Hey @craftycurate, it was me who made the update to the title - I do this quite often to make the forum easy to read (and navigate) but I'm sorry to cause any issues.
I felt that the title change reflected your understandable comments regarding the update and motion graphics. I also felt that this title would mean people would talk more about what motion graphics update they might want in the future.
Given we have just released a new version I also felt it was important for people looking for information about HitFilm Pro 2017 and what might have been changed in the software - notably not that many motion graphics update.
We sometimes change titles on the forum to keep things logical and easy to read but we don't change posts unless someone is abusive. I hope that helps.
Even as only a forum member here, your wording sounded a little too negative to me too. Especially just with new upgrade enthusiasm going around. So wasn't surprised when i saw your thread changed. At least only the tittle changed to a less negative and direct reflecting your problem with it.
It is their forum in the end and a more gentle approach is always good wherever you write or talk in general
There are so many users around who have many different needs and preferences.
For me there were (are?) 2 problems with Hitfilm. Blurry export of files and the weird loading stye of image files. Loading style of files is cumbersome but workable. But blurry export is unacceptable for me.
Yes I realised I had posted out of disappointment and frustration after months of waiting, and hoping that Hitfilm would emerge as a strong MoGraph solution at last, with an improved basic feature set in this area, and that the post title reflected that, but was probably too strongly expressed, though I stand by the sentiment behind it.
HitFilm bills itself as a Motion Graphics solution, so it is positioning itself in a market where other solution exist, with certain feature sets, it has to back up that claim with a strong competitive feature set, which I don't believe it does yet.
It's not that the post was edited per se that gets me. What annoys me is that it was changed to something I don't believe is true and wouldn't have said, so I felt misrepresented. Also, the follow up post was moved onto this thread, perhaps to hide it from public view who knows.
To say Hitfilm 2017 has "not many" new MoGraph features is quite generous, as another poster pointed out. In reality, I scoured the New Features list and struggled to find anything substantial that improves HitFilm's motion graphics capabilities significantly. I have now changed the title again to reflect my own views.
Bottom line is that I want this to be a great product, because Hitfilm's success is all our success if it helps us fulfil our creative goals, and I hope to purchase it when it becomes indispensable for my own work. I wait for future updates to see what direction the company takes it in.
@craftycurate just to be sure that we have everything on our list for future improvements, could you be a bit more precise on why the text handling is very basic and what is missing? I have a few ideas myself but just making sure we're talking about the same things. If you could edit your first post that'd be wonderful
My post in this thread is now not so related if it becomes a mograph thread. 3D artists need exr. No exr, no pro artist will use hf.Well only those that render with png, jpegs, amateurs.
@chibi I have to disagree with that. The real benefit of multi-channel EXR files is easier file management but there are workflows and situations where multi-channel files aren't desirable and you're much better off taking the hit on file management rendering out multiple sequences instead.
@CedricBonnier: from my perspective, things I'm missing most from the text tools:
Have updated OP as requested. For motion graphics, the greater priority by far in Shape Layers and vector import and handling which doesn't seem to exist at all currently.
I would disagree with that at some level. The Hitfilm demo mode certainly has no shape/vector support at all. Shape and vector support in Hitfilm is available via the Boris effects. Boris obviously does not let FxHome demo the Boris code for free in Hitfilm. Boris also provides the more sophisticated text support in Hitfilm.
Native vector support is greatly preferred. It shouldn't be necessary to rely on 3rd party plugins for core functionality.
@craftycurate thanks for your feedback. There is a task in our tracker to improve text but the priority dropped when we introduced BCC9 in HF4 as it allowed to create much more advanced text and titles than what you can do natively in the software. In Pro 2017 we've updated BCC to version 10 and added titler pro, which enhances the text capabilities even further.
Saying that, we do understand that improving our native tools is important as well. We'll get to it eventually, we just need to figure out the right order to tackle each bug / feature in to get the most people pleased. As long as the software goes in the right direction and you guys are happy with it, that's all that matters to us
@craftycurate "Native vector support is greatly preferred. It shouldn't be necessary to rely on 3rd party plugins for core functionality.
