[Joe Marler]

[Oliver Peters] “because this is using a Xeon processor, you do lose the advantage of accelerated H264 encoding that the Core-i chips offer. So in theory, encoding MP4 and other H264 files might actually be slower on this machine than current fully-loaded iMacs and MBPs.”

That is a possible issue. OTOH if you lose 3x performance due to no Quick Sync but gain 4x from 18 cores, it still might be faster. There are also other possibilities — see my comments in this post: https://forums.creativecow.net/thread/335/95701#95782

[Joe Marler]

[Andrew Kimery] “Since the iMac Pro will use Xeon CPUs does that mean it won’t have Quick Sync to speed up H.264 encoding?”

Intel does not currently put Quick Sync on CPUs with more than four cores, so unless this is changed the iMac Pro will not have Quick Sync. This means it will not have hardware-accelerated encode/decode for H.264, H.265, VP9, AV1 or any wrapper format using those codecs such as AVCHD and XAVC-S. The Mac Pro doesn’t have Quick Sync either but that was designed in the pre-4k era and before H.265 was looming on the horizon. H.265 is much more compute-intensive than H.264, therefore needs hardware acceleration even more.

OTOH since Apple makes hardware, OS and (in the case of FCPX) the application, in theory they could rapidly add code to use AMD’s hardware acceleration VCE (Video Coding Engine): https://en.wikipedia.org/wiki/Video_Coding_Engine

VCE like nVidia’s NVENC resides on the GPU but is functionally separate from the GPU — it is a “bag on the side” from an architectural standpoint. Long GOP encode/decode cannot be meaningfully accelerated using traditional GPU methods, it requires fixed-function logic, IOW an ASIC (Quick Sync being one example).

Whether VCE is actually practical would depend on an extensive technical evaluation of performance, image quality, and how many software changes are required at the OS and application layers. It would also depend on how functional and stable the software development framework which supports VCE. If the hardware works but the development libraries and other software does not, or is not available in macOS, then it’s not an option.

The lack of Quick Sync in most Xeon CPUs has long been an issue for the video world. It potentially impacts all editing software but some software like Premiere either doesn’t use it, doesn’t use it on Mac or doesn’t it effectively. FCPX has long used Quick Sync very effectively so the impact is significant.

I’ve never seen a good explanation from Intel or anybody else why Quick Sync is withheld from CPUs with more than four cores. A possible explanation may be it requires GPU on-chip resources, even though the actual encoding logic is separate. This would require Xeon to have the full on-chip GPU which is a significant hit to the transistor budget. Xeon is mostly a server and workstation chip. Servers don’t need an integrated GPU and workstations will have a discrete GPU, so putting the integrated GPU on chip would waste transistors. The only loser is the subset of the video editing community that relies on Quick Sync. It was probably a tradeoff.

[Joe Marler]

[greg janza] ” I have two OWC Thunderbay raids that are pc formatted because I primarily edit on a pc so the raids are connected to my mac through ethernet. Can I use these raids with FCPX? Nope….My raids work just fine on my pc and mac with premiere. They are ntfs…”

macOS cannot write to NTFS volumes, only read from them. Your observed problems may have less to do with FCPX than with your storage configuration. I actually don’t see how Premere is even working in that config, unless your project and scratch files are not on that RAID. In general nobody using a Mac should expect normal read/write app behavior on an NTFS volume.

Mac users needing write access to NTFS will often use Paragon NTFS: https://www.paragon-software.com/ufsdhome/ntfs-mac/

I use Paragon when I infrequently need write access — for any Mac app, not just FCPX — to NTFS volumes and it works fine, including for FCPX.

But you are apparently not directly connecting your Mac to the RAID, but using ethernet. Unless that’s 10 gig, that will cut your I/O performance (for both Premiere and FCPX) to a fraction of what’s available from the RAID, so this would not be a good configuration for evaluation. This is especially so for FCPX which does many database I/O calls.

However using network-attached storage may explain why Premiere works at all. In that case the file I/O calls are translated by the network layer to remote filesystem I/O calls, so this can buffer an app from dependency on a specific file system.

