QLab 3 to 4 compatibility

[James Cooper]

Hi all,

Will QLab 4 open QLab 3 workspaces? (final release)

(Yes, I know, I know... QLab 3 is very old at this point!)

We're due to upgrade our entire playback system (QLab 4, and M1 Mac Mini's as an upgrade to some 2014 Minis) quite soon before our next production, but I'm hoping to program everything in v3, then bundle/carry it across relatively painlessly?

Has anyone got any old workarounds / tips / tricks so I'm not re-making the entire showfile?

I'd be interested in how the migration would handle audio / cue output levels/assignments too - my guess is that might cause issues between machines/software versions.

Many thanks!

[Sam Kusnetz]

Hello James

QLab 4 can open QLab 3 workspaces, and the translation is almost perfect, but you will need to go through everything and double check to be sure. Notably, curves in Fade cues, integrated fades within Audio and Video cues, and MIDI or Network cue fades might be subtly altered because of the way we changed (and improved!) the handling of fade curves in QLab 4.

Additionally, QLab 4 built upon and improved the way that video surfaces work, so video surfaces may need tweaking. My experience was that I needed to make only small adjustments and it’s not something you need to worry about, but it is something you need to watch for.

Audio levels, output labeling, routing, and stuff like that should all carry over exactly.

It’s worth mentioning: QLab 3 will not run on M1-based Macs at all, or if it does, we do not believe it will behave predictably. I encourage you to not even install QLab 3 on your M1 Mac.

You can always contact us at sup...@figure53.com for assistance and advice.

Best
Sam

––
Sam Kusnetz [he/him/his] (what is this?)
Figure 53
https://qlab.app | https://figure53.com

[James Cooper]

Hi Sam,

Thanks for this - that's very encouraging to hear!

A note on QLab 3 on an M1: I'm not going to try it - but if it's just the M1 vs Intel chip compatibility, would this be a scenario the Rosetta tech would be used for?
(I completely get that, due to age and obsoleteness, it won't be something you'll consider testing 🙂)

I noticed the QLab website didn't seem to recommend M1 Macs - has any performance testing been done comparing Intel chips vs M1 performance with QLab?

Best,

James

[Sam Kusnetz]

On May 4, 2021, 4:04 PM -0400, James Cooper <ja...@jcooper.tech>, wrote:

A note on QLab 3 on an M1: I'm not going to try it - but if it's just the M1 vs Intel chip compatibility, would this be a scenario the Rosetta tech would be used for?

It’s more that M1 Macs require Big Sur and QLab 3 is not compatible with Big Sur. Also, Rosetta is less successful translating older apps, and QLab 3 certainly is older.

(I completely get that, due to age and obsoleteness, it won't be something you'll consider testing 🙂)

We’re a small company, and we’re putting all our energy into supporting the current version of QLab and future development.

I noticed the QLab website didn't seem to recommend M1 Macs

Could you help me find the parts of our website that gave you that impression? It’s not something we’re intentionally expressing, so I’d like to adjust it where I can.

The FAQ and QLab manual, for example, both note compatibility with M1.

has any performance testing been done comparing Intel chips vs M1 performance with QLab?

Not really. We test for compatibility, but apple-to-apples performance tests are very challenging, and purchasing reference Macs to do these tests can get really expensive really quickly. It’s also really hard to know what sort of tests to run that provide meaningful and consistent data. Should we test the number of simultaneous Audio cues a system can play without dropping frames? Well, a single dropped frame here and there is almost never audible, so that’s not really a meaningful test. So where do we set the bar? What combinations of cues is reasonable to test?

Because QLab is used in so many different ways and with so many different configurations, our general policy is to ensure compatibility and strive for reliability, and then allow specific real-world use cases to define the boundaries of what’s possible given a particular configuration.

Best
Sam

[James Cooper]

Hi Sam,

Thanks for getting back to me!

On the M1 recommendation on the website: https://qlab.app/docs/v4/general/system-recommendations/

Despite mentioning the M1 compatability at the start, under the "Processor" heading, the page still gives me the impression to choose i7/i9/Xeon processors over the M1 chip - just wondering where the M1 might fit into that lineup - i.e. perhaps not as good as a Xeon, but maybe better than an i7.

I think you make a very good point on performance tests - exactly how you'd benchmark is beyond me... likely something to do with video codec performance, and multiple surfaces... in my experience, that's usually where QLab has chugged (usually running multiple hi-res video feeds, and mapping/corner pinning them down to 10% scale on the output - which isn't the most efficient thing anyway)

Simply because of how diverse the software is, it's almost impossible to benchmark... but a stress test or something might be useful - that said, that info is generally being released by the hardware enthusiast community all the time - as well as us end-users finding it out as we use it.

Anyway, this is off topic now - thanks for addressing my original query!

Best,

James