Precisely, the K-x sensor is better at high iso they don't need to apply much noise reduction
It's not a big difference, but it is enough to Dpreview give 9.0 in image quality, others have 8.5
--
--
The experts seem to agree that at 3200 and higher the K-x is applying some software NR to its raw files. This was the case for almost all previous Pentax cameras (at least starting from the K10D) and starting at varying ISO (as low as 800 on cameras based on the 10.2MP CCD). Fortunately the NR applied is rather light.
That being said, when you inspect the results visually (as opposed to numerically for DxO) you can see how this affects noise AND details. The way Dpreview presented the results, the K-x seems to be a big winner IMHO, NR or not.
--
Manu
http://flickr.com/photos/ensh/
Réflex Pentax: http://www.flickr.com/groups/pentaxfr/
My PPG: http://www.pentaxphotogallery.com/home#section=ARTIST&subSection=1312871&subSubSection=3929608
Personal for me but, the lack of visible AF points would be very off putting for me to use, and the AA batteries are also (for me) a big negative. Plus the unit is a bit on the small side for me. I would also add the inability to turn off in camera NR completely as a potential deal breaker (say for astrophotographers). All minor (and easily fixable) things, just a shame to even be able to list them.
Honestly I'd prefer it that way too but unfortunately Pentax doesn't care about it (and maybe they are right from a marketing point of view). But I got over it because the NR is fortunately not intrusive enough to be bothersome (like for example it was for the Sony A700 pre-v4 firmware).
Yes, but we aren't all vampires running around at night just because all SLRs have good low ISO performance are we? IF low ISO is no different across the vendors, then comparing IQ is rather pointless. I would wager very few people venture up into 3200 ISO land. This isn't to poo-poo having the functionality, but it's like ranking cars based on how many colors you can choose for the interior. Important to some, but to the vast majority of people completely and utterly irrelevant.
Well some are more raw than others. I'd like to see how the K-x engine does in some real world situations before I bless the approach entirely. And frankly, I don't even care if they ARE doing in camera raw NR, but I do so very much wish that:
Honestly I'd prefer it that way too but unfortunately Pentax doesn't care about it (and maybe they are right from a marketing point of view). But I got over it because the NR is fortunately not intrusive enough to be bothersome (like for example it was for the Sony A700 pre-v4 firmware).
--
Manu
http://flickr.com/photos/ensh/
Réflex Pentax: http://www.flickr.com/groups/pentaxfr/
My PPG: http://www.pentaxphotogallery.com/home#section=ARTIST&subSection=1312871&subSubSection=3929608
AA and compactness are personal preferences. I like these features myself.
AA and compactness are personal preferences. I like these features myself.
The lack of visible AF points can be bothersome to some extend. I also use the K-7 which has the same AF layout but with visual indications. In use it's not so much a big deal it appears to be, you get used to it pretty quickly IMHO but again this could differ from one person to the other.
NR can be turned off up to ISO 3200. I'm not familiar with astrophotography but I'd guess you would rarely use ISO 3200 and up for this use.
--
Manu
http://flickr.com/photos/ensh/
Réflex Pentax: http://www.flickr.com/groups/pentaxfr/
My PPG: http://www.pentaxphotogallery.com/home#section=ARTIST&subSection=1312871&subSubSection=3929608
I tend to disagree. The main reason most people don't consider ISO 3200 as useful is because on most cameras it is not usable . For indoors photography (which is a common need for most people), high-ISO is a far better choice than direct flash, and in many cases even better than bounce flash. The more high-ISO becomes usable, the more people find it useful. I bet nowadays many users are just unaware of high-ISO being such a good alternative to flash. This is probably true for people who don't frequent websites such as dpreview and base their knowledge on experience with previous cameras, sometimes small-sensor cameras.
True, but how can you tell? The EOS-7D provides better high-ISO performance than the EOS-50D, but how much of it can be attributed to a better sensor, and how much of it can be attributed to better "cooking" of the RAW files? Without inside information, there's no way to know.
I would too, but since this isn't likely to happen then it's important to note that whatever Pentax does to the RAW files, it's doesn't result in worse files than what one can get from the competition. And as stated, there's no way to know what the competition does to their RAW files, just how good the final result is.
This would be great, but since other vendors don't do it, don't expect Pentax to.
--
Fair enough but it maybe just a bit far fetched to say that Pentax "f-ed up".Oh I agree many of my bitches are personal, but they are what they are
From what I could understand, Pentax did it for space and cost savings. The body is the same as the K-m, they apparently could not add the AF points in the viewfinder without redesigning at least part of the body.
