[falconeye]

Ok, I know it is off topic here. But it may be of interest for the DSLR users.

German "digital lifestyle" magazine SFT (they do DSLR tests as well) featured the

Pentax Optio A30 (second, black edition)

as follows: "outclasses entire P&S competition" (editor's choice). With:


PRO: Stunning IQ -- In ISO 100-200 best image quality in its class only rivaled by a good DSLR with good glass!

CON: Slow AF


Sounds familiar

Are those findings in line with common beliefs? Does Pentax have the same reputation in P&S as in DSLR (good IQ but slow AF)? Is A40 even better IQ wise?

[*isteve]

Ok, I know it is off topic here. But it may be of interest for the DSLR users.

German "digital lifestyle" magazine SFT (they do DSLR tests as well) featured the

Pentax Optio A30 (second, black edition)

as follows: "outclasses entire P&S competition" (editor's choice). With:


PRO: Stunning IQ -- In ISO 100-200 best image quality in its class only rivaled by a good DSLR with good glass!

CON: Slow AF


Sounds familiar

Are those findings in line with common beliefs? Does Pentax have the same reputation in P&S as in DSLR (good IQ but slow AF)? Is A40 even better IQ wise?

Perhaps one of the reasons why the IQ is so good is because they take the time to focus properly?

I have an A30 - its not that slow unless you use wide area of face recognition. On spot AF its quite fast.

In fact if you compare it to any Nikon P&S is blindingly fast!

[k100d]

how about high iso?

[mermozjg]

PRO: Stunning IQ -- In ISO 100-200 best image quality in its class only rivaled by a good DSLR with good glass!

excellent IQ at low ISO is possible with a tiny sensor. FF will not give you better IQ quality at low ISO than APS-C , only 1 stop better noise profile and swallower DOF.
"quod erat demonstrandum"

[Deni]

how about high iso?

I'm not sure about current cameras but nothing could touch the fuji's f20/30/31 high iso performance.

Like everybody else fuji got into the mp wagon and high iso performance is not the same anymore.

[*isteve]

I'm not sure about current cameras but nothing could touch the fuji's f20/30/31 high iso performance.

Like everybody else fuji got into the mp wagon and high iso performance is not the same anymore.

Possibly not, but the A30 pulled far more detail out of a shot than the Fuji's at low ISO and up to about ISO 200/400. So much so I sold my F30 and bought an A30, and I am very pleased with it. On the other hand if you want a camera to take pictures at parties and post them on the internet, get the Fuji.

[falconeye]

Actually, what got me triggered, wasn't that a P&S can do well at low ISO.

It was that they picked the A30 and no other (so, not even a A40) which stands out of the P&S so much that you won't be able to tell the difference to DSLR from a low iso image.

So, I was wondering if anybody ever did a A30/K10D ISO100 image comparison at 100% (both are 10 MP)?

[24X36NOW]

excellent IQ at low ISO is possible with a tiny sensor. FF will not give you better IQ quality at low ISO than APS-C , only 1 stop better noise profile and swallower DOF.
"quod erat demonstrandum"

Nonsense; a practical demonstration showing better IQ even at ISO 100 is here:

The Full-Frame Advantage

Keep drinking the KoolAid, if you like; to each his own.

[*isteve]

Actually, what got me triggered, wasn't that a P&S can do well at low ISO.

It was that they picked the A30 and no other (so, not even a A40) which stands out of the P&S so much that you won't be able to tell the difference to DSLR from a low iso image.

So, I was wondering if anybody ever did a A30/K10D ISO100 image comparison at 100% (both are 10 MP)?

I could but the weather is terrible so give me a few days. The A30 pulls some impressive resolution at the centre its true and the colour is very good indeed.

[mermozjg]

Nonsense; a practical demonstration showing better IQ even at ISO 100 is here:

The Full-Frame Advantage

Keep drinking the KoolAid, if you like; to each his own.

You don't have another reference ?
KR is well known for HIS nonsense. KoolAid will do better than KR's recommendations.

[falconeye]

I could but the weather is terrible so give me a few days.

Every time you need The results will be worth the wait, I'm sure.
Ideally, same effective 35mm length and same subject. For both, a 20% full size and 100% crop, please (and not telling us which is which...)


