Board index FlightGear Release candidates 3.0

3.0.0 MacOS first impressions

This is the archive of topics about the 3.0 release candidates.

3.0.0 MacOS first impressions

Postby someguy » Sun Jan 19, 2014 5:57 am

I grabbed the first RC from Jenkins very soon after it was built. Living dangerously, you betcha...

So far, so (mostly) good. The Go To Airport dialog works, yay! If/when METAR loads, the clouds look better than ever, dare I say terrific, and now actually extend way into the distance, if that's how you set them. I say if/when, because often it needs a nudge from the apply button in the weather dialog.

However, the red glideslope boxes still dive underground well short of the runway, just as in 2.12. Also, navdata often wants to reload at spawn, which takes several minutes (5-10?), and if terrasync is enabled, it goes into an endless download even if you already have the scenery, not spawning your plane until scenery download is disabled from the menu. That's a pretty big bug, unless some new undocumented feature is going on that I don't understand.

I don't see the shallow water color around tropical islands, though enclosed lagoons are turquoise now. Maybe that's another Mac driver issue, like Rembrandt. Frame rates don't seem different from 2.12. It was late, and I only saw two other planes (777s) on the ground, but both were at least half submerged in the tarmac. I was not.

Shift-escape resets are still buggy. If I spawn onto AI Nimitz and do a reset, my plane tries to respawn above the middle of KSFO, typically crashing when it buries its nose into the grass. 2.12 did that, too.

Hope this helps. I'm sure I'll find more stuff, both good and not so much, over the next few days.
User avatar
someguy
 
Posts: 1650
Joined: Tue Nov 25, 2008 5:54 am
Location: USA
Version: 2019.1.1
OS: Mac OS X 10.11.6

Re: 3.0.0 MacOS first impressions

Postby Thorsten » Sun Jan 19, 2014 7:31 am

I don't see the shallow water color around tropical islands, though enclosed lagoons are turquoise now. Maybe that's another Mac driver issue, like Rembrandt


They're not available at all incarnations of the water shader / all quality levels. So that's very unspecific.

and if terrasync is enabled, it goes into an endless download even if you already have the scenery, not spawning your plane until scenery download is disabled from the menu.


I *think* from my own observation what it tries to do is phone home to the server to see if there need to be updates (it displays downloading scenery, but no data rate, and system monitor shows no incoming data stream), and if the connection is not too good, this takes a bit. It's the same thing which I get when accessing GIT sometimes. Technically the terrain is an SVN repository, so I think this makes sense. I don't think it's a bug as such, but it clearly doesn't behave as the user expects it. Well, once of the reasons why I'm personally not so much into terrasync.
Thorsten
 
Posts: 12001
Joined: Mon Nov 02, 2009 8:33 am

Re: 3.0.0 MacOS first impressions

Postby zakalawe » Sun Jan 19, 2014 12:51 pm

someguy wrote in Sun Jan 19, 2014 5:57 am:However, the red glideslope boxes still dive underground well short of the runway, just as in 2.12.

That's an, ahem, 'unmaintained' feature, if you want it fixed probably best to look at it yourself.

someguy wrote in Sun Jan 19, 2014 5:57 am:navdata often wants to reload at spawn, which takes several minutes (5-10?)

It should rebuild on the first start with the new version (3.0), please be more specific about 'often'. Remember if you change scenery paths it has to rebuild too.

someguy wrote in Sun Jan 19, 2014 5:57 am:if terrasync is enabled, it goes into an endless download even if you already have the scenery, not spawning your plane until scenery download is disabled from the menu.

This is a real bug, but needs further information - do you have a working 'normal' Internet connection at this point, or something that might look like one? I ask because I've tested both 'offline' (no internet connection, or the terrasync site not available) and with an Internet connection. If there is a problem with the connection, it should timeout, so it shouldn't spin for more than 20-30 seconds, usually much less. The only thing I can think of that would make it seem to take forever would be if you have a connection, but it's really, really slow - like, dial-up modem slow. If you could run with --log-level=info and post a bug report in the bug tracker I will take a look.

someguy wrote in Sun Jan 19, 2014 5:57 am:Shift-escape resets are still buggy. If I spawn onto AI Nimitz and do a reset, my plane tries to respawn above the middle of KSFO, typically crashing when it buries its nose into the grass. 2.12 did that, too.

