Sneak peeks for 1.7

[Hokusai]

9 hours ago, Ben said:

Ok, Here are some videos to demonstrate all the snapping that I've added for handles.  

Ben,

 

Thanks for sharing the videos and thanks for the hard work! It is much appreciated. The videos looks great, I can't wait to give it a try. 

 

Hokusai

[Alfred]

27 minutes ago, Ben said:

You should make that request in a new thread.  Sounds like a big enough feature to warrant closer attention.

 

@Peregrin originally made the request quite some time ago, Ben, and it’s already been much discussed.

 

Here’s a more detailed thread on the same subject:

 

[Ben]

12 hours ago, Bri-Toon said:

I think I already figured out just how useful the features in those sneak peeks will be. I'm currently drawing a dining room whereas the room is in a slanted angle, and I want the windows to match the perspective. Rather than skewing the parts of the square shape of the windows individually, wouldn't that mean you can skew both square shapes at the same time if the selected nodes are put in a bounding box?

 

Er, give me a diagram and I'll figure out how best to achieve it.  If we are missing an obvious tool, then I can also look into that.

[Ben]

19 hours ago, Pšenda said:

 

The examples show use on a single-segment curve.

However, I infer that angle snapping will controlled both adjacent handle even in multi-segment curves, or am i wrong?

 

Angle/direction snapping for handles comes from two sources:

1) the tangent of the curve leading into the handle being moved (usually the angle of the other handle sharing the same node)

2) the node and handle at the other end of a curve segment (using either the line between the two curve nodes, or the direction of the other handle).

 

1 - gives you the snapping to smooth (following the curve tangent), snap to match the incoming direction (without converting to smooth), and snap to 90 degrees to the tangent

2 - gives you all the other snaps in those videos I posted - some use the 'inline' direction, others use the direction of the other handle

 

It's a little bit involved to explain how each direction snap is derived, but hopefully you can see what they are based on when the snapping indicators appear.  I will always give some sort of visual cue to show what a snap is doing.

[Bri-Toon]

5 hours ago, Ben said:

Er, give me a diagram and I'll figure out how best to achieve it.  If we are missing an obvious tool, then I can also look into that.

 

Oh why thank you, Ben. I was actually implying that these new features will really come in handy, but I can give you an example if you'd like.

 

 

On the door in this picture, I have rectangles inside each other. The bottoms of the two on the top have a 0 degree slope and doesn't match the perspective of the room by reaching to a vanishing point. Normally to fix this, I would use the Node Tool and bring the bottom right corners up and then touch them up a little. However, if selected nodes are put in a bounding box, wouldn't that mean that both corners can just be skewed up rather than having to click the nodes and drag them up?

 

I made a drawing without using the Pen Tool and just basic shapes, and I'm realizing that in many cases it can be easier of having to just skew, resize, or rotate shapes. You don't have to watch the whole thing, but just this section on the clock might help show what I mean. Simple transformations is a time saver for me. So I guess what I'm interested in is to skew, resize, and rotate direct selections rather than whole objects.

[Nana]

On 17/01/2018 at 4:09 AM, Ben said:

Ok, Here are some videos to demonstrate all the snapping that I've added for handles.  Please note, I'm not talking about snapping to grid, I am talking about snapping to useful construction angles relative to the opposite curve point and handle.

 

So, the following snaps are shown:

 

1) Snapping to smooth a cusped node - SnapToSmooth.mov

2) Snapping to inline, and 90 degrees to inline - SnapToInlineAndNormal.mov

3) Snapping to reflected angle - SnapToReflected.mov

4) Snapping to parallel direction - SnapToParallel.mov

5) Snapping to 90 degrees of parallel - SnapToParallelNormal.mov

6) Snap to logical triangle (useful for square corners?) - SnapToTriangle.mov

 

The useful thing about these snaps is that they will enable you to perform common curve construction, completely independent of grid or set axis.  You can use them in tandem with a second action while holding Shift to snap the handle lengths to match the preceding or following handle (while maintaining the direction you already snapped).

 

Wow.  This is exciting.  Much appreciated, Ben.  I am looking forward to testing the 1.7 Beta.

