Jump to content

jaymonkey

Members
  • Posts

    15
  • Joined

  • Last visited

Everything posted by jaymonkey

  1. @InsanelyDeepak Many thanks the clear and concise guide on resolving issues with combo jacks, I think this is about the clearest guide i've seen on the subject. https://osxlatitude.com/forums/topic/11316-how-to-fix-static-noisedistortioncrackling-sound-and-combo-jack-on-laptops/ One question :- Towards the end of the guide you say the following :- Could you please provide an example of sending commands using AppleALC rather than using hda-verb + codec commander. There is no mention of the feature in the AppleALC documentation, and i can't find any examples on-line. Many users now use Lilu + AppleALC so being able to resolve combo jack issues without having to install HDA-Verb + Codec Commander would be fantastic. Thanks again Cheers Jay
  2. @EMlyDinEsH All is working great, I do not believe that the problem was with your firmware uploader kext. As has become apparent, Skvo's patch was the missing piece of the puzzle - for some people at least. If anybody wants a ioreg dump please let me know . . . I think i have figured out what has been going on .. Without Skvo's patch, the total number of available channels for the radios on the card was very much reduced. As we know there was no 5Ghz channels and just a few 2.4Ghz channels. Since Bluetooth operates in the 2.4Ghz band it has to share the available channels in the 2.4Ghz radio spectrum with the wifi section of the card. As we all know 2.4Ghz wifi is very widespread and its very common for nearby routers and bluetooth devices to 'compete' for channels sometimes reducing the available channels even more resulting in reduced channel bandwidth . I believe that the poor wifi transmission speed and intermittent bluetooth performance was all being caused by the card being crippled and not able to use the full bandwidth triggering some sub channel throttling to occur .... I think that this is a low level feature of the cards chipset/radio and the different BT firmware versions show slightly different symptoms/behaviours in OSX. I am now using FW=0, so now running the latest firmware which is what I think the latest OSX drivers/kext's are optimised for. Everything is working fast and stable. Am streaming 1080P from my Synology NAS over 5Ghz @ 300Mbs watching a movie with bluetooth headphones with my Apple Magic Touch Mouse, Apple MagicTouch Pad and my iphone all connected via BT 4 - and all working without any issues or delays, I don't think i've ever experienced such amazing STABLE wireless performance in any OS till now ..... at least in my house ... upgrading to this card has been money well spent. Update: I created a step by step guide for installing this card here. Cheers Jay
  3. To all following this thread, I took the plunge and updated my Sony SE2 to 19.9.2, it took a little bit more work than the usual combo updates but once 10.9.2 was installed, WiFi and bluetooth continued to function as per my last post. I have now applied Skvo's patch and can confirm that WiFi is now working correctly (both 5Ghz and 2.4Ghz) and I am getting full dual channel link speeds. As of now my Card is 100%, BT4.0 and full WiFi - very happy and pleased with this card now. A big thank you to Toledia, EMlyDinEsH and Skyo for the final solution, it never ceases to amaze me how determined and dedicated the hackongtosh community is, solving this issue has been possible by the effort of everyone involved and goes to show the power of an open development community, things actually get fixed ..... To Summarise my final solution :- PID=0x3403 VID=0x13d3 - BCM94352 - Azureware AW-CE123H Install following kext's:- toledaARPT.kext BCM4352.kext (wifi & BT version) BTFirmwareUploader.kext (FW=0) Apply Skvo's patch, rebuild system caches reboot - Done Update: I created a step by step guide for installing this card, see here. Cheers Jay
  4. @EMlyDinEsH - Further Info After a bit more experimentation I found that by using both kext's in the above configuration, it is possible to get a working solution of sorts. Install both BCM4352.kext and V2.3 of BTFirmwareUploader.kext (FW=3) On initial cold boot BTFirmwareUpdater will fail to upload BT firmware. After cold boot disable bluetooth using icon menu, right click. Put system to sleep. On wake BTBluetoothFirmwareUploader seems to work correctly and uploads the correct firmware. After this both bluetooth 4.0 and Wifi are working correctly, until next power down Not perfect but so very close . . . Let me know if there is anything else I can test/debug for you. Cheers Jay Update: sometimes it works ok on a cold boot (both kext appear to run ok during osx load), I have not changed anything