JohnSnr

It didn't take too long for photorealistic as it's not photorealistic unless you've never seen a photo 😉but it still looks good, a Bond Seamaster. I did look at those before getting the Planet Ocean. The only thing is you have the light coming from the left on the body and the bezel showing it coming from the right. With the profile of the bevel I don't think it can look like that but you only have to flip it.

Now that summer is truly over I've just started some drawing again. This is for a portrait, a bit more than head and shoulders shot. It will be my first attempt, and possibly last (it's going very slowly), at a realistic one or pretty much any person come to that. I've bits to do around the edge to blend in but will do that when the skin is done.
I guess it will show up here pretty much full screen, it obviously won't be that large when on the face, so zoom out or stand back.  I've included the AD file just in case you're interested. 
Lips.afdesign

It's really fiddly and some places it's just a mass of noise or blocky that you have to make sense of. So although I've worked on it at least for a little while every day, if I'm being optimistic then I'm perhaps a fifth of the way through.
I don't much like the reflections on the eye so might leave them out. I don't want to do days of just drawing all the hairs so I'm doing some as I go. Pretty much each one has its own gradient so that's slow. If you're having your photo taken you should shave 
I'll get there eventually.
I was thinking of sharing the ad file when I'm done but then I'm not sure you'll be able to learn much from it.

Indeed, without surface texture it would have looked rubbish, well a lot worse. You see that often when people do vectors faces especially men and children, they don't do any texture, pores etc, and it looks like plastic. Which is why most pick a lady in full makeup which hides all of that and makes it so much easier.

Crikey! (Actual words used were a little cruder and rhyme with “suddy bell”.)
I would suggest that Serif contact you to make some kind of Spotlight article about this, if you were amenable – preferably with an animation showing a swipe between the finished drawing and the vector-only version.
I think it’s reasonable to say that if something as good as this doesn’t show the power of the application in the right hands then nothing will.
You consistently produce wonderful artwork but this is really special and deserves way more exposure than what it will get here.

Transferring images to canvas has been around since the early pin hole camera obscura's. Vermeer later used a lens based camera obscura for his works. Going in to the 1960's some photorealistic artists used projectors to project a slide on to their canvas. So the idea isn't new it's just that the way you go about it has changed.  
Now having said that there are still varying degrees of talent involved, my way is at the bottom end, it's more technique and patience than anything artistic, whereas  just tracing outlines and doing the rest yourself is at the other end. My stuff is more like the Adobe Illustrator vector mesh illustrations, which you don't see so much these days, except it's done with just the pen tool.
Not everyone will like it and some will think it's cheating or just tracing so nothing clever. But then I never say it's art or clever and often say anyone can do it, it's just for fun. So play along if you like.
The Subject
I saw a huge dragonfly fly past me last week when I was out cycling and it's something I've thought about doing before but I've never got around to it. So I thought that would be a good one to do next, it's fiddly with all the mottling and colour blends and the depth of field on a macro shot makes it even more so. I had a look for a decent size reference photo and came across this, it's not the same sort as I saw but it looks good. Unfortunately when you zoom in it has lots of artefacts but just gives you more work so that's ok (I think). The million trillion hairs will be fun (nope).
https://en.wikipedia.org/wiki/Yellow-winged_darter
Below is how I’m starting this project. Sometimes I do the background last rather than first if it's a simple one or I don't know what to use but it's basically the same process for anything.
Adjustments
So the first thing I do are any adjustments I think it needs like levels or curves, just like any photo. Then  depending on the size I enlarge it, I use Photoshop's Preserve Details Enlargement option, I don't know what that would be in AP but someone probably will. So for this one I doubled the size to just over 5kx3k
Backups
I personally don't use the built in AD snapshots, I backup to another drive. It's up to you but, for example, if the file became corrupt so do your snapshots (correct me if I'm wrong).
Tools , Toolbar and Panels
I turn Tools and Toolbar off (not the context toolbar as you will use that sometimes) to give me a bit more space. I'm only using Pen (P), Move Tool (V), Node Tool (A) and Colour Picker Tool (I) so I use the keyboard shortcuts.  The only two panels I have are Colour and below it Layers. 
I also have Edit>Preferences>User Interface>Show Selection in Layers Panel checked. So that when you click on an object it's shown in the layers panel rather having to scroll through to find it. Although it doesn't always work.
Create The AD Document
Open the image in AD. I rename the photograph layer to Photo as I call my background group Background. For this I set the File>Document Setup>Colour>Colour Format to RGB/16 but sometimes I leave it at the default RGB/8. Then save it as a AD file.
Blurs
You soon get used to how much blur you need to use, so how much for a little bit up to how much to blend it in. Obviously it varies depending on resolution. One thing is you do need a blur on the edges or when it’s on the background it will look like it’s been out out and stuck on. Sometimes it might be only a tiny bit but do you do need some.
Create The Background
It's just a blurry background, You can either try and copy the photo or do it a bit differently but it makes sense to keep the light and dark areas in the same places. I didn't like the light green band across the top so basically got rid of that and added a few bits of brown. It might need some little adjustments at the end.
I created a rectangle with snapping temporarily turned on (that's about the only time I use it unless I add a border) and a bit of a left to right gradient using the dark colours from the photo. I then created the shapes in a sub-group, I set the group to 100px blur and then the individual objects to 100px blur too. I think it's just my imagination but I thought it showed less banding than just setting the group to 200px blur. So you you could just do that instead.
Oh remember to lock you groups.


