[bdery]

Does it really add processing time in camera

In short, yes, extra processing is never without impact. In many cases it won't be measurable, but not always. For instance, adding clarity to a JPEG inside a K-1 takes a significant time, it's certainly measurable.

[gatorguy]

In short, yes, extra processing is never without impact. In many cases it won't be measurable, but not always. For instance, adding clarity to a JPEG inside a K-1 takes a significant time, it's certainly measurable.

Thanks.

Wouldn't processing JPEGS have more of an impact on resources and write times than raw only? I ask because, while I haven't actually timed it to verify, it feels as tho the burst and buffer clearing does better when I write raws to only card slot 1 and avoid JGEG's altogether. If writing to card 2 at the same time doesn't change the processing times I won't bother doing so.

[house]

Wouldn't processing JPEGS have more of an impact on resources and write times than raw only?

The raw actually contains a jpeg, in pentax case it think it's full size but fairly compressed? The camera relies on this jpeg processing for histograms etc as well so it's doing most of the processing regardless if you're saving a separate jpeg or not.

[gatorguy]

The raw actually contains a jpeg, in pentax case it think it's full size but fairly compressed? The camera relies on this jpeg processing for histograms etc as well so it's doing most of the processing regardless if you're saving a separate jpeg or not.

Correct, there is. My understanding it's more of a thumbnail, and not usable by itself for reproduction or sharing. Do I have that right? If so wouldn't processing a full-size JPEG and writing it as a separate file in addition to the raw slow down the buffer clearing and reduce the number of burst shots before slowing down? If I'm correct that's why when I anticipate needing to do bursts of an undermined length I write only to slot 1, and raw.

[stevebrot]

Wouldn't processing JPEGS have more of an impact on resources and write times than raw only?

The preview JPEGs (plural) included in the DNG/PEF will include the lens correction (the largest is full-size), if such actually exists for the particular lens. (I don't have access to the current list, but a few years ago correction was only available for a relative handful of auto-focus Pentax lenses.)


Steve

[luftfluss]

Does it matter how the lens gets there, as long as it gets there? For example, if the 24MP Nikon Z6 + Z 24-70/2.8 lens creates a sharper, more detailed image with less CA than a 24MP D750 with the AF-S 24-70/2.8, who cares if the former uses more software correction than the latter?

[house]

My understanding it's more of a thumbnail, and not usable by itself for reproduction or sharing.

It's not terrible quality and as mentioned full size so I'd say it can be used for a lot of scenarios. You have to extract it first though and it's not a feature available in most software.

Does it matter how the lens gets there, as long as it gets there? For example, if the 24MP Nikon Z6 + Z 24-70/2.8 lens creates a sharper, more detailed image with less CA than a 24MP D750 with the AF-S 24-70/2.8, who cares if the former uses more software correction than the latter?

If you're a raw shooter correction profiles exist or can be accurately home made for all lenses. If the milc lens is better than the dslr lens *after* corrections have been done to the raw file I can see your point. It being better sooc is irrelevant to me as a raw shooter. Some discussions put forward this idea that the official corrections are sophisticated, complicated and very good while in reality the are often the opposite and half an hours work can make a better one.

It's also important to understand that corrections are a moving target. The software is continously getting better. One lens may suddenly fare better than another because a new algorithm was devised and it's better suited to that particular expression of the aberration. Only one thing is for sure and that is that all software corrections leave some trace but it's also true that optical corrections may sacrifice qualities that can't be software corrected.

[Rondec]

I don't think Pentax's lens profiles affect RAW write times, do they? They don't do anything to the RAW file, they only affect the displayed jpeg when you review the image, but if you open it in Lightroom or Raw Therapee, you won't see that same image at all. On the other hand, my experience with shooting jpegs is that their write times is slowed considerably when you have lens profiles applied.

(I wonder if the RAW files are big enough that the time it takes to write them is about the same with or without those profiles).

---------- Post added 01-10-22 at 06:37 PM ----------

I don't think Pentax's lens profiles affect RAW write times, do they? They don't do anything to the RAW file, they only affect the displayed jpeg when you review the image, but if you open it in Lightroom or Raw Therapee, you won't see that same image at all. On the other hand, my experience with shooting jpegs is that their write times is slowed considerably when you have lens profiles applied.

(I wonder if the RAW files are big enough that the time it takes to write them is about the same with or without those profiles).

[stevebrot]

if the 24MP Nikon Z6 + Z 24-70/2.8 lens creates a sharper, more detailed image with less CA than a 24MP D750 with the AF-S 24-70/2.8, who cares if the former uses more software correction than the latter?

First of all, that is a big "IF". Secondly, application of AI to supply sharpening and detail (assuming the Z6 does this) is hardly an optical process. As a result basing a lens test on camera output may not be testing the lens at all.


