FCPX And El Capitan

Why the angst if I have not updated??? Very simple, I have noticed a disturbing trend with FCPX development, and I am only trying to understand what is going on.

Not for the first time has an update been released having been through internal testing, then into limited in field public testing, for this to do some sort of mischief to the end user system, as was the case with the repeating updates to Video Codecs, or now with audio issues/pug ins together with GUI issues.

I fully grasp that FCPX is complex, with loads of lines of code, very smart brains at work, at times during my life I have not seen wood for trees, editing at high speed, I have not found the shot I was looking for, I missed it, this happens.

Surely logic dictates that it would be far simpler to issue small updates to get rid of tiny errors, if you try and patch too many things at once would this not cause a problem down the line?

How can I explain this? If you iron out the tiny issues, this would lead to a clearer less complex patch later? Am I wrong in thinking this..If so I am so sorry, maybe I am not the most logical person around, but updating from lest complex reduces the complex patching later?

We of course have no idea about the size or duties of each person at Apple that works on FCPX coding for improvements and those on debugging and coding for error correction.

So really what is we want as a user group? Constant updates that are at times buggy, or a stable error free stable system? I would prefer a stable error free stable system, that I am more than willing to wait for a once a year, massive update...

Adding more features without debugging the errors leads to frustrated users having to find workarounds, that don't always work. I think it is more than reasonable to take to foot off the progress pedal, take time to really iron out the bugs, ind what is causing errors, possibly open into debate with users about what they want to see...

"Surely logic dictates that it would be far simpler to issue small updates to get rid of tiny errors"

No, there are no tiny errors with software programming. Change one line of code, it can potentially have a rippling effect to hundreds or thousands of other lines of code. There are no simple fixes like you imagine it to be. That is simply not how programming works.

If I have a pile of 10,000 pebbles, all relying on the placement of all others to retain the structural integrity of the stack, moving one, be it in a good or bad location, will endanger the stability of all others.

If it were possible to program as easily as you imagine it to be, software technology would be vastly beyond where it is now. The point we keep trying to get across you to, Justine, is that you over simplify what programming takes. You've never done it, you've no experience in it, yet you defy those of us who have, when we try to explain. The world is not flat, no matter how much a pope claims it to be so.

ALL software development companies have multiple programmers, each working on specific features, or bugs, or interface designs. Many things can, and MUST be done at the same time, or that developer will not make enough progress fast enough to keep up with their customer base's needs. That's the fact of software life.


"I have noticed a disturbing trend with FCPX development"

Nope, there's no disturbing trend with FCPX development. There is no evidence of this, is simply does not in fact exist. No clue why you continue to insist on some nonexistent administrative flaw.

Let me ask this question, Justine; Can you give us an example of a application and software developer who does things the way you imagine they should be done?

MsJustine wrote: I have noticed a disturbing trend with FCPX development, and I am only trying to understand what is going on.

Sorry, but I do not concur with your assessment. There is little if any evidence to support your premise, aside from the current El Capitan compatibility which for the most part is very minor in scope.

Under Yosemite, there are few issues with FCP X 10.2.2 ... at least there doesn't seem to be any major universal discrepancies or issues that preclude the majority of users from accomplishing their tasks.

Once again, I will point out that you, nor I have incurred zero additional expenditure whatsoever for any update or maintenance to FCP X since the original purchase. I reiterate ... all the updates since v10.0 have cost us exactly zero, zip, zilch ... nada ... absolutely nothing ... which is all the more perplexing why you seem so annoyed.

Perhaps you should seek a refund from Apple since you have experienced an issue trying to use FCP X on El Capitan .... Err ... wait a minute ... you haven't tried OS X 10.11 yet ...

At least wait until Apple has issued an update for El Capitan compatibility. Then, I may just join you in your effort if Apple would fall short in a significant manner. Until then, this is merely a tempest in a teapot ....

MsJustine wrote: ...I have noticed a disturbing trend with FCPX development...Not for the first time has an update been released having been through internal testing, then into limited in field ...audio issues/pug ins together with GUI issues.

What release of FCP X are you talking about? 10.2.2 was released before El Capitan -- it was the OS X release which caused problems, not FCP X. A source code freeze happens long before the final release. When that happened for FCP X, OS X was even further back so the FCP X team would have had to make a speculative fix for an unreleased OS version still being changed.

MsJustine wrote: ...Surely logic dictates that it would be far simpler to issue small updates to get rid of tiny errors, if you try and patch too many things at once would this not cause a problem down the line?

While it is true that making a small change entails less risk than a large change, even the tiniest change requires a full test cycle -- if it's going to be widely released. This cannot be fully automated and requires a lot of people. Apple would have to pull the test team off of 10.2.3 or 10.3 to test a release with the timecode cosmetic fix. That would then delay the next major version for no substantial benefit.

The only exception to this is some product development groups release "hotfixes" which are less-tested versions that are distributed to a single site or limited sites. The limited distribution helps contain the risk and is typically only done for critical problems.

