Latest opinions on Software for enlarging???

esther_edmond · March 19, 2011

Hi, I'm looking for input on the latest software programs used for significantly upsizing images (ie for canvas printing etc). As these programs continually upgrade and improve, I'm interested in hearing some current opinions and experiences.

I've read a lot of reviews and tried several programs with varying results. I've downloaded trial versions of Perfect Resize (latest version of Genuine Fractals), PhotoZoom Pro 4 (BenVista), and Alien Skin BlowUp. I also have CS5. It's easy enough to compare features and ease of use. But it gets a little more complicated when comparing image quality on final output. You really need to understand the best settings to use for each program to fairly compare results. I sent an email to each company asking their recommended settings for enlarging to canvas and received varying replies of helpfulness. The "industry standard" (GF) provided the least guidance, saying "check our tutorials." I'd already done that, they were pretty basic.

For anyone interested in this topic, I did find an interesting tutorial on enlarging at Cambridge: <a href="http://www.cambridgeincolour.com/tutorials/digital-photo-enlargement.htm">http://www.cambridgeincolour.com/tutorials/digital-photo-enlargement.htm</a> It explains some of the different interpolation methods used by some of these programs.

I've also read several comments suggesting QImage. It's really more of a printing software, but apparently does a nice job on resizing as well--at about half the cost.

I also wonder if CS5 can do just as good a job if I knew how to employ better sharpening techniques after enlarging--thus alleviating the need to buy a plug-in altogether. But so far, I haven't been able to get the CS5 output to look as good as the others.

In experimenting, it's clear that results depend on how adept you are at using the appropriate settings for each program and for the individual image. On the first go-around, (without too much knowledge of what I was doing) Perfect Resize (GF) seemed to provide the best rendition without being too sharp in areas that didn't need it--however it was also slow as molasses (not when rendering the final output, but each time I'd tinker with a setting it would take forever to respond while the other programs responded almost immediately). It didn't seem to be a system issue as my set-up and video card exceed their recommendations. The clunky response time made the program very frustrating to use and their phone tech was not much help either. The other programs were MUCH faster so you could easily evaluate each change to their settings.

So other than suggesting the purchase of a 5D Mark II or stitching together several images, what has been your experiences with interpolation programs and which do you prefer? I'd also sincerely appreciate any suggestions on the right post-sharpening techniques in CS5 to possibly accomplish the same thing. Thanks for your thoughts and time to reply!

john_a5 · March 19, 2011

At one time I owned Genuine Fractals--version 1 and 2. At the time, I think it did a better job than PS did--probably around PS4, not PSCS4! Anyway, over time I didn't see a need for GF as photoshop seemed to do a very good job. Last year, just to compare the current state of the art, I downloaded the then new GF and compared it with PSCS5. I enlarged images from 60mb (8bit) to 850+mb file sizes. I did find a very negligible difference in some places, but that was looking at the file at 100%, at the print size (40x60 @ 360dpi), there were no differences noticeable at all.

I don't know if it would help if you were shooting really small files or not, but I doubt it would be all that different to use one of these other products.

digitaldog · March 19, 2011

See:http://www.digitalphotopro.com/technique/software-technique/the-art-of-the-up-res.html

garrison_k. · March 19, 2011

I also wonder if CS5 can do just as good a job if I knew how to employ better sharpening techniques after enlarging--thus alleviating the need to buy a plug-in altogether. But so far, I haven't been able to get the CS5 output to look as good as the others.

Esther, it's in my experience that Adobe is a step behind the great plugin makers. I've witnessed Adobe forever playing catch-up and then bring us, for example, pano and hdr tools after great software has already brought them to market. Adobe, it seems to me, purchases the rights from others or reverse engineer.

Andrews' link is extremely dated. I'd practice any of the methods mentioned with a grain of salt or just seek more current tutorials all together.

robert lee · March 20, 2011

I feel that the best looking resizing results come from programs using the Lanczos algorithm. Qimage is one. I generally use <a href="http://www.dl-c.com/">PictureWindows </a>, and ImageMagik on the command line when batching or when specific control is beneficial.