BTW, Are you able to share any information on innovations coming in Photo 1.7 as well??

[Ben]

17 hours ago, Bri-Toon said:

On the door in this picture, I have rectangles inside each other. The bottoms of the two on the top have a 0 degree slope and doesn't match the perspective of the room by reaching to a vanishing point. Normally to fix this, I would use the Node Tool and bring the bottom right corners up and then touch them up a little. However, if selected nodes are put in a bounding box, wouldn't that mean that both corners can just be skewed up rather than having to click the nodes and drag them up?

 

I made a drawing without using the Pen Tool and just basic shapes, and I'm realizing that in many cases it can be easier of having to just skew, resize, or rotate shapes. You don't have to watch the whole thing, but just this section on the clock might help show what I mean. Simple transformations is a time saver for me. So I guess what I'm interested in is to skew, resize, and rotate direct selections rather than whole objects.

 

Yes - you could do that.  Select all the poly curves, select the individual points, then use the transform mode to shear/rotate your points.  It will only be a shear though - it won't give you vanishing point convergence.

 

I am thinking about what we might do about vanishing perspective guides/tools.  That might be some time off yet though (given the size of my to-do list).

[Bri-Toon]

6 hours ago, Ben said:

I am thinking about what we might do about vanishing perspective guides/tools.  That might be some time off yet though (given the size of my to-do list).

 

Well I look forward to seeing what you have planned. Keep it up.

[TonyO]

While you guys are making updates to the Node tool, maybe consider this nifty little action as a time saver 

 

 

[Ben]

I have another sneak peak for you.

 

This one might be a little hard to follow, but I'll try explain it best as I can.  Something that has always bothered me is when I put a linear or radial gradient fill on something, and then try scaling or shearing the object.  The gradient fill is unstable in that it doesn't track the object transform.  This is because linear and radials fills are positioned relative to two key points, forming a line through which the gradient is rendered.  Elliptical (and bitmap) fills don't suffer from this because they have three key points, forming a rectangle in two axis.

 

So, I decided we need to fix that. See this video: CorrectedFills.mov

 

The top objects have a legacy linear, radial and elliptical gradient fill.  You'll see that when I shear the first two objects, the fill kind of moves around inside the object and doesn't conform to the shear.

The bottom objects have a corrected fill.  When I shear these objects you'll see that the fills continue to adhere to their placement relative to their object.

 

I then show you some new handles that appear in the Fill tool, which show you the "correction" points of the linear and radial fills, connected with dashed lines.  (You'll notice that the elliptical fills appears the same as they need no correction).  These new handles show you when a fill is being corrected, and give you opportunity to place the extra correction points if you so chose.  You can double click the correction point handles to adjust the fill into a conventional linear or radial fill in document space.

 

Why is this important?

 

If I create an asset or a symbol, I can then place it, and transform it, and maintain the visual appearance without the fills distorting.  This is then especially useful for 2.5D drawing - I can use assets and symbols that were designed in 2D, transform them to grid plane (using the new planar tools coming in 1.7 which make use of shear and scale), and they will maintain their fills.

Of course, it is just generally useful, in that it kind of maintains a more WYSIWYG approach to gradient fill placement.

 

 

 

Just, don't ask about conical fills - they are going to take a bit more figuring out.

[befehr]

Absolutely BRILLIANT!

Assuming there will not be any issues with exporting to other formats? I guess they’ll rasterize?

[Ben]

No more than we currently have.  Some vector formats don't support elliptical, so we do a "fix" to radial.  So, they will be no worse than they currently are.  The linear gradient correction can easily be applied as it is done by fixing the key line.

 

All raster formats will be fine as we do the rasterisation.

 

[befehr]

Sounds good, Ben. Thanks

 

[A_B_C]

Absolutely brilliant, as he said … 

[MEB]

Hi Ben,

One small improvement that would be nice to have for the Bitmap Fill type is the ability to constrain/lock the image used to its 1/1 dimensions. This is particularly useful to fill with small bitmap patterns where we don't want any resample applied. Would be nice if this could be added to v1.7.