Steve

[stevebrot]

if the 24MP Nikon Z6 + Z 24-70/2.8 lens creates a sharper, more detailed image with less CA than a 24MP D750 with the AF-S 24-70/2.8, who cares if the former uses more software correction than the latter?

First of all, that is a big "IF". Secondly, application of AI to supply sharpening and detail (assuming the Z6 does this) is hardly an optical process. As a result basing a lens test on camera output may not be testing the lens at all.


Steve

[luftfluss]

First of all, that is a big "IF". Secondly, application of AI to supply sharpening and detail (assuming the Z6 does this) is hardly an optical process. As a result basing a lens test on camera output may not be testing the lens at all.


Steve

First of all, that is a big "IF". Secondly, application of AI to supply sharpening and detail (assuming the Z6 does this) is hardly an optical process. As a result basing a lens test on camera output may not be testing the lens at all.


Steve

I have only the words of several Nikonians who have used the both the Z 24-70/2.8 and a previous 24-70/2.8.

If you have a link to information regarding "application of AI to supply sharpening and detail" I would be appreciative.

Here is a link to one individual's thoughts regarding the Z7 and the sharpness of output: The Nikon Z7’s Insane Sharpness | Strolls with my Dog

Here is an imperfect comparison between the two lenses: Nikon NIKKOR Z 24-70mm f/2.8 S on Nikon Z7 vs Nikon AF-S NIKKOR 24-70mm f/2.8E ED VR on Nikon D850

I believe the article linked in the OP is, indeed, about testing the overall output of a specific body + lens combination, rather than the lens "in a vacuum" as it were. But this is simply an evolution of how lenses have been tested for a long time in the Digital Age - hence the admonishment by Optical Limits, "Please note that the tests results are not comparable across the different test systems!"

[mee]

Correct, there is. My understanding it's more of a thumbnail, and not usable by itself for reproduction or sharing. Do I have that right? If so wouldn't processing a full-size JPEG and writing it as a separate file in addition to the raw slow down the buffer clearing and reduce the number of burst shots before slowing down? If I'm correct that's why when I anticipate needing to do bursts of an undermined length I write only to slot 1, and raw.

I think one should also take into consideration the context of when these "rules" were set in place -- a decade or more ago when computer processors weren't so powerful as they are today. So sites such as DPR probably wanted to show off the lens as it is but also because performance might degrade on camera reviews with all the corrections enabled. Today it might not be as big of a deal with processors many times over more powerful in our cameras, at least on the flagships (such as the K-3 III).

That said the only way to be sure would be to run tests. But I'd only do that if you really have the itch to find out. Seems tedious to me.

[house]

I don't think Pentax's lens profiles affect RAW write times, do they? They don't do anything to the RAW file, they only affect the displayed jpeg when you review the image, but if you open it in Lightroom or Raw Therapee, you won't see that same image at all. On the other hand, my experience with shooting jpegs is that their write times is slowed considerably when you have lens profiles applied.

(I wonder if the RAW files are big enough that the time it takes to write them is about the same with or without those profiles).

---------- Post added 01-10-22 at 06:37 PM ----------

I don't think Pentax's lens profiles affect RAW write times, do they? They don't do anything to the RAW file, they only affect the displayed jpeg when you review the image, but if you open it in Lightroom or Raw Therapee, you won't see that same image at all. On the other hand, my experience with shooting jpegs is that their write times is slowed considerably when you have lens profiles applied.

(I wonder if the RAW files are big enough that the time it takes to write them is about the same with or without those profiles).

Settings should affect raw write speed as well. As mentioned a full size, full settings JPEG is saved in the dng/pef. The camera has to do the processing to create the file no ways around it.

[Kunzite]

Does it matter how the lens gets there, as long as it gets there? For example, if the 24MP Nikon Z6 + Z 24-70/2.8 lens creates a sharper, more detailed image with less CA than a 24MP D750 with the AF-S 24-70/2.8, who cares if the former uses more software correction than the latter?

Of course it does, particularly in the context of lens reviews, where we compare and decide "winners" and "losers".

How about penalizing the lens with less distortion, because it's not set to correct it by default?
How about penalizing the sharper lens, because it doesn't use extra software sharpening ("diffraction correction" or how they're calling it)?
How about giving the "win" to an inferior but more aggressively "corrected" lens because it's somewhat cheaper and/or lighter?

And, if that's the logic, why not applying it to cameras as well?
Sorry, but this is nothing more than using double standards.

[Rondec]

Settings should affect raw write speed as well. As mentioned a full size, full settings JPEG is saved in the dng/pef. The camera has to do the processing to create the file no ways around it.

Maybe so. Regardless, I have them turned off. They don't do any good to the RAW image and if I want to apply lens corrections it is pretty simple to do so in Lightroom.