Go see the samples at <a href="http://www.americaswonderlands.com/digital_photo_interpolation.htm">AmericanWonderlands</a>. They have side by side images processed with Lanczos, Genuine Fractals, and a bunch of others.

Ed_Ingold · March 20, 2011

Photoshop resampling works just fine up to 2x. You are kidding yourself if you enlarge any further. If you want bigger, shoot bigger.

digitaldog · March 20, 2011

Andrews' link is extremely dated. I'd practice any of the methods mentioned with a grain of salt or just seek more current tutorials all together.

</blockquote>

What should be taken with a grain of salt is the quote above. Clearly Mr. K once again made a comment without doing his homework! He should practice scientific testing.

Point in fact, if he had attempted to upsize an image in Photoshop CS2 (which was the current version of Photoshop when Jeff Schewe’s articles was written and mentioned as being used in the article) and CS5, he would have seen that they are identical! Upsizing the same image using BiCubic Smoother in both, then using the Apply Image command>subtract command, as described in this tutorial (http://digitaldog.net/files/Apply_Image.pdf), shows they are pixel for pixel IDENTICAL.

Are the upsizing in ACR which Jeff recommends identical between ACR 4 and ACR 6? Nope, ACR 6 processing is actually improved! So the techniques described in Photoshop are identical should you go that route, the upsizing and ACR pipeline are different and improved in more modern versions of ACR (certainly ACR 6.X). There’s a bit less flexibility (there are only 3 fixed upsizing options in the ACR Workflow Options dialog). And ACR as well as LR have undergone a totally new and improved Process Version (PV 2010).

In addition, the PhotoKit sharpening routines Jeff recommends are identical today as they were in 2007. So you can dismiss the piece Jeff wrote if you wish, but its not due to its age, its still as relevant 4 years later as it was when published.

Jeff’s conclusions are still valid:

The top image was up-resed in Photoshop CS2 using Bicubic Smoother to get from a 6-megapixel native resolution to a nominal 25-megapixel interpolated file size. The bottom image was up-resed in Camera Raw 3.1 using the 25-megapixel interpolation option. Both images are being displayed at a Photoshop screen zoom of 1200% to show the results of each method. They're essentially identical for all practical purposes. (AR Comment: today, the ACR version should be superior.)

Well, I told you that this whole thing was going to be problematic, remember? But I think you get the point. If you start with a technically correct image and do the proper image processing, you can get a good result up-resing even up to 400%.

</blockquote>

garrison_k. · March 20, 2011

Is USM still in CS5? Kidding aside Esther, look for up-resizing tutorials that have sharpening done in Lab mode and not ones that carpet-bomb the whole image with USM. Edward's advice is bang on, too.

Robert, I use the firefox extension "WOT", or Web of Trust, and the link you provided for americas wonderlands claims the site has a poor reputation in terms of trustiness, privacy, and vender reliability. A malware/spyware caution to anyone clicking there.

http://www.americaswonderlands.com/digital_photo_interpolation.htm

digitaldog · March 20, 2011

No reason to convert to Lab to sharpen (time and data loss results), when you can simply do any sharpening routine you want and Fade with Luminosity. The idea here is to sharpen without affecting color, plus you have control over the strength with the opacity slider.

garrison_k. · March 20, 2011

No reason to convert to Lab to sharpen...

There's a whole internet out there that will argue that one.

No reason to convert to Lab to sharpen (time and data loss results),

I've never noticed data loss and this is the first time I've ever heard that it might be an issue. I'm also not sure how you can complain about time loss (by going to Lab) when you suggest fiddling with sliders? Which is a good idea, no argument. I'm just unsure of the point you are trying to make in terms of shaving a couple tenths of a second off that it takes to get into Lab mode. The topic is enlarging which implies time will be spent tweaking and experimenting the image for best results. I know I do, anyways.

