a99 buffer too small?

TrojMacReady said:

EinsteinsGhost said:

VirtualMirage said:

EinsteinsGhost said:

You’re comparing cRAW (D600) to RAW (A99), just as people have done cRAW (A77) to RAW (A99) before you.

While at it, why don't we throw in Nikon D800 and Canon 5DMkIII into the mix as well?

Click to expand...

Do you enjoy spreading false information?

The A99's outputted files are in a form of cRAW, which is why they are around 24-25MB in size when outputted to the memory card instead of closer to the 42-45MB mark (depending on sidecar file size and any extra data Sony stores in the file).
--
Paul

Click to expand...

I am the one spreading wrong information? Prove it. Show me a RAW file out of A99 that is 14-bit/24-25MB

Click to expand...

Click to expand...

Thanks TrojMacReady!

EinsteinsGhost,

I have proved it numerous times in the past even with links to download the RAW files yourself but you continuously failed to acknowledge it. With what TrojMacReady has also shown you that should be enough, but I'll go a step further.





Here is the information from an A99 RAW file. Notice how it too says it is a compressed RAW. And if you are curious as to where I got this RAW file, it is from the same site I have listed over and over to you to go look for yourself and see that the RAW file sizes for the A99 are what I originally stated. Just in case you forgot that website or forgot our past conversations even said website, here it is again for reference:

http://www.quesabesde.com/noticias/sony-a99-analisis-fotos-muestras,1_9088

It just so happens to be the same site TrojMacReady must have gotten their A99 RAW example.

So, I think that is proof enough to show that I am right.

--
Paul

johnbee said:

EinsteinsGhost said:

Wise to replace 6 FPS mode with 4 MPS, so buffer has a greater clearing time? I see many already crying about "only" 6 FPS.

Click to expand...

6FPS is fine for short bursts I'm sure. Though if you look closely you'll find that I suggested that they "add" a mid-range option to help make better use of the data write times. Which could look something like: High 6FPS, MED 3-3.5 FPS, and :LOW 1FPS, type of thing. This could prove useful for studio and location type work or in cases where workflow matters most.

Click to expand...

Not sure if you know this or not, but the A99's Lo continuous capture rate is around 2.5fps when in full frame mode.
--
Paul

VirtualMirage said:

Not sure if you know this or not, but the A99's Lo continuous capture rate is around 2.5fps when in full frame mode.

Click to expand...

No I didn't actually
Good to know.

VirtualMirage said:

TrojMacReady said:

EinsteinsGhost said:

VirtualMirage said:

EinsteinsGhost said:

You’re comparing cRAW (D600) to RAW (A99), just as people have done cRAW (A77) to RAW (A99) before you.

While at it, why don't we throw in Nikon D800 and Canon 5DMkIII into the mix as well?

Click to expand...

Do you enjoy spreading false information?

The A99's outputted files are in a form of cRAW, which is why they are around 24-25MB in size when outputted to the memory card instead of closer to the 42-45MB mark (depending on sidecar file size and any extra data Sony stores in the file).
--
Paul

Click to expand...

I am the one spreading wrong information? Prove it. Show me a RAW file out of A99 that is 14-bit/24-25MB

Click to expand...

Click to expand...

Thanks TrojMacReady!

EinsteinsGhost,

I have proved it numerous times in the past even with links to download the RAW files yourself but you continuously failed to acknowledge it. With what TrojMacReady has also shown you that should be enough, but I'll go a step further.





Here is the information from an A99 RAW file. Notice how it too says it is a compressed RAW. And if you are curious as to where I got this RAW file, it is from the same site I have listed over and over to you to go look for yourself and see that the RAW file sizes for the A99 are what I originally stated. Just in case you forgot that website or forgot our past conversations even said website, here it is again for reference:

http://www.quesabesde.com/noticias/sony-a99-analisis-fotos-muestras,1_9088

It just so happens to be the same site TrojMacReady must have gotten their A99 RAW example.

So, I think that is proof enough to show that I am right.

Click to expand...

Not necessarily. Unless A99 is only outputting cRAW. Even then, I'm curious as to how A99 with its 14-bit cRAW can be the same size as 12-bit cRAW out of A77.