This is the first report I've heard of this, if it happened in 2.12 too, did you already mentioned it somewhere that I missed?
zakalawe
 
Posts: 1257
Joined: Sat Jul 19, 2008 4:48 pm
Location: Edinburgh, Scotland
Callsign: G-ZKLW
Version: next
OS: Mac

Re: 3.0.0 MacOS first impressions

Postby sa7k » Sun Jan 19, 2014 12:57 pm

someguy wrote in Sun Jan 19, 2014 5:57 am:It was late, and I only saw two other planes (777s) on the ground, but both were at least half submerged in the tarmac. I was not.

If you mean in multiplayer that is because other users have different scenery, this gets fixed as all people updates to the new scenery.
sa7k
 
Posts: 380
Joined: Fri Mar 16, 2012 2:24 pm
Location: SA7K
Callsign: LV-EPM
IRC name: sa7k
Version: git
OS: debian

Re: 3.0.0 MacOS first impressions

Postby Johan G » Sun Jan 19, 2014 1:16 pm

sa7k wrote in Sun Jan 19, 2014 12:57 pm:...this gets fixed as all people updates to the new scenery.

..which will not happen until all users run FlightGear 2.8 or 2.10 and later as most if not all tiles with an apt.dat version 8.50 or later airport layout (curved edges etc.) will crash older versions of FlightGear. :|
Low-level flying — It's all fun and games till someone looses an engine. (Paraphrased from a YouTube video)
Improving the Dassault Mirage F1 (Wiki, Forum, GitLab. Work in slow progress)
Johan G
Moderator
 
Posts: 6243
Joined: Fri Aug 06, 2010 5:33 pm
Location: Sweden
Callsign: SE-JG
IRC name: Johan_G
Version: 2020.3.4
OS: Windows 10, 64 bit

Re: 3.0.0 MacOS first impressions

Postby someguy » Sun Jan 19, 2014 3:57 pm

Thorsten wrote in Sun Jan 19, 2014 7:31 am:They're not available at all incarnations of the water shader / all quality levels. So that's very unspecific.


My water shader is always maxed, since it has almost no effect on frame rate.

(it displays downloading scenery, but no data rate, and system monitor shows no incoming data stream), and if the connection is not too good, this takes a bit.


It shows the normal data rate (300-325) for my DSL connection. iStat Pro (a monitor widget) confirms the data rate.
User avatar
someguy
 
Posts: 1650
Joined: Tue Nov 25, 2008 5:54 am
Location: USA
Version: 2019.1.1
OS: Mac OS X 10.11.6

Re: 3.0.0 MacOS first impressions

Postby someguy » Sun Jan 19, 2014 4:44 pm

zakalawe wrote in Sun Jan 19, 2014 12:51 pm:]It should rebuild on the first start with the new version (3.0), please be more specific about 'often'. Remember if you change scenery paths it has to rebuild too.]


Hmm. I just tried spawning twice at the last place I had the problem last night. First time, it showed "downloading" for about a minute, then spawned. I quit and relaunched, and there was no downloading delay at all. I guess I'll just have to be more patient, and let it redownload all my scenery.

BTW, the new scenery download panel's Log window shows downloads, but the viewport is too narrow and the paths are truncated.

Here's an info-level log of a spawn to a location visited last week, but not yet in 3.0:

