JPEG or RAW - a new look.

Messages
5,001
Edit My Images
Yes
With the various arguments about whether Topaz Labs new product "JPEG to RAW" actually can turn a JPEG into a 16 bit TIFF or DNG file.

The usual argument in favour of using RAW is that a JPEG, being only 8 bits, has lost too much digital information for it to become a 16 bit file.

So here is a little test - I have posted 2 images on my Flickr site, can you tell the difference?

A little hint - both were processed in J2R then edited exactly the same way.

They are both 100% crops.

View: https://www.flickr.com/photos/20926615@N05/46303221754/in/dateposted-public/


View: https://www.flickr.com/photos/20926615@N05/46303183134/in/dateposted-public/


Look forward to your replies.
 
I suppose one was taken as a jpeg and the other as a raw file? That is totally irrelevant in this comparison; because the raw file had to be converted into a jpeg and the data discarded in order for J2R to even open the file... they both started the process as 8bit jpegs.

That's not to say raw is always better... a 16bit file is only (potentially) useful if your camera is putting out more than 8bit data.
 
I can turn a 1 bit image into a 24 bit image so obviously J2R *can* take an 8 bit and covert to 16 bit.

I can also take a 24 bit image, convert to a 1 bit image, and then convert back to a 24 bit image, but the result won't be anywhere near as good as the original 24 bit image.

The question is should I? I'm sure there are some edge cases when this might be useful, and applied appropriately, it might make sense. What I'm not getting is the point of routinely chaining together a bunch of unnecessary conversions as a matter of course?
 
Last edited:
I can turn a 1 bit image into a 24 bit image so obviously J2R *can* take an 8 bit and covert to 16 bit.

I can also take a 24 bit image, convert to a 1 bit image, and then convert back to a 24 bit image, but the result won't be anywhere near as good as the original 24 bit image.

The question is should I? I'm sure there are some edge cases when this might be useful, and applied appropriately, it might make sense. What I'm not getting is the point of routinely chaining together a bunch of unnecessary conversions as a matter of course?

A 1 bit image would consist only of black and white.
 
I suppose one was taken as a jpeg and the other as a raw file? That is totally irrelevant in this comparison; because the raw file had to be converted into a jpeg and the data discarded in order for J2R to even open the file... they both started the process as 8bit jpegs.

That's not to say raw is always better... a 16bit file is only (potentially) useful if your camera is putting out more than 8bit data.

Wrong.
 
Can't you just keep this to a single thread?

We get it. You think topaz whatever is fantastic. I'm still waiting to see anybody else with your enthusiasm.

In the same way that you and others like PhotoShop and Lightroom?

This is a single thread with a different purpose.
 
In the same way that you and others like PhotoShop and Lightroom?

This is a single thread with a different purpose.

Yes, but we don't create thread after thread after thread to try and convince everybody else.

Sorry, but this thread is no different to the others you've started. Just a different image.

For me, and I suspect most other people here, photography is and art and that includes the post processing. I enjoy working with my best images crafting them into something special.

Why the hell would I want to just shove them through some meat grinder type software with very little control to accept whatever it spits out.
 
Last edited:
Yes, but we don't create thread after thread after thread to try and convince everybody else.

Sorry, but this thread is no different to the others you've started. Just a different image.

For me, and I suspect most other people here, photography is and art and that includes the post processing. I enjoy working with my best images crafting them into something special.

Why the hell would I want to just shove them through some meat grinder type software with very little control to accept whatever it spits out.

So you know all about it without actually trying it?

Or what this thread is actually about?
 
So you know all about it without actually trying it?

Or what this thread is actually about?

You've posted so much about it, yes I know what this thread is about. And no I won't try the software.

Its targeted at people that just want a quick fix for their sub par photos shot with poor technique, poor cameras and poor lenses. It's not my thing and I'm willing to bet I'm not the only one that thinks this.
 
You've posted so much about it, yes I know what this thread is about. And no I won't try the software.

Its targeted at people that just want a quick fix for their sub par photos shot with poor technique, poor cameras and poor lenses. It's not my thing and I'm willing to bet I'm not the only one that thinks this.

So you're wrong on all counts - you don't know what the thread is about - you condemn without a trial - literally - and you assume that any who use it are crap amateurs producing crap photos with below par equipment - WOW!
 
So you're wrong on all counts - you don't know what the thread is about - you condemn without a trial - literally - and you assume that any who use it are crap amateurs producing crap photos with below par equipment - WOW!

So what is this thread about? How is it different to any of the others?
What exactly is the difference between your two image?

I can't see any difference, other that some sharpening which I can do in any software. One thing is for certain, IT'S HAS NOT BEEN CONVERTED TO RAW.

But look, if it works for you, that's great, just stop trying to convince the rest of us.
 
