maemo.org - Talk

maemo.org - Talk (https://talk.maemo.org/index.php)
-   Maemo 5 / Fremantle (https://talk.maemo.org/forumdisplay.php?f=40)
-   -   [Closed - Leaked PR1.2] Info and Discussion thread (https://talk.maemo.org/showthread.php?t=51713)

ndi 2010-05-23 00:37

Re: [Leaked PR1.2 RC] Info and Discussion thread
 
Could you take a pic with same settings but outside in full sun? We could run a comparison.

lucas777 2010-05-23 01:07

Re: [Leaked PR1.2 RC] Info and Discussion thread
 
Heres when i take a pic outside with pr.1.2 leaked

http://img36.imageshack.us/img36/1543/20100521001.jpg

lemmyslender 2010-05-23 01:18

Re: [Leaked PR1.2 RC] Info and Discussion thread
 
Quote:

Originally Posted by ndi (Post 670990)
PR 1.2 updates radio modem so reverting might let you phoneless(, or so I've read). If W7 beta flashed my MB BIOS and it would no longer run XP, that beta would be laughed off the stage.

A beta FW would be OK as long as it would be available OTA and as long as a button in the menu could do it back to 1.1.1 OTA. As in, you know, "revertable".

I'm still going to wait because so many people complain, and I simply can't be without my N900.

That beta might b laughed off the stage, but there'd still be people that did it. If I wasn't planning on upgrading to 1.2, I'd be worried.

If a beta was OTA, everyone would flash and you'd get all sorts of complaints and bad press. If you have to flash it, far fewer people will flash and you get a lot less complaints from people who shouldn't be beta testing.

In a real beta, I'm sure re-flashing the modem firmware would be possible.

Fact is Nokia missed out on a chane to get another "300" and improve their product. I believe it's a matter of too few people left working on it. I'm running it, haven't had a single problem, and I love it.

deny_winarto 2010-05-23 02:04

Re: [Leaked PR1.2 RC] Info and Discussion thread
 
Quote:

Originally Posted by jflatt (Post 670089)
EXIF data shows digital zoom ratio 2.8. Is that right? Surely digital zoom will trash the image

Hmm.. yes but i don't think thats the problem, i've taken numerous shot without zooming and the results were about the same.. I'll upload the one without zooming later

Quote:

Originally Posted by nikolag (Post 671043)
Similar results here. This is most likely a software optimization issue since my E72 takes uncomparably better photos and it doesn't even have Zeiss lens. There is too much noise in the photos from the N900 even in perfect outdoor lighting. I really don't see any improvements in the camera over PR1.1. The night mode takes the same really bad photos.

Even my N73 takes better pics than this :(
Oh man, so no improvements? It could be hardware issue though because the result is ALOT BETTER for outdoor pics, it seems like there's something wrong with the way N900 handles lighting..
I'm gonna take it to Nokia Center tommorow and see what they think about it..

Quote:

Originally Posted by olighak (Post 670609)
I see a hint of a blue-greenish hue on this image as well as the digital zoom (which will make it look bad on any device).

The only fix to a blue-greenish hue is to buy a black permanent marker and take off the battery cover. Apply the permanent marker on the blue-green edge of the lens cover which is closest to the lens when taking pictures. You might want to mask the chrome frame with paint tape or similar to avoid marking that black as well.

Once dry, a few seconds after appliation, your pictures will be void of this blue-green hue and will look much better.

I'll try this, but i'm gonna use small black tape instead..
To be honest i'm quite pessimist because some people already tried this on other thread and it makes no difference..

Quote:

Originally Posted by lucas777 (Post 671066)
Heres when i take a pic outside with pr.1.2 leaked

http://img36.imageshack.us/img36/1543/20100521001.jpg

This is what i was talking about, the pics taken is better outdoor..
Could you take indoor pics at night for comparison? Inside a room with good lighting, just like my case?

deny_winarto 2010-05-23 02:12

Re: [Leaked PR1.2 RC] Info and Discussion thread
 
Quote:

Originally Posted by ndi (Post 671050)
Could you take a pic with same settings but outside in full sun? We could run a comparison.

Here's one i just took outside my house :

http://tinypic.com/m/ab181w/2

Same settings, autofocus, macro mode and hi res.
See the difference?

lucas777 2010-05-23 02:49

Re: [Leaked PR1.2 RC] Info and Discussion thread
 
yeah when i get home i will take some inside.

