Douglas F Watt
Senior Member
Lots of high expectations for this first FF camera from Sony in 4 years . . . Does it live up? Well, based on my brief and limited experience with this (a week plus with my own camera) mostly yes (and in low light - YES), but there are some issues and puzzling design decisions, esp. around the lack of a decent hi-speed buffer for continuous shooting. Overall, this is an appealing choice for those looking to upgrade from Sony APS-C models or other subframe DSLRs to full frame cameras - if one is willing to pay the hefty cost and weight penalties, in order to get the extra picture quality at high ISO. Dynamic range appears to be somewhat better than corresponding APS C models - and appears to be very close to the very high benchmark set by the Nikon D800 (and its Sony sensor). Image quality though is consistently high for with little decline as ISO goes up, unlike the A65/77. For those coming from any current or recent Sony SLT, the menus, operating system and controls will feel immediately comfortable and very familiar, speeding the learning curve for this new camera. But . . . .
Is this really worth the $2000 premium on the already excellent Sony A65 or the nearly $1200 premium on the Sony A77? That's a tough question, and mostly for a typical consumer, the answer is going to be 'probably not' - but for professionals (or enthusiasts with lots of money that they are eager to part with) needing the high ISO capabilities of a full frame model, the answer is probably 'yes', as picture quality on APS-C chips can't compete with a full frame chip of the same resolution as light declines. But in bright light, the differences are just too subtle to see, even for the most obsessional pixel peeper. At ISO 100 in RAW with roughly comparable lenses set to mid aperture values, I can't tell the difference between a 24MP image this camera generates (even blown up to 100%), and what a Sony A65 can generate. But at ISO 1600, the difference is fairly clear, and by 3200, it's not even close - the A65 is a pretty noisy mess, and prints are only usable at small sizes like 4x6. At ISO 3200, the A99 is capable of generating remarkably clean and smooth pictures with minimal noise and much detail. ISO 6400 on the A99 is roughly equivalent to ISO 1600 on the A65 (the last really useable RAW setting for the A65 before being forced to use heavy-handed post-processing NR). So it is WAY better in low light - but it should be. Sports shooting will be worlds better compared to even the best m4/3 and APS-C sensor performance. But again, this is what a FF sensor should deliver for its cost (and weight) penalties.
Pros:
1) Very good video for a DSLR - and likely well ahead of Canon and Nikon on this one. Video shot in 60p is particularly impressive, and the camera maintains tight focus on high-speed subjects and challenging scenes with minimal hunting. Uncompressed HDMI output to external recording devices is another theoretical plus, although not one I will use (shared with D800 but not on the 5DMIII).
2) Great overall image quality for photos - esp. as light declines - equals the great low light performance of the Canon 5DMIII and just a touch more noisy than the Nikon D600. Two full F stops better at least than the A65/77 in terms of noise.
3) High-resolution EVF gives the photographer much relevant control information (but see cons for flip side of this).
4) High resolution 24 MP sensor with good to excellent dynamic range and very accurate color equaling or exceeding resolution of every 35 mm camera on the market save one (again see cons).
5) Able to use Alpha mount APS-C lenses (via sensor crop - but see cons again).
6) Roughly equal low light performance compared to its two main FF competitors. Even ISO 6400 is pretty clean and 12,800 is useable up to full page prints with modest post-processing.
7) Configurable menu system similar to Sony's popular A77 that is both deep and intuitive in its layout, fostering quick mastery. Excellent control/operating system structures, and lots of flexibility for the advanced user.
8) Having image stabilization built into the sensor instead of the lens makes lenses potentially less expensive and lighter, and giving you full IS with any lens in your bag.
9) SLT approach enables full-time live view and phase detection autofocus even when shooting movies, and saves weight too, compared to more conventional DSLR technology.
Cons:
1) Price - at least $400 above where it should be.
2) EVF is not for everybody - some people simply can't transition from optical viewfinders (what do they know
!)