Last edited:
So you know all about it without actually trying it?

Or what this thread is actually about?

I tried it but after all your posts I was expecting different results. Not sure I want to use Wet Transfer, besides I've already furnished topaz with my email and not keen to give it out to more companies.

If you still want to have a go with my photo is there another way to get the file to you?
 
So what is this thread about? How is it different to any of the others?
What exactly is the difference between your two image?

I can't see any difference, other that some sharpening which I can do in any software. One thing is for certain, IT'S HAS NOT BEEN CONVERTED TO RAW.

But look, if it works for you, that's great, just stop trying to convince the rest of us.

don't jump to assumptions.

The fact that you can't see any difference IS the difference!

But the actual purpose of the 2 test images was simple - how much infomation is there in a JPEG that could be used by JPEG to RAW?

A lot more than most people realise.

They were both JPEGs from my Canon 1Ds MkII which gives 16 MP images.

Test A is the original JPEG, just over 10MB in size.

Test B is a reduced version of that - 833Kb in fact!

The original JPEG was reduced by Easy Thumbnails and kept the image size but reduced the filesize (JPEG Quality set to 60%)

Yet it still produced a good image despite the huge loss of data.

So if it is possible to produce a good image from such a small filesize it means there is a huge amount of available data in JPEGs.

So I see no reason why Topaz Program cannot use this data to do exactly what it says.

Both full size images remained at 16MP regardless of the filesize.
 
I tried it but after all your posts I was expecting different results. Not sure I want to use Wet Transfer, besides I've already furnished topaz with my email and not keen to give it out to more companies.

If you still want to have a go with my photo is there another way to get the file to you?

It's actually Wetransfer and nobody wants your e-mail - just upload the image and let me have the URL - simples!
 
don't jump to assumptions.

The fact that you can't see any difference IS the difference!

But the actual purpose of the 2 test images was simple - how much infomation is there in a JPEG that could be used by JPEG to RAW?

A lot more than most people realise.

They were both JPEGs from my Canon 1Ds MkII which gives 16 MP images.

Test A is the original JPEG, just over 10MB in size.

Test B is a reduced version of that - 833Kb in fact!

The original JPEG was reduced by Easy Thumbnails and kept the image size but reduced the filesize (JPEG Quality set to 60%)

Yet it still produced a good image despite the huge loss of data.

So if it is possible to produce a good image from such a small filesize it means there is a huge amount of available data in JPEGs.

So I see no reason why Topaz Program cannot use this data to do exactly what it says.

Both full size images remained at 16MP regardless of the filesize.
Therefore proving that the jpeg compression algorithm is particularly effective and selecting the less relevant information to discard. That’s all.
 
Therefore proving that the jpeg compression algorithm is particularly effective and selecting the less relevant information to discard. That’s all.

And also proving that Topaz' AI can probably use this information to achieve exactly what it says on the tin.
 
don't jump to assumptions.

The fact that you can't see any difference IS the difference!

But the actual purpose of the 2 test images was simple - how much infomation is there in a JPEG that could be used by JPEG to RAW?

A lot more than most people realise.

They were both JPEGs from my Canon 1Ds MkII which gives 16 MP images.

Test A is the original JPEG, just over 10MB in size.

Test B is a reduced version of that - 833Kb in fact!

The original JPEG was reduced by Easy Thumbnails and kept the image size but reduced the filesize (JPEG Quality set to 60%)

Yet it still produced a good image despite the huge loss of data.

So if it is possible to produce a good image from such a small filesize it means there is a huge amount of available data in JPEGs.

So I see no reason why Topaz Program cannot use this data to do exactly what it says.

Both full size images remained at 16MP regardless of the filesize.

Sorry, but you are wrong. Clearly you have zero understanding of how jpg compression works. The software is using so called AI (another misrepresentation in my opinion) to add data based on what it thinks should be there.

Here's a novel idea. Shoot raw.
 
And also proving that Topaz' AI can probably use this information to achieve exactly what it says on the tin.
No. That’s not true. It simply shows that at 60% ‘percent quality’ the image degradation is acceptable on that jpeg image.

You ran both images though the same J2R, so there’s no control sample. You can’t draw any conclusions about the impact of J2R.
 
Can't you just keep this to a single thread?

We get it. You think topaz whatever is fantastic. I'm still waiting to see anybody else with your enthusiasm.

I share Pete's enthusiasm for Topaz in general but the JPG-RAW utility doesn't offer me anything as I always have a raw file to start with.

Bob
 
Regarding JPG 'quality'

Here's a picture of my cat at jpeg quality 100, saved width 1024px, no output sharpening - 414KB

DSC00188 1-100.jpg

Here's the same picture at Quality ZERO, same settings - 34KB
DSC00188 1-000.jpg

jpeg is very effective at discarding information selectively - specifically retaining areas of significant contrast variation (ie. edges).
 
