PC Processors

[walvest]

I’ve discovered that the essential licensing software for music production, iLok License Manager, does not run on the 9900X. Additionally, the DAW program Cakewalk, while it installs on the 9900X, does not function properly when executed. UA Connect, which is necessary to use UAD plugins, also fails to run. Since many programs, virtual instruments, and plugins rely on iLok License Manager, it is currently impossible to produce music on the 9900X.

I’ve seen other users on Reddit experiencing the same issues with the 9900X and 9950X. It’s surprising to see such problems with well-established programs like iLok and Cakewalk, which have been consistently updated and have run smoothly on any computer until now. I hope AMD can provide a solution to this issue soon.

[misterj]

Thanks, OregonVideoPhotography hope it helps and maybe the software developers learn to test on available HW. John.

[AP2]

I just discovered this issue from a Facebook post. I've been back and forth with iLok and Soundtoys support for weeks and they keep sending me 'fixes' that don't work. They never mentioned the compatibility issues with the 9950x, although now I realize they have known about it the whole time. Very frustrating.

[Lukas108]

Yes, in my case the same ... no responsibility.  All my plugins works, Sound Toys not. Time to abandon this bundle? 

[OregonVideoPhotography]

I just sent a bunch of info to AMD. I will keep posting what I learn here. Don't want to escalate this yet, but in reality if it is not fixed it is a denial of service and there are other ways to fix this. Every keep requesting info every week fro ilock, pace, and AMD. It's the squeaky wheel that gets fix. Lukas are you using software or USB? it is very interesting that your vst work - are they iLock Protected? 

[elstaci]

Reading this thread which AMD says it is a CPU "Bug" or "Flaw" I find it odd.

The reason why, Personally, I would believe a CPU Bug or Flaw would affect a wide range of programs yet the only Users complaining are those that use iLok Manager and its supported programs and other types of professional programs that seems to be related, at the moment, with Professional Music type programs.

I haven't read any other Users mentioning about incompatibilities of the Zen 5 processors with other programs.  Of course I could be completely wrong in that assumption.

If other Non-iLok Manager type programs are working with Zen 5 it seems like it is a iLOK Manager programming issue rather than a Zen 5 bug or flaw as one User in this thread mentioned.

If a hot fix was able to make iLok Manager somewhat compatible but not 100% compatible proves that it is a programming issue rather than a CPU firmware issue.

If it is due to a Zen 5 Security protocol or firmware feature I would  believe it would affect a lot more professional programs like iLok Manager. 

iLok Manager will need to collaborate with AMD to find out how to make iLOK Manager compatible with whatever CPU Firmware feature is causing it to not work correctly.

just my opinion.

[misterj]

What makes you think AMD said this?

'... which AMD says it is a CPU "Bug" or "Flaw"...'

John.

[martel80]

I might be wrong but I think Universal Audio Plugin of the Native type are affected as well. And as far as I understand, they are not relying on Ilok. I might be wrong though

[Lukas108]

Reply from Sound Toys. It seems like are waiting for AMD firmware update. 

 

[AP2]

I just received the exact same response (copy/pasted) from iLok...

[OregonVideoPhotography]

Posted on the iLock site:

"

Unfortunately, we have discovered that this is an issue with the AMD Zen 5 CPU microarchitecture that affects iLok-enabled software. This issue prevents the software from running correctly on Windows systems with these CPUs.

The problem is due to a bug in the CPU itself, which impacts the way certain protection features in iLok-enabled software work. We have reported the issue to AMD and are waiting for their response.

Currently, there is no immediate solution or workaround. We are hopeful that AMD will release a firmware update to fix the issue. In the meantime, we are also exploring potential workarounds, but these would require updates to the software from your software provider.

For now, we recommend monitoring updates from your software provider for any patches or updates that address this issue. Unfortunately, until a fix is available, the affected software may not function properly on your system with an AMD Zen 5 CPU.

Please check with your software provider or AMD for the latest updates on this issue. We will also provide updates as we receive more information from AMD."

"Bug in the CPU itself" impacts protection features in iLok-enabled software", wow could they be any more selfserving? It's always been the software developers developing software, not CPU developers. And now they are passing off possible work-a-rounds if the software providers will modify their software.

