AMD Radeon Software Crimson ReLive 17.3.3 Download & Discussion

There are two interesting variables in your comparison.

1. Windows Build # (because you did first comparison test in July 2016)

2. you jump from 16.x to 17.3.3


What you can try to get a better comparison of the drivers is:
Reinstall your old driver on your system and re-run the benchmark, to get the old driver result with current Windows Build

Also if you have a dual boot system, try the other Windows (7 / 8.1) for comparison.

 

Sadly I didn't uploaded some past results, but I remember with 17.2.1 i got the first visible improvement in DX11 dc, but DX12 was still at 19million while mantle was 20-21million and DX12 gaming was rock solid.
To me this result seems like a really bad regression and it's maybe an explanation of 17.3.X instabilities in DX12 games.

 

That doesn't entitle him to call people clowns when they have legit support issues. He has been doing this for some months on top.

 

Since when your friend's GTX 760 became yours?

I own a GTX 770 4GB Windforce at home as I need Cuda's Bruteforcing, which I have stated many times before. And I have access to many more GPUs from both camps at work. Yes, I "work" and not just game with GPUs, so basic functionalities like switching between clock states is a necessity for me. Leaving a PC on standby that has a Fury X equipped for 10 minutes only to find a garbled display to greet me that forces me to replug the monitor or reboot the PC at times isn't something I am allowed to talk about, right?

Your friend's GTX 760 with a fan failure makes nvidia a failure, amiright? :3eyes: Pity you've no idea how Sapphire Vapor-X cards are prone to fan failures.

Who's everybody, btw?

The way I see it, you and Agonist are the only two guys in this section of the forum who has an issue with anyone that shares their findings of problems they have with AMD's GPUs. And both of you are like if you don't like the support, go buy nvidia, it's as if they bought the cards with your money and not their own.

Very nice gains on those 3DMark benches mate. :infinity: Keep up the good work.

 

People are this forum from which he sought a reaction, they ironically make him a clown

 

Yeah AMD drivers or AMD self not that great getting weird desktop and game performance which I did not get when I was using Nvidia.

 

For those who were waiting on a solution for Andromeda CF, they updated the information on the driver page, confirming that Bioware needs to release a game patch before you can get CF working.

http://support.amd.com/en-us/kb-art...mson-ReLive-Edition-17.3.3-Release-Notes.aspx

"New Multi GPU profile:

Mass Effect™: Andromeda (DirectX®11)
Multi GPU support is pending an application patch."

 

LOOOL :banana:


Anyway I used to have quite a few errors in my Event Viewer. I used this to get rid of Event ID 10016 (the only proper fix on web), and now my Even Viewer is clean. The only thing left is:

me, me, me... elp plx

Code:

Activation context generation failed for "c:\program files\amd\cim\bin64\SetACL64.exe". Dependent Assembly Microsoft.VC80.MFC,processorArchitecture="amd64",publicKeyToken="1fc8b3b9a1e18e3b",type="win32",version="8.0.50608.0" could not be found. Please use sxstrace.exe for detailed diagnosis.

 

btw... clean, fresh install. go figure...

 

Not delusional, but based on my experience, since I actually brought the card, I tested more than 500 games, since I perform benchmarks on youtube with both Nvidia and AMD.

Call me fanboy whatever you want, but go take a look too on the nvidia bugs that are on the known issues list for more than 2 years already.

Problems exist on both sides. AMD is having problems with ESO, so what?
Nvidia can be having problems with other games, the difference is that you end up playing ESO with an AMD hardware and unluckly there is problems with AMD hardware.

Plus, it's not solved, and I would not hold my breath, since it's not even listed on the known issues, and the drivers branch didn't change. So it's pretty obvious that until a new branch is out or it is listed on Known Issues, that problem isn't going to change, no need to keep spamming every thread with the Grass problem of the ESO, just keep reporting, until it appears on the list of things to solve.

Cheers

 

I think I solved this problem but I won't be 100% sure until tomorrow when I start up my machine.

What I did was remove StartCN from auto starting. The run entry is located at HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows\CurrentVersion\Run.

Doing this doesn't seem to affect things. I don't overclock so it's possible it won't overclock at startup. I can still run AMD Radeon Settings from the desktop context menu. If you overclock and what to try this let me know if your O/C takes affect on a cold startup just to be sure.

 

I had a very similar setup to yours a while back (2 7970s) and the gtx 680s I had in my nvidia setup at the same time was nearly always the far more consistent platform. But bear in mind I was heavy into the newest big name titles and AMDs drivers for such games were very often lacking for stability and consistent performance.

Unless every one of those 500 games you played was mature (e.g. had multiple driver and game patches, played many months after release) I'm really finding it hard to believe that you had no major issues with AMD's drivers. (aka your statement "every game works good").

The ESO bug is not an issue for me since I turned off grass for performance purposes on both my AMD and Nvidia systems (although he also mentioned that some screen corruption STILL occurs in the background), but it was an issue somebody brought up in this very thread that illustrates a rather consistent nagging issue with AMDs drivers throughout the years. They have these sort of problems often in my experience, and it can take FOREVER for them to fix it, especially if the game in question isnt a recent release.

 

There you go.
Like I said judging by my experience.
I don't use Multi-GPU, I always used Single GPU, and probably that's the reason why I've never had any problem.

CF in the past was a pretty piece of Turd comparing to SLI, that is why when benchmark websites started to use Nvidia's FCAT, it was discovered a ton of problems in the drivers.
Today the story it is much different, but I still believe that CF are less supported than Nvidia.
Cheers

 

StarCN is realtek control panel? I run my realtek with stock Windows drivers, so no StarCN.exe for me.

My latest Event ID 33 error is 20 hours old, and my PC has been turned off and on and booted up in the meantime... im hoping maybe no more Event ID 33 for me.
What I did is I reinstalled/repaired those Visual Basic(C++?) Runtimes from AMD driver package subfolder.

Though I dont have high hopes tbf

 

It's StartCN which executes RadeonSettings.exe. The startup location has been moving the last few releases between the Registry entry I gave or it might be a scheduled task.

 

I have my Radeon Settings (ie. radeon control panel) running in my tray, but no StartCN for me, nowhere:

https://www.dropbox.com/s/suq0n4w2hlv9afy/Screenshot 2017-03-23 22.42.41.png?dl=0
https://www.dropbox.com/s/jj5eefylvkn0445/Screenshot 2017-03-23 22.43.48.png?dl=0

All I have been able to find are instances of StartCN.xml in my Registry
"C:\Program Files\AMD\CNext\CNext\StartCN.xml"

 

Ah ok... found it.

https://www.dropbox.com/s/yf9pfz1onfjxcwr/Screenshot 2017-03-23 22.57.37.png?dl=0

 

If your Radeon Settings start up when Windows starts it's got to be somewhere. Look for where RadeonSettings.exe gets started. I didn't install ReLive so I wonder if that's a factor with this issue

 

No. I have Relive.
Its the issue with Microsoft Visual runtimes.

Now I installed this:

https://www.microsoft.com/en-us/download/confirmation.aspx?id=29
https://www.microsoft.com/en-us/download/confirmation.aspx?id=15336

 

Let me know if it works. If the Runtime version number is not the same as what's in the event log message then it might not work.