OS X compatible motherboard -> QUO

[vegasloki]

 

 

Especially so, because if you are in a professional industry that needs more than 32GBs RAM, more CPU power than an i7-Quad Core, or lots of PCI-e slots (or might need these in the next 3-4 years) then you are shooting yourself in the foot with this motherboard or even with a Haswell socket 1150.  Those are iMac's in disguise.  

 

You should have bought a socket 1366 in 2009-mid 2011, then bought a socket 2011 if purchasing in 2011 Q3 until Q4 2014.

 

 

Expansion chassis have been part of our landscape for about 10 years or so.  It's not about the lanes, it's about fitting that many cards in the system with our I/O counts.  I'm in the audio dept, not the picture dept and deal primarily with music.  A typical large scale Pro Tools HD system I use is based on a octo Mac Pro though some of the Nuendo rigs are i7 950s and they're plenty big.  There is more than just drives with interfacing with Thunderbolt though we use drives like we used tape in the olden days.  It's about portability of interfaces in being able to interface different devices, in our apps particularly MADI and fibre channel.  You're thinking drives only but there are more advantages to Thunderbolt than connecting drives.  An example would be Hammerfall cards in a Thunderbolt enclosure, some sort of drive array and a Mac Book Pro.  Fits in a Pelican case and will be dumped to the SAN via fibre channel when we get back so the guys that mix it or put it to picture can have it or we mount it in one of our rooms if we are finishing it.  Much of our content capture happens remotely and until recently we'd have rigs racked up and ready to ship out  (or send a mobile) but using Thunderbolt as a means to interface what are normally PCIe devices we can have the same capability for many apps in a much smaller foot print and logistically be able to support it better.  We still need a truck for live to air but for many things a good flyaway rig that is compact and gets clean tracks is all we need.  For the studio rigs we'd need a card to interface the Magma chassis so we might as well use TB.  We've been implementing it as needed over the last year and it's been handy (though not cheap).

 

We could make a Pro Tools or Nuendo system out of the Quo board and have it deliver pro results.  But we don't because of several reasons I'm not going to delve into right now but our main obstacle isn't the resource capability of the board so much as no hot swap TB is a deal breaker. Part of it is the support (that we wouldn't get) from using non approved hardware and the grief from Avid for using it.  For us the Mac Pros and nice MPBs are some of the least costly items.

[maleorderbride]

Expansion chassis have been part of our landscape for about 10 years or so.  It's not about the lanes, it's about fitting that many cards in the system with our I/O counts.  I'm in the audio dept, not the picture dept and deal primarily with music.  A typical large scale Pro Tools HD system I use is based on a octo Mac Pro though some of the Nuendo rigs are i7 950s and they're plenty big.  There is more than just drives with interfacing with Thunderbolt though we use drives like we used tape in the olden days.  It's about portability of interfaces in being able to interface different devices, in our apps particularly MADI and fibre channel.  You're thinking drives only but there are more advantages to Thunderbolt than connecting drives.  An example would be Hammerfall cards in a Thunderbolt enclosure, some sort of drive array and a Mac Book Pro.  Fits in a Pelican case and will be dumped to the SAN via fibre channel when we get back so the guys that mix it or put it to picture can have it or we mount it in one of our rooms if we are finishing it.  Much of our content capture happens remotely and until recently we'd have rigs racked up and ready to ship out  (or send a mobile) but using Thunderbolt as a means to interface what are normally PCIe devices we can have the same capability for many apps in a much smaller foot print and logistically be able to support it better.  We still need a truck for live to air but for many things a good flyaway rig that is compact and gets clean tracks is all we need.  For the studio rigs we'd need a card to interface the Magma chassis so we might as well use TB.  We've been implementing it as needed over the last year and it's been handy (though not cheap).

 

We could make a Pro Tools or Nuendo system out of the Quo board and have it deliver pro results.  But we don't because of several reasons I'm not going to delve into right now but our main obstacle isn't the resource capability of the board so much as no hot swap TB is a deal breaker. Part of it is the support (that we wouldn't get) from using non approved hardware and the grief from Avid for using it.  For us the octo Mac Pros and nice MPBs are some of the least costly items.

I think we might be talking past each other, or about different things.

 

Audio guys have it the easiest as PT, CuBase and Logic consistently needing less PCI-e, less RAM, and less CPU than video guys.  If anyone could get by on a Haswell or IB i7-Quad Core then it is you guys, but I don't really see you making that argument here.

