Discussion in 'Videocards - AMD Radeon Drivers Section' started by Hilbert Hagedoorn, Jan 8, 2013.
Ive never had to use it personally, but its a nice feature to have just in case.
No spoofing being used. Your idea about spoofing got me curious and I tried a few models, out of desperation with no luck.
But it seems this has been a two part issue. Originally it was the permissions levels on shortcuts and executables inhibiting Radeon Pro.
The crashing/performance was down to having no page file set on the SSD drive. Have amended to 8192 and all problems have stopped.
RP can crash or refuse to start if it's not installed with admin rights along with the RP support service. I've had the same thing happen a couple of times.
John told me the fix which was to reinstall RP as admin and make sure RP service starts automagically.
Ive seem him post that in the past, but not a problem ive ever had.
RP has been acting weird as of late on my end, none of my profiles would save. I just reinstalled it like 10mins ago and now it's working again....
The missing options in Tomb Raider definitely seem to be related to the the game only running in DX9 mode for some reason. I couldn't replicate your issue by spoofing a DX9 only GPU but it can be caused by using this registry tweak.
Let me guess it was BF4 crashing?
You prob have 8GB RAM which isn't enough to disable PF in BF4 while running CF/SLI, especially w/ 4GB vram. 12GB or 16GB you shouldn't have any trouble when PF is turned off.
I don't know claim to understand the inner workings of Frostbite3 but it's easy to get a broad idea on how it handles memory.
The amount of vram on the card is the same as it will tell DX to allocate for GPU data caching in system memory which seems rather greedy imho when there is 4GB vram available.
Most likely why SLI/CF rigs have the highest commit charge and crash the most.
Players who insist BF4 doesn't eat system memory generally don't understand Windows memory management, or where to look to see what's really being used.
Data which is not sharable and/or various mapped files not currently in use by the OS don't show in the Task Manager graph. But it still counts against Commit limit since none of it can be dumped.
I'd be interested to hear from someone who does understand how BF4 woks and how to reduce it's massive memory footprint.... Ironic since I don't even play, but still curious.
Tomb Raider will not give those options if I have API monitoring enabled. Maybe then it defaults to DX9?
BF4 was one of them, though it affected several games in different ways - all bad of course. Someone else mentioned that 16GB is the optimal RAM size to be able to disable page file without negative side effects. And I am tempted to upgrade, though the cost doesn't seem worth it.
Yes would be very interesting though doubt you'd get that information unless you knew someone on the inside
In all honesty I think the creator of RadeonPro is an ******* for accepting donations for continued development of RadeonPro only to ditch us without any recent release and refusal to release the source code.
It's the only decent tool us AMD users have to boot and it's already starting to break from the continued driver releases, what are we to do when it completely stops working?
The man has to feed himself.
Ive donated, and ive received s ton more value from RP than ive given in funds.
It works fine, a new release is not nessisary. John said if drivers break something, he will do his best to fix it.
Thats perfectly reasonable to me.
He has said from the very beginning, he will NEVER release the source, nor would i expect him to. Its his blood sweat and tears in there.
You come off as one of those immature entitled boobs.
Go try to find Ray and have him release his source for ATT.
Yeah, i didnt think so.
You're the only ******** I see around here.
It's not crowd funding.
Well, we just have to hope that someday, someone else comes around and does an even better tool than RadeonPro
If I was an *******, I would have done what this guy did. :banana:
That was uncalled for... you should know that I pay to keep the website and all other online resources. If that's not a good reason to keep accepting donations then maybe I should shutdown everything about RadeonPro that I am keeping online right now.
By the way, donations are and always were minimal and I never complained about it... and that never stopped me from keep updating both the software and the documentation. I am very grateful for all you guys who donated any amount, but keep in mind that I've always been clear about RP being closed source from the very beginning, you can find that in FAQ section of user's manual - http://www.radeonpro.info/manual/frequently-asked-questions/where-can-i-get-the-source-code
You should ignore retards with 5 posts posting retarded ****.
I don't think people care about the program being closed source or the existance of donations. Rather, they, ahem... are frustrated from the program being abandoned, because everyone had huge expectations and now is left with CCC, basically.
That can be an issue with detection code, you can fix it by creating an entry for TombRaider.exe in compatibility list and tick "Ignore D3D9"
Hi, congrats on the new job...
I installed Raptr recently out of curiosity and tbh it was a disappointed.
The features which make RP so great don't appear to be implemented with Rptr which really surprised me since RP development brought you to AMD's attention to begin with - correct me if I'm wrong?
Do AMD realize that now RP is no longer under development many users will switch to Nvidia just because of nVIDIA Inspector?
They need to keep every user they have....
^ Giving him a job does not automatically make them owned RP. There's a bigger price to pay if they want the source code imo.
I didn't say that.
If John develops software for AMD using his own code, that doesn't mean AMD own the source.
RP was impressive enough for AMD to headhunt him in the first place, so logically RP features should be turning up in Raptr.
But if he leaves AMD can't use his code in any future releases without Johns permission.