Would anyone have some ideas why my 2.4 GHz 8-Core Intel Core i9 MacBook Pro would average 4 minutes 15 seconds to run a project in debug and my Intel Pentium Gold 7505 @ 2.00 GHz averages 1 Minute 50 Seconds (without making changes to the project between runs)?
Thanks for any help you can in saving my greying hair!
Both machines are SSD
The screen resolution on both computers is the default setting for them. 32 Bit (windows from the Mac targeting Parallels, but it actually takes longer again to run the same project with 64 bit compiled for Mac)
iCloud is switched ON - I didn’t realise this would be a factor?
How is your internet connection to Apple (i.e. not just to your isp)?
If the application is being code signed on debug, it may need to communicate with Apple (hardened runtime for instance).
Also macOS BS likes to check applications on every single launch, if it has to contact Apple and something in the pipeline isn’t working very well…
Thanks very much Sam. I figure my internet is ok (well downloading apps, etc, doesn’t seem any slower than with other activities) - also is the code signing standard now when building an app in Xojo? I thought this would be done in something like your own App Wrapper. The operation in your app doesn’t take long (maybe 40 seconds) - though I hadn’t timed notoriety, but that does take 4 or 5 minutes, I’d say.
When running debug on the Mac, the process is slower throughout the build process - if you start the Mac first and then the pc, you’ll see the pc quickly catch up and pass the Mac out. Testing on a 16 core iMac Pro shows similar results to the MacBook Pro by the way…
It is possible that your connection to the world is fine, but a hop in the chain to Apple’s time stamping services is broken or really slow. Last week it was failing for me.
You’d have to check with Xojo. I personally use App Wrapper to code sign (and do other things) during debug builds via a build script).
40 seconds… That is a long time. If you don’t mind, at some point (no rush), after you’ve processed with App Wrapper, can you select “Contact Ohanaware” from the Help menu and send me all the details, I’d love to take a peek into why it takes that long.
Could be a whole bunch of things under the hood that causing the slow down, or might just be marketing…
Hi DerkJ and thanks for replying. Maybe I’m misunderstanding you or I haven’t explained correctly. Profile Code is for when the app is running, correct? I am finding the issue in the Build stage when you run the project in debug mode
From the thread, it’s not entirely clear that we’re talking about the amount of time between when you press Run in the IDE to when your app launches, but now that it is…
Signing, if you have an Apple Silicon system, the IDE will apply an ad-hoc signature to your app whenever you build or run because it’s needed for the app to actually run now. It’s good enough for running locally, but not for distributing to other machines or long time use.
I’m assuming that the Intel Pentium Gold 7505 machine is Windows or Linux. The number of cores may be a factor here as more cores might mean that compiling is faster (although we stop allocating cores at a certain point because of diminishing returns). And if it’s Linux, the file system is significantly faster than macOS and can save a lot of time in saving temp files.
I am curious about the length of time though. Even the IDE project doesn’t take this long on my i9 8-core Mac running Big Sur. Is your Mac running low on RAM maybe?
The iMac is 3 GHz 10-Core Intel Xeon W 32 GB 2666 MHz DDR4 (I actually said 16 earlier) and does very little (MySQL and FileMaker Server running, little else) and the MacBook Pro’s spec is MacBook Pro (16-inch, 2019) 2.4 GHz 8-Core Intel Core i9 32 GB 2667 MHz DDR4. The Intel Pentium Gold 7505 is Windows 11 - it’s only 2 Core and 4 GB. Incidentally, I also Run the same Project in Debug mode on an M1 MacBook Pro and the compile takes a fraction of the time the 16"
ah but the difference between the Xeon and the Core i9 are drastic in terms of performance. I’ve got a 24 core Xeon Mac Pro here that takes 80 minutes to compile the macOS Universal IDE in aggressive mode whereas my 8 core i9 Macbook Pro does the same work in about 35 minutes. So I’m not really surprised at your statistics.