Photoshop's SUPER RESOLUTION vs GIGAPIXEL AI

Video Statistics and Information

Video
Captions Word Cloud
Reddit Comments
Captions
hey guys this is anthony morgante i am mr photographer.com last week i did a video demonstrating a new feature of photoshop's camera called super resolution if you haven't seen that video i'll have a link to it in the description below this video super resolution is meant to be used on an image that is either a heavily cropped or b shot on a camera that was relatively low megapixels when you have an image that is heavily cropped you're limited by what size print you could make from it and this is where you would use something like super resolution to increase the resolution of the image so you could create a larger print with it now there are a lot of different applications on the market that do this most notably is one by topaz labs called gigapixel ai and for those of you that have followed my channel know the past year year and a half i've done several videos on gigapixel ai well i thought it would be interesting if i did a head-to-head comparison we'll take the same image run it through super resolution and we'll take another that same image i should say and run it through gigapixel ai now i have the image here in lightroom and i have it in lightroom because i want to show you i did a significant crop on it so it really was a high resolution image at one time but i cropped it down to one thousand seven hundred a little 1796 pixels by 2692 pixels and i should have did the math ahead of time but i think that's around four and a half to five megapixels now i did do some processing it in lightroom you could see i did some basic processing and what i didn't do is i didn't do any sharpening and i didn't do any luminance noise reduction i just did some color noise reduction the default value of 25. i kind of want to see if specifically super resolution like enhances the noise because we know or those of you that have are follow my videos know that gigapixel ai has noise reduction built in so i don't think it will be an issue there so we're going to send this image into both of those applications now both of them of course photoshop works as a standalone application so does um topaz labs gigapixel ai and both of them of course work as plugins in lightroom but if i open they both read raw files if i open the raw file into photoshop it of course will open up into camera raw which is what we want but if i open the raw file up into topaz labs gigapixel ai it will not recognize the lightroom edits unless i use it as a plug-in and when you use it as a plug-in it doesn't load raw files it will load a tiff file i hope that made sense so if i use gigapixelai as a standalone product and load this raw file into it it won't recognize the lightroom edits so i would have to use it if i wanted to read those lightroom edits including that crop i would have to use it as a plug-in so to try to make everything equal is i exported this image twice from lightroom each of them are tiff files and they're identical so in gigapixel ai i have the tiff file open up in it and in photoshop which essentially now is adobe's camera raw it's open there as well now i want to make note here and i mentioned this in that video where i introduced super resolution that again i'll have linked in the description below this video i mentioned in that video that super resolution is only available in the full non-filter version of camera raw meaning if you have an image open up into photoshop and you go up to the filter menu and down to camera raw you won't find super resolution there you have to open an image directly into camera raw then you'll find super resolution and to get to it all you need to do is right click right on the image and go down to enhance when you do that you'll see that there's one checkbox super resolution it's as simple as that just make sure that check box is checked then click enhance so there's really no control here you're just doubling the length and doubling the width of the image and that in effect quadruples the megapixel size so down here you'll see in the film strip we now have two images this is the original image you can see at the bottom it's 1797 by 2692 4.8 megapixels next to it is our enhanced image it's 3594 by 5384 19.4 mega megapixels so there is our super resolution image now by default camera raw saves this as a dng file but i think most people when they send an image through super resolution or they use gigapixel ai or any other app they're using it because they want to print it and typically you would print from a jpeg directly or you would send that jpeg to a professional photo lab for them to print it so i'm going to open this super resolution image up into photoshop and then from photoshop i'm going to save it as a jpeg and then we'll compare jpeg to jpeg so we're trying to make this as much of an apples to apples comparison as possible so i'm going to go up to file down to export export as and we're going to save it as a jpeg with 100 quality i'm not going to adjust the width or height at all so it's the full scale hundred percent we're going to convert a test rgb though and we're going to export this it's going to be exported to the same folder and remember i mentioned that adobe camera raw creates a dng file and you can see it's right there that's our image in the dng version there's the original tiff file that i opened up into photoshop it's camera raw and we'll just save this so it's saving it and now i'm really done with photoshop now um it's when i go to close it's going to ask me to save it but it's already saved enough so we'll click don't save now we'll deal with gigapixel ai now again i have that same exact image opened up into gigapixel ai now right off the bat gigapixel gives you a lot more control um with super resolution you only get 2x twice the resolution here i could make it smaller i could go 2x 4x 6x or i could just put in my own multiplier so i can make a 10x 20x if i want i also could come in and just put a specific width in that i have in mind and it will be proportional it will adjust the height accordingly or i could