Sorry, but you are wrong. Clearly you have zero understanding of how jpg compression works. The software is using so called AI (another misrepresentation in my opinion) to add data based on what it thinks should be there.

Here's a novel idea. Shoot raw.

So I shoot RAW - then have to convert it to TIFF or DNG or use something like ACR to work on it, sharpen it, reduce noise if necessary etc - And finally I end up with - guess what?

A JPEG!

This new software makes that almost redundant and I only pay for something when I really think it's excellent and I can use it.

I bought it this morning - with a 20% reduction.

And my Z800 is already churning away, batch converting a lot of older JPEGs to TIFFs ready for further editing.

And regardless of what anyone says I think real professional will be buying it in their droves.

Because for them money is time!
 
So I shoot RAW - then have to convert it to TIFF or DNG or use something like ACR to work on it, sharpen it, reduce noise if necessary etc - And finally I end up with - guess what?

A JPEG!

Umm - only because you saved it as a JPG!

It doesn't have to be JPG, it can be TIF or any other format your editing software offers as an option.
 
So I shoot RAW - then have to convert it to TIFF or DNG or use something like ACR to work on it, sharpen it, reduce noise if necessary etc - And finally I end up with - guess what?

A JPEG!

This new software makes that almost redundant and I only pay for something when I really think it's excellent and I can use it.

I bought it this morning - with a 20% reduction.

And my Z800 is already churning away, batch converting a lot of older JPEGs to TIFFs ready for further editing.

And regardless of what anyone says I think real professional will be buying it in their droves.

Because for them money is time!

Or... shoot raw, open up in any modern editor (Lightroom, Capture One, On1 etc) and save as jpg when you want to upload etc. It's one step. Couldn't be easier, and importantly, you are in control and you can apply your own present and sync related images etc - certainly easier than unnecessarily piping through a minimum of three separate programmes before you even start to 'edit'.

Or, shoot jpeg and do pretty much the same, obviously with the option to avoid the final step in your happy with your OOC image.

I can absolutely understand when you have that one killer shot that hasn't gone quite right, that something like Topaz AI suite could be used to rescue it, but for 99% of the images that are well exposed, focussed etc, it's detracting from the image by making it look overly artificial because it's making choices for you (edges too contrasty, over sharpened look etc).
 
Last edited:
Umm - only because you saved it as a JPG!

It doesn't have to be JPG, it can be TIF or any other format your editing software offers as an option.

Yes bit it can't normally be viewed as such unless you use unless you use something like Windows Image Viewer, I was talking about what most of use do every day on the web - view JPEGs
Or... shoot raw, open up in any modern editor (Lightroom, Capture One, On1 etc) and save as jpg when you want to upload etc. It's one step. Couldn't be easier, and importantly, you are in control and you can apply your own present and sync related images etc - certainly easier than unnecessarily piping through a minimum of three separate programmes before you even start to 'edit'.

Or, shoot jpeg and do pretty much the same, obviously with the option to avoid the final step in your happy with your OOC image.

I can absolutely understand when you have that one killer shot that hasn't gone quite right, that something like Topaz AI suite could be used to rescue it, but for 99% of the images that are well exposed, focussed etc, it's detracting from the image by making it look overly artificial because it's making choices for you (edges too contrasty, over sharpened look etc).

In fact it doesn't over sharpen as you would know if you tried it.

On the other hand I may over-sharpen.
 
Yes bit it can't normally be viewed as such unless you use unless you use something like Windows Image Viewer, I was talking about what most of use do every day on the web - view JPEGs

We've been here before. Try viewing a jpeg without the right software to decode it. TIFF, PNG, GIF are no different.

At the end of the day, it if works for you, and you're happy, who am I to say. Personally, I don't think it's a miracle cure all. Machine Learning is not infallible, and to trust it to blindly try to improve images that frankly don't need it, is going to cause unnecessary degradation. I project I recently worked on, despite being trained by literally 100's millions of images, still decided a truck was a pillow!

Like all tools, it's best used only when needed.
 
Yes bit it can't normally be viewed as such unless you use unless you use something like Windows Image Viewer, I was talking about what most of use do every day on the web - view JPEGs

You could argue that ALL images are captured only to be viewed!

I produce the majority of my images for printing - why would I capture in JPG? Makes little sense to me but if it works for you......
 
don't jump to assumptions.

The fact that you can't see any difference IS the difference!

But the actual purpose of the 2 test images was simple - how much infomation is there in a JPEG that could be used by JPEG to RAW?

A lot more than most people realise.

They were both JPEGs from my Canon 1Ds MkII which gives 16 MP images.

Test A is the original JPEG, just over 10MB in size.