 

You indicate that there is some sort of problem with fitting that many cards in an audio build and maintaining "your I/O count." But you can fit that many in.  Just plug them in.  Your latency won't suffer because you have a full house and IRQ issues are so 2006.  You will not have lower performance in any way.  If you want a Fibre card, a 10GbE card, a PT Native card, a quad-monitor GPU, and any other PCI-e card you just plug them in internally simultaneously.  Instead of spending that 1K+ on a Magma chassis you can now put it towards a computer that is not obsolete.

 

Your next example is about using a (granted oldy but goody) 9 year old PCI RME Hammerfall card along with a MBP.  How is that relevant?  We are talking about towers here, not MBP's reliance upon TB.  I completely grant that TB is aimed at MBP, iMac, and Mac Mini users as they have no internal options.  Your example is off target and completely irrelevant to any discussion about a tower's motherboard as you can choose to have PCI-e slots.

 

What could make sense as an advantage is that the modular nature of TB allows you to have fewer fully "built up" computers as you just connect TB devices to whatever you need.  That might be a valid argument, but I would challenge you to run the numbers.  It might take less organization on the part of your IT guy, but it probably costs you considerably more money in the final tally.  TB chassis and Magma chassis are not cheap.  Plus, having a tower with a bunch of cables running off it is not exactly a streamlined workstation.

 

When you do start addressing towers, and studio rigs, you say that you still "need a card to interface the Magma chassis so we might as well use TB."  What?  Why can't you just plug whatever you have in that Magma into the PCI-e lane directly?  Am I missing something here or are you?  

 

You end by saying that no hot swap TB is a deal breaker for a workstation, but you have not yet shown why TB is even needed in an audio workstation.  There is exactly one audio interface that uses TB and it also support USB 3.0 as both supply many times the bandwidth it needs.

 

Oddly enough there is not hot swap TB (or any TB) on either of the two workstations towers that you say that you do use routinely.

 

 

[genzai]

wow, i had decided to stay out of the conversation because it had gotten so silly but now i feel compelled to respond. Clearly you are talking about a "different thing". I'll try to address several points, but please for the love of god don't let your bias for wanting TB to go away affect your logic so much... Even though it was not my post you were responding to i understand completely where he is coming from.

Also I apologize in advance since this has strayed a fair ways off topic.

 

 

Audio guys have it the easiest as PT, CuBase and Logic consistently needing less PCI-e, less RAM, and less CPU than video guys.  If anyone could get by on a Haswell or IB i7-Quad Core then it is you guys, but I don't really see you making that argument here.

 

 

So dead wrong. The fact that you put cubase and logic in the same sentence as PT is a flag right there, but i will elaborate.

Audio post production and high end tracking has ALWAYS been a more demanding process than the video side. In large film mixes, historically be it 40 highspeed dubbers in the back room, magma chassis full of HD DSP cards, or the more modern equivalent using HDX; there is a lot of gear involved. the extreme track count and DSP needs of audio requires very special setups. Track counts can reach into the hundreds, spread across 3+ DAW systems with a dedicated re-recording system to boot. With current versions of PT the entire session gets loaded into ram which has typically been far more demanding on the system then anything our video department machines have to handle which for the most part is ram previews and such. If your working native, system core count and overall power is even more important.

 

You indicate that there is some sort of problem with fitting that many cards in an audio build and maintaining "your I/O count." But you can fit that many in.  Just plug them in.  Your latency won't suffer because you have a full house and IRQ issues are so 2006.  You will not have lower performance in any way.  If you want a Fibre card, a 10GbE card, a PT Native card, a quad-monitor GPU, and any other PCI-e card you just plug them in internally simultaneously.  Instead of spending that 1K+ on a Magma chassis you can now put it towards a computer that is not obsolete.

 

 

In G5/ Mac pro terms it was very easy to exceed the palsy slot count internally. But even with a hypothetical 7 PCIe that isn't enough for many situations. Magma chassis are important to our industry. They produce no measurable latency. In addition to X amount of DSP cards you also need to provide tons of I/O, possibly several network interfaces, and depending on the need any number of other cards. Our hero systems have to have 6 monitors and though we usually drive video on a dedicated machine, some setups use blackmagic cards in the hero systems. regardless of what interface you are using, avid or third party, its also very easy to exceed I/O in many cases and need additional cards just to facilitate more audio i/o. I agree that i prefer investing in fast current machines and have no qualms building hackintoshes, we haven't bought a mac pro in a few years now and we use less magma chassis than we used to due to most of our hacks having enough slots but in many cases its still required or desirable.

