My first impressions of Titler+
- Wow great at last, over 20 years after Unicode became a standard, we at last have a modern Title Tool, rewritten, with a modern(ish) interface. So I can add titles in all scripts, with diacritics, in Cyrillic, in arabic…. No more fiddling around with photoshop, exporting, importing. Hurray!
- Toolbox nicely laid out. Not too many bells and whistles which are unnecessary most of the time. For fancy 3D titles I’ll use another tool - here we have a modernised replacement for the old Title tool, usable for lower thirds, simple static text and end-credits (the primary uses of the Title tool) with the possibility for soft discreet shadows which are very important.
- It is directly accessible from within the timeline, no second window, no plugins, it is integrated.
Great work…
But then I started to use it….
OK, now I expect a few issues after all this is a first release (but not an alpha version), a few unforeseen use cases etc…. What I found to be honest disappoints me greatly. Certainly my expectations were high, expecting not only a slick, easy to use welcome update of the >20 year old Title tool for the final year of the decade, but with the niggles with the old title tool sorted out (that came up very frequently and well known amongst seasoned editors and AEs), namely:
1. Possibility to add left, center and right justifying tab-stops allowing two column rolling titles (for end-credits).
2. The possibility to copy and paste images into rolling titles (used to work reliably a long time ago…)
3. The possibility to export all titles on a track as text with timecode (as is possible with subcap) for reports required by production (would be fantastic)
4. Full Unicode ie. accented / non-latin character / right-to-left support.
5. Or maybe even something amazing like importing an RTF file making a rolling title with tabs, images, fonts… or constraining vertical movement in rolling credits to a fixed number of pixels to avoid flicker… I digress… :-)
The impression I have is that it was not thought out from a user perspective, the result is nowhere near as intuitive as it should be for software today.
In a sentence, Titler+ - it is a bit like a sliding door with a pull-handle and a sign saying Push…. ;(
So here we go :
a) You click in the monitor, a caret appears, you can type. Double click a dotted border appears, so far so good. All is nice and works as expected. A good start. It has an ugly black drop shadow that I’d prefer off by default, but I can turn it off. Fine. Looks good so far.
b) Click on the T in the Titler+ « toolbox », you get a square - You type and the square doesn’t change size, the text overflows - no wrapping…. No means (that I found) to have a text area that will wrap, not that it’s that important but the UI leads to expect that behaviour. Resize the square and the text changes size, but the square is still smaller than the text??? Can’t change the size of the square without deforming the text. It turns out the T creates a text layer (zone) not simply allowing text to be added, this is unclear….. hmmm it’s starting to feel a bit quirky……
c) Playing around with it I understand that there seems to be two types of ‘object’ - a ‘layer’ = zone that is created in the safe area that you can resize, resizing all the ‘child’ objects .. and text created by clicking, making ‘child’ objects of a layer….. Phew…. Seems far too complicated and unclear - problem with nomenclature and UI. Maybe reduce opacity on all other layers other than the selected one to indicate the current layer, but still UI badly thought out.
The whole layer / text object concept is very unclear in use - should be straightforward and evident. I create a text object and that should be it, they have an order that I can change. Rather than have confusing layers (this is an implementation detail that should be hidden from me), allow me rather to group objects together, with safe-area guides as before. I should be able to group and ungroup text or graphic elements and reorder them at will. Or keep all objects independent as in the old Title tool. The parameters in « layer » - position, scale, rotation, skew should simply apply to the current object or group, with the possibility for global positioning/scaling. Simple and easy to understand, and closer to the way much other software operates. No need to have layers, particularly as the UI has no window showing the hierarchy of objects and what layers exist (justified in eg. Photoshop but not here). Something reminds me of TitleDeko with less UI... hmmm...
d) Can’t get rolling titles to work for the life of me…. That is to say…. No vertical scroll bar. Rolling titles are used 99% of the time for end-credits, which are typically 10-20 times the height of the frame. I can see the rest by moving the position in the clip and selecting ‘effect mode’ in the effect editor, but then I can’t edit. And « Effect mode » in the effect palette this doesn’t mean anything at all!! I am in effect mode..!! This is misnamed. « Preview » maybe? And there is no none, just Crawl, roll etc... Not clear.
