Lens Decisions and Coming to Terms With an Unpleasant Fact

[Brownie]

I came to realize last season that I was under-equipped for late night shooting at the track. 2.8 is fine when there's no movement, but freezing action at night results in crazy hi ISOs. There's more than one way to skin a cat, as they say, so I set about doing some testing. At the very end of the season I took a 35/1.4 and 50/1.7 out. It was a day shoot, so this was just to check composition and framing.

In this particular situation I am standing next to a safety wall about 10-12 feet from the cars. Moving away from the wall with a longer FL results in the lower part of the car being obscured. I can move up or down the wall along its length, and I can move away from the wall a little bit when standing down-track to change the angle with a narrow lens, so the wall isn't in the frame. Cropping has limited success in removing the wall due to angles.

Both the 35 and the 50 did ok. With the 35 I couldn't get close enough to keep the wall out and still get the entire car in the frame, or move away too far, so it was limited to one spot along the wall slightly in front of the cars. The 50 was about the same, but it could fill the frame pretty well on a car in the far lane. Neither one was perfect, one of the reasons hate primes for shooting in these situations.

I seriously considered one of the 35-150/2-2.8's. But it really only gets me a stop at the wide end, the rest of the range puts me right back in 2.5-2.8 territory. With no other fast zooms to look at, primes are the only choice. That in mind I picked up an 85/1.4 over winter.

This past Friday night was a late event. I purposely left my 2.8 zoom lenses at home and took the 35 and 85. I wanted to confirm what I found with the 35 last season as well as try it at night, plus determine if the 85 would be as usable as expected.

The 35 results were consistent. Same limited range to shoot, not really close, not down track. Meh. The 85 is a whole different story. It's long enough to let me move around a bit, but not so long that I can't get closer and keep the wall out. It's a really fun lens to shoot with and will become a go to for those circumstances. As much fun (and as useful) as it is, I am thinking about a 135. I'm going to take my Minolta 28-135 out and test it as a prime during the day to see how well the FL works out before biting on a prime.

During the day I tested some different FLs with the 24-105. I had been looking at 24 and 28 primes as a fast/close solution. I almost bought a 28, thinking that was what I needed, but decided to wait. Good thing I did, it's too narrow to get the length of a car in at that distance. I decided on a 24, which will be delivered today.

Now to the unpleasant fact. In this setting, the A7RIII is a liability. That sucks because I really love the camera, but no amount of love can overcome the issues. I originally bought the RIII as a backup camera, but ended up using it as a second camera, if that makes sense. As a backup it would get me through in a pinch, but as a second it falls short. I realized this the first time I took it out there and have used it primarily in the pits where the shots are more static, or with a shorter lens standing right next to the cars. That works better than down track since they aren't moving as quickly and the camera has more time. Still, the camera can't keep up. It will get the first shot or two, then hesitate while it tries to focus, then hunt, etc. etc. Sometimes it rips off a burst flawlessly, but those instances are far fewer than required. Typically there will be a shot completely missing while the camera thought about it, or two shots OOF, or something else.

The problems are primarily due to the slow write speed, several generations old AF, and the older processor. Even with the slow-ish write speed of the A7 IV I can shoot a 6-7 frame burst and end up with as many usable shots from which to choose.

I've been talking about the A9III for some time, hoping Sony will update it extensively, including a higher resolution sensor. I've come to the conclusion that something in the 40MP range is perfect for this. Now Nikon has released the Z8, with 45MP and a very fast burst rate. At $3900 I have my doubts as to whether Sony will have a reasonable response.

I'll live with the A7RIII for now and keep an eye on Sony's next move. If it turns out to be another $5000 24MP camera, I'll pass. I'd be just as far ahead with another A7 IV at that point.

 

[MeowZzz]

Can you hire a 135/1.8 to try? (sorry if already addressed)

As to A9iii, funnily enuff, I am not sure I would want the 33mp sensor. The images from my a7iv allow more cropping, but usually the a9's images at night start off a bit nicer. So, if I don't crop, then it's all good.


I have spent a ton of time on n off circuits, and most of them had very little close to the traffic viewing. One track downunder was a handshake from pit wall to the bikes, but most of them, and especially in Euro and Asia are like telephoto distance away.