 

Your next example is about using a (granted oldy but goody) 9 year old PCI RME Hammerfall card along with a MBP.  How is that relevant?  We are talking about towers here, not MBP's reliance upon TB.  I completely grant that TB is aimed at MBP, iMac, and Mac Mini users as they have no internal options.  Your example is off target and completely irrelevant to any discussion about a tower's motherboard as you can choose to have PCI-e slots.

 

 

There is nothing old about RME interfaces, some of the PCIe solutions are quite new and even the older I/O devices, as you mention, work great if you buy the newer interface cards. It is relevant because one of the big shifts since apple abandoned the traditional mac pro is people want to be able to move the gear from one system to another, including real macs with no PCIe slots. It is very relevant. 

 

Further i would really like to point out that apples decision to base the "new mac pro" completely around thunderbolt may have major repercussions. We obviously have no idea how successful the new mac pro will be. But the fact is there is now NO apple computer with a pcie slot. The implication is that we will become more and more dependent on TB whether we like it or not as drivers for older or newer PCI devices will be less and less likely to be updated or written for OS X.

 

Whatever your reasons for building hacks, this becomes an extremely relevant issue and we really should be taking TB very seriously at this point as OS X users.

 

What could make sense as an advantage is that the modular nature of TB allows you to have fewer fully "built up" computers as you just connect TB devices to whatever you need.  That might be a valid argument, but I would challenge you to run the numbers.  It might take less organization on the part of your IT guy, but it probably costs you considerably more money in the final tally.  TB chassis and Magma chassis are not cheap.  Plus, having a tower with a bunch of cables running off it is not exactly a streamlined workstation.

 

I'm not sure what you're on about here so i'll just skip it.

 

When you do start addressing towers, and studio rigs, you say that you still "need a card to interface the Magma chassis so we might as well use TB."  What?  Why can't you just plug whatever you have in that Magma into the PCI-e lane directly?  Am I missing something here or are you?  

 

 

I think i addressed this above, magma chassis are still relevant to us in certain situations and especially on hero systems. They also (as an aside) in some cases allow us to leverage older, perfectly useful equipment such as PCI-X HD cards, etc. Obviously to interface to a magma card we need a PCI(e) card in the computer and the point is valid that if this can be done over thunderbolt (which it can) then its a better solution in many ways and it also allows connecting to a real modern mac. We ARE talking about heterogeneous working environments right? We do have lots of real modern macs around you know.

 

You end by saying that no hot swap TB is a deal breaker for a workstation, but you have not yet shown why TB is even needed in an audio workstation.  There is exactly one audio interface that uses TB and it also support USB 3.0 as both supply many times the bandwidth it needs.

 

I'm not sure which one your talking about, but avid has a native TB interface which has been quite popular, UAD Apollo also comes to mind, and again we are going to see more of this now that all real macs are TB dependent.  

 

Hopefully this explains things now. Maybe we work on different levels or in different worlds but the main point is OS X and TB are sort of in lock step now and it will be a hugely unfortunate thing if current and future hacks continue to have semi-crippled thunderbolt support.

 

My 2 cents plus 50.

g\

 

 

[theconnactic]

Genzai, thank you: you just spared me to post something in the lines you did (but perhaps less summarized). The difference here is we use Logic (so I could elaborate how that's nothing wrong in putting Logic in the same sentence as PT - I don't know about Cubase - but that would be way off topic).

 

All the best!

[maleorderbride]

My apology.  I got carried way and overstated things, but the core issue is still (supposedly) this motherboard with exactly 2x PCI-e x16 and 2x PCI-e x1.  With a full-sized video card installed that drops you down to 1x PCI-e x16 and 1x PCI-e x1.

 

Even with two fully working TB1 ports, this specific motherboard that sparked this conversation, would be less able to function as an high-end professional workstation (not just audio) than something like an ASUS P9X79 WS (Full size video card+ 5x PCI-e x16).  This Quo motherboard maxes out at 32GBs of RAM.  A P9X79 WS can have more RAM than OS X can address. 