The Reed
I created the reed shape, I added a little bit of brown to the basic gradient as it was bit blown out and bright in the photo. I'll add few bits of detail every time I want a break from the main image.
Abdomen
I never do it as one big object I break it in to parts. I decided to start with the abdomen to make sure I could get it looking anything like the photo. I wasn't sure how best to manage the objects in this part but decided to do a group per segment and under there a group for the details and a group for the basic colour bands.  I created the main shape filled with a gradient colour picked from the darkest part from each segment.  
So for my first test segment the basic background gradients looked like this at which point you think it's never going to work but it will

Then you add as much detail as you want or need. It perhaps doesn’t look like much but it takes me quite a time.

 

If you flip the dark fill over to stroke (with the background and reed turned off) then you can see how well it blends in with the photo, or doesn’t. You can then also toggle the segment and see if you’ve missed any obvious highlights or shadows.

(cropped)
Next
I might do the head next then move on to the near side legs but at this rate it's not going to be for a while.
So that should be enough to get you started if you really want to try it out. I'll try and upload the AD file if anyone's interested even though there's not a load in it (I just tried but it kept failing). 
 

Red Eyed Tree Frogs are native to Mexico, Central America and go all the way down to Colombia (perhaps on holiday). Its scientific name is A. callidryas, which comes from Greek words kalos (beautiful) and dryas (a tree or wood nymph).
I was chatting with someone about drawing and she liked to do Victorian style flora and fauna illustrations (I think there’s a proper name for it but I forget). I said that I usually try for realistic when I mess around for fun and she wanted to see something. One her favourite things are frogs, she has lots of ornaments etc in her house. So I did this, I think it might be the first living thing that I’ve done properly although I'll probably do a bit more to it sometime.
Just the pen tool, no textures.

 

Each time that the app icon gets updated it gets uglier so I go back to the very original one which I think might have just been used in the 1.0 beta, I can't remember.  I think it looks far nicer in the dock. Anyway, if it's not been said before then it's still included in the resources (mac: show package contents > contents/resources) as designer_filetype_icon.png. On the mac I use Liteicon (https://freemacsoft.net/liteicon/) to change app icons but you can do it through Get Info without any app, just drag it over the current one. I can't remember what I used to use under Windows to change icons.
My brother said that AP has the original version in there too (I don't use it), photo_filetype_icon.png

This was mainly to try a few different ways of doing things to see if the extra effort was worthwhile or not. The main thing was not to use gradients but draw a bunch of objects to make pseudo-gradients so you've a bit more control over things. There a few rough bits like the leaves but it's ok. As usual it took longer than I expected, I should really have learnt by now.

I've just seen the new AD 1.7 samples, well done 

I was searching for something unrelated and came across some 3d mandalas which looked fun, different from the usual flat coloured ones. The person had used pens so I added quite a bit of grain to make it looks more pencil like. I didn't plan it out (should have done) but it's a nice little project for an evening or so.

You should think about prints of some of your work.

If you know your bikes then this isn’t the first one that comes to mind when you think of the iconic Cinelli Laser but this is a much rarer Laser although there are more types than you perhaps think including a track tandem. I heard that only 5 were made and only 2 are known to have survived. One is in private hands and I’m guessing the other will be in the Cinelli museum. It seems that there is only one photo of this bike on the net. Although the largest version of it is a decent size it’s of poor quality as it’s a scan of a small’ish photo in an old Cinelli brochure. 
All in all this is made up from about a dozen different reference photos of the components mainly taken at different angles.
At first glance it looks quite simple but was actually a decent sized project and took far longer than I expected. It’s A0 I size. Ignore any interference on the wheels it only shows in the uploaded version.