It looks like we are going to see some fireworks on this. "Bug in the CPU"....they better have real proof to publish that and feel that they've written the right code for it. Someone put their foot in their mouth with this "notice". I am sure AMD will respond once they see this notice.

[misterj]

Thanks, OregonVideoPhotography. I was about to ask for evidence for this "Bug in the CPU itself." I really hope that AMD responds soon, very soon. If iLock reported this to AMD I would expect they would post what they said to AMD so others can try to reproduce the problem. Like, why does this only seem to effect the music industry? John.

[OregonVideoPhotography]

Yes, I'm running Photoshop, Resolve PRO, and Mercalli Pro all at the same time and no issues. iLock is the issue IMHO. You hear of a new CPU and as a developer, you don't try it out to see if your program/vst will work on it? As I said before the word "bug" means a real issue in the firmware: I don't see it in doing heavy duty video with sound on Resolve PRO. AMD is a 277 billion dollar firm, I'd be surprised if iLock breaks 50 million as it is privately held; careful if you wake the Giant. It appears to me that the statement on the iLock website is trying to blame software vst developers and AMD for an issue that is directly caused by the new iLock coding. Whenever has it been up to the CPU manufacture to test every program; it is up to the software developers to say what hardware is needed to run their programs. I know for a fact no one told SweetWater about the issue so they could start to issue warnings. It will be very interesting to see how this turns out. It will be interesting to see how many of the music vst/programs providers stay with iLock. East West was proactive in trying to help me (I own OPUS and it did not work, Play did work with the "Modified" hot fix iLock) and had great communications. The issue is now before all of the key players, and we can only wait until action is taken. I will start looking at non-iLock options. What are the best non-iLock vst orchestras?

[Dotje]

Softube response on 16 of stepember 2024 on my issues with ILOK (29 july 2024):

This is the first case where I see this issue. Another came in over the weekend, so this is either rather new, or very uncommon.

Roadmap:
Until we have a clear picture on what is happening, and we can reproduce this in-house it is very hard to give you any estimates there.

It though now seems like iLok have identified the issue on their side, this likely means that we can properly look into it as well. Again, we can't nor should be able to correct code on the iLok side of things.

My response:

Dear ....

I am deeply disheartened by your response concerning the iLok compatibility issues with AMD's Ryzen 9 9900X and 9950X CPUs. It appears my concerns, raised eight weeks ago, have not received the seriousness they deserve.

Confirmation of the Problem:

The fact that an iLok engineer (XXXX of PACE Anti-Piracy Technical Support, Incident XXXXXXX) identified a quick fix on my system confirms this issue is widespread. Similarly, multiple users on AMD platforms and potentially your internal testing should have flagged this problem.

A simple communication with iLok within the past eight weeks could have verified the issue and initiated support efforts. Their website itself acknowledges the AMD Zen 5 compatibility concern (link: https://help.ilok.com/faq_ilm.html#amd_zen_5).

The Impact on My Work:

This incompatibility has severely affected my workflow and lectures. I rely heavily on 39 Softube plugins, and losing access to them has crippled my project templates, including Console 1 hardware and the Flow subscription. Furthermore, older projects may require significant workarounds or complete replacement of Softube plugins. Additionally, my workflow and lectures centered around Softube synthesizers have been entirely disrupted.

Wasted Resources and Lack of Support:

Eight unproductive weeks have forced me to unnecessarily replace my hardware (CPU, Memory & Motherboard) and shift to alternative VST brands that avoid iLok or offer different protection solutions. Despite this significant financial and logistical burden, I received minimal support or acknowledgement from Softube.

Desired Resolution:

Moving forward, I expect a more proactive approach. Please collaborate with iLok to develop a permanent solution addressing this compatibility issue and ensure my continued access to the Softube products I have invested in.

Furthermore, I request clear communication regarding the timeline for a fix and a plan to rectify the impact on affected users.

I look forward to a satisfactory response and a solution to regain access to my essential music production tools.

[misterj]

Thanks, Dotje. John.

[walvest]

I received this email today. It seems to be an issue with AMD itself. It's a desperate situation where we have to keep waiting for the patch.

Hello,

Unfortunately, we have discovered that this is an issue with the AMD Zen 5 CPU microarchitecture that affects iLok-enabled software. This issue prevents the software from running correctly on Windows systems with these CPUs.