since the above update but now sometimes BT4 and WiFI seem to be working perfect cold booting OSX using both your kext and BCM4352. I have attached a new archive with ioreg and system_profile. Not sure whats going on and why sometimes it works now, perhaps the kext's are loading at different times or different order ? It does seem to be a a bit hit and miss though. But if Wifi starts playing up then then disabling BT, putting system to sleep and then waking from sleep and re-enabling BT seems to restore BT and wifi functionality. Will report any further strangeness ... If only there was a solution to enable 5Ghz Dual stream .... Perhaps the just released 9.2 Update will help found this in the list of changes:- Improves ExpressCard compatibility for some MacBook Pro 2010 models Full list of updates here :- http://support.apple.com/kb/HT6114 Might have a go at installing tomorrow and report back Cheers Jay Cold Boot OSX - Both Kext-FW=3 - Now Working.zip
  5. Hi EMlyDinEsH, Many thanks for your reply, I downloaded and installed version 2.3 of your bluetooth firmware uploader kext and now cold booting into OSX, Bluetooth 4.0 does seem to perform much better than the previous version, however it seems to have introduced a new issue:- Initial connection to the WiFi network is painfully slow sometimes almost 2 mins !!, same happens when waking from sleep, BT connects almost instantly but wifi takes an age, when it does connect the transmission rate is very slow between 15-30 with BTFirmwareUploader removed from the system WiFi connects within 10 seconds of boot and connects at a transmission rate between 130-145. I use WiFiSpy which displays the wifi transmission rate in the menu bar, very useful little app when working with these issues, or you can hold the option key down while clicking on the wifi icon to see extra connection info. As reported in my last post, this behaviour was also apparent with version 2.1 of your kext but the issue seems to be worse with version 2.3 Has anybody else seen this issue ?, As discussed in my last post I am using toledaARPT.kext injector to enable wifi on the BCM4352 IOBluetoothFamily.kext is vanilla and bluetooth transport control seems to be enabled with both toledia's and your injector's correctly. The issue with the wifi only occurs when using BTfirmwareuploader, removing your kext solves the wifi issue but obviously stops bluetooth from working. As requested I have attached a zip file with system profile and ioreg dumps for each of the following cold boot conditions. 1. Cold Boot Win 8.1 - BCM4352 - then soft reboot back into OSX Mavericks 9.1 - (BT FW=V14.5452) - in this mode I am using Toleda's BCM4352.kext which apparently 'patches' the BT firmware. This combo boot gives the best BT & WiFI Performance, Wifi connects within 10secs after cold/soft boot/reboot, BT seems to behave exceptionally well - but its a pain in the arse when i forget to first boot windows. 2. Cold Boot OSX - BCM4352 - (BT FW=V14.C4096) - this is exactly the same OSX configuration as above but without first booting into windows 8.1, Wifi works just the same but BT does not work very well at all. 3. Cold Boot OSX - BTFirmwareUploader - (BT FW=V14.5575)- This is my latest OSX configuration, i removed Toleda's BCM4352.kext and installed version 2.3 of your kext, in this configuration Bluetooth 4.0 seems to working well but does not 'feel' as stable as condition 1 (win8.1) and has introduced the wifi issues explained above. All of the above conditions are completely repeatable every time, it is not an intermittent problem. I am thinking that this might be some sort of (low) power saving mode that the card is being put into ? , as all three conditions all use the same wifi enabler, but BT and wifi performance seem to be affected with the introduction of your kext .... When using your Bluetooth kext, Disabling Bluetooth via the Icon->context menu, resolves the issue with WiFi, connections are almost instantaneous on a wake after sleep and transmit rate is always 130 + I hope there is enough debug info in the attached zip file, please let me know if you require anything else. Appreciate all your hard work, I was beginning to regret buying the BCM4352 but your work has given me hope that there is a solution very close. Cheers Jay Update: Ok i went through my ioreg dumps and the worked through the different FW versions that your injector uses (0-3) PID=0x3403 VID=0x13d3 - BCM94352 - Azureware AW-CE123H Results using Toleda BCM4352.kext only Cold Booting Win 8.1 then soft reboot OSX - Reported firmware version is V14.C5452 - BT4.0 and Wifi work perfect - Cold Booting OSX - reported firmware version is V14.C4096 - BT4.0 does not work, WiFi work perfect Results using