 

That (edit: I mean Genzai) thinks that the most important thing holding this specific motherboard back from being an audio workstation is a lack of hot-swappable TB is a severe misunderstanding of hardware.  That is the main point I tried to address. 

 

I do not think that TB has no utility, or needs to die.  But I do place it below picking out the proper motherboard and CPU socket in order of importance as you will be able to add a TB PCI-e card to your system very easily down the road.  While you could replace your CPU/Motherboard/OS Install, that is certainly less straightforward than installing a single PCI-e card.  I imagine we can agree on that--yes?

 

Magma chassis are not irrelevant in your industry, but they are an expensive way to arrive at what you want.  No one wants to spend $3K to make up for 3x PCI-e lanes they could have purchased for $110 more than this motherboard costs.

 

So, hyperbole aside, does that make sense?

 

 

I will need to do some reading on high-end audio post production as I have never heard or spoken with an audio guy that faces the same hardware limitations as someone say color grading a 5K film, or waiting 12 hours for their 12-core to render less than 10 seconds of effects work.

 

Hopefully this explains things now. Maybe we work on different levels or in different worlds but the main point is OS X and TB are sort of in lock step now and it will be a hugely unfortunate thing if current and future hacks continue to have semi-crippled thunderbolt support.

 

My 2 cents plus 50.

g\

 

I agree--it would be unfortunate if current and future hackintoshes require a restart in order to use TB devices that are attached to them.  I don't see that as semi-crippled, but depending on your workflow I can see how that could be the case for some professionals.

 

I do not quite see it as lockstep so much as Apple doing what Apple does--making a Maginot Line just because they can--because they largely have a captive user base.

But unless Apple/Intel/Motherboard manufacturers inadvertently fix that for us, then I do not see it as likely that much effort will be funneled into fixing it on hacks.  Hopefully I am wrong!  But for the vast majority of (pro) users there is already a solid option that is less expensive and has many times more bandwidth.  It's called PCI-e.

[ILL]

I think who wants hot swap TB (for PCIe) devices never try to use a TB expansion box on a regular MBP.

We're using a Sonnet box with a RED Rocket and 2 SSD in RAID 0 for render out (or 2x Aja Kona 3G) with a MBPr (late'12) and there's no chance to have them working if pluged after the boot. You can't even let the MBP to sleep just ON or OFF, for this reason the Sonnet box doesn't have a switch

We have a TB to FW800, a TB to Ethernet and a TB to HDMI that works after boot but no way to have PCIe lanes.

I believe (please correct me if I'm wrong) that the PCIe funcionality via Thunderbolt is simply not hot swap.

For me is not an issue with a "less than 10" boot time" and I hope QUO will boot as fast as a MBP with the system installed in a SATAIII SSD.

Than if QUO decide to ship my KS motherboard may be I can write something more than thoughts....

[joe75]

http://support.apple.com/kb/ht5219#22

 

Get over it already..

[^Andy^]

http://support.apple.com/kb/ht5219#22

 

Get over it already..

So broken by design then - well that's handy!

[toastie]

http://support.apple.com/kb/ht5219#22

 

Get over it already..

 

Seems to be a limitation on TB when running Boot Camp.  Or does this limitation extend to OS X as well?  I would hope not.

[maleorderbride]

Seems to be a limitation on TB when running Boot Camp.  Or does this limitation extend to OS X as well?  I would hope not.

Looks like it extends a least partially to OS X as well as you cannot sleep and wake with Thunderbolt devices on a Mac running OS X.

 

26. Do Thunderbolt devices stay connected if my Mac hibernates?

If the battery in your portable Mac is depleted enough to cause your portable to hibernate, all Thunderbolt devices will be disconnected. After connecting to power and waking the system, restart your computer to reconnect your Thunderbolt devices.

 

 

 

OK, found some info about it and it looks like the issue ILL has is specific to the RedRocket card.  It is not hot-swappable on TB.  

http://www.reduser.net/forum/showthread.php?80613-Thunderbolt-RED-Rocket

 

So, any device that is on the "officially supported list" should presumably be hotswappable on a Mac's TB port with OS X.

[joe75]

25. If I eject a Thunderbolt device using the Taskbar tool or disconnect the Thunderbolt cable, can I reconnect it again without restarting?

No. A directly-connected Thunderbolt device that is removed with the Taskbar tool or by disconnecting the Thunderbolt cable will not be recognized until you restart the computer.