Any answer to this question? I can only imagine it to be true if A99's RAW is even more compressed than A77.

ck3 said:

phixel said:

TrojMacReady said:

KM Glassman said:

In reading a review of the a99 at Expertreviews.com, I was surprised to see how small the buffer is in the a99:

"In our test it slowed to 2.3fps after 13 frames for JPEGs, and to 0.9fps after 11 frames for RAW." *
http://www.expertreviews.co.uk/dslrs/1294723/sony-alpha-a99-review-hands-on

In comparison, the Nikon D600 with the same Sony 24MP sensor does much better before filling the buffer:
JPEG Fine Large: 57
RAW: 16 (14-bit lossless compressed) to 27 (12-bit compressed)
Source: http://www.bythom.com/d600.htm

Seems odd that Sony's a99 flagship FF camera underperforms Nikon's D600 entry level FF camera that sells for $700 less. Any explanations for this?

• Phil

Click to expand...

No idea, maybe because of certain settings or because it's a pre production camera?

The Imaging Resource measurements of the D600 are almost an exact copy of the A99 specs as stated on the Sony website:

D600 RAW 14 bit: 14
A99 RAW 14 bit: 15

D600 RAW 14 bit +jpeg: 12
A99 RAW 14 bit + jpeg: 12

D600 jpeg fine: 28
A99 jpeg fine: 24

Click to expand...

I think I may have found the reason (bolded by me), for the discrepancy between Thom Hogan's D600 real world performance figures and Imaging Resources results.

"The Nikon D600's buffer depths are good for a prosumer SLR, considering the file sizes. The D600 managed to capture 28 L/F JPEGs, 14 RAW frames, or 12 RAW+JPEG frames before slowing down. (You'll likely do better with real-world subjects, as the target image we use for our tests is designed to be difficult to compress.) "- ImagingResource. com

It will be interesting to see if Sony's stated a99 performance specs hold up to the Imaging Resource's torture test as well.

• Phil

Click to expand...

Let's not forget Thom Hogan's caveats:

"Like all Nikons there are a lot of caveats. The buffer will never go beyond 100. You'll need to set JPEG Size Priority (default) to get the JPEG numbers; JPEG Optimal Quality will cut into those numbers a bit. Long Exposure NR will take an image off the buffer size. Auto Distortion Control will take as much as half the buffer. HI ISO values will steal some buffer. Remember that if you're not using state-of-the-art UHS-1 SD cards, you probably won't get these numbers either (they're from my 8GB SanDisk Extreme Pro). There may be other things that impact buffer but I haven't discovered them yet.

I don't think JPEG shooters are going to have any trouble with the buffer size. Best case they'll be shooting continuously for 10 to 18 seconds. Worst case, they'll still have probably a five second buffer. Raw shooters may need to watch settings carefully. I can see the camera getting to a buffer of 7 (r7 reported in viewfinder) when I set the camera in certain ways, which is barely over a second worth of buffer. I'm guessing that raw shooters will want to set 12-bit Lossless Compressed and turn off all the things that take buffer space (especially Auto Distortion Control), which should get them a bit short of four seconds of continuous shooting."

Click to expand...

I feel the buffer limitations are more problematic for JPEG shooters, with some or all of these JPEGs setting (NR, CA removal, distortion correction), turned on. I think Thom missed one memory hog though - Auto-CA removal. For this reason, RAW shot buffer capability is more consistent compared to JPEGs. Of course the manufacturers always use the optimal JPEG settings scenario to gin up their camera's performance numbers. Nothing new there.

• Jon

EinsteinsGhost said:

Not necessarily. Unless A99 is only outputting cRAW. Even then, I'm curious as to how A99 with its 14-bit cRAW can be the same size as 12-bit cRAW out of A77.

Any answer to this question? I can only imagine it to be true if A99's RAW is even more compressed than A77.

Click to expand...

You refuse to admit that you are wrong.

From what we have seen so far the A99 doesn't give an output option that exclusively states RAW or cRAW. It's RAW output is a cRAW file, just like how the A77 labels their files. The screenshots show that, the RAW files show that. People that have used the camera have reported back that they didn't see an option in the menu. Sony's spec sheets don't even show an option between the two like they did for their older cameras that did (as well as how Nikon and Canon does with their cameras).