Esther, take what you wish in terms of many methods of skinning your cat. Just research them and best of luck. I've done what I can here...

digitaldog · March 20, 2011

Any color space conversion causes data loss. If you do this on a high bit file, its moot. If you do it on a 24 bit image, its not moot and in fact, banding can occur upon output especially if more edits are applied after the conversion (you’ll have to apply yet another color space conversion to make a print).

The facts are, the math proves that any time a conversion to Lab is produced, the rounding errors and severe gamut mismatch between the two spaces can account for data loss, known as quantization errors. The amount of data loss depends on the original gamut size and gamma of the working space. For example, if the working space is Adobe RGB, which has 256 values available, converting to 8- bit LAB reduces the data down to 234 levels for neutrals. The net result is a loss of 22 levels. Doing the same conversions from ProPhoto RGB reduces the data to only 225 values, producing a loss of 31 levels. With 256 original values pre color, without factoring in previous and additional edits, this one edit can result in banding on output depending on the output device and the image content. Why even go there when the move is completely unnecessary?

This is simple math. Then there’s the time hit converting into and out of Lab when you can produce the same benefits (a lack of color shifting after sharpening) by using Fade which produces more control options and is far faster too.

robert lee · March 20, 2011

Robert, I use the firefox extension "WOT", or Web of Trust, and the link you provided for americas wonderlands claims the site has a poor reputation in terms of trustiness ...

</blockquote>

That's too bad.

Here's the overview image. The author American Wonderlands author interpolated from 1407x612 pixels to 9000x3915

The following is the claimed result from Genuine Fractals 3.0

The following done with Lancosz (no detail though on whose implementation and other parameters.)

garrison_k. · March 21, 2011

Robert, thank you for going to the trouble of posting for us. WOT has been great for keeping spyware and malware off of my PC.

patricklavoie · March 21, 2011

talking about sharpening, i dont either see the why of converting to Lab personnaly... hell, i dont see the why of converting to Lab for anyhting.. I really know what Lab is, what people do with it.. but most of it if not all can be done by using different blending mode using layer.. anyway that was not the point of the original question. ; )

I also ear of data lost when converting form rgb to lab, lab to rgb but i never personnaly noticed it...

As for interpolation, i have made multiple test myself, and find that Photoshop alone can do amzing result and fast one up to 200%.. at 400% you can start seing some artifact appearing up close but nothing that you can really put your gfinger on from a distance.. at 800% all my test show that most of all the image i test with different software look crappy from close, and all look good from a distance.

What i also discover is many software give you the impression of a beter smooth finish, but looking at the result it look like aquarel painting.. all the detail seem kind of have a watercolor paint effect...

___

edit*

as you can see in Robert example, looking at Genuine Fractal result.. you can see what i mean about painting effect really well there.

garrison_k. · March 21, 2011

Patrick, you work in 8 bits and you don't notice data loss? Like you and everyone else, I only "hear" of data loss once in awhile and is always in terms of math and theories typed out in forums. Also like you, I never visually seen it on the monitor and never on print. The net is filled with articles on the benefits of Lab mode and its preference over RGB. Maybe I'm stuck in my ways, but it works for me and I've yet to come across anyone saying it is a mistake. For those interested in Lab, start with reading/watching Dan Margulis.

esther_edmond · March 22, 2011

Hey all, thanks for the suggestions and lively debate. I've actually visited several of the posted links, including the comparison shots. The tough thing about these comparisons (as mentioned) you need to understand the settings for each of these different programs-- one may look better only because the settings were handled more appropriately with one program vs. another.

My eyes and brain are tired--too much pixel peeping in comparing programs. After experimenting with different sharpening techniques in CS5, things look better than my first attempts, but still not completely satisfied with the results. The interpolation of Photoshop just doesn't seem to be as good as the plug-ins unless you know how to carefully manipulate sharpening afterwards with all kinds of "fancy maneuvering"--and I still see stair-stepping that I don't see in some of the other programs. HOWEVER--I do agree, some of the other programs have kind of a "watercolor" look about them which doesn't seem all that satisfying either.