The problem is due to a bug in the CPU itself, which impacts the way certain protection features in iLok-enabled software work. We have reported the issue to AMD and are waiting for their response.

Currently, there is no immediate solution or workaround. We are hopeful that AMD will release a firmware update to fix the issue. In the meantime, we are also exploring potential workarounds, but these would require updates to the software from your software provider.

For now, we recommend monitoring updates from your software provider for any patches or updates that address this issue. Unfortunately, until a fix is available, the affected software may not function properly on your system with an AMD Zen 5 CPU.

Please check with your software provider or AMD for the latest updates on this issue. We will also provide updates as we receive more information from AMD.

Best Regards,

Katie Bosserman

iLok.com Technical Support

[OregonVideoPhotography]

Anyone want to guess how long a "work-a-round" or fix is going to take. Is Spitfire the best non-iLok orchestra?

[OregonVideoPhotography]

Anyone else find iLok's post disingenuous for various reasons?

[giacomopasquali]

Same reply they sent me.

I have quite a lot of doubts that this is really related to AMD (And I write to them too).

I'm not a developer so I might be wrong of course but, how is it possibile that with the same protection some plugins load and some not? Even some of the same brand!

This is my situation at the moment:

NOT WORKING:

• Sonnox (NO ONE)
• SoundToys (NO ONE)
• Softube (crash + not finding license in the softube central app)
• NeuralDSP Rabea (app shutdown, Error 2001), Cory Wong (Asking for activation), GojiraX (freeze), ToneKing (app shutdown)
• Slate FG-X2

WORKING:

• LiquidSonics
• NeuralDSP Morgan Amp Suite, Soldano SLO100X, Nolly X, Mateus Asato, Plini X, Fortin Nameless X, Mesa Boogie IIC+
• Pulsar 8200, Massive,
• Slate Trigger,
• Empirical Labs Arousor
• KIT Plugins
• UAD Spark

[Lukas108]

Hi. Did you test it on fresh system or system that is previously installed in other AMD CPU? 

I noticed that all project created on 5950X crashes with Sound Toys, but when I create new project and insert any of SoundToys then it does not crash.  

Obviously ... I can't open any of previously created project. 

[giacomopasquali]

My system was previously on a 14900K. Switched MOBO and CPU to a 9950X, of course with the driver work needed.

[misterj]

giacomopasquali, I have failed to remind all to get their AMD drivers from here. Please do not get your AMD drivers from the MB vendor. They are usually down level as are the drivers from Microsoft. Thanks, John.

[Dotje]

Hauptwerk

AMD Zen 5-series CPU bug prevents iLok working
Sat Sep 14, 2024 4:29 am

PACE have notified us that a bug has been found in the microarchitecture of the AMD Zen 5-series CPUs that were released in July (branded under the “Ryzen 9000”, “Ryzen AI 9”, “Ryzen AI 300”, and possibly more names), which prevents any iLok-licensed products from working (including Hauptwerk).

They have reported it to AMD and are awaiting their response (hopefully a CPU firmware update to fix it).

In the meantime we'd strongly recommend avoiding buying those particular CPUs for use with Hauptwerk (or for use with any other iLok-licensed products).

Best regards, Martin.
Hauptwerk software designer/developer, Milan Digital Audio.

[misterj]

Dotje, you got me here. Am I just not knowledgeable? I cannot find any reference to AI in Ryzen 9900X. What am I missing? This does point out to me that all users posting here, Please post the HW components you are running. Maybe there is some similarities. Thanks, John.

[Dotje]

UVI took 1 month (from 1 august ) to responce om y bug report on Sep 19, 2024

We have just been notified by PACE that there is indeed a bug in the AMD Zen 5 CPU microarchitecture that affects the PACE License Support.

There is no immediate solution or workaround for this bug unfortunately, PACE is currently working with AMD to resolve the issue as soon as possible and hopefully AMD will release new firmware that will fix the issue for end users at some point.