I’m not really much of a motorbike person but fancied doing one for a long time and Peter’s Ducati drawing reminded me that it’s about time I got around to it.
I didn’t want to do the whole thing this time, I wanted a small’ish project. There are a few crops that looked good but I wanted the Ducati name in it. This is a slight crop of a crop to make it A paper size. Adding depth of field, so some things at varying degrees of out of focus, was a bit fiddly at times but not too bad.
The petrol cap was sort of interesting as the reflections made it look very vector like in the photo. One of my favourite bits is the green petrol (I assume) pipe and clip (the clip took just over an hour) which doesn’t jump out at you but I really like it. Of course I also like my brake pivot screw that I showed the other day, together with the horn section. 
There is one bit I was lazy about, on the fork crown the black paint in the photo had texture which I didn’t do. However photos of other bikes show it smooth painted so I’m OK with that. I think the brown stripe on the tank was actually yellow when new but I’ve yet to see one where it’s still that colour other than on a model. It could be the models are wrong though.
Near the end it was killing my MacBook so I split it and used a master for the finished stuff. Ignore any banding on the tank it only shows up on the upload. There are a few tiny bits that I might do as I get time but I’m quite happy with it. The main image is half size.
I did think to look at how much they go for these days, they’re not cheap.

that sometimes make me happy.  I managed to spend a good couple of hours on a screw but wanted it to look just like the original (the background's just blocked in).  The wireframe doesn't even look that complicated, well I suppose it does for a screw, but it still takes time. It's full size from a 7216x5412 canvas. I was intending this one to be a quickie but somehow it doesn't often turn out that way with me. Now it's time to do a bit of work in the dark so I can see where the feint parts are.

This is as far as I could get this week.  Just the sea, shell, some reeds and the flower. Not the people or land I just quickly masked them in so I could see if my stuff blended in. The sea and shell still need more work (obviously), the sea mainly below the shell which is tricky. I started on a few other bits like the sky and the flowers below the lady on the right, it's hard to make out what's what with them, But not finished them either.
I used the google 30k image for the sea colours but will add a colour or gradient overlay on the sea background to brighten it up a bit and probably add a touch of blue to it too. It's not easy trying to approximate the dry brush strokes, I'm just making them quite transparent then adding a blur. Oh, I was doing the sea and shell in the same doc. but it started to kill my macbook sometimes, cursor was jumpy, so I split them.
Each bit is a project on its own. Even something like the falling flowers, you have a quick look and think there's only a few but there's 30 of those.Only 29 to go.
The outlines is just the sea and shell.

Just fancied doing it for some reason, skipped the lettering as it was a quickie.

Thought I had better finish it or I probably never would and it didn't take too long (although I kept fiddling with it). Added the cap too. It's not perfect but it will do. The paper in the background is just part of a photo vectorised.
 
 

I'm cheating a bit with gradients especially on the waves. If they look about the same that's good enough, I do one and then copy & paste the style. I just have a few basic wave ones in a palette. Like you say they can become confusing and naming them often doesn't really help, by the time you've found the right one sometimes you may as well have done it from scratch.
You soon instinctively check the box when the fx panel comes up but it would be nice not to have to.
Oh that screen grab shows how I often do the initial block of colour, I looks for a gradient and use that. Although in this case I think the shell gradient might be completely replaced with detail. I can't decide if the edging around the shell is colour or gold.
Time for obvious tip of the day. Sometimes you want to blur something, for example for a highlight or shadow, but not go over part of the surrounding area. You might want to blur to one side of a thin line and you can't hide the unwanted blur under things.
A dumb example (probably other ways to do it but I do it this way)... You don't want the blur to go on to the the yellow, you want a hard edge and just blur away from it.

I duplicate the shape and use the duplicate as a container, so clear out the stroke/fill etc.  Then make the original the sub-object and drag out the container so that the blur isn't clipped.

 
If you want the edge to be less sharp add some blur to the container. Also adding blur to the container means you can also use this to have a different amount of blur each side.
 