Again, if anyone has specific suggestions about how enlarge with PS and get better results with certain techniques, please share or send a link. I tried a few suggestions found on Thom Hogan's site (edge and contrast sharpening) which got me a bit closer but still not there yet. Also, if anyone has specifics on sharpening for canvas output, that would be much appreciated as well, seems like I did read that a larger radius is required, any ideas?

Oh, and any thoughts on why manipulating in Genuine Fractals is so much slower than the rest, I'd probably go that route and be done with this, if it weren't for the aggrevation factor!

Can't wait for an affordable, full frame, higher MP from Nikon; 12MP doesn't quite cut it if you want to print large, (yes, I know I can stitch). With the events in Japan, I have a feeling the arrival of new models may be quite delayed. The situation over there is such a tragedy, definitely puts my angst about this subject in perspective.

digitaldog · March 22, 2011

I've actually visited several of the posted links, including the comparison shots. The tough thing about these comparisons (as mentioned) you need to understand the settings for each of these different programs-- one may look better only because the settings were handled more appropriately with one program vs. another.

</blockquote>

IF the output is to a printer which I assume it is, you really have to make a print. Viewing the differences on screen isn’t telling. Printing the iterations identically on a good printer is.

digitaldog · March 23, 2011

Also like you, I never visually seen it on the monitor and never on print.

</blockquote>

Here’s a section of a high rez file, Adobe RGB (1998), 8-bit and its cousin, the same but converted to Lab and back. Both images have identical edits: a dose of USM and a slight Hue/Sat move. Its zoomed in to 300% to pixel peep. Its pretty obvious which is which. Its not night and day, but it shows what happens when you post edit images after a Lab conversion vs. just doing the same edits without that move into Lab. We also have to conduct an output color space conversion to print either and we have no idea what other edits prior or after would be applied to this data to reach our goals. Its these kinds of smooth gradients that suffer the most. Going from Adobe RGB (1998) to Lab and back is going to toss about 22 levels alone, that’s why you see this effect in such areas. sky, chrome bumpers, glass like this etc.

http://digitaldog.net/files/8-bit_vs.8-bit_to_LAB_Back.jpg

garrison_k. · March 23, 2011

...but it shows what happens when you post edit images after a Lab conversion vs. just doing the same edits without that move into Lab.

Interesting. Is this your own conclusion? I ask because it's the first I've heard and would appreciate an article or two showing the world isn't flat as I think by now I'd come across an argument or two against Lab mode sharpening. In the meantime, the result you should have had is smoother gradients and less artifacts while applying more aggressive sharpening. Anyone can do this test on their own and come up with a different result than you have here. Also, I'm curious why you did it in 8-bit and used USM instead of Smart Sharpen in 16 bit? Did you do this in Lightness Channel when in Lab mode, or just applied USM after then mode change into Lab?

digitaldog · March 23, 2011

I ask because it's the first I've heard and would appreciate an article or two showing the world isn't flat as I think by now I'd come across an argument or two against Lab mode sharpening.

</blockquote>

I suspect there’s a lot you haven’t heard of but there are all kinds of posts and pieces that explain why its unnecessary to convert to Lab for sharpening to remove the results of color additions from sharpening by using Fade Luminosity. Prior to this functionality in Photoshop (which goes back many, many versions), converting to Lab to just the Lstar channel may have been necessary if the goal was not to produce an alteration of color values due to the use of USM.

But OK, since apparently doing your own testing, or viewing mine isn’t concrete enough, since you must have an article, here you go:

http://lightroom-news.com/lightroom-11-update/sharpening/

In the early days of Photoshop, people sometimes converted an RGB image to Lab mode and sharpen the Luminosity channel separately. This technique allowed you to sharpen the luminance information without sharpening the color content. More recently (from Photoshop 3.0 onwards), it has been easier (and less destructive) to sharpen in RGB mode and use the Luminosity blend mode to restrict the sharpening to the luminance information.

</blockquote>

Or: http://bythom.com/sharpening.htm

A better way to sharpen just the luminosity channel without changing to Lab Color is to perform sharpening as you usually would (e.g., using Unsharp Mask), then choose Fade Unsharp Mask from the Edit menu. In the dialog that appears, change the Mode fromNormal to Luminosity and click on the Okay button.