Psymastr 2010-05-23 18:51

Re: [Leaked PR1.2 RC] Info and Discussion thread
 
So, its almost june... when will this 1.2 become official?

casper27 2010-05-23 18:56

Re: [Leaked PR1.2 RC] Info and Discussion thread
 
Quote:

Originally Posted by Psymastr (Post 672008)
So, its almost june... when will this 1.2 become official?

The leaked RC version that is been discussed here will never become official. It has been superceded already.
The official pr1.2 should have about 4 weeks extra bug fixes in but there is also speculation of a PR1.2/Meego release very soon.
Might even be one and the same ;)

ndi 2010-05-24 01:50

Re: [Leaked PR1.2 RC] Info and Discussion thread
 
Quote:

Originally Posted by deny_winarto (Post 671091)
Here's one i just took outside my house :

http://tinypic.com/m/ab181w/2

Same settings, autofocus, macro mode and hi res.
See the difference?

I get the same quality as you do here. One can still see the poor CMOS sensor artifacts and JPEG compression, but it's much lower. Bug in post-processing?

Ok, I'll say it. We have a pretty bad camera. I also suspect some upscaling, as it's distinctly lacking in detail, knowing it has optics and a 5 MPx sensor. Reducing images on the PC I go below 3MPx for a clear, sharp image. And about 1.5MPx for beautiful. I'm not saying it's Nokia, but it sure seems like someone upscales.

I did a side-by-side comparison to some pics I took with my old N80, pinhole 3.2 MPx, later admitted to be upscaled from 2 MPx IIRC. They are strikingly similar. Gain is small. Also, the "smaller" N80 sensor also has less noise in good light (more in low).

Ever since I bought a DSLR and some glass I've begun to see defects I never noticed before in photos. It is my personal opinion, as an amateur photographer, that N900 either cheats (upscaling/post processing - median noise) or has some sort of processing bug. And it seems to be worse in 1.2 (still on 1.1.1 here). Not to mention the flash myself in the eye design of the case. Also, the compression has some kind of problem. Fitting 5 MPx in 1 MB in a second without the aid of a nitrogen-cooled quad core means that losses are going to be quite high.

Oh well. Anyone have a test pattern handy?

deny_winarto 2010-05-24 03:45

Re: [Leaked PR1.2 RC] Info and Discussion thread
 
Quote:

Originally Posted by ndi (Post 672525)
I get the same quality as you do here. One can still see the poor CMOS sensor artifacts and JPEG compression, but it's much lower. Bug in post-processing?

Ok, I'll say it. We have a pretty bad camera. I also suspect some upscaling, as it's distinctly lacking in detail, knowing it has optics and a 5 MPx sensor. Reducing images on the PC I go below 3MPx for a clear, sharp image. And about 1.5MPx for beautiful. I'm not saying it's Nokia, but it sure seems like someone upscales.

I did a side-by-side comparison to some pics I took with my old N80, pinhole 3.2 MPx, later admitted to be upscaled from 2 MPx IIRC. They are strikingly similar. Gain is small. Also, the "smaller" N80 sensor also has less noise in good light (more in low).

Ever since I bought a DSLR and some glass I've begun to see defects I never noticed before in photos. It is my personal opinion, as an amateur photographer, that N900 either cheats (upscaling/post processing - median noise) or has some sort of processing bug. And it seems to be worse in 1.2 (still on 1.1.1 here). Not to mention the flash myself in the eye design of the case. Also, the compression has some kind of problem. Fitting 5 MPx in 1 MB in a second without the aid of a nitrogen-cooled quad core means that losses are going to be quite high.

Oh well. Anyone have a test pattern handy?

Sorry for being a noob, but what's upscaling? :p
Is it optimizing image result?
So if it is indeed causing problem,
maybe we can propose them to make another camera program to solve this problem..

I think we need opinions from digital camera expert,
especially the one that knows the structure & coding of N series camera,
that way we can determine the exact cause of this problem..

Being a normal user, i can only hope this is software problem,
not a hardware problem, because it'd be harder to fix :(
Hopefully this is fixed in the "official PR 1.2" :)

btw i've tried removing the back cover,
i thought it'd better than marking the lens cover,
still get same result though :(


All times are GMT. The time now is 18:18.

vBulletin® Version 3.8.8