3) EVF really needed a contrast adjustment on A77/65, as images sometimes either had blown highlights in the viewfinder or areas where image too dark to see details - Sony failed to include this adjustment in their new flagship camera - a major omission.
4) only 24 MP - clearly out-resolved by Nikon D800.
5) only 6 FPS with autofocus is disappointing for a SLT design - APS C designs were all class-leading in terms of FPS. Canon does 6 FPS number while still having to move the mirror up and down, and even the D800 manages 4 frames per second with its much higher megapixel throughput demands on imaging pipeline, while the $800 Sony A65 does 10 FPS . . . . so what happened here?
6) Sony APS-C lenses (but not 3rd party lenses) will automatically result in a significant drop in resolution - no option to simply allow full frame vignetting (which some users might not mind in some contexts). Only lenses with the 'AF-D' designation (not that many in the Sony stable) will get the full benefit of the sensor phase detection augmentation of autofocus.
7) Shorter battery life relative to its two main competitors (the cost of the excellent EVF)
8) Despite the flagship technology and price, the A99 still has too shallow a buffer for meaningful high-speed shooting - if shooting RAW plus JPEG, buffer fills up in 18 images or just 1.5 seconds. Can't believe that with all the noise about this issue in previous APS-C models that Sony did so little differently in their flagship camera.
9) USB 3.0, Wi-Fi support and built-in flash all omitted from feature set - no option currently even for add-on Wi-Fi adapter (could just get Wi-Fi SD card?). (OK - pro cameras usually don't have built-in flash . . . but this makes remote triggering of slave units difficult).
10) Weight, esp. with the best available mid-range zoom (the CZ 24-70 2.8) - this combo weighs in around 1.7 kilos (almost 4 lbs).
UPDATE OCTOBER 17th - The D600 now has studio test images on DPR, and this new Nikon FF model clearly (but only slightly) exceeds the Canon 5DMIII in low light ability, and is close to the Nikon D4 - suggesting that the D600 will be formidable competition for the A99 (the D600 has basically the same sensor, but the A99 is only about ½ F stop of poorer, due to the SLT design taking ~ 30% of sensor light). Low light ability is really the only major/substantive advantage that full frame pro cameras have over 24 MP APS-C models like the Sony A77/65 (putting aside the EVF/OVF debate). This unfortunately has only re-amplified my concern about the price point - the D600 is really impressive at least in its early returns, and, it's $700 less (and with a more complete pro-camera ecosystem around it). On the other hand, early returns suggest that Nikon has had some initial quality control problems with dust and perhaps other junk on the sensors.
November 24th Update
Having had a bit more time with the camera, I have some better sense of its strengths and weaknesses. I love the low light ability, but I am finding that it really isn't enough better outside of that area than an A65 for me to take it everywhere - and I find myself carting the moderately heavy A65 (with its 16-50 2.8 lens) around when I am willing to tote major hardware (over my Galaxy S II cellphone) instead of the REALLY heavy A99 (with the monster CZ 24-70 2.8) except when I know I will be shooting in low light. I find myself struggling with the curious concern about the camera that as great as it is, it just isn't THAT much better than the A65 - except again in low light - to feel wildly happy with it after coughing up $2800. Its video is only marginally more capable than the A65's, unless I am willing to add all kinds of pro accessories. So I'm finding myself wondering if I am going to get enough meaningful usage from this, if in good light, I find myself preferring the A65 with its faster shooting speed, identical EVF, and slightly more compact form factor/weight. Neither the A65/77 or A99 has an decent-sized buffer for real high speed shooting (which would have tipped the scales in favor of the A99), and I am still stunned that Sony (after all the flack voiced about this issue in relationship to the other SLTs) didn't fix this issue on the A99.