</blockquote>

Or: http://www.ronbigelow.com/articles/sharpen4/sharpen4.htm

Once the OK button is clicked, the sharpening will be applied to the tonal values only. This produces a sharpening that is almost identical to LAB sharpening, but it does not require any conversions (for those interested in the difference between lightness and luminosity, see the Lightness and Luminosity box). Thus, the use of the Fade tool does not cause the type of image quality loss that is concomitant with the use of LAB sharpening.

</blockquote>

I’m sure there are many other such references you can find using “the Google”.

In the meantime, the result you should have had is smoother gradients and less artifacts while applying more aggressive sharpening.

</blockquote>

No because the trip to Lab removed 22 levels from the original 255 per channel. The USM isn’t necessary to produce similar results, try it. The facts are, the higher the bit level, the smoother gradients can be, especially after additional processing of the data. Its WHY our cameras and scanners operate in high bit. Its WHY Photoshop has so many high bit functions. Or is it your take that all this high bit functionality is nonsense, the engineers at Adobe, Nikon, Canon, Phase1 etc all give us high bit capture and editing solely for marketing reasons?

Also, I'm curious why you did it in 8-bit and used USM instead of Smart Sharpen in 16 bit?

</blockquote>

First, the idea here is to illustrate the effects of editing 8-bit per color images (and the negative effects on smooth gradients after tossing away 22 levels just to sharpen the Lstar channel). 2nd, based on a lot of serious work by my late colleague Bruce Fraser, when working on PhotoKit Sharpener, he find nothing at all useful in Smart Sharpen to incorporate into our product.

Did you do this in Lightness Channel when in Lab mode, or just applied USM after then mode change into Lab?

</blockquote>

Both files have identical processing. The USM was applied after conversion to Lab and back to RGB. It is not possible to have an apples to apples comparison any other way, the original data is in RGB (Adobe RGB to be exact), and applying USM the Fade Luminosity which would eliminate color artifacts as doing this on Lstar would produce differences in said processing compared to applying USM on the Lab file. So again, the effect of the loss of smoothness in the gradients is the effect of losing 22 levels just going from Adobe RGB to Lab and back on an 8-bit per color file. Additional edits were applied because many people will continue to edit their images and the more they do so, the more the compounded data loss piles up. At the very least, they have to apply an output profile to convert the data for printing. There is ZERO reason to convert to Lab to sharpen! You can retain all the data (well the original 22 levels) AND produce the same benefits of Lstar sharpening by using Fade Luminosity and do so faster with more control. Its WHY our Adobe engineer friends provided this very useful blend mode!

But at least NOW you HAVE seen the effects of which you said you never saw on-screen or on the print. I would encourage you (and others) to try this with other images and edits. OR at the very least, if you just have to convert into and out of Lab, do it on a high bit file.

garrison_k. · March 24, 2011

But OK, since apparently doing your own testing, or viewing mine isn’t concrete enough, since you must have an article, here you go:<a rel="nofollow" href="http://lightroom-news.com/lightroom-11-update/sharpening/" target="_blank"> http://lightroom-news.com/lightroom-11-update/sharpening/</a>

That's LR 1.1. Sure brings back memories. And it's even older than your 2007 blog post you first pointed to. The link from Thom is from 2003. The Ron Bigelow article is just as badly dated. Seems your "tests" and your references are all based on 8-bit from the previous century.

But OK, since apparently doing your own testing...

My testing? Right. I'm not smart enough to come up with this Lab sharpen thing on my own, publish it, have the guru's and god's out there test and confirm and have the results become a standard. But I am smart enough to do what the expert suggest and follow procedures and voila, my results are just like theirs. Anyone can do their own testing and they'll find that one can apply more aggressive levels of sharpening in Lab than RGB before you introduce artifacts, halo's and ugly gradients etc. So in the meantime, I'm sticking with Lynda.com, Dan Margulis and Adobe et al instead of the grumpy old "Digital Dog" here on PN that's pointing to a 5 year old blog (that I found hard to believe had fuzzy un-readable screen captures that accompanied the tutorial on how to up-rez).