I sorry but this just not make sense. So if the Boris name is not on the functionality then your opinion somehow changes? Who cares where the function comes from. What matters is the function provided.
A real problem with the 3rd party thing is that they do not allow demo so one cannot fully evaluate the as delivered Hitfilm platform. One has to include Mocha in that situation also.
Craftycurate, I will note that NO software has every possible feature natively. There's a reason third-party plug ins exist (for example, to get Hitfilm-like particle sims in Ae requires Trapcode Particular--a plug-in that costs more than Hitfilm Pro. Then to add 3D models to Ae you need Element 3D--a third party plug-in that's half the cost of Hitfilm Pro).
Incidentally Hitfilm Ignite is a plug-in suite to add Hitfilm effects to other software. Running HFI in Ae gives Ae functions that Ae does not have.
A general note on Hitfilm development: FxHome is a small company--around 20 people. They have multiple product lines. Once you account for back-end staff (general office, accounting, marketing) and staff for PhotoKey, it's unlikely there are more than four or five programmers assigned to Hitfilm. This staff would love to do every feature every user asked for, and do it this week. Sadly, they can't.
Cedric has a very valid point when he says improving Hitfilm's native text became lower priority once BCC was licensed. Having BCC bundled in HFP or as an add on for HFE gives a very good alternative for users who need more advanced text (yes, I also would like improvements to Hitfilm's native text), while allowing the dev team to work on other things.
As I said in an earlier post, there are plenty of improvements to Hitfilm I want that aren't implemented yet, but, trying to look past what *I*want, the HFP 2017 improvements to the speed of response in the editor, the full audio panel (finally audio editing in Hitfilm isn't a bad joke!) , the new search filters, the increase in allowed media sizes--these are much needed improvements that also happen to affect the highest number of users.
And these changes to core functionality open up future options. Example: we have users hoping for HDR support. HDR requires a 32-bit color depth. Now that Hitfilm has that, it's possible to develop HDR support. 32 bit color is a fundamental engine enhancement--now implemented--that had to happen before specific implementations could be done.
Btw, you say there are no new motion graphic tools in HFP 2017? I point to Neon Path--mograph artist who have wanted to do strokes on text or custom paths just got a feature requested for years!
Now, to you and @chibi a gentle admonition: careful to confuse what you (or I) want with what's vital. Yes, native support for vectors would be awesome. Right now Boris is the only option. Or, rasterize your vector at a high resolution--telling Illustrator to rasterize an element 4000px wide will give you something usable in Hitfilm. Is it as convenient? Nope, but you can still use the element. Wanna draw a vector shape in Hitfilm? Create a huge plane and use the (vector) layer mask to draw the shape!
Deep data in OpenEXR would be nice, but that's a November 2013 addition to a 2001 file spec. You can render out multiple passes to composite just like everyone did for 12 years. Deep data makes things a bit faster since all passes are in a file, but it's not essential at all since the passes can still render to multiple files to be composited. No functionality is "missing," you just have to use a slightly slower method!
And Scopes were the absolute most vital missing tool for "professionals." Those were absolutely necessary for those of us (raises hand) who prepare media for television or motion picture broadcast and have to actually conform to SMPTE standards. @Aladdin4d @Stargazer54 me, and anyone else in the forum who make their living in broadcast whooped with joy over scopes. In comparison, OpenEXR Deep Data ain't nothing.
@NormanPCN of course one could download the demos of BCC 10 or Mocha Pro 5 (plug in), which accept Hitfilm as host. Again, it would be easier if Boris/Imagineer let their plug-ins function in the HFP demo, but there's still a workaround.
Final note: craftycurate, as I stated on page one of this thread, *I* am the mod who combined your thread about the title of this thread with, well, this thread. I didn't do it to hide the thread, I didn't do it to censor you. I did it because you started a subdiscussion in this thread on the topic, and having two threads about the same topic fragments discussion and spreads out responses. My intent was to make things easier for you and others to follow. If you prefer, I can split all discussion on thread titles back out to a new thread. If so, I will move ALL comments in this thread relating to editing of existing threads to the new discussion and leave this thread focused on motion graphics. Still, pretty much anytime users create multiple threads on an identical topic they'll end up merged. Again, that's to prevent fragmentation of a discussion. :-)
"I will note that NO software has every possible feature natively"
This goes beyond features you touch and feel. Critical portions of the guts of apps we the user do not directly see are 3rd party code and I am not talking about GUI encapsulation code. That is obvious.
