I can only give you feedback from a couple of people at work who have used both and they both prefer pxl smart.
user interface was much better and one said it was more efficient(speed)
Why waste the extra money if you have the new photoshop. and if you don’t have it, why not save your money and use it to upgrade to photoshop. You will be better off in the long run.
I just attended a recent seminar. Two pix of the same subject enl. to 16×19 – one enl. with GF and one with PS "stairstep" interpolation. In blind observation, most people could not see any difference, while two or three thought the "stairstep" pix was sharper. I agree with Buko, spend your money on PSCS.
Unless I’m mistaken, I believe the Geniune Fractals format has very good compression ratio. That could be a factor in deciding to use the .stn format rather than .psd.
Genuine Fractals is way too slow for anything other than occasional use. It doesn’t work on 16 bit images either.
I agree, stick with Photoshop.
Gary Smith
Also PSCS has the Bicubic Smoother option for upsampling and Bicubic Sharpening for downsampling.
Asked this once before but no one responded. I’m not following the logic of using Smoother for upsampling and Sharper for down. If upsampling’s gonna "soften" my image by interpolating pixels, my pea brain says SHARPEN as I interpolate, not smoothe. Why wouldn’t I want to sharpen? The only answer I’ve conjured on my own is that Smoother somehow protects against artifacts. Is this it? And if it is, why wouldn’t I want to use Smoother going down as well?
I also understand (and see in my own experiments) that these new options produce subtle differences at best. Nothing dramatic here. Are my eyes seeing correctly?
Thanks.
From color theory list:
#1
1200 x 800 @ 300 ppi to 2126 x 3190, 6 steps at 110% ea…Bicubic.
#2
from same original to 2126 x 3190, single step… Bicubic Smoother.
#3
from same original to 2136 by 3190, Genuine Fractals, two steps – first save in STN, reopen and resize.
Compared all three at 100%, 200% and even at 1200%
and could not see much of a difference in any of them. Some minor color shift measured with GF and Smoother Bicubic being the closest.
Bicubic Smoother will do the trick in one step.
Nothing wrong with your eyes. Personally I prefer the Sharper method for 200%, 400%, 800% etc. I have not tried downscaling, but I’m told that extreme scaling, either way, will benefit from one algorithm or the other.
I am actually very impressed with the scaling options. The difference between Bicubic, for example, and Nearest Neigbor is remarkable for something that is so easy to use.
When I saw Buko’s suggestions last night I was in the process of down-sampling a few photos from about (5400 x 3600) pixels to (800 x 533) pixels. Bicubic Sharper made a noticeable, but not huge improvement.
If I were to downsample images and forget to use Bicubic Sharper I wouldn’t loose any sleep over it and not do the work over. I think it is worth the effort if I do think of it.
Gary Smith
Thanks, guys. But I was assuming there was some DIGITAL PRINCIPLE behind Smoother when upsampling (and Sharper when downsampling). In other words, even if in practice the differences are going to be subtle, I assume there’s still a reason that recent expert counsel on this subject recommends the use of Smoother for upsampling. Anyone know it?
Perhaps it’s because sharpening is one of, if not the, last things you should do to an image before declaring it finished. So if you want to do further work on an image after upsampling it, you would be better off not sharpening it and then use USM when ready. But then you’ve lost the benefit of any temporary data that could have been using during the upsampling stage. Just a guess; I’ve no idea if it’s remotely accurate or not.
Better guess than I could make, Graham. But then, why use Sharper when downsampling? Hmm. Who do we talk to about this?
Scott Byer or Chris Cox probably knows. In fact I think Scott explained it a while ago but I can’t remember where. I do remember that the standard Bicubic method (not the smoother or sharper variants) was improved (increased quality at the expense of a small performance hit) between 7 and CS.
OK. Gonna start a new thread and see if we can entice either one or both to respond.
Pardon me, maybe I’ve missed something. Since when can you enlarge a photo in Photoshop from a 72 d.p.i. image to a 3 ft. x 10 ft. banner image and have no pixel distortion on the final output?
Since when can you enlarge a photo in Photoshop from a 72 d.p.i. image to a 3 ft. x 10 ft. banner image and have no pixel distortion on the final output?
Who said anything about doing that?? also what size 72 ppi image 3ft by 10ft?? no problem.