Overall, this is clearly a great camera but perhaps just not great enough for this much money? I have decided after much thought to send it back, and wait for the A109, which will have the extra resolution of the 36MP sensor, and probably an even better EVF - and I suspect Sony will find a way to make its low light ability roughly equal to the Nikon D800 while improving its speed and AF). In the meantime, I will live with the mediocre low light ability of the A65. For $2000 less than the A99, it seems like the better value even if it's not the better camera. Perhaps Sony has just set the bar so high with their subframe SLTs that they are competing not just with CaNikon, but now with themselves to some extent. I look forward to seeing what Sony does with their next FF model, and whether they can improve the low light performance of their next generation of subframe sensors. Even this generation of EVFs is (at least in my experience - most of the time) superior to even full frame optical finders - the next generation should be really something and give Canikon even more to think about.
Is this really worth the $2000 premium on the already excellent Sony A65 or the nearly $1200 premium on the Sony A77? That's a tough question, and mostly for a typical consumer, the answer is going to be 'probably not' - but for professionals (or enthusiasts with lots of money that they are eager to part with) needing the high ISO capabilities of a full frame model, the answer is probably 'yes', as picture quality on APS-C chips can't compete with a full frame chip of the same resolution as light declines. But in bright light, the differences are just too subtle to see, even for the most obsessional pixel peeper. At ISO 100 in RAW with roughly comparable lenses set to mid aperture values, I can't tell the difference between a 24MP image this camera generates (even blown up to 100%), and what a Sony A65 can generate. But at ISO 1600, the difference is fairly clear, and by 3200, it's not even close - the A65 is a pretty noisy mess, and prints are only usable at small sizes like 4x6. At ISO 3200, the A99 is capable of generating remarkably clean and smooth pictures with minimal noise and much detail. ISO 6400 on the A99 is roughly equivalent to ISO 1600 on the A65 (the last really useable RAW setting for the A65 before being forced to use heavy-handed post-processing NR). So it is WAY better in low light - but it should be. Sports shooting will be worlds better compared to even the best m4/3 and APS-C sensor performance. But again, this is what a FF sensor should deliver for its cost (and weight) penalties.
Pros:
1) Very good video for a DSLR - and likely well ahead of Canon and Nikon on this one. Video shot in 60p is particularly impressive, and the camera maintains tight focus on high-speed subjects and challenging scenes with minimal hunting. Uncompressed HDMI output to external recording devices is another theoretical plus, although not one I will use (shared with D800 but not on the 5DMIII).
2) Great overall image quality for photos - esp. as light declines - equals the great low light performance of the Canon 5DMIII and just a touch more noisy than the Nikon D600. Two full F stops better at least than the A65/77 in terms of noise.
3) High-resolution EVF gives the photographer much relevant control information (but see cons for flip side of this).
4) High resolution 24 MP sensor with good to excellent dynamic range and very accurate color equaling or exceeding resolution of every 35 mm camera on the market save one (again see cons).
5) Able to use Alpha mount APS-C lenses (via sensor crop - but see cons again).
6) Roughly equal low light performance compared to its two main FF competitors. Even ISO 6400 is pretty clean and 12,800 is useable up to full page prints with modest post-processing.
7) Configurable menu system similar to Sony's popular A77 that is both deep and intuitive in its layout, fostering quick mastery. Excellent control/operating system structures, and lots of flexibility for the advanced user.
8) Having image stabilization built into the sensor instead of the lens makes lenses potentially less expensive and lighter, and giving you full IS with any lens in your bag.
9) SLT approach enables full-time live view and phase detection autofocus even when shooting movies, and saves weight too, compared to more conventional DSLR technology.
Cons:
1) Price - at least $400 above where it should be.
2) EVF is not for everybody - some people simply can't transition from optical viewfinders (what do they know
!)3) EVF really needed a contrast adjustment on A77/65, as images sometimes either had blown highlights in the viewfinder or areas where image too dark to see details - Sony failed to include this adjustment in their new flagship camera - a major omission.
4) only 24 MP - clearly out-resolved by Nikon D800.
5) only 6 FPS with autofocus is disappointing for a SLT design - APS C designs were all class-leading in terms of FPS. Canon does 6 FPS number while still having to move the mirror up and down, and even the D800 manages 4 frames per second with its much higher megapixel throughput demands on imaging pipeline, while the $800 Sony A65 does 10 FPS . . . . so what happened here?