Hope you got ear plugs wherever you are!

 

[Brownie]

I guess you use the a74 then

I think there still is a significant market for a high speed lower mp camera, namely the photo journalist. For news organizations that are buying 10s of thousands of these cameras the price difference between an A9 and A1 is a big deal. For the general consumer I think there is a smaller market for the A9 but still probably large as a couple thousand is still a lot for many.

I hadn't considered that, but wouldn't a less expensive A7 IV or even a III work for those situations? Unless they're shooting sports it seems it'd be enough.

 

[Brownie]

I guess you use the a74 then

I do use the A7 IV, as my main camera. The A7RIII is the second camera. I keep one on a Peak Design clip with a short lens so I can move around without having to change lenses. It not only saves time but it saves my sensors from getting all kinds of airborne crap on them. I have to restrict the RIII to specific situations because of the limitations I've found, and it still has a hard time keeping up. That is the main point of the discussion.

Can you hire a 135/1.8 to try? (sorry if already addressed)

As to A9iii, funnily enuff, I am not sure I would want the 33mp sensor. The images from my a7iv allow more cropping, but usually the a9's images at night start off a bit nicer. So, if I don't crop, then it's all good.


I have spent a ton of time on n off circuits, and most of them had very little close to the traffic viewing. One track downunder was a handshake from pit wall to the bikes, but most of them, and especially in Euro and Asia are like telephoto distance away.

Hope you got ear plugs wherever you are!

I can try my Minolta Maxxum 28-135/4-4.5. It delivers fantastic results. I'll try it out at full extension on one of my trips out there to see how I like the FL.

As one of the track photographers at the dragway, I have an all-access media pass and get right up close and personal with the cars.

Milan Dragway Heads Up!

It sure seemed like a long time between last season and this, but we're off to a great start. Got out there about 4:30 and stayed until 10:00 or so. They only had a few divisions of semi-finals and finals left so I bailed. It was starting to get chilly! DSC05152 by Shotglass Photo, on Flickr...

www.alphashooters.com

 

[spudhead]

I hadn't considered that, but wouldn't a less expensive A7 IV or even a III work for those situations? Unless they're shooting sports it seems it'd be enough.

Quick answer no, until you have had and used the a9 you do not have a clue how it betters many lesser models even the newer ones with more pixels, in so many ways and it is just not about speed although 20 frames is more than I ever use or need most of the time. I think many assume that a9 users shoot full tilt locked on all the time in fact locked on is a quick way to lock the focus and exposure and end up with what many users on here complain about 1 shot in focus and the rest off, regardless of camera body, anyway that is what I have found after 4years plus with the a9 and extensive testing. if we had a cut down a74 it would probably be little better than the a 73 which I still have and can take many of the shots I take now with ease the speed of the a9 sensor and the electronic shutter and no blackout are more than worth the smaller pixel count that many complain about

 

[Brownie]

Quick answer no, until you have had and used the a9 you do not have a clue how it betters many lesser models even the newer ones with more pixels, in so many ways and it is just not about speed although 20 frames is more than I ever use or need most of the time. I think many assume that a9 users shoot full tilt locked on all the time in fact locked on is a quick way to lock the focus and exposure and end up with what many users on here complain about 1 shot in focus and the rest off, regardless of camera body, anyway that is what I have found after 4years plus with the a9 and extensive testing. if we had a cut down a74 it would probably be little better than the a 73 which I still have and can take many of the shots I take now with ease the speed of the a9 sensor and the electronic shutter and no blackout are more than worth the smaller pixel count that many complain about

Right, but the comment was made in the context of photojournalism. Is that kind of speed needed for the application? Sports, yes, but seems like anything would work for the remainder.

 

[spudhead]

Right, but the comment was made in the context of photojournalism. Is that kind of speed needed for the application? Sports, yes, but seems like anything would work for the remainder.

I think I will bow out here since this thread is so far away from where it started Tim

 

[Brownie]

I think I will bow out here since this thread is so far away from where it started Tim

No it's not, it's a normal continuation of a conversation. Conversations evolve.