Code: Select all
Enabling ATI viewport hack
Starting automatic scenery download/synchronization. Using built-in SVN support. Directory: '/Users/[me]/Library/Application Support/FlightGear/TerraSync'.
2014-01-19 10:38:04.045 fgfs[799:d07] Error loading /Library/ScriptingAdditions/Adobe Unit Types.osax/Contents/MacOS/Adobe Unit Types:  dlopen(/Library/ScriptingAdditions/Adobe Unit Types.osax/Contents/MacOS/Adobe Unit Types, 262): no suitable image found.  Did find:
   /Library/ScriptingAdditions/Adobe Unit Types.osax/Contents/MacOS/Adobe Unit Types: no matching architecture in universal wrapper
fgfs: OpenScripting.framework - scripting addition "/Library/ScriptingAdditions/Adobe Unit Types.osax" declares no loadable handlers.
Multiplayer mode active!
environment init
Nasal runtime error: non-numeric string in numeric context
  at /Applications/FlightGear3.app/Contents/Resources/data/Nasal/view.nas, line 671
  called from: /Applications/FlightGear3.app/Contents/Resources/data/Nasal/view.nas, line 707
  called from: /Applications/FlightGear3.app/Contents/Resources/data/Nasal/view.nas, line 684
  called from: /Applications/FlightGear3.app/Contents/Resources/data/Nasal/view.nas, line 732
Loading local weather routines...
Initializing electrical system...
...Done.
Nasal runtime error: No such member: views
  at /Applications/FlightGear3.app/Contents/Resources/data/Nasal/view.nas, line 215
  called from: /Applications/FlightGear3.app/Contents/Resources/data/Nasal/view.nas, line 774
  called from: /Applications/FlightGear3.app/Contents/Resources/data/Nasal/globals.nas, line 105
Inititialisation DA42 ...done
zkv1000 loaded
failed to load sound buffer: Warning: STEREO files are not supported for 3D audio effects: /Applications/FlightGear3.app/Contents/Resources/data/Aircraft/Diamond-Da42/Sounds/gear.wav


The "downloading scenery" part took about a minute, but KINL is pretty barren. The first spawn at KSFO took much, much longer.
User avatar
someguy
 
Posts: 1650
Joined: Tue Nov 25, 2008 5:54 am
Location: USA
Version: 2019.1.1
OS: Mac OS X 10.11.6

Re: 3.0.0 MacOS first impressions

Postby someguy » Mon Jan 20, 2014 6:01 pm

Just for grins, I tried enabling Rembrandt today, with MP and AI models disabled. To my surprise, if I reduce the shadow map size to 1024, I can get frame spacing of 180-220, or 4-5 fps. That's enough (barely) to fly the C172P at night and see how cool the runway looks when illuminated by the landing light on approach! This was the first time I'd seen Rembrandt run on my iMac at a near-usable speed, and it made me smile. Someone is doing good work on the code, and I hope they continue to whittle away at it so Mac users can enjoy the experience, too.

I played with the shader settings, and found that enabling the Model shader (there are only two positions, on and off) runs the frame spacing up to 2500-3000, obviously not flyable. On top of that, disabling the Model shader doesn't take effect until I reboot the iMac. Restarting FG is not sufficient. FG/Rembrandt apparently does something deep inside the Mac's video system and doesn't reset it properly, not even on exit.

I'm still having poor luck getting METAR to load completely without a manual poke of the Apply button. It does work occasionally.

Nothing crashed, BTW, in over an hour of experimentation. Kudos to the Stability Fairy! :)

Let me be clear, here: I'm reporting my experiences with the initial RC, not whining. I'm delighted with its stability and cosmetics. There are apparently some bugs, which is to be expected, but I'm finding workarounds that make this first release a real improvement over 2.12, even as-is. I can't see any reason to keep 2.12 around...but I think I'd better not trash it until I've done another week of testing.
User avatar
someguy
 
Posts: 1650
Joined: Tue Nov 25, 2008 5:54 am
Location: USA
Version: 2019.1.1
OS: Mac OS X 10.11.6

Re: 3.0.0 MacOS first impressions

Postby someguy » Sun Jan 26, 2014 9:44 pm

Just had a panic in fgfs. Here's the report, if anyone cares:

Code: Select all
Sun Jan 26 15:37:36 2014
panic(cpu 1 caller 0xffffff800ca54c4d): "a freed zone element has been modified: expected 0xffff3f8037f76fa0 but found 0xffff008037f76fa0, bits changed 0x3f0000000000, at offset 0 of 160 in zone: ipc ports"@/SourceCache/xnu/xnu-2422.1.72/osfmk/kern/zalloc.c:461
Backtrace (CPU 1), Frame : Return Address
0xffffff816bd63410 : 0xffffff800ca22f69
0xffffff816bd63490 : 0xffffff800ca54c4d
0xffffff816bd634d0 : 0xffffff800ca5242f
0xffffff816bd63530 : 0xffffff800ca51d09
0xffffff816bd63630 : 0xffffff800ca196b9
0xffffff816bd63650 : 0xffffff800ca26fdf
0xffffff816bd636a0 : 0xffffff800ca5d14e
0xffffff816bd63730 : 0xffffff800ca23ee6
0xffffff816bd63ee0 : 0xffffff800ca24060
0xffffff816bd63f20 : 0xffffff800cadc671
0xffffff816bd63fb0 : 0xffffff800caf350b

BSD process name corresponding to current thread: fgfs

Mac OS version:
13B42

Kernel version:
Darwin Kernel Version 13.0.0: Thu Sep 19 22:22:27 PDT 2013; root:xnu-2422.1.72~6/RELEASE_X86_64
...
Kernel slide:     0x000000000c800000
Kernel text base: 0xffffff800ca00000
System model name: iMac11,2 (Mac-F2238AC8)

System uptime in nanoseconds: 115805580165005
last loaded kext at 83950721438305: com.apple.driver.AppleUSBCDC   4.2.1b2 (addr 0xffffff7f8eb7b000, size 20480)
last unloaded kext at 84079739719631: com.apple.driver.AppleUSBCDC   4.2.1b2 (addr 0xffffff7f8eb7b000, size 16384)
loaded kexts:
com.uModel: iMac11,2, BootROM IM112.0057.B01, 2 processors, Intel Core i3, 3.06 GHz, 12 GB, SMC 1.64f5
Graphics: ATI Radeon HD 4670, ATI Radeon HD 4670, PCIe, 256 MB
Memory Module: BANK 0/DIMM0, 4 GB, DDR3, 1333 MHz, 0x01BA, 0x202020202020202020202020202020202020
Memory Module: BANK 1/DIMM0, 4 GB, DDR3, 1333 MHz, 0x01BA, 0x202020202020202020202020202020202020
Memory Module: BANK 0/DIMM1, 2 GB, DDR3, 1333 MHz, 0x80AD, 0x484D54313235533654465238432D48392020
Memory Module: BANK 1/DIMM1, 2 GB, DDR3, 1333 MHz, 0x80AD, 0x484D54313235533654465238432D48392020
AirPort: spairport_wireless_card_type_airport_extreme (0x168C, 0x8F), Atheros 9280: 4.0.74.0-P2P
Bluetooth: Version 4.2.0f6 12982, 3 services, 15 devices, 1 incoming serial ports
Network Service: Built-in Ethernet, Ethernet, en0
Network Service: AirPort, AirPort, en1
Serial ATA Device: ST3500418AS, 500.11 GB
Serial ATA Device: OPTIARC DVD RW AD-5680H
USB Device: Hub
USB Device: USB-PS/2 Optical Mouse
USB Device: USB2.0 Hub
USB Device: T.16000M
USB Device: MT1806
USB Device: BRCM2046 Hub
USB Device: Bluetooth USB Host Controller
USB Device: Internal Memory Card Reader
USB Device: Hub
USB Device: iMate, USB To ADB Adaptor
USB Device: Built-in iSight
USB Device: IR Receiver
FireWire Device: unknown_device, Unknown
Thunderbolt Bus:
User avatar
someguy
 
Posts: 1650
Joined: Tue Nov 25, 2008 5:54 am
Location: USA
Version: 2019.1.1
OS: Mac OS X 10.11.6

Re: 3.0.0 MacOS first impressions

Postby zakalawe » Mon Jan 27, 2014 1:08 pm

someguy wrote in Sun Jan 26, 2014 9:44 pm:Just had a panic in fgfs. Here's the report, if anyone cares:

Unfortunately these are basically impossible to action for anyone outside Apple. I think they get submitted back to the mother-ship automatically on the next boot, I certainly hope so. Even if the cause is ultimately us doing something wrong in FG, there is just no way to correlate what that might be, with these reports.