BTFirmwareUploader V2.3 only FW=0: V14.C5575, BT=Yes, WiFi=Slow Connect & Transmit Speed FW=1: V14.C4096, BT=No, WiFi=Perfect (same as when using BCM4352.kext) FW=2: V14.C5271, BT=Yes, WiFi=Slow Connect & Transmit Speed FW=3: V14.C5452, BT=Yes, WiFi=Slow Connect & Transmit Speed Not sure what this is telling us except that when FW=1 with BTfirmwareloader then result is the same as cold booting OSX with BCM4352.kext, wifi works BT does not, all other FW versions seem to interfere with the WiFi connection ??? Obviously some thing is different when using your kext, or is BCM4352.kext introducing some other setting which is not set by your kext ? Conclusion You would think that from the above experimentation using BTFirmwareUploader V2.3 with FW=3 and using Toleda BCM4352.kext should give the same result as booting first with Win 8.1 but i think the two kext interfere with each other at boot time and it would appear that it stops BluetoothFirmwareUploader from installing any firmware as it shows as V14.C4096 which gives the same result as cold booting with just BCM4352.kext, BT Does not work but WiFi is perfect. I have added another set of dump files with FW=3, with and with out BCM4352.kext to the attached archive. The last config (both kext with FW=3) includes console log for the bluetooth devices - is it possible that BCM4352.kext is changing the identity so BTFirmwareloader is no longer injecting the right FW as ID has changed ? I think if these two kext would behave together then i think all the issues would be resolved . . . . . . Hope this helps Cheers Jay MonkeyBook-Pro BT Debug.zip
  6. @EMlyDinEsH Just wanted to say a big thanks for this post, and thought it might be useful to share my experience Like many here, I bought a BCM94352 (Azureware AW-CE123H) Combo BT4.0 / WifI ABGn-ac Mini PCIe. BLUETOOTH PID=0x3404 (13316) VID=0x133d (5075) WIFI PID=0x43B1 (17329) VID=0x14E4 (5348) I've been using it in my Sony Vaio SE2 MonkeyBook Pro, (i7 8GB RAm 128GB-SSD, 15.5" IPS Intel HD3000) Like many I was hoping to use the 5Ghz and BT4 with OSX and like many i found it didn't work too well. My system is quad boot, and after loading drivers for Windows 7 & 8.1 and Ubuntu all of these OS's work perfectly with the BCM94352 - 2.4 & 5.0 GHz Dual band @ 300Mbs ( i don't have ac router so can't comment on that), Bluetooth seems to work fine with no issues with these OS's as well. Under OSX, using Tolida's injector 2.4Ghz wifi works perfect but only single stream (~145Mbs) , As I understand it, there is currently no support in OSX for Broadcom 5Ghz radios so no 5Ghz - didn't find that out until after the event Bluetooth 4.0 under OSX using tolidia's BC4352.ketx injector at first seemed to work but after the first pairing no other devices could be connected correctly no matter what although they would get listed in BT preferences. Investigations .... If I booted into windows 8.1 and paired all my BT devices with windoze and then soft-reboot into OSX then i could connect to all of my bluetooth devices and it all seemed to work fine until a power down, if cold booted into OSX I could no longer connect to my BT devices, but if i booted into windows 8.1 and then soft-reboot (didn't even login) back to OSX my devices would automatically reconnect and all is well with bluetooth working perfect across all devices, flawless as far as i can tell until a power down. Interestingly if i cold booted into OSX first and then soft-rebooted into windows 8.1 then Windoze suffered exactly the same Bluetooth issues we have all be seining in OSX ?. Very strange but it must tell us two things. 1. During the the loading and initialisation of the device under OSX it brakes something, the thing that it brakes must be some sort of volatile memory status bit or firmware that does not survive a power cycle. 2. If cold booted into Win 8.1- the OS correctly initialises or updates the firmware / volatile memory / status bits, which can survive a soft reboot, as OSX initialises it does not bother to initialise the card as it already has the firmware / status-bit or whatever it is set and therefore works perfect in OSX. This procedure is repeatable with always the same out come. And thats how i have been using for the last few weeks it until i came across this post (also linked from other posts and forums) initially i had no better luck with your bluetooth firmware loader, but after checking the console log, it showed that it wasn't being successfully injected, after deleting tolida's BC4352.kext it seemed to load ok and all my BT devices now seem to work, i'm using the 2.1 version i have tried firmware 0 and 2 and both seem to work ok'ish for most of the time when using apple magic touch mouse or trackpad, however