Sorry you didn't realize it but look and listen again. I'm zoomed in to 100% initially and zoom out for an overall view (time at 1:19) and as I both show visually and explain verbally (starting at time 1:27), invoking AI Clear automatically zooms to 100% (again) You can verify this by looking at the image size controls at the top right. Pixel peeping (at 100% or more) is fine to satisfy yourself of the changes but you also need to consider what you are doing with the final image. If you're going to make a large print then viewing at 50% or so (depending on your monitor) will be more analogous to scrutinizing a large final print. If you're going out to Instagram or Facebook then it is far less critical. Happy shooting, Joel

It's situational but for most people I'd give the nod to A.I. Clear. A.I. Clear is so much easier to use and does a great job most of the time so it is my go-to for noise reduction AND it sharpens at the same time. If you want to have a lot of manual controls and you don't mind spending time learning it you may like DeNoise better. Remember that DeNoise is a plug-in so the integration isn't as elegant as A.I. Clear. It's nice to have both but if you made me pick one, it would be A.I. Clear. Hope this helps

JoelWolfsonPhoto thank you sir💫

I haven't run into this with portraits (or otherwise.) Feel free to send me a sample file that is giving you this problem. I'd be interested in seeing it. Be sure to send me the unprocessed file so I can try it on my end. Also if you are truly unhappy with it I would contact Topaz about it, along with some samples of your issues. They may give you refund even though it's beyond their return window. It's worth a try.

I had same problems. A lot of artifacts along the entire frame (not only portraits). Note that the AI Clear preview shows excellent results, which don't match those of the saved file. I've tried to follow the suggestions of the help service (running the progam as an admin, saving the image to a different folder, saving with a different color profile and saving to a different file type,...) but no way to improve the results. I have a powerful 2018 Dell XPS13, no problem at all with other software like PS or Lightroom. AI Clear is totally useless for me at this time.

I haven't run into this with portraits (or otherwise.) Feel free to send me a sample file that is giving you this problem. I'd be interested in seeing it. Be sure to send me the unprocessed file so I can try it on my end. Also if you are truly unhappy with it I would contact Topaz about it, along with some samples of your issues. They may give you refund even though it's beyond their return window. It's worth a try. -Joel

Yes and it works very well. First open any image in Studio and run A.I. Clear. Once it completes go to the menus at the top and choose File > Batch Process... You'll get a dialog where you can choose your source folder, destination folder, as well as naming, file format, and sizing options. Then just click OK and you're good to go. I find it to be fairly fast on my 2017 iMac.

Hi Denise, Yes, A.I. Clear is a Pro Adjustment for Topaz Studio. Studio itself is free. One needs to pay for A.I. Clear though they have a 30 day trial and you can use "Effects" (sometimes called presets by other companies) that incorporate A.I. Clear without actually buying it. Because Studio has batch processing it's easy enough to add to a workflow with Lightroom and/or Photoshop. Hope this helps, Joel

I agree with the "awesome" assessment. It still never ceases to amaze me. It may have move into the number one spot on my favorite Pro Adjustment list.

Topaz is going to roll out more A.I. products. There's a new one coming out next week. I work with several companies that include A.I. in their products. I'm not a programmer but I try to pick their brains about capabilities when I talk with the product folks. There are different types of A.I. for different purposes. The identification of dust and scratches has been around a long time. The best one I saw (not A.I.) was an Apple product from the 90s that no longer exists. The rest of the good dust and scratch removal tech (not necessarily A.I.) resides in scanner software like that in SilverFast. I think there will come a day when all the things you mentioned are contained in one program/plug-in/Pro Adjustment but maybe not as soon as we hope.

A.I. Clear's purpose is to reduce noise/grain while maintaining sharpness and detail. Because it uses artificial intelligence you can also use it on photos without noise and it will still sharpen it. I'm not sure what you mean by "cleaning up very old photos." If you have scanned an old photo or negative and it has a lot of grain then yes, it will help. It won't however clean up stains, cracks, or correct color. The idea of a an A.I. product that can fix all these other issues often seen in very old photos is a good one but I'm not aware of a product like that available or in-the-making. Perhaps some other folks out there might know of something like this. Other than fixing my own very old family photos, it's not really on my radar.