 

[Brownie]

 

[AlphaWorld]

I just read the opinion piece on DPreview about why the writer, owning the Z7 and Z9, is not going to buy a Z8 - the EVF is only 3M dots, but the Panasonic they also have is 5M dots. (Must admit to thinking, this person has a Z7, Z9, and a Panasonic S1H - why not add a Z8? )

I hadn’t realised that Sony was way ahead on that front - the “ancient” (well, it’s SO last decade…) A7RIV was already 5M, and the A1 and A7RV are both 9M - and I really appreciate the detail we can see in those viewfinders.

I wonder if the next A9 will also use the 9M dot EVF, or if it will use something like an ultra fast 5M dot EVF at 200+Hz?

 

[liggy]

Quick answer no, until you have had and used the a9 you do not have a clue how it betters many lesser models even the newer ones with more pixels, in so many ways and it is just not about speed although 20 frames is more than I ever use or need most of the time. I think many assume that a9 users shoot full tilt locked on all the time in fact locked on is a quick way to lock the focus and exposure and end up with what many users on here complain about 1 shot in focus and the rest off, regardless of camera body, anyway that is what I have found after 4years plus with the a9 and extensive testing. if we had a cut down a74 it would probably be little better than the a 73 which I still have and can take many of the shots I take now with ease the speed of the a9 sensor and the electronic shutter and no blackout are more than worth the smaller pixel count that many complain about

Agreed - for my uses the benefits of a stacked sensor > more megapixels. Otherwise I would consider trading in the A9II for an A7R V for the enhanced AF capabilities.

 

[Brownie]

I fully understand and agree with what you guys are saying about stacked sensors, but I don't understand the speed Vs Pixel count discussion anymore. Those days are behind us, you can now have both.

Z8:

45MP stacked sensor
20FPS Lossless Compressed RAW + jpeg
30FPS jpeg only
60FPS DX Crop
120FPS 11MP jpeg

For $3900, US.

The only question left (at least in my mind) is if Sony will follow suit or stick with the status quo.

 

[spudhead]

I fully understand and agree with what you guys are saying about stacked sensors, but I don't understand the speed Vs Pixel count discussion anymore. Those days are behind us, you can now have both.

Z8:

45MP stacked sensor
20FPS Lossless Compressed RAW + jpeg
30FPS jpeg only
60FPS DX Crop
120FPS 11MP jpeg

For $3900, US.

The only question left (at least in my mind) is if Sony will follow suit or stick with the status quo.

That is not what I am saying, from my point of view if Sony stay with a low megapixel camera in terms of the a9iii, and the price point is 5g then I already have a camera that does what I need, while I would like more megapixels it not the be all and end all, I have no interest in the ai crap or video, and in terms of the hobby how long will it be before we sit the camera down and just tell it to take a shot. I am a huge believer in high quality glass and would possibly put the 5g to a used 400 2.8 at this point in time. If I were nearer retirement or shooting more than the once a week I seem to get now that is a different ball game, new body regardless of price and the high end glass for sure, and that is Sony target customer, the ones who buy at any cost

 

[liggy]

I fully understand and agree with what you guys are saying about stacked sensors, but I don't understand the speed Vs Pixel count discussion anymore. Those days are behind us, you can now have both.

Z8:

45MP stacked sensor
20FPS Lossless Compressed RAW + jpeg
30FPS jpeg only
60FPS DX Crop
120FPS 11MP jpeg

For $3900, US.

The only question left (at least in my mind) is if Sony will follow suit or stick with the status quo.

Glad that Nikon is competing. Those of us that stay with Sony will hopefully benefit in terms of either features or pricing. From a purely financial standpoint I completely get that Sony is probably seeing that the $$$ is in moving upmarket. If you're going to carry a camera around it had better justify itself by being more capable than the every improving high end phone cameras.

Speed? For me the 20/30 fps in the A1 is more than adequate.

 

[Brownie]

Glad that Nikon is competing. Those of us that stay with Sony will hopefully benefit in terms of either features or pricing. From a purely financial standpoint I completely get that Sony is probably seeing that the $$$ is in moving upmarket. If you're going to carry a camera around it had better justify itself by being more capable than the every improving high end phone cameras.

Speed? For me the 20/30 fps in the A1 is more than adequate.