Both files have identical processing. The USM was applied after conversion to Lab and back to RGB.

Well there you go. You're not doing it the lightness channel of Lab where sharpening is applied, you're doing it in 8-bit then mucking around with hue and saturation afterwards. Can you break any more rules in your quest for validity?

I’m sure there are many other such references you can find using “the Google”.

And so I did. Reference/instrcution here

<a href="http://www.lynda.com/Photoshop-CS5-tutorials/prepress-and-printing/Sharpening-in-Lab/65778-4.html">www.lynda.com/Photoshop-CS5-tutorials/prepress-and-printing/Sharpening</a>

<a href="http://www.lynda.com/Photoshop-CS3-tutorials/mastering-lab-color/Sharpening-the-Lightness-channel/36969-4.html">www.lynda.com/Photoshop-CS3-tutorials/mastering-lab-color/Sharpening</a>

<a href="http://www.photoshopsupport.com/tutorials/cb/sharpening.html">www.photoshopsupport.com/tutorials/cb/sharpening</a>

<a href="http://www.photoshopsupport.com/photoshop-cs3/forensic/sharpening-in-lab-mode.html">www.photoshopsupport.com/photoshop-cs3/forensic/sharpening-in-lab-mode</a>

It's a simple test anyone can do to make up their own mind. Open a 16-bit file and keep it in 16-bit. Then image>mode>lab, then get off layers and go to channels, select on lightness, and because we are testing and wish to make it obvious, lets apply an obscene amount of Smart Sharpen, like 150 and 3.5, then back we go to rgb, so it's mode>rgb. That one's done (took a real long time eh). Now open the original a second time but we'll stay in rgb and apply the same amount of Smart Sharpen. Maybe it's best to make a layered psd file for quick comparison? Lets put that rgb file on top of the first lab file in order for us to click the eye icon and view the layers easily. The empirical evidence is enough for me or one can seek out some of the (modern) tutorials on Adobe Tv etc.

digitaldog · March 24, 2011

That's LR 1.1. Sure brings back memories.

</blockquote>

No, its Photoshop! Read it again, Martin is referring to converting to Lab vs. using Fade Luminosity in Photoshop (as are all the other references I supplied). LR has nothing to do with, nor is the age of the article(s) much like the first post you made incorrectly dismissing the resampling in CS2 and you smartly ignored once I pointed out that nothing has changed since then. Nothing has changed in terms of Fade vs. Lab sharpening since LR1!

I'm not smart enough to come up with this Lab sharpen thing on my own, publish it, have the guru's and god's out there test and confirm and have the results become a standard.

</blockquote>

That’s possible, I will not argue the point.

Well there you go. You're not doing it the lightness channel of Lab where sharpening is applied, you're doing it in 8-bit then mucking around with hue and saturation afterwards. Can you break any more rules in your quest for validity?

</blockquote>

You are missing the point but I fully expected that! But lurkers are most likely not missing the point which is the only reason I continued. The post processing HAS to be identical or the test is invalid. Forget USM, the point is, you tossed away 22 levels just going RGB to Lab and back. Then you apply additional edits and you see banding because you tossed away 22 levels for no reason. It illustrates the damage done to the data solely due to the lab conversion where the original data was left in RGB and suffered none of this banding. Get it? Remember what you wrote a few days ago (Patrick, you work in 8 bits and you don't notice data loss?Like you and everyone else, I only "hear" of data loss once in awhile and is always in terms of math and theories typed out in forums...I never visually seen it on the monitor and never on print.)? Well now you see it. If you look.

The test and debate has ZERO to do with Lstar vs. Luminosity routines on color. We can go there but you first need to get your head around the concept of data loss just from the RGB to Lab to RGB conversion on 24 bit data. Its unnecessary and has been since Photoshop 3.

Open a 16-bit file and keep it in 16-bit.