Feel free to ask if you have any further questions.
All the best from Paris,
Support Team

[stegeocan]

Hi, another 9900X user here unable to use plugins after upgrading.  I've updated to the latest AGESA 1.2.0.2, tried all sorts of random things (including disabling one CCD and changing almost all security/memory settings) and had given up hope of getting my Neural DSP plugins working soon, but spotted this:

"WORKING:

• LiquidSonics
• NeuralDSP Morgan Amp Suite, Soldano SLO100X, Nolly X, Mateus Asato, Plini X, Fortin Nameless X, Mesa Boogie IIC+"

Plini X is one the Neural  DSP plugins I have and it behaves the same as the others and crashes with a BEX64 error whenever I try to open it (standalone or in FL Studio).  I get that same error with Tim Henson, Gojira X and Petrucci.

That is also the same error I got when trying to start the PACE service, until I installed the hotfix (v5.10.0 B1 (b123,12)) which resolved that.

I can get the Plini (non-X) version to work if I go all the way back to the 32 bit version, but all the latest 64 bit plugins all give me the same problem.  

Are you able to open the standalone plugin?  I'm using Win 11 Pro (clean install 23H2 latest updates, latest drivers from AMD), X670E 9900X, 2x32Gb RAM. RTX3080. 

You also say Gojira X freezes for you whereas for me I immediately get the BEX64 error (same as all the others) and it doesn't open (standalone) or crashed FL Studio (plugin).

If I deactivate plugins in the iLok License Manager, then opening the plugin (standalone or in FL studio) will prompt me to activate the plugins, and that works fine, except that once activated I again get the BEX64 error.

Did you reinstall windows after your HW upgrade?

Faulting application name: Archetype Plini X.exe, version: 1.0.1.0, time stamp: 0x66059948
Faulting module name: unknown, version: 0.0.0.0, time stamp: 0x00000000
Exception code: 0xc0000005
Fault offset: 0x00000000001416a8
Faulting process id: 0x0x3CC0
Faulting application start time: 0x0x1DB0F0DC6F5BDD8
Faulting application path: C:\Program Files\Neural DSP\Archetype Plini X\Archetype Plini X.exe
Faulting module path: unknown
Report Id: 072464d8-6fc4-4f0d-9e6b-668967ed3b61
Faulting package full name:
Faulting package-relative application ID:

Thanks!

[misterj]

Thanks, stegeocan. Great to have the Event Viewer to look at. Please look in C:\Windows\Minidump and if it is not empty, please compress it and up load it somewhere I can get it. I think your application provider would like to see it also. Please tell me what the BEX64  means. Thanks, John.

[stegeocan]

Thanks for looking into this.

There is no minidump in C:\Windows\Minidump as the system doesn't actually crash, but I have the application crash dumps.  There are three different ones in here:

https://drive.google.com/drive/folders/1-3QN8cHp_0rJdvJhiA_17DM8FSQriuM6?usp=sharing

BEX64 is a memory buffer overflow exception:

"The thread tried to read from or write to a virtual address for which it does not have the appropriate access."

I do have DEP (Device Execution Protection) enabled at the moment but have previously tried disabling it (bcdedit.exe /set nx AlwaysOff ) and got the same problem.  It is as if there is some hardware level DEP going on which I can't disable.  I've disabled core isolation and every other type of protection I can find and that doesn't change things either.

I have been in contact with Neural DSP and iLok.  iLok were going to do a remote session with me but cancelled it and said they don't need to anymore. 

It is infuriating as it feels as if there is a workaround just one setting tweak away!

Thanks for your time looking at this

[misterj]

Thanks, stegeocan. DLed all three and will take a look. John.

EDIT: Need to install visual studio.

[misterj]

stegeocan, VS did not help me much but the three dumps you sent of three different .exe files ended in ntdll.dll with the same error-'reading or writing at an address that is not permitted'. The module before the crashing one is always kernel32.dll and the path always looks the same to me. There are a bunch of .exe's with names ending in 32. This may mean a lot of code is original 32 bit code. Maybe worth considering by the developers. Sorry I could not glean much from the dump files. John. 

[stegeocan]

No worries John, thanks for looking.

ntdll.dll in System32 folder is actually the 64 bit version (the 32 bit version is in SysWow64). 

That's the same for all DLLs although I realise that sounds the wrong way around.

https://stackoverflow.com/questions/949959/why-do-64-bit-dlls-go-to-system32-and-32-bit-dlls-to-sysw...

ntdll.dll is the interface iLok uses to access the hardware via the kernel.  I guess it is doing something 'clever' to identify your hardware here.

