Why is the size/resolution different when using Place Embedded vs making Smart Object manually?

[jjwithers]

If you drag an image (A) into Photoshop onto image (B) where it creates a SO layer, or choose Place Embedded, the file comes in at one size, and indicates 100% when you hit the Transform tool. 

If you open the file (A) up, drag it into document (B) as a normal pixel layer, then convert it to a Smart Object layer, the 100% scale is actually the true resolution relationship betweeen file (A) and File (B). 

Why does the Embed function have a complete disconnect to actual resolution relationships between two documents? 

This is a nightmare for dealign with students and entry level Photoshop users who don't know this is happening.  Is it a Bug? 

I can't see how it is logical in any way. 

See the attached image for how it changes the resolution, and in this case interpolates way too much in the scaling. 

In my opinion, if the Embed function is being used,  and the SO fits into the image it is being embedded in, When you hit command T - you should have the actual percentage/scale of that SO based on it's resolution relationship to the 'host' document.  

[lambiloon]

Hi place embed is good for new students as when we use it becomes SO it can be resize easily because it prevents the details, other than we do drag process and do manually .... and this is not bug when switch it off from preferences....regards

[jjwithers]

There is no preference that I am aware of that has to do with the resolution of a smart object layer.  

There is a preference to create a SO when placing, but beginners would have NO idea how to turn that off or what that means. 

What you are not addressing is the fact that a placed SO layer is not using it's resolution correctly vs a SO layer created after the contents were added as a layer.  

[JJMack]

Place may degrade images based of the two Print resolution. However, the Smart object has the actual  image pixels  so you can recover the original image quality by scaling the layer to have the same number of pixels  as the placed in file canvas size.

If the Document you are placing a file into has the same print resolution as the file being  places in and the placed in file canvas size is smaller or equal the the document its being placed into the place in layer will not be degraded the smart object layer pixels will be the same as the file being placed in pixels as will the hardened smart object pixels. 

[jjwithers]

This video explains the error better.  

A placed embedded SO comes into the master document at nearly double the size, but it states 100%. The interpolation looks terrible. 

The SO created by putting the content as a layer, and then converting the layer to a SO is accurately showing the resolution without upscale-inerpolation. 

[JJMack]

You are most like using Place editing a document the has print a resolution and are placing in Image Files that have other Print resolution and you do not like or understand the scaling Place is  doing. 

You are also most likely confused by the Photoshop preferences Resize during Place.  Resize During place is not about the scaling  Place does to images.  It just effect smart object that are created larger them the current documents canvas.  Larger smart object layer will be scaled down in size to fit on canvas if resize during place is checked.   

Place is not like Copy Past, duplicate layer or dragging layer between documents.    Place will scale the image based on document and files print resolutions.

For me this is a Problem. IMO Adobe should have implemented Places like Copy Pastes duplicate layer dragging layer between documents. Just let the user scale the new layer for the current documents use.  The users still need to do this with Smart object layers. 

I think someone at Adobe thought that scaling based are print resolution would help user scale the new layer.  If Adobe used some AI to analyze object in the document and object in the smart object they may be able to help with scaling. Scaling based on print pixel sizes has no bearing on image content its just a print size setting.   

The good news is the  harden smart object has your image quality and when you scale the smart object layer you will be scaling from you harden image quality not from the scaled layer image pixels place created.  Always have Place create Smart layers do not let place destroy your image quality.

This should demo that you image quality is not lost that the Object is you image.  Here is an extreme stupid demo.  Here I create a new document that canvas size has the number of Pixels my camera captures.  I'm going to be using this on the web so resolution does not matter  so I set the resolution stupidity to 1.  In my Preferences I check Skip transform during place. I do not want to be brothered by Place's interactive transform. Lets Place create the smart object layer I'll transform the layer if it needs to need be transform. 

It created a 8px by 12px image layer from my raw 8.2MP 300dpi conversion.  The smart Object create has my 8.2MP image it is fine.

[D Fosse]

It's not a bug. It's how smart objects are defined.

Smart objects honor physical print dimensions, not pixels. In effect, it treats raster images like vector objects. It's done this way for compatibility with vector applications like Illustrator. It's the only possible way to make it work.

Yes, it's counter-intuitive, and yes, it overrides the file's native behavior. Proceed with caution to avoid accidental resampling.

[JJMack]

And it brother many that Place does this for it not helpful. Users still have to scale the smart object layer created.  The scaling Places does is not useful in most cases. I even removed the Frame Tool from my tool bar and disabled its shortcut for the Frame Tool uses Place and mess up image it will not scale image to fill a frame when place scale the image small then frame area. Adobe told me they will not fix this issue.  "Adobe print pixels sizes should change print print sizes not image sizes". So I removed the frame tool it was not needed in the first place.  Oh there that bad word again PLACE...