Odd that the computer nerd (me) is the old schooler here
. A bounce flash and iso 1600 almost do me well enough for indoor shooting. The exception is those times when you a) can't bounce and b) can't flash period. As I said it's nice to have, but not the be all and end all some make it out to be (and that's my real issue with everyone fixating on it, when 90% of shots I see (and take) are between ISO 100 and 400, with another 5% being 800. I haven't seen any of those shots needing more ISO.Well with a little knowledge you can guess the extent of the cooking that goes on. Harder with Canon since they are the only ones that use their sensors, and some Nikon's too. But for 'off the shelf' sensors you can guess.
Not disagreeing at all, great output from the little guy (and I've never said otherwise)
This wasn't just aimed at Pentax. That is an 'across the board' comment, sorry if I wasn't clear in the intent.
To me they are significant enough to fall into the f-ed up category. Again, it's personal obviously, but combined they are deal breakers for me. Having said that I would rather Pentax fixed their distribution and support network more than any of those features.Fair enough but it maybe just a bit far fetched to say that Pentax "f-ed up".Oh I agree many of my bitches are personal, but they are what they are
I wouldn't have thought a little light would have been that hard. The 7D, I believe, has a transparent LCD on the VF, which results in almost no space used, and the cost is probably pretty darned insignificant. Agree on a second dial too. Nikon's lack of drive motor is annoying and I think that will fade in importance over time (I see Nikon moving to lens driven AF across the board eventually). As for Canon they finally stopped crippling spot metering in their amateur cameras.
I like using bounce flash, but it has limitations. For one, if you bounce the light backwards the light on the subject is flattering, but the background is too dark (especially in a large room). If you bounce it forward and use a reflector card (or another small attachment that is far smaller than a wall), then the light on the background may look nice, but the subject is going to be harshly lit. Also, the flash removes the ambient light, which can actually look nicer and/or more interesting than bounce flash light.Odd that the computer nerd (me) is the old schooler here
Try using f/4.5 and ISO 800 indoors and you'll see that the shutter speeds you get are simply not fast enough for anything that moves. Heck, even for shooting static subjects it's a problem without IS.
So what sensors do the K-x, A500 and D90 use? Do you know for sure that it's the same sensor? Can you prove that it hasn't been modified to fit specific requirements of Pentax, Nikon and Sony? Can you tell what affect do the surrounding electronic circuits have over the results? These are obviously different between different cameras, and they can affect the results as much as different RAW cooking can, see the D3X vs. A900 differences.
Agreed, but that's where your glass choice comes in. Personally I would rather have the faster glass for other reasons besides low-light. To be clear, I'm not suggesting people who shoot a lot in doors should be using kit lenses are 4.5 or 5.6, etc. Just as the right camera is part of the equation, so are the lenses on the front.
Well I can pretty much guarantee that the sensors are not custom modified for the customer. Coming from a background in IC design and fabrication the costs involved would be enormous and the volumes far to small to warrant that. And yes there are other factors, but that's why I said guess, and not know .
There are two problems with this suggestion:
Does that guarantee/guess also stand for the difference between the D3X and the A900 not being related to a different sensor? Do you believe it's only different "cooking" of the RAW file?
Well I'm a believer in right tool for the job. I see your point and I don't disagree with it on some level, but also I look at things like the 50mm 1.8 that Canon has.. dirt cheap and great for those situations. Canon is probably the low light price/performance leader if your look at an entire system price. However, again, the new Pentax unit is a great camera for low-light. Not arguing (nor have I ever argued) that on bit, and well done to Pentax for it.
I would attribute some of it to overall cleanliness of the signal readout lines and general shielding, grounding, heat control etc. And some of it I would attributed to processors (i.e. cooking). Sony's sensor guys are better than their SLR guys IMHO. If you want I can really take a look at the output of both and tell you what I think, but I've not more than casually glanced at either.
I think that the well regarded Tamron 17-50mm f/2.8 is now available in stabilized versions for Canon and Nikon. A bit more expensive than the non-VC version but still much cheaper than the OEM offers.
This lens is 80mm equiv, so it may be useful for some portrait work, but it isn't flexible enough for much else.
I don't think Canon is the "low light price/performance leader". Pentax fast lenses aren't as inexpensive as they used to be a year ago (the 50/1.4, 16-50/2.8 and 50-135/2.8 are much pricier than they used to be), but (1) Canon equivalent lenses are just as expensive, and (2) there are some very good third party lenses that become stabilized on the K-x and further extend the low-light lead of this camera. Some examples: Sigma 20/1.8, 24/1.8, 28/1.8, 30/1.4, 50/1.4, 70-200/2.8 and Tamron 17-50/2.8 and 70-200/2.8.
Sure, I'd be very interested to read your comments about this. I believe imaging-resource.com offers RAW files for download for both cameras, taken under the same lighting and settings.