It'd be more than adequate for me too, but I don't want 50MP. I am still hoping that the A9III comes in around 35-40. Even the Nikon is more than I want, but it's also $2500 less than the A1 for similar performance. I'd love an updated stacked version of the 42 in the A7RIII.

My desire to stay more than 24MP comes from the fact that my images get printed in larger sizes quite a bit under circumstances in which I have no control. It just makes me a bit uneasy to send a file out to someone knowing full well they could crop the hell out of it for a 'closeup', or to remove my logo, or whatever. If it starts a little bigger, then the crop ends up bigger. I still haven't ruled out an A9X with a lower res sensor, just not comfortable with it yet.

 

[liggy]

It'd be more than adequate for me too, but I don't want 50MP. I am still hoping that the A9III comes in around 35-40. Even the Nikon is more than I want, but it's also $2500 less than the A1 for similar performance. I'd love an updated stacked version of the 42 in the A7RIII.

My desire to stay more than 24MP comes from the fact that my images get printed in larger sizes quite a bit under circumstances in which I have no control. It just makes me a bit uneasy to send a file out to someone knowing full well they could crop the hell out of it for a 'closeup', or to remove my logo, or whatever. If it starts a little bigger, then the crop ends up bigger. I still haven't ruled out an A9X with a lower res sensor, just not comfortable with it yet.

Totally understood. I've had good results with Topaz Gigapixel / PhotoAI for upscaling images from my 16mp Fujis so as software improves I'm prioritizing AF/Tracking more than MP and low noise as the software enhancements will IMO outpace hardware developments.

 

[Brownie]

Totally understood. I've had good results with Topaz Gigapixel / PhotoAI for upscaling images from my 16mp Fujis so as software improves I'm prioritizing AF/Tracking more than MP and low noise as the software enhancements will IMO outpace hardware developments.

I’ve done a little upsizing, sometimes it makes a difference, sometimes I can’t see any at all.

 

[iamdlewis]

Right, but the comment was made in the context of photojournalism. Is that kind of speed needed for the application? Sports, yes, but seems like anything would work for the remainder.

It isn't about the fps for them, though that can help them from time to time. The reason for going with the A9 over the A7 will be the processing performance and the AF performance. The A9 has nearly no risk of rolling shutter and an AF system that incredible even when compared to the great A7 ones. The no blackout EVF I am sure is nice for them as well.

 

[Gaz]

I love my 135mm GM and find myself reaching for it when I go looking for birds. I’d expect it to be pretty reasonable on the track, even attached to my A7iv. On the other side, I’ll test my 20mm F1.8 prime on Teslas parked in the street out the front of my apartment and you can be the judge.

The A9iii needs to match the A7iv pixel count. Those extra pixels are really useful for cropping fast moving objects… and I’ve been stunned by the evening’s sky shots on A7iv.

 

[Brownie]

So, my trip to the speedway last weekend (link below) was an eye opener. Both cameras did pretty well, The A7IV was expected, the A7RIII was better than I imagined. I need to collect my thoughts on this.

Flat Rock Speedway, first time doing oval track!

Was invited to shoot at a local ARCA track this weekend. Most of this was from the infield, some from the pit wall. I shot shutters from 100-1000, and was surprised just how slow it took to get decent motion blur. Depending on the lens, anywhere from 100-160, sometimes could get a little with...

www.alphashooters.com

 

[Brownie]

While shooting the ARCA event last week I noticed a few things that are inconsistent with previous experience. As stated above, the A7RIII performed very well, much better than expected. The main difference was the shooting location and type of event. Where at the dragway the cars launch and drive away from (or toward) me pretty quickly, the oval track lets you to pan for a long time with little change in distance when standing on the inside of the curve, which allows for a longer burst. The minimal change in distance resulted in a lot less focus hunting as they circled me and yielded a much higher percentage of good images. In some cases the camera still hesitated and I couldn't figure out why at the time, but I think I've nailed it down.

I believe my current SD card, the Sony Tough 128G with a 150MB/second write speed is adding to this. Between the slower sensor/processor, the higher pixel count (compared the A7 IV), and the card, the camera is bogging down while it writes. This was even apparent with the A7 IV slowing its shutter as the camera tried to write to the card. This isn't a problem with a 1 second burst, but in this case there were some ridiculously long bursts (because I could). I'm thinking about picking up 250-300MB/second card to see how much it improves, if any. This is why I asked for confirmation of the A7RIII's RAW burst rate in the other thread.