</blockquote>

No, open an 8-bit per color file and run the two images as I did, then you can run us down a different rabbit hole and move into Smart Sharpen. The three articles I referenced, my tests, the feedback from Patrick are on topic which is, there is no reason to convert to Lab to sharpen and doing so on a 24 bit document shows visible data loss from edits done after the fact. See if you can stay on this simple topic, then digress into other subjects if you must. Or just ignore what are mathematical facts and images that illustrate the data loss from the Lab conversions.

FWIW, with the somewhat poor images you posted in line (no ability to see larger images as I did), it appears that the RGB smart sharpen has smoother gradients than the Lab version. But without better viewing methods, I’ll let that slip for now. Since it was presumably done in 16-bits per color, it has nothing to do with this topic (data loss on 24-bit data from conversions to Lab and back when the routine is unnecessary).

digitaldog · March 24, 2011

And so I did. Reference/instrcution here <a rel="nofollow" href="http://www.lynda.com/Photoshop-CS5-tutorials/prepress-and-printing/Sharpening-in-Lab/65778-4.html" target="_blank">www.lynda.com/Photoshop-CS5-tutorials/prepress-and-printing/Sharpening</a> <a rel="nofollow" href="http://www.lynda.com/Photoshop-CS3-tutorials/mastering-lab-color/Sharpening-the-Lightness-channel/36969-4.html" target="_blank">www.lynda.com/Photoshop-CS3-tutorials/mastering-lab-color/Sharpening</a> <a rel="nofollow" href="http://www.photoshopsupport.com/tutorials/cb/sharpening.html" target="_blank">www.photoshopsupport.com/tutorials/cb/sharpening</a> <a rel="nofollow" href="http://www.photoshopsupport.com/photoshop-cs3/forensic/sharpening-in-lab-mode.html" target="_blank">www.photoshopsupport.com/photoshop-cs3/forensic/sharpening-in-lab-mode</a>

</blockquote>

Adobe Creative Suite 3 (CS3) was announced on March 27, 2007, Adobe announced that Lightroom 1 would be shipping on February 19, 2007. So should I dismiss these old URLs of yours to be consistent with you dismissing the content I posted? Don’t do as I do, do as I say?

Andrews' link is extremely dated. I'd practice any of the methods mentioned with a grain of salt or just seek more current tutorials all together. That's LR 1.1. Sure brings back memories. And it's even older than your 2007 blog post you first pointed to. I'm sticking with Lynda.com, Dan Margulis and Adobe et al instead of the grumpy old "Digital Dog" here on PN that's pointing to a 5 year old blog.

</blockquote>

esther_edmond · March 25, 2011

Goodness boys, it's getting mighty hot in here!

I could probably spend the next 4 weeks experimenting and reading all the posts presented in the last few go-arounds. I can see there are no simple answers--but then, nothing good in life is ever easy.

Maybe after experimenting with some of these suggestions, I'll start a new thread to share a few thoughts regarding the enlarging software programs I've been comparing--the original post, BTW :) ! I am sure my insights will not approach the PhD level thinking displayed here, but might provide a few helpful tidbits to my fellow undergrads. :)

Thanks fo much for the "back and forth," it's been very informative--and fun!

garrison_k. · March 25, 2011

Let me get this straight. The rule is to shapren in Lightness channel. But you're trying to discredit this method and a whole whack of respected professionals by doing something (sharpen in A&B in Lab instead) that they say we're not supposed to do in the first place?

I have an idea! We'd like to see you do as I have done and post a string of images here comparing both methods. Do yours in RGB and another exactly how sharpening in Lab mode is supposed to be done. Maybe you can draw circles around the smoother gradients in your RGB and the artifacts in the Lab version?

FWIW, with the somewhat poor images you posted in line...

You're a first class guy, Andrew Rodney. It's also pretty funny coming from the owner of that gallery of yours.

Sign In

Latest opinions on Software for enlarging???

Recommended Posts

Link to comment

Share on other sites

Top Posters In This Topic

Popular Days

Top Posters In This Topic

Popular Days

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Create an account or sign in to comment

Create an account

Sign in