[Chris Kenny]

[Gustavo Bermudas] “I’m sure Blackmagic not only hurts developers, but it also hurt a lot of post houses that use Resolve as well (and paid for it), prices has gone down incredibly low for post services, and I remember a few years ago we were eager to buy new stuff that came out, now we’re like “do I really need this?”
I really love Resolve to death, but I despise Blackmagic for what they’re doing, and right now we’re tied to it because there are no other options, why should I invest in a 6 figure system when clients want to pay a figure less?”

Would you rather have post production facilities competing on the basis of how much capital they have access to, or competing on the basis of efficiency and talent? Expensive tools give you the former. Cheap give you the latter. It seems like when evaluating the effects of Resolve’s pricing you’re counting the harm done to long-established high-cost facilities, but not counting the benefits reaped by new facilities (e.g. my company), by clients, or by consumers who ultimately get access to more and better content.

Anyway, there are still fairly expensive things clients will pay more for. Grading off of a cinema projector in a DI theater instead of crowding around a 24″ monitor in some guy’s basement, for instance. Setting up a serious room can still run to six figures even with Resolve at $995.

—
Digital Workflow/Colorist, Nice Dissolve.

You should follow me on Twitter here. Or read our blog.

[Chris Kenny]

[Walter Soyka] “I agree that I/O hardware is necessary with respect to “serious” grading (are we saying serious now because it’s less incendiary than professional?) — but I think there’s a lot of serious editorial that does not need I/O in 2014.

I also think we are not that far away from “semi-serious” grading happening under a display LUT, obviating the need for video I/O.”

I agree that the technical benefits of ‘real’ video I/O over GUI display monitoring are receding (though it would be nice if Apple would finally enable 10-bit output), but again, in Resolve, it’s not just about those technical benefits. You cannot display your program image full screen on one GUI display while interacting with the application on another display. The only way to get a full-screen program view while interacting with the UI is to buy BMD hardware. I think that’s a pretty critical capability.

[Walter Soyka] “BMD should start making display probes.”

Yeah, they could easily adapt the OEM version of the i1Display Pro like many other vendors do (HP with the Dreamcolor probe, SpectraCal with the C6) and have a whole calibration system built-into Resolve rather than just the test pattern generator it presently has. It’s a good idea.

—
Digital Workflow/Colorist, Nice Dissolve.

You should follow me on Twitter here. Or read our blog.

[Chris Kenny]

[Walter Soyka] “I’m not arguing that Resolve wiping out the editorial market is inevitable — just pointing out that BMD is capable of destabilizing the market by sloshing around some revenue from ancillary markets. Not everyone competing in the space enjoys this luxury — small developers especially — and it may (or may not!) be negative in the long run for the industry.”

Buying BMD’s I/O hardware is basically a requirement for using Resolve Lite for serious grading/editing. Without this, you can’t monitor a full-screen image while simultaneously interacting with the UI. As far as I can tell (and someone correct me if I’m wrong here) this is still true in Resolve 11; the new dual-screen UI can split bins/scopes to one screen and viewers/timeline to another, but you still can’t monitor full screen while interacting with the timeline unless you have BMD I/O hardware.

This means that in practice Resolve Lite is ‘free’ as a transcoding and media management tool, but if you want a useful grading or editing tool, it still costs money.

—
Digital Workflow/Colorist, Nice Dissolve.

You should follow me on Twitter here. Or read our blog.

[Chris Kenny]

[Frank Gothmann] “They may be developing it much more (and I hope it’s the guys from Davinici that are still there doing it), question is how stable and bug-free the outcome is.”

My experience with software has been that the more obscure a product is, the flakier it tends to be. The extreme case of this is that software developed for in-house ‘enterprise’ use is notoriously terrible, but the same phenomenon applies in less extreme cases as well. In general, a larger user base means more bugs are discovered more quickly, and if moving more copies at a lower price leads to higher total revenue (which is likely the case with Resolve, particularly counting all the video interfaces, etc. it helps to sell) then there are also more resources available to fix those bugs.

—
Digital Workflow/Colorist, Nice Dissolve.

You should follow me on Twitter here. Or read our blog.