e) The relationship between the effect palette and the titles is unclear. Eg. Select a few characters, layer/scale has no effect as it only applies to layers (zones) (no visual cue as to which parameters apply), Layer/Add Text adds a new independant Text Layer, not text within a layer…. Should be named Layer/ Add text layer… I understand the principle, but there is a problem of organisation and naming, making it very unintuitive (related to c)
f) The mixture of parameters that apply to the currently selected layer and operations‘select layers’, ‘add text’ in the same place is misleading and seems poor design. Effect palette should be essentially for parameters and related operations, especially as there is a dedicated toolbox window with this effect for everything else - much more convenient (great that they this is the first AVX2 effect with a dedicated interface - everything does not have to be shoehorned into the effect palette). Also no means to add a tracker as was possible previously.
g) Accented characters can not be inputted from the keyboard. É, ç à from the keyboard are ignored. This wasn’t tested it would seem with non-US keyboards (I’m on MacOS 10.12…6, with a French keyboard). But you can copy and paste anything you like including non-latin and right to left scripts, which is great news.
h) Selection of text. Every text editor, be it in Word, Photoshop, Title tool… changes the background of selected text…. Here, you get a confusing rectangle, making you think that you have created an new object…. The difference between a click creating a new object and selecting an existing object is not always correctly handled -> there should be an arrow tool for selecting and a text tool for creating/editing and a way to make the caret disappear. Feels quirky.
i) No means to select and align multiple text objects (left, right, center), a) with respect to each other and b) within the frame. Multiple object selection impossible via Cmd+click or Shift+click.
j) Select previous/next layer icons look like alignment icons….
k) Resizing of rectangles should be unconstrained by default, and aspect constrained with shift. The opposite is the case.
l) No polygons (often useful for basic lower thirds), or lines (there is a convention in France to place a large full frame cross 240f before FFOA - line tool useful).
m) Tools/Avid title+ refers merely to the tool palette, allowing you to find if it is hidden behind other windows. This is not clear. Does nothing in Src/Rec mode.
n) To create a title you need to drag+drop from the effect editor, add title in the timeline doesn’t always work (either no response or ERROR: unable to add an Avid Titler+ effect) - impossible to create a title as the only thing in the sequence as you can’t create the space in the timeline to drop the effect.
o) Alt+Cmd after Cmd+L to move around doesn’t bring up the hand cursor (but still works - this is a minor issue)
p) The two icons in the top-right of the effect palette (drag points, select areas) are identical, wasn’t able to figure out what they do. Related to (f) - if is changes interaction with the objects in the monitor it should be in the toolbox.
q) With my developer hat on, under the hood, looking at the AAF, for interoperability... doesn't look amazing... some of the data (text/font names) is hidden as "Private data" in a proprietary binary format (seemingly confirming some TitleDeko heritage...), the rest is as AAF Constant Values, with many repeated property names. Maybe just use XML as a text value, that other downstream software can parse and use. Avid (and other companies) uses XML elsewhere in AAF (and MXF for that matter) where things don't easily fit in to the AAF datamodel. Interoperability is what sets Avid apart and where other systems fall way short. AAF is documented - everything is there, once other manufacturers properly implement import (not always the case) - a clear interface that is easily parsed reduces bug probablity elsewhere.
You sort of get used to how it operates, but it is very very quirky. The Avid design team should put someone who has never used the Title+ tool but with experience of editing software as a user and observe how they get on without a manual - a lot would be learned…. It practically breaks most UI rules in the book…. Poor affordance, hidden hierarchy, no feedback from some user operations, unconventional behaviour. Yet it was certainly beta-tested… (eg. with respect to the accented character issue reproduced on two machines - no French beta testers???)
Maybe I’m a bit harsh, but I do think that the first impression that you have on these sorts of things is the right one. It is possible that some of the things I was unable to do, are in fact possible - but this is the point, it should be intuitive - the manual should not be necessary, not for something as straightforward as this. I teach in a film school and using this, I’m thinking, this is going to be a fun one to explain to first-time users…. :-(
It is a pity that with this opportunity to rethink things from scratch we didn’t get better than this. Should be clean, well thought out and elegant.
Don’t get me wrong, I’m testing out the other features of 2018.12 and so far very impressed, not least the possibility to make a multigroup from a sequence AT LAST with cameras that start and stop whenever!!! Super work Avid!! but this titler….. :-(
Curious to see what everyone else makes of it!
Janusz