Also the notion that it's a small, safe fix may not be correct. The garbled timecode bitmap might indicate it's related to deeper changes in the OS X graphical subsystem.

MsJustine wrote: ...If you iron out the tiny issues, this would lead to a clearer less complex patch later?

The current source tree is probably hugely different from 10.2.2. They likely have major structural changes underway due to Metal and other features. They would have to fork the source tree back on 10.2.2, make the minimum fixes for the timecode display problem, then later when 10.3 is finished remember to go back, pick up that fix and roll it into the latest source tree. Or if the fix is already in 10.3, extract that and see if it can be retrofitted into the 10.2.2 source tree. Oftentime there are non-obvious dependencies. To quote Buckaroo Banzai: "Don't tug on that. You never know what it might be attached to."

MsJustine wrote: ...We of course have no idea about the size or duties of each person at Apple that works on FCPX coding for improvements and those on debugging and coding for error correction.

For business or professional software, the industry standard is roughly a 1:1 ratio between software developers and test engineers but this can vary from 2:1 to 1:2. I don't know the size of the FCP X dev team but a rough guess would be 30-50 with an equal number of testers.

I found this on the world wide web

Which, once again, proves that you never can trust anything that people post on "the world wide web"

FCP X is not broken at all with El Capitan, I'm running it very comfortably on my new iMac 5K test system. As a matter of fact FCP X is very snappy and fast, all plugins work as expected now (I have updated iZotope to RX 5 Advanced) and the little tc UI glitch is hardly noticeable on this superb screen.

I still have reasons why I won't update our production systems to EC yet, but FCP X is certainly not one of them.

- Ronny

Ronny I think the Image is applying when updating the OS make sure the App is compatible with the OS and not the other way around.

I still find the meme hilarious!

I absolutely agree the picture itself is hilarious, and thanks for posting it

- Ronny

How about this:

ronny courtens wrote: Which, once again, proves that you never can trust anything that people post on "the world wide web"

FCP X is not broken at all with El Capitan, I'm running it very comfortably on my new iMac 5K test system. As a matter of fact FCP X is very snappy and fast, all plugins work as expected now (I have updated iZotope to RX 5 Advanced) and the little tc UI glitch is hardly noticeable on this superb screen.

I still have reasons why I won't update our production systems to EC yet, but FCP X is certainly not one of them.

- Ronny

I recently took advantage of iZotope's upgrade offer myself and upgraded to RX 5 and Ozone 6 (Ozone 7 when available), both of which will be supported in El Cap.

Glad to know in advance of upgrading to El Cap of the things that will not be compatible. The iZotope apps are big for me and I'd never upgrade to El Cap until they are ensured to be compatible.

edit to add:
The Ozone 7 registration key was e-mailed to me shortly after posting this message. A bit earlier than I thought.

Re: TC Window.. How come Motion isn't affected? Fonts looked the same to me.

Interesting Stingray, no cosmetic issue in my Motion.

Two very different sets of coding. One line can effect things all over the place in an application. It may not be the code for the Dashboard, it may be somewhere else in the app that effects the Dashboard. Very easy to understand. Well, if you've ever worked for a software development company before.

taught programming in schools and universities as well as working for top level Government agencies for over 22 years Ben.. think I can understand it thanks!

Just answering a question. I was an IT engineer for over 30 years.

So now that Sony has fixed their SxS card issue, and all but iZotope and Persons have upgraded AU products, what is left to fix in FCPX with EC? Besides the trivial Dashboard cosmetic?

Hey Ben,

- I just got an e-mail from a user who has issues with titles crashing 10.2.2 since he updated to El Capitan. I don't have this problem on my new 5K iMac, but this does not mean it isn't possible. I have advised him to do the test with a new admin user account to see if this solves the issue or not. If it does, he can easily restore his user directory and this should fix things.

- One very important bug report I have received yesterday is with El Capitan on shared storage systems. Mads Larsen Nielsen from Metronome has reported that he had connected two new iMacs with EC to their shared storage network and he got significant bandwidth drops on these machines, making them unusable. Both machines had a clean install of EC and only these machines show problems, also on their second shared storage network. This bug has been reported and confirmed.

Will keep you updated if I have more news. EC is still a big no-go for me.

- Ronny

Why did Apple allow plug ins to crash FCPX?

Where is the quality control? Surely Apple has to test and confirm that plug ins are safe and will not negatively affect FCPX?

MsJ
Essentially NO.
Apple does not approve 3rd party plug-ins. In OS, unlike iOS, you are free to download plug-ins from anyone/anywhere.
Apple provides APIs and even software development environments like Swift for creating apps/plug-ins, etc… but that doesn't prevent developers from creating bad software. It is up to the developers to make sure their software is compatible. If the developers are part of the Apple Developer Program, they get advanced access to new OS software via a Non Disclosure Agreement (NDA) to fix/improve their software.

Hope this helps, Greg