As for your explanation, as I have mentioned before, the files between the two are not identical in size. But they are close in size. My A77 RAW files I have seen range from 23.5-24.6MB with the average size being around 24MB. With the A99 files I have been seeing them range from around 24-25MB, and that is just the few that we have seen so far. If the range in size is about the same as the A77, then 24-25MB should be pretty close to the files size range for the A99, give or take a fraction of a MB.

With compression, just because the A99 has about a 17% higher bit depth doesn't mean you should assume the compressed file is going to grow by 17% too. That isn't how compression works. Compression may not work as linearly as you may think it does. Other factors come into play that can keep the file size at a minimum, such as a reference table of fixed variables that the compression scheme can simplify greatly since it's compression codec will know exactly how to decode and uncompress that data.

--
Paul

Thats about standard for all the Sony cameras starting with the A57 and moving upwards.
The figures are correct as i have mentioned in earlier threads .
The Sony cameras just can't keep up with the Fastest cards on the market.

One day the penny will drop in this forum and members who don't believe this will have to swallow some pride.
As I said before faster cards are a waste of money .The cameras are the problem.

Ive read over all of these posts and have come to believe that no one here understands what a fast frame rate camera is all about... its not about capturing a bunch of shots... its about capturing "ONE" shot ... if you want a bunch of shots, shoot a movie.

the whole idea of fast frame rate is to adjust for the speed of the movement which can't easily be predicted at 1/2000 of a second... IN the old days there was alot of luck involved and the discipline to get as close to the moment as possible before firing the burst.. a good film photog can get the ball off the bat in three or four frames... If you do not know what IM talking about then you don't know what fast frames rates are designed around.

Sports shooter like myself are looking for the one Money shot so they shoot a burst.. its not shooting a spray machine gun style.. its shooting a burst to capture a designed shot.. like the ball off the bat, or the puck off the stick.... I know when the pitcher throws the ball that 60'6" at 86 MPH it will reach the catchers glove in 0.48 sec. As you can see getting the money shot happens in less than a half second.. if you want to get the ball off the bat you need to fire a half second burst to cover the flight of the ball into the catchers glove.. a half second burst is 6 frames... 1/3 of the buffers capacity. you could take back to back throws and still not fill the buffer, but at a 1/3 buffer capacity, those 6 frames are removed from the buffer in less than 3 seconds leaving the buffer empty again... I shoot sports all the time and have never had to shoot 17 frames ( a full buffer) to get any shot. High speed shooting is about getting one shot, you throw the rest away..that's how its taught in college, and that's the way most sport shooter I know do it... I don't know what all these people here think high frame rates are all about... its specifically surrounding a specific shot with enough images to get the actual money shot.. The buffer for the A77 and a99 is more than enough for someone knowing what fast frame rates are all about. That is getting a single shot. in my scenario, the ball will travel 10 ft per frame, this does not insure getting the money shot but if you have a 6 fps camera you only get three frames in the delivery making the each frame represent 20 feet of travel per frame. chances are much higher getting the shot with 6 frames as oppose to trying to get the shot with 3 frames. That's why you use digital see if you got the money shot, and if not just shoot again.. there is also no guarantee the batter will swing, or make contact... this is why sports shooting is still luck, but with 12 fps and a good camera your chances are much better to get the shot...
--
Bill aka EO

--Personally, I would be embarrasssed to use a full buffer to get a single shot. getting the shot in three or four frames is the stuff that sport shooter brag about...
Bill aka EO

For the most part, I agree with what you are trying to say here.

I think there have only been a handful (at most) situations where the buffer of a camera has ever got in my way. But for 99% of the rest of the time it has never been an issue. And those times where the buffer interrupted my shot it didn't necessarily prevent me from getting the shot either, it only cut short a sequence of motion I was trying to capture. But in the end it didn't ruin what I had already captured and a few seconds later I was ready to shoot again, no loss.

Now what has gotten in my way more times than I can count, which has a far greater impact to me than buffer, is available light and shutter speed. And without enough of either, good luck trying to even reach your max frame rate to even fill your buffer.
--
Paul

VirtualMirage said:

EinsteinsGhost said:

Not necessarily. Unless A99 is only outputting cRAW. Even then, I'm curious as to how A99 with its 14-bit cRAW can be the same size as 12-bit cRAW out of A77.

Any answer to this question? I can only imagine it to be true if A99's RAW is even more compressed than A77.