Uhhhh. its more than filemanagement...is that what you think about multi layer exr??
And I'm interested to hear which workflow would exr not be desirable working as a 3d artist.Do you render in pngs? :P
Currently working on Unity3d for DCC and 3D VFX I find the hub and spoke model good. Unity / HitFilm provide a solid foundation where to add third party features. In case of HF you even didn't need to pay for it. When a feature is really appreciated the team can insurce it in the core, using third party to offload some risk.
From my part too want more VFX in HF, but I did understand the model, and more I suggest sooner or later to open some API and create a marketplace for extensions.
@chibi No it isn't much more than easier file management so it is absolutely what I think about multi-channel OpenEXR and I never said OpenEXR wasn't desirable. What I said was there are times when multi-channel OpenEXR isn't desirable because there are times depending on the workflow following whatever you do as a 3D artist where you are much better off exporting multiple passes as single channel sequences. Some examples off the top of my head would be if you intend to batch process a channel or two with Photoshop or something like it or even though they can technically handle multi-channel life is still easier in both After Effects and Fusion if you stick to single channel sequences not to mention gaining some added flexibility going that route. Heck multi-channel OpenEXR in Fusion was a total nightmare until pretty recently.
On top of all that as @Triem23 already pointed out what you're after is a fairly recent spec and you are still free to do things the same way they were done before that spec just like many many other people are doing everyday. If you can render out to multi-channel OpenEXR it's simple as making a few quick changes to your render queue to render out to single channel OpenEXR sequences.
P.S. If you really want an example where OpenEXR wouldn't be desirable at all it would be in a pipeline using DPX or TIFF sequences.
You make some good points, but some points of disagreement too:
>>"Craftycurate, I will note that NO software has every possible feature natively."
No-one's suggesting it does or should, myself included. But vector support is a bread and butter feature for a package that bills itself as a motion graphics solution I would have thought. Does anyone else share my view or am I out on a limb on this one?
I'm strongly suggesting that full vector import and editing support is badly needed esp where vector artwork has been prepared in another package in some standard format, and needs to remain editable as vectors after import. The two partial workarounds do not fully compensate at present (1) Prepare artwork elsewhere, and rasterise on import, losing editing capability (2) Try to recreate it in Hitfilm as a vector mask. I wait for future updates, or plugin support for this.
Of course, support for AI or EPS which allows round trip editing would be Nirvana but just importing would be fine to start with.
To be honest, reflecting on others' comments, I don't care whether text is delivered natively, or by a 3rd party plugin or an alien spacecraft in low earth orbit connected by satellite - it should FEEL native (i.e. integrate smoothly) which as others have pointed out I haven't been able to test in the demo, so I hope I would be pleasantly surprised if I did have that opportunity.
>> "Now, to you and @chibi a gentle admonition: careful to confuse what you (or I) want with what's vital."
This is bizarre logic and a false dichotomy. Delivering what your customers want is by definition "vital" because that's why they buy your product and keep you alive as a company. Anyway, I've only requested features that I believe fall within the scope of the market that Hitfilm is aiming to penetrate?
>> "Btw, you say there are no new motion graphic tools in HFP 2017?"
>> "trying to look past what *I*want"
What's with the *I* thing? I am perfectly aware that what *I* would like is only one customer speaking, and you have to prioritise and rank the requests on a cost-benefit basis or whatever, but I can only represent my own angle on things, which is what I've done. I don't think my feature requests are bizarre for this kind of software solution but if enough people tell me I'm way off then I'll assume that I've lost the plot somewhere. I'd encourage others to throw their hat into the ring and say what *they* want too. Otherwise how will you get consensus about what matters to your (potential) customer base?