Why FCPX doesn’t work perfectly in that case, I don’t know. If using libraries on an SMB share, this requires 10.3.x to work properly: https://www.studionetworksolutions.com/final-cut-pro-libraries-with-smb-shared-storage-first-look/

If you want to evaluate FCPX from a feature or performance standpoint, it’s probably better to put those files on locally attached storage, then after that evaluation you can investigate how best to make that work for network-attached storage. It is definitely possible and people do this frequently. But using both Premiere and FCPX with local storage would simplify the evaluation and ensure you aren’t being thrown off track by items unrelated to the product.

[Joe Marler]

[Bob Zelin] ” iMac Pro’s top of the line ATI card…Radeon…will be the equivalent of the
NVidia 1060, which is fine. But since the iMac Pro starts at $4999, are we talking about an iMac that is the same price as a blown out HP Z840, “

I don’t think anybody knows the exact performance of the Radeon Pro Vega GPU to be used in the iMac Pro. The only number they mentioned was 11 teraflops which is about equal to an nVidia 1080 Ti. We won’t know for sure until the exact model of the GPU and specs are released and maybe not until the iMac Pro itself is released and benchmarked. Apple’s WWDC video said the Radeon Vega GPU was 3x faster than the current Mac Pro GPU — don’t know if that is single or dual cards.

Re price, Apple’s video said it was $2000 cheaper than a tower PC configured to equal specs. I don’t know about that, but the cheapest Dell Precision workstation I could configure with 8-core Xeon, possibly similar GPU and 1TB SSD was over $5k, didn’t remotely approach 3 gigabytes/sec SSD performance, plus it didn’t include a monitor.

[Joe Marler]

[KB Burnfield] “Joe, you were right. I opened the captured video in VLC and the sharpness and quality were much better but the interlacing was very apparent.

Are there any Deinterlace filters / plugins anyone would recommend? In the one video there’s a fair amount of motion from a handheld camera.”

Interlacing is not by itself wrong if the playback chain properly deinterlaces. E.g, in VLC pick Video>Deinterlace>On, and then pick Video>Deinterlace Mode>Yadif, or try several different algorithms. In some cases Youtube will automatically deinterlace uploaded content.

FCPX does a pretty good job deinterlacing by itself. I haven’t tested 3rd-party plugins. BorisFX had one for Legacy FCP but I don’t think it works in FCPX. To deinterlace in FCPX, add that content to a progressive project, then in Inspector, click the Info button at top, then at bottom click the drop-down menu and select Settings, then set the Deinterlace check box. Then export that as (say) 720p and compare it in VLC to the original content.

Always examine the output very closely using VLC or other tools which allow selective deinterlacing. It is very easy to accidentally “bake in” deinterlacing artifacts, not notice it, then the material is permanently degraded for downstream viewers.

[Joe Marler]

[Wu Zhengjiang] “if I want use old footages, I couldn’t use search anymore, because all the footages are packaged in FCPX libraries.”

Two possible options are FinderCat Pro: https://intelligentassistance.com/findrcat-pro.html

https://www.youtube.com/watch?v=f_njT4qmwV4

And KeyFlow Pro: https://www.keyflowpro.com/

Some contents or functionalities here are not available due to your cookie preferences!

This happens because the functionality/content marked as “Google Youtube” uses cookies that you choosed to keep disabled. In order to view this content or use this functionality, please enable cookies: click here to open your cookie preferences.

[Joe Marler]

[Scott Thomas] “looking at the DV footage through whatever deinterlacing is in Quicktime player? If you play the footage back out to a SD broadcast monitor, it would probably look like your memory of it. “

This is a very important point. DV is interlaced and the deinterlacing behavior of various playback methods will vary. Quicktime Player will deinterlace sometimes, based on the metadata in the video file header. VLC by default does not deinterlace but has several user-selectable deinterlace algorithms. If DV content is imported to a progressive FCPX project, the viewer will usually display that as deinterlaced but if exported it may or may not be deinterlaced. If imported to an interlaced project the export will never be deinterlaced — even if the deinterlace checkbox is set in Inspector — since by definition the project is interlaced.