Click to expand...

You refuse to admit that you are wrong.

From what we have seen so far the A99 doesn't give an output option that exclusively states RAW or cRAW. It's RAW output is a cRAW file, just like how the A77 labels their files. The screenshots show that, the RAW files show that. People that have used the camera have reported back that they didn't see an option in the menu. Sony's spec sheets don't even show an option between the two like they did for their older cameras that did (as well as how Nikon and Canon does with their cameras).

As for your explanation, as I have mentioned before, the files between the two are not identical in size. But they are close in size. My A77 RAW files I have seen range from 23.5-24.6MB with the average size being around 24MB. With the A99 files I have been seeing them range from around 24-25MB, and that is just the few that we have seen so far. If the range in size is about the same as the A77, then 24-25MB should be pretty close to the files size range for the A99, give or take a fraction of a MB.

With compression, just because the A99 has about a 17% higher bit depth doesn't mean you should assume the compressed file is going to grow by 17% too. That isn't how compression works. Compression may not work as linearly as you may think it does. Other factors come into play that can keep the file size at a minimum, such as a reference table of fixed variables that the compression scheme can simplify greatly since it's compression codec will know exactly how to decode and uncompress that data.

--
Paul

Click to expand...

I just downloaded one of the RAWs from A99, opened it in IDC and looked at EXIF. It says... "ARW 2.3 (Compressed) Format".

Does that mean there is only cRAW compressed that is about the same size as A77 despite 14-bit depth as opposed to 12-bit? I would not bet against that. Would you?

EvilOne said:

--Personally, I would be embarrasssed to use a full buffer to get a single shot. getting the shot in three or four frames is the stuff that sport shooter brag about...

Click to expand...

That's fine for amateur work. Though most people shooting for money will look for convenience. That being said, my wife and I cover various events in our area and we rely heavily on continuous shooting.

TBH. I don't have an issue with the FPS of the A99 myself(6 is plenty enough for anything I do). Though the continuous shooting speeds and buffer could turn-out to be an issue. - will have to try a body out and see I guess.

johnbee said:

EvilOne said:

--Personally, I would be embarrassed to use a full buffer to get a single shot. getting the shot in three or four frames is the stuff that sport shooter brag about...

Click to expand...

That's fine for amateur work. Though most people shooting for money will look for convenience. That being said, my wife and I cover various events in our area and we rely heavily on continuous shooting.

TBH. I don't have an issue with the FPS of the A99 myself(6 is plenty enough for anything I do). Though the continuous shooting speeds and buffer could turn-out to be an issue. - will have to try a body out and see I guess.

Click to expand...

--So as a professional please tell me what camera you use??? here is a quote from you that someone asked the same thing that went unanswered

Member said:

johnbee wrote:

While I really don't know much about the A77, I'd say that the A99 is not the right > tool for the job. IMO. If you want to cover sports, then you'd be better off getting a > camera designed to shoot sports.

Click to expand...

Member said:

That being said you can get some very good deals on past models that are very good> (leagues ahead) of the A99 in terms of FPS and buffer.

Click to expand...

Member said:

Member said:

Such as?

Click to expand...

Click to expand...

You obviously do not know much about the A77 because the camera was designed around sport shooting, I bought that camera specifically because of the frame rate, its the fastest fully automatic focus frame by frame 12 fps camera in the world... there are some that claim fast frame rates but the focus is locked after the first frame... as a professional you should know that... again what is the magic camera that you use at your events to capture faster frames rates than 12 fps and with a bigger buffer... ?

The camera pushes 12 frames into one second of action allowing 12 stop action frames in that second... thats a real big deal when shooting sports where incramental frames can actually give you the best chance to capture the money shot moment... If you shoot events like " grass growing , or chess matches, 12 fps mean much but try to capture a 200 mph shuttle **** in an olympic batmitton match and 6 fps isint going to get it.

Bill aka EO

thubleau7 said:

Thats about standard for all the Sony cameras starting with the A57 and moving upwards.
The figures are correct as i have mentioned in earlier threads .

Click to expand...

You mean those threads where you again and again have shown a severe lack of understanding of the difference between an UHS-I enabled camera, and a non-UHS-I enabled camera.

Your measurements are only valid for a non-UHS-I enabled camera. The a99 and a77 supports UHS-I.