PPS No need to mess about with the threads - thanks for explaining. It's fine as it is
"Does anyone else share my view or am I out on a limb on this one?"
The answer depends on things like what you consider to be motion graphics vs what you don't, are you going for a look that can only be achieved with vector graphics etc etc. Some things are bitmap only, some are vector only and there are some things out there that can do both. What's best or vital is relative.
To many being able to include 3D models and particle sims is absolutely vital for mograph work. For others the vital thing is going to be 3D text and for still others like you it's going to be vector support but simply because a package doesn't include one of those things doesn't make it useless for mograph work.
Just as a different viewpoint my personal experience over the years has been that the promise of using vector tools for animation in an NLE is overrated. There's problems with importing vector art from app to app (or plugin) and what remains editable as a vector and the tool set available to do any vector editing is severely limited compared to Illustrator or Corel DRAW or even Inkscape. I've generally had better and more consistent results limiting vector work to creating bitmap assets and working with those so for me vector support isn't even on the radar as a feature I would want nor would it be something I'd ever consider as vital.
"Simply because a package doesn't include one of those things doesn't make it useless for mograph work."
Yes it does depend very much on what end result one is after. The projects I've landed have always been text\vector heavy but that's just one branch of the MoGraph tree.
"I've generally had better and more consistent results limiting vector work to creating bitmap assets"
For me the best results by far are with vector import, and leaving the frame-by-frame rasterisation to native After Effects functionality. Pre-rasterising your vectors means taking care with scaling and so on to avoid pixellation other artefacts.
"There's problems with importing vector art from app to app (or plugin) and what remains editable as a vector and the tool set available to do any vector editing is severely limited"
I'd settle for any vector import at this point! Editing with a node editor, would be a bonus but I could live without it.
Google's meta description attribute still has the original (negative title) - ironically that's what made me click it and look at this thread.
@CedricBonnier to belatedly add to your request about suggestions on how to improve text:
1) Make text selectable on the Layer Tab. Just this one thing would make it easier to change text, when you've done something with one layer in a comp and want to duplicate it on another, but with new text retaining the animation you just made, but the text is either:a) multilayered with some other text (e.g. multiple outlines, or offset layers for style) or other Layers, making selection very difficult, b) Scaled, at an angle - 2D rotated or in 3D - which makes it well nigh impossible to select. When/If you do manage to select it, you have a 'ghost' selection box in the middle of the screen where the text used to be which you are supposed to navigate by touch and luck to change the text. As it's all too easy to fall off this 'selection' and click on something else in the scene (no soloing of layers...yet ) and have to try again, this one thing would make text manipulation about....oh, 10,000,000% easier to use. At a conservative estimate.
Using 3rd party plugins to integrate sophisticated functionality isn't a down side, IMO; rather it's an up side. It enables a small company to offer sophisticated functionality along side stuff that they're developing themselves. There's really no reason for FXHome to reinvent the wheel when they can make an agreement with a company who already has a great wheel to provide that wheel to their users -- Mocha Pro is a great example of this; it's a very sophisticated and very specialized tool, one of the best in the business, and we get access to it as part of the package.
IMO it's a win-win for us, because rather than duplicating Imagineer's extensive expertise in motion tracing, FXHome is working on other more important stuff from our (their customers') point of view, like the foundational update that enables 8K and 32-bit color support.
HitFilm still comes up short for team projects because its interoperability with other post tools is rather limited; there is for example still not a robust way to share an edit between Premiere and HitFilm (sure, you can render a high quality DI to use in HitFilm, but then you have no way to work with transitions, for example).
Stay in HitFilm, and you get lots of capabilities; the editor is turning out to be quite good so far, and the scopes are really well implemented. I don't expect to need motion tracking on the current edit, but it will come up sooner or later.
A lot of what's new in HF 2017 is stuff that I know I've been hoping for, so I'm pretty happy with it, even though there are a few more things that I'm still hoping for that would make it possible for me to integrate into more projects. It's getting there.