[D Fosse]

"Adobe told me they will not fix this issue. "

That's because it already works as intended. It has to work this way, or vector smart objects simply wouldn't be possible. Pixel-based smart objects is fine as long as you're within Photoshop only, but there's no way to make it work with Illustrator. It has to be based on size, and resolution/ppi is the only way to translate between them.

[JJMack]

That may be Adobe BS. Do vector files even have a print resolution they are not pixels images. Vector smart object are vector files that Photoshop does not support there are no vectors in Photoshop for vector smart object. Vector smart object processing is different then none vector smart objects. Photoshop can only import vector file as a single pixels layer.  If you double click on a vector smart objects layer content thumbnail  in the layers palette to open the layers  vector object. The vector work file Photoshop creates will not open in Photoshop.  If you have AI installed the vector work file will open in AI. On my machine if  vectot work file is a .svg file  Chrome will display the SVG work file. If the work vector file is an .ai vector file it opens in Notepad. Photoshop is not a vector editor. It does not support vector files, it can import vector files as a single raster layer. Vector file associated system application is not Photoshop. When you scale a vector smart object the smart is replaced with an import of the vector file at the transform size.  Only vector smart object work this. I do not know if vector image file even have a print resolution setting they are not pixels images they are metadata text files.  When you transform other smart object Photoshop changes the smart object transform that is currently used to transform the pixels hardened for the Smart object by Photoshop to produce the Layers Pixels.  Other smart objects are resize via interpolation. 

Photoshop is a pixels editor it not a vector editor,  Ai is a vector editor does AI support all of Photoshop layer kinds can AI edit Photoshop PSD files.  Adobe has provides some interapplication interfaces for passing paths between AI and PS. 

[D Fosse]

Yes, you're confirming my point. It has to work the way it does now. It's not because of any Photoshop programming, it's because of the basic nature of raster vs vector data.

[JJMack]

Because it work the way  the frame tool has and issue as do users. We must manually correct the stupid scaling places does. I see no reason why Place has to be implemented like it has been. The scaling it does is not used the user must correct places scaling places saving graces is the smart object is good. So always have create smart object checked when placing in your preferences so you  can correct what place messes up.

[mukfe5FF4]

please, anyone, help to create create a boll I want to upload on my website.

[Bojan Živković]

@jjwithers "I can't see how it is logical in any way."

The only logic here is that document can not have two resolutions. You can not have one layer with 96ppi and one above it at 300ppi.

Here is my understanding and explanation:

What must happen is to match resolutions of object which is placed and parent document. How deed is done? Perhaps worst scenario for many of us, Photoshop is correcting resolution of placed image in background with Resize turned on! If only that is ommited, to change resolution with Resample turned off! That is why you have much smaller of expected image 300ppi placed into parent document which has 96ppi assigned.

[jjwithers]

PPI should be irrevelant. 

This should be based on total number of pixels within each document.  If you have a document that is 4000 pixels wide, and you place an embedded image that is only 1000 pixels wide, then it should come in at 1/4th the size and read 100% when using the transform tool. 

Instead, the SO comes in at some interpolated larger size that reads 100% when using the transform tool but it is actually more pixels than 1000px across. 

To me, this is an error and can cause a destructive workflow. 

[D Fosse]

OK, I'm repeating this for the third time in this thread: it has to be this way for compatibility with Illustrator.

Yes, in Photoshop we should always be thinking in pixels. I say that all the time. But vector smart objects are simply not possible if based on pixels. Hence this special case.

This is one of those things that once you know it, it ceases to be a problem. Smart objects always honor print size, not pixels. It's pretty easy to relate to.

[jjwithers]

How can print size and pixels not be related? 

Unless you take the ppi out of the equation and go strictly on WxH settings ? 

[D Fosse]

Pixels and print size are related through the ppi setting. That's the link.

The point I'm trying to make the whole time here, is that smart objects treat raster image files as if they were vector files. To do that, they need to calculate a print size, and for that the ppi number is used. Ppi translates pixels to physical size.

And now Photoshop and Illustrator speak the same language. This means an Illustrator file can be placed in Photoshop as a smart object, and their relative sizes defined.

In Photoshop a smart object then maintains the print size whatever the original ppi, even if it means resampling the pixel data. That's the price to make this work at all.

This should in fact be familiar already. Font sizes in Photoshop work the same way, only in reverse. Font size isn't defined by pixels, but by physical size. It comes from lead typesetting, where one point = 1/72 inch. So the only way to make it work in Photoshop, is to use the ppi number to translate from size to pixels.