There are cases where DV content can be imported to a progressive project and if not manually deinterlaced, the export will discard alternate scan lines. FCPX behavior seems to vary based on the particular file metadata. As Scott said, the important thing is be aware of these areas, especially that the playback method can alter your perception of video quality. The quality issues do not arise from how DV is imported but how it’s handled after the import, esp. regarding interlacing. Always check the exported quality vs the original files using different playback methods. VLC is good since it provides manual control over deinterlacing and various algorithms.

In general most content delivery is progressive except for ATSC TV which is mostly 1080i/29.97, although Fox and ESPN are 720p/59.94. Because of this it can make sense to “hard deinterlace” interlaced content such as DV. However it is really easy to “bake in” artifacts from poor deinterlacing, and then nobody downstream can fix it.

DV is also 4:3 so that’s another complication. Some DV cameras could shoot 16:9 by matting the 4:3 frame. That in turn discarded scan lines which degraded resolution, so hopefully the content is at least 4:3 non-matted.

[Joe Marler]

[Ray Sherman] “In FCPX’s Media Import window it shows H.264 Codec & MPEG-4 as File Type. Basically, all I want to do is copy my video natively without any quality loss. I was always under the impression that H.264/MPEG-4 is less quality over 4K. “

All you need to do is import it. H.264 is your camera’s native codec. That is the best available quality for that camera at 4k. You do not need to transcode to optimized media for quality reasons. If you need to transcode to proxy for performance reasons, that is OK.

Your only real decision is whether to import from the card, or make an archive then import, or copy the folder tree to disk then import from there, or copy just the video files to disk and import from there. None of those affects image quality.

In theory the best practice is keeping the entire folder tree (which an archive or import from card also does) because some metadata is stored there, but FCPX will not import from a tree using “leave files in place”. It will import from a copy of the video files using “leave files in place”. I would never do this for AVCHD but for XAVC-S it seems to work OK and it saves space and makes import faster.

[Joe Marler]

[Ray Sherman] “I re-checked my RX100v settings and created a new clip and tried importing directly off the SD card into FCPX to make an archive. Once again, it shows my video as being a H.264 codec instead of XAVC S. “

XAVC-S and AVCHD both use some variant of the H.264 codec. So they are all H.264, which is why a codec inspection tool will show H.264 or AVC. XAVC-S and AVCHD are just two different wrapper formats. XAVC-S is fine for the RX100V. If you have a choice of bit rates and resolution, pick the highest one. On our A6500 and A7RII cameras we use UHD 4k at 29.97p at 100 megabits/sec.

We copy the video files to a hard drive, then import using “leave files in place”. That is not a good practice for AVCHD but for XAVC-S it works fine. You can also import from the card or make an archive copy like you described.

To obtain smooth 4k editing performance (whether Premiere or FCPX) it is often necessary to transcode to proxy. Optimized media is not needed, but proxy usually is. This can be done during import or afterward.

[Joe Marler]

[Jenn Lindsay] “If the iPhone is filming at 30fps but the lights in Greece are 50Hz, how come there’s no flicker with the iPhone camera? It only happened with the C100. “

Jenn, that is a good question. The problem is interaction between shutter speed (not frame rate) and power line frequency. Admittedly maintaining the 180 degree shutter rule will require a certain shutter speed which in turn is determined by frame rate: https://www.red.com/learn/red-101/shutter-angle-tutorial

We don’t know what shutter speed the iPhone uses at 30 fps. Unlike stills, EXIF data is not as standardized for video so it’s often difficult or impossible to analyze the file and determine video shutter speed. Oftentimes lower-end automatic video cameras will use a variable shutter speed to try and maintain exposure.

However a more likely reason you don’t see this on an iPhone or small-sensor camcorders is they use a “global shutter”, whereas large-sensor camcorders like the C100 use a “rolling shutter”. Note these are both electronic not mechanical. A global shutter exposes the entire sensor at once, whereas a rolling shutter sequentially scans the sensor. A rolling shutter is an unavoidable technological limitation on many camcorders with large CMOS sensors, although a few do have global shutters.

For material shot by large-sensor rolling shutter cameras in regions with differing power line frequencies, there is no easy answer. It’s probably best to use the locale-specific frame rate and 2x shutter speed, then in post figure out the best way to conform the footage. Normally you want everything shot at the same frame rate but in this case you don’t have a choice. Conforming different frame rates is a lot easier than fixing a flicker or banding problem.