Hi Patricia, AI Clear only runs within Topaz Studio. Of course you can launch into Studio from Lightroom. 300 series NVIDIA cards are supported with Studio so it can't hurt to try it. Studio is a free download and A.I. Clear is free to try (topazlabs.com/trials/ref/67/) Let me know how it goes. Cheers, Joel

Good question. That will depend on your camera type and images. Since using A.I. Clear I've found that I rarely use denoise. If you have been using Precision Detail for capture sharpening your need for it will lessen considerably. However I still find Precision Detail useful for other purposes such as selective/creative sharpening, enhancing skies, etc. Topaz lets you try A.I. Clear free for 30 days ( topazlabs.com/trials/ref/67/ ) Try it on some different types of images. If you end up buying it, you can save 15% using code JWOLFSON

Excellent, thanks for the response. I look forward to giving this a try.

From the Edit module, Just add a layer with your photo of the trees and use the masking tools. To add the layer, click on the + sign at the bottom of the layers palette or via the menu: Layer > Add layer from file... Either way you can navigate to your photo add it as a layer. And now it's all non-descructive editing.

I agree- it should be in its own layer so that you can have multiple layers and ability to adjust opacity. As a bare minimum the latter. I will definitely mention it to them. Also I would post it to their photo raw project site if you haven't already. They add new features based on user requests. If it's already there just give it a rating and your input will be counted. Cheers, Joel

It's possible it doesn't look right on the video as you don't always get the full resolution with video. That's why I also include a photo comparison in the written review. Among all the model strengths and opacity slider you should be able to get whatever amount of smoothing you want on the skin. Also, if portraits are a main thing for you and check out the new ON1 2019 Photo Raw with their amazing portrait panel with numerous controls for skin, eyes, and mouth. joelwolfson.com/on1-photo-raw-2019-is-here/

Hi Tom, Thanks for responding. Yes and I am happy they have that viewing option. In fact that part of it is much better than LR where you have to go into preferences and then synchronize the folder to see the jpegs...and then remember you did that the next time you import. My issue is that I like to have the RAW and the JPEGs to have identical names except the suffix .jpg and .arw (or whatever cam I'm using). This is how I've imported my images for years. It's also just a good DAM (digital asset management) practice to have a sensible and consistent naming convention. I use text_YYYYMMDD_seq# and name the containing folder the same as the batch of photos inside (w/o the sequence # of course). This isn't possible in the ON1 import module without a workaround. When the raw and jpeg have different names due the sequence number you can't use the group function once they're imported either. You have to use filtering to separate raw and jpeg and be very careful that you have each group in the exact same sequence and then rename them to match. If you happened to have shot a couple raw only or jpeg only you'll mess things up. I don't think it would take much for ON1 to fix this. I plan to mention it to ON1. If you don't organize your photos this way or don't shoot raw+jpeg or don't care then it's all a moot point.

@@JoelWolfsonPhoto True, but I have always used the download app that Nikon put out that correctly sequences the NEF and JPG images the same so both versions of the image have the same sequence number. If I had to rely on another process, I would be unhappy for sure.

Right now it is easier for me to import into Lightroom and then it's just one click to catalog it in ON1 and then the raw+jpeg grouping works too. When I have the time I will try ON1's new LR migration tool.

@@MrMasPick Tom, FWIW I've found a pretty good workaround. Because both camera systems I use give both raw and jpeg the same filename internally I skip the whole import module and just copy my image folder from my SD cards to a folder on my hard drive. Then from ON1 they can be grouped together, at which point I rename, keyword, etc and place in an appropriately named folder on my image server. Of course, as I mentioned in my newly expanded review (joelwolfson.com/on1-photo-raw-2019-is-here) this defeats the purpose of the import module for anyone who shoots raw+jpeg and wants them named the same.