Test B is a reduced version of that - 833Kb in fact!

The original JPEG was reduced by Easy Thumbnails and kept the image size but reduced the filesize (JPEG Quality set to 60%)

Yet it still produced a good image despite the huge loss of data.

In my own experiments on the best settings for jpeg quality I found that generally speaking I very rarely could tell the difference between 100% (no compression) and 80%. I could, with difficulty, often tell the difference between 75% and 100%. I could easily tell the difference between 65% and 100%. It varied a lot depending on the lens and the kind of photograph.

So if it is possible to produce a good image from such a small filesize it means there is a huge amount of available data in JPEGs.

So I see no reason why Topaz Program cannot use this data to do exactly what it says.

The conversion to jpeg does a number of different things to the RAW information. It adjusts the colours based on the selected (or default) white balance. If the white balance setting is close to correct then nothing of importance is lost in that. But if white balance is badly off some important information needed for the new changes in white balance will have been lost. The same goes for dynamic range. The combination of selected exposure, contrast, etc. settles the chosen dynamic range and how it is represented. Highlight detail outside the selected white limit is lost for good. Dark detail under the selected black limit is lost for good. And the resolution of dynamic range gradients with those limits is reduced.

In short, the kind of RAW file a jpeg can be converted to is one which has already been precooked according to the chosen or default jpeg conversion. This will not be any more useful for further post processing than the jpeg itself unless it cleverly guesses some extra things, such as adding stuff that looks like more strands of hair to places which look as though they were probably hairy, adding some woven cloth texture to places which looks as though they were probably cloth, some stubble to something which looks like an unshaven male chin, and so on. This information doesn't come from extra information in the jpeg. It comes from (artificially) intelligent guesswork, based on looking at lots of what it thinks are similar images.

No doubt it will keep on getting better at this. I look forward to the day when you'll be able to say to it "Please add some clouds and half a dozen seagulls to the sky, change the pebbly beach to sand, remove the parked car on the promenade, and shorten the women's skirts."
 
You could argue that ALL images are captured only to be viewed!

I produce the majority of my images for printing - why would I capture in JPG? Makes little sense to me but if it works for you......

If you produce for printing you wouldn't!
 
Here is my view. You can shoot jpg and it will give you a good picture. You can shoot RAW and capture everything. The choice is yours. RAW is like a negative it has everything so you can develop it the way you want to. Nothing new in that from film days
 
No doubt it will keep on getting better at this. I look forward to the day when you'll be able to say to it "Please add some clouds and half a dozen seagulls to the sky, change the pebbly beach to sand, remove the parked car on the promenade, and shorten the women's skirts."

That's already being done - not the women's skirts (so far) but it's now possible to use AI to improve a sky, even place someone in another setting.

Oh and create totally realistic faces:

https://www.theregister.co.uk/2018/12/14/ai_created_photos/

or attempt to create original art:

https://blogs.scientificamerican.com/observations/can-ai-create-true-art/

Those who deny the incredible possibilities of this new technology are just Canutes trying to stop the encroaching waves.
 
Here is my view. You can shoot jpg and it will give you a good picture. You can shoot RAW and capture everything. The choice is yours. RAW is like a negative it has everything so you can develop it the way you want to. Nothing new in that from film days

JPG Images do look great up to a point. I have recently worked on 3 raw files, enlarged to several feet in width and length and saved as TIFF files for the company printing the vehicle wrap. If I would have done the same process starting with a JPG, there would be a marked difference in the end result. All depends what you want from the image.
 
Exactly. It’s what works for you. I personally shoot RAW very rarely jpg
 
Yes bit it can't normally be viewed as such unless you use unless you use something like Windows Image Viewer, I was talking about what most of use do every day on the web - view JPEGs
Have something like Adobe Bridge and you can see RAW files prior to processing and Jpegs and Tiffs. Just because Jpegs are what is used as the final result most of the time doen't mean that that should be what you start with. It is a choice obviously.

In fact it doesn't over sharpen as you would know if you tried it.
Any software that can make a lot of changes to an image without user intervention doesn't seem like a good idea to me. No matter how 'intelligent' software is it can make mistakes. What if an image doesn't need sharpening at all! Again it is a choice obviously, but I like to edit/process each image with regards to what I think needs changing, if indeed anything needs changing. ;)

Those who deny the incredible possibilities of this new technology are just Canutes trying to stop the encroaching waves.
And there are those that don't seem to want to do what most people do that gives great results, and seem to go out of their way to do something different which may or may not be an improvement for whatever reason. And then seem to want to convert others to their way of thinking. :rolleyes:

It is not denying that technology can do some amazing things, as photographers we use technology all the time, but sometimes things are embraced either before they are ready, or they are 'fixing' a problem that didn't need 'fixing' for most people.
 
Back
Top