(And yes, it's frustrating for all concerned)

If you start seeing these consistently with 3.0 where a previous version was fine, we might have a big problem, but there's still no way to know what the culprit might be. Of course if you're using a feature you didn't previously use, such as Rembrandt in particular, then it's probably a pre-existing issue.
zakalawe
 
Posts: 1257
Joined: Sat Jul 19, 2008 4:48 pm
Location: Edinburgh, Scotland
Callsign: G-ZKLW
Version: next
OS: Mac

Re: 3.0.0 MacOS first impressions

Postby someguy » Fri Jan 31, 2014 12:45 am

zakalawe wrote in Mon Jan 27, 2014 1:08 pm:If you start seeing these consistently with 3.0 where a previous version was fine, we might have a big problem...


It never happened before, in any version. I've also gotten a few scary crashes that began with loud static noises, followed in a few seconds by a black screen and extremely loud 3-tone alarms that could only be stopped by pulling the power cord. That started with 2.12 and continued through two builds of 3.0.0. I can't reproduce this crash, it seems completely random. I thought it might be related to a recent RAM upgrade (now 12 GB total), but I ran memtest for hours with zero errors. There are no console messages.
User avatar
someguy
 
Posts: 1650
Joined: Tue Nov 25, 2008 5:54 am
Location: USA
Version: 2019.1.1
OS: Mac OS X 10.11.6

Re: 3.0.0 MacOS first impressions

Postby Hooray » Fri Jan 31, 2014 12:47 am

please enable logging and redirect everything to a file, then try to reproduce the issue and post the last 100-200 lines of the log file - you may need to enable debug/bulk debugging if the info is not conclusive
Please don't send support requests by PM, instead post your questions on the forum so that all users can contribute and benefit
Thanks & all the best,
Hooray
Help write next month's newsletter !
pui2canvas | MapStructure | Canvas Development | Programming resources
Hooray
 
Posts: 12173
Joined: Tue Mar 25, 2008 8:40 am

Re: 3.0.0 MacOS first impressions

Postby F-JJTH » Fri Jan 31, 2014 1:33 am

someguy wrote in Fri Jan 31, 2014 12:45 am:I thought it might be related to a recent RAM upgrade (now 12 GB total), but I ran memtest for hours with zero errors. There are no console messages.

Last year I did RAM upgrade (4Go => 12Go), one of the RAM was wrong and gave me similar symptome than the one you are describing.

I invit you to remove your recent new RAM and re-test, then add your RAM one by one (once per day for example) that way you will know which one is bad (if there is a bad RAM of course)

Regards,
Clément
User avatar
F-JJTH
 
Posts: 696
Joined: Fri Sep 09, 2011 11:02 am

Re: 3.0.0 MacOS first impressions

Postby someguy » Fri Jan 31, 2014 5:36 am

Thanks, Clément, but it's only happened four times in a month at irregular intervals. Something I noticed tonight may offer insight: I spawned a C172P on the AI Nimitz (doesn't everyone do that?) and made two take-offs and landings. Then I took off once more and flew to San Rafael, the nearest airport. Just after spawning, I was using around 3 GB of RAM out of my 12 total; when I landed, I checked my RAM usage again, and only 605 MB was free. The only other app running was Firefox, to display mpmap. A massive memory leak?
User avatar
someguy
 
Posts: 1650
Joined: Tue Nov 25, 2008 5:54 am
Location: USA
Version: 2019.1.1
OS: Mac OS X 10.11.6

Re: 3.0.0 MacOS first impressions

Postby zakalawe » Fri Jan 31, 2014 10:55 am

That memory use certainly seems a little high - are you using any settings which are particularly memory intensive (random trees with high density, 3D clouds with large draw distance *and* high density). Except you mention flying from the carrier, you sure don't get many trees in the ocean.

An application exhausting system memory should not cause a kernel/driver panic of course, I'd expect OS-X to warning you about low-memory before reaching that point.
zakalawe
 
Posts: 1257
Joined: Sat Jul 19, 2008 4:48 pm
Location: Edinburgh, Scotland
Callsign: G-ZKLW
Version: next
OS: Mac

Next

Return to 3.0

Who is online

Users browsing this forum: No registered users and 0 guests