Thanks, the leaves and blossoms are the easy bits though. I changed my mind and started working on the sea as it's one on the make or break bits (just started the shell too). Not as easy as I hoped but that's not a surprise. AD's not too thrilled with exporting it, it takes about 7 minutes for JPG, think it must be the large blurs. That makes me sort of wonder if it will manage to export it at all when it's finished but it should be ok..
It does start to work harder. I'm on the mac and use Turbo Boost Switcher to turn off turbo boost which keeps the CPU from running away. There's ways to do under Windows too, I think the Intel CPU tweak utility lets you.
Yes the same size documents does make things simple, I did that on the Jag where I did the wheel and tyre in a different doc, but with this image being huge I thought the registration way works better for this particular one. 
I do use the group effects sometimes (if that's what you meant) but it depends on what I'm working on as often pretty much everything's different so I end up doing them for each object. I do get a bit fed up of having to check the Scale with Object option a million times, have I missed where you can set that as default?
The other one is not being do an average colour pick for gradient stops. I'm having to use some rectangles as temporary "stops" as below, use the eye dropper with 5x5/17x17 average to set their colours then transfer that to the gradient stops, bit long winded if you have to do it often. Perhaps I'm stupid and missing something but I'm not sure AI's any better.
I'm going to be starting some pretty big icon projects for people so this might well take even longer than I thought.
 

Probably obvious tip of the day. Often an object that's too regular doesn't look right. Whilst it's quicker to add any roughness as you draw it, if you're like me, you'll sometimes need to add some afterwards. This demo object is too regular at the bottom.

 
So all I do is add nodes between the original nodes (I usually do 2 or 3 but it depends how far apart the original nodes are) and twist the middle one using the handles, you can twist it either way and it will look slightly different, pull in or push out.

I know but that's not too easy when you're doing vector with just the pen tool. If you start trying to emulate every brush stroke, basically a photorealistic version, then I could do it but I'd be doing it forever. But I'll see what I can do when I get to the figures.  Some film grain always helps making it look less sterile, I don't think adding a texture over the whole lot would work. Also I now have the 30K wide photo to work from, although it kills my macbook so I'm chopping bits out as I need them. So it means that I don't have to guess what some things look like now, sometimes even what colour they are.  The old one was 10K wide which is a very good size for a single object but pretty low res for something like this.
This weekend's job is to work on the orange blossoms, there are a lot of them and you spot extra ones that have almost faded away. I'll work on the branches at the same time. When I get bored of doing the same thing I do something else for a break, like last night I did a few of the waves and thought about the sky, then I go back to it.
This is the orange blossom palette I set up. As always it's something to use as a base, you often want to apply one then alter something, normally either the highlight or the shadow part. Of course you can then add that to the palette.
I've also been using paste style a lot which I don't normally. So you copy the object you want the style of and then select the object you want to apply it to and do a paste style (I always use Shift-Cmd-V, which is Shift-Ctrl-V on Windows). That copies all of its attributes, so fill, stroke, effects... but not the actual object. 
Orange Blossom.afpalette
 
 

I did look to see if someone had done something trying to show what the original colours would have looked like, but couldn't find anything. Shame as I doubt they sky looks like it does now.
The main reason why doing this might not be the brightest idea is that I've never done people before, apart from low poly that doesn't really count. I've been trying not to think about it but, as I've said before, it's best to do the hardest bits first if you're not sure you can do them, or at least do enough to find out if you can or can't.
So I thought I'd better try or I'll do a ton of work and find out I can't do people. So today's task was to have a go at Venus' face.I didn't have time to finish it properly but gave me an idea of what to do. Whether I'll use it or redo it I'm not sure but for a 1st attempt at a face I was quite happy with it. I found the thought of doing it harder than actually doing it, it's just shapes and shading like everything else just need to be more careful.
 

You all probably know what it looks like but perhaps not in detail. You can see a decent size version on wikipedia, https://en.wikipedia.org/wiki/The_Birth_of_Venus
I've just finished the foliage on the orange trees, not yet done the blossom, branches or anything else. As you can see from the original it has lost most of its green, although you can see some up the top where a frame had been. It made it a bit tricky to even see what was going on in places. I could have done extra shading in them but think it looks ok like this, can always add some if I decide I want to. In the end I only used the darker leaf gradients. It was about 20h of work, yeah I'm slow.
Now I've one section finished I'll copy it and use that as the master image. I'll empty out the layers on the work copy but not delete them. If you delete the layers then it's harder to work out where to put new ones in the master. Doing it this way means AD doesn't need to work so hard plus you won't accidentally do some something silly with a completed section as locking doesn't really lock. So as new bits are finished they get copied in to the master.  Plus I do a backup or two a day (and exports). I just copy it as I trust that more than the snapshot, it can't really go wrong.
I need to create a much larger blossom palette, I'll upload it, and that will probably be next unless I fancy doing something else. 
 
 

TB, or what they used to call consumption, was a common way to die. All the Bronte sisters died from it too and that was 350 years later. They still haven't eradicated it, just not so many die from it now.
Here's the palette of golds that I'm using so far.
 
 
Gold.afpalette