Wow Bill seems like your a pretty accomplished sports shooter. How about posting some of you're action shots? Love to see some.

EvilOne said:

--Personally, I would be embarrasssed to use a full buffer to get a single shot. getting the shot in three or four frames is the stuff that sport shooter brag about...
Bill aka EO

Click to expand...

EvilOne said:

--So as a professional please tell me what camera you use???

Click to expand...

I don't shoot sports and so my own gear likely won't be helpful in this case. Though if you want to talk low light event shooting then I might have something useful to contribute.

EvilOne said:

You obviously do not know much about the A77 because the camera was designed around sport shooting...

Click to expand...

I'm fairly certain the A77 wasn't the issue(as indicated). Though I'd point-out the A77 has limitations of its own against the professional camera's(13s data writes @ 15 image burst). That being said, in the FF format realm, I'd say something along the lines of a 1Ds III or 1D III/IV is better suited for the task. Keeping in mind that the issue here isn't bound by FPS alone(that much we know), but more importantly... with a camera's ability to sustain a continuous frame rate before topping-out during a shoot.

Hope this helps.

can change this. I remember when Minolta had this issue with the 7d. People complained and Minolta replied. The buffer speed was like night compared to day.

VirtualMirage said:

For the most part, I agree with what you are trying to say here.

I think there have only been a handful (at most) situations where the buffer of a camera has ever got in my way. But for 99% of the rest of the time it has never been an issue. And those times where the buffer interrupted my shot it didn't necessarily prevent me from getting the shot either, it only cut short a sequence of motion I was trying to capture. But in the end it didn't ruin what I had already captured and a few seconds later I was ready to shoot again, no loss.

Now what has gotten in my way more times than I can count, which has a far greater impact to me than buffer, is available light and shutter speed. And without enough of either, good luck trying to even reach your max frame rate to even fill your buffer.
--
Paul

Click to expand...

Michaels7 said:

can change this. I remember when Minolta had this issue with the 7d. People complained and Minolta replied. The buffer speed was like night compared to day.

Click to expand...

While I fail to understand how this applies to my comment, you also have to remember that the firmware that the Minolta 7D was released with was horribly in need of optimization.

That camera when released didn't even support proper USB 2.0 cable transfer speeds when it came out, only offering about a quarter of what they initially advertised. File flushing times when it first came out barely reached 1.2 MB/s, the firmware update pushed it up to 6MB/s. The camera also didn't even offer highlight blinks when it released.

All of that showed signs that the hardware was rushed out the door before the firmware was ready. The firmware was functional, but severely handicapped to what was being advertised.

While there are a lot of possibilities for improvements via firmware updates, the A99 in its current state with pre-release firmware does not seem to be in the situation the Minolta 7D was.

--
Paul

I'm very aware of what the 7D specs that firmware had. I know because I'd purchased it when it first came out. I'd didn't mentioned the other updates because it was irrelevant to the topic. What I do remembered was the slow buffer speed and Minolta addressing that. What Sony's plans are with the current A99 buffer speed is hard to call right now. They may and they may not address it. Time will tell. Frankly, I think that people are jumping the gun (negative) before the camera has hit the streets.

Edit: I just realized that you'd mentioned the buffer size. Error in reading.

VirtualMirage said:

Michaels7 said:

can change this. I remember when Minolta had this issue with the 7d. People complained and Minolta replied. The buffer speed was like night compared to day.

Click to expand...

While I fail to understand how this applies to my comment, you also have to remember that the firmware that the Minolta 7D was released with was horribly in need of optimization.

That camera when released didn't even support proper USB 2.0 cable transfer speeds when it came out, only offering about a quarter of what they initially advertised. File flushing times when it first came out barely reached 1.2 MB/s, the firmware update pushed it up to 6MB/s. The camera also didn't even offer highlight blinks when it released.

All of that showed signs that the hardware was rushed out the door before the firmware was ready. The firmware was functional, but severely handicapped to what was being advertised.

While there are a lot of possibilities for improvements via firmware updates, the A99 in its current state with pre-release firmware does not seem to be in the situation the Minolta 7D was.

--
Paul

Click to expand...

I would like to see someone check all these cards on an A77 an A99 a Canon 6D and a Nikon D600 or D800 otherwise I can't believe any of this rubbish.