One thing for sure, I do not want/need any more than the 10FPS the A7RIII can provide. There isn't enough difference in auto racing from frame to frame to make it worth the additional uploading and culling time. If I do end up with a speed camera it'll be set to one of the slower burst rates anyway.

 

[iamdlewis]

While shooting the ARCA event last week I noticed a few things that are inconsistent with previous experience. As stated above, the A7RIII performed very well, much better than expected. The main difference was the shooting location and type of event. Where at the dragway the cars launch and drive away from (or toward) me pretty quickly, the oval track lets you to pan for a long time with little change in distance when standing on the inside of the curve, which allows for a longer burst. The minimal change in distance resulted in a lot less focus hunting as they circled me and yielded a much higher percentage of good images. In some cases the camera still hesitated and I couldn't figure out why at the time, but I think I've nailed it down.

I believe my current SD card, the Sony Tough 128G with a 150MB/second write speed is adding to this. Between the slower sensor/processor, the higher pixel count (compared the A7 IV), and the card, the camera is bogging down while it writes. This was even apparent with the A7 IV slowing its shutter as the camera tried to write to the card. This isn't a problem with a 1 second burst, but in this case there were some ridiculously long bursts (because I could). I'm thinking about picking up 250-300MB/second card to see how much it improves, if any. This is why I asked for confirmation of the A7RIII's RAW burst rate in the other thread.

It might be good to change the shutter release to "Release" since there is little distance change between each shot and at the range the DoF should still be good if the AF is off by a couple inches. This would stop the hunting which could be caused by the AF being confused by some decals or something like that.

Since you will be basically doubling the speed you will obviously notice a difference but it won't be twice as fast. I have been using a mix of Sony (both types), Lexar and ProGrade for 5 years and can't say I really notice a difference between them.

 

[Brownie]

It might be good to change the shutter release to "Release" since there is little distance change between each shot and at the range the DoF should still be good if the AF is off by a couple inches. This would stop the hunting which could be caused by the AF being confused by some decals or something like that.

Since you will be basically doubling the speed you will obviously notice a difference but it won't be twice as fast. I have been using a mix of Sony (both types), Lexar and ProGrade for 5 years and can't say I really notice a difference between them.

This was more an observation. I don't expect to return to this track anytime soon, maybe never. For one thing it was by invitation, and to be honest it's not my cup of tea. I found the small oval track a bit boring. I suppose if I were a regular it may be different.

I'm hoping the faster write speed will reduce the problems I've experienced at the drag strip. If it's safe to assume that even a portion of the hesitation is attributable to a slow card, a faster one may help get me through in the near term. It's looking more like the A9III won't show up until late summer at the earliest, and I'd like to improve things for the current season if possible.

 

[Brownie]

Ok, maybe @spudhead is right and I'm missing something in the setup. I just did a reset (painful) and built up a new custom Mode 1. I also ordered a Kingston Canvas React card. Not quite as fast as the best Sony, but @ 260MB/s write faster than what I'm using now, and for $82 there's not much risk if it doesn't improve anything.

Burst is set on 'Hi', which is supposed to give 8FP/s in RAW Uncompressed with no blackout. Strangely enough that's a 2-frame improvement over the A7 IV, and a little closer to what I 'think' is the ideal 10FP/s for this kind of shooting. Tracking sensitivity set to 5. Focus area is Expanded Flexible Spot with 'center' toggled on C1.

One thing I had forgotten about with this camera is that it locks the focus on first frame if you're using an aperture over f/8. That can be a pain in the bright sunshine out at the track, so I'll need to keep it in mind.

I need to get out to the track on a non-critical day and put it to the test as the 'first' camera instead of the second and see how it does. It's all going to boil down to how quickly the camera can refocus as the cars move. I'm not expecting a huge improvement, but it'd be nice to see a little bump in hit rate.

 

[spudhead]

Ok, maybe @spudhead is right and I'm missing something in the setup. I just did a reset (painful) and built up a new custom Mode 1. I also ordered a Kingston Canvas React card. Not quite as fast as the best Sony, but @ 260MB/s write faster than what I'm using now, and for $82 there's not much risk if it doesn't improve anything.