You're welcome and glad you liked it. I always have glasses on when I'm working on my computer. Though it couldn't hurt to have the font size a bit bigger but certainly a minor thing in light of all the nice improvements. I really love the speed and new interface/workflow. Happy Shooting, Joel

It might be a minor detail, but it's not if you struggle to read it... I agree with you as to the rest of the Prog.. It's a good well thought out one...

@@anthonyhazlewood5788 Perhaps you are already aware that Photo Raw is based on customer requests and feedback- If you don't mind taking the time I would let them know about the small font size on the filter selector. ( www.on1.com/products/photo-raw/ideas/ ) The company is full of young people with good eyesight. We don't want to wait until they are old enough to need reading glasses :) Cheers, Joel

Hi Joel, actually I already did that.....

First of all, yes you can mask whatever you want. There are extensive masking capabilities in every adjustment/layer in Studio. I think what you are seeing as a darker and lighter eye is during the split view of my video. Her left eye is in a shadow from her hair and looks darker than her right eye. Take a look at the part of the video when I first apply the A.I. Clear and you'll see the brightness doesn't really change. I was intrigued by your observation so I went back to the files and measured the RGB values of both before and after and they are the same. So bottom line is that A.I. Clear shouldn't change your brightness levels on its own but of course you can control/change that through masking. Cheers, Joel

Thanks! I'm glad you found it useful.

Thanks Peter. Interesting. Since I did the review I have found myself using it as a first step for capture sharpening for almost every image I bring into Studio. If there is noise, it reduces that but as you say, it also sharpens the high frequency areas. The intelligence seems to work well. Shortly after I did the review Topaz changed the controls and labeling to a more logical format allowing one to achieve desired results more easily. If I go out on a limb for a rule of thumb, I've found the default to be good for landscapes but need to use medium or high model strength for portraits and then tweak with the opacity slider. I showed this in my most recent Topaz webinar.

Clean, Denoiser 6, InFocus Sharpen Ai, Clear Ai, Nik's deNoise, this and that... how much of everything is needed to take a good photo?! I think some people are more interested in the geek technology than the actual art of photography.

No, I don't think it's better. It's easier and faster, especially for those who don't already own or know how to use DeNoise 6. But if you are already familiar with the controls in DeNoise 6 you can continue to use that. Both can be accessed within Topaz Studio.

I would agree with Joel that it's not as good as DeNoise. For an inside wedding picture I'm testing with, I have to strain to notice a difference between the original and processed image. Maybe the slider just needs to go further to apply more of an effect, but for this image, it's basically useless.

Thanks!

You're welcome.

I'm not sure I'd assume it's an awful product. I don't know about the white dots- I haven't run into that but I would certainly bring it up with Topaz. If you send me an example I'll take a look it. In my blog post about AI Clear (where this video is originally posted; link above) I addressed the batch processing in more detail. "I have yet to put the batch processing to a real test. But because it uses AI to evaluate each image individually, it should work much better than batching images with non-AI noise reduction. I spoke with the folks at Topaz and they recommend batching similar types of images. For example, batching a bunch of images of same/similar subjects shot at night with high ISO rather mixing drastically different images. This may be obvious from the settings I demonstrate here in the video versus the portrait example above but I wanted to point it out nonetheless." I'm unable to edit this video on KZbin or I would update it to reflect my conversation with Topaz about batch processing I've used Lightroom every day for my professional work since 2008 and I would disagree that it's better at noise reduction. The chroma noise reduction is equally good but not the luminance noise reduction. On any given image it may be possible to get similar results but in my experience I don't find LR all that useful. This is why I've used Topaz DeNoise. I want to spend more time on AI Clear but so far it seems to be a faster and less laborious method to get to a similar result as DeNoise. Cheers, Joel

curious what is much better about the DxO product?

The final results of NR on images shot with my D810 appear IMHO any way, to be closer to how the same image should look if shot at a lower ISO