[ILL]

I finally received the tracking from QUO, hopefully I will start testing it sometimes next week.

I know that the rocket is absolutely not hot swap but I still believe that is not possible even in OSX without bootcamp (never used it so I can't check it, only hacks with multi OS).

I've got a TB drive, some adapters and the expansion box with more PCIe cards than the Rocket and a MBPr with 10.8.4. I'll test tomorrow or friday (CET) and I'll post my results, I do not have a TB Apple display to try.

[Lanceomni]

Is anyone else using Corsair CMY32GX3M4A1600C9 on their board?

 

I have not seen anyone else reference using this particular 32GB set.

 

I am working to narrow down an issue. 

[IronManJFF]

I was planning on using it (2x8GB)

[Lanceomni]

I am trouble shooting some random kernel panics with my build. I have experienced dock and safari crashes as well as kernel panics.

 

My current setup is a clean install (4th since I started):

1. Windows 8 (UEFI Install)
   
2. 10.8.4 (Vanilla install on H3A. Patched the audio)
   
3. 10.9 DP7
   

No additional hardware with the exception of memory, CPU and hardwired network.

 

My most recent KP yielded the following:

panic(cpu 4 caller 0xffffff80002b8655): Kernel trap at 0xffffff805248c600, type 14=page fault, registers:

CR0: 0x0000000080010033, CR2: 0xffffff805248c600, CR3: 0x000000003a6cf033, CR4: 0x00000000001606e0

RAX: 0x0000000000000000, RBX: 0xffffff8046e89854, RCX: 0x7fffffffffffffff, RDX: 0x0000000000000080

RSP: 0xffffff839735bef8, RBP: 0xffffff800068510f, RSI: 0x0000000000000046, RDI: 0x0000000000000000

R8:  0x0000000000000001, R9:  0x0000000000000000, R10: 0xffffff839735bcac, R11: 0x0000000000000000

R12: 0x0000000000000000, R13: 0x0000000000000001, R14: 0xffffff8046e89854, R15: 0xffffff839735bf50

RFL: 0x0000000000010296, RIP: 0xffffff805248c600, CS:  0x0000000000000008, SS:  0x0000000000000010

Fault CR2: 0xffffff805248c600, Error code: 0x0000000000000011, Fault CPU: 0x4 Kernel NX fault


Backtrace (CPU 4), Frame : Return Address

0xffffff839735bb90 : 0xffffff800021d626 

0xffffff839735bc00 : 0xffffff80002b8655 

0xffffff839735bdd0 : 0xffffff80002ce17d 

0xffffff839735bdf0 : 0xffffff805248c600 

Unaligned frame

Backtrace terminated-invalid frame pointer 0xffffff800068510f


BSD process name corresponding to current thread: mdworker

Boot args: slide=0

Ive ran Rember several times without an issue as well as maxing out each core with "yes > /dev/null" in several terminal windows. Niether caused a KP or resolved any errors.

 

I have been pulling down all of my Adobe applications from CC to induce some real world situations.

 

My next step will be to boot in SUM and run memtest individually on each memory module. I was under the impression that running Rember in OSX would only cause false positives. Ill give it a try anyway.

 

Under Windows 8, I ran IntelCPUBurn and LinX without issue. I did try to get HWmonitor to work but it crashed. No system crashes on the most recent build.

 

EDIT: I also noticed "org.netkas.FakeSMC 1.0.0d1" I always ran version 2.5. Not sure if that matters. I believe 2.5 changes were arbitrary but were there bugs that were fixed in the version(s) between. Not sure.

[maleorderbride]

Are you using the custom BIOS and clover?

 

If not, why do you have slide=0 as a boot arg?

 

 

 

Instead of Rember in OS X or even -s just do Prime95 (in Windows) for at least 8 hours, with custom, max threads, and RAM amount set to be 90% of your total RAM.  If you have bad RAM (or RAM settings/compatibility) that will cause a crash at stock CPU speeds.

 

If no crash there, then I would definitely start looking closer at software.

 

If you really want to test RAM then memtest86+ or memtest86 is still more trustworthy in my book than Rember in any mode.  Isn't Remember just a GUI for old versions of memtest86?

 

Why is my FakeSMC (CF Bundle)version 4.2 and you have version 1. whatever?

[Lanceomni]

Are you using the custom BIOS and clover?