put a specific height that i want and again it's proportional it will put the correct width in when i do that i could crop it here if i wanted to and as far as that we'll stay with the 2x because we're comparing apple's tables and we have four different modes ai modes the standard mode is meant for most images and you can see that whenever i just move it slightly it's going to generate a new preview so it has to do this every single time but just looking at it because there is noise reduction built in and if you look in the original image on the left and i'll zoom in in post production so you can see it there is a significant amount of noise in the background of this image and it's cleaned up perfectly in the processed image so that is an advantage to at least so far we'll look at the jpegs in a minute but that's an advantage to gigapixel ai now we'll go over look at the other modes architectural mode is meant for real estate photography or cityscapes something like that and i have all four of these modes by the way set to auto so it's automatically uh putting suppressed noise and remove blur on a specific setting that it feels is good for this image now right here with the architectural mode it looks a little sharper it did remove most of the noise just like the other ones so it's as far as noise reduction it's identical to standard it seems to be a little sharper though compressed mode is meant for jpegs jpegs usually are compressed and when you compress an image there's often what they call artifacts jpeg artifacts you might have heard and this will help suppress any artifacts on a compressed image and this actually does look sharper than standard mode as well of course i could come in and move sliders around and do it myself now art mode is meant for what it means art so if you have a drawing or a painting something you've done or something you've done digitally or something scanned you'll probably want to use art mode now as far as the modes are concerned i could go to standard mode because it's a standard image it's not really any of the other three and i could come in and try to move around the sliders remove more blur suppress less noise maybe that'll make it a little sharper but you know what i'll just go to architectural mode it looked a little sharper and the noise reduction was fine and i'll just leave it there i could again come in and move those sliders around you also have some additional settings if you do usually art sometimes you have colors bleed especially reds on greens and things like that you would flip this switch and hopefully that will take care of that also if you have a person in the image typically you don't want their skin sharpened as much so you would uh turn on face refinement it will recognize the person in the image and it will make sure that it doesn't enhance anything you don't want to enhance like pimples and things like that but i'll just stay with the straight architectural mode so you can see though in gigapixel ai there's a lot more uh options here uh to either confuse you or to hopefully uh help you create a better enlargement so we'll click save image and when it does that i'm going to again save it as a jpeg with maximum resolution i'll leave that same file name we'll save it to the source directory and we'll use that color profile srgb because that's what i used when i exported the photoshop image so we'll click save now one thing about gigapixel it takes a long time to save so it's going to take a little while so i'll pause the video and we'll come back once it's fully saved okay it's fully saved in gigapixel ai i'll just close that down we'll minimize lightroom we don't need that anymore now here are the two folders i mentioned uh in the first folder the photoshop folder the tiff file is the original file the jpeg is our exported image and the dng file over here we have our exported jpeg and the original tiff file so what we want to do is we want to compare the jpegs all right so we'll open that up and there is our photoshop jpeg and we'll open this up and there is our gigapixel jpeg we'll put that on the right and we'll try to look at them side by side and uh the gigapixel one is sharper uh definitely just a little bit sharper and there is um no noise in it so i'm going to blow this up so you can get a better view first i'll blow up the photoshop version then i'll blow up the gigapixel version and you can decide which one you like better but you know really they both did a really good job if you ask me so um i think gigapixel because it's more versatile and you know if you're a wildlife photographer and you're off often cropping a lot and you often will be shooting at higher isos and things like that gigapixel might be the better choice whereas if you're a travel photographer just you know uh taking images from your travels your vacations your holidays things like that you're probably not cropping quite as much and if you do the odd crop here or there i think photoshop works great you could use that and not have to invest in gigapixel at all so that's my comparison it was i tried to make it as apples to apples and pos as possible but it was kind of difficult to do but either way i hope that helps you decide which application would work best for you thank you everyone watches my videos i really do appreciate it i'll talk to you guys soon you
Info
Channel: Anthony Morganti
Views: 41,111
Rating: undefined out of 5
Keywords: anthony morganti, morganti, Sky Replacement, Photoshop Skills, How to replace the sky, Adobe Photoshop, Photoshop tutorials, Photography, Easy photoshop tutorials, Easy Photo Tutorials, adobe camera raw, adobe camera raw tutorial, adobe camera raw 2021, adobe camera raw 13, acr, raw images, raw files, raw files in lightroom, raw files in photoshop, super resolution, gigapixel ai, gigapixel versus super resolution, gigapixel versus photoshop, enlarging images, enlarging photos
Id: b6Mmope9eh0
Channel Id: undefined
Length: 13min 10sec (790 seconds)
Published: Tue Mar 23 2021
Related Videos
Note
Please note that this website is currently a work in progress! Lots of interesting data and statistics to come.