Contribute
Register

**Possible** freezing issues with 10.11.2, Z97X UD5H and 280X

Status
Not open for further replies.
Ya it's an odd one to be sure, and you and I have different 280x cards as well with what sounds like the same issue. Mine is an ASUS Direct CU II card and I'm starting to wonder if the factory overclock may have something to do with it, but I am just too damn lazy to put the card in my PC to reflash/down clock it. lol

I am running my 280X cards absolutely stock and as they came out of the box. no refreshing, no over clocking. I don't think thats the issue here, I think its a dodgy driver in the new upgrades.

However by all means try a reflash and report back how it goes. The more information the better.

Thanks,

Rob
 
Hi, I will start building my first ever Hackintosh, based on some post, got almost all the parts together (just waiting for a SSD Samsung evo and a Seagate Barracda to come in a couple of days)

So far my parts are:

Gigabyte Z97X-UD5H Motherboard
Intel i7 4790K Processor
Crucial Ballistix Tactical 16 GB Ram
Sapphire Vapor-X R9 280X Graphic Card
Corsair CS750M Power Supply
Corsair 330R Case

As I realized you have a similar build, just wanted to ask (still can't send p messages) am I on the right track? should I be aware of specific difficulties and if so, eventual fixes? Any commentary or suggestion will be greatly appreciated. Many thanks

Rob

I'd rather you didn't hijack this thread as it is specific for the freezing issue.

However I'll reply quickly, since you have much the same parts as mine then it should be fine, the freezing issues noted with 10.11.2 and 10.11.3 notwithstanding. Check the memory again and again and again. Check every single letter and digit of the memory SKU (and they are loooong) and check again and again and again that it is explicitly listed for this exact motherboard.

Any other questions please raise in a separate thread as I want to keep the focus of this thread clear.

Rob
 
Well I gave up, I found a deal on a 750ti and put it in...I had to use SwitchresX to get full resolution on my UltraWide LG display but this machine is now solid and sleep works perfectly! Same reason I stopped using the AMD card on my PC, driver support is just not there with AMD. Oh well, the 280x can be wall art or something I guess.
 
Well I gave up, I found a deal on a 750ti and put it in...I had to use SwitchresX to get full resolution on my UltraWide LG display but this machine is now solid and sleep works perfectly! Same reason I stopped using the AMD card on my PC, driver support is just not there with AMD. Oh well, the 280x can be wall art or something I guess.

That is one solution but sadly for anybody using FCPX, its not a good one as the 750ti is nowhere near as fast. I suspect it will get fixed in 10.11.[4/5/6]. Until then I'll stay with 10.11.1 as it works well.

I have to say though that Apples QA is getting steadily worse, I have fond memories of Snow Leopard and its stability (though that may be through rose tinted glasses).

Rob
 
Agreed on the apple QA, it's sad to say but it seemed to start a downward trend shortly after a certain CEO passed away. Most of my work is light photoshop and illustrator so the 750ti seems to hold up just fine for my purposes, the only reason I used the 280x in the first place was it was surplus from when I upgraded my gaming PC. I will keep an eye on the thread though as I'm curious if the drivers do improve.
 
FWIW: My issue seems to have been caused by heavy SMB usage. I've been systematically removing hardware/software to narrow it down.

I had been using CrashPlan to backup a Drobo5N Share via symbolic link & persistent mount (smbfs). I was also using the Drobo5N as a Crashplan target (meaning storing backup data there) also with a persistent mount. The process completed frequently, and errors didn't appear in any logs at crash time. However, immediately after stopping that process my crashes have disappeared. Crashplan 4.5.2 is still running on OSX and backing up the local machine without issue.

I've gone 6 days of otherwise heavy system usage without an OSX restart.
 
FWIW: My issue seems to have been caused by heavy SMB usage. I've been systematically removing hardware/software to narrow it down.

I had been using CrashPlan to backup a Drobo5N Share via symbolic link & persistent mount (smbfs). I was also using the Drobo5N as a Crashplan target (meaning storing backup data there) also with a persistent mount. The process completed frequently, and errors didn't appear in any logs at crash time. However, immediately after stopping that process my crashes have disappeared. Crashplan 4.5.2 is still running on OSX and backing up the local machine without issue.

I've gone 6 days of otherwise heavy system usage without an OSX restart.

Your problem and solution is sadly unrelated to this thread. You have / had an issue with SMB performance. Apples networking code has been going downhill for years They've dumped their own networking protocol and now use SMB. Unfortunately Apple can't make SMB work reliably, that and the fact that many NAS vendors have junk network implementations as well means many products simply do not work with Apple os's. I have given up on SMB and just use NFS and.very careful with when I use it.

Anyway pleased you have sorted your problem out but the 280x issue still exists.

Rob
 
Did another 'upgrade' to 10.11.3 yesterday after removing Keyboard Maestro and still had lockups.

I have noticed a few lockups with 10.11.1 now.

Its very odd, if I use Remote Desktop to connect back to my Hack from my Macbook it immediately clears the spinning beachball. Don't even have to click on Remote Desktop, the act of connecting clears the beachball.

Very odd. The investigation continues.

Rob
 
Would you mind popping over to that thread and posting your system there and the fact that you are NOT having issues with 10.11.3. I'm trying to keep this thread for video stuff (for which your post here is absolutely fine and thanks for posting) and a separate thread for issues. It would be great if we could get more information about what you have and see if there is some reason your system is fine and other peoples fails.

After testing, it isn't stable if there are two displays connected. It doesn't matter if I plug displays to different graphics cards or both displays in same graphics card.

It's stable with only one display and dual graphics cards.

I've also tested to enabling IGP, works somewhat, doesn't freeze, but performance degrades with FCPX (BruceX: 2x280x = 17sec, IGP+2x280x = 26sec.), but in Luxmark it gains more points.. So useless for me.

I just ordered USB3-to-DVI display adapter from Amazon, as I don't need another monitor for anything but email, calendar, FCPX bin, etc..
 
After testing, it isn't stable if there are two displays connected. It doesn't matter if I plug displays to different graphics cards or both displays in same graphics card.

It's stable with only one display and dual graphics cards.

I've also tested to enabling IGP, works somewhat, doesn't freeze, but performance degrades with FCPX (BruceX: 2x280x = 17sec, IGP+2x280x = 26sec.), but in Luxmark it gains more points.. So useless for me.

I just ordered USB3-to-DVI display adapter from Amazon, as I don't need another monitor for anything but email, calendar, FCPX bin, etc..

Ok, you appear to have come to the conclusion I was (very) slowly thinking about, that multiple displays are unstable in 10.11.3. I'd seen some other threads with hints of this, the common factor being multiple displays.

The reason for the slowdown in FCPX is that FCPX seems to favour the onboard graphics over the external graphics cards. Even so, 26 secs is quite respectable :) but I understand what you mean.

Did you try active display port adaptors at all? I have passive ones but until now have never needed active adaptors. I may have a look on Amazon and see how much they are now. I can't see them working but it might be worth a shout.

Please let us know how the USB3-DVI adaptor works.

Thanks,

Rob
 
Status
Not open for further replies.
Back
Top