No, I am using H3A and a vanilla UEFI install.

 

If not, why do you have slide=0 as a boot arg?

I did not set slide=0 as a boot argument manually. It is not a clover specific argument. It has existed for a while to turn off kernel ASLR. However, it may be needed if H3A has something that is occupying those addresses.

 

I did not zero out the drive; however, I have never had issues with boot loader residue from previous installs. Ill double check.

 

Instead of Rember in OS X or even -s just do Prime95 (in Windows) for at least 8 hours, with custom, max threads, and RAM amount set to be 90% of your total RAM.  If you have bad RAM (or RAM settings/compatibility) that will cause a crash at stock CPU speeds.

Ill give that a try.

 

If no crash there, then I would definitely start looking closer at software.

Same exact image runs fine on Gigabyte board.

 

If you really want to test RAM then memtest86+ or memtest86 is still more trustworthy in my book than Rember in any mode.  Isn't Remember just a GUI for old versions of memtest86?

I thought memtest86+? was a bit outdated for newer hardware and larger quantities of memory?

 

Why is my FakeSMC (CF Bundle)version 4.2 and you have version 1. whatever?

 

That is why I thought it was odd. It makes me curious about your comment about the slide=0. Although I doubt it is, I may do a clean install to a new hard drive or zero out my SSD and do a reinstall. Just to check.

 

Can anyone else confirm the version of FakeSMC?

[joe75]

org.netkas.FakeSMC 1.0.0d1 is from their list of hacked kexts in KextsExcludeList

[Lanceomni]

org.netkas.FakeSMC 1.0.0d1 is from their list of hacked kexts in KextsExcludeList

 

I just read about that. Interesting.

 

Can others who are running H3A and a vanilla install confirm the version of FakeSMC that is running?

 

Running the following in terminal should bring it up:

kextstat | grep -v com.apple

[toren]

I'm running H3A with a vanilla ML installation.  kextstat shows FakeSMC 1.0.0d1.

[Lanceomni]

I'm running H3A with a vanilla ML installation.  kextstat shows FakeSMC 1.0.0d1.

 

Well if others have the same then I guess that is what they used. That will likely have to be changed. Reading up on joe78's comment.

 

I am hoping for a source release for ozmosis. xpamamadeus' post (http://www.insanelymac.com/forum/topic/291655-ozmosis/) is pretty interesting. I might try it for one of my other boards.

[joe75]

the kextexcluded list shows fakesmc 1111 actually and thats in mavericks so yours might be what they added in firmware

[maleorderbride]

 

Same exact image runs fine on Gigabyte board.

 

 

 

 

How odd.  Nevermind about it being software then.

 

 

I thought memtest86+? was a bit outdated for newer hardware and larger quantities of memory?

 

 

You are correct, use memtest86, not memtest86+.  I couldn't remember which one of those two was outdated.  Pretty sure Rember is even more outdated though.  

 

That said, several times now I have found sticks of RAM which pass memtest86 for 5+ passes but cause instability in Prime95.  Swapping to a new stick and the problem goes away with Prime95.  If you have a RAM problem with Prime95 and run at 90% RAM load it often is displayed as a worker failure and your CPU load goes down to 88% or whatever--just not 100% anymore.  You probably will not get a bluescreen.

 

That could also indicate a problem with the BIOS settings though--so you could raise the VCCSA, VCCSA LLC, or just RAM voltage even.  Those are all ASUS terms and cannot remember what they are called in a GB BIOS, so sorry >_<

 

 

That is why I thought it was odd. It makes me curious about your comment about the slide=0. Although I doubt it is, I may do a clean install to a new hard drive or zero out my SSD and do a reinstall. Just to check.

 

Have you tried using Chameleon and legacy boot?  I understand you would want to use UEFI boot, but just to see if there is a link?

[Lanceomni]

maleorderbride,

 

Ill let this run all night and see what comes out of it.

 

It may very well be a memory compatibility problem. Nobody else has confirmed on the exact same model memory.

 

As far as using chameleon, id rather achieve the same success as others have with the board and H3A firmware.

[maleorderbride]

Well, after messing with the Ozmosis bootloader and adding my own kexts to the BIOS image, I would say you should change the FakeSMC version to a different one if you want and see if that helps.

 

Also, Ozmosis seems to use the boot-arg slide=0 by default.  It does so on my non-Quo motherboard with Ozmosis installed.