And this is not meant to be another "why bigger is better" thread. Please hurry back to the corresponding FF threads, please

[*isteve]

Every time you need The results will be worth the wait, I'm sure.
Ideally, same effective 35mm length and same subject. For both, a 20% full size and 100% crop, please (and not telling us which is which...)


And this is not meant to be another "why bigger is better" thread. Please hurry back to the corresponding FF threads, please

OK but I will have to disguise the image format by cropping both to the same dimensions, so what I will do is make a collage so you see parts of the pic side by side at 25% and 100%. I will include a shot of the "big picture" for reference

Frankly I have no idea how they will compare. I know the A30 is softer round the edges, but then so are some lenses on the K10D. I will use my FA 24-90 on the K10D as this is nearest to the equivalent focal length of the A30 and it represents a resonable mid quality Pentax zoom.

Tomorrow looks possible.

[Geekybiker]

You don't have another reference ?
KR is well known for HIS nonsense. KoolAid will do better than KR's recommendations.

Okay, KR isn't the best, but the idea is right. A larger sensor can use a poorer lens and achieve better detail within limits. The LP/mm of the lens must exceed the LP/mm of the sensor.

Noise aside, sensor size does matter once you get high enough MP. Small sensors will get diffraction limited much quicker.

[*isteve]

Okay, KR isn't the best, but the idea is right. A larger sensor can use a poorer lens and achieve better detail within limits. The LP/mm of the lens must exceed the LP/mm of the sensor.

Noise aside, sensor size does matter once you get high enough MP. Small sensors will get diffraction limited much quicker.

Do you know the formula for the resolution of lens + sensor?

Actually you have it the wrong way round. If the lens can resolve more than the sensor, you get moire and need an AA filter, further reducing resolution. If you have more MP on the sensor you can use a weaker AA filter. Double bonus points.

Also even if diffraction affects you at wider F stops, you have much greater DOF so you dont need to stop down as much. On a 645 camera you regularly have to shoot landscapes at F16 or F32. They only look sharp because a big negative needs less enlargement, but in terms of LPMM they were anything but. With a sensor, a 12MP sensor will provide the same resolution on an A3 print whatever the sensor size.

True on a very tiny sensor, diffraction limits the absolute resolution but looking at a Canon G9 its still resolving more than a G7 with the same lens or for that matter, the same as a Canon 40D with a 50mm prime lens!

G9
Horiz Absolute 1950 Extinction 2400 (with Moire)
Vertl Absolute 1900 Extinction 2400 (with Moire)

G7
Horiz Absolute 1775 Extinction 2225 (with Moire)
Vertl Absolute 1850 Extinction 2350 (with Moire)

40D
Horiz Absolute 2100 Extinction 2300 (no Moire)
Vertl Absolute 1800 Extinction 2300 (no Moire)

So, apart from some moire artifacts, I doubt you would be able to tell the difference between a well exposed crop of a 40D shot (10MP APS) and a G9 shot (12MP digicam) and if you put a cheap old zoom lens on the 40D you may even get them the wrong way round.

[Geekybiker]

Do you know the formula for the resolution of lens + sensor?

Actually you have it the wrong way round. If the lens can resolve more than the sensor, you get moire and need an AA filter, further reducing resolution. If you have more MP on the sensor you can use a weaker AA filter. Double bonus points.

Not offhand.

I think you misunderstand me. I was trying to say that with the same lens, a larger sensor will support more resolution (in MP) than a small sensor. IE say you have a fantastic lens that resolves 2400 lp/mm. If you built a sensor around that lens you'd want to match that lp/mm for best results from what you said about AA filters. The APS-C sensor would of course end up with more resolution. (Alot of other things would change like FOV, etc)

I'm probably not expressing it best. But larger sensor = greater theoretical resolution anyways.

Awhile ago I did the math and a 1/1.8" sensor from a PS would support about a max of around 12mp at f2.8 (That close to wide open on most PS, right?) Assuming my math was right we're close to the limit for PS. Awhile ago I read that the max on aps-c is around 32mp (not my math so I'm not sure what they used as a basis)