After several days of constant bombing in the forums by all kind of people in search of provocation or being interesting, the terrain smokes and indeed you can have a pretty thick skin
but it hurts. So why people should bother doing some advanced testings if they know they gona be immediatly hang-up on the public square?

At least, the Lu-La S2 testing was done in conditions closed to a fashion shot. Yes, there were no time stress, neither a nervous young and arrogant AD involved, but to be fair
that was the closest real world gear testing I have seen so far. Unperfect? of course. But closer to anything that has been done, and instead of having receiving first, and before
the critics showers, some congratulations (and I'm not talking about receiving awards but simply recognition of the time spent for us), what generally happens are missiles fired
in a question of hours and the forum filled by erasmus scientist deliriums datas, more or less serious, that would put to shame the entire Silicon valley.

This S2 is probably better that we think and worse that what was expected.

But again, let me allow to repeat once again a Guy's statement that I like, knowing that is it shared fortunatly by many: trust the eyes not the numbers.

[michael]

Watch for Mark Duboboy's S2 "First Impressions" review here on Monday.

As for Puts, while I have great respect for his lens testing, I and a number of others feel that he misses the mark when it comes to testing digital cameras.

With regard to the M9, I know of no one other than Mr. Puts who thinks that the M9 is anything other than stellar, and far superior to the A900 or other 24MP DSLRs. Not even close.

Ps. I own and regularly use both an M9 and an A900 and can tell you that as good as the A900 is, the M9 smokes it in IQ. Similarly when it comes to lenses. The Sony / Zeiss lenses are excellent, but not in the same league as Leica's M glass.

BTW; readers should not infer from this that these differences are huge. They're not. They're subtile, and possibly not even obvious to someone not looking for them. At this level of performance we are talking about small and incremental differences that may not be critical to all users. To some though, they will.

Michael

[telyt]

As for Puts, while I have great respect for his lens testing, I and a number of others feel that he misses the mark when it comes to testing digital cameras.

Seconded.  My experience with Leica-R lenses parallels Puts' tests but in testing digital cameras he's still on the learning curve.

[ErikKaffehr]

Michael,

I'm much aware of your experience with the M9. Another poster on this forum is Eleanor Brown whom I also trust, and she seems to be quite enthusiastic about the image quality from the M9. I'm much looking forward to Mark Dubovoy's first impression.

I don't know what the issue is with Mr. Puts testing. Something I'm aware of is that it is not the same thing to test in the lab and in real life. Still, I would prefer to have an explanation why Mr. Puts's testing differs from other authors experience. It sort of make no sense, you have a Leica enthusiast with great experience of testing and he still arrives to findings at odds with many renowned photographers.

I was considering LR3 to be part of the explanation, it has a new processing pipeline giving much better sharpness, but I know that you are one of the beta testers of LR3, so I presume that you already have eliminated that variable.

Regarding the test by Erwin Put's I just want to share the information. So I'm not particularly pro or contra Leica. Also, I really appreciate that Erwin Puts shares his information. If a systematic error is found or is known in Mr. Puts testing it would be nice to have it known.

Some things that are obvius to me:

- Contrast range is limited in test
- Flat field targets are photographed at relatively close distance

Focusing and camera vibrations may be an issue, but Erwin Puts says that he is very careful with focusing. At least one of the pictures show some weakness, probably camera vibration, though.

My take is, Mr. Puts tests are interesting but don't necessarily paint the entire picture. He does publish detailed comparisons and that is a good thing.

Best regards
Erik

Watch for Mark Duboboy's S2 "First Impressions" review here on Monday.

As for Puts, while I have great respect for his lens testing, I and a number of others feel that he misses the mark when it comes to testing digital cameras.

With regard to the M9, I know of no one other than Mr. Puts who thinks that the M9 is anything other than stellar, and far superior to the A900 or other 24MP DSLRs. Not even close.

Ps. I own and regularly use both an M9 and an A900 and can tell you that as good as the A900 is, the M9 smokes it in IQ. Similarly when it comes to lenses. The Sony / Zeiss lenses are excellent, but not in the same league as Leica's M glass.

BTW; readers should not infer from this that these differences are huge. They're not. They're subtile, and possibly not even obvious to someone not looking for them. At this level of performance we are talking about small and incremental differences that may not be critical to all users. To some though, they will.

Michael

[RobSaecker]

With regard to the M9, I know of no one other than Mr. Puts who thinks that the M9 is anything other than stellar, and far superior to the A900 or other 24MP DSLRs. Not even close.