[Ben]

Do you mean independent of the size/scale/transform of the object into which the bitmap fill is being made?

[MEB]

Yes. In other words, the size of the image used in the fill would be the same as the size of the external image we picked from the Finder. So if we select a pixel pattern with just 10x10 pixels to fill a 100px square shape it would fill it 10 times horizontally horizontally and 10 times vertically (no scaling/no resampling). Basically lock it to the original image size in pixels.

[A_B_C]

Yes, that would be very useful … 

[Clayton]

On 1/16/2018 at 6:24 PM, Frank Jonen said:

 

You have no idea for how long I wanted these sorts of snapping features.

The only app ever offering extensive snapping so far was ViaCAD with the downside that almost everything was line vector based. 

 

1.7 is going to be one awesome release.

 

 

 

This is one of those things that sounds boring, but is absolutely huge for day-to-day workflow.

 

When I saw the feature list for 1.7 my first reaction was "aw man, my pet feature still isn't there?" but then I saw the videos and quickly changed my tune to "no, this is definitely more important!"

[A_B_C]

(What was your pet feature again, Clayton? )

[Ben]

There is a fair amount of other stuff coming in 1.7 that I'm not going into in this thread - some of it not so obvious just from the UI. The tool refactoring work was big, but is opening the door to new features.  So, 1.7 should be a big push forward.

 

As we get closer to 1.7 Beta time, I'll be releasing more details.  As ever, we will only release complete features - anything not up to scratch will be held back until it is up to our exacting standards.

[JET_Affinity]

Quote

There is a fair amount of other stuff coming in 1.7...

As soon as the axonometric grids functionality is in place, dimension tools is going to look like a glaring omission.

Take all the time you need...you know, 10-15 minutes.

JET

[Ben]

Dimensioning is going to be interesting, especially if it needs to conform to the scaling of the grid axis.  As I mentioned before - there is no spoon - your objects/layers don't exist in a plane, they are just drawn and transformed to maintain plane perspective. It's all done at tool time with reference to the active grid plane, and no extra information needs to be persisted with the object.

 

So, this will take some considerable working out (on my part).

 

I'm also not saying when we'd add dimensioning tools.  We have longer term plans, beyond what is on the road map.

[Frank Jonen]

The join shapes approach from Astute Graphics is an interesting approach for dimensioning vector objects. Take SketchUp's UI for extrusion as how to operate it and Affinity's own Symbols to spawn a target shape and you could have a solid start for dimensioning. 

[JET_Affinity]

5 hours ago, Ben said:

Dimensioning is going to be interesting, especially if it needs to conform to the scaling of the grid axis.

So, this will take some considerable working out (on my part).

I'm also not saying when we'd add dimensioning tools. We have longer term plans, beyond what is on the road map.

You might want to take a look at a tidy little program called DrawPlus for that. Maybe you've heard of it.  

But seriously, the fleshed-out grids feature by itself is going to be light years ahead of the practically non-existent support for axonometric drawing in other mainstream drawing programs. I'm just saying dimension tools will naturally follow as an expectation.

To this day, Illustrator, for one example, still provides no dimensioning tools whatsoever. One has to spend half-again the price of the host program for a third-party plug-in, (for which compatibility chronically breaks when Adobe auto-installs random updates), or resort to cheezy scripts with very limited capability.

So dimension functionality like that of DrawPlus would be another major competitive advantage. (Even completely dumb dimension objects dependent upon manually-entered values would be more than Illustrator provides.)

Though not ideal, it wouldn't break my heart to see dimensioning implemented in stages like the grids feature has been. While waiting for full-functionality is never fun, for anyone who has been paying attention, the staged implementation of Affinity's grids in light of this "sneak peek" should at least be taken as a sign of assurance of Serif's intentions toward continuous improvement. Consider how many not-quite-there Illustrator features added years ago have received no improvements (3D Effect, for example).

Here's a contradiction for you: I wish Serif continued rapid success with the Affinity line, but hope it will forever continue to "act small" in terms of customer communication, reasonable pricing, and improvement development. So far, all that's been very refreshing.

JET