Sony A700 replacement - SLT, 25MP APS-C, >100,000 ISO?

Soldato
Joined
10 Feb 2010
Posts
3,248
Saw this on TP, and while most of us are Canon heads, this looks like Sony could really be pushing with this and figured most of you would be interested :)

http://www.amateurphotographer.co.u...ent_at_CES__more_information_news_304848.html

A lot is speculation, so I'd be very surprised if a 25MP SLT camera on APS-C could get anything other than chroma noise at D3s/1DMk4 sensitivities.

However it /is/ Sony, and they certainly have the money to back an underpriced product like this in exchange for really breaking into the market - something that technology alone is unlikely to do... Wishful thinking maybe, but I'll wait until I hear more and see samples before I kill off the dream ;)
 
Actually... With a higher-end processor and developments in their SLT technology... The handheld twilight could be expanded to general noise reduction using ridiculously high fps bursts to shoot at say 3 1/4000 shots, 5/4000?
 
I was just about to drop £700 on two Sigma Nikon mount primes but I'll give those a wait for now and just hire them if need be - funds may well be directed towards this, depending on how much of this hype it lives up to :)
 
The more competition the better, pushes down the price of the kit that I actually want.

Have to say that I'm a little concerned that they're still pushing the "more pixels is better" mantra, as other groups will feel obliged to follow.
 
What an utterly stupid sensor. 12Mp is the sweet spot on DX, going to 16 adds very little.
Considering on the Nikon D3x full frame 24MP sensor, only a handful of the best primes stopped down are ever close enough to out-resolve the sensor, 24MP on DX will mean there is not a single practical lens in existence that wouldn't suffer from diffraction limitation.

Heck, 6MP is already the theoretical limit for DX sensor, f/11, green light. The gains above this are much more modest, usually resolving artifacts from the colour array and AA filters etc. Going from 6MP of my D70 with weak AA filter to 12MP with a stronger AA filter certainly did not double the signal resolution, and the linear resolution obviously much less.

As long as Nikon doesn't follow the stupid MP war of Canon and Sony.
 
Is a higher MP count detrimental to IQ though? If not i guess its no biggie, and is a marketting gimmick without consequence (i dont think it would be any cheaper if they were to drop a few MP from the sensor in real terms).
 
Is a higher MP count detrimental to IQ though? If not i guess its no biggie, and is a marketting gimmick without consequence (i dont think it would be any cheaper if they were to drop a few MP from the sensor in real terms).

Very much so, lower pixel density on the sensor means much less noise and improved low light performance given equivalent sensor technology. That's why the D3s has substantially better high ISO performance than D3x...

The claim of that kind of ISO values at such a high resolution on DX would mean either a massive advance in sensor and processing technology or terrible image quality basically. Or they're cheating on those numbers...
 
Is a higher MP count detrimental to IQ though? If not i guess its no biggie, and is a marketting gimmick without consequence (i dont think it would be any cheaper if they were to drop a few MP from the sensor in real terms).

But it is detrimental to image quality! Let alone functionality
 
Very much so, lower pixel density on the sensor means much less noise and improved low light performance given equivalent sensor technology. That's why the D3s has substantially better high ISO performance than D3x...

The claim of that kind of ISO values at such a high resolution on DX would mean either a massive advance in sensor and processing technology or terrible image quality basically. Or they're cheating on those numbers...

Those ISO numbers wont be base ISO but a boosted version, probably the sensor operates to ISO3200 and the rest are boosted values (Hi1,Hi2). Furthermore, it is probably pixel-binned.
 
Those ISO numbers wont be base ISO but a boosted version, probably the sensor operates to ISO3200 and the rest are boosted values (Hi1,Hi2). Furthermore, it is probably pixel-binned.

Actually, I think it'll be worse still - the same way the A55 uses a multi shot NR to fake ISO25600.

It might be effective but it's not ISO25600 and it's misleading to suggest it is as you loose other capabilities (burst mode speed, shooting in RAW) to shoot at that ISO...
 
The resolution is just a rumour, mind; the only thing a Sony employee has confirmed has been the ISO numbers, aside from the mag alloy, price point etc.
 
Back
Top Bottom