[Chris Kenny]

[Anish Prithviraj] “Hi Prathvish/Chris,

Please make sure that you disable “Use Red Rocket if available”. If that is enabled and a rocket is available, it is using the rocket instead of GPU (even if GPU for Red Debayer is enabled). “

Unchecked here (and the Thunderbolt Rocket is disconnected). I’m curious if something might be wrong here and I should be seeing better performance.

—
Digital Workflow/Colorist, Nice Dissolve.

You should follow me on Twitter here. Or read our blog.

[Chris Kenny]

[Rohit Gupta] “We’re getting 5K full-res in RT on the 12-core/d700.”

Is that with full frame footage or is that 5K widescreen? Mine seems to settle down at ~18-20 fps with full res full frame 5K material. Same in ‘half res premium’ mode. (I copied a couple of clips to the internal SSD to rule our storage bottlenecks.)

—
Digital Workflow/Colorist, Nice Dissolve.

You should follow me on Twitter here. Or read our blog.

[Chris Kenny]

[prathvish hegde] “while there isnt much difference between the redrocket and titan but i think resolve is not using the second titan for debayeringing r3d also half resolution and full resolution premium are playedback at the same speed 🙁 “

Seeing the same thing on a 12c/D700 Mac Pro — Half Res Premium with 5K Epic footage seems to be basically the same speed as full-res (similar speeds to what you’re seeing). I wonder what’s up there.

Half Res Good is silky smooth now though, and full-res decodes are actually a little faster than what we got from the Rocket over Thunderbolt, so the Rocket is officially getting retired from our grading suite. I suppose it will still have its place for on-set transcoding (hooked up to a MacBook Pro). Although if buying a system for on-set transcoding today we’d probably just figure out how to pack up a Mac Pro now that it’s so tiny, and forget the Rocket.

—
Digital Workflow/Colorist, Nice Dissolve.

You should follow me on Twitter here. Or read our blog.

[Chris Kenny]

[Marcus Moore] “Playing back undebayered R3Ds without a RedRocket card is a wonderful thing. Even if you can only do one stream (Higher Quality). I hope we’ll see some progress from RED on their 3rd party GPU support at NAB.”

Third-party GPU decoding support is starting to show up in the wild. DaVinci Resolve and SCRATCH just shipped updates with it in the last couple of days.

—
Digital Workflow/Colorist, Nice Dissolve.

You should follow me on Twitter here. Or read our blog.

[Chris Kenny]

[Craig Seeman] “Mercury Elite Pro Qx2 can do USB3 RAID.
https://eshop.macsales.com/shop/hard-drives/RAID/Desktop/“

But you’d need to hook up at least two of them and do simultaneous data access in the same direction across both of them before you’d feel the difference between one PCIe lane for all the ports and one PCIe lane per port. Actually, even that might not do it. Those units benchmark below 200 MB/s each over USB 3. At that data rate you’d need three of them, with simultaneous data access in the same direction across all three, before you’d be bottlenecked by that 500 MB/s PCIe lane.

That’s a pretty contrived situation. In general, Apple knew they’d need to accept bottlenecks in some places given the number of PCIe lanes available from the CPU and the chipset… and they were pretty smart about locating those bottlenecks so they’d cause the fewest compromises in real-world performance.

—
Digital Workflow/Colorist, Nice Dissolve.

You should follow me on Twitter here. Or read our blog.

Tweets by NiceDissolvePro

[Chris Kenny]

[Brian Mulligan] ““Apple’s newer Final Cut X release did not have the capabilities we needed to execute a news-oriented workflow,” Koetter explained, “so we went through a rigorous process with Adobe to vet Premiere for our needs and they responded with a very large number of modifications to their NLE, including features in their ‘editing finesse’ category and closed captioning tools. To us, Premiere appears to be the best-in-class craft editing system for our purposes.””

Doesn’t seem like there’s any real surprise here, unless it’s that Avid’s collaborative editing platform lost this one. FCP X doesn’t have an equivalent of Anywhere, which clearly played a major part in their decision. On the other hand, Anywhere, at least in its current form, is basically irrelevant to probably 95% of NLE customers, so the takeaway from this for most people is approximately nil.

—
Digital Workflow/Colorist, Nice Dissolve.

You should follow me on Twitter here. Or read our blog.