if i use my BT headset it disconnects every time. This does not happen when 1st booted with windows 8.1, it just seems to work prefect then. The only other observation i have seen is that wifi seems to take much longer to connect after boot in osx and sometimes does not work at all after sleep, this did not happen when using tolida's BC4352.kext, i realise that the BT4 and wifi are completely separate devices despite being on the same combo card, but something odd is going on.... Could you please explane how to dump the BT firmware / volatile memory in both windows and OSX, i'd like to see if there is any difference between the firmware after running your BT loader or having it initialised by win 8.1, could be a useful pointer in putting this issue to bed once and for all ....... Thanks so much for your work, Cheers Jay
  7. EMlyDinEsH, Its the least I can do given the amount I've learned from using your guides and fixes. Attached archive contains everything (and more) you asked for including Patched DSDT & SSDT Specific CPU Type: Intel i7 2640M @ 2.8Ghz / Max turbo 3.5ghz Cheers Jay Sony_SE.zip
  8. @EMIyDinEsH Spoke to early, i got a bit of a strange issue with your EAPDfix. If I put my system to sleep by closing the lid, and then wake it by opening the lid and pressing a key, your EAPDfix does its job and sorts the speaker sound out (speaker graphics shows on the screen briefly) However, if I let the system automatically go to sleep via the energy saver or if i manually put the system to sleep by clicking on the apple menu and selecting sleep, then your fix does not seem to work ? although checking the system log shows that it did in fact run with no errors:- Nov 3 03:33:38 jaysmonkeybookpro kernel[0]: EAPD: Sending Command 0x1470c02 to Speaker Node#20 Nov 3 03:33:38 jaysmonkeybookpro kernel[0]: EAPD: Sending Command 0x1570c02 to Headphone Node#21 Nov 3 03:33:38 jaysmonkeybookpro kernel[0]: EAPD: Command(s) Sent Successfully No speaker graphic appears on the screen and the system has no speaker sounds (just like before your EAPDfix was installed), however if i then close the lid and let it go to sleep and then wake it, your EAPDfix works again ? Wried ehh ? i'm running OSX 10.9.0 - Mavericks - As long as EAPD is working All sound I/O works perfect inc. HDMI Audio. Any ideas or suggestions ?, i've attached a zip with my DSDT, Patched AppleHDA, ioreg & codec dump. Cheers Jay PS: Many thanks for your patching guide on AppleIntelSNBGraphicsFB, have now got HDMI audio working so very happy ​Update 3-Nov 2013: I've managed to fix this behaviour by adding the kernel flag 'darkwake=0', all seems perfect now. Sony_SE.zip
  9. Hi EMlyDinEsH, I have Sony Vaio SE which uses RealTek ALC275 codec, and was having to using an old patched SL IOAudioFamily.kext to get around the issue of no sound after wake from sleep but you little fix seems to have solved that for me so now I can run native IOAuidioFamily.kext and finally have HDMI Audio. Changes to plist :- <key>EAPDFix</key> <dict> <key>Audio Sense</key> <true/> <key>AudioNodes</key> <dict> <key>Headphones</key> <integer>21</integer> <key>Speaker</key> <integer>20</integer> </dict> This should work for Sony SA & SB that use the same codec, if anybody needs it I've attached my pre-configured version. Great Work, thanks again. Cheers Jay EAPDFix-Sony-ALC275.zip
  10. I got it working, I said it was close ..... I just worked through all the edits again in the end it was one silly little mistake in the info.plist for AppleHDAHardwareConfigDriver.kext :- <key>CodecID</key> <integer>283902581</integer> <key>ConfigData</key> <data> ASccAAEnHQABJx6gAScfkAFHHBABRx0BAUce FwFHH5ABVxwgAVcdEAFXHiEBVx8DAUcMAg== </data> <key>FuncGroup</key> <integer>1</integer> <key>LayoutID</key> <integer>275</integer> The layout ID is 275 !! , should be 28 to match the 1C id in the DSDT .. etc We both missed that one but all working now..... Jay
  11. I tried adding the additional EAPD verb data to the config data section of the plist in AppleHDAHardwareConfigDriver.kext as per EMlyDinEsH's instructions (steps 7-8) at the start of this thread. Question: My verb data is in hex blocks just like the example but the config data in the AppleHDA is in alphanumerics in one big block is there some sort of compile necessary ? Anyway after adding the new verb data I tried AppleHDA but there is still no sound from the speakers, headphones work great as does internal MIC so the CODEC is working, also speakers are listed in the audio section of system Information as default output. I think the additional EAPD verb data also needs to be added to the layout and platform xml files but i'm not sure how to go about this. darkera13 - Thank you very much for helping me get this far can you think of anything else to try ? ----------- EMlyDinEsH - I know you must get a lot of requests for help but this is so close now, the attached archive contains the codec dump, verb info, my DSDT and the last version of AppleHDA (OSX10.8.5 S1) that darkera13 patched for me, but i don't think the EAPD is correct or missing as no sound from speakers, output switches to headphones and works fine but no sound from speakers. Any advice or help would be very much appreciated. Cheers Jay ALC275 HDA.zip