Burst is set on 'Hi', which is supposed to give 8FP/s in RAW Uncompressed with no blackout. Strangely enough that's a 2-frame improvement over the A7 IV, and a little closer to what I 'think' is the ideal 10FP/s for this kind of shooting. Tracking sensitivity set to 5. Focus area is Expanded Flexible Spot with 'center' toggled on C1.

One thing I had forgotten about with this camera is that it locks the focus on first frame if you're using an aperture over f/8. That can be a pain in the bright sunshine out at the track, so I'll need to keep it in mind.

I need to get out to the track on a non-critical day and put it to the test as the 'first' camera instead of the second and see how it does. It's all going to boil down to how quickly the camera can refocus as the cars move. I'm not expecting a huge improvement, but it'd be nice to see a little bump in hit rate.

Hopefully it might help Tim

 

[Brownie]

Since you will be basically doubling the speed you will obviously notice a difference but it won't be twice as fast. I have been using a mix of Sony (both types), Lexar and ProGrade for 5 years and can't say I really notice a difference between them.

You called it. Difference is hardly noticeable.

 

[Gaz]

Sony didn’t mind copying the flagship A7siii to introduce a cheaper high spec model, even knocking off parts from its flagship blog camera to do it.

The Nikon Z8 is a cheap Z9.

The A9iii would sit nicely with me if it had specs that competed with the A1.

Then the new release A1 can be something that is industry leading enticing both existing A1 owners, as well as others to upgrade at a ridiculously expensive rate.

You mark my words.

 

[Brownie]

At this point it's safe to say that the A7RIII's pausing and stuttering comes down to the AF, as compared to the A7 IV.

The A7RIII has a higher burst rate, a higher resolution sensor, and seems to allow more images in the buffer during a burst. But, it can't keep up with the A7 IV's focus.

If I could design my new camera, it'd be an A7RIII with the latest and greatest AF capability. It has all the speed and resolution I want.

 

[AlphaWorld]

At this point it's safe to say that the A7RIII's pausing and stuttering comes down to the AF, as compared to the A7 IV.

The A7RIII has a higher burst rate, a higher resolution sensor, and seems to allow more images in the buffer during a burst. But, it can't keep up with the A7 IV's focus.

If I could design my new camera, it'd be an A7RIII with the latest and greatest AF capability. It has all the speed and resolution I want.

I think you are not alone - I have seen more than a few people express the same feelings about the sensor in the A7RIII. I'm delighted with the A7RV - maybe if Sony put a new, faster, version on the A7RIII sensor into an otherwise unchanged A7RV you'd get everything you'd want - I suspect you'd really enjoy the much faster processor, the new rear screen, and the beautifully detailed EVF.

 

[Brownie]

I think you are not alone - I have seen more than a few people express the same feelings about the sensor in the A7RIII. I'm delighted with the A7RV - maybe if Sony put a new, faster, version on the A7RIII sensor into an otherwise unchanged A7RV you'd get everything you'd want - I suspect you'd really enjoy the much faster processor, the new rear screen, and the beautifully detailed EVF.

Funny, I was just thinking the same thing.

But in all honesty, is the A1 all that far off? I mean, when you boil it all down, the A1 seems pretty close, at least in Sony's lineup.

I want to get the RM3 out to the track with my 'new' setup and see how it does. I have a feeling the AF will be improved a little.

 

[Brownie]

I went to the track yesterday just to test the new settings on the R3 and try out the Samyang 24/1.8, which as it turns out is even more top-drawer than previously thought. (I'll post a couple shots in that thread). The R3 testing results were mixed. Without going into a lot of detail:

• Longer than normal bursts were used for testing to check AF accuracy and stuttering.
• Hit rates were better, but still not as good as the M4. The AF lock is not as good nor as fast.
• I did not experience any lock-up or stuttering with the new settings.
• I still love the sensor on the R3, but there is noticeably more noise than the M4.
• I do not need 10FPS+ for this kind of shooting. Even the 8FPS from the R3 resulted in more culling than I want. Anywhere from 6-8 will be ok.

Based on this, a speed camera is likely off the table. It almost appears another M4 is the right answer.