BTW; readers should not infer from this that these differences are huge. They're not. They're subtile, and possibly not even obvious to someone not looking for them. At this level of performance we are talking about small and incremental differences that may not be critical to all users. To some though, they will.

Michael,

these two statements seem to me to be contradictory; either the M9 is "far superior" or the differences are "small and incremental", and may not be obvious. I'm confused.  

[fredjeang]

Apparent contradictions generally reveal big truth.

I like those.

In fact I have the feeling that Michael tells at the same time the truth, and then put a wised calming pill knowing how can react the forums.  

I still have a lot to learn in my english but I'm generally not too bad reading in between lines, so my interpretation (don't take it too seriously) is:

-1) The truth: With regard to the M9, I know of no one other than Mr. Puts who thinks that the M9 is anything other than stellar, and far superior to the A900 or other 24MP DSLRs. Not even close.
Ps. I own and regularly use both an M9 and an A900 and can tell you that as good as the A900 is, the M9 smokes it in IQ. Similarly when it comes to lenses. The Sony / Zeiss lenses are excellent, but not in the same league as Leica's M glass.

-2) The Calming pill or joker:  readers should not infer from this that these differences are huge. They're not. They're subtile, and possibly not even obvious to someone not looking for them. At this level of performance we are talking about small and incremental differences that may not be critical to all users. To some though, they will.

And a totally personal interpretation of what could have been the Michael's post if politically uncorrect:

Puts is a mess when it comes to gear testing and what he should really do is keeping testing lenses instead or go out and have a beer.
With regard to the silly M9 evaluation, I've been walking on this gear planet enough to tell you that this Leica is not for the kids but for the real men and smokes any other dslr to date,  
and I know you guys are not going to beleive me anyway, even if Ansel Adams could confirm that fact from where he rests, so as I'm cool and relax here in Mexico and I don't want any forum hassle,
I'm ready to admit that those differences are minimum and in the end just go to the bloody store and try by yourself.

Again, don't take it seriously, Im kidding...or, Am I telling the truth?

[michael]

Fred,

You are now officially hired as my interpreter.  

Michael

[RobSaecker]

Fred,

You are now officially hired as my interpreter.  

Michael

Fred,

just make sure he pays you well. He can afford it.  

[Guy Mancuso]

Apparent contradictions generally reveal big truth.

I like those.

In fact I have the feeling that Michael tells at the same time the truth, and then put a wised calming pill knowing how can react the forums.  

I still have a lot to learn in my english but I'm generally not too bad reading in between lines, so my interpretation (don't take it too seriously) is:

-1) The truth: With regard to the M9, I know of no one other than Mr. Puts who thinks that the M9 is anything other than stellar, and far superior to the A900 or other 24MP DSLRs. Not even close.
Ps. I own and regularly use both an M9 and an A900 and can tell you that as good as the A900 is, the M9 smokes it in IQ. Similarly when it comes to lenses. The Sony / Zeiss lenses are excellent, but not in the same league as Leica's M glass.

-2) The Calming pill or joker:  readers should not infer from this that these differences are huge. They're not. They're subtile, and possibly not even obvious to someone not looking for them. At this level of performance we are talking about small and incremental differences that may not be critical to all users. To some though, they will.

And a totally personal interpretation of what could have been the Michael's post if politically uncorrect:

Puts is a mess when it comes to gear testing and what he should really do is keeping testing lenses instead or go out and have a beer.
With regard to the silly M9 evaluation, I've been walking on this gear planet enough to tell you that this Leica is not for the kids but for the real men and smokes any other dslr to date,  
and I know you guys are not going to beleive me anyway, even if Ansel Adams could confirm that fact from where he rests, so as I'm cool and relax here in Mexico and I don't want any forum hassle,
I'm ready to admit that those differences are minimum and in the end just go to the bloody store and try by yourself.

Again, don't take it seriously, Im kidding...or, Am I telling the truth?

Fred have to admit that was pretty well said and to be politically correct I won't say it but just give it a color. LOL

I tested the M9 as well and it does smoke anything out there under the 35mm blanket. No AA filter and some great glass will do it every time in the 35mm arena. End of story but on the same hand the S2 and the P40+ test I would grab the P40+ any day over it for Professional use. The S2 just needs to age a whole bunch before a lot of Pro's put there neck out on it. We just need a COMPLETE system top to bottom and that includes dedicated software.