As an editorial and VFX suite for a small indie production, it's a great solution; being able to "round trip" from editorial to VFX by clicking a tab is pretty sweet
Grade layers, expressions, nodes... those would be pretty nice, I'll grant. So would RedCode import, though I suspect that I'm in a small minority of HitFilm users who have ready access to Red cameras. At least I have a Red camera that can capture in DNxHD and ProRes, though there are times when it's very advantageous to use RedCode for the detail (e.g. chroma key).
Palacono These are some of the usability frustrations that make working with text difficult in Hitfilm.
Another one that is counterintuitive and awkward is not being able to resize the text box manually, but having to edit it using pixel dimensions in a properties panel. Being able to drag the text box visually is far better, as you can with other layer objects. I hope this will change in a coming update.
You can drag a text box to resize it. Make sure you're in Text Mode (click the little "A" icon in the preview window, then look at the lower-right corner of your text box. You should see a tiny rectangle that turns white when you mouse over it. Drag that to resize your text box.
"batch process a channel or two with Photoshop"
I don't know any 3d artists that uses photoshop for image sequence editing. That's why there's fusion, ae, nuke. Photoshop is mainly for image editing. Your example doesn't make sense for 3d artists.
As of right now, no 3d forum is interested with Hitfilm. It just won't work without multi channel exr support. And its not new its been around for several years.
@chibi Sorry you feel that way but I know some 3D artists that do in fact batch process sequences with Photoshop on occasion which is why I said it. There's even tutorials on how to do it from vfxforge.com. I'll grant you the tutorial is working with multi-channel files and the 3rd party ProEXR plugin but I chose this one to prove another point - there are 3D artists that use Photoshop for compositing instead of a compositor like HitFilm, Nuke, AE or Fusion. Just because it doesn't make sense to you does not automatically mean nobody else in the world does it that way.
I'm not sure what your definition of "interested" is but when random forum members take the time and effort to develop tools and scripts just to facilitate workflows between 3D programs like Blender and Lightwave it sure seems like there's forum interest to me and when there's commercial tools like RocketComp for Maya to set up a Hitfilm - Maya workflow then I think maybe forum interest isn't even important.
Now for you to say "It just won't work without multi channel exr support" is frankly bordering on the ridiculous. It can work and there are people doing it everyday even when using compositors other than HitFilm like Nuke and Fusion. Before you get going again Fusion has never.had good support for multi-channel EXR files with the Python OpenEXR libraries and bindings in scripts being the main way to make multi-channel EXR files usable. A very common recommendation for Fusion is "EXR's are great. Multichannel EXR's are horrible. Do not do use them." Nuke does better and AE does ok with the ProEXR plugin but there's a couple of hidden gotchas with all of them or anything else that does handle multi-channel EXR files.
The first gotcha has to do with the difference between multi-part vs deep data. Deep data channels are actually interleaved meaning in order to access and/or process one channel every channel in the file has to be read so if you have 10 channels, all 10 channels get accessed and read into memory just so you can work on that one channel. This is a massive hit to the processing overhead needed and it's universal to any program that can work with these files because it's part of the file spec. The overhead for working with a 10 channel sequence is always as high or higher than 10 separate EXR sequences. This is one reason why some places still prefer to work with multi-layer TIFF's and yet another very good reason to go with multiple EXR sequences instead of cramming everything into a single multi-channel EXR - You can work with and process 10 separate EXR sequences faster than you can process a single 10 channel sequence. Don't believe me then check the forums for Nuke and Fusion. Separate sequences always equal faster pipelines especially when you only need to make changes with a few channels.
Multi-part images are different in that the channels aren't interleaved and you can access a single channel independently of the others but they still have something in common with deep data channels that's the second hidden gotcha. Some channels are to be handled in pairs (example - forward/backward motion vector channels). Why is that a problem? I'm glad you asked. It's a problem for almost every OFX host regardless of it being a layer based compositor, a node based compositor or an NLE like Vegas because the current OFX 1.4 api and hence every OFX 1.4 plugin can only handle one channel period, end of discussion. The Foundry has added an unpublished extension to Nuke to handle multiple channels and they've submitted it to become part of OFX 1.5 but so far the only other program to even attempt to provide that kind of support is Natron and note it's currently just the program and not Natron's native plugins. Currently the only plugins supporting the unpublished extension are The Foundry's own Furnace plugins. What this means in real world terms is you can't dump everything into a single multi-channel EXR sequence and expect everything to work downstream from you unless everybody else in the world starts using Nuke.