One thing I haven't tried yet is using an different TPM and turning off the fTPM in the cpu.  Will probably try that at some point. 

Out of frustration I ordered an iLok USB key to try as I am hoping that gets checked differently.  Unfortunately after ordering what was apparently 'in stock' I now have a lead time of 2 weeks.  I suspect I will regret that purchase.

Will update here if I make (or fail to make) progress.

Thanks again

BTW, is it me or is this forum really awkward to use.  It feels like it is designed to be used for very short threads only

Thanks again.

[AP2]

It's not you. It's getting impossible to navigate this thread.

[misterj]

Sorry, stegeocan, I failed to answer your BTW. I agree with stegeocan below. If you don't realize to click all the LOAD MORE REPLIES scattered throughout the thread then searches mean nothing. Please remember that AMD is a great hardware company but not so much when it comes to software and documentation. John.

[OregonVideoPhotography]

I was curious as to what Microsoft's Copilot would say on the issue, I ask what is the issue and could it really be called a Bug:

"

You make a valid point. The responsibility for ensuring software compatibility typically lies with the software developers, who should design their programs to work seamlessly with the operating system. The term “bug” might indeed be misleading in this context if the issue stems from how the software interacts with the firmware rather than a flaw in the CPU itself.

In the case of iLok, it seems the software might be relying on specific firmware functionalities that are not behaving as expected with the new Ryzen processors. This can create compatibility issues that appear to be hardware-related but are actually due to the software’s reliance on certain firmware behaviors.

It’s a complex interplay between hardware, firmware, and software, and resolving such issues often requires collaboration between CPU manufacturers, firmware developers, and software companies. Keeping an eye on updates from all involved parties is crucial."

No response yet from iloc or AMD about what is the plan to fix the issue.

[AP2]

I know there are relatively few of us who use the new 9000 series CPUs, but we need to find a way to make more noise about this or it will never get fixed. AMD, iLok, and the plugin developers seem to be content either ignoring the problem or blaming someone else.

If iLok and the devs truly think AMD is to blame, why are they not sending out an email to all of their customers warning them to not buy or build PCs using these CPUs? And when is AMD going to respond to all of this? This is truly a s-show.

[desmond_dales]

For me, this is pretty obvious. Ilok is still 32bit, installer looks like from xp times...i think they literally haven't changed or updated their software for more than a decade. This cannot be a cpu bug. The thing is that their pretty old software cannot properly address instructions in new processors. Maybe they fired all programmers long ago and now keep on posting this cpu bug nonesense. On the other hand, nowadays in these online days, there is no need to use ilok. Those plugin vendors who are not lazy has already developed their own licensing management. iLok was a good idea back then, but not for 20's. I only feel sad for SSL plugins...and still i hope pace hired some programmers to make a new ilok for 21st century

[Lukas108]

I think we need to write to news portals. They will willing to write about that. 

[OregonVideoPhotography]

Well here is the bigget issue I think. If Chat was right, the 277 billion company (AMD) might want to make the little company pay for 1. issuing a statement that wasn't true, and 2. Not happy that at a roll out point the little company tried to blame AMD for something in firmware that works with the operating system but not with that firms product (which may or may not be authorized to use a specific part of the firmware to make their product run). Hard to tell if AMD and iLoc are working on a solution or if the lawyers are lining up. The issue for us users is something call denial of service. All in all this could either be solved by iLoc and its customers changing things in software that rely only on the operating system, AMD and iLoc working together, or the courts deciding the issue who is responsible for the denial of service to use software the customers who bought the product assuming the product would working in the operating system (after all did iLoc publicize that their product would work only with specific firmware in CPU's - don't think so).  We are all small fish, however iLoc's customers (software providers) are the ones with a lot to lose in this situation. As long as we keep the issue alive by posting tech support requests as well as having discussions, well that is about all we can do.

[AP2]

I'm trying to think of ways to reach the big PC hardware channels, like Hardware Unboxed, JayzTwoCents, Linus Tech Tips, etc., as well as any big music production channels that would warn producers not to use the 9000 series AND not to buy iLok products. I think that's the only thing that can light a fire under those companies.

[Dotje]

I pitched this a while back, but crickets.

[OregonVideoPhotography]

Just keep communicating with the firms involved, that way they know the issue won't go away.