  12. Ok, i tried the new AppleHDA but still no sound from the speakers, internal mic & headphones all seem to work ok. So i went back and took a long look at my source data and re-read EMlyDinEsH's guide at the start of the thread and i think i found something, as i explaned i used a script to process the codec_dump and produce the verbs.txt file, i went through the script to understand how it uses its rules to workout the verb data and guess what, it does not have a rule for EAPD. After looking at the the guide again it says that if you have a EAPD on node 14 (speaker) then you must calculate separate verb data for the EAPD otherwise no sound through the speakers ? Do you think this could be the problem ?, i've attached the new verb.txt with the additional EAPD verb value Cheers Jay verbs.txt
  13. Thank you for your post, its almost there, The audio only works on headphones not speakers, the on screen volume changes and speaker icon are all working as if the speakers are working but there is no sound from the speakers, as i said headphones work great and the internal mic gain is bang on now, it must be something to do with the speakers gate /pin switching ? The audio through the headphones is working great now, perfect, no looping and is very clear with no noise which is a big improvement on the SL kexts i was using, it also seems to work fine after sleep and wake up .. etc., Thank you so much for your help, this is so close now, just the speakers, is there anything i can do to gather any more info to debug this ? Cheers Jay
  14. Hi there, Many thanks for taking a look at my sound issues and having a go at trying to get it to work, I've just finished trying out the DSDT and AppleHDA with the following results. 1. Once OSX is booted, there is the speaker icon in the top menu bar and i can adjust the volume level with the appropriate fn keys and the on screen display shows the volume level going up and down, however there is no sound. playing a mp3 or a youtube stream also results in no sound. 2. On the sound preference plane the output device is correctly identified as internal speakers with no other options, if i plug in headphones the output device changes from internal speakers to headphones correctly and back to speakers when i unplug them. 3. Playing a sound with headphones plugged in (volume change sound) does indeed make the correct sound through the headphones however it continuously loops the last ~1 second of sound until i unplug the headphones. 4. HDMI audio is not working, no HDMI device in sound pref, HDMI video is working ok but it was before. 5. The internal microphone seems to be working ok. This is a major improvement on my last attempt which resulted in nothing working at all, is there anything that can be done ?, I've attached a i/O reg dump file. Cheers and thank you for your assistance Jay Jay.ioreg.zip
  15. Hi EMlyDinEsH, Thanks for posting your AppleHDA patching guides, My codec uses ALC275 not ALC260, I dumped my codec info from Ubuntu and tried to follow this guide but i just cant get things to work, I recently created an all new DSDT and got all the patches for audio in there and its working but with a very old SL AppleHDA & modified IOAudioFamily which is not nice. I am trying to get back to current revisions of OSX kexts (10.8.5-S1) which is where it all falls apart as i don't seem to be able to take the info in your guide and apply it to AppleHDA. I tried taking the verb info from the old patched (for ALC275) SL AppleHDA but i think there must be a big difference with the way things are patched in 10.8.x I'm getting stuck when creating the layouts in AppleHDA and the layout.xml as i don't have layout.xml in the SL version so I cant use it as a reference . . and i just cant seem to make sense of it. I could really do with some help, the attached archive contains the codec dump from linux and modified verbs using Signal64's script (takes codec-dump text file and converts it to modified verbs) along with my DSDT, I have also included the old (but working) SL kext's as reference, and untouched AppleHDA (10.8.5-S1) I'd appreciate any help or advice you can give. Many Thanks Jay Archive.zip AppleHDA.kext.zip
×
×
  • Create New...