[micek]

Another poster on this forum is Eleanor Brown whom I also thrust

[ErikKaffehr]

Yeah,

She shoots Phase One (P45 ?) and makes very good pictures. She certainly qualified to have an opinion about the M9 and she certainly likes it.

Best regards
Erik

[Rob C]

Fred,

You are now officially hired as my interpreter.  

Michael

This is a worrying trend, Michael.

What with taking yourself off to lotus-eat, indulge in local tequila etc., it makes me nervous for the future of LuLa. Have you considered a line of succession in this connection?

I only half-jest. Taking the self out of the urban jungle does strange and unexpected things to the drive of the inner man... you could find he throws away his spiritual permiso de conducción. Be careful.

Rob C

[ErikKaffehr]

Very sorry!

Corrected...

[michael]

Rob,

Fear not. I was just in Lotus Land for two weeks and am now on my way back in dangerous Toronto, with its earthquakes, tornados and urban riots.

I won't be back in Mexico again 'till December, and between now and then there are a number of exciting developments coming to LuLa.

Michael

[Rob C]

Rob,

Fear not. I was just in Lotus Land for two weeks and am now on my way back in dangerous Toronto, with its earthquakes, tornados and urban riots.

I won't be back in Mexico again 'till December, and between now and then there are a number of exciting developments coming to LuLa.

Michael

Isn't normality refreshing?

;-)

Rob C

[Nick Rains]

We just need a COMPLETE system top to bottom and that includes dedicated software.

Agreed about the system, but what dedicated software do you need? It comes with LR3 and shoots DNG.

[Guy Mancuso]

I have gone over this more times than I can count but I actually quote myself here on a thread. Bottom line nothing will beat a dedicated back to a dedicated sensor software package. Hassy and Phase have this and anyone shooting either system that knows how to raw process will almost always turn to there dedicated software package to squeeze the best out of those files.

I was replying to someone but you get it what I am talking about here.

 Lets point out a few things almost every cam has dedicated software even Nikon and Canon. Hassy and Phase do within there programs. Now my reasoning is very simple and i know this all too well. Just for example bring a Phase file in LR and it is not very good at all and the same with Hassy files will always be better in Phocus. The point here is the software is finely tuned and dedicated to that sensor. This takes in many things that simply are overlooked or other programs can't do. It is not just about a better profile but a completely written program that squeezes every drop out of those files. This is in regard to tonal range, sharpness, noise, DR, color, color tone, corrections if needed, exif data, tethering, color editors, styles and the list goes on. Secondly if your going to spend that kind of money and all your competitors are providing dedicated software than they will have the advantage. Trust me there is nothing special whatsoever about a DNG file. It was a good concept from Adobe but almost all OEM's said sorry to bad we want our own and we want to put our own special sauce in our files.

Now sure one can say it is a open format and any program can process it but it will never ever be as good as a dedicated software program BUILT for that cam and there algorithms and firmware. Example just noise control alone my Phase files is Sooooooooo much better in C1 than in LR or any other program. To me this kind of reasoning is on a Professional level that takes us past the general processing routines and profiles made for cams. Honestly IMHO this is where men go to work and the boys go play.

Also don't you think you deserve a dedicated software to squeeze every drop out of the money you spent on it when everyone around you has there dedicated software. Maybe better said in all this it uses generic raw processing. Also from what I seen and by your own tests barrel distortion and light falloff and you don't have corrections for that when the OEM says this stuff is built into the design. Maybe it is and i believe they did BUT I did not see that.

Just look at the M9 which IMHO looks far better from a C1 file than a LR files. Not saying C1 is the king of the roast but if your seeing differences coming from different programs than your not getting consistent or tuned images from the sensor. This is what Phocus for Hassy and C1 for Phase, Leaf Capture for Leaf, Sinar not sure of the name are all about. Even Nikon people see differences with Nikon Capture over other programs and Canon red(Color) has always been better out of DPP than any other program. Although this may have changed.

Now lets be clear I love Leica products and owned many but I have always seen different quality coming from different software packages and after extensive testing the S2 is no different just like any other cam. For Leica to succeed with this S2 program they can't be like any other cam out there and be generic. Anyway I agree we certainly may have different opinions on this which is great but I will not buy into this system if I can't squeeze every drop out of that sensor with processing. Everything so far I have seen is a compromise between DR, Color artifacts , over sharpening, color , tone etc etc. Sure you can get one program and get the best you can think out of it but do you really know what you might be leaving behind.
__________________