6) Sony APS-C lenses (but not 3rd party lenses) will automatically result in a significant drop in resolution - no option to simply allow full frame vignetting (which some users might not mind in some contexts). Only lenses with the 'AF-D' designation (not that many in the Sony stable) will get the full benefit of the sensor phase detection augmentation of autofocus.
7) Shorter battery life relative to its two main competitors (the cost of the excellent EVF)
8) Despite the flagship technology and price, the A99 still has too shallow a buffer for meaningful high-speed shooting - if shooting RAW plus JPEG, buffer fills up in 18 images or just 1.5 seconds. Can't believe that with all the noise about this issue in previous APS-C models that Sony did so little differently in their flagship camera.
9) USB 3.0, Wi-Fi support and built-in flash all omitted from feature set - no option currently even for add-on Wi-Fi adapter (could just get Wi-Fi SD card?). (OK - pro cameras usually don't have built-in flash . . . but this makes remote triggering of slave units difficult).
10) Weight, esp. with the best available mid-range zoom (the CZ 24-70 2.8) - this combo weighs in around 1.7 kilos (almost 4 lbs).
UPDATE OCTOBER 17th - The D600 now has studio test images on DPR, and this new Nikon FF model clearly (but only slightly) exceeds the Canon 5DMIII in low light ability, and is close to the Nikon D4 - suggesting that the D600 will be formidable competition for the A99 (the D600 has basically the same sensor, but the A99 is only about ½ F stop of poorer, due to the SLT design taking ~ 30% of sensor light). Low light ability is really the only major/substantive advantage that full frame pro cameras have over 24 MP APS-C models like the Sony A77/65 (putting aside the EVF/OVF debate). This unfortunately has only re-amplified my concern about the price point - the D600 is really impressive at least in its early returns, and, it's $700 less (and with a more complete pro-camera ecosystem around it). On the other hand, early returns suggest that Nikon has had some initial quality control problems with dust and perhaps other junk on the sensors.
November 24th Update
Having had a bit more time with the camera, I have some better sense of its strengths and weaknesses. I love the low light ability, but I am finding that it really isn't enough better outside of that area than an A65 for me to take it everywhere - and I find myself carting the moderately heavy A65 (with its 16-50 2.8 lens) around when I am willing to tote major hardware (over my Galaxy S II cellphone) instead of the REALLY heavy A99 (with the monster CZ 24-70 2.8) except when I know I will be shooting in low light. I find myself struggling with the curious concern about the camera that as great as it is, it just isn't THAT much better than the A65 - except again in low light - to feel wildly happy with it after coughing up $2800. Its video is only marginally more capable than the A65's, unless I am willing to add all kinds of pro accessories. So I'm finding myself wondering if I am going to get enough meaningful usage from this, if in good light, I find myself preferring the A65 with its faster shooting speed, identical EVF, and slightly more compact form factor/weight. Neither the A65/77 or A99 has an decent-sized buffer for real high speed shooting (which would have tipped the scales in favor of the A99), and I am still stunned that Sony (after all the flack voiced about this issue in relationship to the other SLTs) didn't fix this issue on the A99.
Overall, this is clearly a great camera but perhaps just not great enough for this much money? I have decided after much thought to send it back, and wait for the A109, which will have the extra resolution of the 36MP sensor, and probably an even better EVF - and I suspect Sony will find a way to make its low light ability roughly equal to the Nikon D800 while improving its speed and AF). In the meantime, I will live with the mediocre low light ability of the A65. For $2000 less than the A99, it seems like the better value even if it's not the better camera. Perhaps Sony has just set the bar so high with their subframe SLTs that they are competing not just with CaNikon, but now with themselves to some extent. I look forward to seeing what Sony does with their next FF model, and whether they can improve the low light performance of their next generation of subframe sensors. Even this generation of EVFs is (at least in my experience - most of the time) superior to even full frame optical finders - the next generation should be really something and give Canikon even more to think about.