Multi-channel EXR files of the type you've been asking to support are new. The ability to create them at all first appeared with the public beta of OpenEXR 2.0 on June 18, 2012 with the official release being April 9, 2013. That's recent enough that you still run into problems like the second gotcha above and why major programs like AE and Fusion are still relying on 3rd party plugins or scripting hacks to do anything with them. It takes more than three years for something like this to work it's way fully through the industry. It's also why, being brutally honest, your insistence and snide remarks about PNG's are wearing very thin. Support is not an absolute requirement for 3D work. Depending on what needs to be done there are going to be times you can't use multi-channel EXR files at all. There are going to be times where you can use them but doing so would be a monumentally stupid idea and times where it's just simply a poor choice. There are also going to be times where it would be hands down the absolute best choice and as a feature request it's a great one but in no case would it ever be an absolute requirement.
Funny. I know 3d artists who work for The Mill, Pixar, Dreamworks, and several post houses who, yes, use Photoshop image sequences. You're overlooking tons of working pros who work in concept, storyboard and animatics (i.e. low-res and early draft stuff) where everything is being laid out before needing final versions. In these cases 16-bit multi channel EXR is a total waste of time and resources, where a photoshop batch is a lot faster, and gets the storytelling across before working on the final versions. In fact, some of the artists at studios I know use photoshop sequences or near final renders. Photoshop can be used to quickly batch scenes with variant grades using tools identical to AE in less time in you're doing multiple tests of a short couple of seconds.
Yup, 3d forums aren't terribly interested in Hitfilm. Then again A) most artists on these forums are single users, not in an actual production environment--a one-man-band workflow is different from a team workflow--and Hitfilm isn't a dedicated 3D program... It's a compositor that happens to have some 3D functions as a nice bonus.
Now, as far as Hitfilm being "new?" Yeah, the OBJ format is from 1987. Lightwave, 3DS MAX, Photoshop, After Effects... All date back to the late 1980s or early 1990's. Hitfilm launched in 2011. A five year old program is certainly "new" compared to software with a 25-to-30 year life.
OpenEXR goes back to 2001. OpenEXR aov to November 2013. As I have already said to you before, there is only small benefit to OpenEXR aov, and, as @Aladdin4d correctly notes the benefit is soley file management, since it's a little bit easier to have a single file holding 10 or 12 alpha channels than 10 or 12 files with a single channel. Otherwise there is zero functional difference in compositing layers loaded in a single file vs layers in different files. Since render times are identical (saving as a single file or a multi-layer file doesn't change how much time it takes to calculate a shadow map, etc...) the only time advantage is in not opening and closing multiple disc I/O sessions--a matter of a few seconds per frame. Admittedly, a few seconds a frame eventually adds up, but it's still a convenience feature, nice to have, not a critical function. Period.
But, hey, what do I know? I've only been working in multimedia since 1997.
When you consider how long HitFilm's been around, FXHome's progress has been impressive.
It's true, HitFilm isn't all that popular in higher end productions, but it's so new the market that isn't surprising; the priority seems to have been on small, independent work which makes sense, but FXHome has added some stuff that's appropriate for higher end productions in every release. 8K support for example is clearly geared toward higher end productions; a filmmaker on a tight budget isn't likely to be shooting with a $50,000 camera package.
One thing I like about the way that FXHome has been developing HitFilm is that they aren't excluding the one-person indies while adding higher end functionality.
@WhiteCranePhoto Hitfilm won't break the hold Ae has on higher studios, of course. But that's why there's Ignite, bringing many of the more unique Hitfilm effects into Ae. We do know from some of the FxHome blogs that Ignite is being used on Hollywood productions and Hitfilm itself is getting used for previz. Nothing wrong with that!
Sign in to comment